ZEN-3694

ZEN-3694 is Zenith Epigenetics' oral BET bromodomain inhibitor, a class of drugs that block proteins tumors use to keep their growth programs switched on. The lead Phase 2 program combines ZEN-3694 with enzalutamide and pembrolizumab in metastatic castration-resistant prostate cancer (mCRPC) that has progressed on androgen-signaling drugs (NCT04471974) [1]. At least five other combination trials, most run by NCI-sponsored investigator groups rather than by Zenith itself, are testing the drug in triple-negative breast cancer, squamous lung cancer, colorectal cancer, and solid tumors paired with talazoparib [2][3][4][5][6]. The commercial context is mixed. The BET class has produced repeated efficacy and thrombocytopenia failures over a decade, and no BET inhibitor has yet reached FDA approval. The one bright spot is pelabresib (CPI-0610, MorphoSys, now Novartis), whose Phase 3 MANIFEST-2 trial in JAK-naive myelofibrosis hit its primary SVR35 spleen-volume endpoint but where Novartis paused filing over leukemic transformation signals and launched MANIFEST-3 to gather more data before regulatory submission [10]. Pelabresib is proof the mechanism can move a hard endpoint in the right disease, but it is not yet an approval. ZEN-3694 is the surviving asset of a program that was originally spun out with the thesis that a slightly less potent, more forgiving PK profile could break the class deadlock. The Phase 1b/2a signal in mCRPC published in 2020 was modest but interpretable, which is why NCI kept building trials around it [7].

ZEN-3694 is a novel small molecule that has never been approved anywhere. It sits at Phase 2 in its most advanced indication, mCRPC, with the enzalutamide combination based on published Phase 1b/2a data from Aggarwal and colleagues showing a manageable safety profile and durable disease control in a subset of enzalutamide-resistant patients [7]. No FDA breakthrough, fast track, orphan, or accelerated approval designations have been publicly announced for ZEN-3694 in any indication. The active trial matrix listed in ClinicalTrials.gov as of this writing includes NCT05327010 (talazoparib combo, active not recruiting, n=88, ORR primary endpoint) [2], NCT05607108 (squamous NSCLC, recruiting, n=13, MSK-sponsored) [3], NCT05422794 (TNBC with pembrolizumab and nab-paclitaxel, Phase 1, n=57) [4], NCT06102902 (colorectal with cetuximab plus encorafenib, Phase 1) [5], and NCT05803382 (capecitabine combination, Phase 1) [6]. The mCRPC Phase 2 (NCT04471974) is sponsored by Rahul Aggarwal at UCSF with Zenith and Merck as collaborators, enrollment 61 actual (active, not recruiting), with a composite response rate primary endpoint defined as either RECIST 1.1 objective response or a confirmed 50% or greater decline in serum PSA, measured up to 3 years [1]. No publicly disclosed readout date. Zenith is privately held, so pipeline updates come from investigator presentations at ASCO, ASH, and AACR rather than from earnings calls. Zenith's last publicly disclosed financing was a modest private round several years ago, and no recent up-round has been announced, which reinforces the operational dependence on NCI-funded trials.

BET stands for bromodomain and extra-terminal, a family of four proteins (BRD2, BRD3, BRD4, and BRDT) that read chemical tags on chromatin, the DNA-plus-protein material that packages the genome [8]. Specifically, BET proteins recognize acetylated lysine residues on histones, the spool-like proteins DNA wraps around. Think of them as adapters that dock onto activated regions of the genome and recruit the machinery that reads out genes. Tumors abuse this system: many cancers depend on high-output transcription of oncogenes like MYC and, in prostate cancer, on androgen receptor splice variants such as AR-V7 that drive resistance to enzalutamide and abiraterone. Blocking BET proteins collapses these hyperactive transcriptional programs. The genetic case is strong. MYC amplification, NUT midline carcinoma (a BET-fusion cancer), and AR-V7-driven mCRPC all rest on transcription addictions that BET inhibitors interrupt in preclinical models. AR-V7 positivity is detected in roughly 20 to 40 percent of enzalutamide-resistant mCRPC patients depending on the assay and disease timing, which frames the addressable subpopulation for a targeted BET strategy. Preclinical work has also shown ZEN-3694 restores CDK4/6 inhibitor sensitivity in ER-positive breast cancer models [9]. The problem has never been the biology. It has been finding a therapeutic window where the same mechanism that shuts down tumor transcription does not also crash platelets and gut epithelium.

