This is the accessible text file for GAO report number GAO-09-866 
entitled 'New Drug Approval: FDA Needs to Enhance Its Oversight of 
Drugs Approved on the Basis of Surrogate Endpoints' which was released 
on October 26, 2009. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as part 
of a longer term project to improve GAO products' accessibility. Every 
attempt has been made to maintain the structural and data integrity of 
the original printed product. Accessibility features, such as text 
descriptions of tables, consecutively numbered footnotes placed at the 
end of the file, and the text of agency comment letters, are provided 
but may not exactly duplicate the presentation or format of the printed 
version. The portable document format (PDF) file is an exact electronic 
replica of the printed version. We welcome your feedback. Please E-mail 
your comments regarding the contents or accessibility features of this 
document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

Report to the Ranking Member, Committee on Finance, U.S. Senate: 

United States Government Accountability Office: 
GAO: 

September 2009: 

New Drug Approval: 

FDA Needs to Enhance Its Oversight of Drugs Approved on the Basis of 
Surrogate Endpoints: 

GAO-09-866: 

GAO Highlights: 

Highlights of GAO-09-866, a report to the Ranking Member, Committee on 
Finance, U.S. Senate. 

Why GAO Did This Study: 

Before approving a drug, the Food and Drug Administration (FDA) 
assesses a drug’s effectiveness. This assessment may be based on 
evidence showing that a drug has a positive impact on a surrogate 
endpoint—a laboratory measure, such as blood pressure—instead of more 
direct clinical evidence, like preventing strokes. After approval, FDA 
often requires or requests a drug sponsor to further study the drug. 
Concerns have been raised about FDA’s reliance on surrogate endpoints 
and its oversight of postmarketing studies. This report provides 
information on (1) all drug applications approved based on surrogate 
endpoints in FDA’s accelerated approval process, (2) a subset of 
applications for potentially innovative drugs approved based on 
surrogate endpoints under FDA’s traditional process, and (3) FDA’s 
oversight of postmarketing studies. GAO identified drugs approved based 
on surrogate endpoints, obtained the status of related postmarketing 
studies, and reviewed FDA’s oversight of a sample of 35 studies it 
required under its accelerated approval process, selected to include 
studies which were at varying levels of completion. 

What GAO Found: 

FDA approved 90 applications for drugs based on surrogate endpoints 
through its accelerated approval process from the creation of the 
process in 1992 through November 20, 2008, and about two-thirds of 
postmarketing studies have been closed. FDA created the accelerated 
approval process to expedite the approval of drugs which are designed 
to treat serious or life-threatening illnesses and are expected to 
provide meaningful therapeutic benefits compared to existing 
treatments. Under this process, 79 of the 90 applications were approved 
for drugs to treat cancer, HIV/AIDS, and inhalation anthrax. Because of 
the need to expedite approval, FDA approves drugs under this process 
based on surrogate endpoints which are not yet proven substitutes for 
clinical endpoints, but does require that drug sponsors complete 
postmarketing studies to confirm the drug’s clinical benefit. FDA had 
required drug sponsors to conduct 144 postmarketing confirmatory 
studies associated with these 90 applications, and as of December 19, 
2008, classified 64 percent as closed—meaning that drug sponsors had 
met FDA’s requirements for these studies or FDA determined the studies 
were no longer needed or feasible. However, several of the remaining 
studies have been classified by FDA as open for an extended period. 

FDA approved 69 applications on the basis of surrogate endpoints for 
new molecular entities (NME)—potentially innovative drugs containing 
active chemical substances that have never been approved for marketing 
in the United States in any form—through its traditional approval 
process from January 1998 through June 30, 2008. These 69 NME drugs 
accounted for about one-third of the 204 applications for NME drugs 
which FDA approved through its traditional process during this period, 
many for drugs to treat cancer, heart disease, and diabetes. Unlike 
surrogate endpoints used in the accelerated process, FDA considers 
those used in the traditional process as valid substitutes for 
demonstrating the clinical benefit of drugs, and thus does not require 
sponsors to complete postmarketing confirmatory studies. However, FDA 
requested that sponsors complete 175 postmarketing studies to obtain 
other information on many of these NME drugs, and as of February 13, 
2009, FDA classified about one-half as closed. 

Weaknesses in FDA’s monitoring and enforcement process hamper its 
ability to effectively oversee postmarketing studies. FDA has not 
routinely been reviewing sponsors’ annual submissions on the status of 
studies in a timely manner. It has little in the way of readily 
accessible, comprehensive data to monitor studies’ progression and does 
not consider such oversight a priority. FDA is implementing initiatives 
to improve its oversight, but it is too early to tell if they will be 
effective. Although FDA has authority to expedite the withdrawal of a 
drug from the market if a sponsor does not complete a required 
confirmatory study with due diligence, or if a study fails to confirm a 
drug’s clinical benefit, it has not specified the conditions that would 
prompt it to do so. It has never exercised its authority, even when 
such study requirements have gone unfulfilled for nearly 13 years. 

What GAO Recommends: 

GAO recommends that FDA clarify the conditions under which it would 
utilize its authority to expedite the withdrawal of drugs under its 
accelerated approval process. FDA disagreed with the need to develop 
such clarifying guidance. GAO believes doing so would enhance FDA’s 
oversight. 

View [hyperlink, http://www.gao.gov/products/GAO-09-866] or key 
components. For more information, contact Marcia Crosse at (202) 512-
7114 or crossem@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

FDA Has Approved Many Applications Based on Surrogate Endpoints through 
Its Accelerated Approval Process and about Two-Thirds of Postmarketing 
Studies Have Been Closed: 

FDA Approved about One-Third of NME Drug Applications Based on 
Surrogate Endpoints through Its Traditional Process and about Half of 
the Postmarketing Studies Requested Have Been Closed: 

FDA's Oversight of Postmarketing Studies Is Hindered by Weaknesses in 
Its Monitoring and Enforcement: 

Conclusions: 

Recommendations: 

Agency Comments and Our Evaluation: 

Appendix I: Applications for Drugs Approved under FDA's Accelerated 
Approval Process Using Surrogate Endpoints: 

Appendix II: Applications for Drugs Approved under FDA's Traditional 
Process Using Surrogate Endpoints: 

Appendix III: Applications Selected and Questions Regarding FDA's 
Oversight of Required Postmarketing Studies: 

Appendix IV: Sales for Selected Drugs Approved Based on Surrogate 
Endpoints under the Accelerated Approval Process: 

Appendix V: Comments from the Department of Health and Human Services: 

Appendix VI: GAO Contact and Staff Acknowledgments: 

Tables: 

Table 1: Summary of Surrogate Endpoints for Accelerated Application 
Approvals, from June 19, 1992-November 20, 2008: 

Table 2: Status of Postmarketing Studies Requested under the 
Accelerated Approval Process, June 19, 1992-November 20, 2008: 

Table 3: Summary of Surrogate Endpoints for NME Drug Application 
Approvals, from January 1, 1998-June 30, 2008: 

Table 4: Status of Postmarketing Studies Requested under the 
Traditional Approval Process, January 1, 1998-June 30, 2008: 

Table 5: NDAs and BLAs Approved Based on Surrogate Endpoints under the 
Accelerated Approval Process, from June 19, 1992-November 20, 2008: 

Table 6: NDAs Approved Based on Surrogate Endpoints under the 
Traditional Approval Process, from January 1, 1998-June 30, 2008: 

Table 7: List of 15 Accelerated Approval Applications Selected for 
Review: 

Table 8: Total U.S. Sales for Selected Drugs Approved under the 
Accelerated Process: 

Figures: 

Figure 1: Applications Approved Using Surrogate Endpoints under FDA's 
Accelerated Approval Process, June 19, 1992-November 20, 2008: 

Figure 2: Percentage of Approved Applications Granted Accelerated 
Approval for Cancer and HIV/AIDS Drugs, June 19, 1992-November 20, 
2008: 

Figure 3: Status of Postmarketing Studies FDA Required under Its 
Accelerated Approval Process, June 19, 1992-November 20, 2008: 

Figure 4: Elapsed Time from Drug Approval to Fulfillment for 
Postmarketing Studies Required under the Accelerated Approval Process, 
June 19, 1992-November 20, 2008: 

Figure 5: Percentage of Postmarketing Studies Requested under the 
Accelerated Approval Process by Disease, June 19, 1992-November 20, 
2008: 

Figure 6: Percentage of Applications for NME Drugs Approved Using 
Surrogate Endpoints under FDA's Traditional Approval Process, January 
1, 1998-June 30, 2008: 

Figure 7: Percentage of Postmarketing Studies Requested under the 
Traditional Process, by Disease, January 1, 1998-June 30, 2008: 

Abbreviations: 

ASR: annual status report: 

BLA: Biologic License Application: 

DARRTS: Document Archiving, Reporting and Regulatory Tracking System: 

FDA: Food and Drug Administration: 

FDAAA: Food and Drug Administration Amendments Act: 

FDAMA: Food and Drug Administration Modernization Act of 1997: 

HHS: Department of Health and Human Services: 

NDA: new drug application: 

NME: new molecular entity: 

OIG: Office of Inspector General: 

[End of section] 

United States Government Accountability Office:
Washington, DC 20548: 

September 23, 2009: 

The Honorable Charles E. Grassley: 
Ranking Member: 
Committee on Finance: 
United States Senate: 

Dear Senator Grassley: 

The Food and Drug Administration (FDA), an agency within the Department 
of Health and Human Services (HHS), is the federal agency responsible 
for ensuring the safety and effectiveness of medical products, 
including drugs and biological products.[Footnote 1] Before a new drug 
can be marketed in the United States, a drug sponsor must demonstrate 
that it is safe and effective for its intended use, and obtain approval 
from FDA.[Footnote 2] Sponsors can demonstrate safety and effectiveness 
by conducting studies, known as clinical trials, on human volunteers 
and then submitting the results, as part of an application, to FDA for 
review.[Footnote 3] As part of its approval process, FDA reviews the 
data in the application, including the results of the clinical trials. 
If FDA determines that the drug's benefits outweigh its risks, it may 
approve the sponsor's application to market a new drug. 

FDA generally prefers that when conducting clinical trials, sponsors 
demonstrate the effectiveness of a new drug by showing its impact on a 
clinical endpoint--a direct measure of how a patient feels, functions, 
or survives. Demonstrating the effectiveness of a new drug, however, 
can require a sponsor to study the drug on thousands of patients over 
several years, potentially costing hundreds of millions of dollars. As 
an alternative to demonstrating a drug's effectiveness by its impact on 
a clinical endpoint, sponsors may submit, and FDA may approve 
applications based on clinical trials that demonstrate a new drug's 
impact on a surrogate endpoint--a laboratory measure or physical sign 
used as a substitute for a clinical endpoint--that reasonably predicts 
a clinical benefit. For example, demonstrating that a drug can lower 
blood pressure may be used as a surrogate endpoint to predict whether 
the drug is effective in preventing strokes. Through the use of 
surrogate endpoints, a drug sponsor can demonstrate the effect of a new 
drug on a surrogate endpoint based on smaller and shorter trials than 
would be required to prove the drug's effectiveness on a clinical 
endpoint. Unlike establishing clinical effectiveness, however, 
demonstrating the effect of a new drug on a surrogate endpoint does not 
always directly prove any benefit to a patient. Thus, reliance on a 
surrogate endpoint can create uncertainty because a drug's effect on a 
clinical outcome may not be known until after the drug is approved and 
further studied in patients. 

FDA allows the use of surrogate endpoints in both the accelerated and 
traditional approval processes. In 1992 FDA established an accelerated 
approval process to expedite the approval of applications for certain 
new drugs that are designed to treat serious or life-threatening 
illnesses and which are expected to provide a meaningful therapeutic 
benefit over existing therapy.[Footnote 4] Due to the need to expedite 
approval of such drugs, under this process, FDA may accept, as a basis 
for approval, evidence that demonstrates the drug's impact on surrogate 
endpoints which are reasonably likely to predict clinical benefit, and 
have not yet been demonstrated to be valid substitutes for clinical 
endpoints. Because FDA's approval of a drug based on these surrogate 
endpoints rarely establishes the drug's clinical benefits in relation 
to a clinical endpoint, FDA has developed additional regulatory 
requirements when sponsors use surrogate endpoints under the 
accelerated process. Specifically, when FDA approves a drug based on a 
surrogate endpoint under the accelerated approval process, FDA requires 
a sponsor, as a condition of approval, to conduct postmarketing 
confirmatory studies to validate that a drug's impact on a surrogate 
endpoint also leads to clinical benefits for patients. 

In contrast, under the traditional process--by which FDA reviews most 
drugs--FDA recognizes the surrogate endpoints as valid substitutes for 
clinical endpoints, and thus there are no such postmarketing study 
requirements. In addition, under both the accelerated and traditional 
approval processes, FDA may request--and sponsors may agree--to conduct 
additional postmarketing studies to address other matters that FDA has 
determined are worthy of further examination.[Footnote 5] 

Regardless of whether a postmarketing study has been required or 
requested by FDA, sponsors conducting such studies must comply with 
provisions in the Food and Drug Administration Modernization Act of 
1997 (FDAMA) and implementing regulations to report annually to FDA on 
the status of postmarketing studies.[Footnote 6] According to FDA's 
regulations, sponsors must continue to submit these reports each year 
until FDA notifies the sponsor, in writing, that it has determined that 
the study has been fulfilled or that the study is either no longer 
feasible or would no longer provide useful information.[Footnote 7] 

While FDA has long accepted surrogate endpoints to support drug 
approval, the use of such endpoints can be controversial. Although the 
use of surrogate endpoints can expedite drug approvals, it can also add 
uncertainty when the relationship between a surrogate endpoint and 
clinical benefit or endpoint has not been fully established. Recently, 
concerns have surfaced about some drugs which FDA approved based on 
surrogate endpoints. For example, in 2008 FDA approved the drug Avastin 
to treat breast cancer based on its ability to limit tumor growth; 
however, studies used to support approval also showed that the drug did 
not improve overall survival. You asked us to examine FDA's oversight 
of drugs approved based on surrogate endpoints. In this report we: 

1. identify applications FDA has approved based on surrogate endpoints 
through its accelerated approval process, including the surrogate 
endpoints used for approval, as well as the status of any associated 
postmarketing studies; 

2. identify applications FDA has approved for selected new drugs based 
on surrogate endpoints in its traditional approval process, including 
the surrogate endpoints used for approval, as well as the status of any 
associated postmarketing studies; and: 

3. evaluate FDA's oversight of postmarketing studies. 

To identify the applications for new drugs that FDA has approved on the 
basis of surrogate endpoints under the accelerated approval process, we 
obtained information from FDA, which included a list of all such 
applications the agency approved from June 1992[Footnote 8] through 
November 20, 2008 (appendix I contains a list of individual drugs, 
application numbers, and specific surrogate endpoints used for approval 
through the accelerated approval process). We analyzed this information 
to determine the number of approvals per year, the endpoints used for 
approval (e.g., measures of viral load), and the diseases that drugs 
were approved to treat based on surrogate endpoints. 

To identify the number and status of postmarketing studies FDA has 
required or requested for those drugs approved based on surrogate 
endpoints under the accelerated approval process, we obtained a list of 
such studies from FDA. Because FDA may approve multiple applications 
for the same drug, the same study can be associated with multiple 
applications;[Footnote 9] therefore, we utilized a unique numerical 
identifier assigned by FDA to each study to ensure that we did not 
count the same study more than once. For those postmarketing studies 
required under the accelerated approval process, FDA also provided 
certain key dates related to the progress of these studies, including 
the dates drug sponsors submitted their final study reports to FDA, and 
the dates FDA notified sponsors it had approved these reports. We then 
used this information, which was current as of December 19, 2008, to 
determine, among other things, the number and percentage of 
postmarketing studies by status and the average time it took sponsors 
to fulfill their postmarketing study requirement. To provide the most 
current information on the status of these studies, FDA had to research 
electronic and paper files because the information was not readily 
available in the database FDA uses to track the status of postmarketing 
studies. To assess the reliability of the information FDA provided, we 
compared the status and key dates for a sample of studies to the source 
data contained in FDA's files. We determined the data were sufficiently 
reliable for the purpose of our review. 

Similar to the postmarketing studies FDA required under its accelerated 
approval process, FDA did not have readily available data on studies it 
requested, such as certain key dates related to progress of the studies 
including the dates FDA approved final study reports. Because FDA 
officials indicated that it would be too time consuming to generate 
these time line data, as they did for the required studies, they only 
provided us with status information from their database used to track 
postmarketing studies. They provided us with data as of January 6, 
2009. For these requested postmarketing studies, we also used a unique 
numerical identifier assigned by FDA for each study to ensure that we 
did not count the same study more than once. We then identified the 
number and percentage of studies by status and by the length of time 
studies have remained open. We could not determine how long it took 
sponsors to fulfill studies, because the dates FDA determined the 
studies were fulfilled were not readily available. We did not verify 
the accuracy of this information. One limitation of using these data 
was that status information may not be current if FDA had not updated 
it in a timely manner. However, this represents the best information 
available and is what FDA uses to track the progress of requested 
postmarketing studies. 

To identify applications for selected new drugs that FDA approved on 
the basis of surrogate endpoints under the traditional process, we 
limited our scope to a subset of drugs. Because FDA could not readily 
identify the extent to which it approves applications based on 
surrogate endpoints through its traditional process, we reviewed only 
those applications FDA approved for new molecular entities (NMEs)-- 
potentially innovative drugs containing active chemical substances that 
have never been approved for marketing in the United States in any 
form.[Footnote 10] Although applications for NME drugs represented only 
about 10 percent of all applications FDA approved during this period, 
we limited our review to NME drugs because they represent the newest 
and potentially most innovative drugs. As a result, we believe our 
analyses would capture many of the key concerns related to the use of 
surrogate endpoints for new drug approval. One limitation of our 
analysis is that the percentage of NME drugs approved based on 
surrogate endpoints cannot be projected to the other drugs FDA approved 
through the traditional process during this period. 

Specifically, we reviewed information on all 219 applications for NME 
drugs that FDA approved from January 1, 1998, through June 30, 2008, 
and determined the proportion of applications FDA approved that were 
based on surrogate and clinical endpoints. This time period provided 
the most recent 10-year approval window at the time of our review. Of 
the 219 NME drugs, we excluded 15 because they were not approved to 
treat a specific disease.[Footnote 11] For the remaining 204 NME drugs, 
we examined documents summarizing the results of clinical trials for 
each of them. These documents, accessed from FDA's Web site, included 
the drugs' original labeling and FDA's original medical reviews. 
[Footnote 12] We reviewed these documents and determined whether or not 
the primary endpoint--the principal measure used to determine whether 
the drug was effective--for each of the 204 NME drugs was a surrogate 
or clinical endpoint. We then submitted our analyses to FDA to confirm 
that we correctly identified the endpoints as surrogate or clinical for 
each of the 204 applications we reviewed (appendix II contains a list 
of individual drugs, application numbers, and specific surrogate 
endpoints used for approval through the traditional approval process). 
After identifying the number of applications for NME drugs in our 
sample approved on the basis of surrogate endpoints through the 
traditional approval processes, we determined the number of approvals 
per year, the specific endpoints used for approval, and the diseases 
for which drugs were approved. 

To identify the number and status of postmarketing studies FDA 
requested for the approved NME drugs, we obtained a list of such 
studies from FDA, including information on the status of each study at 
the time of our review. FDA provided us with data on the status of 
studies as of February 13, 2009. We then used the same approach for 
analyzing information on these requested postmarketing studies as we 
used for those studies requested by FDA under the accelerated approval 
process, subject to the same limitations. 

