Abstract 3136: A novel algorism evaluating dominant clonality in Chinese NSCLC patients with concomitant EGFR and ALK variation

Background: Patients harboring co-mutations of epidermal growth factor receptor (EGFR) mutation and anaplastic lymphoma kinase (ALK) rearrangements may have difficulty in choosing appropriate tyrosine kinase inhibitor (TKI) therapy. Herein, a real-word cohort with more than 20,000 NSCLC patients was retrospectively established. Characteristics of co-mutations were analyzed extensively. Method: Formalin fixation and paraffin embedded (FFPE) tumor samples of non-small lung cancer cases from 12 clinical centers were enrolled. A multi-gene PCR panel (Amoy multi-gene PCR kit v.1), which includes EGFR, ALK, ROS1, RET, KRAS, NRAS, BRAF, HER2, and PIK3CA was applied to analyze driver gene mutation status. The cycle threshold (CT) values of positive mutation were recorded. CT ranking score (CTS) of a specific mutation type in a certain sample is calculated as 1-Rx (Rx: The quantile of the CT value of sample x, taking the CT value of the unique mutation sample in this study as a reference). Results: Co-mutations of EGFR/ALK were observed in 41 patients,including EGFR 19del/EML4-ALK (n=19) and EGFR L858R/EML4-ALK (n=10). Median CT values (25.51, n=19) of EGFR 19del in EGFR/ALK co-mutation samples were significantly higher than EGFR 19del mono-mutation (23.01, n=3973, p < 0.01). Higher median CT values of ALK-fusion in EGFR/ALK co-mutated samples were also observed when compared with mono-ALK fusion (26.1 vs 18.5). Median CTS of EGFR 19del, L858R and ALK fusion were 15.5% (n=19), 30.5% (n=10), and 3.2% (n=29) respectively. When defined CTS > 33.3% as positive cut-off of value, 41% (n=12) of EGFR/ALK co-mutation samples were EGFR CTS positive, 17.2% (n=5) were ALK CTS positive. No EGFR/ALK CTS co-positive samples was observed. Compared with L858R, 19del has a higher CTS positive rate of ALK when co-mutated with ALK (26.3% vs 0%, p < 0.01). Conclusion: Compared with EGFR or ALK mono-mutation, abundance of dominant clone comprising either mutated EGFR or ALK-fusion in co-mutation samples is lower, which was proved by the reported poor TKI efficacy in co-mutated patients. CTS can evaluate dominant clone for NSCLC with co-mutated driver genes. Indicated by CTS, tumor with EGFR/ALK co-mutation tends to form a single dominant clone. L858R is more likely to be the dominant subclone than 19del. Therefore, CTS algorism is a promising tool providing clone estimation. And it should be further validated in future perspective clinical trials. Citation Format: Yang Liu, Xianghong Ya