WCLC 2025 | Biomarker predictors of response to neoadjuvant chemo-IO in early NSCLC

We know that chemo-IO reshaped the treatment landscape of early-stage non-small cell lung cancer and that currently predictors such as biomarkers have not been completely tested in a real-world setting and patient selection might be challenging in this specific setting. Especially patients who achieve PCR pathological complete response are the ones who achieve the best survival outcomes in the long term so our study aims to determine what are the clinical genomic features of patients who achieved a pathological complete response or who did not achieve a pathological complete response. This is a multicentric study involving six academic centers across the United States and the European Union. 221 patients have been involved in this study and we looked at three main domains which are radiologic response in correlation to pathologic response, genomic correlates in correlation of pathologic response and the whole thing together in order to dissect which of these features could be the most predictive of this important endpoint. We’ve seen that radiologic response doesn’t really correlate with pathological response. In fact, stable disease patients, according to RECIST, might still achieve pathologic complete response, whereas patients who achieve a complete pathologic, a complete radiological response or a partial response might not achieve a pathological complete response. Therefore, we asked ourselves how the conventional predictors of ICI benefit in metastatic setting place themselves in this setting. We’ve seen that patients with high TMB levels, PD-L1 positive tumors and mutations in KRAS in conjunction with TP53 mutations are the ones who achieve the best pathological complete response rates, whereas patients with mutations in KRAS-11, KRAS-12 who have been known in the metastatic setting to be ICI resistance genes achieved the worst outcomes. But how to take all this information together? Some patients with KRAS and KEAP1 mutations in conjunction in our study achieved pathological complete response. Therefore, we asked ourselves, are there some other predictors we might use in these very difficult patients? The answer is high TMB levels and high PD-L1 levels might still represent a benefit for this subset of patients. In fact, patients with high TMB levels and PD-L1 levels, who had also KRAS and KEAP1 mutations, achieved pathological complete responses. Also it’s quite conventional to look at single predictors, whereas we try to dissect a little bit more predictors together and genes together. In fact we saw that patients who had similar impacting mutations in other genes of the same pathway as KEAP1 such as NFE2L2 did not achieve pathological complete responses. So in conclusion you need to look at predictors together, TMB levels, PD-L1 levels, KRAS and mutational status in TP53, STK11 and KEAP1 genes and choose the best regimen for your patients. Probably patients with worse prognostic factors such as KRAS and KEAP1 mutations or KRAS and STK11 mutations need to be enrolled in clinical trials in order to see if they could benefit from combos instead of single agent therapies.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.