Who should be our next FDA Commissioner?

And what exciting ideas would they bring to the table?

Recently Trump nominated Robert F. Kennedy Jr. to be the Secretary of Health and Human Services. This is obviously a terrible pick given his wrongheaded views on vaccines and his peddling of conspiracy theories about COVID-19.

Trump can compensate a bit for this terrible wrong by appointing a great new Commissioner to lead the FDA. Three potential candidates are receiving a lot of attention right now on social media and in the blogosphere. Here are my thoughts on these candidates and the ideas, backgrounds, and skills they bring to the table:

Joseph Gulfo, MD

My #1 pick is Dr. Joseph Gulfo, MD, MBA. He has 25 years of experience in the pharmaceutical and medical device industries. As President and COO of Anthra Pharmaceuticals, he was instrumental in the 1998 approval of Valstar, a treatment for bladder cancer. Then, at Cytogen Corporation he contributed to the development of the monoclonal antibody ProstaScint. Finally, and most famously, as President and CEO of MELA Sciences he fought the FDA to get their MelaFind diagnostic device approved. Gulfo’s battle with the FDA led to a congressional hearing and is documented in his book Innovation Breakdown, which I reviewed previously. Gulfo has worked closely with all three major branches of the FDA - the Center for Drug Evaluation and Research (CDER), the Center for Biologics Evaluation and Research (CBER), and the Center for Devices and Radiological Health (CDRH).

Gulfo’s battle with the FDA while at the helm of MELA Sciences took a drastic toll on his health. Thus, after that ordeal he decided to retire from industry and focus full time on writing and research. He has written detailed policy documents on how to improve the FDA in conjunction with the Niskansen center and Mercatus center. He is currently a senior fellow at the Progressive Policy Institute.

According to Dr. Gulfo, the FDA must return to its original mandate, which emphasized promptly and efficiently taking action in a timely manner. As I’ve written about previously, the FDA currently moves way too slowly, which costs lives. Gulfo also points out that the original mandate states the FDA should promote the public health. Today, the FDA always describes itself as protecting the public health. This linguistic distinction may seem like nit-picking, but it reflects a deep change in philosophy that has taken hold over the years at the FDA. An over-focus on safety and protection comes at the cost of innovation and means many valuable drugs and treatments never make it to market.

Gulfo also points out that the FDA's legal mandate is to focus solely on safety and effectiveness. Over time the FDA has moved far beyond this, as officials have sought to increase the power of the FDA to micromanage healthcare in the US. Today, the FDA often insists that new drugs and devices must be better and cheaper than existing alternatives. While this sounds good, there are several major problems with this:

First, such requirements rule out personalized medicine in advance. Clinical trials only measure how good a drug works on average among a chosen group of trial participants. However, a drug may be worse on average than existing drugs but still be better for certain individuals. In some cases, the drug may be better for entire subpopulations that were not included in the original trial. By setting and requiring a “one size fits all” standard for what constitutes “better”, the FDA greatly limits what drugs can get to market. In the process, choice and the ability of doctors to personalize treatment are greatly reduced.

Another fact we need to consider is that clinical trials are never perfect sources of information. Larger trials may actually determine that drug X is better than drug Y, on average. Sometimes rare side effects and interactions are only discovered post-approval. If a patient is experiencing a rare side effect or interaction, then it can be helpful to have an alternative available, even if that alternative is not better on average.

Finally, demanding that drugs must be better than existing alternatives rules out combination therapies. A drug may be "worse" on its own, but useful in combination with existing drugs. Currently, combination therapies are generally not studied in FDA-approved trials. The discovery of combination therapies is almost entirely done post-market. This is especially important when tackling complex diseases like cancer.

The FDA also bars drugs and therapies that it thinks would be too expensive. However, considering cost is not in their mandate. Trying to predict prices in advance is, to borrow a phrase from Hayek, a “fatal conceit.” We simply do not have the ability to do so. In the end, drug prices can only be sorted out by the market.

According to Gulfo, the only criteria should be whether a drug, device, or therapy is safe and effective in some manner. After that, post-market surveillance, additional studies, and consensus practices among doctors will figure out which drugs have the most clinical utility and under what specific circumstances.

Gulfo is also mad at the FDA for overstepping their authority by banning perfectly valid tests over concerns about whether the information the tests provide is “clinically useful”. If a test gives accurate information, it should be approved. Doctors know best if and how to use any information a test provides. As an example, the FDA shut down 23andMe’s genetic testing program over concerns over how the results might be used. However, how to structure care based on test results is best decided by doctors working in conjunction with their patients, not bureaucrats in Washington, DC. Baring companies from sharing valid test results also violates the First Amendment!

Finally, I want to highlight Gulfo’s plan for clearer standards around what “effectiveness” means. Right now, the FDA sets the bar for “effectiveness” in a very haphazard way. The FDA’s ever-changing and often capricious definitions of effectiveness make it hard for companies to plan their R&D. Clearer standards around effectiveness would also increase venture capitalists’ ability to invest in drug and medical device startups.

