Sotorasib plus either pembrolizumab or atezolizumab revealed promising efficacy and relatively low rates of hepatotoxicity with a lead-in dosing strategy in patients with KRAS G12C–positive non–small cell lung cancer.
Sotorasib (Lumakras) plus either pembrolizumab (Keytruda) or atezolizumab (Tecentriq) revealed promising efficacy and relatively low rates of hepatotoxicity with a lead-in dosing strategy in patients with KRAS G12C–positive non–small cell lung cancer (NSCLC), according to efficacy and safety data presented at the 2022 World Conference on Lung Cancer.1
Across all treated patients (n = 58), the objective response rate (ORR) was 29% (95% CI, 18%-43%), comprised of 2 complete responses and 15 partial responses, and the disease control rate was 83% (95% CI, 71%-91%). The median duration of response was 17.9 months (95% CI, 5.6-not evaluable). Of note, responses were similar in immunotherapy-pretreated and -naïve patients. Overall survival with any sotorasib/PD-1 inhibitor combination was 15.7 months (95% CI, 9.8-17.8).
Due to the durable efficacy and safety of low-dose sotorasib as lead-in therapy followed by concurrent use with pembrolizumab, this regimen will be studied further as frontline treatment in patients with KRAS G12C–mutated NSCLC.
Sotorasib was granted accelerated approval by the FDA as monotherapy for patients with previously treated KRAS G12C–mutant NSCLC as determined by an FDA-approved test based on results from the phase 1/2 CodeBreak 100 trial (NCT03600883).2 Preclinical evidence has suggested that PD-1 inhibition may synergize with sotorasib to enhance CD8-positive T-cell infiltration and inhibit tumor growth.3
In the current analysis, patients from the phase 1/2 CodeBreak 100 and CodeBreak 101 (NCT04185883) trials who were administered sotorasib plus either atezolizumab or pembrolizumab were examined for efficacy and safety. The primary end point was safety with secondary outcomes of ORR, duration of response (DOR), disease control rate (DCR), and pharmacokinetics.
Patients (N = 58) were included if they had advanced KRAS G12C–mutated NSCLC, received or refused prior standard therapy, no prior experience with a KRAS G12C inhibitor, and no active brain metastases. All participants were randomized to an oral daily dose of sotorasib ranging from 120 mg to 960 mg. Half of the patients were first treated with a 21- or 42-day lead-in of sotorasib followed by treatment in combination with atezolizumab at 1200 mg (n = 10) or pembrolizumab at 200 mg (n = 19) every 3 weeks; the other half were treated with combination therapy of atezolizumab/sotorasib (n = 10) or pembrolizumab/sotorasib (n = 19) from the outset.
The median patient age was 66 years (range, 29-86), most had a history of smoking (93%), and the median number of prior therapies was 1 (range, 0-7). There were 12 patients (21%) who were receiving the combination as frontline therapy and 39 (67%) had been previously treated with an anti–PD-1/L1 agent. Most patients had an ECOG performance score of 1 (81%), 31% had brain metastases, and 26% had liver metastases. PD-L1 expression was under 1% in 10 patients (17%), between 1% and 49% in 16 (28%), and 50% or above in 21 (36%); the PD-L1 expression status was unknown for 11 patients.
For the patients who received concurrent pembrolizumab, sotorasib resulted in a high rate of hepatotoxic events, most of which were from liver enzyme elevation. Grade 3 hepatotoxicity across dose levels occurred at a rate of 80% in the 120 mg group (n = 5), 75% in the 360 mg group (n = 8), 100% in the 720 mg group (n = 2), and 75% in the 960 mg group (n = 4). Although sample sizes were small, lower starting doses of sotorasib trended toward fewer events of enzyme elevation leading to experimentation with the sotorasib lead-in dosing strategies.
For patients receiving pembrolizumab and a safety lead-in of sotorasib, grade 3 or greater treatment-related AEs (TRAEs) occurred in 100% of those in the 120 mg group (n = 3), 20% in the 240 mg group (n = 5), and 55% on the 360 mg group (n = 11). The rate of grade 3 hepatotoxicity in each respective group was 67%, 20%, and 45%. Overall safety data from both the lead-in and concurrent administration groups suggested a lower starting dose of sotorasib and lead-in administration is preferable for better tolerability.
Looking at all treatment arms, lead-in treatment strategies resulted in fewer overall and grade 3/4 TRAEs, discontinuations, and hepatotoxicity vs concurrent therapy. The first occurrence of grade 3/4 hepatotoxicity was outside the dose-limiting toxicity window in 88% of patients with 97% of cases resolving with corticosteroids or treatment modification and/or discontinuation. Notably, pretreatment with immunotherapy did not appear to influence the rate of treatment-related hepatotoxicity.
For the patients treated with atezolizumab/sotorasib, grade 3/4 TRAEs occurred at rates of 30% and 0% in those treated with lead-in therapy vs 50% and 10%, respectively, with the combination. Treatment discontinuation of either atezolizumab or sotorasib due to TRAEs occurred in 10% of those treated with lead-in therapy vs 50% on a concurrent regimen. Median treated duration was longer in the lead-in group at 6.5 months (range, 1-18) vs 4.4 months (range, 1-14) with concurrent therapy, whereas the median duration of combination therapy specifically was shorter at 1.5 months (range, 0-18) vs 2.5 months (range, 1-14), respectively. Median onset of hepatoxicity was shorter with the lead-in regimen at 50 days (range, 28-93) vs 67 days (range, 36-147) with concurrent treatment.
For the patients treated with pembrolizumab/sotorasib, grade 3/4 AEs occurred at rates of 53% and 0% in those treated with lead-in therapy vs 74% and 5%, respectively, with the combination. Treatment discontinuation of either atezolizumab or sotorasib due to TRAEs occurred in 32% of those treated with lead-in therapy vs 53% on a concurrent regimen. Median treatment duration was longer in the lead-in group at 2.8 months (range, 1-15) vs 4.9 months (range, 2-30) with concurrent therapy, whereas the median duration of combination therapy specifically was shorter at 0.7 months (range, 1-15) vs 2.3 months (range, 1-9), respectively. Median onset of hepatoxicity was shorter with the lead-in regimen at 73 days (range, 45-127) vs 51 days (range, 29-190) with concurrent treatment.