To evaluate FDA's oversight of postmarketing studies, we reviewed 
relevant laws and regulations, and policy documents describing FDA's 
program oversight and enforcement authority. We also examined internal 
control standards, which include the need to establish policies and 
procedures to help ensure effective and efficient operations.[Footnote 
13] We interviewed FDA officials to identify the oversight activities 
it engages in to monitor the postmarketing studies it has required or 
requested sponsors to conduct. We also reviewed the enforcement tools 
FDA uses or can use to ensure sponsors conduct these studies. To review 
specific instances of FDA's monitoring and enforcement activities, we 
selected a judgmental sample of 15 applications approved based on 
surrogate endpoints under the accelerated program and the 35 
postmarketing studies that FDA required drug sponsors to complete for 
these drugs. These applications were selected to generate a sample that 
included a variety of drugs and a range of studies at various stages of 
completion. We provided FDA with a standard series of questions for 
each of the 35 studies, and requested that FDA's medical reviewers, who 
are responsible for monitoring these 35 studies, provide specific 
information on them, including a description of the studies, FDA's 
efforts to monitor these studies, and applicable enforcement actions 
taken, if any, to prompt sponsors' compliance (appendix III identifies 
the 15 applications which we selected, and provides the standard series 
of questions we provided to FDA for each of the 35 postmarketing 
studies). FDA officials indicated that several individuals were 
involved in completing the questions provided for each of the 15 
applications. FDA staff completed the first half of the questions 
related to time lines, current statuses, and ASR submissions. The 
medical reviewers or other staff provided information related to any 
study completion problems, underlying issues, and monitoring 
activities. To obtain a better understanding of FDA's oversight of 
postmarket studies, we reviewed reports issued by HHS's Office of 
Inspector General (HHS-OIG) and Booz Allen Hamilton, a contractor 
retained by FDA in 2006 to conduct an independent analysis of the 
agency's postmarket oversight processes and procedures. Finally, for 
certain drugs approved under the accelerated approval process, we 
obtained annual U.S. sales data from the time the applications for 
these drugs were approved through December 2008 (appendix IV identifies 
the applications and drugs and the total U.S. sales since approval). As 
of December 19, 2008, these applications had not been converted to full 
approval, and more than 5 years had elapsed since they were initially 
approved. We obtained the annual sales data from IMS Health, a provider 
of market information to the pharmaceutical and health care industries. 

We conducted this performance audit from June 2008 through September 
2009 in accordance with generally accepted government auditing 
standards. Those standards require that we plan and perform the audit 
to obtain sufficient, appropriate evidence to provide a reasonable 
basis for our findings and conclusions based on our audit objectives. 
We believe that the evidence obtained provides a reasonable basis for 
our findings and conclusions based on our audit objectives. 

Background: 

FDA's responsibilities for overseeing the safety and effectiveness of 
drugs begin before a product is brought to market and continue after a 
drug's approval. Its premarket responsibilities include reviewing a 
drug sponsor's proposal for conducting clinical trials, as well as 
reviewing applications. If FDA determines that a drug is safe and 
effective--that its clinical benefits outweigh its potential health 
risks--and that other requirements are met, it will approve the 
application.[Footnote 14] Once it approves a new drug, FDA is charged 
with monitoring the safety and effectiveness of the drug, which 
includes overseeing a sponsor's progress in completing postmarketing 
studies. 

The Benefits and Risks of Surrogate Endpoints in Drug Approval: 

Reliance on surrogate endpoints to predict clinical benefit can bring 
treatment benefits to patients years before definitive information on a 
drug's effect on clinical outcome is available, and at relatively low 
cost. This is because a drug sponsor may be able to demonstrate the 
effect of a new drug on a surrogate endpoint based on smaller and 
shorter trials than would be required to prove the drug's effect on a 
clinical endpoint. In order to further enhance drug development, some 
have called for their expanded use as a basis for drug approval. For 
example, FDA's 2004 report on the slowdown of innovative medical 
therapies calls for the acceptance of new surrogate endpoints to guide 
drug development.[Footnote 15] Based on this report, FDA cited the need 
to clarify the conditions for the use of new surrogate endpoints in the 
drug approval process. To facilitate this effort, FDA planned to 
develop an inventory of surrogate endpoints which had been used as the 
basis for approval of drugs through the traditional and accelerated 
approval processes. 

Despite the potential benefits of using surrogate endpoints, reliance 
on these endpoints may introduce uncertainty regarding the risks and 
benefits of a drug, and may lead to the adoption of useless or even 
harmful therapies.[Footnote 16] This can arise if the effect on a 
surrogate endpoint does not accurately predict whether treatments 
provide benefits to patients, or if the drug has a smaller than 
expected benefit and a larger than expected adverse effect, which might 
not be recognized without large-scale, long-term clinical trials. For 
example, several large trials assessing drugs based on surrogate 
endpoints found those drugs to be clinically ineffective, and in some 
cases, identified unexpected adverse effects such as increased rates of 
death.[Footnote 17] 

FDA Role in the Drug Development and Application Review Process: 

Once a drug sponsor identifies a promising chemical compound or 
biologic organism capable of curing or treating diseases, the sponsor 
may decide to test it on humans. Before doing so, a sponsor must submit 
the data that have been collected on the compound or organism in prior 
studies and outline its plans for clinical trials. The clinical trial 
stage gradually introduces experimental drugs to increasingly larger 
numbers of patients to determine the drug's safety and efficacy. It is 
during this stage that a sponsor determines whether it will evaluate a 
drug's effectiveness using a clinical or surrogate endpoint. 

Once a drug sponsor completes clinical trials, it may submit an 
application to FDA for review. The application contains scientific and 
clinical data intended to demonstrate that the drug is safe and 
effective for its proposed use. Depending on factors such as the types 
of disease the drug is designed to treat, whether other effective 
treatments are available, and whether the sponsor assessed the drug 
based on a surrogate or clinical endpoint, FDA will review the 
application under either the accelerated or traditional approval 
process. If the application is for a drug designed to treat serious or 
life-threatening illnesses and the drug is expected to provide 
meaningful therapeutic benefits compared to existing treatments, and 
the sponsor evaluated the drug's impact on a surrogate endpoint that 
only reasonably suggested clinical benefit, FDA will review the 
application under its accelerated approval process. In general, if the 
sponsor assessed the drug's impact on a clinical endpoint, or a 
surrogate endpoint that FDA considers to be a valid substitute, FDA 
will review the application under its traditional approval process. 

According to FDA officials, they do not have specific criteria for 
determining when they will accept a surrogate endpoint as a valid 
substitute for a clinical endpoint, and such decisions are made on a 
case-by-case basis. This determination is dependent on factors such as 
the type of drug being approved and the disease being treated. However, 
FDA generally considers surrogate endpoints as valid substitutes for 
clinical endpoints when these endpoints have been shown to accurately 
predict a clinical benefit over time through definitive studies. For 
example, FDA considers lowering blood pressure as a valid surrogate 
endpoint to establish a drug's clinical effectiveness in reducing the 
risk of stroke. 

Regardless of which process FDA uses, the application will be reviewed 
by one of FDA's medical review divisions, depending on the disease 
being treated by the drug. The medical review division evaluates data 
contained in the application to determine whether the drug should be 
approved. If the medical review division determines the sponsor has 
demonstrated the drug is safe and effective for its intended use, and 
has met other applicable requirements, FDA will issue an approval 
letter. The approval letter outlines any postmarketing studies that FDA 
has required or requested the sponsor to conduct while the drug is 
being marketed for sale in the United States--including any associated 
postmarketing study time frames a sponsor may need to meet.[Footnote 
18] As a condition of approval under its accelerated approval process, 
FDA requires that sponsors conduct postmarketing studies known as 
confirmatory studies. FDA requires sponsors to conduct these 
postmarketing studies to verify and describe the drug's clinical 
benefits and thereby resolve any remaining uncertainty regarding the 
drug's clinical benefit.[Footnote 19] In addition, FDA may request 
sponsors to conduct postmarketing studies when it determines that 
additional information, while not essential for approval, is important 
in improving the prescribing, use, and quality of a drug or consistency 
in drug manufacturing. For example, FDA may request sponsors, and 
sponsors may agree, to continue to evaluate a drug's safety, 
effectiveness, pharmacology, toxicology, or manufacturing controls. FDA 
may request these studies under both the accelerated and traditional 
review processes. 

FDA's Oversight of Postmarketing Studies: 

FDA's oversight of postmarketing studies consists of a variety of 
monitoring and enforcement activities. After it outlines required and 
requested studies in the approval letter, FDA is responsible for 
monitoring sponsors' progress in completing the studies, and taking 
actions to help ensure studies are completed. A key component of FDA's 
oversight is the requirement that sponsors of all approved drugs report 
annually on their progress or status towards completing postmarketing 
studies. According to the implementing regulations, drug sponsors must 
report on the status of required and requested postmarketing studies in 
annual status reports (ASR), which are due within 60 days of the drug 
application's approval anniversary date.[Footnote 20] Federal 
regulations also require ASRs to contain, in part, information on the 
description of the required and requested postmarketing studies, along 
with the schedule for their completion.[Footnote 21] 

FDA has established a goal of reviewing ASRs within 90 days of receipt 
to confirm the accuracy of the information provided. FDA may use 
information obtained from the ASRs to update a database that includes 
information on postmarketing studies. The database, which was designed 
to allow FDA to track and monitor the status of required and requested 
postmarketing studies, and identify those studies which are falling 
behind established time frames, includes information on dates the 
studies were required or requested in the approval letter, a 
description of the study, and its status. FDA also makes certain 
information from this database regarding the status of postmarketing 
studies available to the public, and is required to report certain 
information annually in the Federal Register.[Footnote 22] 

Based on its review of an ASR, FDA designates a status of either "open" 
or "closed" for each required and requested postmarketing study every 
year. FDA further classifies open studies into one of the following 
five categories:[Footnote 23] 

* Pending - Those required and requested studies that have not been 
initiated, but are not yet delayed. Generally, the first patient has 
not been enrolled in the study. 

* Ongoing - Those required and requested studies that are proceeding 
according to or ahead of any original schedule. 

* Delayed - Those required and requested studies that are proceeding, 
but are behind their original schedule. 

* Terminated - Those required and requested studies that were ended 
before their actual completion, but the sponsor has not yet submitted a 
final report to FDA. A postmarketing study may be terminated because 
the study would no longer provide useful information or the study is no 
longer feasible. In some instances, there may be another postmarketing 
study that the drug sponsor must conduct in lieu of the terminated 
study. 

* Submitted - Those required and requested studies that have been 
completed--that is, the last patient has finished the protocol--or 
terminated and a final study report has been submitted to FDA, but FDA 
has not yet notified the drug sponsor that the study has been fulfilled 
or released. Drug sponsors are required to continue to annually submit 
an ASR on the status of its study until FDA provides written 
confirmation that the obligations have been met. 

FDA further classifies "closed" studies into one of the following two 
categories: 

* Fulfilled - Those required and requested studies that have been 
completed and a final study report has been submitted and reviewed by 
FDA. When FDA completes its review of a study and finds that the 
postmarketing study has satisfied FDA's request or requirement, the 
agency will issue a written confirmation to the drug sponsor that it 
considers the study requested or required in the approval letter to 
have been fulfilled. 

* Released - Those required and requested studies that have not been 
completed and have been found to be no longer needed or feasible. For 
example, FDA may release a sponsor from the need to compete a study 
because a new drug has been developed which renders the drug being 
studied obsolete, thus eliminating the need to conduct the original 
study. 

Once a drug sponsor has completed a postmarketing study, it submits a 
final study report to FDA for review. FDA's goal is to review the final 
study report within 12 months of receipt and notify the sponsor whether 
FDA considers the study closed. If the postmarketing study is a 
confirmatory study required under the accelerated approval process, the 
final study report should provide information confirming the drug's 
clinical benefit. If it provides confirming information, and FDA agrees 
that the report satisfies the confirmatory study requirements, then FDA 
converts the drug from accelerated to full approval.[Footnote 24] 

FDA's oversight of postmarketing studies may be aided by enforcement 
tools, including administrative action letters and expedited withdrawal 
procedures of a drug from the market, in certain cases.[Footnote 25] If 
a sponsor fails to submit or is late in submitting an ASR, or if FDA 
determines a sponsor is not making adequate progress in completing a 
required or requested study, FDA may issue an administrative action 
letter--commonly referred to by FDA as "Dunner" letters. FDA uses these 
letters to remind sponsors about their need to meet federal 
requirements related to either submission of ASRs or completing a 
required or requested study. Also, under the accelerated approval 
regulations, FDA may initiate procedures to withdraw a drug from the 
market through an expedited process if required postmarketing studies 
to confirm and verify a drug's clinical benefit are not performed with 
due diligence.[Footnote 26] In contrast, FDA does not require that drug 
sponsors complete requested postmarketing studies, regardless of 
whether a drug has been approved under the accelerated or traditional 
process. Such studies are typically related to safety, effectiveness, 
pharmacology, toxicology, or manufacturing controls. 

FDA Has Approved Many Applications Based on Surrogate Endpoints through 
Its Accelerated Approval Process and about Two-Thirds of Postmarketing 
Studies Have Been Closed: 

FDA has approved 90 applications based on surrogate endpoints under its 
accelerated approval process since the process was established in 1992. 
During this period, FDA required or requested sponsors to conduct over 
450 postmarketing studies associated with the approval of applications 
for these drugs, and the majority of these studies have been classified 
by FDA as closed--meaning that drug sponsors had met FDA's requirements 
for these studies or FDA determined the studies were no longer needed 
or feasible. However, several have been classified by FDA as open for 
an extended period of time. 

FDA Approved 90 Applications Based on Surrogate Endpoints under the 
Accelerated Approval Process, the Majority for Drugs to Treat Cancer 
and HIV/AIDS: 

From June 19, 1992, through November 20, 2008, FDA approved a total of 
90 applications based on surrogate endpoints, or an average of 5 per 
year. FDA approved these applications for a total of 64 different 
drugs.[Footnote 27] Over this period, there was variability in the 
number of applications FDA approved annually based on surrogate 
endpoints. For example, in 1992 and 1994, FDA approved 1 application 
based on surrogate endpoints under the accelerated approval process, 
while in 2004 it approved 10 applications (see figure 1). 

Figure 1: Applications Approved Using Surrogate Endpoints under FDA's 
Accelerated Approval Process, June 19, 1992-November 20, 2008: 

[Refer to PDF for image: vertical bar graph] 

FDA approved an average of 5 applications per year. 

Year: 1992; 
Applications approved: 1. 

Year: 1993; 
Applications approved: 3. 

Year: 1994; 
Applications approved: 1. 

Year: 1995; 
Applications approved: 6. 

Year: 1996; 
Applications approved: 9. 

Year: 1997; 
Applications approved: 3. 

Year: 1998; 
Applications approved: 9. 

Year: 1999; 
Applications approved: 8. 

Year: 2000; 
Applications approved: 9. 

Year: 2001; 
Applications approved: 3. 

Year: 2002; 
Applications approved: 6. 

Year: 2003; 
Applications approved: 7. 

Year: 2004; 
Applications approved: 12. 

Year: 2005; 
Applications approved: 4. 

Year: 2006; 
Applications approved: 6. 

Year: 2007; 
Applications approved: 3. 

Year: 2008; 
Applications approved: 6. 

Source: GAO analysis of FDA data. 

[End of figure] 

Almost all--79 of the 90 applications--were for drugs to treat three 
diseases. Specifically, 38 of the applications were for drugs to treat 
cancer, 30 were for drugs to treat HIV/AIDS, and 11 were for drugs to 
treat inhalation anthrax. The remaining 11 applications were for drugs 
to treat a variety of other diseases (appendix I contains a list of 
individual drugs, application numbers, and specific surrogate endpoints 
used for approval through the accelerated approval process). 

Since FDA began using the accelerated process in 1992, there has been a 
general shift in approvals based on surrogate endpoints from 
applications for HIV/AIDS drugs to applications for cancer drugs. In 
the first 9 years of the accelerated approval process, from 1992 
through 2000, applications for drugs to treat HIV/AIDS made up 48 
percent of the approvals, while applications for drugs to treat cancer 
made up 26 percent of these applications. Conversely, from 2001 through 
2008, applications for drugs to treat cancer made up over half--59 
percent--of the applications approved, while drugs to treat HIV/AIDS 
accounted for only 18 percent of approved applications (see figure 2). 

Figure 2: Percentage of Approved Applications Granted Accelerated 
Approval for Cancer and HIV/AIDS Drugs, June 19, 1992-November 20, 
2008: 

[Refer to PDF for image: vertical bar graph] 

Year: 1992-1996; 
Cancer applications approved: 28%; 
HIV/AIDS applications approved: 50%. 

Year: 1997-2000; 
Cancer applications approved: 25%; 
HIV/AIDS applications approved: 46%. 

Year: 2001-2004; 
Cancer applications approved: 68%; 
HIV/AIDS applications approved: 12%. 

Year: 2005-2008; 
Cancer applications approved: 47%; 
HIV/AIDS applications approved: 26%. 

Source: GAO analysis of FDA data. 

[End of figure] 

Consistent with the types of applications approved under the 
accelerated process, the specific surrogate endpoints most frequently 
used to obtain approval were those used to demonstrate the 
effectiveness of cancer and HIV/AIDS drugs. Specifically, FDA approved, 

* 38 applications for cancer drugs based on how the drugs impacted 
tumors, as measured by various tumor assessment surrogate endpoints 
such as response rate (e.g., tumor shrinkage), length of time until the 
cancer spread, or length of time until the drug no longer worked; 

* 30 applications for HIV/AIDS drugs based on the drugs' ability to 
lower the viral load of HIV in the blood stream, as measured by the 
drugs' impact on the surrogate endpoint HIV-RNA; and: 

* the remaining 22 applications for drugs to treat other diseases-- 
including inhalation anthrax and various bacterial infections--based on 
a variety of surrogate endpoints (see table 1). 

Table 1: Summary of Surrogate Endpoints for Accelerated Application 
Approvals, from June 19, 1992-November 20, 2008: 

Disease: Cancer; 
Surrogate endpoint(s) used for approval: 
* Tumor assessment; 
* Response rate; 
* Disease-free survival; 
* Progression-free survival; 
* Time to treatment failure; 
* Cytogenic and/or hemotologic response as measured in marrow or blood; 
* Other; 
Number of applications: 38. 

Disease: HIV/AIDS; 
Surrogate endpoint(s) used for approval: 
* Viral load (HIV-RNA); 
* CD4 count; 
* p24 antigen existence; 
* Other; 
Number of applications: 30. 

Disease: Inhalation anthrax; 
Surrogate endpoint(s) used for approval: 
* Drug exposure in monkeys compared to human plasma concentrations; 
* Serum CIPRO concentrations; 
* Other; 
Number of applications: 11. 

Disease: Other diseases; 
Surrogate endpoint(s) used for approval: 
* Various; 
Number of applications: 11. 

Total: 
Number of applications: 90. 

Source: GAO analysis of FDA data. 

[End of table] 

FDA Required 144 Confirmatory Postmarketing Studies under the 
Accelerated Process, and About Two-Thirds Have Been Closed: 

From June 1992 through November 20, 2008, FDA required drug sponsors to 
conduct 144 postmarketing confirmatory studies associated with drugs 
approved based on surrogate endpoints under the accelerated approval 
process. Consistent with the types of drugs FDA approved under the 
process, FDA required the majority of these studies--119 or 79 percent--
for drugs approved to treat cancer or HIV/AIDS. Specifically, FDA 
required 82 studies for drugs to treat cancer, and another 37 studies 
for drugs to treat HIV/AIDS. The remaining studies were for drugs to 
treat a variety of other diseases. 

At the time of our review, we found FDA had classified 92, or 64 
percent, of the 144 required studies as closed--meaning that drug 
sponsors had met FDA's requirements for these studies or FDA determined 
the studies were no longer needed or feasible. In contrast, FDA had 
classified 52 of the 144 studies, or 36 percent, as open, and that 
sponsors had made varying levels of progress in completing them (see 
figure 3)[Footnote 28]. 

Figure 3: Status of Postmarketing Studies FDA Required under Its 
Accelerated Approval Process, June 19, 1992-November 20, 2008: 

[Refer to PDF for image: illustration with 3 pie-charts] 

All required postmarketing studies (total=144): 
Closed: (92): 64%; 
Open (52): 36%. 

