FDA to Akums: 'Analyst Error' Is Not a Root Cause

Akums Drugs and Pharmaceuticals Ltd. of Haridwar, India received a CDER warning letter dated February 25, 2026 for a failure FDA has seen too many times before: out-of-specification results were being pushed aside as "analyst error" or "laboratory error" without scientific proof that any analyst or laboratory error actually occurred.

That detail matters more than the individual failed test. FDA did not frame this as a one-off documentation miss. The agency said the firm "failed to thoroughly investigate" unexplained discrepancies and "did not adequately extend" the investigation to other batches and products that may have been affected. That is a system finding, not a lab anecdote.

The regulatory anchor is explicit. Under 21 CFR 211.192, any unexplained discrepancy or failure of a batch or any of its components to meet specifications must be thoroughly investigated, whether or not the batch has already been distributed. That investigation must include a review of relevant laboratory records. FDA cited Akums because the record did not support the invalidation decisions, and because the surrounding review appears to have accepted those decisions anyway.

This is the enforcement pattern quality teams keep underestimating. FDA is not only asking whether an OOS result was real. FDA is asking whether the firm had a defensible, scientific process for deciding that an OOS result was invalid. If the answer is no, the invalidation becomes evidence of weak quality oversight.

Akums is not a marginal site. It is one of India's largest contract manufacturers, and it supplies multiple global markets, including the United States, the European Union, and regulated African markets. That scale changes the significance of the warning letter. If FDA were to move toward an import alert, the operational impact would not stop at one plant or one domestic market. Customers across several regulated supply chains would feel it.

The warning letter also makes clear that recurrence is part of the story. FDA said the CAPA system had already failed to prevent the same OOS invalidation pattern from reappearing after a prior inspection cycle. That is where agency patience tends to disappear. Once a firm has already been told the weakness exists, another round of unsupported invalidations stops looking like poor judgment and starts looking like a quality system that never actually changed.

That is why this letter belongs in the same enforcement conversation as recent actions involving Indoco Remedies, Outin Futures, and Brands International. The products differ. The immediate defects differ. The recurring failure mode does not. Investigations are kept narrow, suspect results are explained away too quickly, scope is not extended to connected batches or products, and corrective actions close records without proving the problem is under control.

FDA has been especially blunt on one point: "analyst error" is not a root cause just because it is convenient. It is only a root cause if the record shows what happened, how it happened, why the conclusion is scientifically justified, and what objective evidence supports invalidating the result. Without that chain of proof, the phrase functions as a label for uncertainty, not an explanation.

The remediation FDA demanded from Akums follows directly from that logic. The agency called for a comprehensive review of all OOS investigations conducted in the past two years, an independent assessment of the overall investigation program, and a global CAPA that addresses both root cause and scope. In other words: do not fix one file. Rebuild the decision system that produced the file.

For any manufacturer reading this, the practical question is not whether your SOP mentions OOS. The question is whether your last invalidated result could survive challenge from an investigator who asks for raw records, original reasoning, scope justification, and evidence that similar lots, methods, analysts, and products were assessed. Pull one invalidated OOS this week and test it against 21 CFR 211.192. If the file cannot prove why the result was set aside and how the risk was contained, treat that gap now, before FDA treats it for you.