• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

Safety

  • Print
  • Share
  • E-mail

Appendix A

Table of Contents: Managing the Risks From Medical Product Use: Creating a Risk Management Framework

Previous Section : Appendices

Comparison of Postapproval Risks for Drugs
Approved Before and After the Implementation of PDUFA

INTRODUCTION

 

The drug review and approval process in the United States has undergone significant changes in the last few years. Complaints about FDA's drug approval process at the beginning of this decade challenged the length of time it took to get a product reviewed by the Agency. Some critics argued drugs were too often available in other parts of the world sooner than in the United States. With the implementation of the Prescription Drug User Fee Act (PDUFA) in 1992, the Food and Drug Administration Modernization Act of 1997, and Agency managerial initiatives, major improvements in review time have occurred for both priority and standard drugs. Several recent news articles have indicated a belief that the Center for Drugs is now approving drugs too fast, without having adequate safety information available at the time of approval.

The General Accounting Office (GAO) published a report, FDA Drug Review -- Postapproval Risks 1976-1985, in April of 1990 that addressed safety concerns related to risks that were not uncovered until after drug approval and marketing. This allowed the Task Force to compare drugs approved during the PDUFA period with those of drugs approved in a prior time frame. Although the PDUFA program began officially in 1993, many drugs reviewed that year had entered the system pre-PDUFA. Thus, our study only looked at drugs approved during calendar years 1994 through 1997. Drugs approved in 1998 were not included since many of them have not yet been on the market long enough for unforeseen problems to have been identified and characterized. Both the GAO report and our analysis only looked at new molecular entities (NMEs).

This review had two goals: (1) to determine the rate of occurrence of serious postapproval risks in new drugs approved since changes in FDA review processes were made under PDUFA, and (2) to determine whether the rate of occurrence of serious postapproval risks in new drugs, as determined above, has changed since PDUFA became effective.

PROCEDURE

The supervisory consumer safety officers from the fifteen medical divisions were given a list of the 142 NME approvals during the period and asked to provide the following information

1. A copy of the approved labeling at the time of approval

2. A copy of the current approved label

3. Copies of all Dear Healthcare Professional letters related to the drug

4. summary of label changes that met the GAO criteria for serious postmarketing label changes based on the identification of significant postapproval risk (see discussion under Criteria)

5.Further details, if the drug was removed from the market due to serious postmarketing adverse events

 

One concern with this process was the fact that these products had not been on the market as long as the products studied in the GAO report. However, the GAO report states that"most unexpected adverse reports, particularly those that are serious, are expected to emerge within three years of approval." The GAO report did not specifically look at the time between drug approval and the first event leading to the identification of serious postapproval risk, so this statement cannot be verified.

GAO CRITERIA FOR A SERIOUS LABEL CHANGE

The criteria for assessing label changes were specified for each section of the drug label and are given in Table 1. If any criterion was met, the drug was classified as having a serious postapproval risk. The addition of a new indication, even though the inclusion of a new patient population might bring with it major risks for that group, was not included. The increase in risk had to be associated with the indication(s) for which the drug was originally approved.

 

Table 1

Criteria for Label Changes Reflecting Serious Postapproval Risk

Section of Label Criterion for Serious Change
Description None
Clinical pharmacology None, although increased understanding of pharmacokinetics may reflect or lead to changes in other sections of the label.
Indications and usage A limitation put on a drug's use or the removal of an indication because of adverse reaction reports' suggestion that use of the drug may lead to hospitalization, increases in the length of hospitalization, or severe or permanent disability or to death. The limitation on use must correspond to the indications for which the drug was originally approved.

Contra-

indications

The addition of a group of patients for whom the drug is contraindicated because it may lead to hospitalization or to increases in the length of hospitalization, severe or permanent disability, or death.
Warnings The identification of a concern not listed in the original labeling, a much greater concern for a condition recognized before approval, or the addition of a subclass of patients (e.g., those who already have some serious illness or some other characteristic) for whom the drug may pose substantial danger that may lead to hospitalization, increases in the length of hospitalization, or severe or permanent disability or death.
Precautions Changes that specify the need for increased diligence by the prescribing physician (e.g., in detecting underlying conditions or because of possible drug interactions that might pose a significant threat to the patient), the addition of a subsection providing information to alert the patient to watch for signs of a life-threatening adverse effect, or changes in other sections that are needed to forestall use of the drug that may lead to hospitalization, or severe or permanent disability or death.
Adverse reactions The addition of newly identified adverse reactions with a high frequency or an increase in the frequency of previously identified adverse reactions to a high level that may lead to hospitalization, increases in the length of hospitalization, or severe or permanent disability or death.
Drug abuse and dependence None
Overdosages The addition of overdose effects at recommended dosages that may lead to hospitalization, increases in the length of hospitalization, or severe or permanent disability or death.
Dosage and administration A reduction made to the recommended dosage because of concerns that a higher dose may lead to hospitalization, increases in the length of hospitalization, or severe or permanent disability or death.
How supplied None

