The phase 1b Study 116/KEYNOTE-524 trial of lenvatinib and pembrolizumab revealed the RAS gene signature to possibly be associated with progression-free survival in unresectable hepatocellular carcinoma.
The RAS gene signature is potentially associated with progression-free survival (PFS) from combination treatment with lenvatinib (Lenvima) and pembrolizumab (Keytruda) in patients with unresectable hepatocellular carcinoma (HCC) when accounting for T-cell–inflamed gene expression profile (GEP), according to an exploratory analysis from the phase 1b Study 116/ KEYNOTE-524 trial (NCT03006926).1
Otherwise, no statistically significant associations were reported between gene expression signatures and PFS or objective response rate (ORR) before adjusting for T-cell–inflamed GEP.
Study authors, led by Richard S. Finn, MD, professor of medicine at David Geffen School of Medicine at University of California, Los Angeles, presented the results in a poster at the 2022 International Liver Cancer Association Conference.1
The multicenter, open-label Study 116/KEYNOTE-524 trial was a dose-finding and dose-expansion study exploring the use of lenvatinib plus pembrolizumab in patients with unresectable HCC who had no prior systemic treatment and had Barcelona Clinic Liver Cancer stage B or C disease. Patients (n = 104) received 8 or 12 mg of daily lenvatinib, depending on the patient’s weight, and 200 mg of intravenous pembrolizumab once every 3 weeks.
The ORR by independent imaging review was 36.0% per RECIST 1.1 criteria and 46.0% per modified RECIST criteria. The median duration of response was 12.6 months by RECIST criteria and 8.6 months by modified RECIST criteria. The median PFS was 8.6 months and 9.3 months by RECIST and modified RECIST criteria, respectively. The median overall survival was 22 months.2
Ninety-five percent of patients experienced 1 or more treatment-related adverse event, with the most common being hypertension (36%), diarrhea (35%), fatigue (30%), decreased appetite (28%), and hypothyroidism (25%). The only grade 4 event reported was leukopenia/neutropenia.
After receiving a breakthrough therapy designation for the combination, the developers submitted an application for the regimen. Despite the clinically meaningful efficacy seen with the regimen in the Study 116/KEYNOTE-524 trial, the FDA issued a complete response letter for the indication because it followed shortly after the approval for atezolizumab (Tecentriq) plus bevacizumab (Avastin).3
Exploratory analyses have since sought to identify possible biomarkers of response to the treatment regimen. In the analysis from Finn et al, investigators analyzed RNA sequencing data in 40 patients from the study for an 18-gene T-cell–inflamed GEP and 11 different noninflamed signatures to determine possible associations between these gene expression signatures and posttreatment clinical outcomes.
Analysis was completed using logistic regression and Cox proportional hazard regression models with a 1-sided P value for the T-cell–inflamed profile and 2-sided for all other signatures. The prespecified level for significance was set at α = 0.05.
Of the evaluable patients, the median age was 65.5 years (range, 47-81) and 75% of patients were male. Sixty percent of the patients were White, 10% were Asian, 5% were Black or African American, and the remainder were unknown, other, or not reported. The ECOG performance status was 0 in 48% of patients and 1 in 53%. Thirty-five percent of the group were responders to the treatment regimen, with an ORR of 35% (95% CI, 21%-52%), including a complete response in 10%. The median PFS in the group was 7.7 months (95% CI, 4.4-9.9).
Evaluated P values for associations with gene expression signatures are included in the TABLE1 both before and after adjustment for T-cell–inflamed GEP. Most notably, the P value for association with RAS and PFS was statistically significant at .003 after adjustment for T-cell–inflamed GEP.
The patient-level area under the receiver operating characteristic curve (AUROC) score was 0.55 (95% CI, 0.36-0.75) by response status. Per gene expression signature, the AUROC by response rate was 0.34 (95% CI, 0.15-0.53) for the angiogenesis signature, 0.35 (95% CI, 0.16-0.54) for the granulocytic myeloid-derived suppressor cell signature, 0.46 (95% CI, 0.26-0.67) for the monocytic myeloid-derived suppressor cell signature, and 0.41 (95% CI, 0.22-0.61) for the glycolysis signature. With the hypoxia gene expression signature, the AUROC was 0.36 (95% CI, 0.17-0.55), 0.55 (95% CI, 0.5-0.75) for the microvessel density signature, 0.34 (95% CI, 0.16-0.52) for the MYC signature, and 0.44 (95% CI, 0.24-0.64) for the proliferation signature. The RAS signature showed an AUROC of 0.38 (95% CI, 0.18-0.58), 0.48 (95% CI, 0.28-0.68) for the stroma/epithelial-to-mesenchymal transition/transforming growth factor ơ signature, and 0.51 (95% CI, 0.31-0.71) for the WNT pathway signature.
Finn et al noted in their poster that findings from the analysis require further validation in larger cohorts, and that efforts are ongoing to identify biomarkers of response for patients with advanced HCC.