Open in interactive viewer → charts, metric popovers & call review

Relmada: Stronger Science, Thinner Disclosure

18-month durability lands, but the CIS question gets deferred a third time — and the IND is still the only gate that matters
RLMD · Earnings Call · 2026-11-12

Relmada Therapeutics reported its third quarter on 12 November into a tape that had already made up its mind and then changed it: the stock fell 52% over the thirteen weeks to mid-October, then rallied 33% in the last three weeks into the print, against a $449 million market cap and roughly $202 million of cash. That is a name where the market has stopped pricing hope and started pricing gates. And the call was, almost line by line, an inventory of gates.

What actually changed

The clinic delivered. Twenty-three high-risk NMIBC patients in complete response at twelve months rolled into a long-term extension; of fourteen evaluable, twelve — 86% — remain in CR at eighteen months, and the BCG-unresponsive subgroup held 7 of 8, or 88%. The twelve-month pool has grown to 51 patients with an anytime complete-response rate of the sort that makes urologists pay attention: 94% overall, 96% in BCG-unresponsive. Two patients progressed to muscle-invasive disease at month 15, both T1 patient cases that were BCG-naive — and Brian Ritchie's explanation was disarmingly specific: “these patients, after the 12-month CR, did not comply fully with their maintenance dosing between months 12 and 15. So that could have been 1 factor, we don't know.” — Brian Ritchie, Chief Medical Officer · 2026-11-12 No radical cystectomies to date. No Grade 3+ treatment-related adverse events, no related SAEs, no related discontinuations, across the whole Phase 2.

But here is the piece of context the press release buries: the numbers came down as the dataset matured. In March 2026, the chief medical officer at the time described a twelve-month CR of 80% in BCG-unresponsive and 84% overall: “our 12-month CR in BCG-unresponsive is 80%, and our 12-month CR in the overall population is 84%. That includes four patients with CIS.” — Raj S. Pruthi, Chief Medical Officer · 2026-03-19 Today the same endpoints read 73% and 77%. That is the correct direction for a growing cohort and it is honest reporting — but it is a quieter number than the “best in class” framing that carried the first half of the year.

The genuinely new fact is temporal. In August, Sergio Traversa said of the eighteen-month cut: “we have not seen the data after the 12 months.” — Sergio Traversa, Chief Executive Officer · 2026-08-06 Three months later that data is the headline. Something the company could not describe in August now anchors the entire story. That is a real change of state, not a talking point.

Three quarters, one unanswered question

Kelsey Goodwin of Piper Sandler has now asked essentially the same question in March and again in November: how does this dataset compare to competitors that skew toward CIS? The answer this quarter:

We had a mix of CIS with pure CIS, CIS Ta, and CIS with T1 patients. We look forward to sharing additional results as the dataset matures and believe a future medical meeting may be a more appropriate forum to provide further details on the CIS subgroup.

Brian Ritchie, Chief Medical Officer · 2026-11-12

That is a deferral, not an answer, and it matters more than it looks. The Phase 3 single-arm study in second-line BCG-unresponsive CIS is the registration pathway most exposed to the CIS/papillary mix, and the one place where a competitor's cleaner population could make Relmada's headline CR look better than it is. When Andres Maldonado asked how many Phase 2 patients actually match that second-line SPECTRUM population, the answer was blunter still: “we have not conducted that analysis.” — Brian Ritchie, Chief Medical Officer · 2026-11-12 The Phase 2 was run in Israel, where “most of the first-line treatments were not available until very recently.” A cleaner, less pre-treated Phase 2 population is a gift to efficacy and a liability to extrapolation.

This is the same pattern the keyword history shows. The company's two most-used clinical anchors in prior quarters — “open label” and the “RESCUE program” — have both fallen away. The Phase 3 has been renamed SPECTRUM, framed as breadth across the “entire NMIBC disease spectrum,” and the old name is gone from the vocabulary entirely. Rebranding a program before the first patient is dosed is not sinister, but it is a signal that the company wants the conversation to be about the spectrum rather than the specific, hard-to-enroll, CIS-heavy slice that the first registration pathway depends on.

The CMC gate, and a nice problem to have

The most substantive disclosure was manufacturing. The Phase 2 product was, in Traversa's words, “literally a prototype made in the pharmacy of the hospital” and autoclaved at 120°C — a process structurally incapable of scaling to an IND. Relmada rebuilt it from zero and now has two parallel sterilization paths: Piramal in the US and Scinai in Israel. “We are waiting for stability data. And we have significant stability data already, and they look good... it's a nice problem to have, and it looks like we'll have both products will meet the FDA requirements for the IND.” — Sergio Traversa, Chief Executive Officer · 2026-11-12 The bottleneck has moved from scheduling, which is what August's call was about, to stability data, which is the last physical gate before an IND filing. IND clearance is guided for year-end, with first patient dosed within Q1 2027 on a 30-45 day post-clearance window including the holidays.

Note what this means for the global read: IND enabling study work has been one of the better-performing themes in the market over the trailing year. Relmada is not riding a wave here — it is late to its own.

Cash, and what the tape is rewarding

The balance sheet is genuinely transformed: $202.1 million versus $93 million at 31 December 2025, funding operations through 2029 including completion of the Phase 3 SPECTRUM studies. The filed numbers lag the call — the last 10-Q showed $116 million of effective net cash and a 22.7-quarter cash runway before the raise; the call's $202 million is not in a filed statement yet. Burn is real: R&D roughly doubled year over year to $9.2 million and net cash used in operations widened to $12.6 million, while the net loss of $13.9 million, or $0.12 per share actually narrowed per-share because G&A fell to $3.9 million from $6.3 million and the share count ballooned. R&D at $9.2 million this quarter is the number to watch: a two-arm Phase 3 does not run on that.

The macro tape offers an odd confirmation. Over the trailing ninety days, one of the better-performing clinical themes in the market has been precisely the one Relmada is selling — durability of response has been a top-quartile advancer, carried by VIR, IOVA and AZN, while the thirty-day global decliners are wall-to-wall AI data center and high-bandwidth memory names. Capital is rotating from compute to clinical catalyst. Relmada is a small expression of that rotation — but the market only pays for durability when it can size the population, which is exactly the question the company has now deferred three times.

The verdict: the science got stronger and the disclosure got thinner in the same quarter. What to watch is not the 18-month number — that is now in the deck forever — but whether the IND clears, and whether Relmada ever puts a CIS/papillary split next to that 94%.