Evicenter
MarkdownMedical informationP&T meetings
Darovasertib (IDE196)

First-line metastatic uveal melanoma, in combination with crizotinib

Also known as LXS196
Manufacturer
IDEAYA Biosciences
Launch
Not announced
Last updated
206 sources

Executive summary

Darovasertib efficacy and safety

6 evidence topics · 17 sources

Summary: OptimUM-02 design and populations

OptimUM-02 (NCT05987332) is a phase 2/3, multi-arm, multi-stage, open-label study in HLA-A∗02:01-negative participants with metastatic uveal melanoma who have had no prior systemic therapy in the metastatic or advanced setting. The phase 2a dose optimization stage randomized participants 2:2:1 to darovasertib 300 mg BID plus crizotinib 200 mg BID, darovasertib 200 mg BID plus crizotinib, or investigator's choice; in December 2024 IDEAYA announced the Independent Data Monitoring Committee recommendation of a move-forward dose and the completion of Part 2a dose optimization. In the phase 2b and phase 3 portions, participants were randomized 2:1 to darovasertib 300 mg twice daily plus crizotinib 200 mg twice daily or investigator's choice of pembrolizumab, ipilimumab plus nivolumab, or dacarbazine, stratified by baseline LDH and by investigator's choice of comparator treatment (ipilimumab plus nivolumab vs single-agent therapy). Treatment began with a 7-day run-in of darovasertib monotherapy before crizotinib was started. The primary endpoints are PFS by BICR per RECIST 1.1 in phase 2 and OS in phase 3, which IDEAYA states will be used to support a potential accelerated approval and full approval in the United States, respectively. The phase 2b analysis required 130 PFS events to detect a hazard ratio of 0.55 with 90% power (stratified log-rank test, 1-sided α of 2.5%), assuming a median PFS of 3 months in the control arm.

The full enrollment release reports 435 participants enrolled in December 2025, the annual report for 2025 reports 437, and the ClinicalTrials.gov record gives an estimated enrollment of 420. The primary analysis (data cutoff January 23, 2026) included an efficacy population of 313 participants, the intent-to-treat population of the phase 2b portion (210 darovasertib plus crizotinib; 103 investigator's choice). The ASCO 2026 abstract reports 338 participants who received at least one dose (safety population); the presentation reports safety populations of 239 participants treated with darovasertib 300 mg plus crizotinib 200 mg and 100 controls across phases 2a and 2b. Median follow-up was 7.4 months. In the control arm, 79 participants (76.7%) received ipilimumab plus nivolumab and 24 (23.3%) pembrolizumab, and none received dacarbazine; the topline release gives 78 (76%) and 25 (24%). In the combination and control arms, respectively, mean age was 61.1 and 61.5 years, 101 (48.1%) and 52 (50.5%) participants were female, 48 (22.9%) and 16 (15.5%) had ECOG performance status 1, 18 (8.6%) and 5 (4.9%) had a largest metastatic lesion of 8.1 cm or more, 127 (60.5%) and 65 (63.1%) had hepatic disease only, and 120 (57.1%) and 62 (60.2%) had baseline LDH at or below the ULN.

Summary: OptimUM-02 efficacy

At the January 23, 2026 data cutoff, the PFS analysis of the 313 participants was based on 159 events. Median PFS by BICR was 6.9 months (95% CI, 5.6 to 8.3) with darovasertib plus crizotinib and 3.1 months (95% CI, 1.8 to 4.2) with investigator's choice (hazard ratio 0.42; 95% CI, 0.30 to 0.59; p<0.0001), a 58% reduction in the risk of disease progression. By investigator assessment, median PFS was 6.7 months (95% CI, 5.6 to 8.2) and 2.7 months (95% CI, 1.7 to 4.1) (hazard ratio 0.36; 95% CI, 0.26 to 0.50).