Closed postmarketing studies: 
Fulfilled (73): 79%; 
Released (19): 21%. 

Open postmarketing studies: 
Ongoing (17): 33%; 
Submitted (16): 31%; 
Delayed (10): 19%; 
Pending (7): 13%; 
Terminated (2): 4%. 

Source: GAO analysis of FDA data. 

Note: Status as of December 19, 2008. 

[End of figure] 

Of the 92 closed studies, sponsors had fulfilled requirements for 73 of 
them. This means that sponsors had completed these studies and 
submitted them to FDA, and upon review, FDA determined sponsors had 
fulfilled their study requirements. There were an additional 19 studies 
which FDA classified as released, meaning it had released the sponsors 
from the study requirements because it determined the studies were 
either no longer feasible or would no longer provide useful 
information. 

Based on our analyses of the 71 fulfilled studies, we found that in 
general, sponsors were able to fulfill about two-thirds of their study 
requirements in less than 5 years, with time frames ranging from 7 
months to more than 12 years. In contrast, nearly one-third, or 23, of 
these studies took over 5 years to fulfill (see figure 4).[Footnote 29] 

Figure 4: Elapsed Time from Drug Approval to Fulfillment for 
Postmarketing Studies Required under the Accelerated Approval Process, 
June 19, 1992-November 20, 2008: 

[Refer to PDF for image: vertical bag graph] 

Elapsed time: Less than 1 year; 
Number of studies: 2. 

Elapsed time: Between 1 and 2 years; 
Number of studies: 12. 

Elapsed time: Between 2 and 3 years; 
Number of studies: 15. 

Elapsed time: Between 3 and 4 years; 
Number of studies: 11. 

Elapsed time: Between 4 and 5 years; 
Number of studies: 8. 

Elapsed time: Over 5 years; 
Number of studies: 23. 

Source: GAO analysis of FDA data. 

Note: Status as of December 19, 2008. 

[End of figure] 

The amount of time needed to fulfill study requirements was generally 
longer for those studies involving drugs to treat cancer. Specifically, 
61 percent of the postmarketing studies that took over 5 years (14 of 
23) were studies for drugs to treat cancer. Conversely, only 9 percent 
of studies which took over 5 years (2 of 23), were studies for drugs to 
treat HIV/AIDS. According to FDA officials, the greater length of time 
needed to fulfill required confirmatory studies for drugs to treat 
cancer has resulted from the approach generally used to approve many 
cancer drugs under the accelerated process. In particular, most of the 
cancer drugs gained accelerated approval based on single arm clinical 
trials,[Footnote 30] which studied these drugs' impacts in small 
numbers of patients with resistant tumors. According to FDA officials, 
this approach has been used to approve drugs targeting resistant tumors 
because these patients have no other effective therapy and it would be 
too difficult and time consuming for sponsors to conduct more in-depth 
randomized control studies.[Footnote 31] Because patients need access 
to these drugs as soon as possible, FDA deemed this approach sufficient 
for accelerated approval. However, in order to establish clinical 
benefits of the drugs through the required confirmatory postmarketing 
studies, the sponsors needed to conduct randomized control trials 
comparing the drugs to either a placebo or another drug(s). To do this, 
sponsors must design and conduct new randomly controlled clinical 
trials and recruit new patients, many of whom may be reluctant to sign 
up for a clinical trial when an approved drug is already on the market. 
In contrast, according to FDA officials, it is typically easier for a 
sponsor to complete a required confirmatory study for HIV/AIDS drugs 
because sponsors may simply continue the study which led to the 
original accelerated approval of the drug and do not have to design new 
studies to fulfill the postmarketing study requirements. For example, 
sponsors typically obtain accelerated approval for HIV/AIDS drugs based 
on a 24-week randomized clinical trial, and to meet the confirmatory 
study requirement they continue the same study for an additional 24 
weeks, using the same patients and the same endpoints. 

The 52 studies FDA classified as open covered a variety of statuses, 
and thus were at varying levels of completion. Specifically, 7 were 
pending and had not yet begun, 10 were delayed and thus behind the 
sponsor's original schedule, and 17 were ongoing, and thus on or ahead 
of schedule. Additionally, 2 were terminated, meaning the sponsor had 
stopped the study, but had not submitted results to FDA. The remaining 
16 had been submitted by the sponsor to FDA, and were awaiting FDA's 
review. Based on our analyses of these 52 open studies, we found the 
majority--approximately 65 percent--were for drugs approved to treat 
cancer. In contrast, only 6 percent of the open studies were for drugs 
approved to treat HIV/AIDS. Based on information provided to us by FDA, 
we determined that the average age of the 52 open studies was just over 
4 years and the majority--37 of the 52 studies, or 71 percent--had been 
required since 2004. However, we found that 15 studies had been open 
for more than 5 years, including several open for more than 8 years. 
For example, on May 17, 2000, FDA approved an application for the drug 
Mylotarg to treat certain patients with acute myeloid leukemia based on 
its ability to control cancer in blood cells. As a condition of 
accelerated approval FDA required the drug sponsor to conduct one 
confirmatory study, and as of December 19, 2008, more than 8 years 
later, the study was ongoing, with an anticipated completion date of 
October 2014. 

FDA Requested over 300 Postmarketing Studies under the Accelerated 
Process, and About Two-Thirds Have been Closed: 

In addition to the 144 confirmatory studies FDA required from June 1992 
through November 20, 2008, FDA also requested--and sponsors agreed to 
conduct--317 other postmarketing studies associated with drugs approved 
through the accelerated process based on surrogate endpoints. FDA 
requested the majority of these studies--90 percent--for drugs approved 
to treat HIV/AIDS and cancer. Specifically, FDA requested 194 studies 
for drugs to treat HIV/AIDS, and another 91 studies for drugs to treat 
cancer. FDA requested the remaining studies for drugs to treat a 
variety of other diseases (see figure 5). 

Figure 5: Percentage of Postmarketing Studies Requested under the 
Accelerated Approval Process by Disease, June 19, 1992-November 20, 
2008: 

[Refer to PDF for image: pie-chart] 

HIV/AIDS (194): 61%; 
Cancer (91): 29%; 
Other (32): 10%. 

Source: GAO analysis of FDA data. 

[End of figure] 

Based on the status FDA assigned each study at the time of our review, 
FDA had classified 203 of the 317 requested studies, or 64 percent, as 
closed, meaning that drug sponsors had satisfied FDA's request for 
these studies or FDA determined the studies were no longer needed or 
feasible.[Footnote 32] Additionally, we found that FDA had classified 
114, or 36 percent, as open, and sponsors had made varying levels of 
progress in completing them (see table 2). These percentages are 
similar to those for studies FDA required sponsors to complete. 

Table 2: Status of Postmarketing Studies Requested under the 
Accelerated Approval Process, June 19, 1992-November 20, 2008: 

Status of open studies: Pending; 
Number of studies (% of Total): 57 (18%). 

Status of open studies: Ongoing; 
Number of studies (% of Total): 21 (7%). 

Status of open studies: Delayed; 
Number of studies (% of Total): 6 (2%). 

Status of open studies: Terminated; 
Number of studies (% of Total): 0 (0%). 

Status of open studies: Submitted; 
Number of studies (% of Total): 30 (9%). 

Total open: 
Number of studies (% of Total): 114 (36%). 

Status of closed studies: Fulfilled; 
Number of studies (% of Total): 166 (52%). 

Status of closed studies: 
Released; Number of studies (% of Total): 37 (12%). 

Total closed: 
Number of studies (% of Total): 203 (64%). 

Total: 
Number of studies (% of Total): 317 (100%). 

Source: GAO analysis of FDA data. 

Note: Status as of January 6, 2009. 

[End of table] 

Based on information FDA provided, we found that the length of time 
studies have been open varies by status. Specifically, those studies 
classified as pending had been open, on average, for about 5.5 years, 
with more than 40 percent pending for over 8 years. In addition, 
studies classified as ongoing and delayed had been open, on average, 
for 5.3 and 4 years respectively, and those classified as submitted 
have been open on average about 5.6 years.[Footnote 33] We could not 
calculate the amount of time it took sponsors to close a requested 
study, because, according to FDA, the dates studies were fulfilled or 
released were not readily available. 

FDA Approved about One-Third of NME Drug Applications Based on 
Surrogate Endpoints through Its Traditional Process and about Half of 
the Postmarketing Studies Requested Have Been Closed: 

From January 1, 1998, through June 30, 2008, FDA approved about one- 
third of NME drug applications based on surrogate endpoints under its 
traditional process. Many of these applications were for drugs to treat 
cancer, heart disease, and diabetes. FDA requested 175 postmarketing 
studies associated with these NMEs, and about one-half have been 
classified by FDA as closed. 

About One-Third of the Applications for NME Drugs Approved under the 
Traditional Process Were Based on Surrogate Endpoints, Many for Drugs 
to Treat Cancer, Cardiovascular Disease, and Diabetes: 

From January 1, 1998, through June 30, 2008, FDA approved 204 
applications for NMEs to treat diseases through the traditional 
approval process.[Footnote 34] Of these 204 applications, FDA approved 
69, or about 34 percent, on the basis of surrogate endpoints.[Footnote 
35] The percentage of NME approvals based on surrogate endpoints per 
year varied, ranging from 17 percent in 1999 to 54 percent in 2005. 
Additionally, in most years from January 1998 through June 2008, NME 
applications that were approved based on surrogate endpoints comprised 
less than half of all NME approvals in any given year (see figure 6). 

Figure 6: Percentage of Applications for NME Drugs Approved Using 
Surrogate Endpoints under FDA's Traditional Approval Process, January 
1, 1998-June 30, 2008: 

[Refer to PDF for image: vertical bar graph] 

Year: 1998; 
Percentage of Applications Approved: 35. 

Year: 1999; 
Percentage of Applications Approved: 17. 

Year: 2000; 
Percentage of Applications Approved: 36. 

Year: 2001; 
Percentage of Applications Approved: 24. 

Year: 2002; 
Percentage of Applications Approved: 43. 

Year: 2003; 
Percentage of Applications Approved: 39. 

Year: 2004; 
Percentage of Applications Approved: 32. 

Year: 2005; 
Percentage of Applications Approved: 54. 

Year: 2006; 
Percentage of Applications Approved: 25. 

Year: 2007; 
Percentage of Applications Approved: 50. 

Year: 2008; 
Percentage of Applications Approved: 40. 

Source: GAO analysis of FDA data. 

Note: 2008 data include NME drug approvals only through June 30, 2008. 

[End of figure] 

The most frequently used surrogate endpoints in the traditional process 
were those to establish the effectiveness of drugs to treat cancer, 
cardiovascular disease, and diabetes. Specifically, 13 of the 69 
applications--or 19 percent of applications approved--were for drugs to 
treat cancer, using various tumor assessments as surrogate endpoints, 
including response rates, similar to those used in the accelerated 
process. In addition, 11 of the 69 applications, or 16 percent, were 
approved for drugs to treat cardiovascular disease, based on their 
ability to decrease blood pressure or control cholesterol levels. 
Furthermore, 10 of the 69, or 14 percent were for diabetes drugs, based 
on their ability to lower blood sugar levels (see table 3). Drugs to 
treat a variety of other diseases--including renal disease and 
hepatitis B--accounted for the remaining 35 applications, and their 
approval was based on a variety of surrogate endpoints (appendix II 
contains a list of individual drugs, application numbers, and specific 
surrogate endpoints used for approval of NME drugs through the 
traditional approval process). 

Table 3: Summary of Surrogate Endpoints for NME Drug Application 
Approvals, from January 1, 1998-June 30, 2008: 

Disease: Cancer; 
Surrogate endpoint(s) used for approval: 
* Tumor assessment; 
* Time to progression; 
* Response rate; 
* Progression-free survival; 
* Other (e.g., serum testosterone levels); 
Number of applications: 13. 

Disease: Cardiovascular conditions; 
Surrogate endpoint(s) used for approval: 
* Blood pressure; 
* Lipid levels (cholesterol and triglycerides); 
Number of applications: 11. 

Disease: Diabetes mellitus; 
Surrogate endpoint(s) used for approval: 
* Blood sugar; 
* Fasting plasma glucose levels; 
* HbA1c levels; 
Number of applications: 10. 

Disease: Other diseases; 
Surrogate endpoint(s) used for approval: 
* Various; 
Number of applications: 35. 

Total: 
Number of applications: 69. 

Source: GAO analysis of FDA data. 

[End of table] 

Our analysis of NME drug applications represents only a small subset of 
all applications which FDA may have approved based on surrogate 
endpoints under the traditional approval process during this time 
period. From January 1, 1998, through June 30, 2008, FDA approved about 
2,000 other applications; however, it does not track whether they 
approved these applications based on surrogate endpoints. Thus the 
extent to which FDA used these and other surrogate endpoints as a basis 
for approval is unclear.[Footnote 36] 

FDA Requested 175 Postmarketing Studies under Its Traditional Process, 
and About One-Half Have Been Closed: 

From January 1, 1998, through June 30, 2008, FDA requested 175 
postmarketing studies for NMEs approved based on surrogate endpoints 
under the traditional process. During this time frame, FDA requested 
studies for drugs to treat a variety of diseases. Of the 175 studies, 
FDA requested 43, or 25 percent, for drugs to treat cancer. In 
addition, FDA requested 31 studies, or 18 percent, for drugs to treat 
hepatitis B. The remaining studies were for drugs to treat a variety of 
other diseases, including renal disease, diabetes, HIV/AIDS, and 
cardiovascular disease (see figure 7). 

Figure 7: Percentage of Postmarketing Studies Requested under the 
Traditional Process, by Disease, January 1, 1998-June 30, 2008: 

[Refer to PDF for image: pie-chart] 

Cancer (43): 25%; 
Hepatitis B (31): 18%; 
Renal disease (25): 14%; 
HIV/AIDS (22): 13%; 
Diabetes (19): 11%; 
Cardiovascular disease (10): 6%; 
Other (25): 14%. 

Source: GAO analysis of FDA data. 

[End of figure] 

Based on the status FDA assigned each study at the time of our review, 
FDA had classified 94 of the 175 requested studies, or 54 percent, as 
closed, meaning that drug sponsors had satisfied FDA's request for 
these studies or FDA determined the studies were no longer needed or 
feasible. Additionally, we found that FDA had classified 81, or 46 
percent, as open, and sponsors had made varying levels of progress in 
completing them (see table 4). 

Table 4: Status of Postmarketing Studies Requested under the 
Traditional Approval Process, January 1, 1998-June 30, 2008: 

Status of open studies: Pending; 
Number of studies (% of Total): 33 (19%). 

Status of open studies: Ongoing; 
Number of studies (% of Total): 21 (12%). 

Status of open studies: Delayed; 
Number of studies (% of Total): 7 (4%). 

Status of open studies: Terminated; 
Number of studies (% of Total): 4 (2%). 

Status of open studies: Submitted; 
Number of studies (% of Total): 16 (9%). 

Status of open studies: Total open; 
Number of studies (% of Total): 81 (46%). 

Status of closed studies: Fulfilled; 
Number of studies (% of Total): 87 (50%). 

Status of closed studies: 
Released; Number of studies (% of Total): 7 (4%). 

Status of open studies: Total closed; 
Number of studies (% of Total): 94 (54%). 

Status of open studies: Total; 
Number of studies (% of Total): 175 (100%). 

Source: GAO analysis of FDA data. 

Note: Status as of February 13, 2009. 

[End of table] 

Based on information FDA provided, we found that the length of time 
studies have been open varies by status. Specifically, those studies 
classified as pending had been open, on average, for 3.8 years, ranging 
from less than 1 year to 10.9 years. In addition, studies classified as 
ongoing and delayed had been open, on average, for 2.9 and 5.8 years 
respectively, and those classified as submitted have been open on 
average 4.2 years.[Footnote 37] We could not calculate the amount of 
time it took sponsors to close a requested study, because, according to 
FDA, the dates studies were fulfilled or released were not readily 
available. 

FDA's Oversight of Postmarketing Studies Is Hindered by Weaknesses in 
Its Monitoring and Enforcement: 

FDA has not been routinely monitoring the status of postmarketing 
studies, primarily because oversight of these studies is not considered 
a priority. Regarding its enforcement of postmarketing study 
requirements, we found FDA has not fully utilized its available 
enforcement tools, even when sponsors have failed to complete required 
studies. 

FDA Has Not Been Routinely Monitoring the Status of Postmarketing 
Studies, although Initiatives to Improve Its Monitoring Are Underway: 

FDA has not been routinely reviewing ASRs to confirm the accuracy of 
information and verify the status of studies within its goal of 90 
days. Our review of information provided by FDA for 35 specific 
required postmarketing studies showed that just over half of the ASRs 
provided for 19 of these studies had never been reviewed by FDA medical 
reviewers, and that for those that had been reviewed, about half were 
not done in a timely manner (see appendix III for a copy of the 
information request we submitted to FDA for these postmarketing 
studies). FDA medical reviewers indicated that they were not able to 
complete scheduled reviews of ASRs, and according to FDA officials, 
this task was a lower priority compared to their other 
responsibilities, such as review of applications for new drugs. 

This finding is consistent with the results reported by the HHS-OIG in 
2006[Footnote 38] as well as FDA's contractor, Booz Allen Hamilton, in 
2008.[Footnote 39] Both found that FDA was not completing reviews of 
ASRs in a timely manner and had not made such reviews a priority. For 
example, the HHS-OIG 2006 study, which examined FDA's review of ASRs 
submitted during fiscal year 2004, found that FDA did not meet its goal 
of reviewing ASRs in 90 days 55 percent of the time. For 26 percent of 
these ASRs, it took FDA more than 180 days to complete its review. 
Similarly, the contractor's study, which examined FDA's review of ASRs 
for requested studies, found that FDA reviewers were missing the 
agency's goal of reviewing ASRs in 90 days 47 percent of the time. Both 
HHS-OIG and the contractor reported that the monitoring of 
postmarketing studies is not a top priority of the agency. 
Specifically, the HHS-OIG reported that FDA officials indicated that 
other tasks, including reviewing drug applications and documenting FDA/ 
industry meetings, are higher priorities than reviewing postmarketing 
studies and ASRs. In addition, the contractor similarly reported that a 
reason for missing review deadlines was heavy reviewer workload and 
that postmarketing study review-related tasks, including review of 
ASRs, were often given a lower priority as compared to application 
reviews. 

Without routine monitoring, FDA's database does not contain complete 
and reliable information, including key dates and study information 
needed to track the progression of postmarketing studies. When we 
requested information on time frames associated with the progression of 
postmarketing studies, FDA could not provide it to us from its 
database. This is because FDA had not entered into the database the key 
dates such as when studies started, are scheduled for completion, and 
their current status. To provide us with the information we requested, 
FDA had to comb through multiple data systems and paper files to 
recreate milestones and status outcome by study. This problem is not 
new. The HHS-OIG reported in 2006 that FDA was not entering into the 
database the information from ASRs, noting, for example, that for 
primarily requested studies, the study start dates were present for 
only 6 percent and original projected completed dates were present for 
only 21 percent. As a result, the HHS-OIG recommended that FDA improve 
its database for monitoring postmarketing studies so that it provides 
timely, accurate, and useful information and ensure that studies are 
being monitored. 

Given that FDA's medical reviewers are not routinely reviewing ASRs and 
do not have reliable information readily available to track the status 
of postmarketing studies, the agency cannot effectively monitor these 
studies. FDA does not know the current status of many postmarketing 
studies or whether they are progressing towards completion. In 
addition, FDA does not know whether drug sponsors are submitting 
complete and accurate ASRs in a timely manner. As a result, FDA lacks 
current, reliable, and easily accessible information on the status of 
open postmarketing studies, and meeting federal reporting requirements 
is difficult.[Footnote 40] The information available regarding 
postmarketing studies can only be as accurate and complete as the data 
contained in FDA's postmarketing database. 

