Entinostat

Entinostat is an oral pill that blocks HDAC enzymes - proteins that help cancer cells hide from the immune system and resist hormone therapy. Syndax Pharmaceuticals (with NCI cooperative groups) ran it in Phase 3 for hormone receptor-positive breast cancer (E2112/NCT02115282, n=608) in combination with exemestane after a Phase 2 signal (ENCORE 301) failed to hold up under Phase 3 testing. The Phase 3 missed both PFS and OS. Newer Phase 2 work pairs it with PD-1 blockers in pancreatic cancer and lymphoma, hunting for a niche where HDAC inhibition unmasks tumors to checkpoint therapy [1][2].

Entinostat is a novel small molecule, never approved anywhere despite ~20 years of clinical work. It carries FDA Breakthrough Therapy designation (granted 2013) for HR+ breast cancer based on the ENCORE 301 Phase 2 readout, plus Orphan Drug status in myelodysplastic syndromes. The BTD was driven by ENCORE 301's overall-survival signal (median OS 26.9 vs 19.8 months, HR 0.59, 95% CI 0.36-0.97, p=0.036) in a 130-patient unselected HR+ population [7] - a striking benefit for a Phase 2, but one that did not replicate in the larger E2112 Phase 3. The flagship Phase 3, E2112 (NCT02115282), enrolled 608 women with HR+/HER2- advanced breast cancer progressing on a non-steroidal aromatase inhibitor, randomized to exemestane ± entinostat. It missed its co-primary PFS endpoint (median 3.3 vs 3.1 months, HR 0.87, 95% CI 0.67-1.13, p=0.30) and, in the final 2021 readout, missed overall survival as well (median 23.4 vs 21.7 months, HR 0.99, 95% CI 0.82-1.21, p=0.94) - essentially null on both endpoints. The trial is listed as ACTIVE_NOT_RECRUITING for follow-up only [3][8]. Syndax has since shifted its lead asset focus to revumenib (menin inhibitor, approved as Revuforj November 15, 2024 for KMT2A-rearranged acute leukemia [6]) and axatilimab (Niktimvo, CSF1R, approved August 2024 for chronic GVHD) - entinostat is now a partnered/legacy asset rather than a near-term commercial bet. Current entinostat trials are investigator-initiated or cooperative-group sponsored rather than Syndax-led registration studies, consistent with passive maintenance rather than active development. A 2025 meta-analysis of entinostat + endocrine therapy in HR+ disease found no statistically significant survival benefit across pooled trials [1]. Entinostat's composition-of-matter IP dates to the early 2000s and is approaching or past expiry in major jurisdictions, which limits partnering economics even if a future biomarker-selected signal emerges (caveat: specific patent dates not independently verified here).

HDACs are enzymes that strip acetyl groups off histones - the spools that DNA wraps around. When acetyl groups come off, DNA winds tighter and genes get silenced. Cancer cells abuse this to switch off tumor-suppressor genes and immune-recognition genes. Block HDACs and you re-open those silenced genes, in theory waking up tumor suppressors and making tumors visible to T cells. Entinostat is a benzamide that selectively hits class I HDACs (HDAC1, 2, 3) [4] - narrower than approved pan-HDAC drugs like vorinostat or panobinostat, which is meant to give a cleaner safety profile. The mechanism is genetically validated: HDAC1 knockouts derail tumor growth in mouse models, and four HDAC inhibitors are already approved in T-cell lymphomas and myeloma. The harder question is whether HDAC inhibition does anything useful in solid tumors. The hypothesis behind the breast cancer program - that entinostat reverses endocrine resistance by re-sensitizing ER signaling - looked plausible in ENCORE 301 but did not survive Phase 3 [3][8]. The immune-priming rationale is the basis for current checkpoint combinations: HDAC inhibition increases MHC class I expression (MHC class I proteins are the flags tumor cells display so T cells can find and kill them) and reduces myeloid-derived suppressor cells (MDSCs are immune cells that tumors recruit to suppress T-cell attacks; reducing them is meant to let checkpoint drugs work) [2].

