1477P - STK11 mutations predict poor prognosis for advanced NSCLC treated with first-line immunotherapy or chemo-immunotherapy according to KRAS, TP53, KEAP1, and SMARCA4 status

Background

The upfront treatment of non-oncogene addicted NSCLC relies on immunotherapy single-agent (IO) or in combination with chemotherapy (CT-IO). Genomic aberrations such as KRAS, TP53, KEAP1, SMARCA4, or STK11 may impact the survival outcomes. The prognostic role of STK11 mutations under upfront IO or CT-IO is still unclear.

Methods

We performed an observational study of 145 patients (pts) treated with first-line IO or CT-IO for advanced non-squamous (nsq) NSCLC at our institution tested with an extensive lab-developed NGS panel. The primary objective was to assess the clinical outcomes of STK11 mutated (mut) pts. Then, we performed an external validation through the public OAK/POPLAR dataset, including nsq NSCLC pts treated with single-agent IO or CT.

Results

Most pts were male (59.7%), former smokers (61.1%), with ECOG PS 0-1 (84%), and received first-line CT-IO (58.6%). 44.8% had a mutation in KRAS, 21.4% in KEAP1, 50.3% in TP53, 13.1% in SMARCA4, and 14.4% in the STK11 gene. In 14/21 pts STK11 and KRAS mutations co-occur (p=0.053). The median overall survival (OS) was 13.2 months (mo.) (95% CI, 8.6-19.6), while the median progression-free survival (PFS) was 6.5 mo. (95% CI, 4.8-8.9). The mOS was 8 mo. (95% CI, 5-16.7) for STK11 mut pts and 17.3 mo. for STK11 WT pts (95% CI, 8.9-24.4) (p=0.038). TP53 (8.3 vs 17.3), KRAS (9.2 vs 15.9), and KEAP1 (8.9 vs 15.9) mut pts evidenced a trend for dismal mOS. SMARCA4 status had no impact on mOS. STK11 mutations were detrimental to OS in the univariate (HR 1.74, p=0.041) and multivariate model (HR 1.97, p=0.025) after adjusting for sex, age, ECOG PS, treatment (IO vs CT-IO), KRAS, KEAP1, TP53, SMARCA4 status. Genomic alterations did not impact the mPFS in our cohort. Within the OAK/POPLAR dataset, STK11 mutations (60/818 pts) were significantly associated with increased death risk in the univariate (HR 2.01, p<0.001) and multivariate model (HR 1.66, p=0.001) after adjusting for age, sex, treatment (IO vs CT), KRAS, KEAP1, TP53, SMARCA4 status.

Conclusions

STK11 aberrations hampered the mOS of nsq NSCLC pts treated with first-line IO or CT-IO. The negative prognostic impact seems to be unrelated to IO administration.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.