FDA has three initiatives in place to address the agency's oversight 
weaknesses. First, to ensure FDA has current information on the status 
of open postmarketing studies and facilitate the timely review of ASRs, 
FDA retained the contractor in 2008 to review ASRs for all 
postmarketing studies classified as open. As part of this effort, the 
contractor will support the work performed in FDA's medical review 
divisions and conduct an initial review of all newly submitted ASRs. 
This initial review is intended to help ensure that ASRs are reviewed 
within FDA's 90-day review goal. Based on its review, the contractor 
will determine whether the status of the postmarketing studies that 
sponsors listed in the ASR is correct, and meet biweekly with FDA staff 
to discuss ongoing issues. FDA staff will use this information to 
determine the appropriate status of open studies and update its 
tracking database.[Footnote 41] This year FDA renewed its contract with 
Booz Allen Hamilton, which is now scheduled to expire in 2014, to 
continue review of outstanding and new ASRs associated with 
postmarketing studies. 

The second initiative is the creation of a new tracking coordinator 
position within each medical review division with responsibility for a 
variety of tasks related to the tracking of postmarketing studies. 
First, the tracking coordinators will ensure that reviewers are kept 
informed of key postmarketing schedule dates. They will also verify the 
accuracy of postmarketing study information and monitor whether 
expected activities are conducted within specified time lines. Further, 
the coordinator will serve as a key point of contact to interface with 
the contractor and ensure that status information provided by the 
contractor is entered into the postmarketing database. Currently, 
tracking coordinator responsibilities are shared between other existing 
positions within the divisions. FDA officials indicated that they were 
not sure whether the tracking coordinator position would be filled by 
one person in each medical review division or if the duties would be 
divided among many individuals in each division. 

To the extent that the ASR reviews and tracking coordinator initiatives 
are successful, FDA's third initiative--the Document Archiving, 
Reporting and Regulatory Tracking System (DARRTS)--could facilitate the 
monitoring of postmarketing studies for NDAs.[Footnote 42] DARRTS is a 
Web-based system, which according to FDA officials should allow FDA 
staff greater access to information and provide enhancements over the 
current database, such as creating management reports on specific drugs 
and their respective studies. 

FDA Has Not Fully Utilized Its Enforcement Tools Related to Required 
Postmarketing Studies: 

FDA has not fully utilized its two enforcement tools--issuing 
administrative action letters and withdrawing a drug from the market in 
certain cases--to encourage and compel drug sponsors to complete 
required confirmatory postmarketing studies. FDA has the discretion to 
issue administrative action letters to drug sponsors if 1) sponsors are 
late or fail to submit ASRs, or 2) FDA determines that sponsors are not 
sufficiently progressing in completing their studies. 

The extent to which FDA has issued administrative action letters 
related to its oversight of postmarketing studies required under the 
accelerated approval process is unclear. According to FDA officials, 
they do not have a centralized database which tracks the letters they 
have issued to sponsors when sponsors were late or failed to submit 
ASRs, or when FDA determined that sponsors were not sufficiently 
progressing in completing their studies. Our review of FDA's oversight 
of a sample of postmarketing studies required under the accelerated 
approval process suggests its use may be limited when sponsors do not 
submit their ASRs within required time frames. Specifically, we found 
that sponsors were late in submitting ASRs for 12 postmarketing studies 
in our sample. However, FDA issued an administrative action letter to 
the sponsor of only 1 of these 12 studies. 

In addition to sending administrative action letters for drugs approved 
under the accelerated approval process, FDA also may begin expedited 
proceedings to withdraw a drug's approval in a number of situations 
including if it determines that: 1) sponsors are not completing 
required confirmatory postmarketing studies with due diligence, or 2) a 
study failed to confirm the drug's clinical benefit. According to FDA 
officials, the agency has never withdrawn from the market a drug 
approved through the accelerated process due to either of these issues. 
Our review of the 90 applications approved based on a surrogate 
endpoint under the accelerated approval process revealed several 
circumstances that appeared to meet the regulatory conditions for 
withdrawal, but FDA was hesitant to use its enforcement authority. 
Specifically, we found that for 36 of the 90 applications, drug 
sponsors had not fulfilled their confirmatory study requirements by 
establishing the clinical effectiveness of those drugs. This includes 
several applications for drugs that FDA had approved more than 10 years 
ago and for which sponsors had not yet completed all of their required 
studies, and others where the studies failed to confirm the drug's 
clinical effectiveness. An example of sponsors not completing 
confirmatory studies includes the case of ProAmatine: 

* FDA approved ProAmatine in 1996 to treat individuals with low blood 
pressure based on the surrogate endpoint of raising 1-minute standing 
systolic blood pressure. As a condition of approval, it required the 
sponsor, Shire Pharmaceuticals, to conduct a confirmatory study to 
validate long-term clinical benefits. However, the sponsor was not able 
to design and conduct a sufficiently adequate clinical study, and the 
clinical benefit of the drug has never been established. Despite this, 
the sponsor benefited from market exclusivity between 1996 and 2003, 
and sales of this drug generated millions of dollars for the company 
(see appendix IV for total U.S. sales since approval for ProAmatine and 
six other drugs for which applications had not converted to full 
approval as of December 19, 2008). Furthermore, once ProAmatine's 
market exclusivity ended in 2003, FDA approved five generic versions of 
the drug, although the clinical benefit of the drug was never 
confirmed. Recognizing that the study requirement related to ProAmatine 
was still outstanding; in August 2007 FDA posted a letter on its Web 
site inquiring about certain legal and regulatory issues related to the 
generic manufacturers' potential completion of the confirmatory study. 
In August 2008, according to officials, FDA posted another letter in 
which it threatened to withdraw approval of ProAmatine and the generic 
drugs based on its accelerated approval authority if no company 
completed the required studies.[Footnote 43] At the same time it 
indicated that 3 years of market exclusivity would be available if a 
company completes the study and it confirms clinical benefit. As of 
June 2009, nearly 13 years after it approved ProAmatine, FDA had not 
initiated the withdrawal of ProAmatine or the generic versions, but FDA 
officials indicated that it planned to issue a final administrative 
action letter to the sponsor and generic manufacturers in a final 
effort to obtain completion of the required study. 

An example of sponsors completing confirmatory studies which failed to 
confirm the drug's clinical effectiveness includes the case of Iressa: 

* FDA approved Iressa in May 2003 to treat patients with non-small cell 
lung cancer based on the surrogate endpoint that showed that it causes 
significant shrinkage in tumors in about 10 percent of patients. As a 
condition of approval, FDA required the sponsor to conduct a 
postmarketing study to verify the expected clinical benefit. In 
December 2004, FDA announced that the results of the clinical trial in 
1,700 patients indicated that the drug did not prolong survival. 
Despite this, FDA did not utilize its authority to withdraw the drug 
from the market. However, it did take the step of restricting Iressa's 
use to a subset of patients who had already taken the medicine and 
whose doctor believed it was helping them. FDA directed new patients 
wanting to take Iressa to two other available therapies shown in 
studies to improve survival in patients whose cancer has progressed 
while on previous therapies. 

According to FDA officials, they have not developed guidance to specify 
the conditions under which they would exercise their authority to 
withdraw approval of a drug that the agency approved based on surrogate 
endpoints under the accelerated approval process. Under the 
regulations, FDA can initiate expedited withdrawal procedures for drugs 
approved based on surrogate endpoints when sponsors fail to perform 
required confirmatory postmarketing studies with due diligence, or a 
study failed to confirm the drug's clinical effectiveness. Although the 
regulations outline conditions under which FDA could utilize expedited 
withdrawal authority for drugs approved under the accelerated approval 
process based on surrogate endpoints, withdrawal is not required and 
the agency has latitude in determining when to exercise this authority. 
The officials recognized that they have not specified criteria for 
defining how they would implement the due diligence requirement of the 
regulations, such as determining how long it should take a sponsor to 
complete a study or when a sponsor is taking too long, which could 
result in a drug's withdrawal from the market. Without such guidance, 
officials indicated that it is not clear as to when or how to enforce 
the due diligence criteria for withdrawal, but acknowledged that 
ProAmatine may have been a case where they could have fully utilized 
their withdrawal authority. Officials further stated that it would be 
difficult to develop specific criteria for due diligence for withdrawal 
that could be generally applied to the wide range of diseases treated 
by drugs approved under the accelerated approval process. Additionally, 
they questioned the need for such specific criteria, because other than 
the case of ProAmatine, they have rarely faced circumstances where drug 
sponsors were reluctant to work toward completing their postmarketing 
study. Regarding cases when a confirmatory study failed to demonstrate 
a drug's clinical benefit, agency officials indicated they would be 
hesitant to withdraw a drug in such cases. For example, they said that 
despite the confirmatory study results, Iressa may still be effective 
for a number of patients. 

Conclusions: 

The use of surrogate endpoints has been accepted by FDA as a means of 
demonstrating the efficacy of drugs approved through both its 
traditional and accelerated approval processes. While the use of 
surrogate endpoints can expedite the approval of drugs, reliance on 
these endpoints also introduces uncertainty regarding the risks and 
benefits of a drug because the clinical effectiveness is not directly 
measured. Thus their use can lead to the adoption of useless or even 
harmful therapies if the effect on a surrogate endpoint does not 
accurately predict whether treatments provide benefits to patients, or 
if the drug has a smaller than expected benefit and a larger than 
expected adverse effect. 

With the creation of the accelerated approval process in 1992, FDA 
expanded the use of surrogate endpoints, thus expediting the approval 
of certain drugs for serious or life-threatening diseases. While the 
availability of these drugs provided new treatment options to patients, 
it also introduced new elements of uncertainty, as FDA allowed the use 
of surrogate endpoints which had not yet been shown to be valid 
predictors of a drug's clinical effectiveness. Thus, patients could 
potentially find themselves taking drugs approved under the accelerated 
process that may not be clinically effective in treating their illness. 
Recognizing the need to mitigate this uncertainty and risk, FDA 
required that drug sponsors conduct postmarketing studies with due 
diligence to confirm that such drugs actually have clinical benefits. 
It also implemented a process for monitoring the progression and 
completion of these studies, including establishing a database to track 
the progression of the studies, and instituting expedited withdrawal 
procedures for drugs when studies are not completed or if they failed 
to confirm clinical benefits. 

Despite these tools, weaknesses in FDA's monitoring and enforcement 
process hamper its ability to effectively oversee postmarketing 
studies. FDA has not routinely and consistently reviewed the ASRs 
submitted by drug sponsors, and information the agency needs to ensure 
that sponsors are completing required confirmatory studies in a timely 
manner was not consistently entered into its tracking database. 
Therefore, the agency has lacked an effective management information 
system capable of generating data needed to effectively monitor the 
progress of such studies. FDA has recently implemented several 
initiatives to enhance its oversight of postmarketing studies, which 
may provide the agency with an opportunity to develop reliable data 
needed to adequately monitor the progress of studies and improve 
oversight. Additionally, FDA's plans to designate personnel to serve as 
focal points for monitoring postmarketing studies may provide a greater 
emphasis on oversight of such studies. It is too early to tell if these 
initiatives will be successful. While FDA's new system, DARRTS, may 
have features superior to the tracking database it is replacing, the 
root cause of many problems associated with the previous system was not 
the system itself, but the failure to enter information into it. The 
ultimate success of FDA's new initiatives is largely dependent on the 
timely review and prompt entry of ASRs into DARRTS. 

While FDA has implemented measures to enhance its monitoring of 
postmarketing studies, the agency has taken a passive approach to 
enforcing confirmatory study requirements. It has never exercised its 
authority to withdraw a drug it approved based on surrogate endpoints 
under the accelerated approval process, even when such studies have 
been outstanding for nearly 13 years. Further, FDA has not attempted to 
clarify the circumstances under which it would exercise this authority 
and has never developed specific time frames for sponsors to complete 
their confirmatory studies. The combination of its ineffective 
monitoring and lack of criteria outlining when a drug should be 
withdrawn from the market make it difficult for FDA to utilize its 
enforcement authority. Consequently, drugs which have not proven to be 
clinically effective may remain on the market while their associated 
confirmatory studies remain incomplete. 

Recommendations: 

To clarify FDA's enforcement authority under the accelerated approval 
process, we recommend that the Commissioner of FDA take the following 
action: 

* Clarify the conditions under which the agency would utilize its 
authority to expedite the withdrawal of drugs approved based on 
surrogate endpoints under the accelerated approval process if sponsors 
either fail to complete required confirmatory studies with due 
diligence, or if studies are completed, but fail to demonstrate the 
clinical effectiveness of the drugs. 

Agency Comments and Our Evaluation: 

We provided a draft of this report to HHS for review. HHS provided 
written comments from FDA, which are reprinted in appendix V. In its 
comments, FDA disagreed with our recommendation to clarify the 
conditions under which it would utilize its authority to expedite the 
withdrawal of drugs approved based on surrogate endpoints under the 
accelerated approval process. FDA also said that it thought we 
minimized the success of the accelerated approval program. The agency 
stressed that this program has provided millions of patients access to 
new treatments sooner than would have been possible under the 
traditional approval process. In addition, FDA described its efforts to 
improve monitoring of the completion of required and requested 
postmarketing studies. FDA also provided technical comments, which we 
incorporated as appropriate. 

Regarding our recommendation, FDA indicated that it would be difficult, 
if not impossible, to provide further clarification as to when it might 
utilize its authority to expedite withdrawal of a drug approved on the 
basis of surrogate endpoints. FDA acknowledged that there have been 
cases where confirmation of the clinical benefit of drugs approved 
under the accelerated approval program did not occur in a timely 
manner. However, FDA stressed that, unless there are clear safety 
concerns emanating from the confirmatory trial, it must carefully 
assess each case and consider the underlying reasons and consequences 
of all regulatory options, including their potential impact on 
patients. In expressing its disagreement and the need for case-by-case 
assessments, FDA cited the examples of two drugs mentioned in our 
report--ProAmatine and Iressa. In the case of ProAmatine, for which 
adequate clinical trials have never been completed to establish the 
drug's benefit, FDA stated that it would not be appropriate to initiate 
expedited withdrawal, as ProAmatine is the only approved therapy for 
the condition that it treats. FDA stated that rather than providing an 
example of the agency failing to exercise its authority to withdraw 
approval, ProAmatine provides a good example of the complex issues it 
must consider when a clinical benefit has not been confirmed and the 
drug approved remains the only FDA-approved treatment for a serious or 
life-threatening condition. Regarding Iressa, FDA noted that, despite 
the fact that the clinical trials failed to confirm the drug's clinical 
benefit, they nonetheless suggested there were some positive outcomes 
for certain patients. Therefore, rather than withdraw approval of 
Iressa, FDA indicated that it took an appropriate and balanced approach 
by restricting access to those patients who were already receiving 
treatment and whose physicians felt they were benefiting from Iressa 
and to other patients based on physician assessments. 

We recognize that FDA faces challenges in determining whether it should 
initiate the expedited withdrawal of a drug approved under the 
accelerated approval process. Acknowledging that the clinical benefit 
of these drugs is uncertain at the time of their approval, FDA's 1992 
regulations provided that expedited withdrawal of drugs may be 
appropriate when confirmatory studies fail to show clinical benefits. 
Indeed, FDA's ability to require sponsors to complete clinical trials 
with due diligence and its authority to undertake the expedited 
withdrawal of a drug if a clinical benefit is not confirmed are 
specifically cited by FDA in its comments as two important safeguards 
intended to minimize the risks inherent in the accelerated approval 
process. Although nearly 17 years have elapsed since FDA issued the 
accelerated approval regulations, the agency has not attempted to 
define the circumstances under which the authority that this important 
safeguard provides would be used. 

While FDA commented that it would consider exercising its withdrawal 
authority if a clear safety concern emerged from a confirmatory trial, 
which would certainly be appropriate, the regulations governing the use 
of surrogate endpoints and the purpose of the required studies are 
focused on establishing the clinical benefit of a drug, not its safety. 
We agree that it may be challenging for FDA to develop guidance to 
clarify the conditions under which it would utilize its expedited 
withdrawal authority. However, we do not agree that it is impossible-- 
or even too difficult--to do so, nor do we believe that such guidance 
would have to be so prescriptive as to prevent FDA from considering the 
unique circumstances of individual cases. As the scientific experts 
charged with overseeing the use of drugs it approves, FDA should be in 
a position to implement this recommendation. In our view, this would 
serve two purposes. First, this would clarify within FDA when this 
option should be considered in order to mitigate risks to patients 
taking drugs which FDA has approved. Second, it would serve to clarify, 
for drug sponsors, FDA's expectations regarding performance 
requirements related to completing required confirmatory studies, and 
the consequences of not meeting requirements. This would put drug 
sponsors on notice that, although FDA has not utilized this authority 
to date, it remains a viable option, and would enhance the agency's 
ability to effectively carry out its oversight responsibilities. 
Without specific parameters governing the use of this authority, such 
as definitive time frames or other requirements for when sponsors need 
to complete confirmatory studies, there appears to be little likelihood 
that FDA would ever utilize its authority, thus potentially diminishing 
the value of what FDA considers an important safeguard. 

Regarding the examples of ProAmatine and Iressa, we recognize that FDA 
faced numerous scientific and ethical issues in overseeing the use of 
these drugs, and that it ultimately needed to balance such factors when 
determining the best approach to take in overseeing the completion of 
required confirmatory studies. However, we chose to discuss these drugs 
because the very circumstances surrounding them illustrated specific 
oversight issues which FDA considered when it developed the 1992 
accelerated approval regulations. The circumstances involving these 
drugs also highlighted the challenges that FDA faces in determining how 
to conduct its oversight responsibilities. As delineated in the 
regulations, the purpose of these confirmatory studies is to verify and 
describe the clinical benefits of these drugs. When sponsors fulfill 
these requirements, they establish the clinical evidence similar to 
what would ordinarily have been required were the drugs reviewed under 
the traditional approval process. In neither of these two instances has 
this level of evidence been established--no adequate study has ever 
been completed to confirm the clinical benefits of ProAmatine and, 
while the studies for Iressa were conducted, they failed to confirm any 
clinical benefit. 

Specifically, in the case of ProAmatine, as we noted in our report, FDA 
officials acknowledged that this matter may have gone on too long 
without being resolved. Although FDA's comments indicate that it 
believes the sponsor conducted trials with due diligence, 13 years have 
elapsed, and although the drug is also available generically, 
sufficient confirmatory trials have not been completed. As FDA 
indicates in its comments, obtaining the completion of the trials has 
become a complex matter and it is still seeking ways to encourage one 
or more of the generic manufacturers to conduct the confirmatory 
clinical trials. Thus, we believe that this example raises questions 
about whether the sponsor had displayed due diligence in meeting the 
confirmatory study requirements as well as what circumstances would 
lead the agency to exercise its withdrawal authority. We provided the 
example of Iressa to illustrate the challenges that FDA faces in 
overseeing the accelerated approval process. In developing the 1992 
regulations, FDA specifically anticipated a situation similar to that 
posed by Iressa--the completion of a confirmatory study that failed to 
demonstrate a drug's clinical benefit. Although 17 years have passed, 
FDA has not established guidance clarifying how it would exercise its 
expedited withdrawal authority in such circumstances. While FDA's 
actions in these two cases may have been appropriate, they nonetheless 
serve to illustrate the need to clarify use of the expedited withdrawal 
authority, consistent with our recommendation. 

Regarding FDA's statement that the accelerated approval process has 
been very successful, and has resulted in the early approval of many 
significant therapeutic advances for patients, we believe that this 
conclusion is a matter of scientific judgment, and beyond the scope of 
our review. However, our report clearly identified both the purpose of 
the accelerated approval process, and the specific drugs that FDA has 
approved using this process. Thus, we believe we provide an accurate 
factual description of the drugs approved on the basis of surrogate 
endpoints through this process since it was instituted in 1992. 
Further, we do not believe that our emphasis on examples of FDA's 
decisions regarding specific drugs and the results of their associated 
confirmatory studies, in any way minimizes the benefits provided by the 
accelerated process. Instead they are meant to illustrate challenges 
associated with FDA's oversight of such an important program. 

