An Evaluation of US Food and Drug Administration's Program to Register HIV Drugs for Use in Resource-Constrained Settings

Abstract

Importance: The US Food and Drug Administration (FDA) program to review antiretroviral drugs for use in low-resource settings via the US President's Emergency Plan for AIDS Relief (PEPFAR) now supports treatment of more than 14 million patients with HIV. However, an in-depth evaluation of the program has not been undertaken.

Objective: To conduct a quantitative analysis of the FDA-reviewed antiretroviral drug applications in order to assess the contributions of PEPFAR and to identify areas for improvement.

Design, setting, and participants: A cross-sectional study was conducted of all PEPFAR applications submitted to the FDA from December 1, 2004, to May 31, 2018. The analyses were conducted between October 2018 and February 2019.

Main outcomes and measures: Numbers and types of applications reviewed, how long it took for applications to obtain approval or tentative approval (time to registration), how often the FDA issued a complete response letter (CRL) identifying deficiencies precluding application approval or tentative approval and their reasons, and the association between CRLs and time to registration.

Results: Overall, 260 PEPFAR applications for 327 antiretroviral therapies were reviewed by FDA, of which 216 applications (83%) for 272 drugs were authorized for use. Of the 216 authorized applications, 184 applications for 231 drugs remain in active status and, thus, are available for use. Twenty-six percent (56 of 216) of the applications were for pediatric-specific formulations or strengths; the remainder were for adults. For all 216 applications, the median (interquartile range) time to registration was 10.0 (7.0-17.5) months. Thirty-seven percent (95 of 260) of the applications received 1 or more CRLs, resulting in a total of 172 CRLs; most applications received 1 CRL, whereas some were issued up to 6 CRLs. Among all CRLs, 264 deficiency reasons were identified; the most common deficiencies were associated with manufacturing processes (155 [44%]), followed by product labeling (62 [23%]), and failing facility inspections (54 [20%]). Complete response letters were associated with an increased time to registration. Applications without CRLs had a median (interquartile range) time to registration of 9.0 (5.5-12.0) months, whereas those with at least 1 CRL took a median (interquartile range) of 22.0 (14.0-38.0) months (P < .001).

Conclusions and relevance: The FDA's PEPFAR program has made many antiretroviral drugs available for global use. However, FDA and the pharmaceutical companies could take steps to improve the quality of applications submitted to prevent avoidable deficiencies in manufacturing processes and labeling. Further efforts to develop better, easier to use pediatric-specific therapies are needed.

Figure 1.. US President’s Emergency Plan for AIDS Relief Applications Submitted to the US Food and Drug Administration (FDA) Since 2004, by Application Type

The data for 2018 are partial, up to May 2018. Applications are shown regardless of current regulatory status. ANDA indicates Abbreviated New Drug Application; ART, antiretroviral therapy; and NDA, New Drug Application.

Figure 2.. Time to Registration and Time to Decision for Applications

The data for 2018 are partial, up to May 2018. A, Box plots show the median time to registration (overall median [interquartile range], 10.0 [7.0-17.5] months), which may include multiple review cycles and is inclusive of all time from the day the US Food and Drug Administration (FDA) receives an application for review to the day it is made available for use by the US President’s Emergency Plan for AIDS Relief. As such, the time includes FDA's review time as well as any time spent by the application holder in providing or responding to FDA questions related to the review. For example, denial of an application will inevitably increase time to registration because the application will require multiple review cycles; other factors, such as submission of new information by the applicant toward the end of review process, may extend the review time to give the FDA more time to consider the new information. B, Box plots show the time to decision by FDA on postregistration changes over time (overall median [interquartile range], 5.0 [2.0-12.0] months). The decision in this analysis is an affirmative decision to allow the requested change. Submissions for postregistration changes started in 2006 and since then the time to decision has consistently decreased from a median of 10.0 to 30.0 months to 5.0 months. Lines within boxes denote medians; bottom and top borders of boxes denote the 25th and 75th percentiles, respectively; vertical lines denote interquartile ranges; and circles denote outliers. CMC indicates Chemistry, Manufacturing, and Controls; REMS, Risk Evaluation and Mitigation Strategy.

Figure 3.. Complete Response Letters Issued to Applications

Scatterplot with applications (circles), trend line, and 95% CIs (shaded areas) shows that as the number of complete response letters issued to applications increases, so does the time to registration.