Sitravatinib

Sitravatinib (MGCD516) is an oral multi-kinase inhibitor originated by Mirati Therapeutics, which Bristol Myers Squibb acquired in January 2024 primarily for the KRAS G12C drug adagrasib, not for this asset [13]. Its pivotal Phase 3 SAPPHIRE trial in previously treated non-squamous non-small cell lung cancer (NSCLC) read out negative at ESMO 2023: the sitravatinib plus nivolumab combination missed overall survival with a median OS of 12.2 vs 10.6 months for docetaxel (hazard ratio 0.86, 95% CI 0.70-1.05, p=0.144), and progression-free survival numerically favored docetaxel (4.4 vs 5.4 months, HR 1.08, p=0.452) [1]. That is a clean null, not a near-miss, and it undercuts the entire TAM plus checkpoint hypothesis in the tumor with the strongest preclinical rationale. What remains active is a set of Phase 2 combinations with BeiGene's PD-1 antibody tislelizumab, sponsored largely out of China and Korea, in triple-negative breast cancer (TNBC), biliary tract cancer, small-cell lung cancer (SCLC), and uveal melanoma [2][3][4][5]. For a Western investor, the commercial thesis is effectively closed. For the drug as a molecule, the question is whether any of the tislelizumab combination signals are strong enough to justify a regional filing in China with the NMPA (National Medical Products Administration).

The drug is investigational, never approved, and the lead Phase 3 program (SAPPHIRE, NCT03906071, n=577) failed on OS in 2023 [1]. BMS has not initiated a replacement Phase 3 since the acquisition closed. The BioCosm node currently anchors to NCT04904302, an M.D. Anderson investigator-initiated Phase 2 in renal cell carcinoma that is listed as terminated on ClinicalTrials.gov; the more defensible anchor for a status conversation is SAPPHIRE. No FDA breakthrough, fast track, orphan, or accelerated approval designations are on record for sitravatinib. The WHO-assigned ATC code L01EX26 reflects classification rather than approval [6]. Active late-phase activity is running through BeiGene (now trading as BeOne Medicines) under the tislelizumab combination umbrella protocol NCT04164199 [7], with disease-specific Phase 2 readouts published across 2025 and 2026 in TNBC [2], biliary tract cancer [3], SCLC maintenance [4], and post-checkpoint-inhibitor RCC [8]. None of these have registrational (i.e., Phase 3 designed to support marketing approval) anchors disclosed. The realistic near-term regulatory path is a China NMPA filing on a specific tumor if one of the tislelizumab cohorts shows differentiated activity, not a U.S. or EU submission. Composition-of-matter patent protection extends to at least 2029 in the U.S. before any regulatory extensions, per Mirati's last standalone 10-K, giving any potential China filer a workable exclusivity window [14].

Sitravatinib blocks a spread of receptor tyrosine kinases at once. The scientifically interesting subset is the TAM family: TYRO3, AXL, and MERTK. Think of TAM receptors as "stand down" signals cancer uses on nearby immune cells. Tumors secrete a protein called GAS6 that binds AXL and MERTK on macrophages, pushing those macrophages into a wound-healing, immunosuppressive mode instead of a tumor-killing mode [9]. Blocking TAM should, in theory, retrain the tumor microenvironment so that PD-1 antibodies like nivolumab or tislelizumab work again in patients who have progressed on checkpoint blockade. That is the entire rationale for the combination studies. Sitravatinib also hits VEGFR2, PDGFR, and KIT, which is standard blood-vessel and stromal kinase coverage similar to cabozantinib and lenvatinib. The mechanism has real preclinical support but weak clinical validation: no TAM inhibitor is approved for any oncology indication, and the SAPPHIRE failure is the largest randomized test of the "reverse ICI resistance" hypothesis to date. Bemcentinib (BerGenBio), the most direct AXL-selective competitor still in clinical development, has also had a rocky path with mixed NSCLC readouts and no approval, which further weakens class-level enthusiasm. Cabozantinib remains the closest approved multi-kinase comparator, but its efficacy is generally attributed to VEGFR2 and MET, not TAM. So the drug's biological differentiation from cheaper multi-kinase generics is unproven.