The defining trial, E2112 (NCT02115282), was a 608-patient Phase 3 run by ECOG-ACRIN with NCI sponsorship: HR+/HER2- advanced breast cancer, post-aromatase-inhibitor progression, randomized 1:1 to exemestane + entinostat 5 mg weekly vs exemestane + placebo. Co-primary endpoints were PFS and OS. The design was a direct successor to the Phase 2 ENCORE 301 (PFS 4.3 vs 2.3 months; OS 26.9 vs 19.8 months, HR 0.59) [7] - a reasonable bet at the time, though enrolling on a Phase 2 PFS/OS signal in breast cancer has a poor historical track record. The trial missed PFS at interim (HR 0.87, p=0.30) and missed OS at final analysis (HR 0.99, p=0.94) - the OS curves were essentially superimposable, eliminating any 'near-miss' interpretation [3][8]. Currently active trials are smaller and combination-focused: NCT02453620 (entinostat + nivolumab + ipilimumab, n=57, HER2- breast and solid tumors, Phase 1), NCT04708470 (HPV+ and GI cancers triple immunotherapy, n=55, Phase 1/2), and a Phase 1b dose-finding with the BET inhibitor ZEN-3694 (NCT05053971, n=49) [5]. The pancreatic Phase 2 with nivolumab (n=27) reported in Nat Commun 2024 showed ORR ~3% - a clean miss [2].

Our model puts the odds of this drug eventually getting approved at 20%. That figure starts from a historical baseline - about 48% of Phase 3 drugs in this area reach approval - then shifts based on ten facts about the trial and sponsor. The estimate is nudged upward by the trial's non-randomized design, but pulled back by the sponsor's weak approval track record, enrollment smaller than typical for this phase, and few secondary endpoints. The remaining facts fell close to average, so they didn't move the number much.

Efficacy is the dominant risk and it is largely already realized. E2112 missed both co-primary endpoints in HR+ breast cancer with hazard ratios essentially at 1.0 - not a near-miss [3][8]. The mechanism appears to work in T-cell lymphoma (where vorinostat, romidepsin, and belinostat are approved) but has repeatedly failed to translate to solid tumors. Safety is manageable but not trivial: class I HDAC inhibition causes fatigue, neutropenia, thrombocytopenia, and GI toxicity - tolerable in T-cell lymphoma where there are few options, harder to justify in HR+ breast cancer where CDK4/6 inhibitors and elacestrant have raised the bar. Commercial risk is now severe: the HR+ breast cancer market is dominated by Verzenio, Ibrance, Kisqali, and now elacestrant for ESR1-mutant disease - even a marginal positive readout would face poor pricing use. Execution risk for Syndax is mainly opportunity cost; the company's value is now in revumenib (menin inhibitor, approved Nov 2024 [6]) and axatilimab (CSF1R, approved Aug 2024), not entinostat. The realistic risk is that entinostat quietly winds down to investigator-initiated combination work without a clear registrational endpoint [5].

Mostly noise at this point. The signal-generating event for entinostat already happened - E2112 read out negative in 2020-2021 - and the asset has effectively been deprioritized as Syndax pivoted to revumenib and axatilimab, both now approved. The interesting story here is about Syndax the company, not entinostat the drug: SNDX's 2025-2026 8-K cadence (five filings between May 2025 and April 2026) is about revumenib launch dynamics and axatilimab uptake in chronic GVHD, not entinostat readouts. A check-back trigger would be a positive Phase 2 readout from one of the active immunotherapy combinations - particularly NCT02453620 (entinostat + nivo + ipi) or the HPV-associated Phase 1/2 (NCT04708470) - that identifies a biomarker-selected responder population. Absent that, treat entinostat as a legacy asset with optionality value, not a near-term catalyst. With composition-of-matter IP aging out, even a partnering deal would have limited upside. The broader read is that class I HDAC inhibition outside hematologic malignancies has accumulated enough negative Phase 3 data that future programs will need a much sharper biomarker hypothesis to be worth running.