SGEM Xtra: Tell Me Lies, Sweet Little Lies – FDA Approved & Ineffective

Subscribe: RSS

Date: August 12, 2025

Article: FDA Approved and Ineffective by Jeanne Lenzer and Shannon Brownlee. June 5, 2025. The Lever.

Jeanne Lenzer

Guest Skeptic: Jeanne Lenzer is a long-time medical investigative journalist and returning SGEM guest. Her previous work, including the book The Danger Within Us, explored how conflicts of interest and weak evidence can endanger patient care. In this new project with The Lever, Jeanne analyzes how the FDA approves drugs that often don’t meet basic efficacy standards.

I think many people assume that if a treatment is FDA-approved, surely it must work. However, people may be shocked to find out about the FDA’s drug approval process and how ineffective or harmful medications make it to market.

Most patients and doctors have no idea that the FDA has quietly flipped the drug approval process on its head by putting most drugs on the market before they are shown to be safe or effective, with the promise that they will do those studies after they are on the market.

Studies have shown that once a drug is on the market, patients do not want to enter or remain in clinical trials because they are convinced they’re being deprived of a proven treatment, and failure to enroll sufficient patients is one reason postmarket studies are delayed or never conducted.

Once a drug is on the market, the FDA can require postmarket testing, but the due dates are often set so many years in the future that the patent will have run out. Take the controversial new Alzheimer’s drug, Leqembi, which causes brain bleeds, swelling and death. The manufacturer is required to conduct safety studies and report annually to the FDA. However, neither the FDA nor the company will release the safety data until the final report is due. This will be in 2036, 13 years after it’s been on the market.

This can be considered a guaranteed win for Pharma and a bad deal for patients. The FDA has fallen so low that the agency introduced a new term, “dangling approvals” for drugs approved before clinical benefit was proved and then allowed to remain on the market even after the post-marketing studies fail. Richard Pazdur, head of oncology at the FDA, defended the agency’s refusal to order certain cancer drugs off the market even after their post-market studies failed, saying “A failed trial, doesn’t mean a failed drug.” Well, true and true, but the idea is that drugs were supposed to be proven to work before they go on the market, not after. And with Pazdur’s and the FDA’s reasoning, any bad drug could remain on the market forever. And a few have.

A two-year investigation by Lenzer and Brownlee found 429 FDA approvals (2013–2022) where the majority of drugs were authorized on inadequate evidence of effectiveness, with heavy reliance on surrogate outcomes, frequent lack of replication, and slow/absent confirmatory trials. The highlighted real-world harm and cost and call for regulators and clinicians to re-center on patient-oriented outcomes (POO) before widespread adoption.

The article gives a few stories about FDA-approved drugs that turned out to cause harm without any benefit. One example was the drug Elmiron:

Elmiron (pentosan polysulfate sodium) was approved in 1996 by the FDA for interstitial cystitis (chronic bladder pain)
FDA approval was given without solid evidence of effectiveness
FDA allowed approval on the condition that a follow-up study be done to confirm efficacy
The follow-up study took 18 years to complete and showed no benefit over placebo
Despite this, Elmiron remained on the market
Not just a lack of benefit, but there were reported cases of harm
Elmiron was associated with cases of a rare eye disease called pigmentary maculopathy that resulted in vision loss or blindness.
It was also associated with severe colitis and dozens of deaths
Bottom line is a drug with no proven benefit and significant harms stayed on the market for nearly 30 years
A cautionary tale of systemic FDA failure to protect patients

There were other examples in the articles:

Avastin (bevacizumab): Initially approved for metastatic breast cancer via progression-free survival (PFS) surrogate; later trials failed to show survival benefit; approval was withdrawn for that indication, but surrogate-based oncology approvals have continued apace.
Copiktra (duvelisib): Approved on surrogates; subsequent data suggested earlier death vs comparator and serious harms; FDA restrictions only years later.
ProAmatine (midodrine): Repeatedly rejected; ultimately approved under Subpart H with promised post-market evidence that never convincingly materialized.

Special Guest Skeptic: Dr. Jerome Hoffman. He is a longtime friend of the SGEM, a mentor and a true thought leader in evidence-based medicine (EBM).

Dr. Hoffman has critically appraised stroke studies, challenged the tPA (alteplase) hype, and now serves as the analytical lead behind this FDA investigation.

Four Questions for Jeanne & Jerry

Listen to the podcast to hear their answers to these and other questions.

What are the four standards the FDA claims to use when approving drugs?
- Control Group: Patients taking the drug were compared to a control group that was given a placebo or a comparator drug.
- Replication: At least two “well-controlled” trials showed the drug was eﬀective.
- Blinding (Masking): Subjects in the studies and the doctors who cared for them don’t know which patients are on the drug and which are in the control group.
- Clinical Endpoint: The studies measured the drug’s eﬀect on patients’ survival or function rather than a surrogate measure.
Of the hundreds of drugs analyzed, how many meet all four?
- Therefore, 44% of these drugs met two or fewer of the FDA’s four core criteria for approval.
- 73% percent of drugs approved by the FDA did not meet the agency’s four foundational standards required to show they work as expected.
- More than half of drug approvals were based on preliminary data rather than sound evidence that patients had fewer symptoms, improved function, or lived longer.
- Fifty-five of the 429 drugs approved met only one of the four standards needed to show that a drug is safe and effective; 39 drugs met none of them.
Which category of drugs seems to be the worst offender, and why?
- Oncology was the standout concern: of 123 cancer drugs, 2.4% (3/123) met all four criteria; 29 met none.
- 81% of cancer approvals used surrogates (progression-free survival) rather than overall survival.
What changes are needed? Do we fix the FDA, or start from scratch?

Key Take Home Points:

Even giving them the benefit of the doubt, nearly ¾ of drugs failed to meet one or more of the standards.
Cancer drugs are among the worst, often approved on flimsy surrogate endpoints like tumour shrinkage.
Elmiron, for example, stayed on the market for nearly 30 years despite never being shown to work and causing blindness in some patients.
The FDA is funded largely by industry through user fees, leading to deep structural conflicts of interest. Clinicians must remain skeptical, dig deeper into the evidence, and resist the assumption that “FDA-approved” means “safe and effective.”

SGEM Bottom Line: Being approved by the FDA doesn’t mean a drug is effective or even better than a placebo. All we can conclude is that the FDA approved the drug. We need to be skeptical, look at the primary evidence, and be willing to question the system.

People may not know, but the SGEM tagline came from a lecture given by Dr. Hoffman. He concluded the lecture by encouraging the audience to be skeptical, even of him.