Ifinatamab deruxtecan

Ifinatamab deruxtecan (I-DXd, MK-2400) is a B7-H3-directed antibody-drug conjugate from Daiichi Sankyo and Merck, now in Phase 3 (NCT06925737, IDeate-Prostate01) for metastatic castration-resistant prostate cancer with overall survival as the primary endpoint, a 1,440-patient enrollment target, and docetaxel plus prednisone as the active comparator [1]. It pairs an antibody against B7-H3, a surface protein turned up on many solid tumors, with deruxtecan, the same topoisomerase I inhibitor payload that powers Enhertu. The bet is whether Enhertu-style ADC chemistry can crack a target that has frustrated every prior B7-H3 program.

This is a novel compound and the first Phase 3 trial against B7-H3 in prostate cancer. The mCRPC readout from NCT06925737 (IDeate-Prostate01) will not come fast: 1,440 patients on an OS primary endpoint typically means four to five years from a 2025 enrollment start, putting mature data in the 2028 to 2029 window. The IDeate-Prostate01 trial-in-progress abstract was presented at ASCO GU 2026 but no patient data was reported [11]. A parallel Phase 1b combination trial, IDeate-Prostate02 (NCT06863272), is recruiting 360 patients to test I-DXd alongside other agents in mCRPC [2]. Outside prostate, the molecule has its own Phase 3 in second-line small cell lung cancer (IDeate-Lung02) [3] and a Phase 1 SCLC combination with gocatamig (NCT07227597). The pan-tumor Phase 1/2 dose-escalation paper (IDeate-PanTumor01) appeared in Lancet Oncology 2026: maximum tolerated dose was not reached, antitumor activity was seen across multiple solid tumors, and one treatment-related interstitial lung disease death was reported [4]. Earlier IDeate-PanTumor01 disclosures from 2022 ESMO had shown a 33% confirmed objective response rate in the heavily pretreated mCRPC cohort (n = 54), the strongest prostate-specific signal available for this molecule to date. The Phase 2 SCLC IDeate-Lung01 primary analysis (JCO 2026) was the highest-profile readout of 2025-2026: at the selected 12 mg/kg dose (n = 137), confirmed objective response rate was 48.2% (95% CI, 39.6-56.9) with a median duration of response of 5.3 months [5]. On the back of those results, the FDA granted I-DXd Breakthrough Therapy Designation and Priority Review for previously treated extensive-stage SCLC [12]. No FDA breakthrough or fast track designations have been disclosed specifically for the mCRPC program. Merck took over commercial rights in 2023 through a roughly $22 billion Daiichi Sankyo ADC deal covering three molecules; ifinatamab deruxtecan is the one with the broadest solid-tumor opportunity given B7-H3 expression across prostate, lung, and head and neck cancers [6].

The drug works like a guided missile. The antibody half locks onto B7-H3 (also called CD276), a cell-surface protein found at high levels on many solid tumors including prostate, lung, breast, and head and neck cancers, while staying mostly absent on healthy tissue. Prostate cancer in particular shows among the highest B7-H3 transcript levels of any solid tumor type, and B7-H3-high tumors associate with worse survival [13]. When the antibody binds, the cancer cell internalizes the whole construct, enzymes inside the cell cleave the linker, and deruxtecan is released. Deruxtecan is a topoisomerase I inhibitor: it traps the enzyme that uncoils DNA during replication, causing DNA breaks and cell death. The payload is the same molecule used in Enhertu (trastuzumab deruxtecan), the HER2 ADC that generated approximately $4.98 billion in combined global sales for AstraZeneca and Daiichi Sankyo in FY 2025 (up from $3.75 billion in FY 2024) and rewrote treatment for HER2-positive and HER2-low breast cancer [7]. Deruxtecan has a drug-to-antibody ratio of about 8 and a bystander effect, meaning the released payload can also kill neighboring tumor cells that lack the target, which matters when target expression is patchy [8]. B7-H3 has also been proposed to suppress immune responses, though the receptor it signals through is still debated (TIM-3 has been proposed but not confirmed). If that immune biology is real, hitting B7-H3 with an ADC might deliver both cytotoxic payload and a secondary immune benefit, but this remains speculative. The mechanism is well validated as a delivery system. What is still being established is whether B7-H3 itself is productive enough as a target to drive responses across tumor types [9].

NCT06925737 (IDeate-Prostate01, MK-2400-001) is a Phase 3 randomized, open-label trial enrolling roughly 1,440 men with metastatic castration-resistant prostate cancer who have progressed after androgen receptor pathway inhibitors [1]. Patients are randomized to I-DXd 12 mg/kg every three weeks versus docetaxel 75 mg/m² every three weeks plus prednisone 10 mg daily. Primary endpoint is overall survival. The size, OS endpoint, and active comparator point to a registration-quality design where a positive result would support full approval rather than accelerated. Recruitment is open across multiple regions, with Merck Sharp and Dohme as sponsor. The patient population fits an ADC swing: late-line mCRPC patients have few options after enzalutamide, abiraterone, docetaxel, and cabazitaxel, and the current alternatives (Pluvicto for PSMA-positive disease, olaparib for BRCA-mutant subsets where the tumor carries inherited or acquired mutations in the BRCA1 or BRCA2 DNA-repair genes) only reach defined subgroups. Two design concerns stand out. First, there is no biomarker selection at entry: patients are enrolled regardless of B7-H3 expression level, which raises the bar on efficacy because the trial has to win on an unselected population. The protocol does plan exploratory B7-H3 expression analyses, which will matter for label language even if the primary win covers everyone. Second, docetaxel as the comparator is a high bar in patients who may already have had a taxane, and the open-label design introduces ascertainment bias on subjective endpoints though OS is unbiased. The OS endpoint is also unforgiving: ADCs with strong response rates have sometimes failed OS in mCRPC because post-progression therapies wash out the survival signal. A response-rate-plus-PFS co-primary would have been a softer landing.