The lead Phase 2 study NCT04471974 tests ZEN-3694 plus enzalutamide plus pembrolizumab in mCRPC patients who have progressed on prior androgen receptor-signaling therapy [1]. The trial is investigator-initiated out of UCSF (sponsor-investigator Rahul Aggarwal, MD), not sponsored by Zenith directly, which matters commercially because it means Zenith is not paying to run it. Actual enrollment is 61 patients, and the primary endpoint is composite response rate defined as either objective response by RECIST 1.1 or a confirmed 50% or greater decline in serum PSA (PSA50, a prostate cancer blood-marker drop used as an early response signal), measured up to 3 years. The three-drug combination is aggressive: enzalutamide continues to pressure AR signaling, ZEN-3694 disrupts AR-V7-driven bypass transcription, and pembrolizumab attempts to convert an immunologically cold tumor hot. The scientific bet is that BET inhibition remodels the tumor microenvironment enough to unlock checkpoint response in a cancer where PD-1 monotherapy has essentially failed. Design concerns are real. Triple combinations in mCRPC have historically struggled to show a clean efficacy signal attributable to any single agent, especially without a randomized comparator arm. The single-arm composite endpoint of ORR or PSA50 sets a low bar for a positive signal and will not distinguish ZEN-3694 contribution from enzalutamide-plus-pembrolizumab activity. A cleaner adjudication would require a randomized comparator on rPFS (radiographic progression-free survival, which measures how long until scans show tumor growth). The parallel NCI trials in TNBC, CRC, and lung are dose-finding or single-arm Phase 2, appropriate for a small compound but insufficient for registration.

Our model estimates a 8% chance this drug is eventually approved. It starts from the historical base rate for Phase 2 drugs in this area (about 13%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by a non-randomized design and its light or open-label blinding; it is held back by 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 largest single failure mode. BET inhibitors have consistently produced tumor shrinkage in a minority of patients across indications but have not translated that into randomized survival benefit outside myelofibrosis. Without a biomarker that reliably identifies BET-dependent tumors, the class keeps sputtering in Phase 2 single-arm data and disappointing in Phase 3. Safety risk is mechanism-based: BET inhibition causes dose-limiting thrombocytopenia because megakaryocyte differentiation depends on BET-regulated transcription, and this has forced dose reductions and holidays across every clinical-stage BET program to date. GI toxicity and fatigue also cluster around BET dosing, and the pelabresib leukemic-transformation signal in MANIFEST-2 is a new class-level flag regulators will now look for [10]. The three-drug prostate combination adds enzalutamide fatigue and pembrolizumab immune adverse events on top. Execution risk is elevated because Zenith is a small privately held company running lean, with much of the current trial matrix funded by NCI rather than internal capital. That is efficient but fragile: a private financing shortfall or a strategic reprioritization could stall the program even if data are clean. Commercial risk assuming approval is severe. mCRPC affects roughly 40,000 to 50,000 US patients per year, and the market is crowded with generic enzalutamide and abiraterone on the horizon, enzalutamide currently priced near $18,000 per month, PARP inhibitors owning HRR-mutant patients (tumors with broken homologous recombination DNA repair genes that make them sensitive to PARP inhibitors), and Pluvicto (a lutetium-177 radioligand) owning PSMA-positive disease (tumors expressing prostate-specific membrane antigen on the cell surface) at roughly $42,000 per cycle. A three-drug regimen would need to beat sequential monotherapy on both efficacy and cost, and payers will scrutinize incremental benefit hard. Patent protection status for ZEN-3694 has not been publicly disclosed in detail; composition-of-matter coverage typically dates from the original 2015 to 2016 filings, which would put nominal expiry in the mid-2030s absent extensions.

Worth watching, with skepticism. ZEN-3694 is the last credible BET asset in solid tumors with an active development program that keeps generating investigator interest, and pelabresib's Phase 3 hit reframes the class as mechanistically validated but regulator-scrutinized rather than dead. The signal to look for is not another single-arm ORR number. It is a randomized readout, even a small one, where the ZEN-3694 arm beats a control arm on rPFS (radiographic progression-free survival) or duration of response in a defined patient subset such as AR-V7 positive mCRPC. Absent randomization, another Phase 2 update in the enzalutamide combo becomes noise in the pattern the BET class has already established. The parallel TNBC and talazoparib trials are the more interesting scientific bets because the synthetic-lethal rationale with PARP inhibition and the immune-remodeling rationale in TNBC are cleaner than the mCRPC transcription-addiction story. Check back at ASCO GU 2027 for mCRPC updates and at ASCO 2027 for the ComBET talazoparib readout (NCT05327010), which is already active-not-recruiting and should mature. If either produces a clean randomized signal, this becomes a real asset. If both produce more Phase 2 tea leaves, the solid-tumor BET thesis is functionally dead, regardless of what happens with pelabresib in myelofibrosis.