I’m Dr. Diana Zuckerman, president of the National Center for Health Research, a patient-centered and consumer-oriented public health think tank. Our Center is very involved in FDA issues pertaining to the safety and efficacy of medical products, and I appreciate the opportunity to share my views today.
PDUFA performance measures have focused on speed, but in addition PDUFA performance measures should evaluate whether patients are protected from ineffective or unsafe products being approved! As Commissioner, Peggy Hamburg said innovation needs to mean products are better, not just new. The performance goals we’ve heard about today fall short, because they emphasize speed and ease of approval, not on the quality of the outcome of FDA reviews or of the outcome for the patients using these products.
PDUFA have resulted in more and faster approvals, but not all those approvals have helped patients, and some have seriously harmed them.
Premarket performance should also include evaluations of the percentage of applications that were rejected or withdrawn because there was a lack of evidence proving safety or efficacy. And the specific reasons why they were rejected or withdrawn.
When post-market surveillance works, it should sometimes result in FDA warnings, recalls, or withdrawals. FDA should provide the percentage of these for 5 years post approval and the reasons for those actions.
Performance should also include the percentage of products approved based on at least two well-designed studies providing solid scientific evidence. As someone trained in epidemiology, I love big data, but since most applications are for new products not yet on the market, clinical trials will still be the best data available. We want to know how many approvals were based on at least two phase 3 randomized, controlled trials demonstrating robust evidence of safety and efficacy and favorable benefit-risk profiles.
Performance should also be based on the percentage of approved products for which FDA mandated post-marketing studies and the percentage where those obligations were fulfilled – started and ended on time, conducted as required, and whether they did or did not confirm safety and efficacy. As you know, FDA recently had a meeting on widely used cancer drugs that were approved for certain indications through accelerated approval but failed to provide data confirming that they worked for those types of cancer. And yet the indications remained approved for years after it was obvious they did not work. Those delays are harmful to patients and should be considered a performance goal worthy of user fee support.
A newly published study indicates that too often a rejected application is subsequently resubmitted and approved when FDA ignores their own criticisms of the original application, even when those criticisms remain valid. The controversial approval of Aduhelm is just the most salient example of that.
Specific Changes to Commitment Letter
I have a general concern about the Commitment letter, because it changes policies that should be publicly debated by Congress and should include input from patient, consumer and public health advocates as part of any negotiations. Policies should not be negotiated behind closed doors at meetings that exclude those important perspectives.
I have time to recommend 5 specific changes to the Commitment letter:
- I was glad to hear about FDA’s new efforts to include patient preferences and involvement. The Commitment letter should specify that these activities should always include harmed patients, not just patients recruited by industry, who are often patients desperate for treatment. All patient perspectives are important, we all are concerned about patients who urgently need a treatment that works, but harmed patients have too often been excluded from FDA meetings and committees.
- Voluntary REMS strategies are rarely proven to work. The REMS program needs a complete overhaul or REMS should be avoided. Instead, most safety concerns should be resolved before products are approved. A good example is the REMS for prescription opioids, which FDA learned was not working. Few doctors took the voluntary training, fewer finished the voluntary training, and even those who were trained did not learn all the important issues that were included.
- The letter should implement the National Academies’ public health framework for regulatory oversight of opioids.
- In-person manufacturing inspections remain the most effective way to determine problems. We all understand that remote inspections were needed during the pandemic, but the Commitment letter should specify that remote inspections should be the exception.
- User fees should fund independent, objective studies to assess and quantify the harms that resulted or were avoided due to approval decisions.
And I will just add one other issue, since enhancements to the Sentinel program were discussed earlier in this meeting. I have been a strong supporter of the Sentinel post-market surveillance program, but it has been in place for years and cost an enormous amount of money. So, it was distressing to hear this morning that the FDA is still trying to figure out how to use those data so that they can provide usable information about safety and efficacy.
In conclusion, those of us who respect and admire the FDA know how important it is as a public health agency. We must make sure that industry user fees do not interfere with that essential mission. When performance goals and the Commitment letter are made behind closed doors, it is difficult for the public to have confidence that the FDA is a public health agency.