The objective response rate (ORR) by BICR was 37.1% (78 of 210; 95% CI, 30.6 to 44.1) with darovasertib plus crizotinib and 5.8% (6 of 103; 95% CI, 2.2 to 12.3) with investigator's choice (odds ratio 10.8; 95% CI, 4.4 to 26.4), with five complete responses (2.4%) in the combination arm and none in the control arm. By investigator assessment, the ORR was 39.5% (83 of 210) and 1.9% (2 of 103). Median duration of response by BICR was 6.8 months (95% CI, 5.5 to 11.3) in the combination arm and not estimable in the control arm. The disease control rate (complete response, partial response, or stable disease for at least 12 weeks) by BICR was 73.3% (154 of 210) and 31.1% (32 of 103). Progressive disease was the best overall response by BICR in 13 participants (6.2%) and 29 participants (28.2%). As reported from the presentation, waterfall plots showed tumor reduction in “nearly 90% of patients” in the combination arm and “about 35% in the control arm”.

In the ESMO 2026 subgroup analysis (BICR), median PFS with darovasertib plus crizotinib and with control was 8.4 and 4.2 months with normal LDH, 5.7 and 1.8 months with LDH above the ULN and below 2 times the ULN, and 3.1 and 1.7 months with LDH 2 times the ULN or higher (n = 25 and 12); the corresponding ORRs were 46.7% and 8.1%, 27.7% and 3.4%, and 16.0% and 0%. Median PFS was 8.3 and 2.9 months with hepatic metastases, 4.9 and 2.8 months with hepatic and extrahepatic metastases, and 6.7 and 8.3 months with extrahepatic metastases only (n = 11 and 7). With a largest lesion of 8 cm or larger (n = 18 and 5), median PFS was 4.2 and 1.7 months and the ORR 11.1% and 0%. By control stratum, median PFS was 6.9 and 3.1 months in the ipilimumab plus nivolumab stratum and 5.8 and 2.8 months in the pembrolizumab or dacarbazine stratum. The poster states that for LDH 2 times the ULN or higher, target lesion 8 cm or larger, and extrahepatic-only metastases, the small numbers of participants gave wide 95% confidence intervals that included the null value.

OS, the phase 3 primary endpoint, was not mature at the primary analysis, and no OS estimate has been reported; the presentation reports “an early trend toward improved OS”, and IDEAYA states that the next OS update will be part of the prespecified interim analysis. The quarterly report for the first quarter of 2026 describes “a statistically significant improvement in the secondary endpoint of overall survival rate as assessed by blinded independent central review”.

Summary: OptimUM-02 safety

In the safety populations of the presentation (239 participants receiving darovasertib 300 mg plus crizotinib 200 mg and 100 receiving investigator's choice, phases 2a and 2b), treatment-related adverse events (TRAEs) of any grade occurred in 235 (98.3%) and 89 (89.0%) participants; the table of the most common TRAEs gives 88 (88.0%) for the control arm. Grade 3 or 4 TRAEs occurred in 97 (40.6%) and 37 (37.0%) participants, and treatment-related serious adverse events in 22 (9.2%) and 25 (25.0%); the ASCO 2026 abstract gives 9.2% and 25.3%. One grade 5 TRAE occurred in each arm (0.4% and 1.0%), and the abstract states that one treatment-related fatal adverse event occurred in each arm; the presentation describes the grade 5 serious adverse event in the combination arm as hepatic failure in a participant with extensive bulky liver metastases, considered unrelated by the sponsor, and the event in the control arm as gastrointestinal hemorrhage.

The most common TRAEs with darovasertib plus crizotinib were diarrhea (84.9%), nausea (74.9%), peripheral edema (66.9%), vomiting (49.8%), dermatitis acneiform (41.8%), fatigue (39.3%), and hypotension (33.1%); with investigator's choice, diarrhea (29.0%), increased alanine aminotransferase, increased aspartate aminotransferase, fatigue, and nausea (25.0% each), and pruritus (24.0%). Grade 3 or 4 TRAEs with the combination were mainly diarrhea (10.0%), syncope (7.1%), and hypotension (3.8%); with investigator's choice, increased alanine and aspartate aminotransferase (7.0% each), diarrhea (6.0%), hepatitis (5.0%), and colitis (4.0%). For treatment-emergent adverse events of any cause, the abstract reports diarrhea (89.5%), nausea (78.2%), peripheral edema (70.7%), and vomiting (53.6%) with the combination, and grade 3 or 4 diarrhea (10.9%) and syncope (7.9%). Hypotension and syncope were managed with hydration, holding blood pressure medications when appropriate, and the 7-day darovasertib run-in.

