Prospective Biomarkers of Pembrolizumab Combo Response Show Feasibility in Frontline Advanced NSCLC

T-cell inflamed gene expression profile and tumor mutational burden assessment was found to be a feasible approach to study the clinical activity of 3 pembrolizumab-based combination regimens in treatment-naïve patients with advanced non–small cell lung cancer.

T-cell inflamed gene expression profile (T-cellinf GEP) and tumor mutational burden (TMB) assessment was found to be a feasible approach to study the clinical activity of 3 pembrolizumab (Keytruda)-based combination regimens in treatment-naïve patients with advanced non–small cell lung cancer (NSCLC), according to interim results of the phase 2 KEYNOTE-495/KeyImPaCT trial (NCT03516981) that were presented during the 2021 SITC Annual Meeting.

Results showed that in patients with TcellinfGEPlow/TMBlow tumors, the objective response rate (ORR) was 12% in those who received pembrolizumab/lenvatinib (Lenvima), 12% in those who received pembrolizumab/quavonlimab, 0% for patients for patients on pembrolizumab/favezelimab at 200 mg and not available for pembrolizumab/favezelimab at 800 mg.

In the subgroup of patients with TcellinfGEPlow/TMBhigh–tumors, the ORRs for the pembrolizumab-based combination regimens were 22%, 33%, 33%, and 50%, respectively. Furthermore, in those who had TcellinfGEPhigh/TMBlow–tumors, the ORRs were 30%, 9%, 25%, and 11%, respectively.

However, in patients with TcellinfGEPhigh/TMBhigh NSCLC, the ORR was 39% with pembrolizumab/lenvatinib, 40% with pembrolizumab/quavonlimab, 60% with pembrolizumab/favezelimab at 200 mg, and 42% with pembrolizumab/favezelimab at 800 mg.

“Various response rates were observed among the 4 biomarker subgroups,” lead study author Martin E. Gutierrez, MD, director of Drug Discovery and chief of Thoracic Oncology at Hackensack Meridian Health Hackensack University Medical Center, said in a virtual presentation during the meeting. “The TcellinfGEPhigh/TMBhigh subgroup showed a trend of the best response rates among the biomarker subgroups for all 3 therapies, and the TcellinfGEPlowTMBlow subgroup showed a modest response when pembrolizumab was combined with quavonlimab or lenvatinib.”

PD-1/PD-L1–based combination regimens have demonstrated both efficacy and safety in patients with advanced NSCLC. However, it is unclear which patients are most likely to respond to combinations including pembrolizumab; several patients do not respond to this treatment approach due to resistance mechanisms.

T-cellinfGEP and TMB are both considered clinically validated biomarkers of response to pembrolizumab. In the phase 3 KEYNOTE-495/KeyImPaCT trial, investigators prospectively investigated the feasibility of T-cellinfGEP and TMB as a way to assess the ability to predict outcomes in the advanced NSCLC patient population who have not received prior treatment.

Investigators explored 4 biomarker subgroups defined by TcellinfGEP and TMB, along with the estimated percentage of those on study based on The Cancer Genome Atlas data: TcellinfGEPlow/TMBhigh (17%), TcellinfGEPhigh/TMBhigh (33%), TcellinfGEPlow/TMBlow (21%), and TcellinfGEPhigh/TMBlow (29%). The cutoff of –0.16 was used to define Tcellinf high and low; 175 mutations/exome was used to define high and low TMB.

In the KEYNOTE-495/KeyImPaCT study, patients aged 18 years or older with advanced NSCLC cannot have received prior systemic therapy for advanced disease, must have measurable disease per RECIST v1.1 criteria, an ECOG performance status of 0 or 1, and EGFR-/ROS1-/ALK-/BRAF-negative tumors.

Patients then underwent tumor biomarker screening and were stratified into the 4 biomarker subgroups: TcellinfGEPlow/TMBlow (group I, n = 66), TcellinfGEPlow/TMBhigh (group II, n = 66), TcellinfGEPhigh/TMBlow (group III, n = 66), TcellinfGEPhigh/TMBhigh (group IV, n = 90).

Upon stratification, 208 patients were randomized 1:1:1 to receive pembrolizumab at 200 mg every 3 weeks plus lenvatinib at 20 mg daily (assigned, n = 72; on treatment, n = 25); pembrolizumab at the same dosage plus the CTLA-4 inhibitor quavonlimab at 75 mg every 6 weeks (assigned, n = 72; on treatment, n = 26); or pembrolizumab at the same dose plus the LAG-3 inhibitor favezelimab at 200 mg (assigned, n = 30; on treatment, n = 3); or 800 mg (assigned, n = 34; on treatment, n = 14) every 3 weeks.

The primary end point is ORR; secondary end points are progression-free survival, overall survival, safety, and follow-up. The first interim analysis for each therapy after at least 10 patients have had 12 or more weeks of follow-up.

Baseline characteristics were similar between the 4 treatment arms. The median age was 67.25 years (range, 27-85), 66.75% of patients were male, and 67% of patients had an ECOG performance status of 1. Additionally, most patients (84%) were considered current/former smokers.

Seventy-nine percent of patients had nonsquamous histology, more than half (52.75%) of patients had a PD-L1 tumor proportion score of 1% or higher, 38.75% had TMB-high tumors, and 58.75% of patients had TcellinfGEP-high status.

Regarding safety, adverse events (AEs) were mostly similar across the 4 treatment arms. Grade 3 to 5 AEs occurred in 81%, 50%, 43%, and 59% of those on pembrolizumab/lenvatinib, pembrolizumab/quavonlimab, pembrolizumab/favezelimab at 200 mg, and pembrolizumab/favezelimab at 800 mg, respectively. Serious AEs occurred in 22%, 14%, 7%, and 15% of patients respectively. AEs led to discontinuation in 32% of patients on pembrolizumab/lenvatinib, 18% in patients on pembrolizumab/quavonlimab, 27% on pembrolizumab/favezelimab at 200 mg, and 15% on pembrolizumab/favezelimab at 800 mg.

Treatment-related AEs occurred in 88%, 65%, 57%, and 59% of patients on pembrolizumab/lenvatinib, pembrolizumab/quavonlimab, pembrolizumab/favezelimab at 200 mg, and pembrolizumab/favezelimab at 800 mg. However, grade 3 to 5 AEs were highest with pembrolizumab/lenvatinib (63%) compared with 18% with pembrolizumab/quavonlimab, 7% with pembrolizumab/favezelimab at 200 mg, and 24% with pembrolizumab/favezelimab at 800 mg.

Immune-mediated AEs or infusion reactions occurred in 44%, 32%, 40%, and 32%, respectively. The most common occurring one was hypothyroidism (33%) in the pembrolizumab/lenvatinib arm.

Guiterrez concluded that the trial is still enrolling patients.

“Further validation is needed to determine additional efficacy and may be addressed as more mature data become available,” he added.

Reference

Gutierrez M, Lam WS, Hellmann MD, et al. KEYNOTE-495/KeylmPaCT: interim analysis of a randomized, biomarker-directed, phase 2 trial of pembrolizumab-based combination therapy for non-small cell lung cancer (NSCLC). Presented at: 2021 SITC Annual Meeting; November 10-14, 2021; virtual. Abstract 457.