番茄社区

The single-arm, phase II, trial was enough to nab sotorasib (Lumakras) accelerated FDA approval in 2021, as a treatment for patients with KRAS G12C-mutant advanced non-small cell lung cancer (NSCLC) who had disease progression on previous systemic therapy.

The agent turned in a 37% objective response rate (ORR), 6.8-month median progression-free survival (PFS), and 12.5-month median OS in the trial, all of which were fairly meaningful given that "until the past decade or so, efforts to generate an inhibitor of KRAS were unsuccessful," wrote Neal Rosen, MD, PhD, of Memorial Sloan Kettering Cancer Center (MSKCC) in New York City (NYC), in a editorial.

But as with any treatment that does well in a tightly controlled clinical trial, the question remained: Is it as effective "in the wild," so to speak? Kathryn C. Arbour, MD, also of MSKCC, and colleagues, assessed the agent in a multicenter retrospective study "to identify factors associated with real-world progression free survival (rwPFS)," OS, and toxicity.

Their study was done at three New York City-based institutions, and Arbour's group reported in that among 105 patients, sotorasib treatment led to a 5.3-month median rwPFS, 12.6-month median OS, and 28% real-world response rate. The team also found that KEAP1 co-mutations were linked with shorter rwPFS and OS, although there were no significant differences in either of those outcomes for those with TP53 co-mutations.

Finally, almost all patients who developed grade ≥3 treatment-related adverse events (TRAEs) had previously been treated with anti-PD-L1 therapy. Arbour and colleagues advised that "[a]wareness of these findings may assist clinicians with optimal timing and sequencing of these therapies in the clinic and devising appropriate follow-up plans after sotorasib initiation."

In general, the researchers highlighted that single-agent sotorasib was tied to modest activity in the real-world setting, and that assessment of KRAS co-mutation status and recent exposure to anti-PD-L1 therapy are warranted.

Of course, clinical trials forge on, including the phase II SCARLET trial that tested sotorasib plus carboplatin-pemetrexed (carbo-pem) in patients with KRAS G12C-mutated, advanced NSCLC. Hiroaki Akamatsu, MD, PhD, of Wakayama Medical University in Japan, shared details at the ASCO's 2023 annual meeting.

What was the impetus for the trial "The primary endpoint analysis of study: A single-arm, phase II study of sotorasib plus carboplatin-pemetrexed in patients with advanced non-squamous, non-small cell lung cancer with KRAS G12C mutation (WJOG14821L)?"

Akamatsu: Through 30 years of drug discovery, KRAS G12C has become a druggable target in advanced non-squamous NSCLC. Standard treatment for this population consists of ICI [immune checkpoint inhibitors] with or without platinum doublet sotorasib, a cytotoxic chemotherapy monotherapy.

However, the ORR ranges from 10% to 50%. The efficacy and safety of sotorasib plus carbo-pem in KRAS G12C-mutated advanced non-squamous NSCLC has not been investigated. Additionally, the incidence of KRAS G12C is lower in East Asians -- less than 5% -- compared with Caucasians, around 15%. That's why we conducted this study as a single-arm trial.

What are some of the study details?

Akamatsu: Key inclusion criteria are as follows: patients who are diagnosed as advanced non-squamous NSCLC with KRAS G12C mutation; naive for cytotoxic chemotherapy and KRAS inhibitors; should have measurable lesions and ECOG performance status of 0 to 1. Brain mets [metastases] were allowed unless symptomatic.

Study treatment consists of induction-phase sotorasib 960 mg, carboplatin AUC5, pemetrexed 500 mg/m² up to four cycles, followed by maintenance phase with sotorasib plus pemetrexed until PD [progressive disease] ... from October 2021 through July 2022, 30 patients were enrolled.

One did not receive protocol treatment due to deterioration, so 29 patients were analyzed for safety. Two patients were excluded for the efficacy analysis; one found to not fulfill inclusion criteria after enrollment. The other had an anaphylactic shock at course one, day 1, so this patient received only one dose of sotorasib and soon after withdrew from the study. Finally, 27 patients were analyzed for efficacy.

What were some of the main findings?

Akamatsu: At the time of this analysis, median follow-up time was 4.1 months. [For all enrolled patients], median age was 70. Most of the patients were male, had a smoking history, and adenocarcinoma histology.

Half of the participants had >50% of PD-L1 expression, and 10 were 1% to 49%, and five were negative. Seven patients had brain metastases at the time of enrollment, and only two had prior history of ICI. Both received durvalumab [Imfinzi] as a consolidative treatment after chemoradiotherapy.

This is a primary endpoint, ORR rate by BICR [blinded independent central review]. Of 27 patients, 24 had partial response, one, stable disease, and two, PD. Thus, ORR reached 88.9%, with 80% CI 76.9-95.8%, indicating that this study met its primary endpoint.

This is preliminary PFS, based on the median follow-up time of 4.1 months. By BICR, median PFS was 5.7 months, with 6-months PFS rate as 49%. By investigators' assessment, median value was 7.6 months, with 6-months PFS rate as 56% ... we are providing additional analysis later this year.

This is efficacy outcome by PD-L1 expression level. Consistently higher OR was observed across all PD-L1 expression levels. The result of PFS should be interpreted in caution because each of them had a very limited number of patients.

Median OS was not reached, with 6-months OS was 87%. At the time of the analysis, eight patients discontinued the study treatment. And of those, four received post-study chemotherapy, three received post-study radiotherapy, and none had surgery.

What about AEs?

Akamatsu: TRAEs occurred in ≥15% or any severe cases. Overall, 72% had grade ≥3 AEs. However, most of them are hematological toxicities, like anemia, platelets decreased, or neutrophil decreased, etc. Second [most] common AEs were GI toxicity and fatigue, and most of them are mild, grade 1 or 2. Thirteen percent to 70% of patients had hematological toxicity, and they are also mild. Unfortunately, one patient died due to pneumonia, grade 5.

What were some of the findings based on genomic status?

[Patients who were] positive at baseline for KRAS G12C were shedders and KRAS G23C were not detectable at 3 weeks. Non-shedders who were KRAS G12C positive at baseline were also positive at 3 weeks. Overall, KRAS G12C in plasma was detected in 70% of patients at baseline.

The second common mutation was TP53, detected in 44% of patients. Others include SDK11, CDKN2A, Kip1, BRAF, EGFR amplification, and MET amplification ... for the dynamics of allele frequency of KRAS G12C ... most of the patients had a decrease of all the frequency ... However, at 3 weeks, it was still detectable among 50% of cases. And their ORR seemed to be lower compared with baseline negative population or baseline positive shedder population. ORR did not differ by TP53 status.

Read the study by Arbour and colleagues here and expert commentary about it here.