A closely watched clinical trial testing a new two-drug combination against metastatic uveal melanoma, the deadliest form of eye cancer, has met its primary endpoint, IDEAYA Biosciences and partner Servier announced in spring 2025. The phase 2/3 OptimUM-02 study found that pairing darovasertib (IDE196) with crizotinib as a first-line treatment improved progression-free survival compared with physician-selected standard therapies in patients whose tumors had spread beyond the eye.
“We are pleased with the positive outcome of OptimUM-02, which reinforces the potential of darovasertib in combination with crizotinib to address a critical unmet need for patients with metastatic uveal melanoma,” IDEAYA stated in its corporate press release announcing the top-line results.
The result clears a critical hurdle for the companies, which have said they plan to use the data to support a regulatory filing with the U.S. Food and Drug Administration. If approved, the combination would become the first targeted therapy available to roughly half of all metastatic uveal melanoma patients, those who are HLA-A*02:01-negative and therefore ineligible for tebentafusp (Kimmtrak), the only drug currently approved for this disease.
Why this cancer is so hard to treat
Uveal melanoma starts in the pigmented cells of the eye and is the most common primary intraocular cancer in adults, with about 2,000 new cases diagnosed in the United States each year. The primary tumor can often be controlled with radiation or surgery, but up to half of patients eventually develop metastases, most frequently in the liver. Once the cancer spreads, outcomes are stark: median overall survival has historically hovered around 10 to 13 months, according to published analyses in peer-reviewed oncology literature.
Unlike cutaneous melanoma, which is commonly driven by BRAF mutations and responds to checkpoint immunotherapy, uveal melanoma is fueled by mutations in the GNAQ and GNA11 genes. That distinct molecular wiring explains why drugs that transformed skin melanoma treatment have largely failed in the eye cancer setting. Tebentafusp, approved by the FDA in January 2022, was a breakthrough, but it works only in patients who carry the HLA-A*02:01 tissue marker, roughly half the population. The other half has been left cycling through chemotherapy regimens and off-label immunotherapies with limited benefit.
What OptimUM-02 tested
The trial, registered as NCT05987332 on the federal ClinicalTrials.gov database, enrolled HLA-A*02:01-negative patients with metastatic uveal melanoma who had not received prior systemic treatment for advanced disease. It used an open-label design, meaning both patients and doctors knew which treatment each participant received.
Patients in the experimental arm took darovasertib, a protein kinase C (PKC) inhibitor designed to block a signaling cascade that is hyperactive in GNAQ/GNA11-mutant tumors, alongside crizotinib, a MET/ALK inhibitor. Crizotinib was originally approved for certain lung cancers but is now off-patent and is used off-label in this trial setting, a distinction that matters for cost and access because generic versions of the drug are available. The scientific rationale rests on preclinical work showing that targeting both the PKC pathway and the MET receptor can overcome resistance mechanisms that blunt the effect of either drug alone. The comparator arm allowed investigators to choose from available standard options, reflecting the reality that no single regimen is considered a definitive standard of care for this patient group.
The primary endpoint was progression-free survival, a measure of how long patients live without their cancer worsening.
What the data shows, and what it does not
IDEAYA and Servier have stated that the trial met its primary endpoint, indicating a statistically significant improvement in progression-free survival for the darovasertib-crizotinib arm over the comparator. The companies disclosed this result through a corporate press release. Because publicly traded companies face securities-law liability for materially misleading statements about clinical results, that top-line claim carries legal weight.
However, the granular numbers that oncologists and regulators need to fully evaluate the finding have not yet been made public. Key details still outstanding as of May 2026 include:
- The hazard ratio, median progression-free survival in each arm, p-values, and confidence intervals.
- Overall survival data, which will determine whether delaying tumor progression translates into patients living longer.
- Detailed safety results. Both drugs carry known toxicities. Crizotinib, in particular, is associated with liver enzyme elevations, visual disturbances, and gastrointestinal side effects, concerns that are amplified in a patient population where liver metastases are common. Because crizotinib is off-patent and used off-label here, its generic availability could help contain costs for the combination regimen, but insurers may scrutinize coverage for an off-label indication.
- Subgroup analyses showing how outcomes varied by age, extent of liver involvement, or geographic region.
- Patient-reported quality-of-life data, which increasingly influences both FDA decisions and real-world prescribing.
Until the full dataset is presented at a major oncology conference or published in a peer-reviewed journal, independent experts cannot compare the combination’s benefit directly against existing options or assess whether the effect size is clinically meaningful beyond statistical significance.
The road to a potential approval
Positive top-line results are a necessary but not sufficient step toward FDA approval. The companies must compile a complete data package, hold pre-submission meetings with the agency, and file a New Drug Application or supplemental application. The FDA then typically takes 10 to 12 months for a standard review, or six months under a priority designation. Neither company has disclosed a specific filing date or confirmed whether they have requested or received any expedited review pathway, such as Breakthrough Therapy or Priority Review designation.
IDEAYA, a South San Francisco-based biotech focused on synthetic lethality and precision oncology, has built much of its pipeline around rare, genetically defined cancers. Servier, a privately held French pharmaceutical company, brings global commercial infrastructure and oncology experience, including its existing hematology portfolio. For both partners, a uveal melanoma approval would validate a strategy of targeting small but molecularly well-defined patient populations.
Investors have been watching the program closely. IDEAYA’s stock has been volatile around clinical milestones, and the OptimUM-02 readout is considered a pivotal catalyst for the company’s valuation. Broader market reaction will depend on the detailed data, which analysts expect to be presented at a medical meeting in the coming months.
How OptimUM-02 fits into the metastatic uveal melanoma treatment landscape
For HLA-A*02:01-negative patients with metastatic uveal melanoma, the OptimUM-02 result represents the most significant clinical advance directed specifically at their disease. If the data holds up under regulatory scrutiny and the combination reaches the market, it would offer a biologically rational first-line option where none currently exists.
But calibration matters. Metastatic uveal melanoma remains an aggressive disease, and even a statistically significant improvement in progression-free survival does not mean a cure. Most patients will eventually need additional lines of therapy. Practical questions about drug access, cost of a two-agent targeted regimen, and insurance coverage will also shape how widely the combination is used in practice.
Patients currently considering treatment options should discuss the OptimUM-02 results with their oncologists and check the ClinicalTrials.gov listing for updates as the full data package becomes available. The information announced so far is encouraging and grounded in a well-designed trial, but the complete picture will only emerge once regulators and independent researchers have examined every layer of the evidence.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
