Sitravatinib

Sitravatinib (MGCD516) is a once-promising oral multi-kinase inhibitor from Mirati Therapeutics — now part of Bristol Myers Squibb [8][11] — that lost its lead Western indication when the Phase 3 SAPPHIRE trial in second-line NSCLC missed overall survival in 2023 [6]. The drug has found a second life in China, where investigator-initiated Phase 2 trials are pairing it with tislelizumab (BeiGene/BeOne's PD-1 inhibitor) across triple-negative breast cancer (SPARK, NCT04734262, n=67, final data published Nov 2025) [1], biliary tract cancer (NCT04727996, DCR 69.2% reported) [2], extensive-stage small-cell lung cancer maintenance [3], renal cell carcinoma [4], and uveal melanoma [10]. Whether any of these reach registration, or whether BMS commits real capital to the asset post-acquisition, is the open question.

This is not a novel compound. Sitravatinib has been in clinical development since 2014. The Phase 3 SAPPHIRE trial (NCT03906071, n=577) in pretreated non-squamous NSCLC after immune checkpoint inhibitor failure read out negative in 2023: sitravatinib plus nivolumab did not beat docetaxel on overall survival [6]. Mirati was acquired by Bristol Myers Squibb for $4.8 billion (closed January 2024) primarily for KRAS G12C drug adagrasib, not sitravatinib [8][11]. The current 'Phase 3' status in our database is anchored to ChEMBL's max_phase=3 from the SAPPHIRE program, but no active Phase 3 sitravatinib trial is enrolling in the West. Active development is concentrated in Phase 2 academic studies in China testing combinations with tislelizumab across TNBC (SPARK, NCT04734262) [1], biliary tract cancer (NCT04727996) [2], extensive-stage SCLC maintenance [3], renal cell carcinoma [4], and uveal melanoma (NCT05542342) [10]. NCT04904302 — the MDACC clear cell RCC investigator trial originally anchored to this node — is terminated [9]. The drug carries no active FDA breakthrough, fast track, or orphan designations for any current indication. RxNorm CUI 1888907 and UNII CWG62Q1VTB confirm the molecule identity.

Sitravatinib blocks a bundle of kinases at once. The headline targets are the TAM family — TYRO3, AXL, MERTK — three closely related receptors tumors exploit to keep the immune system quiet. The clearest biology is on MERTK: on tumor-associated macrophages, MERTK recognizes phosphatidylserine externalized on dying tumor cells via Gas6 and Protein S bridges, and signals that the cell is routine cleanup — efferocytosis. That signal actively suppresses pro-inflammatory cytokine output and pushes macrophages into a tolerogenic phenotype. AXL extends the same theme on tumor-associated macrophages and additionally drives epithelial–mesenchymal transition and acquired therapy resistance on tumor cells themselves [13]. AXL and MERTK on tumor-associated macrophages recognize dying tumor cells and respond by suppressing — rather than amplifying — the inflammatory alarm; this keeps the immune system quiet when it should be attacking. Blocking AXL and MERTK should, in theory, convert 'cold' tumors that don't respond to checkpoint inhibitors into 'hot' ones. Note: this is distinct from CD47–SIRPα ('don't eat me') signaling, which is the phagocytic checkpoint and a different axis. Sitravatinib also hits VEGFR2 (the blood-vessel growth receptor that lenvatinib, sunitinib, and cabozantinib target), plus PDGFR and KIT, giving it overlapping anti-angiogenic activity. The combination logic with PD-1 inhibitors is biologically reasonable: shut down the immunosuppressive tumor microenvironment with TAM/VEGFR blockade, then unleash T cells with the checkpoint drug. The mechanism is plausible but the validation is weak: no AXL-selective inhibitor has been approved anywhere, and clinical proof that TAM inhibition meaningfully potentiates PD-1 response remains elusive after a decade of trying. Bemcentinib (BerGenBio) failed to deliver registrational signal in NSCLC (BGBC008, NCT03184571) [12] and the company subsequently discontinued its 1L STK11-mutant NSCLC program and announced exploration of strategic alternatives. AVB-S6-500/batiraxcept (Aravive) failed in platinum-resistant ovarian cancer. SAPPHIRE failed in NSCLC [6]. Independent target validation for AXL in oncology remains weak relative to its EMT/fibrotic and neurodegenerative biology — but the specific Open Targets evidence scores cited in earlier drafts were not directly verifiable here and have been removed pending direct citation.