Finally, regarding FDA's ongoing efforts to improve its oversight of 
postmarketing studies in both the accelerated and traditional approval 
processes, FDA acknowledged that its oversight of postmarketing studies 
in general has been inadequate, and agreed that improvements are 
needed. The agency indicated that because of inadequate staffing and 
information technology resources and competing priorities, its past 
tracking of postmarketing studies has not been timely. FDA stated that 
it has begun to implement a number of improvements to ensure 
appropriate oversight, more efficient tracking, and expeditious review 
of postmarketing submissions from sponsors, but stated that these 
efforts were not fully reflected in our report. We believe our report's 
discussion of FDA's efforts to improve the monitoring of postmarketing 
studies sufficiently addresses FDA's ongoing initiatives. Specifically, 
we discuss its contract with Booz Allen Hamilton, the establishment of 
new tracking coordinator positions which are outlined in its revised 
Manual of Policies and Procedures, and planned improvements in data 
reporting capabilities through the DARRTS system. In particular, FDA 
highlighted the results of efforts from its contractor, Booz Allen 
Hamilton, which FDA said resulted in extensive updating of the data in 
the agency's tracking database. However, the results of these updates 
to the database by FDA's contractor became available subsequent to our 
work, and thus could not be reflected in the data we presented in this 
report. The fact that extensive updates needed to be made, however, 
confirms our assessment that FDA's monitoring and oversight has been 
ineffective. We are encouraged that FDA is taking steps to facilitate 
timely oversight of postmarketing studies through improved tracking 
provided by its contractor and a new database. 

As agreed with your office, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. At that time, we will send copies to the 
Commissioner of FDA and appropriate congressional committees. The 
report also will be available at no charge on GAO's Web site at 
[hyperlink, http://www.gao.gov]. 

If you or your staff have any questions about this report, please 
contact me at (202) 512-7114 or crossem@gao.gov. Contact points for our 
Offices of Congressional Relations and Public Affairs may be found on 
the last page of this report. GAO staff who made major contributions to 
this report are listed in appendix VI. 

Sincerely yours, 

Signed by: 

Marcia Crosse: 
Director, Health Care: 

[End of section] 

Appendix I: Applications for Drugs Approved under FDA's Accelerated 
Approval Process Using Surrogate Endpoints: 

Since the Food and Drug Administration (FDA) began approving drugs 
under the accelerated approval process in 1992, it has approved 90 
applications based on surrogate endpoints. In several cases, FDA 
approved multiple applications for the same drug; as a result, as of 
November 20, 2008, FDA had approved 64 drugs associated with these 
applications under the accelerated approval process. Table 5 provides a 
description of each of these 64 drugs, their new drug application (NDA) 
or biologic license application (BLA) numbers, and the surrogate 
endpoints used for approval. 

Table 5: NDAs and BLAs Approved Based on Surrogate Endpoints under the 
Accelerated Approval Process, from June 19, 1992-November 20, 2008: 

Drug name: Hivid; 
NDA/BLA number: 20199; 
Approval date: June 19, 1992; 
Approved indication: Monotherapy and combination therapy for treatment 
of Human Immunodeficiency Virus infection in specific patients; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and p24 existence (antigen which indicates 
HIV infection). 

Drug name: Betaseron; 
NDA/BLA number: 103471[A]; 
Approval date: July 23, 1993; 
Approved indication: Treatment of relapsing-remitting multiple 
sclerosis; 
Surrogate endpoint(s) used to approve application: MRI evaluations of 
brain lesions[B]. 

Drug name: Biaxin; 
NDA/BLA number: 50697; 
Approval date: Dec. 23, 1993; 
Approved indication: Treatment of disseminated mycobacterial infections 
due to specific bacteria; 
Surrogate endpoint(s) used to approve application: Clearance of 
bacteremia[B]. 

Drug name: Biaxin; 
NDA/BLA number: 50698; 
Approval date: Dec. 23, 1993; 
Approved indication: Treatment of disseminated mycobacterial infections 
due to specific bacteria; 
Surrogate endpoint(s) used to approve application: Clearance of 
bacteremia[B]. 

Drug name: Zerit; 
NDA/BLA number: 20412; 
Approval date: June 24, 1994; 
Approved indication: Treatment of advanced Human Immunodeficiency Virus 
infection in specific adult patients; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and p24 existence (antigen which indicates 
HIV infection). 

Drug name: Casodex; 
NDA/BLA number: 20498; 
Approval date: Oct. 4, 1995; 
Approved indication: Combination therapy for the treatment of advanced 
prostate cancer; 
Surrogate endpoint(s) used to approve application: Time to treatment 
failure[B]. 

Drug name: Epivir; 
NDA/BLA number: 20564; 
Approval date: Nov. 17, 1995; 
Approved indication: Combination therapy for treatment of Human 
Immunodeficiency Virus infection in certain circumstances; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Epivir; 
NDA/BLA number: 20596; 
Approval date: Nov. 17, 1995; 
Approved indication: Combination therapy for treatment of Human 
Immunodeficiency Virus infection in certain circumstances; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Doxil; 
NDA/BLA number: 50718; 
Approval date: Nov. 17, 1995; 
Approved indication: Treatment of Kaposi's sarcoma in specific patients 
with Acquired Immunodeficiency Syndrome; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Doxil; 
NDA/BLA number: 50718-SE1-006; 
Approval date: June 28, 1999; 
Approved indication: Treatment of metastatic carcinoma of the ovary in 
specific patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Invirase; 
NDA/BLA number: 20628; 
Approval date: Dec. 6, 1995; 
Approved indication: Combination therapy for treatment of advanced 
Human Immunodeficiency Virus infection in specific patients; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Norvir; 
NDA/BLA number: 20659; 
Approval date: Mar. 1, 1996; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Norvir; 
NDA/BLA number: 20680; 
Approval date: Mar. 1, 1996; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Crixivan; 
NDA/BLA number: 20685; 
Approval date: Mar. 13, 1996; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection when antiretroviral therapy is warranted; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Ethyol; 
NDA/BLA number: 20221-SE1-002; 
Approval date: Mar. 15, 1996; 
Approved indication: Treatment for the toxicities associated with 
intensive regimens of specific chemotherapy; 
Surrogate endpoint(s) used to approve application: Creatinine clearance 
and response rate to assess tumor protection. 

Drug name: Taxotere; 
NDA/BLA number: 20449; 
Approval date: May 14, 1996; 
Approved indication: Treatment of locally advanced or metastatic breast 
cancer in specific patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Camptosar; 
NDA/BLA number: 20571; 
Approval date: June 14, 1996; 
Approved indication: Treatment of metastatic carcinoma of the colon or 
rectum in certain circumstances; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Viramune; 
NDA/BLA number: 20636; 
Approval date: June 21, 1996; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus infection in specific patients; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Viramune; 
NDA/BLA number: 20636-SE1-009; 
Approval date: Sept. 11, 1998; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection in pediatric patients; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Viramune; 
NDA/BLA number: 20933; 
Approval date: Sept. 11, 1998; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection in pediatric patients; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: ProAmatine; 
NDA/BLA number: 19815; 
Approval date: Sept. 6, 1996; 
Approved indication: Treatment of idiopathic orthostatic hypotension; 
Surrogate endpoint(s) used to approve application: Increase in 1-minute 
standing systolic blood pressure. 

Drug name: Viracept; 
NDA/BLA number: 20778; 
Approval date: Mar. 14, 1997; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection in children when antiretroviral therapy is indicated; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Viracept; 
NDA/BLA number: 20779; 
Approval date: Mar. 14, 1997; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection when antiretroviral therapy is indicated; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Rescriptor; 
NDA/BLA number: 20705; 
Approval date: April 4, 1997; 
Approved indication: Treatment of Human Immunodeficiency Virus-1 
infection; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cells) and viral load (HIV-RNA). 

Drug name: Xeloda; 
NDA/BLA number: 20896; 
Approval date: April 30, 1998; 
Approved indication: Treatment of a specific type of metastasis breast 
cancer in certain patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Sulfamylon; 
NDA/BLA number: 19832; 
Approval date: June 5, 1998; 
Approved indication: Treatment of bacterial infections in certain 
circumstances; 
Surrogate endpoint(s) used to approve application: Necessity to change 
topical antimicrobial treatment during the first 5 days of application 
due to infection or colonization. 

Drug name: Priftin; 
NDA/BLA number: 21024; 
Approval date: June 22, 1998; 
Approved indication: Treatment of pulmonary tuberculosis; 
Surrogate endpoint(s) used to approve application: Negative sputum 
culture at 6 month post-treatment (6 month relapse rate). 

Drug name: Sustiva; 
NDA/BLA number: 20972; 
Approval date: Sept. 17, 1998; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Ziagen; 
NDA/BLA number: 20977; 
Approval date: Dec. 17, 1998; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus infection; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Ziagen; 
NDA/BLA number: 20978; 
Approval date: Dec. 17, 1998; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus infection; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Ontak; 
NDA/BLA number: 103767[A]; 
Approval date: Feb. 5, 1999; 
Approved indication: Treatment of a specific type of persistent or 
recurrent Cutaneous T-cell Lymphoma; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: DepoCyt; 
NDA/BLA number: 21041; 
Approval date: April 1, 1999; 
Approved indication: Treatment of lymphmatous meningitis; 
Surrogate endpoint(s) used to approve application: Response rate 
(complete cytologic response with absence of neurologic 
progression)[C]. 

Drug name: Agenerase; 
NDA/BLA number: 21007; 
Approval date: April 15, 1999; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Agenerase; 
NDA/BLA number: 21039; 
Approval date: April 15, 1999; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Temodar; 
NDA/BLA number: 21029; 
Approval date: Aug. 11, 1999; 
Approved indication: Treatment of refractory anaplastic astrocytoma in 
specific adult patients; 
Surrogate endpoint(s) used to approve application: Progression free 
survival at 6 months and objective response. 

Drug name: Synercid; 
NDA/BLA number: 50747; 
Approval date: Sept. 21, 1999; 
Approved indication: Treatment of serious or life-threatening 
infections associated with vancomycin-resistant enterococcus faecium 
bacteremia; 
Surrogate endpoint(s) used to approve application: Clearance of 
vancomycin-resistant Enterococcus faecium bacteremia. 

Drug name: Celebrex; 
NDA/BLA number: 21156; 
Approval date: Dec. 23, 1999; 
Approved indication: To reduce the number of adenomatous colorectal 
polyps in familial adenomatous polyposis, and as an adjunct to usual 
care; 
Surrogate endpoint(s) used to approve application: Mean reduction in 
colorectal polyp count. 

Drug name: Mylotarg; 
NDA/BLA number: 21174; 
Approval date: May 17, 2000; 
Approved indication: Treatment of CD33+ acute myeloid leukemia in 
specific patients; 
Surrogate endpoint(s) used to approve application: Complete responses 
(blasts, bone marrow, CBC, transfusions)[C]. 

Drug name: Cipro; 
NDA/BLA number: 19537-SE1-038; 
Approval date: Aug. 30, 2000; 
Approved indication: Treatment of inhalation anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Serum CIPRO 
concentrations. 

Drug name: Cipro; 
NDA/BLA number: 19847-SE1-024; 
Approval date: Aug. 30, 2000; 
Approved indication: 
Treatment of inhalation anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Serum CIPRO 
concentrations. 

Drug name: Cipro; 
NDA/BLA number: 19857-SE1-027; 
Approval date: Aug. 30, 2000; 
Approved indication: Treatment of inhalation anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Serum CIPRO 
concentrations. 

Drug name: Cipro; 
NDA/BLA number: 19858-SE1-021; 
Approval date: Aug. 30, 2000; 
Approved indication: Treatment of inhalation anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Serum CIPRO 
concentrations. 

Drug name: Cipro; 
NDA/BLA number: 20780-SE1-008; 
Approval date: Aug. 30, 2000; 
Approved indication: Treatment of inhalation anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Serum concentrations 
(compared to effective animal concentrations). 

Drug name: Kaletra; 
NDA/BLA number: 21226; 
Approval date: Sept.15, 2000; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection in specific patients; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Kaletra; 
NDA/BLA number: 21251; 
Approval date: Sept. 15, 2000; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection in specific patients; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Trizivir; 
NDA/BLA number: 21205; 
Approval date: Nov. 14, 2000; 
Approved indication: Treatment of Human Immunodeficiency Virus 
infection; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Campath; 
NDA/BLA number: 103948[A]; 
Approval date: May 7, 2001; 
Approved indication: Treatment of B-cell chronic lymphocytic leukemia 
in specific patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Gleevec; 
NDA/BLA number: 21335; 
Approval date: May 10, 2001; 
Approved indication: Treatment of chronic myeloid leukemia in certain 
circumstances; 
Surrogate endpoint(s) used to approve application: 
Hematologic/cytogenic response. 

Drug name: Gleevec; 
NDA/BLA number: 21335-SE1-001; 
Approval date:Feb. 1, 2002; 
Approved indication: Treatment of Kit (CD117) positive unresectable 
and/or metastatic malignant gastrointestinal stromal tumors; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Gleevec; 
NDA/BLA number: 21335-SE1-004; 
Approval date:Dec. 20, 2002; 
Approved indication: Treatment of Philadelphia chromosome positive 
chronic myeloid leukemia in specific patients; 
Surrogate endpoint(s) used to approve application: Time to accelerated 
phase or blast crisis. 

Drug name: Gleevec; 
NDA/BLA number: 21588; 
Approval date: April 18, 2003; 
Approved indication: Treatment of newly diagnosed adult patients with 
Philadelphia chromosome positive chronic myeloid leukemia in specific 
patients under certain circumstances; 
Surrogate endpoint(s) used to approve application: 
Hematologic/cytogenic response (CML); response rate (GIST); time to 
accelerated phase or blast crisis (1st line CML)[C]. 

Drug name: Gleevec; 
NDA/BLA number: 21335-SE5-003; 
Approval date: May 20, 2003; 
Approved indication: Treatment of pediatric patients with Ph+ chronic 
phase chronic myeloid leukemia whose disease has recurred after stem 
cell transplant or who are resistant to interferon alpha therapy; 
Surrogate endpoint(s) used to approve application: Cytogenic response. 

Drug name: Gleevec; 
NDA/BLA number: 21588-SE5-001; 
Approval date: May 20, 2003; 
Approved indication: Treatment of pediatric patients with Ph+ chronic 
phase chronic myeloid leukemia whose disease has recurred after stem 
cell transplant or who are resistant to interferon alpha therapy; 
Surrogate endpoint(s) used to approve application: Cytogenic response. 

Drug name: Gleevec; 
NDA/BLA number: 21588-SE1-016; 
Approval date: Sept. 27, 2006; 
Approved indication: Treatment of newly diagnosed Philadelphia positive 
chronic myeloid leukemia in pediatric patients; 
Surrogate endpoint(s) used to approve application: Major hematologic 
and complete cytogenic response. 

Drug name: Viread; 
NDA/BLA number: 21356; 
Approval date: Oct. 26, 2001; 
Approved indication: Combination antiretroviral treatment of Human 
Immunodeficiency Virus-1 infection in adults; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Zevalin; 
NDA/BLA number: 125019[A]; 
Approval date: Feb. 19, 2002; 
Approved indication: Treatment of relapsed or refractory low-grade, 
follicular or transformed B-cell non-Hodgkin's lymphoma in specific 
patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Eloxatin; 
NDA/BLA number: 21492; 
Approval date: Aug. 9, 2002; 
Approved indication: Treatment of metastatic carcinoma of the colon or 
rectum in specific patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Armidex; 
NDA/BLA number: 20541-SE1-010; 
Approval date: Sept. 5, 2002; 
Approved indication: Adjuvant treatment of postmenopausal women with 
specific breast cancer; 
Surrogate endpoint(s) used to approve application: Disease free 
survival. 

Drug name: Fuzeon; 
NDA/BLA number: 21481; 
Approval date: Mar. 13, 2003; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection in specific patients; 
Surrogate endpoint(s) used to approve application: Viral load (HIV- 
RNA). 

Drug name: Fabrazyme; 
NDA/BLA number: 103979[A]; 
Approval date: April 24, 2003; 
Approved indication: Treatment of Fabry disease; 
Surrogate endpoint(s) used to approve application: Intracellular 
substrate accumulation in the vascular endothelium. 

Drug name: Iressa; 
NDA/BLA number: 21399; 
Approval date: May 5, 2003; 
Approved indication: Treatment of locally advanced or metastatic non-
small cell lung cancer in specific patients under certain 
circumstances; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Velcade; 
NDA/BLA number: 21602; 
Approval date: May 13, 2003; 
Approved indication: Treatment of multiple myeloma in specific patients 
under certain circumstances; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Erbitux; 
NDA/BLA number: 125084[A]; 
Approval date: Feb. 12, 2004; 
Approved indication: In combination with other drugs for the treatment 
of EGFR-expressing, metastatic colorectal carcinoma in specific 
patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Truvada; 
NDA/BLA number: 21752; 
Approval date: Aug. 2, 2004; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection in adults; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Alimta; 
NDA/BLA number: 21677; 
Approval date: Aug. 19, 2004; 
Approved indication: Treatment of patients with locally advanced or 
metastatic non-small cell lung cancer after prior chemotherapy; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Luveris; 
NDA/BLA number: 21322; 
Approval date: Oct. 8, 2004; 
Approved indication: Concomitantly use for stimulation of follicular 
development in infertile hypogonadotropic hypogonadal women with 
profound LH deficiency; 
Surrogate endpoint(s) used to approve application: Follicular 
development. 

Drug name: Femara; 
NDA/BLA number: 20726-SE1-011; 
Approval date: Oct. 29, 2004; 
Approved indication: Extended adjuvant treatment of early breast cancer 
in specific postmenopausal women; 
Surrogate endpoint(s) used to approve application: Disease free 
survival. 

Drug name: Femara; 
NDA/BLA number: 20726-SE1-012; 
Approval date: Dec. 28, 2005; 
Approved indication: Adjuvant treatment of postmenopausal women with 
hormone receptor positive early breast cancer; 
Surrogate endpoint(s) used to approve application: Disease free 
survival. 

Drug name: Levaquin; 
NDA/BLA number: 20634-SE1-035; 
Approval date: Nov. 24, 2004; 
Approved indication: Treatment of inhalation anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Drug exposure in 
surviving Rhesus monkeys compared to human plasma concentrations. 

Drug name: Levaquin; 
NDA/BLA number: 20635-SE1-035; 
Approval date: Nov. 24, 2004; 
Approved indication: Treatment of inhalation anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Drug exposure in 
surviving Rhesus monkeys compared to human plasma concentrations. 

Drug name: Levaquin; 
NDA/BLA number: 21721-SE1-003; 
Approval date: Nov. 24, 2004; 
Approved indication: Treatment of inhalational anthrax (post-exposure); 
Surrogate endpoint(s) used to approve application: Plasma 
concentrations. 

Drug name: Levaquin;
NDA/BLA number: 20634-SE5-047; 
Approval date: May 5, 2008; 
Approved indication: Treatment of inhalational anthrax (post-exposure) 
in pediatric patients (over 6 months of age) in certain circumstances; 
Surrogate endpoint(s) used to approve application: PK modeling in 
pediatric patients to extrapolate plasma concentration/exposure from 
adult data. 

Drug name: Levaquin;
NDA/BLA number: 20635-SE5-051; 
Approval date: May 5, 2008; 
Approved indication: Treatment of inhalational anthrax (post-exposure) 
in pediatric patients (over 6 months of age) in certain circumstances; 
Surrogate endpoint(s) used to approve application: PK modeling in 
pediatric patients to extrapolate plasma concentration/exposure from 
adult data. 

Drug name: Levaquin;
NDA/BLA number: 21721-SE5-015; 
Approval date: May 5, 2008; 
Approved indication: Treatment of inhalational anthrax (post-exposure) 
in pediatric patients (over 6 months of age) in certain circumstances; 
Surrogate endpoint(s) used to approve application: PK modeling in 
pediatric patients to extrapolate plasma concentration/exposure from 
adult data. 

