The FDA Didnt Reject Daiichi Sankyos Cancer Drug Because It Didnt Work

The FDA told Daiichi Sankyo the drug wasn't ready. Not because it didn't work — because the factory had a problem.

In June 2024, the agency declined to approve patritumab deruxtecan, a HER3-targeting antibody-drug conjugate that Daiichi was developing with Merck — confirming a manufacturing-related rejection separate from the clinical failure that followed. The Complete Response Letter cited inspection findings at a third-party manufacturing facility. No safety concern. No efficacy concern. Manufacturing. That was the first crack. The second came a year later, in May 2025, when Daiichi and Merck withdrew the drug's approval application after a clinical trial missed its survival target — a separate failure, explicitly described by Merck as unrelated to the earlier manufacturing citation. But by then the manufacturing commitment had already been made.

Now the bill has arrived. Daiichi announced an extraordinary loss of roughly $950 million for the fiscal year ending March 31, 2026, per Endpoints News — the largest single loss in its recent history. The write-off stems from long-term contracts with outside manufacturers for antibody-drug conjugates, or ADCs: precision-guided cancer drugs that combine an antibody — a protein that locks onto tumors — with a cytotoxic payload designed to destroy them. Daiichi had signed these manufacturing agreements, including minimum purchase obligations that committed it to pay for production capacity regardless of clinical outcomes. When pipeline assumptions didn't hold, the company was left holding obligations it couldn't unwind.

The Odawara plant upgrade that would have made one of its ADC programs in-house is now cancelled, producing a ¥19.3 billion impairment charge (roughly $123 million). Daiichi postponed its annual report from April 27 to May 11 and slipped another 10% on the announcement. The stock is down 26.5% year-to-date. And it cannot yet provision for the remaining gap between what it contracted to buy and what it now plans to use, because the medium-term exposure remains too uncertain to quantify.

The manufacturing overcommitment was not an isolated miscalculation. The clinical programs underneath it were already under strain. A second ADC, Datroway, saw its lung cancer rollout delayed — compounding the sense that the pipeline assumptions behind the manufacturing contracts had eroded on multiple fronts simultaneously. Daiichi pivoted to a risk-adjusted manufacturing strategy in mid-2025, but the CMO minimums had already locked in the exposure.

Daiichi is not a failing company. Its base pharmaceutical business grew 12.5% year-over-year, posting ¥2,123 billion in revenue (about $13.4 billion) for the fiscal year. The write-off is a setback to its biologics ambitions, not an existential event. But it illustrates a recurring biotech failure mode: signing long-term manufacturing contracts at peak-optimism valuations, then discovering the clinical and regulatory assumptions underneath them do not survive contact with reality.

The five-year business plan, due May 11 alongside the rescheduled annual report, will be the next signal. Until then, Daiichi is absorbing a loss it can account for, while unable to provision for the one it cannot yet quantify.

This story was reported using public filings, press releases, and regulatory correspondence.