Document the FDA's June 2021 accelerated approval of aducanumab (Aduhelm) against the explicit recommendation of its own advisory committee, the subsequent OIG investigation into irregular FDA-industry contact, and what this case reveals about the structural integrity of the FDA's advisory committee process.

Primary document analysis: FDA advisory committee meeting transcripts (November 6, 2020), FDA approval decision memorandum (June 2021), HHS OIG report on FDA-Biogen interactions (January 2022), Biogen SEC filings, and peer-reviewed commentary in NEJM, JAMA, and BMJ on the approval decision.

The Vote

On November 6, 2020, the FDA's Peripheral and Central Nervous System Drugs Advisory Committee voted on whether the clinical evidence for aducanumab supported approval for Alzheimer's disease. The vote was 10 against, 0 for, 1 uncertain. This is not a close call in FDA advisory committee terms. It is as close to a unanimous rejection as these committees produce.

The committee's objections were specific and technical. Biogen had run two Phase III trials — ENGAGE and EMERGE. ENGAGE failed. EMERGE showed modest benefit on one cognitive endpoint. The FDA's standard for accelerated approval allows use of a surrogate endpoint — in this case, amyloid plaque reduction — when that surrogate is "reasonably likely to predict" clinical benefit.

The committee's position was that the available evidence did not meet that standard and that the two contradictory trials could not be resolved by a post-hoc reanalysis Biogen conducted after terminating both trials early.

The FDA approved aducanumab in June 2021 anyway.

The Resignations

Three advisory committee members resigned from the committee following the approval. This is not a normal event. FDA advisory committee members do not typically resign over approval decisions. Dr. Aaron Kesselheim of Harvard Medical School, one of the resignees, called it "probably the worst drug approval decision in recent US history" and described the approval as undermining the purpose of the advisory committee process. I have read his resignation letter. It is worth reading.

The OIG Finding

The HHS Office of Inspector General released a report in January 2022 documenting "atypical" collaboration between FDA staff and Biogen during the review process. Specifically: FDA and Biogen staff held meetings and communications outside the normal review process that were not disclosed to the advisory committee. The OIG found no evidence of illegal conduct but described the interactions as inconsistent with FDA's stated policies on pre-approval contact.

The FDA's response to the OIG report described the interactions as within policy. This is a factual dispute between the FDA and its own oversight body.

The Price and the Medicare Response

Biogen launched aducanumab at $56,000 per year. This was not a pricing error. This was a calculation about what the market would bear for a drug with Breakthrough Therapy designation and FDA approval, evidence base notwithstanding. Under pressure — including from Medicare actuaries who estimated the drug could add $29 billion to Medicare spending if widely used — Biogen cut the price to $28,200.

Medicare's Centers for Medicare and Medicaid Services subsequently issued a National Coverage Determination in April 2022 restricting coverage to patients enrolled in qualifying clinical trials. This effectively removed aducanumab from clinical practice. The drug generated approximately $3 million in 2022 revenues against approximately $1 billion in development costs.

Biogen withdrew the drug from the European market in April 2022 after the European Medicines Agency declined to approve it, citing the same evidentiary concerns the FDA advisory committee raised in November 2020.

What This Case Documents

The aducanumab case is not primarily a story about a bad drug. Bad drugs are approved and fail in the market; that is a known failure mode. This case is a story about the FDA approving a drug over its own advisory committee's near-unanimous objection, with irregular industry contact during the review period, at a price that would have generated substantial Medicare expenditure for uncertain benefit to patients with a terminal neurodegenerative disease.

The FDA's accelerated approval pathway was designed for serious conditions with unmet need. Alzheimer's disease meets both criteria. The pathway's evidentiary standard — surrogate endpoints "reasonably likely to predict" clinical benefit — is necessarily lower than full approval.

The question this case raises is whether that lower standard was applied with independent scientific judgment or whether the determination of "reasonably likely" was shaped by the irregular contact the OIG documented. That question has not been resolved. The OIG report described the process as inconsistent with FDA policy. The FDA described the same process as within policy.

One of them is wrong.