Physician Expectations Regarding the FDA Approval Process May Differ From Reality

Sanket Dhruva, MD, MHS

When examining the perspectives of physicians regarding the FDA approval process for drugs and high-risk medical devices, investigators found that many would value a more rigorous approval process, according to Sanket Dhruva, MD, MHS.¹

The FDA began granting accelerated approval indications to drugs in 1992, allowing for the approval of an agent based on a surrogate end point and, following the FDA Safety and Innovation Act of 2012, these approvals may occur when agents that fill an unmet need to treat serious conditions affect a surrogate or intermediate clinical end point.² Currently, approximately 59% of FDA-approved drugs hold indications based on changes to surrogate measures.¹

Findings from a national survey published by Dhruva et al in Health Affairs revealed that among board-certified clinicians in internal medicine, medical oncology, or cardiovascular medicine surveyed on drugs (n = 489) and medical devices (n = 479) approximately 50% thought it should be very important to the FDA’s approval decision if a study’s primary end point was surrogate or clinical. Most physicians reported they thought data from 2 or more randomized controlled trials should be required for the approval of drugs (57%) and medical devices (60%). Blinding, randomization, and the number of patients in the trial were also all factors that more than 75% of physicians surveyed thought should be “very important” to the FDA’s approval decision.¹

In an interview with OncLive^®, Dhruva, an assistant professor of Medicine at the University of California San Francisco, detailed the 3 key findings from the study, and in an additional interview he highlighted the implications of the research as well as ways physicians can learn more about the FDA approval process.

OncLive: How was this survey conducted?

Dhruva: We identified a sample of 1450 physicians who are board certified by the American Board of Internal Medicine, and these physicians were distributed among oncologists, cardiologists, and internists. We sent [physicians] a survey via email with email reminders and for those who didn’t reply a mail reminder that we developed. The survey focused on 4 different domains with questions in each domain.

First, we asked about their perception of FDA approvals. Next, we asked them what characteristics they think are important in the trials that support FDA approvals for both drugs and high-risk medical devices. We then asked about expedited FDA development and review pathways. A couple of questions were also focused on moderate-risk medical devices. All survey questions were answered by the board-certified physicians using an electronic survey platform, then the study came into our research database, and all the research was conducted with institutional review board approval. Then, we set off to analyze the results of our survey questions.

What was the first notable finding from the physician survey results?

The first [key finding] of our study [was regarding] overall physician understanding of the FDA approval process. Among the total cohort of physicians—this is self-report in a confidential survey—physicians told us that they understood the FDA approval process, either ‘extremely well’ or ‘moderately well’, [at a] 41% [rate] for drugs and 17% [rate] for medical devices.

A substantial portion of physicians feel that they don’t understand the FDA approval process moderately or better in terms of ‘well’, and you can see that in terms of the ‘not at all well’ [statistics with] 12% of physicians for drugs and 39% of physicians for medical devices [reporting that they thought] they understood the FDA approval process ‘not at all well’. That’s a key finding regarding the overall understanding of FDA approval processes.

How much did physicians value study characteristics for trials supporting approvals?

We asked physicians how important they perceive various study characteristics should be in the trials supporting the FDA approval of drugs or medical devices. Physicians overwhelmingly felt randomization, the use of blinding, and the use of comparators were very important. [They felt] these trials should all have at least 1 primary end point [and] meet the primary end point to support FDA approval, among other characteristics.

This makes sense as these are important characteristics of rigorous clinical studies. When the study is conducted if there is the use of randomization, there is blinding, the primary end point is met, [and] the trial is published in peer-reviewed literature, that gives us greater confidence in the findings of that trial. This is a particularly important finding because it contrasts with the evidence standards that are sometimes used for FDA approval of drugs and high-risk medical devices. For example, there’s not as much as 90% use of randomization [in trials], even though [approximately] 90% of physicians felt that should be an important characteristic. The use of blinding in medical device clinical trials is very infrequent even though more than 70% of physicians felt that this should be an important characteristic. We’re seeing that there is some disconnect between what physicians expect and what the actual realities are of the trials that are used to support FDA approvals of drugs and medical devices.

What has been observed in terms of pre- and post-market trials in the FDA approval process?

In general, there has been greater use of expedited pathways in recent years, such as priority review, breakthrough therapy designation for drugs, and the Breakthrough Devices Program for medical devices. When there is less evidence in the pre-market setting then we need to have more evidence in the post-market setting to close those knowledge gaps or provide more evidence of safety and effectiveness. The pre-market evidence is a bit earlier and [that should] then be complemented in the post-market setting; once the drug or device is approved, we [need] more evidence [to] make sure that it works as intended.

Unfortunately, sometimes post-approval trials are delayed, not conducted at all, [or] show that the drug or device [does not work as expected]. [Perhaps] we initially thought it might work, but in the pre-approval setting we only had a smaller trial, or we used a surrogate end point such as looking at a blood test [and] not measuring how patients feel or if they live longer. We asked physicians, 'What do you expect at that point in time?' They said that if [post-market] trials for an accelerated approval drug or medical device are delayed or not conducted, approximately 60% wanted the FDA to withdraw approval of the drug and an even higher number of physicians wanted the FDA to revoke approval for devices. We know that the FDA extremely infrequently does this.

Similarly, if a post-approval trial is completed and shows that an accelerated approval drug or Breakthrough medical device doesn’t work physicians overwhelmingly—more than 90% for drugs and more than 80% for medical devices—said they want the FDA to withdraw approval. We know that very rarely happens and [there is] somewhat of a disconnect between what physicians were expecting and what has been happening.

References

1. Dhruva SS, Kesselheim AS, Woloshin S, et al. Physicians’ perspectives on FDA regulation of drugs and medical devices: a national survey. Health Aff (Millwood). 2024;43(1):27-35. doi:10.1377/hlthaff.2023.00466
2. Accelerated approval. FDA. Updated February 24, 2023. Accessed January 31, 2024. https://www.fda.gov/patients/fast-track-breakthrough-therapy-accelerated-approval-priority-review/accelerated-approval