A subset of ALK+ non-small cell lung cancer (NSCLC) patients relapse on ALK inhibitor (ALKi) treatment due to on-target resistance mutations affecting the tyrosine kinase domain. Objective: In this study, we investigated the presence of minor resistant clones in pre-treatment tissue samples and assessed their predictive value for subsequent resistance mechanisms. Methods: Using the highly sensitive digital droplet (dd)PCR technique, we analyzed 40 tissue samples obtained from 17 patients who had developed on-target resistance mutations after receiving ALKi between 2013 and 2022. We focused on 10 on-target ALKi resistant mutations identified in our patient cohort. Results: Fifteen ALKi resistance mutations were detected in 13 samples from 11/17 patients. Among these, four mutations were observed as resistance mutations in follow-up biopsies taken after first or subsequent lines of ALKi. Comparison of the test results from two subsequent biopsies, before and directly after therapy, revealed presence of the resistance mutation identified upon relapse in the pre-treatment sample of three cases that were all taken from the same tumor location. In six cases taken from different tumor locations, the resistant mutations were not found in the pre-treatment sample. Conclusions: By using the highly sensitive ddPCR approach, we detected minor clones with on-target resistant mutations in both treatment-naive and relapse biopsies from ALK-positive NSCLC patients. The predictive value of these mutations as the potential resistance-causing mechanism was limited to relapses occurring at the same tumor location as the pre-treatment sample.

EGFR-mutated non-small cell lung cancer (NSCLC) patients can be effectively treated with tyrosine kinase inhibitors (TKI) but frequently present with an EGFR T790M resistance mutation at relapse. We aimed to screen for T790M in pre-treatment formalin-fixed and paraffin-embedded (FFPE) tissue samples of patients with a confirmed T790M mutation at progression. We analyzed 33 pre-treatment DNA samples of NSCLC patients who progressed upon TKI between 2013 to 2019. To establish storage-time dependent formalin fixation-induced background levels for C>T mutations, we analyzed DNA isolated from archival (stored >1 year, n = 22) and recently generated (stored <1 month, n = 11) FFPE samples and included DNA isolated from white blood cells (WBC) (n = 24) as controls. DNA samples were analyzed by droplet digital (dd)PCR, and positivity was defined by outlier detection according to Grubb’s criterion. The T790M background allele frequency levels were 0.160% in DNA isolated from archival-FFPE, 0.100% in fresh FFPE, and 0.035% in WBC. Progression-free survival (PFS) time of the single T790M positive patient was 9 months, while T790M negative patients had a median PFS of 10 months (range 2–27). Proper storage time matched FFPE control samples are essential for reliable detection of T790M mutation at low VAF. The presence of EGFR T790M mutations in pre-TKI samples is rare, even in patients who progressed with EGFR T790M mutations.