Drug name: Bexxar; 
NDA/BLA number: 125011-24[A]; 
Approval date: Dec. 22, 2004; 
Approved indication: Treatment of relapsed or refractory, low grade, 
follicular or transformed CD20 positive non-Hodgkin's lymphoma in 
specific patients; 
Surrogate endpoint(s) used to approve application: Response rate[C]. 

Drug name: Clolar; 
NDA/BLA number: 21673; 
Approval date: Dec. 28, 2004; 
Approved indication: Treatment of pediatric patients 1 to 21 years old 
with relapsed or refractory acute lymphoblastic leukemia after at least 
two prior regimens; 
Surrogate endpoint(s) used to approve application: Complete response 
rate and complete response without platelet recovery[C]. 

Drug name: Aptivus; 
NDA/BLA number: 21814; 
Approval date: June 22, 2005; 
Approved indication: Treatment of Human Immunodeficiency Virus-1 
infection in specific adult patients under certain circumstances; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Arranon; 
NDA/BLA number: 21877; 
Approval date: Oct. 28, 2005; 
Approved indication: Treatment of T-cell acute lymphoblastic leukemia 
and T-cell lymphoblastic lymphoma in specific patients; 
Surrogate endpoint(s) used to approve application: Complete response 
rate[C]. 

Drug name: Exjade; 
NDA/BLA number: 21882; 
Approval date: Nov. 2, 2005; 
Approved indication: Treatment of chronic iron overload due to blood 
transfusions in patients 2 years of age and older; 
Surrogate endpoint(s) used to approve application: Lowering of liver 
iron content. 

Drug name: Sutent; 
NDA/BLA number: 21968; 
Approval date: Jan. 26, 2006; Approved indication: 
Treatment of advanced renal cell carcinoma; Surrogate endpoint(s) used 
to approve application: Response rate[C]. 

Drug name: Thalomid; 
NDA/BLA number: 21430; 
Approval date: May 25, 2006; 
Approved indication: Treatment of newly diagnosed multiple myeloma; 
Surrogate endpoint(s) used to approve application: Response rate (serum 
or urine paraprotein)[C]. 

Drug name: Prezista; 
NDA/BLA number: 21976; 
Approval date: June 23, 2006; 
Approved indication: Treatment of human immunodeficiency virus 
infection in specific patients; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Sprycel; 
NDA/BLA number: 21986; 
Approval date: June 28, 2006; 
Approved indication: Treatment of chronic myeloid leukemia with 
resistance or intolerance to prior therapy including imatinib in 
adults; 
Surrogate endpoint(s) used to approve application: Major cytogenic 
response. 

Drug name: Vectibix; 
NDA/BLA number: 125147[A]; 
Approval date: Sept. 27, 2006; 
Approved indication: Treatment of specific metastatic colorectal 
carcinoma with disease progression on or following fluoropyrimidine 
oxaliplatin and irinotecan containing chemotherapy regimens; 
Surrogate endpoint(s) used to approve application: Progression-free 
survival. 

Drug name: Selzentry; 
NDA/BLA number: 22128; 
Approval date: Aug. 6, 2007; 
Approved indication: Treatment of CCR5-tropic Human Immunodeficiency 
Virus-1; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Isentress; 
NDA/BLA number: 22145; 
Approval date: Oct. 12, 2007; Approved indication: Treatment of Human 
Immunodeficiency Virus infection in specific patients; Surrogate 
endpoint(s) used to approve application: Viral load (HIV-RNA). 

Drug name: Tasigna; 
NDA/BLA number: 22068; 
Approval date: Oct. 29, 2007; 
Approved indication: Treatment of chronic phase and accelerated phase 
Philadelphia chromosome positive chronic myelogenous leukemia in 
specific adult patients; 
Surrogate endpoint(s) used to approve application: Major cytogenic 
response and hematologic response. 

Drug name: Intelence; 
NDA/BLA number: 22187; 
Approval date: Jan. 18, 2008; 
Approved indication: Combination antiretroviral treatment for Human 
Immunodeficiency Virus-1 infection in specific patients; 
Surrogate endpoint(s) used to approve application: Viral load (HIV-
RNA). 

Drug name: Avastin; 
NDA/BLA number: 125085-91[A]; 
Approval date: Feb. 22, 2008; 
Approved indication: Treatment of breast cancer in specific patients; 
Surrogate endpoint(s) used to approve application: Progression-free 
survival. 

Drug name: Promacta; 
NDA/BLA number: 22291; 
Approval date: Nov. 20, 2008; 
Approved indication: Treatment of thrombocytopenia in specific 
patients; 
Surrogate endpoint(s) used to approve application: An increase from the 
baseline platelet count to a count greater than or equal to 50,000/mcL. 

Source: GAO analysis of FDA data. 

[A] Applications for therapeutic BLAs. These therapeutic biologics were 
transferred from the Center for Biologics Evaluation and Research to 
the Center for Drug Evaluation and Research on June 30, 2003. 

[B] In addition to the surrogate endpoint used for approval, there was 
evidence of clinical benefit supporting the approval of this 
application. 

[C] Response rate and complete response surrogate endpoints are tumor 
assessment endpoints which measure the proportion of patients with 
tumor size reduction of a predefined amount and for a minimum time 
period. 

[End of table] 

[End of section] 

Appendix II: Applications for Drugs Approved under FDA's Traditional 
Process Using Surrogate Endpoints: 

Between January 1998 and June 2008, the Food and Drug Administration 
(FDA) approved 204 New Molecular Entity (NME) drugs under its 
traditional process. Of those 204, 69 NME drugs were approved using 
surrogate endpoints. Table 6 provides a description of each of the 69 
NME drugs, their new drug application (NDA) numbers, and the surrogate 
endpoints used for approval. 

Table 6: NDAs Approved Based on Surrogate Endpoints under the 
Traditional Approval Process, from January 1, 1998-June 30, 2008: 

Drug name: Refludan; 
NDA number: 20807; 
Approval date: Mar. 6, 1998; 
Approved indication: Anticoagulation treatment in patients with heparin-
induced thrombocytopenia and thromboembolic disease; 
Surrogate endpoint(s) used to approve application: Time courses of 
platelets and activated partial thromboplastin time for regimen A1, A2, 
and B, and on the time course of Ecarin clotting time (ECT) for regimen 
C. 

Drug name: Lotemax; 
NDA number: 20583; 
Approval date: Mar. 9, 1998; 
Approved indication: Treatment of post-operative inflammation and 
uveitis[A]; 
Surrogate endpoint(s) used to approve application: Measurements of cell 
and flare and resolution of anterior chamber cell. 

Drug name: Actonel; 
NDA number: 20835; 
Approval date: Mar. 27, 1998; 
Approved indication: Treatment of Paget's disease of bone in specific 
patients; 
Surrogate endpoint(s) used to approve application: Serum alkaline 
phosphatase levels. 

Drug name: Azopt; 
NDA number: 20816; 
Approval date: April 1, 1998; 
Approved indication: Treatment of elevated intraocular pressure in 
specific patients; 
Surrogate endpoint(s) used to approve application: Intraocular pressure 
levels. 

Drug name: Zemplar injection; 
NDA number: 20819; 
Approval date: April 17, 1998; 
Approved indication: Prevention and treatment of secondary 
hyperparathyroidism associated with chronic renal failure; 
Surrogate endpoint(s) used to approve application: Intact parathyroid 
hormone levels. 

Drug name: Atacand; 
NDA number: 20838; 
Approval date: June 4, 1998; 
Approved indication: Treatment of hypertension; 
Surrogate endpoint(s) used to approve application: Systolic and 
diastolic pressure levels. 

Drug name: Vitravene; 
NDA number: 20961; 
Approval date: Aug. 26, 1998; 
Approved indication: Treatment of cytomegalovirus retinitis in specific 
patients with Acquired Immunodeficiency Syndrome; 
Surrogate endpoint(s) used to approve application: Median time to 
cytomegalovirus retinitis progression. 

Drug name: Valstar; 
NDA number: 20892; 
Approval date: Sept. 25, 1998; 
Approved indication: Treatment of BCG-refractory carcinoma in situ of 
the urinary bladder in specific patients; 
Surrogate endpoint(s) used to approve application: Time to recurrence 
of disease after treatment compared to recurrence after previous 
courses of intravesical therapy. 

Drug name: Renagel; 
NDA number: 20926; 
Approval date: Oct. 30, 1998; 
Approved indication: Treatment of end-stage renal disease; 
Surrogate endpoint(s) used to approve application: Serum phosphorous 
levels. 

Drug name: Micardis; 
NDA number: 20850; 
Approval date: Nov. 10, 1998; 
Approved indication: Treatment of hypertension; 
Surrogate endpoint(s) used to approve application: Blood pressure 
levels. 

Drug name: Ferrlecit; 
NDA number: 20955; 
Approval date: Feb. 18, 1999; 
Approved indication: Treatment for iron deficiency in specific patients 
undergoing chronic hemodialysis; 
Surrogate endpoint(s) used to approve application: Hemoglobin levels. 

Drug name: Avandia; 
NDA number: 21071; 
Approval date: May 25, 1999; 
Approved indication: Treatment of type 2 diabetes mellitus; 
Surrogate endpoint(s) used to approve application: Blood sugar (fasting 
plasma glucose and HBA1c levels). 

Drug name: Hectorol; 
NDA number: 20862; 
Approval date: June 9, 1999; 
Approved indication: Treatment of secondary hyperparathyroidism in 
specific patients; 
Surrogate endpoint(s) used to approve application: Intact parathyroid 
hormone levels. 

Drug name: Actos; 
NDA number: 21073; 
Approval date: July 15, 1999; 
Approved indication: Treatment of type 2 diabetes mellitus; 
Surrogate endpoint(s) used to approve application: Blood sugar (fasting 
plasma glucose and HBA1c levels). 

Drug name: Aromasin; 
NDA number: 20753; 
Approval date: Oct. 21, 1999; 
Approved indication: Treatment of advanced breast cancer in 
postmenopausal women; 
Surrogate endpoint(s) used to approve application: Objective response 
rate (partial and complete)[B]. 

Drug name: Protonix; 
NDA number: 20987; 
Approval date: Feb. 2, 2000; 
Approved indication: Treatment of erosive esophagitis associated with 
gastroesophageal reflux disease; 
Surrogate endpoint(s) used to approve application: Healing of lesions 
to grade 1 or 0 in the Hetzel-Dent scale. 

Drug name: Lantus; 
NDA number: 21081; 
Approval date: April 20, 2000; 
Approved indication: Treatment of type 1 diabetes mellitus in adults or 
pediatric patients or type 2 diabetes mellitus in specific adult 
patients; 
Surrogate endpoint(s) used to approve application: Blood sugar 
(glycated hemoglobin levels). 

Drug name: Welchol; 
NDA number: 21176; 
Approval date: May 26, 2000; 
Approved indication: Treatment of elevated LDL cholesterol in specific 
patients; 
Surrogate endpoint(s) used to approve application: LDL cholesterol 
levels. 

Drug name: NovoLog; 
NDA number: 20986; 
Approval date: June 7, 2000; 
Approved indication: Treatment of diabetes mellitus in adult patients; 
Surrogate endpoint(s) used to approve application: Glycemic control, 
the rates of hypoglycemia, and the incidence of ketosis. 

Drug name: Trelstar Depot; 
NDA number: 20715; 
Approval date: June 15, 2000; 
Approved indication: Treatment of advanced prostate cancer; 
Surrogate endpoint(s) used to approve application: Achievement of 
castration by Day 29 and maintenance of castration levels of serum 
testosterone from Day 57 through Day 253. 

Drug name: Rescula; 
NDA number: 21214; 
Approval date: Aug. 3, 2000; 
Approved indication: Treatment of open-angle glaucoma or ocular 
hypertension; 
Surrogate endpoint(s) used to approve application: Intraocular pressure 
levels. 

Drug name: Cetrotide; 
NDA number: 21197; 
Approval date: Aug. 11, 2000; 
Approved indication: Treatment of premature luteinizing hormone surges 
in women undergoing controlled ovarian stimulation; 
Surrogate endpoint(s) used to approve application: Luteinizing hormone 
surge. 

Drug name: Trisenox; 
NDA number: 21248; 
Approval date: Sept. 25, 2000; 
Approved indication: Treatment of acute promyelocytic leukemia; 
Surrogate endpoint(s) used to approve application: Cytogenic conversion 
to no detection of the acute promyelocytic leukemia chromosome 
rearrangement. 

Drug name: Starlix; 
NDA number: 21204; 
Approval date: Dec. 22, 2000; 
Approved indication: Treatment of type 2 diabetes mellitus; 
Surrogate endpoint(s) used to approve application: Blood sugar (fasting 
plasma glucose and HBA1c levels). 

Drug name: Foradil Aerolizer; 
NDA number: 20831; 
Approval date: Feb. 16, 2001; 
Approved indication: Treatment of asthma and prevention of 
bronchospasm; 
Surrogate endpoint(s) used to approve application: Forced expiratory 
volume in one second. 

Drug name: Lumigan; 
NDA number: 21275; 
Approval date: Mar. 16, 2001; 
Approved indication: Treatment of elevated intraocular pressure in 
certain populations; 
Surrogate endpoint(s) used to approve application: Intraocular pressure 
levels. 

Drug name: Travatan; 
NDA number: 21257; 
Approval date: Mar. 16, 2001; 
Approved indication: Treatment of intraocular pressure in certain 
populations; 
Surrogate endpoint(s) used to approve application: Intraocular pressure 
levels. 

Drug name: Natrecor; 
NDA number: 20920; 
Approval date: Aug. 10, 2001; 
Approved indication: Treatment of acute decompensated congestive heart 
failure in specific populations; 
Surrogate endpoint(s) used to approve application: Pulmonary capillary 
wedge pressure levels. 

Drug name: Zometa; 
NDA number: 21223; 
Approval date: Aug. 20, 2001; 
Approved indication: Treatment of hypercalcimia of malignancy; 
Surrogate endpoint(s) used to approve application: Complete response 
(lowering of the corrected serum calcium)[B]. 

Drug name: Benicar; 
NDA number: 21286; 
Approval date: April 25, 2002; 
Approved indication: Treatment of hypertension; 
Surrogate endpoint(s) used to approve application: Peak and trough 
blood pressure levels. 

Drug name: Faslodex; 
NDA number: 21344; 
Approval date: April 25, 2002; 
Approved indication: Treatment of metastatic breast cancer; 
Surrogate endpoint(s) used to approve application: Response rate and 
time to progression[B]. 

Drug name: Hepsera; 
NDA number: 21449; 
Approval date: Sept. 20, 2002; 
Approved indication: Treatment of chronic hepatitis B; 
Surrogate endpoint(s) used to approve application: Knodell 
Necroinflammatory Score (i.e. liver biopsy). 

Drug name: Inspra; 
NDA number: 21437; 
Approval date: Sept. 27, 2002; 
Approved indication: Treatment of hypertension; 
Surrogate endpoint(s) used to approve application: Sitting diastolic 
and systolic blood pressure at trough. 

Drug name: Zetia; 
NDA number: 21445; 
Approval date: Oct. 25, 2002; 
Approved indication: Treatment of elevated total cholesterol levels and 
LDL-C; 
Surrogate endpoint(s) used to approve application: Cholesterol levels. 

Drug name: Extraneal; 
NDA number: 21321; 
Approval date: Dec. 20, 2002; 
Approved indication: Single daily exchange for the long dwell during 
continuous ambulatory peritoneal dialysis or automated peritoneal 
dialysis for the treatment of chronic renal failure; 
Surrogate endpoint(s) used to approve application: Ultrafiltration 
rate. 

Drug name: Somavert; 
NDA number: 21106; 
Approval date: Mar. 25, 2003; 
Approved indication: Treatment of acromegaly in specific patients; 
Surrogate endpoint(s) used to approve application: Serum IGF-I levels. 

Drug name: Reyataz; 
NDA number: 21567; 
Approval date: June 20, 2003; 
Approved indication: Treatment of Human Immunodeficiency Virus-1 
infection; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cell) and viral load (HIV-RNA). 

Drug name: Emtriva; 
NDA number: 21500; 
Approval date: July 2, 2003; 
Approved indication: Treatment of Human Immunodeficiency Virus-1 
infection in adults; 
Surrogate endpoint(s) used to approve application: CD4 count (infection 
fighting white blood cell) and viral load (HIV-RNA). 

Drug name: Zavesca; 
NDA number: 21348; 
Approval date: July 31, 2003; 
Approved indication: Treatment of mild to moderate type 1 Gaucher 
disease; 
Surrogate endpoint(s) used to approve application: Liver and spleen 
volume after 12 months of treatment. 

Drug name: Crestor; 
NDA number: 21366; 
Approval date: Aug. 12, 2003; 
Approved indication: Treatment of cholesterol levels; 
Surrogate endpoint(s) used to approve application: Cholesterol levels. 

Drug name: Radiogardase; 
NDA number: 21626; 
Approval date: Oct. 2, 2003; 
Approved indication: Treatment of internal contamination with 
radioactive cesium and/or radioactive thallium; 
Surrogate endpoint(s) used to approve application: Whole body effective 
half life of cesium or thallium. 

Drug name: Plenaxis; 
NDA number: 21320; 
Approval date: Nov. 25, 2003; 
Approved indication: Palliative treatment for men with advanced 
symptomatic prostate cancer for specific reasons; 
Surrogate endpoint(s) used to approve application: Avoid orchiectomy 
and lower serum testosterone levels. 

Drug name: Spiriva Handihaler; 
NDA number: 21395; 
Approval date: Jan. 30, 2004; 
Approved indication: Treatment of bronchospasm associated with chronic 
obstructive pulmonary disease; 
Surrogate endpoint(s) used to approve application: Forced expiratory 
volume in one second, with peak effect occurring within 3 hours 
following the first dose. 

Drug name: Sensipar; 
NDA number: 21688; 
Approval date: Mar. 8, 2004; 
Approved indication: Treatment of secondary hyperparathyroidism in 
patients with chronic kidney disease on dialysis; 
Surrogate endpoint(s) used to approve application: Intact parathyroid 
hormone levels. 

Drug name: Apidra; 
NDA number: 21629; 
Approval date: April 16, 2004; 
Approved indication: Treatment of diabetes mellitus in adult patients; 
Surrogate endpoint(s) used to approve application: Blood sugar 
(glycated hemoglobin and HbA1c equivalents). 

Drug name: Vidaza; 
NDA number: 50794; 
Approval date: May 19, 2004; 
Approved indication: Treatment of myelodysplastic syndrome subtypes: 
refractory anemia or refractory anemia with ringed sideroblasts, 
refractory anemia with excess blasts, refractory anemia with excess 
blasts in transformation, and chronic myelomonocytic leukemia; 
Surrogate endpoint(s) used to approve application: Response rate[B]. 

Drug name: Pentetate Calcium Trisodium; 
NDA number: 21749; 
Approval date: Aug. 11, 2004;
Approved indication: Treatment of known or suspected internal 
contamination with plutonium, americium, or curium to increase rates of 
elimination; 
Surrogate endpoint(s) used to approve application: Ratio of urine 
radioactivity before treatment to the maximum urine radioactivity after 
treatment (excretion enhancement factor). 

Drug name: Pentetate Zinc Trisodium; 
NDA number: 21751; 
Approval date: Aug.11, 2004; 
Approved indication: Treatment of known or suspected internal 
contamination with plutonium, americium, or curium to increase rates of 
elimination; 
Surrogate endpoint(s) used to approve application: Ratio of urine 
radioactivity before treatment to the maximum urine radioactivity after 
treatment (excretion enhancement factor). 

Drug name: Fosrenol; 
NDA number: 21468; 
Approval date: Oct. 26, 2004; 
Approved indication: Treatment of end stage renal disease; 
Surrogate endpoint(s) used to approve application: Serum phosphate 
levels. 

Drug name: Omacor[C]; 
NDA number: 21654; 
Approval date: Nov. 10, 2004; 
Approved indication: Treatment of very high triglyceride levels in 
adults; 
Surrogate endpoint(s) used to approve application: Triglyceride levels. 