Our model estimates a 32% chance this drug is eventually approved. It starts from the historical base rate for Phase 3 drugs in this area (about 48%), 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, its light or open-label blinding, and more secondary endpoints than usual; it is held back by 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 central one. B7-H3 has not been validated as a productive cancer target by any approved therapy. Enoblituzumab (MGA271), a non-conjugated B7-H3 antibody from Macrogenics, posted modest single-agent activity across solid tumors and did not advance to registration. Vobramitamab duocarmazine (MGC018), another B7-H3 ADC with a different payload (duocarmycin), was paused in mCRPC after safety concerns. Ifinatamab deruxtecan is betting that the deruxtecan payload, which has carried weak antibodies to wins before, is potent enough to overcome target ambiguity. The 33% ORR in 2022 ESMO mCRPC dose-escalation cohorts is encouraging but came from a small (n = 54), heavily pretreated subset and at heterogeneous doses; the unselected Phase 3 population at the registration 12 mg/kg dose may read differently. If the response rate in unselected mCRPC patients lands below 15 to 20%, the OS endpoint becomes very hard to hit. Safety risk concentrates on interstitial lung disease (ILD), the signature toxicity of deruxtecan ADCs. Enhertu carries a boxed warning for ILD (the FDA's strongest safety alert, printed prominently on the drug label) with cases ranging from grade 1 to fatal. In IDeate-Lung01 at the 12 mg/kg registration dose, adjudicated ILD occurred in 12.4% of patients overall, with grade ≥3 ILD in 4.4% and grade 5 (fatal) events in 1.5% [5]. The IDeate-PanTumor01 dose-escalation report also documented a treatment-related ILD death [4]. In an older mCRPC population with cardiopulmonary comorbidities, this risk is amplified and could constrain registration doses. Other toxicities to track: neutropenia, nausea, alopecia, and the standard ADC GI profile. Execution risk: 1,440 patients is a large enrollment ask, and the trial competes with Pluvicto expansion studies and other late-line mCRPC trials for the same patient pool. Commercial risk: even with a positive readout, payers will compare it to Pluvicto, which has biomarker selection via PSMA PET imaging (a scan that uses a radioactive tracer binding to PSMA, a protein on prostate cancer cells, to identify patients whose tumors will respond to PSMA-targeted radioligand therapy). Pluvicto carries a wholesale acquisition cost of roughly $42,500 per dose with up to six cycles per course (over $250,000 per full course before discounts), so I-DXd will need to defend pricing in that same band and show meaningful OS benefit versus docetaxel [14].

Worth watching. The signal that matters most before the Phase 3 OS readout (2028 to 2029, per typical timelines for a 1,440-patient OS endpoint) is a prostate-specific objective response rate (ORR, the fraction of patients whose tumors shrink meaningfully on imaging) from IDeate-PanTumor01 cohorts or from IDeate-Prostate02. A clean ORR above 25% in pure mCRPC patients would meaningfully shift the success probability; below 15% would tell you the OS endpoint is in trouble. ASCO 2026 has now passed and produced no new IDeate-PanTumor01 prostate cohort data; the headline I-DXd readouts at the meeting remained the SCLC IDeate-Lung01 primary analysis and the IDeate-Prostate01 trial-in-progress slot at ASCO GU 2026. Next concrete catalysts are ESMO 2026 (October 2026) and ASCO GU 2027 (February 2027) for any prostate cohort updates or correlative biomarker data. Also watch the competitive B7-H3 field: ZL-1310 from MediLink/Zai Lab is advancing in small cell lung cancer with a different B7-H3 ADC payload, and any positive readout there validates the target broadly. For Merck, this is part of a roughly $22 billion deal to diversify beyond Keytruda, which loses US patent exclusivity starting in 2028 [6]. Ifinatamab deruxtecan is one of three Daiichi ADCs Merck licensed alongside Datroway (datopotamab deruxtecan) and raludotatug deruxtecan; the prostate readout is the highest single-asset revenue opportunity given Pluvicto-scale commercial potential. Pluvicto generated approximately $605 million in Q4 2025 alone and Novartis is guiding toward $5+ billion in peak sales [14], setting the commercial reference point a winning I-DXd label would be measured against. For Daiichi, this extends the deruxtecan platform into a third potential blockbuster franchise after HER2 (Enhertu) and TROP2 (Datroway). An mCRPC win would also unlock label expansion into earlier prostate lines and combination biology with AR pathway inhibitors. Next check-in: Q3 2026 Merck/Daiichi earnings calls for enrollment pace commentary on IDeate-Prostate01, then ESMO 2026 in October 2026 for any IDeate-PanTumor01 prostate cohort or IDeate-Prostate02 early signal.