 

RESULTS

During the 4 years studied, 142 NMEs were approved by the Center for Drugs. Of these drugs, two have been withdrawn from the market due to serious postapproval risks. Posicor (mibefradil dihydrochloride) and Duract (bromfenac), both approved during 1997, were withdrawn in June of 1998. One additional drug approved during the period, Redux (dexfenfluramine), has also been withdrawn from the market for safety reasons, but it is not included in the study because it was not a new molecular entity.

Posicor (mibefradil dihydrochloride), approved in August of 1997, is a calcium-channel blocker indicated for use in the treatment of patients with hypertension and chronic stable angina. Reports of serious adverse reactions after taking Posicor with several concomitant drugs led to label changes in December of 1997. The drug was withdrawn from the market by the manufacturer in June 1998 as a result of additional information about potentially harmful interactions with other drugs. Although in many cases drug interactions can be addressed by appropriate labeling changes and public education, the complexity of the prescribing information needed and the seriousness of side effects led to Posicor's withdrawal by the sponsor, Roche Laboratories.

Duract (bromfenac), approved in July of 1997, is a nonsteroidal anti-inflammatory drug (NSAID) approved for short-term management of acute pain (use for 10 days or less). It was never approved as a treatment for longer-term use for chronic conditions. In February 1998, following reports of severe liver failure, the warnings in Duract's labeling were strengthened with the addition of a black box and the sponsor, Wyeth-Ayerst Laboratories, issued a Dear Healthcare Professional letter. Despite these efforts, reports of severe injuries and death associated with liver failure continued to be reported with long-term use of Duract, leading to its withdrawal from the market in June 1998.

Table 2 gives the results of the review of the 142 NMEs approved in 1994-1997. It should be noted that all 24 products that had Dear Healthcare Professional letters also had label changes to reflect the safety concern. There were 19 products that had label changes that were considered serious, using the GAO criteria, that did not have an accompanying Dear Healthcare Professional letter. As should be expected, the numbers show a slight downward trend during the period, with an overall rate of just over 30 percent.

 

 

Table 2

Significant Label Changes for NMEs Approved in 1994-1997

 

Year

No Significant Label Changes

Significant Label Changes

Total NME Approvals

Percent of NME Approvals with Significant Label Changes

Dear HC Professional Letter

No Dear HC Prof. Letter

Total 

1994

13

4

5

9

22

40.9%

1995

19

4

5

9

28

32.1%

1996

39

8

6

14

53

26.4%

1997

28

8

3

11

39

28.3%

Total

99

24

19

43

142

30.3%

 

 

It would seem likely that drugs approved under accelerated approval, which can be approved based on surrogate endpoints reasonably likely to predict clinical benefits, would have more changes as the required phase-4 studies are carried out. In the 4 PDUFA years, 11 drugs have been approved under this provision (see Table 2a). While this 54.6 percent rate is higher than the overall rate of 30.3 percent, the numbers are not large enough to influence the total results.

 

 

 

 

Table 2a

Significant Label Changes for NMEs Approved in 1994-1997 Under Accelerated Approval Provisions

Year

Priority

Yes

No

Total

Percent

1994

1

0

1

100%

1995

2

2

4

50%

1996

2

2

4

50%

1997

1

1

2

50%

Total

6

5

11

55.6%

 

 

 

Under PDUFA, priority drugs -- those that appear to represent an advance over available therapy -- have a 6-month review clock compared to a 12-month review clock for standard drugs. The results for the PDUFA years were examined to see if the decreased time for review might lead to more problems postmarketing. There has not been a significant difference in serious postapproval risk identification between these two categories of products. The results, given in Table 2b, indicate no difference between priority and standard applications except for 1994. This difference is due to both an increase in the rate of label changes for priority applications and to a smaller rate for the standard submissions.

