Association between KRAS/STK11/KEAP1 mutations and outcomes in POSEIDON
An exploratory analysis of outcomes from the phase III POSEIDON study investigated the association of STK11, KEAP1 or KRAS mutations and efficacy outcomes. The trial revealed that patients with metastatic non-small cell lung cancer who received a combined therapy of tremelimumab, durvalumab and chemotherapy experienced longer overall survival compared with those who received chemotherapy alone, regardless of STK11, KEAP1 or KRAS mutational status.
In POSEIDON, tremelimumab (anti-CTLA-4) plus durvalumab (anti-PD-L1) and chemotherapy demonstrated statistically significant and clinically meaningful improvements in both progression-free survival (PFS) and overall survival (OS) vs. chemotherapy alone in patients with first-line EGFR/ALK wildtype metastatic non-small cell lung cancer (mNSCLC). With clinical experience, patient populations with poor outcomes on current standard-of-care treatments are beginning to emerge. For example, improved outcomes following anti-PD-(L)1 immunotherapy (both as monotherapy and in combination with chemotherapy or anti-CTLA-4 therapy) have been observed in patients whose mNSCLC tumours have KRAS mutations versus wild-type KRAS. In contrast, mutations in STK11 or KEAP1 are correlated with poor prognosis and reduced benefit from both anti-PD-(L)1-based immunotherapies and chemotherapy. At WCLC 2022, an exploratory analysis of outcomes from POSEIDON by STK11, KEAP1 and KRAS mutational status was presented.
Study design
Patients (N= 1,013) with EGFR/ALK wild-type mNSCLC were randomised (1:1:1) to first-line tremelimumab + durvalumab + chemotherapy, durvalumab + chemotherapy, or chemotherapy. The trial population included patients with either non-squamous or squamous disease and the full range of PD-L1 expression levels. In the experimental arms, patients were treated with a flat dose of 1,500 mg of durvalumab, or durvalumab and 75 mg of tremelimumab with up to four cycles of chemotherapy every three weeks (Q3W) followed by maintenance treatment with durvalumab Q4W, or durvalumab and a fifth dose of 75 mg of tremelimumab given at week 16. In comparison, the control arm allowed up to six cycles of platinum-based chemotherapy.
In POSEIDON, 954 patients (94.2%) had baseline plasma samples available, of which 902 underwent successful circulating tumour DNA (ctDNA) sequencing using the GuardantOMNI platform. Tissue samples were available for 894 patients (88.3%) at baseline, of which 614 were successfully sequenced using the FoundationOne CDx platform. In total, 973 patients were molecularly characterised by analysis of either tissue DNA or ctDNA samples, with 543 characterised using both platforms. PFS and OS outcomes will be analysed in patients with or without functional somatic mutations in KRAS, STK11 or KEAP1.
Results
The mutation-evaluable non-squamous NSCLC population included 612 patients (96% of the intent-to-treat non-squamous population) whose tumours were molecularly characterised; 30%, 14% and 6% had KRASm, STK11m and KEAP1m, respectively. A clear trend for OS improvement was observed for patients with KRAS, STK11 or KEAP1 mutations when treated with a limited course of tremelimumab added to durvalumab plus platinum-based chemotherapy compared to chemotherapy alone. The exploratory analysis reported a hazard ratio (HR) of 0.56 (95%CI: 0.30-1.03) for STK11-mutated non-squamous NSCLC, a HR of 0.43 (95%CI: 0.16-1.25) for KEAP1-mutated NSCLC, and a HR of 0.56 (95%CI: 0.36-0.88) for KRAS-mutated non-squamous NSCLC. Median OS and two-year survival rates across all mutational subgroups also increased with the immunotherapy combination compared to chemotherapy. Patients with STK11 mutations had a median OS of 15.0 months with the triple therapy compared with 10.7 months in the chemotherapy arm, with estimated two-year OS rates of 32.3% vs. 4.5%. Those with KEAP1 mutations had a median OS of 13.7 months and 8.7 months in the two arms, with two-year OS rates of 35.0% and 0%. There was also a trend towards improved OS in patients with KRAS mutations, with a median OS of 25.7 months vs. 10.4 months in the triple therapy and chemotherapy groups, respectively, with estimated two-year OS rates of 51.7% and 25.6%. Similar trends were observed for PFS. The ORR with triple therapy was 45.2%, 45.5% and 55.0% in the STK11m, KEAP1m and KRASm subgroups, respectively.
Of note, the STK11 and KRAS subgroup analyses were presented for patients with non-squamous histology; while the KEAP1 subgroup analysis was presented for all mutation-evaluable patients irrespective of tumour histology, due to a small sample size. These subgroup analyses should thus be interpreted with caution as they are post hoc, exploratory and of smaller sample sizes.
Conclusion
In this exploratory analysis from POSEIDON, a trend for OS benefit was observed with first-line tremelimumab plus durvalumab plus chemotherapy in STK11m, KEAP1m and KRASm non-squamous mNSCLC. Similar trends were observed for PFS. In these patient subgroups, addition of a limited course of tremelimumab to durvalumab (until disease progression) and four cycles of chemotherapy led to more frequent and deeper responses, including complete responses, with sustained disease control. However, the small sample size of some subgroups limits interpretation. These data suggest the use of tremelimumab plus durvalumab and chemotherapy as a potential first-line treatment for mNSCLC in harder-to-treat patient subgroups, such as those with STK11m, KEAP1m or KRASm.
Reference
Peeters S, Cho B, Luft A, et al. Association Between KRAS STK11 KEAP1 Mutations and Outcomes in POSEIDON: Durvalumab Tremelimumab + Chemotherapy in mNSCLC. Presented at IASLC WCLC 2022; Abstract OA15.04.