Gulfo has a great plan here. He has carefully delineated several tiers or types of effectiveness:

Each type is color-coded, so doctors can easily understand which type of measure was used to approve a drug. Importantly, Gulfo strongly supports biomarker-based measures of effectiveness, which have been controversial at the FDA over the years. Enabling approval based on biomarker change is critical for enabling the approval of drugs and therapies that slow or reverse aging.

Balaji Srinivasan

Let’s now talk about Balaji Srinivasan. Always "the smartest person in the room", he is a polymathic thinker. Over the years he has worn many hats. In 2007 he co-founded the genetic testing company Counsyl. Later he served as CTO of Coinbase and as a General Partner at investment firm Andreessen Horowitz. Recently he wrote a book called The Network State, which envisions a future where technology enables communities to form self-sovereign states connected by shared values.

Balaji spoke brilliantly about the FDA for over an hour on Lex Fridman's podcast. More recently he spoke on the podcast of Vivek Ramaswamy, a close ally of Donald Trump who will soon be co-running the Department of Government Efficiency (DOGE).

Balaji understands the root issue that is drives a lot of problems at the FDA. Citing the book Reputation and Power, Balaji talks about the FDA's incentive structure which leads it to focus on protecting its reputation and increasing its power. Whenever a drug is approved that causes issues post approval, FDA officials are hauled in front of congress and grilled. However, whenever a drug is not approved that should have been, nothing happens. People may die, but their deaths are “invisible” (the “invisible graveyard”). As a result of this, FDA officials focus more on protecting their reputation and minimizing risk rather than doing rational risk-benefit analysis.

Balaji sums this up brilliantly as “minimizing side effects rather than maximizing effect sizes”. Approval thresholds need to be set in a rational way, weighing potential benefits and harms. Due to perverse incentives, the FDA focuses too much on reducing false positives rather than reducing false negatives.

Empirical work backs up these claims. In a 2017 paper in the Journal of Economics by renowned MIT economist Andrew W. Lo and others, it was found via a Bayesian analysis that the FDA's thresholds for approval of cancer drugs are ten times higher than they should be.

In 2016, Balaji was seriously considered by Trump’s team to be FDA Commissioner (alongside Gulfo and O’Neill). At the time, Balaji was branded by mainstream media outlets as a "radical libertarian.” While Balaji certainly has many libertarian views, when it comes to the FDA this is simply not true. In fact, Balaji has said he would not fire anyone from the FDA and he supports higher salaries at the FDA. Higher salaries are necessary for the FDA to be able to attract and keep the top talent they desperately need. Of course, this cannot be done by the Commissioner unilaterally and requires Congress to act.

Furthermore, Balaji wants to create new centers of excellence for regulation, co-located at leading universities and hospitals. These centers could develop new pathways for approval. This system would allow the FDA to scientifically study innovative new ways of doing regulation. Over time, trials could be done comparing the effectiveness of different approval pathways. This is similar to the system used in the EU, where different "Notified Bodies" are authorized to assess and approve medical devices. This allows a greater diversity of approval mechanisms and creates a more competitive system for regulation.

Balaji also has the endorsement of renowned geroscientist Aubrey de Grey, who points out that he will instill much-needed FOMO (“fear of missing out”) at the FDA. Balaji has spoken about how more drugs and therapies will be developed and marketed elsewhere if the FDA continues to lag far behind the cutting edge, like they do now. Having a strong, smart, and capable FDA is critical for our national competitiveness.

One possible issue with Balaji is that he has lived in Singapore the past four years. While he has talked about renouncing his citizenship, it appears he is still a U.S. Citizen, however.

Jim O'Neill

Finally we have Jim O’Neill. Unlike the other two candidates, Jim has experience working in the federal government, first as a speechwriter for the HHS Secretary and later as Principal Associate Deputy Secretary at HHS, between 2007 and 2008. While the precise extent of his accomplishments at HHS are unclear, this is certainly an asset he brings to the table. Later, O’Neill was a Managing Director at Peter Thiel's Mithril Capital. He later co-ran the Thiel Fellowship program. To quote the Foresight Institute, “Thiel Fellows have created more than $90 billion in equity value by founding companies including Freenome, Luminar, Figma, Upstart, OYO, Ethereum, Workflow, Fossa, and the Longevity Fund.” O'Neill was CEO of SENS Research Foundation until his abrupt departure in July 2021.

While I agree strongly with O’Neill’s views on the FDA, life extension, and other matters, I think he is the weakest out of these three candidates. While he occasionally posts on Twitter, O’Neill has not been writing and speaking about FDA policy as much as the other two candidates. Additionally, I have heard troubling things about his character. I won’t repeat what I’ve heard here, since I don’t like sharing gossip when I might be missing important context. Regardless, while O’Neill has held leadership positions, I am more impressed by the leadership accomplishments of Gulfo and Srinivasan.

As a final note, there have been 19 FDA commissioners since the founding of the modern FDA in 1962. 17 out of those 19 have held medical degrees. The remaining two held Ph.D.s in the biomedical sciences. Breaking from this tradition may cause more pushback and resistance from the FDA’s establishment. This is one reason to lean towards Gulfo, who has an MD.

If you liked this post, consider sharing this related thread on X (also on Mastodon).