SAPPHIRE (NCT03906071) is the trial that matters for reading this program's fate. It was a Phase 3 open-label study in advanced or metastatic non-squamous NSCLC after prior checkpoint inhibitor and platinum chemotherapy, randomizing 577 patients to sitravatinib 120 mg daily plus nivolumab 240 mg every two weeks versus docetaxel [1]. Squamous histology was excluded because the SAPPHIRE hypothesis leaned on VEGFR-driven biology where non-squamous tumors have the clearest signal, and because docetaxel-based comparators in squamous NSCLC face different competitor regimens (necitumumab, ramucirumab combinations) that would have complicated the control arm. Primary endpoint was overall survival. At ESMO 2023, sponsor Mirati reported median OS 12.2 vs 10.6 months (HR 0.86, 95% CI 0.70-1.05, p=0.144), median PFS 4.4 vs 5.4 months (HR 1.08, p=0.452), and ORR (objective response rate, the fraction of patients whose tumor visibly shrinks by RECIST criteria) 16% vs 17% (p=0.597) [1]. No confirmatory Phase 3 has been launched. The BeiGene umbrella (NCT04164199, n=404, completed) tested tislelizumab plus sitravatinib, pamiparib, or other partners across multiple tumors as a Phase 2 signal-seeking platform, with safety as the anchor endpoint rather than efficacy [7]. Disease-specific reports include the SPARK Phase 2 in metastatic TNBC (single-arm, n=67 across two dose cohorts) with chemo-free ORRs of 38.1% (70 mg) and 50.0% (100 mg) and median PFS of 8.2 and 5.4 months respectively, plus a chemo-triplet cohort (adding nab-paclitaxel) reporting ORR 75.7% [2]; a Phase 2 in second-line biliary tract cancer (n=43) with disease control rate 69.2% at median follow-up 10.5 months, ORR not fully disclosed in the abstract [3]; two Phase 2 maintenance studies in extensive-stage SCLC [4]; a Phase 2 in uveal melanoma with liver metastases (NCT05542342, n=16) [5]; and a Phase 2 in post-ICI clear cell RCC [8]. All are small, single-arm, and not powered for registration. The Phase 1 nivolumab plus ipilimumab plus sitravatinib triplet in RCC published in Nature Communications is a mechanism study, not a pivotal design [11].

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 one and has already partly materialized. SAPPHIRE tested the exact biological hypothesis (TAM plus VEGFR blockade rescues checkpoint resistance) in the tumor with the strongest preclinical rationale, and the OS HR of 0.86 with a wide CI crossing 1.0 (0.70-1.05) is consistent with essentially no treatment effect rather than an underpowered near-miss [1]. That is a hypothesis failure, not a dose or design failure, and it applies to every downstream indication that leans on the same mechanism. Safety risk is manageable but not trivial: multi-kinase inhibitors hitting VEGFR2 predictably cause hypertension, proteinuria, hand-foot syndrome, and fatigue, and combination with PD-1 layers on immune-related adverse events. The published Phase 2s report treatment-related AE rates consistent with that class profile [2][8]. Execution risk sits mostly with the sponsor question. BMS acquired Mirati in January 2024 for adagrasib and the KRAS franchise, and its 2025 10-K does not highlight sitravatinib as a growth asset [12][13]. Whether BeiGene/BeOne funds a randomized Phase 3 on its own is the pivotal execution unknown, and public earnings commentary has not called out sitravatinib as a top-tier pipeline priority. Commercial risk is severe even in a positive scenario: any indication where sitravatinib might work is already crowded with cheaper generic multi-kinase competitors (cabozantinib, lenvatinib, regorafenib) that payers know how to price, and a novel TAM inhibitor without a differentiated biomarker will struggle to justify premium pricing. Composition-of-matter patent runs to at least 2029 before extensions [14], which is tight for any indication requiring a fresh randomized confirmatory trial.

This is noise for a Western pharma portfolio and a small regional watch item for anyone tracking BeiGene/BeOne. The pivotal readout that mattered already happened and was a clean null, not a near-miss [1]. Market context for the residual indications: biliary tract cancer is roughly 10K U.S. cases per year with no established second-line standard, so a positive readout could support a China filing plus a niche U.S. play; TNBC is roughly 40K U.S. cases per year but crowded with sacituzumab govitecan, trastuzumab deruxtecan in HER2-low, and multiple PD-L1 combinations, so the commercial ceiling for a new multi-kinase plus PD-1 combo is lower despite the larger patient pool. The specific data point that would flip this back to signal is a randomized Phase 3 initiation of sitravatinib plus tislelizumab in a defined tumor, most plausibly biliary tract cancer or TNBC where the recent Phase 2 papers show ORRs above what tislelizumab monotherapy delivers [2][3]. Absent that, the drug is a discontinued Mirati asset sitting inside a BMS pipeline that has bigger fish. Catalyst calendar for the remainder of 2026: ASCO 2026 (May 29 - June 2, Chicago) has likely already passed by the time this writeup is read, so check the published abstracts; ESMO 2026 (October 17-21, Berlin) is the next major oncology congress; SABCS 2026 (San Antonio, December 8-11) is the venue to watch for updated SPARK TNBC data; BeOne Medicines Q3 2026 earnings (~early November) is where a Phase 3 commitment would first be flagged. If none of those trigger a Phase 3 announcement, the asset is functionally dead. The audit-flagged issue on this node (base-rate PoS overwriting the manual Phase 3 failure override) is a systems bug worth fixing so the platform stops presenting a 33% probability on an asset whose pivotal trial has already failed.