TRAEs led to withdrawal of darovasertib in 6 participants (2.5%), of crizotinib in 24 (10.0%), and of investigator's choice treatment in 19 (19.0%), and to dose reduction of darovasertib in 56 (23.4%) and of crizotinib in 63 (26.4%). The abstract reports adverse events leading to discontinuation and dose reduction of darovasertib in 3.8% and 23.8% and of crizotinib in 10.9% and 27.2%, and discontinuation in 19.2% of the control arm. Median relative dose intensities were 91.0% for darovasertib and 77.1% for crizotinib. Median duration of treatment was 6.31 months for darovasertib and crizotinib and 2.87, 2.56, and 2.79 months for pembrolizumab, ipilimumab, and nivolumab.

Summary: OptimUM-01 first-line cohort efficacy and safety

In the phase 2 expansion of OptimUM-01, an open-label phase 1/2 study enrolling participants irrespective of HLA status, 44 participants with no prior systemic therapy for metastatic uveal melanoma received darovasertib 300 mg twice a day plus crizotinib 200 mg twice a day (data cutoff May 28, 2025). Of these, 30 (68.2%) were HLA-A∗02:01-negative, 12 (27.3%) HLA-A∗02:01-positive, and 2 (4.5%) of unknown HLA status; 17 (38.6%) had ECOG performance status 1. Response was assessed by the investigator per RECIST version 1.1. In the 41 efficacy-evaluable participants, the confirmed ORR was 34.1% (14 of 41; 95% CI, 20.1 to 50.6), all partial responses; 23 (56.1%) had stable disease and 4 (9.8%) progressive disease. Median duration of response was 9.0 months (95% CI, 3.8 to 12.0) and median time to response 2.0 months. The disease control rate was 90.2% (37 of 41) at 8 weeks and 53.7% (22 of 41) at 26 weeks. With a median study follow-up of 25 months, median PFS was 7.0 months and median OS 21.1 months in the safety population, with PFS rates of 52% at 6 months and 28% at 12 months; the results section gives 95% CIs of 3.8 to 7.7 months for PFS and 12.5 to 27.1 months for OS, and the abstract gives 5.0 to 8.3 months and 11.9 to 20.7 months. In the 29 evaluable HLA-A∗02:01-negative participants, the ORR was 31.0% (9 of 29; 95% CI, 15.3 to 50.8), the disease control rate 86.2% (25 of 29), and median duration of response 9.03 months. In the 10 evaluable HLA-A∗02:01-positive participants, the subgroup table gives an ORR of 30.0% (3 of 10), and the discussion refers to “the observed ORR of 40%” in this subset. Among 20 participants with GNAQ/11 mutations reported from tumor tissue, 19 (95%) had a molecular response in cell-free DNA.

All 44 participants had at least one adverse event; grade 3 or higher treatment-emergent adverse events occurred in 23 (52.3%), serious adverse events in 9 (20.5%), TRAEs in 43 (97.7%), grade 3 or higher TRAEs in 12 (27.3%), and treatment-related serious adverse events in 4 (9.1%). The most common treatment-emergent adverse events were diarrhea (42 participants, 95.5%; the abstract gives 95.2%), nausea (84.1%), peripheral edema (61.4%), and vomiting (52.3%); the most common grade 3 or higher events were pulmonary embolism (9.1%) and iron deficiency anemia (6.8%). Two participants (4.5%) had grade 5 adverse events, one unrelated to study treatment and one deemed possibly related. Hypotension occurred in eight participants (18.3% in the text; 18.2% in the table), with one event of grade 3 or higher and no discontinuations for hypotension. Adverse events led to discontinuation in 3 participants (6.8%), and related adverse events led to dose reduction of darovasertib in 3 (6.8%) and of crizotinib in 4 (9.1%); 30 participants (68.2%) had at least one dose hold. Mean cumulative dose intensity was 84.9% for darovasertib and 80.0% for crizotinib, and a 2025 report of the same 44 participants (data cutoff 28 May 2025) gives median dose intensities of 92.6% and 88.0%.

