Abstract 1045P
Background
There is an urgent clinical need for highly specific biomarkers to identify non-small cell lung cancer (NSCLC) patients unresponsive to PD-1 blockade. We studied how such specificity could be improved through combination of previously published, complementary biomarkers: the clonal tumor mutational load (cTML), genomic alterations in STK11/LKB1 and KEAP1, and actionable targets for small molecule inhibitors.
Methods
We prospectively collected whole genome sequencing (WGS), RNA sequencing (RNA-seq), PD-L1 immunohistochemistry (IHC) and clinical data of 75 patients with metastatic NSCLC treated with PD-1 blockade monotherapy. The cTML was defined as the genome-wide number of clonal, non-synonymous mutations, with a predefined threshold for low cTML of <300 mutations. Pathogenic alterations in STK11/LKB1 and KEAP1 were defined as non-synonymous mutations plus loss of heterozygosity, or bi-allelic deletions. Actionable, on-label genomic targets for small molecule inhibitors were considered in the genes EGFR, MET, ALK, RET, and BRAF. In low vs high cTML subgroups, we correlated the presence of STK11/KEAP1/actionable alterations to primary resistance, progression-free survival (PFS) and overall survival (OS).
Results
Although low cTML was strongly linked to primary resistance, it had limited specificity. Combining biomarkers effectively improved such specificity, as 20 out of 20 (100%) patients with a low cTML plus an actionable, STK11, or KEAP1 alteration showed primary resistance. These 20 patients represented 47% of all patients showing primary resistance in our cohort, which was a significant enrichment for non-responsiveness vs all other patients (Fisher’s exact P=0.0000017), or vs other patients with low cTML (Fisher’s exact P=0.0010). Individually, these three types of alterations (actionable, STK11, or KEAP1) were all significantly associated with poor PFS in the population with low cTML, but not in the population with a high cTML.
Conclusions
Based on our findings we propose that it may be safe to withhold PD-1 blockade from NSCLC patients with low cTML, in case their tumors harbor an actionable alteration or bi-allelic alteration in STK11 or KEAP1. Large follow-up studies are indicated to confirm this.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
Emile E Voest.
Funding
This study was funded by ZonMW (project number ZonMW 846001002), the Oncode Institute, and the Josephine Nefkens Foundation. This paper and the underlying data have been made possible partly on the basis of the data that Hartwig Medical Foundation and the Center of Personalized Cancer Treatment (CPCT) have made available to the study.
Disclosure
J.G. Aerts: Financial Interests, Personal, Advisory Board: Eli Lilly, Amphera, MSD, Pfizer, BMS; Financial Interests, Personal, Invited Speaker: Eli Lilly, MSD, Takeda, Biocad; Financial Interests, Personal, Other, DSMB member: Biocad; Financial Interests, Personal, Stocks/Shares: Amphera; Financial Interests, Personal, Other, Patent: Amphera, Pamgene; Financial Interests, Institutional, Research Grant, Present participation in >60 clinical trials related to oncology (all compensation paid to institution): Multiple. All other authors have declared no conflicts of interest. All authors have declared no conflicts of interest.