SP-624

SP-624 (forvisirvat) is an oral small molecule from privately held Sirtsei Pharmaceuticals, a wholly owned subsidiary of Arrivo BioVentures, in development for major depressive disorder (MDD). It works by activating SIRT6, a nuclear enzyme that adjusts how DNA is packaged and how neurons handle stress. No SIRT6 activator has ever been approved for any indication, so this is genuinely first-in-class. The earlier Phase 2 (NCT04479852, n=317 treated) missed its primary endpoint in the overall population - MADRS mean treatment difference at week 4 was -1.8 points (95% CI -4.1 to 0.5, p=0.133) - with benefit apparent only in a post-hoc female subgroup analysis [1][2]. The current lead study, NCT06254612 (n=497 target, 20 mg once daily vs placebo, 4 weeks), is classified as Phase 2b/3 and is powered for MADRS change in female patients as the primary endpoint, following the sex-stratified signal [3][4]. Given that framing, this is best treated as a scientifically interesting bet on a specific responder subgroup, not a broad MDD asset.

SP-624 is a novel investigational compound with USAN-approved generic name forvisirvat (the first SIRT6-targeting molecule to receive a USAN, February 2025). It has no FDA approval, and no breakthrough therapy, fast-track, or orphan designation on public record [1][3]. Two Phase 2 MDD studies have completed: the earlier NCT04479852 (n=317 treated, ~67% female) and the ongoing follow-on NCT06254612 (targeted enrollment 456-497, Phase 2b/3), both sponsored by Sirtsei and both using change in Montgomery-Åsberg Depression Rating Scale (MADRS) at 4 weeks as the primary endpoint [1][3][4]. Critically, NCT04479852 did not meet its primary endpoint in the overall population (MADRS delta -1.8, p=0.133); a post-hoc analysis showed statistically significant MADRS separation in female participants, and NCT06254612 was explicitly designed to prospectively test that sex-stratified signal, with MADRS in female patients as the primary outcome [2][4]. A small Phase 1 pharmacodynamic study using brain network analytics in healthy adults and MDD patients has completed (NCT06570369, n=26) [5], along with an earlier Phase 1 exploration in schizophrenia (NCT04510298, n=27) [6]. Phase 1 single-ascending-dose and multiple-ascending-dose (SAD/MAD, standard Phase 1 dose-escalation designs) safety and pharmacokinetic data in healthy adults were published in Clin Pharmacol Drug Dev in 2025 [7]. Sirtsei/Arrivo is private, so financing, cash runway, and pivotal-trial timing are not disclosed in SEC filings. No standalone Phase 3 program is registered as of the most recent enrichment pass; NCT06254612 itself is pivotal-capable if positive.

SIRT6 is an enzyme that sits on chromosomes and quietly does housekeeping. It removes chemical tags from histone proteins (the spools DNA wraps around), which changes which genes are switched on, and it participates in DNA-damage repair and metabolic regulation. In the brain, low SIRT6 activity has been linked to poor DNA-damage repair in neurons, metabolic strain, and inflammatory signaling, patterns that also show up in depressed patients on autopsy and imaging studies. The bet behind forvisirvat is that pushing SIRT6 activity up nudges neurons back toward a healthier resting state, rather than tweaking a single neurotransmitter the way SSRIs or ketamine-derivatives do. That is a genuinely different mechanism from every approved MDD drug, which is both the appeal and the risk. Open Targets, which scores gene-disease associations on a 0-to-1 scale (higher = stronger evidence, with scores above 0.5 generally considered substantial), gives SIRT6 its strongest association with neurodegenerative disease at 0.49, with materially lower scores for depression specifically, so the human genetic support for MDD as an indication is thin [8]. A separate biological rationale involves sex-dimorphic SIRT6/chromatin biology: preclinical evidence suggests SIRT6 activity and downstream inflammatory and DNA-repair readouts differ between males and females, which is the mechanistic frame Sirtsei uses to justify a female-specific efficacy signal. No SIRT6 activator has ever been approved for any indication, so validation rests almost entirely on preclinical models plus the two Sirtsei Phase 2 readouts. The published Phase 2 in MDD is the first controlled human efficacy signal for this class in psychiatry, and the signal was subgroup-specific, not population-wide [2]. Until independent groups replicate or NCT06254612 confirms in a prospectively defined female cohort, the mechanism should be treated as biologically plausible but clinically unproven.

