Faster Drug Approvals, Weaker Data? Study Raises Concerns About FDA Process

An article published Tuesday in JAMA raises concerns that special and accelerated drug approval programs at the FDA in recent decades may have resulted in a process that approves drugs based on weaker data, without reducing overall drug development time.

An accompanying editorial describes the current regulatory process as “a thicket of special programs, flexible review criteria, and generous incentives,” and suggests starting points for reforms, including improving access to biosimilars.

In the article, Darrow et al. describe the evolution of FDA’s approach to drug approval from 1983 to 2018 based on federal laws, FDA regulations, drug approval records, and user fee records.

FDA must balance rigorous testing of new drugs to clearly define benefits and risks against timely approval for drug makers and access for patients. Special development, protection from generic competition, and expedited approval programs, such as Orphan Drug, Fast-Track, Accelerated Approval, Priority Review, and Breakthrough Therapy, were instituted by FDA to support drug development, especially for rare and serious diseases. However, in 2018, 81% of all new drugs won regulatory approval through one or more of the expedited programs, the article noted.

Special approval programs have increased administrative costs at FDA (paid for mostly by user fees), and postponements of generic competition have been costly to the US healthcare system, the authors noted.

Over the time period analyzed by the authors, FDA has accepted more surrogate measures, and as a result harder and more relevant clinical endpoints are studied less often. In 1995-1997, 80.6% of drug approvals were supported by at least 2 pivotal trials, compared to 52.8% in 2015-2017. The authors caution that reliance on surrogate measures may accelerate the approval of drugs that pose significant risk but have little clinical value.

The article reports that although FDA shortened its review times from more than 3 years in 1983 to less than 1 year in 2017, overall drug development time (from beginning human studies to approval) has not changed: approximately 8 years. The rate of new drug approvals (other than generics and biologics) has not increased substantially since 1983.

On the other hand, the authors acknowledge some positive outcomes. The median number of generics has increased following legislation to incentivize and accelerate generic drug development. Plus, biologic approvals are increasing over time, reflecting technological advancement. Although drugs are now supported by fewer studies before approval, the number of patients in these studies has not declined.

In the accompanying editorial, Joshua M. Sharfstein, MD, the former principal deputy commissioner of the FDA, suggests 4 starting points for reforms:

1. Rationalize and update programs to preclude unintended consequences (such as reducing competition) and to limit fast-track and breakthrough status to increase the likelihood that these treatments will offer major advances.
2. Strengthen postmarket safety oversight for drugs with the potential for both major benefits and serious risks, and take advantage of Risk Evaluation and Management Strategies programs.
3. Recalibrate those programs that provide drug makers with special marketing protections to reduce costs and improve access to biosimilars.
4. Promote the generation of definitive evidence via use of clinically relevant end points with patent and pricing incentives.

References

1. Darrow JJ, Avorn J, Kesselheim AS. FDA approval and regulation of pharmaceuticals, 1983-2018. JAMA. 2020;323(2):164-176. doi:10.1001/jama.2019.20288.

2. Sharfstein JM. Reform at the FDA—In need of reform. JAMA. 2020;323(2):123-124. doi:10.1001/jama.2019.20288.