The most relevant trial is the failed Phase 3: NCT03906071 (SAPPHIRE), randomized open-label, n=577, sitravatinib plus nivolumab vs docetaxel in non-squamous NSCLC patients who progressed after platinum chemotherapy and a checkpoint inhibitor [6]. Primary endpoint was overall survival. The trial completed enrollment and read out negative in 2023. The only remaining 'Phase 3' record involving sitravatinib is NCT04164199, a BeOne (formerly BeiGene) umbrella Phase 3 study of tislelizumab plus pamiparib plus other agents including sitravatinib in advanced malignancies, n=404, active but not recruiting [7]. The primary endpoint is incidence of immune-mediated adverse events: this is a registration efficacy trial only by name. It is a basket safety study. Active Phase 2 Chinese investigator trials with tislelizumab provide the meaningful clinical signal: SPARK in TNBC (NCT04734262, n=67, multi-cohort single-arm, data cutoff March 2023, full publication Liu et al. Mol Cancer Nov 2025; updated triplet readout at AACR 2025) [1]; biliary tract cancer second-line (NCT04727996, single-arm Phase 2, DCR 69.2% in ITT after 10.5 months median follow-up; HRD subset had better outcomes) [2]; two SCLC maintenance studies [3]; and NCT05542342 in uveal melanoma with liver metastases (n=16) [10]. NCT04904302 at MD Anderson — listed in our node — is terminated [9]. None of these trials are powered or designed for FDA registration. Single-arm Phase 2s without contemporary controls in heavily pretreated populations make signal interpretation difficult, and none use AXL/TAM biomarker enrichment.

Our model estimates a 23% 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 a non-randomized design; it is held back by smaller-than-typical enrollment for this phase, 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 failure mode and has already materialized once. SAPPHIRE missed OS in the exact patient population (post-checkpoint NSCLC) where the TAM/PD-1 combination hypothesis was strongest [6]. That negative readout suggests either TAM inhibition doesn't reverse checkpoint resistance in unselected patients, or sitravatinib's specific kinase profile generates enough off-target toxicity to prevent effective combination dosing. Safety: multi-kinase inhibitors with VEGFR2 activity carry standard liabilities including hypertension, hand-foot syndrome, proteinuria, fatigue, and bleeding, all of which limit combinability with chemo or immunotherapy. PDGFR and KIT activity add cytopenia risk. None of this is novel territory, but it caps the therapeutic window for combination dosing where most of the active trials live. Execution risk: the active trials are small Chinese single-arm Phase 2s, none powered for registration, and BMS has shown no public commitment to running a Western Phase 3. Commercial risk: even if a Chinese trial generates a striking signal, sitravatinib would enter markets where multi-kinase inhibitors (lenvatinib, cabozantinib, regorafenib) and PD-1 combinations are already entrenched. Lenvatinib (Lenvima/Eisai) loses meaningful US exclusivity around July 1, 2030 under settlement agreements with Sun Pharma and Dr. Reddy's, with full patent coverage extending to February 2036 [14] — so 'imminent' generic erosion is overstated, but generic lenvatinib will likely be on market before any plausible sitravatinib launch. In biliary tract specifically, the competitive frame is harder than 'unmet need' suggests: pemigatinib (Pemazyre, FGFR2 fusion+) and futibatinib (Lytgobi, FGFR2 fusion+) are approved in second-line cholangiocarcinoma; ivosidenib (Tibsovo, IDH1-mutant) is approved in the same setting; sitravatinib's unselected, biomarker-agnostic approach has to outperform biomarker-enriched standards of care to find a niche [15]. There is no obvious clinical niche where sitravatinib beats existing options on efficacy or tolerability, and no companion diagnostic strategy to carve one out.

Mostly noise. BMS paid $4.8 billion for Mirati to acquire adagrasib (KRAS G12C); sitravatinib was a secondary asset that became a failed asset when SAPPHIRE missed [8][11]. The drug is alive only because Chinese investigators are running cheap Phase 2 combination studies with BeOne's tislelizumab, and those studies are not designed to support registration anywhere. **Key catalysts:** SPARK (TNBC) full data already published — Liu et al. Mol Cancer Nov 2025, with an updated triplet readout at AACR 2025 (PS3-01) [1]; readers should consult the publication directly for the ORR/PFS numbers (we are not reproducing them secondhand). Biliary tract Phase 2 (NCT04727996) preliminary signal already in print: DCR 69.2% ITT, HRD-subset enrichment [2] — encouraging but DCR (not ORR) in a single-arm study against biomarker-selected approved alternatives is not registration-grade evidence. Watch for whether BeiGene/BeOne or any Chinese sponsor advances any indication to a controlled Phase 3 — that's the binary question for asset survival. If SPARK or biliary tract follow-up shows an objective response rate north of 30% in a chemo-refractory, biomarker-unselected population that would be a real signal because both indications have brutal unmet need. Absent that, this is a dead program walking. Don't expect BMS to fund a Western Phase 3: adagrasib is the post-Mirati oncology priority, and sitravatinib doesn't fit the post-acquisition portfolio per BMS's 2025 10-K disclosures [8]. The broader read-through — TAM/AXL inhibition has now repeatedly failed to deliver clinically meaningful checkpoint potentiation across bemcentinib [12], batiraxcept, and sitravatinib — is the real signal for portfolio managers evaluating other AXL programs.