Drug name: Symlin; 
NDA number: 21332; 
Approval date: Mar. 16, 2005; 
Approved indication: Treatment of type 1 or type 2 diabetes mellitus; 
Surrogate endpoint(s) used to approve application: Blood sugar (HbA1c 
levels). 

Drug name: Mycamine; 
NDA number: 21506; 
Approval date: Mar. 16, 2005; 
Approved indication: Treatment of esophageal candidiasis or prophalaxis 
against fungal infection; 
Surrogate endpoint(s) used to approve application: Endoscopic 
appearance of the esopageal mucosa. 

Drug name: Baraclude; 
NDA number: 21797; 
Approval date: Mar. 29, 2005; 
Approved indication: Treatment of chronic hepatitis B in adults; 
Surrogate endpoint(s) used to approve application: Knodell 
Necroinflammatory Score (i.e. liver biopsy). 

Drug name: Byetta;
NDA number: 21773; 
Approval date: April 28, 2005; 
Approved indication: Treatment of type 2 diabetes mellitus; 
Surrogate endpoint(s) used to approve application: Blood sugar (HbA1c 
levels). 

Drug name: Levemir; 
NDA number: 21536; 
Approval date: June 16, 2005; 
Approved indication: Treatment of diabetes mellitus in adults who 
require basal insulin for control of hyperglycemia; 
Surrogate endpoint(s) used to approve application: Blood sugar (fasting 
blood glucose and HBA1c levels). 

Drug name: Nexavar; 
NDA number: 21923; 
Approval date: Dec. 20, 2005; 
Approved indication: Treatment of advanced renal cell carcinoma; 
Surrogate endpoint(s) used to approve application: Progression free 
survival. 

Drug name: Vaprisol; 
NDA number: 21697; 
Approval date: Dec. 29, 2005; 
Approved indication: Treatment of euyolemic hyponatremia in 
hospitalized patients; 
Surrogate endpoint(s) used to approve application: Serum sodium 
concentration levels. 

Drug name: Sutent; 
NDA number: 21938; 
Approval date: Jan. 26, 2006; 
Approved indication: Treatment of gastrointestinal stromal tumor and 
advanced renal cell carcinoma; 
Surrogate endpoint(s) used to approve application: Time to progression. 

Drug name: Pylera; 
NDA number: 50786; 
Approval date: Sept. 28, 2006; 
Approved indication: Treatment of helicobacter pylori infection and 
duodenal ulcer disease; 
Surrogate endpoint(s) used to approve application: C-urea breath tests. 

Drug name: Januvia;
NDA number: 21995; 
Approval date: Oct. 16, 2006; 
Approved indication: Treatment of type 2 diabetes mellitus; 
Surrogate endpoint(s) used to approve application: Blood sugar (fasting 
blood glucose, HBA1c levels and post prandial glucose). 

Drug name: Tyzeka; 
NDA number: 22011; 
Approval date: Oct. 25, 2006; 
Approved indication: Treatment of chronic hepatitis B in specific 
populations of adults; 
Surrogate endpoint(s) used to approve application: Composite serologic 
endpoint requiring suppression of HBV DNA to a specific amount in 
conjunction with another serum. 

Drug name: Tekturna; 
NDA number: 21985; 
Approval date: Mar. 5, 2007; 
Approved indication: Treatment of hypertension; 
Surrogate endpoint(s) used to approve application: Seated trough cuff 
blood pressure. 

Drug name: Tykerb; 
NDA number: 22059; 
Approval date: Mar. 13, 2007; 
Approved indication: Treatment of advanced metastatic breast cancer; 
Surrogate endpoint(s) used to approve application: Time to progression. 

Drug name: Somatuline Depot; 
NDA number: 22074; 
Approval date: Aug. 30, 2007; 
Approved indication: Treatment of acromegalic patients who have 
inadequate response to surgery and/or radiotherapy; 
Surrogate endpoint(s) used to approve application: Growth hormone and 
insulin growth factor levels. 

Drug name: Ixempra; 
NDA number: 22065; 
Approval date: Oct. 16, 2007; 
Approved indication: Treatment of metastatic or locally advanced breast 
cancer; 
Surrogate endpoint(s) used to approve application: Progression free 
survival. 

Drug name: Kuvan;
NDA number: 22181; 
Approval date: Dec. 13, 2007; Approved indication: Treatment of 
hyperphenylalaninemia; 
Surrogate endpoint(s) used to approve application: Blood phenylalanine 
levels. 

Drug name: Bystolic; NDA number: 21742; 
Approval date: Dec. 17, 2007; 
Approved indication: Treatment of hypertension; 
Surrogate endpoint(s) used to approve application: Trough sitting 
systolic/diastolic blood pressure. 

Drug name: Treanda; 
NDA number: 22249; 
Approval date: Mar. 20, 2008; 
Approved indication: Treatment of chronic lymphocytic leukemia; 
Surrogate endpoint(s) used to approve application: Objective response 
and progression-free survival[B]. 

Drug name: Durezol; 
NDA number: 22212; 
Approval date: June 23, 2008; 
Approved indication: Treatment of inflammation and pain in the eye 
associated with ocular surgery; 
Surrogate endpoint(s) used to approve application: Complete clearing 
(count = 0). 

Source: GAO analysis of FDA data. 

[A] Also approved for treatment of giant papillary conjunctivitis and 
seasonal allergic conjunctivitis. However, these indications were 
approved based on clinical endpoints and are not included in the table. 

[B] Response rate is a measure of the proportion of patients with tumor 
size reduction of a predefined amount and for a minimum time period. 

[C] In July 2007, the name of this drug was changed from Omacor to 
Lovaza. 

[End of table] 

[End of section] 

Appendix III: Applications Selected and Questions Regarding FDA's 
Oversight of Required Postmarketing Studies: 

To review specific instances of the Food and Drug Administration's 
(FDA) monitoring and enforcement activities, we selected a judgmental 
sample of 15 applications approved based on surrogate endpoints under 
the accelerated program and the 35 postmarketing studies that FDA 
required drug sponsors to complete for these drugs. These applications 
were selected to generate a sample that included a variety of drugs and 
a range of studies at various stages of completion. We then provided 
FDA with a standard series of questions for each of the 35 studies, and 
requested that FDA's medical reviewers, who are responsible for 
monitoring these 35 studies, provide specific information on them, 
including a description of the studies, FDA's efforts to monitor these 
studies, and applicable enforcement actions taken, if any, to prompt 
sponsors' compliance. FDA officials indicated that several individuals 
were involved in completing the questions provided for each of the 15 
applications. FDA staff completed the first half of the questions 
related to description of the studies. The medical reviewers or other 
review staff provided information on the agency's monitoring efforts 
and enforcement actions. The applications we selected and the 
information request are provided below. 

Table 7: List of 15 Accelerated Approval Applications Selected for 
Review: 

Drug name: ProAmatine; 
NDA/BLA number: 19815; 
Approval date: Sept. 6, 1996. 

Drug name: Xeloda; 
NDA/BLA number: 20896; 
Approval date: April 30, 1998. 

Drug name: Sulfamylon; 
NDA/BLA number: 19832; 
Approval date: June 5, 1998. 

Drug name: Priftin; 
NDA/BLA number: 21024; 
Approval date: June 22, 1998. 

Drug name: Remicade; 
NDA/BLA number: 103772; 
Approval date: Aug. 24, 1998. 

Drug name: Viramune; 
NDA/BLA number: 20933; 
Approval date: Sept. 11, 1998. 

Drug name: Sustiva; 
NDA/BLA number: 20972; 
Approval date: Sept. 17, 1998. 

Drug name: Ziagen; 
NDA/BLA number: 20977; 
Approval date: Dec. 17, 1998. 

Drug name: Celebrex; 
NDA/BLA number: 21156; 
Approval date: Dec. 23, 1999. 

Drug name: Gleevec; 
NDA/BLA number: 21588; 
Approval date: April 18, 2003. 

Drug name: Zevalin; 
NDA/BLA number: 125019; 
Approval date: Feb. 19, 2002. 

Drug name: Fuzeon; 
NDA/BLA number: 21481; 
Approval date: Mar. 13, 2003. 

Drug name: Fabrazyme; 
NDA/BLA number: 103979; 
Approval date: April 24, 2003. 

Drug name: Iressa; 
NDA/BLA number: 21399; 
Approval date: May 5, 2003. 

Drug name: Velcade; 
NDA/BLA number: 21602; 
Approval date: May 13, 2003. 

Source: GAO analysis of FDA data. 

[End of table] 

Information Request on the Monitoring and Oversight of Drugs Approved 
Based on Surrogate Endpoints: 

Instructions: Please review the pre-filled information and provide 
information on your monitoring of the drug sponsor's completion of 
postmarketing studies required under the accelerated approval program. 
When you have completed this request, please return it to GAO via email 
at Lichtenfeldd@GAO.gov. Please contact David Lichtenfeld at (312) 220-
7663 if you have any questions. 

NDA Number: 
Approval Date: 
Drug Name: 
Division: 
Number of required studies: 
Commitment IDs and Status: 

Source: FDA-provided data on required postmarketing commitments (as of 
December 19, 2008). 

For study # please provide answers to the following questions: 

1. What is the study description as outlined in the approval letter or 
annual status report (ASR)? 

2. Was there ever an original study schedule or timeline established 
for this study? If yes, when was it established and how was it 
documented (e.g. in approval letter, first ASR etc.)? 

3. If a study schedule was not established, how have you determined the 
current status of the study? 

4. When was this study assigned its current status? 

5. Has/did the drug sponsor submit all required ASRs for this study, 
and were they submitted within required timeframes? 

6. If the sponsor did not submit all required ASRs in a timely manner, 
what action(s) did FDA take to obtain the sponsor's compliance (e.g., 
initiate a teleconference, issue administrative letter)? 

Please provide any available documentation demonstrating FDA's effort 
to follow-up with the drug sponsor regarding ASR submissions for this 
study. 

7. Has FDA reviewed all of the ASRs submitted for this study, and have 
they been reviewed within FDA review timeframes—within 90 days of 
receipt? 

8. If FDA has not reviewed all of the ASRs according to its policy, 
what barriers have prevented FDA from conducting these reviews? In 
addition, when will FDA complete its reviews of any outstanding ASRs 
for this study? 

9. Based on its review of ASRs submitted for this study, has FDA 
identified any problems or concerns with the drug sponsor's progress in 
completing this study? For example, did FDA's reviews reveal that the 
study had been either pending or delayed for an extended period of 
time, or did its reviews provide indications that the sponsor would not 
be able to complete the study within established timeframes? If so, 
what actions did FDA take to compel the sponsor to complete the study?
Please provide any available documentation demonstrating FDA's efforts 
to follow-up with the sponsor regarding progress for this study. 

10. Based on its review of ASRs submitted for this study, has FDA 
identified any significant underlying issues affecting the sponsor's 
ability to complete this study? For example, did the sponsor experience 
any problems designing the study or enrolling patients that affected 
the progress of the study? If so, what actions did FDA take to help the 
sponsor improve the progress of the study? 

Please provide any available documentation demonstrating FDA's efforts 
to follow-up with the sponsor regarding this study's completion. 

11. Outside of ASR reviews, has FDA used any other mechanisms to 
monitor the progress of this study (e.g. routine conference calls or 
other routine communication with the sponsor)? 

12. If the status of this study is fulfilled, what factors contributed 
to the completion of the study? 

13. Please provide any additional comments you believe are relevant to 
the oversight and monitoring of this study. 

Name of FDA Reviewer Who Answered these Questions: 

How long has the Reviewer been responsible for monitoring this study: 
(Please check the appropriate box.) 

1-2 years: 
3-4 years: 
5-6 years: 
7-8 years: 
9-10 years: 
over 10 years: 

[End of section] 

Appendix IV: Sales for Selected Drugs Approved Based on Surrogate 
Endpoints under the Accelerated Approval Process: 

Listed in table 8 are seven applications for drugs approved based on 
surrogate endpoints under the accelerated approval process, and total 
U.S. sales, since approval, associated with those drugs. This listing 
includes applications, which as of December 19, 2008, had not been 
converted to full approval, and more than 5 years had elapsed since 
they were initially approved. 

Table 8: Total U.S. Sales for Selected Drugs Approved under the 
Accelerated Process: 

Application type: NDA; 
Application number: 19815; 
Drug name: ProAmatine; 
Approval date: Sept. 6, 1996; 
Total U.S. sales since approval[A]: $257,574,554. 

Application type: NDA; 
Application number: 19832; 
Drug name: Sulfamylon; 
Approval date: June 5, 1998; 
Total U.S. sales since approval[A]: $72,963,020[B]. 

Application type: NDA; 
Application number: 21024[C]; 
Drug name: Priftin; 
Approval date: June 22, 1998; 
Total U.S. sales since approval[A]: $177,502. 

Application type: NDA; 
Application number: 50747; 
Drug name: Synercid; 
Approval date: Sept. 21, 1999; 
Total U.S. sales since approval[A]: $206,741,816. 

Application type: NDA; 
Application number: 21174; 
Drug name: Mylotarg; 
Approval date: May 17, 2000; 
Total U.S. sales since approval[A]: $206,982,392. 

Application type: NDA; 
Application number: 21399; 
Drug name: Iressa; 
Approval date: May 5, 2003; 
Total U.S. sales since approval[A]: $416,699,000. 

Application type: BLA; 
Application number: 103979; 
Drug name: Fabrazyme; 
Approval date: April 24, 2003; 
Total U.S. sales since approval[A]: $56,308,877. 

Source: GAO analysis of FDA data and National Sales Perspectives (™) 
IMS Health, Inc. 

[A] Sales data through December 2008. 

[B] Annual sales since 1998 for Sulfamylon include sales for this 
approval and an earlier approval, which occurred prior to the 
implementation of the accelerated approval process. 

[C] The Food and Drug Administration converted the application for 
Priftin to full approval on June 1, 2009. 

[End of table] 

[End of section] 

Appendix V: Comments from the Department of Health and Human Services: 

Department Of Health & Human Services: 
Office Of The Secretary: 
Assistant Secretary for Legislation: 
Washington, DC 20201: 

September 8, 2009: 

Marcia Crosse: 
Director, Health Care: 
U.S. Government Accountability Office: 
441 G Street N.W. 
Washington, DC 20548: 

Dear Ms. Crosse: 

Enclosed are comments on the U.S. Government Accountability Office's 
(GAO) report entitled: New Drug Approval: FDA Needs to Enhance its 
Oversight of Drugs Approved on the Basis of Surrogate Endpoints (GAO-09-
866). 

The Department appreciates the opportunity to review this report before 
its publication. 

Sincerely, 

Signed by: 

Andrea Palm: 
Acting Assistant Secretary for Legislation: 

Enclosure: 

[End of letter] 

FDA's General Comments to the United States Government Accountability 
Office's Draft Report Entitled, New Drug Approval: FDA Needs to Enhance 
Its Oversight of Drugs Approved on the Basis of Surrogate Endpoints 
(GAO-09-866): 

The Food and Drug Administration (FDA) welcomes the opportunity to 
comment on the Government Accountability Office's (GAO) findings in the 
draft report and respond to the single recommendation made by GAO to 
FDA. 

FDA believes that the accelerated approval program has been very 
successful. It has resulted in the early approval of many significant 
therapeutic advances for patients with serious or life-threatening 
illnesses who desperately need access to new treatments, in some cases 
years before the drug would have been available under regular approval
procedures. For patients with HIV/AIDS and cancer, especially, the 
accelerated approval program has dramatically expanded the therapeutic 
armamentarium and significantly improved their quality of life and 
survival. 

By focusing primarily on the limited number of cases where the 
postmarketing trials to confirm clinical benefit either have taken 
longer to complete than expected or have not demonstrated clinical 
benefit in the manner that had been anticipated, GAO's report minimizes 
the fact that, as reflected in GAO's own analysis, overall the 
accelerated approval program functions precisely as intended and 
designed. First, through the accelerated approval program, millions of 
patients with serious or life-threatening illnesses have had earlier 
access to new safe and effective treatments. Second, in the vast 
majority of cases, the confirmatory trials have been or are being 
completed in a timely manner, have confirmed clinical benefit, and have 
led to conversion to regular approval. These findings are important for 
patients and also provide validation of the underlying public health 
principles of the accelerated approval program. 

FDA developed the accelerated approval program with full recognition of 
the risk that drugs might be approved and later found not to confer 
clinical benefit to patients. After much input from stakeholders, FDA 
determined that this was a risk worth taking because patients with 
serious or life-threatening illnesses and their physicians are willing 
to accept more risk when making treatment decisions. FDA also sought to 
minimize the risk that patients would be exposed to approved treatments 
that were not safe and effective by incorporating four important 
safeguards into the program. First, the evidentiary standard for 
accelerated approval and regular approval is the same, meaning that 
there must be substantial evidence that the drug has the purported 
effect on the surrogate endpoint and there must be adequate safety data 
for FDA to conclude that the benefits of the drug outweigh the risks. 
Second, the surrogate endpoint must be considered reasonably likely to 
predict clinical benefit. Third, clinical trials to confirm clinical 
benefit are required after approval and must be completed with due 
diligence by the sponsor. Finally, FDA can undertake expedited 
withdrawal procedures if clinical benefit is not confirmed. 

The facts as outlined in GAO's report show that the surrogate endpoints 
used as the basis for accelerated approval have, in most cases, served 
as accurate predictors of clinical benefit. This is demonstrated by the 
fact that two thirds of the required confirmatory studies are now 
considered closed by FDA and over half of the drugs approved under 
accelerated approval have been converted to regular approval. FDA's 
oversight of this program, including careful selection of surrogates 
and extensive efforts to work closely with sponsors to design and 
ensure timely completion of confirmatory trials, has served to make it 
a success. 

FDA has previously acknowledged that its oversight of postmarketing 
requirements (PMR) and commitments (PMC) in general has been 
inadequate; however oversight of postmarketing trials of drugs approved 
under accelerated approval has always been an important FDA priority. 
As noted above, nearly two thirds of the required confirmatory trials 
are now considered complete by FDA. This is clear evidence that both 
FDA and sponsors have taken their postmarketing obligations under this 
program seriously. 

With regard to oversight of the much larger overall portfolio of PMRs 
and PMCs, FDA agrees that improvements are needed. In the past, because 
of inadequate staffing and information technology resources and 
competing workload priorities, the Center for Drug Evaluation and 
Research (CDER) has been unable to thoroughly track PMRs and PMCs or to 
review annual status reports and final study reports in a timely 
fashion. With increased resources from Congressional appropriations and 
statutory requirements, CDER has begun to implement a number of process 
improvements to ensure appropriate oversight, more efficient tracking, 
and expeditious review of postmarketing submissions. These improvements 
were not complete at the time of this GAO investigation and are not 
fully reflected by the data included in the report. These efforts 
include the following: 

* Publication of a new PMR/PMC Development Manual of Policies and 
Procedures (MAPP) and significant revision of the PMR/PMC Tracking 
MAPP; 

* Establishment of the new roles of PMR/PMC Development and Tracking 
Coordinators within each new drug review division; 

* Implementation of new PMR/PMC functions in the document archiving and 
records retention system database (DARRTS), which will improve PMR/PMC 
tracking by enhancing FDA's ability to capture data and generate 
reports; and; 

* Contracting with Booz Allen Hamilton (BAH) to assist FDA staff with 
the review of annual status reports and to conduct an independent 
review of the backlog of all open PMRs and PMCs. 

BAH's audit has led to extensive updates to the data in FDA's tracking 
database and revealed that most of the industry postmarketing studies 
and trials are proceeding according to the agreed-upon timeline. FDA 
plans to continue its contract with BAH to ensure that the new 
processes are implemented and the data in the database are accurate
and up-to-date going forward. FDA also plans to make the BAH audit 
report available to the public. 

