Darovasertib Combo Doubles PFS in HLA-A2-Negative Uveal Melanoma

IDEAYA Biosciences and Servier presented complete Phase 2/3 OptimUM-02 results at the 2026 American Society of Clinical Oncology Annual Meeting in Chicago on Monday morning, showing that darovasertib combined with crizotinib more than doubled median progression-free survival compared with investigator's choice of therapy in first-line HLA-A*02:01-negative metastatic uveal melanoma. The combination produced a median PFS of 6.9 months versus 3.1 months for control, with a hazard ratio of 0.42 (95% CI 0.30–0.59; p<0.0001).

Confirmed overall response rate by blinded independent central review was 37.1% in the darovasertib arm against 5.8% for control (p<0.0001), with 5 complete responses recorded in the combination group. The companies also reported an early trend toward improved overall survival, though the OS dataset remains immature. The most frequent grade 3 or higher treatment-emergent adverse events were syncope, hypertension, and diarrhea, which the sponsors characterized as a manageable safety profile. IDEAYA and Servier intend to file a U.S. NDA in the second half of 2026 in support of accelerated approval.

Why Uveal Melanoma Has Resisted Standard Melanoma Therapies

Uveal melanoma is the most common primary intraocular malignancy in adults, but it shares almost nothing biologically with cutaneous melanoma beyond its melanocyte origin. Where skin melanoma is driven by ultraviolet-induced BRAF and NRAS mutations and carries one of the highest tumor mutational burdens of any cancer, uveal melanoma arises from melanocytes in the choroid, ciliary body, or iris and is initiated in roughly 90% of cases by activating mutations in the G-protein α-subunit genes GNAQ or GNA11. Those mutations sit upstream of protein kinase C and the MAPK cascade, a signaling architecture that has not responded to BRAF inhibitors and that confers low neoantigen load — the likely reason anti-PD-1 monotherapy has produced response rates in the single digits in this disease.

The therapeutic gap has been particularly stark in first-line metastatic disease. Tebentafusp (Kimmtrak), the bispecific gp100×CD3 T-cell engager approved in 2022, demonstrated an overall survival benefit but is restricted to patients carrying the HLA-A*02:01 allele, which leaves roughly half of metastatic uveal melanoma patients globally without a biology-matched first-line option. For that population, investigator's choice has historically meant pembrolizumab, ipilimumab plus nivolumab, or dacarbazine-based chemotherapy — regimens with median PFS in the 2- to 3-month range and limited durable benefit.

How Does the Darovasertib Combination Actually Work?

Darovasertib is an oral protein kinase C inhibitor that targets the downstream effector of GNAQ/GNA11 driver mutations, while crizotinib is a multi-kinase inhibitor of ALK, ROS1, and MET. Preclinical work demonstrated that PKC inhibition alone activates compensatory MET signaling in uveal melanoma cells, and that combined PKC plus MET blockade produces synergistic tumor regression. The OptimUM-02 data is the first Phase 3-stage validation of that dual-pathway hypothesis in a randomized setting.

The mechanism matters because it positions darovasertib + crizotinib as the first targeted regimen in uveal melanoma matched to the disease's founding genetic event rather than borrowed from cutaneous melanoma practice. The doubling of median PFS and the 6.4-fold improvement in confirmed response rate suggest the GNAQ/GNA11 → PKC → MET axis is a durable therapeutic target rather than a transient one, a question that prior single-agent PKC inhibitor trials had not been able to answer.

When Could Darovasertib + Crizotinib Reach Patients?

IDEAYA confirmed an NDA submission to the U.S. Food and Drug Administration in H2 2026 under the accelerated approval pathway, which permits filing on the basis of PFS and response data while overall survival follow-up continues. Assuming a standard 10-month FDA review and no major signals from the agency, the combination could reach U.S. patients in mid-to-late 2027. Servier holds commercialization rights outside the United States and is expected to pursue parallel filings in the European Union and Japan. The data also strengthens the rationale for ongoing IDEAYA programs in earlier-stage disease, including the OptimUM-10 Phase 3 trial of neoadjuvant darovasertib in primary non-metastatic uveal melanoma.

For the broader melanoma research field, the OptimUM-02 result joins a wave of practice-shifting ASCO 2026 datasets — including RP1's 47.8% three-year survival in anti-PD-1-failed cutaneous melanoma presented earlier in the meeting — that collectively widen the targeted-therapy and immuno-oncology options for melanoma subtypes that had reached therapeutic plateaus. Set against the projected doubling of global skin cancer burden by 2050, the readout underscores how aggressively the field is closing biology-matched treatment gaps in melanoma's most lethal subtypes.

Sources: IDEAYA Biosciences ASCO 2026 LBA announcement; IDEAYA/Servier topline results press release; Croce et al., darovasertib mechanism review.