Replimune's FDA date passed with no decision. Silence is not a signal.

The FDA's goal date for deciding on Replimune's RP1 in advanced melanoma was Sunday, 2 August. It passed. As of Tuesday the agency had still not acted. The stock, up more than 120% across three sessions after a favorable advisory committee vote, is now trading on an absence.

An absence is the most misread object in biotech. It's worth being precise about what a missed decision date tells you, and what it doesn't.

A PDUFA date is a goal, not a deadline

PDUFA is the Prescription Drug User Fee Act, the arrangement under which drug companies pay fees and the FDA commits to review timelines in return. The "PDUFA date" everyone circles on a calendar is the agency's own goal date for taking an action on an application. It is a performance target the FDA sets for itself. It is not a statutory deadline, and nothing happens to the agency or to the application when one slips.

So the honest read of 2 August passing is narrow: the review isn't finished, or it is finished and the announcement isn't. Silence is not evidence of approval and it is not evidence of rejection.

You will hear that a late decision means the company and the agency are negotiating a label. You will also hear that it means another rejection is coming. Neither has anything behind it. That retail commentary is currently split between exactly those two readings is a decent illustration of what a genuinely uninformative event looks like.

The panel voted on a narrower question than the headline

On 30 July, the FDA's cellular, tissue and gene therapies advisory committee voted 10-3 in Replimune's favor. The question it was handed reads, in full: "Are the efficacy results from IGNYTE evaluable and clinically meaningful?" Not whether the drug should be approved. Advisory committees advise. The FDA is not bound by them.

The agency has published the meeting's materials but no tally or transcript, so the 10-3 split itself rests on trade reporting from the room.

The reviewers' objections are the substance of this decision, and those are on the record. From FDA's own briefing document for the meeting:

• On survival: "OS data from a single-arm trial without a concurrent randomized control arm are not interpretable for establishing a treatment effect."
• On the combination: "it cannot be determined whether RP1 contributes to any observed effect when administered in combination with nivolumab." IGNYTE was never designed to separate the two.
• On the response assessments: "unreliable response assessment methods confound interpretation of the reported efficacy results, limit FDA's ability to verify the reported results, and do not support the conclusion that this trial is adequate and well-controlled."

A fourth line has circulated widely — a reviewer calling a survival claim "not an evidence-based claim." That phrase appears nowhere in the published briefing document or FDA's slides. It was reported as said aloud at the meeting, and press reporting is the only basis for it.

None of that is new. This is the third pass at the same application. Replimune received complete response letters in July 2025 and again in April 2026, both citing trial design; FDA's regulatory-history slide summarises the second as finding that "issues from first CRL were not addressed." A complete response letter is the FDA saying it cannot approve the application in the form submitted: a rejection of the filing, not necessarily a verdict on the drug.

The number the argument is about

From the company's own three-year analysis presented at ASCO in May: an objective response rate of 33.6%, median duration of response of 24.8 months, median overall survival of 32.9 months, and 47.8% of treated patients alive at three years.

Translated: in patients whose melanoma had already progressed on anti-PD-1 therapy, that share saw their tumors shrink by the threshold the trial defined, and those responses held for roughly two years on median. In a population that has run out of options, that is a real result.

Now the limit, which is the part the FDA keeps returning to. Every one of those figures comes from a trial in which nobody was randomized to anything else — IGNYTE's registry entry records it as non-randomized and unmasked. There is no arm to compare against, so there is no way to establish how many of those patients would have responded to nivolumab alone, or how long a similar group would have survived without RP1. More follow-up doesn't fix this. The missing arm is missing permanently.

There is a second limit, and it is sharper than the first. The 33.6% is the company's count. FDA ran its own primary analysis on the same 140 patients and got 15.7%, after excluding 23 patients it deemed responders whose target lesions had all been injected and two who had no target lesions at baseline. Median duration of response fell from 24.8 months to 14.1. That is not a dispute about what the drug did; it is a dispute about what counts as a response when the drug is injected into the tumour being measured. Which number you use changes the argument entirely, and the agency has told you which one it uses.

What accelerated approval would actually be

If RP1 clears, it clears through accelerated approval — granted on a surrogate measure, in this case tumor response, with the confirming evidence owed later. The confirmatory study is IGNYTE-3, which randomizes patients against physician's choice of treatment, enrolls roughly 400 patients, and has overall survival as its primary endpoint. Those survival data are expected in 2030.

So the question being argued over this week gets settled around the end of the decade. In the interval, a positive decision puts the drug on the market and lets the company sell it. That is the trade accelerated approval makes, and it is worth understanding before treating an approval as the end of the story rather than a provisional milestone with a date attached.

Will they run out of money first?

Replimune held $268.9 million in cash, cash equivalents and short-term investments as of 31 March 2026, and said that funds operations into the first quarter of calendar 2027. That figure already includes scale-up for a potential launch.

Read the two together. Even the good outcome lands the company a few quarters short of the end of its own stated runway, with a commercial launch to fund from there. Approval would improve the terms of a raise. It would not remove the need for one.

The bad outcome isn't hypothetical here either. When the first complete response letter arrived in July 2025, the stock fell 64%. Same drug, same trial, same regulator, opposite direction. A binary that has already re-rated this company hard once is the argument for doing the work before the date rather than reacting after it.

What to watch

• The decision itself, and if it's an approval, the label. The population it covers is what determines the commercial size, not the fact of approval.
• If it's a third complete response letter: whether the FDA names a path this time, or repeats that the trial can't support the filing. Those are very different outcomes for the same headline.
• The terms of the next financing. A raise into an approval and a raise into a rejection are not the same transaction.
• IGNYTE-3 enrollment progress, since that trial is now the whole evidentiary backstop.

The portable lesson is about the calendar. Retail investors treat a PDUFA date as a scheduled event with a scheduled answer, which is why an empty date generates so much noise. It is a goal date on a review that either finished or didn't. When one passes quietly, the correct update to your view is close to zero, and knowing that is worth more than any interpretation of the silence.

This is analysis, not investment advice. It describes what has been disclosed and what has not, and does not recommend any position in any security.