Response to GAO's Recommendation: 

GAO recommended that to clarify FDA's enforcement authority under the 
accelerated approval process, the Commissioner of FDA take the 
following action: Clarify the conditions under which the agency would 
utilize its authority to expedite the withdrawal of drugs approved 
based on surrogate endpoints under the accelerated approval process if 
sponsors fail to complete required confirmatory studies with due 
diligence, or if studies are completed, but fail to demonstrate the 
clinical effectiveness of the drugs. 

FDA acknowledges that there have been cases where confirmation of the 
clinical benefit of drugs approved under accelerated approval did not 
occur in a timely manner. FDA considers the lack of progress or 
completion of confirmatory trials seriously and works closely with 
sponsors to get the required trials back on track for completion. In 
some cases it has been difficult to design or successfully enroll 
patients in the confirmatory trials and FDA and sponsors have worked 
together to find acceptable solutions, including alternate trial 
designs or expanding enrollment to international sites. FDA also has
conducted public reviews of outstanding confirmatory trials for 
oncology drugs to focus attention on the delays and to seek advice from 
its advisory committee experts on ways to get the necessary trials 
completed. 

When trials are not being completed in a timely manner or completed 
trials do not appear to confirm clinical benefit, FDA must carefully 
assess each case and consider the underlying reasons and the 
consequences of all regulatory options, including their potential 
impact on patients. The fact that a trial has not been completed in 
what appears to be a timely manner can reflect the failure of a sponsor 
to exercise due diligence in the conduct of the trial, but it can also 
reflect, for example, difficulty in identifying and enrolling 
appropriate patients. Failure to confirm clinical benefit in a 
completed trial may reflect the possibility that the drug does not in 
fact confer clinical benefit, but it also may reflect, for example, 
unforeseen limitations in trial design, rather than clear evidence of
lack of effectiveness. The most appropriate regulatory approach must be 
governed by the unique factors of the particular case. 

By definition, drugs approved under accelerated approval represent 
significant therapeutic advances for patients with serious and life-
threatening illnesses. FDA must carefully evaluate what other options 
are available to patients at the time it is considering regulatory 
action for failure to confirm clinical benefit. In some cases a drug 
for which clinical benefit has not been confirmed may be the only 
approved therapeutic option for patients with the disease. Removing the 
drug from the market and leaving patients with no treatment may be 
unacceptable. In such a case FDA must consider the benefits of 
continued availability of the drug, which by definition under the 
accelerated approval program was shown to have an effect on the 
surrogate endpoint that was the basis for approval, versus the risk 
that patients may actually be using an ineffective drug and exposing 
themselves only to its risks. FDA must also consider the possibility 
that, despite results from confirmatory studies that may appear to 
indicate that a drug does not provide clinical benefit, there may be a 
subset of patients for whom the drug may nevertheless be effective. 
Further, in addition to withdrawal of approval, FDA has other 
regulatory tools that can be considered and applied as appropriate. 
These include requiring a Risk Evaluation and Mitigation Strategy 
(REMS) under the new authority granted to FDA in the Food and Drug 
Administration Amendments Act of 2009 (FDAAA) or limiting access to the 
drug under an Investigational New Drug application (IND). 

Two specific drugs were mentioned in the report and warrant comment: 

ProAmatine (midodrine hydrochloride) was approved by FDA as a treatment 
for patients with orthostatic hypotension, a condition in which a 
patient's blood pressure decreases upon standing, which can lead to 
syncope (fainting) and falls. Orthostatic hypotension is a serious 
disabling condition and there are no other FDA approved drugs for this 
condition. The surrogate endpoint used for approval was an increase in 
systolic blood pressure on standing. Although FDA routinely uses blood 
pressure as a validated surrogate endpoint for the treatment of 
hypertension, in the case of ProAmatine, FDA faced the opposite 
situation, i.e., a drug intended to increase blood pressure in patients
with hypotension. Here FDA was not certain that the observed increase 
in systolic blood pressure would lead to clinical benefit, and the 
sponsor was required to conduct postapproval clinical trials to assess 
relevant clinical endpoints (e.g., decrease in symptoms). 

The sponsor conducted postapproval trials as required, and while FDA 
did not consider the trials adequate to confirm clinical benefit, there 
were trends observed that suggested the drug may in fact have clinical 
benefit. Since this drug is the only approved therapy for this 
condition, FDA has not concluded that it would be appropriate to 
initiate expedited withdrawal of the approval. We note that the 
ProAmatine case is complicated by the fact that there are approved 
generic versions of the product. FDA has worked through complex legal 
and regulatory issues to find ways to encourage one or more of the 
generic sponsors to conduct the necessary clinical trials. FDA is 
continuing to evaluate options in this case. 

Rather than an example of FDA failing to exercise its authority to 
withdraw approval, ProAmatine is a good example of the complex issues 
FDA must consider when clinical benefit has not been confirmed and the 
drug approved under accelerated approval remains the only FDA-approved 
treatment for a serious or life-threatening condition. FDA must 
carefully balance exercising its regulatory authority versus 
considering the best interests of patients with the disease or 
condition. 
 
The second drug mentioned in the GAO report is Iressa (gefitinib). 
Iressa was approved by FDA as a treatment for patients with advanced 
non-small cell lung cancer, a serious and life-threatening form of 
cancer for which there were few treatment options at the time Iressa 
was approved. The sponsor conducted a large, randomized, controlled 
clinical trial in over 1700 patients with less advanced lung cancer and 
the trial failed to show an improvement in survival time for patients 
treated with Iressa; i.e., clinical benefit was not confirmed. By the 
time the postmarketing trial results for Iressa were available; another 
similar drug (Tarceva) had been approved and had demonstrated an 
increase in survival time in patients with non-small cell lung cancer. 

In considering this case, FDA noted that while the controlled clinical 
trial failed to show a survival benefit for the overall study 
population, there was clear evidence in individual patients of 
significant clinical benefit (e.g., shrinkage of large tumors and 
prolonged survival in patients with end-stage disease) that could not 
be ascribed to factors other than drug effect. There was also a 
suggestion, though not yet proven by controlled clinical trials, that 
certain patients might be responsive to Iressa due to the genetic 
markers on their tumor cells while other patients might not respond. 
This was thought to possibly explain the dramatic individual responses 
seen in some patients and the lack of response seen in others. This 
could also explain the failure to see a clinical benefit in a mixed 
population of patients, many of whom might not be responsive to Iressa 
due to the genetic makeup of their tumors. 

After considering all the available information, FDA believed that it 
was appropriate to direct patients beginning treatment for non-small 
cell lung cancer to treatment with Tarceva, because clinical benefit in 
the form of a survival advantage had been confirmed. FDA also believed 
that patients who were already being treated with Iressa and in whom a 
clinical response was observed might appropriately choose, after 
consultation with their doctors, to remain on Iressa. Rather than 
withdraw approval of Iressa, FDA worked with the sponsor to restrict 
access to those patients who were already receiving treatment and whose 
physician felt were deriving clinical benefit from the drug. Since 
marketing of Iressa was restricted, patients with other diseases whose 
physicians might have prescribed the drug for those different 
indications were accommodated through access under an Investigational 
New Drug Application. 

FDA believes that this combination of regulatory actions was an 
appropriate and balanced approach given the unique circumstances in 
this case. The sponsor of Iressa has continued to investigate the 
factors that might predict response to the drug, and it is possible 
that future clinical trials in properly selected patients will 
demonstrate clinical benefit. 

In both of these examples, the sponsors pursued confirmatory clinical 
trials with due diligence, but despite careful consideration and review 
by both industry and the FDA, the trials did not demonstrate clinical 
benefit. In both cases, review of the data suggests limitations in 
elements of trial design that may have led to the seemingly negative 
results. 

GAO recommends that FDA "clarify the conditions under which the agency 
would utilize its authority to expedite the withdrawal of drugs 
approved based on surrogate endpoints under the accelerated approval 
process if sponsors either fail to complete required confirmatory 
trials with due diligence, or if studies are completed, but fail to 
demonstrate the clinical effectiveness of the drugs." Outside of a 
situation where a confirmatory trial clearly demonstrates harm to the 
patients (e.g., decreased survival for patients with cancer treated 
with the accelerated approval drug), FDA believes that each case must 
considered on its merits and that the criteria in the existing 
regulations and statutory provisions (including the provisions in Title 
IX of FDAAA that authorize civil, criminal, and civil monetary 
penalties for failure to conduct a postmarketing confirmatory trial) 
provide FDA with sufficient authority and flexibility to make balanced 
decisions that protect the program from abuse by sponsors and ensure 
that patients will continue to have access to needed treatments. In 
light of the complexities outlined above and the need for a case-by-
case assessment, FDA believes it would be difficult, if not impossible, 
to provide further clarification as to when it might utilize its 
authority to expedite withdrawal of drug approved on the basis of 
surrogate endpoints. FDA remains committed to closely overseeing 
required postmarketing trials for drugs approved under accelerated 
approval and believes that the improvements being implemented in its 
overall PMR/PMC program will allow for continued success of the 
program. 

[End of section] 

Appendix VII: GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Marcia Crosse, (202) 512-7114 or crossem@gao.gov: 

Acknowledgments: 

In addition to the contact named above, Geri Redican-Bigott, Assistant 
Director; Shaunessye Curry; Krister Friday; Mollie Hertel; Julian 
Klazkin; David Lichtenfeld; Rich Lipinski; and Thanh Lu made key 
contributions to this report. 

[End of section] 

Footnotes: 

[1] Biological products are products derived from living sources--such 
as humans, animals, and microorganisms--that are intended for 
preventing, treating, or curing diseases or conditions. They include 
vaccines, blood products, and proteins. See 42 U.S.C. § 262(i), 21 
C.F.R. § 600.3(h)(2008). For the remainder of this report we use the 
term "drug" to refer to both therapeutic biological products and 
chemically synthesized drugs. 

[2] Drug sponsors typically are the applicants who submit new drug 
applications (NDAs) and biological license applications (BLAs) to FDA 
for review. A drug sponsor may assume responsibility for the marketing 
of a new drug, including responsibility for complying with applicable 
laws and regulations. 

[3] FDA's approval of an NDA or BLA means a sponsor can market the new 
drug. Throughout the remainder of this report we refer to FDA's 
approval of NDAs and BLAs as approval of "drugs". 

[4] 21 C.F.R. pt. 314 subpt. H (2008), 21 C.F.R. pt. 601 subpt. E 
(2008). 

[5] Throughout this report we refer to those postmarketing studies 
which FDA requested and sponsors committed, in writing, to conduct as 
"requested" postmarketing studies. 

[6] See Pub. L. No. 105-115, § 130, 111 Stat. 2296, 2331-2 (codified at 
21 U.S.C. § 356b). 

[7] 21 C.F.R. §§ 314.81(b)(2)(vii), 601.70(a), (b)(2008). The 
regulations also explain that status reports of postmarketing studies 
are required for studies that address (1) clinical safety, (2) clinical 
efficacy, (3) clinical pharmacology, and (4) nonclinical toxicology. 
Drug sponsors, but not those producing biologics, must also submit 
annual reports on postmarketing studies that they have agreed to 
conduct, or that are conducted on their behalf that concern chemistry, 
manufacturing, controls, or product stability. 21 C.F.R. § 
314.81(b)(2)(viii)(2008). 

[8] FDA issued final regulations for the accelerated approval process 
on December 11, 1992, but approved one application under this process, 
prior to issuing the final regulations. We have included this 
application in the scope of our review. 

[9] For example, if a drug can be taken either as a pill or via 
injection, FDA may approve a separate application for each route of 
administration. 

[10] The scope of applications included in this review was limited to 
NDAs for NMEs. This review does not include any new BLAs. 

[11] FDA approved 15 NMEs for drugs used to aid in diagnosing diseases 
or in aiding the absorption of other drugs. Because these 15 were not 
used to actually treat a disease, we excluded them from our analyses. 

[12] We obtained this material from FDA's Web site, [hyperlink, 
www.accessdata.fda.gov/Scripts/cder/DrugsatFDA/]. 

[13] For example, under the standards for internal control, information 
should be recorded and communicated to management and others within an 
entity who need it and within a time frame that enables them to carry 
out their internal control and other responsibilities. See GAO, 
Standards for Internal Control in the Federal Government, [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: Nov. 
1999) and its supplemental guide, Internal Control Management and 
Evaluation Tool, [hyperlink, http://www.gao.gov/products/GAO-01-1008G] 
(Washington, D.C.: Aug. 2001). 

[14] FDA will not approve an application if, for example, the methods 
used in, and the facilities and controls used for, the manufacture, 
processing, and packing of such drug are inadequate to preserve its 
identity, strength, quality, and purity. 21 U.S.C. § 355(d)(3). 

[15] FDA, Innovation or Stagnation: Challenge and Opportunity on the 
Critical Path to New Medical Products (March 2004). 

[16] R. Temple, Are Surrogate Markers Adequate to Assess Cardiovascular 
Disease Drugs? The Journal of the American Medical Association, August 
25, 1999; 282(8):790-795. 

[17] For example, see D.S. Echt, P.R. Liebson, L.B. Mitchell, et al, 
Mortality and Morbidity in Patients Receiving encainide, flecainide, or 
placebo. The Cardiac Arrhythmia Suppression Trial, New England Journal 
of Medicine, 1991; 324:781-788. 

[18] According to FDA officials, prior to 2001, they did not always 
establish time frames for completing postmarketing studies. 

[19] In addition to confirmatory studies, sponsors may be required to 
conduct postmarketing studies in other instances. For example, under 
the Pediatric Research Equity Act of 2003 sponsors may be required to 
study products in children. When a sponsor is required under the act to 
study its product in pediatric populations, FDA can defer the required 
studies until after the product is approved in adults. Pub. L. No. 108- 
55, § 2(a), 117 Stat. 1936 (codified at 21 U.S.C. § 
355c(a)(3)(A)(i)(I)). Further, if FDA approves a drug based solely on 
animal studies, when human efficacy studies are not ethical or 
feasible, it requires the sponsor to subsequently conduct studies to 
verify and describe the drug's efficacy and to assess its safety in 
humans when such studies become feasible and ethical. C.F.R. pts. 314 
subpt. I; 601 subpt. H (2008). Finally, the Food and Drug 
Administration Amendments Act of 2007 (FDAAA) provided FDA with 
additional authority to require postmarket safety studies in certain 
instances. Pub. L. No. 110-85, § 901(a), 121 Stat. 823, 922 (codified 
at 21 U.S.C. § 355(o)). 

[20] 21 C.F.R. §§ 314.81(b)(2), (b)(2)(vii), 601.70(a)-(c)(2008). The 
current reporting requirements became effective on April 30, 2001. In 
2001, drug sponsors were required to submit their first progress report 
by October 30 for those postmarketing studies that addressed clinical 
efficacy, clinical safety, clinical pharmacology, or nonclinical 
toxicology. 

[21] 21 C.F.R. §§ 314.81(b)(2)(vii)(a)(6), (7); 601.70(b)(6), 
(7)(2008). 

[22] 21 U.S.C. 356b(c),(d). 

[23] See 21 C.F.R. §§ 314.81(b)(2)(vii)(a)(8); 601.70(b)(8)(2008). 

[24] According to FDA, conversion from accelerated to full approval 
means the sponsor has demonstrated, through clinical testing, that the 
drug is clinically effective in treating a specific disease or medical 
condition. 

[25] In 2007, FDAAA provided FDA with authority to assess civil 
monetary penalties against those who have not conducted required 
postmarketing studies under the accelerated approval process. Pub. L. 
No. 110-85, §§ 901(a), 902(b) (codified at 21 U.S.C. §§ 333(f)(4), 
355(p)). As of July 2009, FDA officials indicated they have never used 
this new authority. 

[26] 21 C.F.R §§ 314.530(a)(2), 601.43(a)(2)(2008). 

[27] In several instances FDA approved multiple applications for the 
same drug through the accelerated approval process. Thus FDA approved a 
smaller number of drugs than applications. For example, in some cases a 
drug had several routes of administration (e.g., tablet, intravenous 
solution, oral suspension), and FDA approved a separate application for 
each. 

[28] Two of these open studies are for the drug Levaquin, which was 
approved to treat the effects of inhalation anthrax. According to FDA 
officials, these studies will remain in the pending status 
indefinitely, because the sponsor cannot, for ethical reasons, test the 
medication on humans unless there is an anthrax attack or other 
widespread exposure. 

[29] We determined the time to fulfill a study by measuring the amount 
of time that elapsed from the date the study was required in the 
approval letter to the date FDA determined the study was fulfilled. FDA 
was unable to provide the fulfillment dates for 2 of the 73 fulfilled 
studies; therefore the total number of studies used in figure 4 is 71. 

[30] In a single arm trial there is one treatment group of patients who 
all take the drug being studied; there is not a separate group of 
patients taking another drug or a placebo for comparison. According to 
FDA, single arm trials can provide an accurate assessment of tumor 
response in patients with highly resistant tumors. 

[31] A randomized control trial is a study in which participants are 
randomly assigned to a group taking the drug under study, a placebo, or 
a comparison drug. FDA considers a randomized control trial as the most 
reliable type of trial. 

[32] Status information on requested postmarketing studies under the 
accelerated approval process was provided by FDA as of January 6, 2009. 

[33] We determined the amount of time studies had been open by 
measuring the amount of time which elapsed from the date the study was 
requested to January 6, 2009. 

[34] In addition to these 204 NMEs, FDA approved an additional 15 NMEs 
for drugs used to aid in diagnosing diseases or in aiding the 
absorption of other drugs. Because these 15 were not used to treat a 
disease, we excluded them from our analyses. 

[35] These 69 applications include one for a drug that was approved for 
many indications (diseases). This drug was approved for some of these 
indications primarily based on a surrogate endpoint and some were 
approved based on a clinical endpoint. Therefore, we included this drug 
in our scope. 

[36] FDA and others have suggested that the identification of validated 
surrogate endpoints already used in drug approvals, as well as criteria 
for accepting new potential surrogate endpoints, may encourage more 
efficient drug development. See GAO, New Drug Development: Science, 
Business, Regulatory and Intellectual Property Issues Cited as 
Hampering Drug Development Efforts, GAO-07-49 (Washington, D.C.: Nov. 
17, 2006). FDA planned to develop a comprehensive inventory of all 
surrogate endpoints used to approve new drugs, including those under 
the traditional process. FDA officials told us that they were able to 
compile a partial list of such endpoints, but due to other competing 
priorities, this inventory was never completed. 

[37] We determined the amount of time studies had been open by 
measuring the amount of time which elapsed from the date the study was 
requested to February 13, 2009. 

[38] For more information, see HHS' Office of Inspector General's 
report entitled FDA's Monitoring of Postmarketing Study Commitments 
(June 2006). 

[39] FDA retained Booz Allen Hamilton in 2006 to conduct an independent 
analysis of the agency's postmarket processes and procedures. In 
Postmarketing Commitments Study Final Report, issued in January 2008, 
the contractor highlighted further problems, and made recommendations 
designed to improve FDA's monitoring and oversight of postmarketing 
studies. 

[40] FDA is required to report annually in the Federal Register and to 
congressional committees the status of postmarketing studies that are 
the subject of annual status reports submitted to FDA. 21 U.S.C. §§ 
356b(c), 356b note. 

[41] According to FDA officials, their contractor has already 
identified numerous instances where status information was incorrect, 
such as when studies listed as pending were, in fact, ongoing. Based on 
this information FDA will be updating its database with the correct 
information, and this will be reflected in its next annual report in 
the Federal Register. 

[42] DARRTS became operational for NDAs in July 2009. FDA will continue 
to use its current system to oversee postmarketing studies related to 
BLAs. Data related to BLAs are scheduled to be integrated into DARRTS 
in 2010. 

[43] According to FDA officials, the agency did not mail these letters 
directly to the five generic manufacturers, but contacted certain key 
industry officials regarding FDA's posting of these letters. 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the performance 
and accountability of the federal government for the American people. 
GAO examines the use of public funds; evaluates federal programs and 
policies; and provides analyses, recommendations, and other assistance 
to help Congress make informed oversight, policy, and funding 
decisions. GAO's commitment to good government is reflected in its core 
values of accountability, integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 
[hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 
information. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: