1060P - Impact of baseline clinicopathologic and genomic features on outcomes to KRAS G12C inhibitors in patients with NSCLC

Background

Baseline factors that influence efficacy of KRAS^G12C inhibitors (G12Ci) in patients (pts) with KRAS^G12C-mutant non-small cell lung cancer (NSCLC) are largely unknown. We sought to identify clinicopathological and genomic features associated with outcome to G12Ci.

Methods

Among pts at the Dana-Farber Cancer Institute and Massachusetts General Hospital with NSCLC who received a G12Ci as monotherapy, baseline clinicopathological and genomic features were correlated with objective response rate (ORR), progression-free survival (PFS), and overall survival (OS).

Results

Ninety-two pts received a G12Ci (median age 66.5 years, 63% female): 20 (22%) had an Eastern Cooperative Oncology Group performance status (ECOG PS) ≥2, 42 (46%) had history of brain metastases prior to G12Ci start, of which 15 (36%) were untreated before G12Ci initiation. Overall, the ORR was 39% (N=31/80), the median PFS (mPFS) was 5.1 months, and the median OS (mOS) was 9.8 months. Compared to an ECOG PS of 0-1, an ECOG PS of ≥2 was associated with shorter mPFS (5.5 vs 2.5 months, p=0.01) and mOS (12.4 vs 4.9 months, p<0.01). Compared to receipt of a G12Ci in the ≥2^nd line, 1^st-line treatment was associated with a higher ORR (75% [N=9/13] vs 32% [N=22/68], p=0.01), but line of therapy had no significant impact on mPFS or mOS. Co-occurring genomic alterations included STK11 mutation (mut) in 24 cases (27%), KEAP1 mut in 16 (23%), TP53 mut in 40 (44%), SMARCA4 mut in 9 (13%), ATM mut in 11 (13%), and CDKN2A/B deficiency in 21 (25%). ATM wild-type (wt) compared to ATM mut had lower ORR (33% vs 78%, p=0.02) and shorter mPFS (4.1 vs 13.8 months, p=0.04), but no different mOS. Longer mPFS and mOS were observed in SMARCA4 wt compared to mut (mPFS: 5.9 vs 2.8 months, p=0.02; mOS: 11.3 vs 5.3 months, p=0.02) and in CDKN2A/B intact compared to deficient (mPFS: 5.8 vs 2.8 months, p=0.02; mOS: 11.3 vs 5.7 months, p=0.01). There was no difference in mPFS or mOS according to STK11 or KEAP1 mutational status.

Conclusions

Baseline clinicopathological characteristics may help predict outcome to G12Ci in pts with KRAS^G12C-mutant NSCLC. Larger series are needed to confirm if co-occurring genomic alterations may help identify pts who derive differential benefit from G12Ci.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

M. Nishino: Financial Interests, Personal, Advisory Role: Daiichi Sankyo, AstraZeneca; Financial Interests, Personal, Research Grant: Merck, Canon Medical Systems, Daiichi Sankyo, AstraZeneca. L.M. Sholl: Financial Interests, Personal, Advisory Role: GV20 Therapeutics, Genentech, Lilly; Financial Interests, Personal, Research Grant: BMS, Genentech. R. Heist: Financial Interests, Personal, Advisory Board: Daichii Sankyo, Novartis, EMD Serono; Financial Interests, Personal, Invited Speaker: Chugai Roche; Financial Interests, Personal, Other, review of cancer histories: Boehringer Ingelheim; Financial Interests, Personal, Advisory Board, consulting honoraria: AbbVie; Financial Interests, Institutional, Invited Speaker: Daichii Sankyo, Novartis, AbbVie, Roche, Incyte, Mirati, Agios, Corvus, Turning Point, Lilly, Exelixis; Non-Financial Interests, Member: IASLC, ASCO, AACR. M. Awad: Financial Interests, Personal, Advisory Role: BMS, AstraZeneca, Achilles, AbbVie, Neon, Maverick, Nektar, Hegrui, Syndax, Gritstone; Financial Interests, Personal, Research Grant: BMS, AstraZeneca, Lilly, Genentech. All other authors have declared no conflicts of interest.