The current lead study, NCT06254612, is classified as Phase 2b/3 (not exploratory Phase 2), which matters for how the readout should be scored: a positive result on the prospectively defined primary endpoint could support a registration filing rather than require a separate Phase 3 [3][4]. It is a multicenter, randomized, double-blind, placebo-controlled study in adults with moderate-to-severe MDD. Enrollment target is 456-497 patients randomized 1:1 to forvisirvat 20 mg once daily or placebo for 4 weeks. Per Psychiatric Times coverage and consistent with the sex-stratified rationale from the prior Phase 2, the primary outcome is change from baseline in MADRS total score in female patients, driven by the post-hoc finding from NCT04479852 [4]. A 4-week readout is standard for early MDD trials but is on the short end for durability claims; SSRIs typically need 6 to 8 weeks to separate from placebo, while ketamine-class agents move MADRS within days, so forvisirvat sits in the middle. The comparator is placebo only, not an active antidepressant, which limits any head-to-head positioning claim. The earlier NCT04479852 used the same MADRS 4-week design at n=317 treated but with the overall population as the primary analysis [1]. Full results were published in Curr Med Res Opin in 2025: overall MADRS mean treatment difference -1.8 (95% CI -4.1 to 0.5, p=0.133) - primary endpoint not met - with post-hoc separation observed in women only [2]. A pharmacodynamic Phase 1 (NCT06570369) added quantitative EEG-derived brain network analytics as a mechanism-of-action readout, useful for internal go/no-go decisions but not a regulatory endpoint [5].

Our model estimates a 5% chance this drug is eventually approved. It starts from the historical base rate for Phase 2 drugs in this area (about 24%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by larger-than-typical enrollment for this phase; it is held back by heavier-than-usual blinding, the sponsor's thin or weak approval record, and weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk is the dominant concern and is elevated by the fact that the prior Phase 2 already missed its primary endpoint in the overall population. MDD trials routinely fail because placebo response rates run 30 to 40% and drug-placebo separation on MADRS is often only 2 to 4 points even for approved agents; forvisirvat's overall-population point estimate was -1.8, within that noise band. The prospective female-only primary in NCT06254612 rides on a post-hoc subgroup finding, and post-hoc signals frequently fail to replicate. A 4-week endpoint gives limited room to show durability. Safety risk is harder to bound because no SIRT6 activator has been chronically dosed in large populations. SIRT6 touches DNA repair, telomere maintenance, and metabolic pathways [10], so long-term toxicology in Phase 3 and post-marketing will need to rule out oncogenic and metabolic signals that a 4-week trial cannot see. The Phase 1 SAD/MAD study in healthy adults reported an acceptable early safety profile [7], but that is a low bar. Execution risk is meaningful: Sirtsei is private, Arrivo BioVentures is a small parent, and any follow-on registration-supporting program at competitive scale typically runs $150M or more, which likely requires a partner or a financing round. Commercial risk is severe even on success. MDD is dominated by generic SSRIs and SNRIs at pennies per pill, and payers push hard on step therapy. Recent novel-mechanism entrants (esketamine, dextromethorphan-bupropion) have had slow uptake despite approval, so forvisirvat would need a clear differentiation story on speed, safety, or a defined responder population (women). Patent and exclusivity details are not publicly disclosed by Sirtsei; without a clear composition-of-matter runway past mid-2030s, a female-subgroup MDD label may not attract the payer and marketing investment needed to compete with generic first-line therapy.

Worth watching, not yet worth trading around, and the framing matters. SP-624 is one of the few genuinely novel-mechanism antidepressants in late-stage development, and SIRT6 activation is different enough from monoamine, glutamate, and GABA approaches that a clean win would matter beyond Sirtsei. But the honest read is that this is now a bet on a specific responder population (women with moderate-to-severe MDD), not a broad MDD asset - the overall-population primary already failed at -1.8 MADRS points (p=0.133) in NCT04479852. The specific data point to look for is the prospective female-population MADRS delta at week 4 in NCT06254612, plus responder and remission rates. Because the prior post-hoc signal in women was described qualitatively as 'robust' but the overall trial only produced -1.8 points, a prospective female-population delta of 3 points or more with p<0.05 would be a clean confirmation and would justify serious follow-up. A delta in the 2-to-3 point range with borderline significance would be ambiguous. A miss in the prospectively defined female population would functionally end the program. Because Sirtsei is private and Arrivo BioVentures does not file public financials, the next commercial catalyst is more likely a partnership, licensing deal, or Series financing than a public disclosure of expanded Phase 3 plans. Watch for data presentations at ASCP (American Society of Clinical Psychopharmacology) annual meeting, typically late May-June, and NEI Congress in November, which are the natural venues for a US MDD readout.