Summary: OptimUM-01 HLA-A∗02:01-positive expansion

In the ESMO 2026 poster of the OptimUM-01 phase 2 expansion in HLA-A∗02:01-positive participants (data extract date March 23, 2026), 101 participants were treated with darovasertib plus crizotinib, with a median follow-up of 10.4 months for OS. Thirteen had no prior systemic therapy, 88 had received prior systemic therapy, and 15 received the combination in the second line after first-line tebentafusp; 22 (21.8%) had baseline LDH 2 times the ULN or higher. The confirmed ORR was 36.4% (4 of 11) in treatment-naive participants and 22.1% (19 of 86) overall, the disease control rate 90.9% (10 of 11) and 80.2% (69 of 86), median PFS 7.4 months (95% CI, 3.6 to 19.6) and 5.4 months (95% CI, 3.7 to 6.8), and median OS 27.1 months (95% CI, 8.9 to not evaluable) and 13.4 months (95% CI, 9.8 to 26.3). After prior tebentafusp, the confirmed ORR was 35.7% (5 of 14), median PFS 5.7 months, and median OS 12.1 months. TRAEs occurred in all 101 participants, grade 3 or higher TRAEs in 38 (37.6%), and treatment-related serious adverse events in 12 (11.9%); the most common TRAEs were diarrhea (82.2%), nausea (76.2%), peripheral edema (70.3%), and vomiting (57.4%). Adverse events led to discontinuation in 15 participants (14.9%), 9 of them treatment-related, and related adverse events led to dose reduction in 28 (27.7%).

Summary: single-agent LXS196 in the CLXS196X2101 phase 1 study

CLXS196X2101 (NCT02601378) was a Novartis-sponsored, first-in-human, multicentre, open-label phase 1 study of single-agent LXS196 (darovasertib) in metastatic uveal melanoma. Of 68 treated participants (56 in dose escalation, 12 in expansion), 59 (86.8%) had received prior systemic therapy and none had received prior tebentafusp. Doses ranged from 100 to 1000 mg once daily (n = 38) and 200 to 400 mg twice daily (n = 30). The maximum tolerated doses were 500 mg once daily and 400 mg twice daily, and the recommended dose for expansion was 300 mg twice daily. Dose-limiting toxicities occurred in 9 of 55 evaluable participants (16.4%), most commonly hypotension (n = 6; 10.9%).

Response (local RECIST v1.1 assessment) occurred in 6 of 66 evaluable participants (9.1%), with a median duration of response of 10.15 months; the results table, which includes 2 participants without post-baseline assessment in the denominator of 68, gives 1 complete response (1.5%) and 5 partial responses (7.4%), and the authors' comparison with AEB071 gives an ORR of 8.8%. At the recommended dose for expansion (18 participants), the ORR was 11.1% (95% CI, 1.4 to 34.7) and the disease control rate 77.8%. Median PFS was 3.6 months (95% CI, 3.5 to 5.4) in all participants and 3.7 months (95% CI, 2.1 to 5.4) at the recommended dose for expansion. Adverse events of any relationship occurred in 66 participants (97.1%) and were grade 3/4 in 29 (42.6%); the most frequent were nausea (70.6%), diarrhoea (54.4%), and vomiting (38.2%). Serious adverse events occurred in 14 participants (20.6%), most often hypotension (8.8%), and grade 3/4 hypotension occurred in 13.2% with once-daily and 3.3% with twice-daily dosing. The only adverse event leading to discontinuation was hepatic failure in a participant with known liver metastases.