 

 

 

Table 2b

Significant Label Changes for NMEs Approved in 1994-1997

(by Priority or Standard Submission)

Year

Priority

Standard

Yes

No

Total

Percent

Yes

No

Total

Percent

1994

7

5

12

58.3%

2

8

10

20.0%

1995

3

6

9

33.3%

6

13

19

31.6%

1996

5

14

19

26.3%

9

25

34

26.5%

1997

3

7

10

30.0%

8

21

29

27.6%

Total

18

32

50

36.0%

25

67

92

27.2%

 

 

COMPARISON WITH 1976-1985

 

The GAO report indicates that "a total of 209 new chemical entities" name changed to new molecular entities) were approved during the 1976-1985 period, and their report addresses a total of 198 of the 209. They state that "of the 11 other drugs, four were never marketed; two were marketed for only a short time and then withdrawn, apparently for economic rather than safety reasons; two have not been marketed for some time and did not have up-to-date labels; and one was not considered a prescription drug. For the two other drugs, we were not able to obtain suitable labels for comparison."

Table 3, based on Table 3.6 of the GAO report, summarizes the results for 1976-1985 compared with those of 1994-1997. For the 10-year pre-PDUFA period, a total of 51.5 percent demonstrated postapproval increases in risk. The numbers of serious label changes per year during the period are highly variable and do not demonstrate a significant trend. However, a comparison of the first 5 years (1976-1980) with the last 5 years (1981-1985) of the GAO study do show a difference. The average for the first 5 years of the study is 58.5 percent and the average for the last 5 years of the study is 46.6 percent. Although it is likely that drugs approved in 1996 and 1997 will continue to experience additional label changes with continued use of the products, it is highly unlikely that the eventual overall rate for drugs approved under PDUFA will be as high as the 51.1 percent observed in the earlier products.

 

 

 

 

Table 3

Significant Label Changes for NMEs

Comparison of 1994-1997 Results With the 1976-1985 GAO Results


Year

No Significant Label Changes

Significant Label Changes

Total NME Approvals

Percent of NME Approvals with Significant Label Changes

1976

10

13

23

56.5%

1977

10

8

18

44.4%

1978

4

13

17

76.5%

1979

5

8

13

61.5%

1980

5

6

11

54.5%

1081

11

15

26

57.7%

1982

14

13

27

48.1%

1983

7

4

11

36.4%

1984

13

9

22

40.9%

1985

17

13

30

43.3%

Total

1976-1985

 

96

102

198

51.5%

1994

13

9

22

40.9%

1995

19

9

28

32.1%

1996

39

14

53

26.4%

1997

28

11

39

28.3%

Total

1994-1997

99

43

142

30.3%

 

Table 4, based on Table 2.1 of the GAO report, compares the significant label changes by Medical Category. Because of the disparity of the number of drugs within the various categories and the small number of drug products within many of the categories it is difficult to make direct comparisons.

 

 

Table 4

Significant Label Changes for NMEs by Drug Category

Comparison of 1994-1997 Results With the 1976-1985 GAO Results


Drug Class

1976-1985

1994-1997

No

Yes

Total

Percent

No

Yes

Total

Percent

101

Cardiac (1)

5

12

17

70.6%

4

2

6

33.3%

102

Antihypertensive-renal

9

6

15

40.0%

5

5

10

50.0%

201

Neurology

4

1

5

20.0%

9

2

11

18.2%

202

Psychopharmacological

6

9

15

60.0%

5

0

5

0.0%

203

Drug Abuse

2

3

5

60.0%

1

0

1

0.0%

301

Fertility-antifertility

2

3

5

60.0%

0

0

0

C

302

Metabolic-endocrine (1)

7

2

9

22.2%

10

3

13

23.1%

303

Metabolic-endocrine (11)

3

5

8

62.5%

7

0

7

0.0%

401

Antibiotic-systemic

7

18

25

72.0%

5

1

6

16.7%

402

Dermatologic

7

6

13

46.2%

7

0

7

0.0%

403

Anti-infective

4

2

6

33.3%

2

3

5

60.0%

404

Ophthalmic

3

3

6

50.0%

1

5

6

83.3%

405

Antiparasitic

2

2

4

50.0%

2

0

2</

 

Next Section : Appendix B