Detecting therapy-guiding aberrations in platelets and plasma at the transcriptome level in non-small-cell lung cancer

Abstract

Abstract Background The detection of driver aberrations such as EGFR and ALK, in tumour tissue or plasma has become important for treating lung cancer patients. The aim of this study is to explore the feasibility of detecting therapy-guiding aberrations in RNA, isolated from platelet and plasma samples. Methods We applied a hybridization capture-based NGS approach on 10 platelet-derived RNA samples with 13 known variants. For validation, we applied RNA-based ddPCR focussing on T790M, L858R and E19del in EGFR, codon 12 and 13 of KRAS and the ALK-EML4/KIF5B fusions on both cell free plasma RNA and platelet RNA. Results For none of the platelet-derived RNA samples we detected the known variant with targeted RNA sequencing despite having sufficient unique reads (1.23x10^5 to 1.4x10^6). We next applied ddPCR to determine presence of mutated transcripts at a higher sensitivity (sensitivity of the assays reached down to 0.07%). No ALK fusion droplets were detected in platelet and plasma samples derived from two patients with a known ALK fusion gene in tumour tissues detected by our assay despite having high droplet counts for the GUSB housekeeping gene indicating that RNA quality was good. For EGFR total droplet counts were very low in both platelets and plasma, indicating low abundance of EGFR transcripts. No EGFR mutant droplets were observed in platelet and plasma samples derived from 8 and 6 patients with known variants, respectively. None of the expected KRAS mutations were detected in 5 plasma-derived RNA samples, with again only very low numbers of wild type KRAS droplets. In RNA derived from platelets, KRAS mutant positive droplets were detected in 3 out of 9 samples with variant allele frequency of 0.07%, 0.11% and 0.55%. Two of the three samples were derived from one patient, but obtained at different time points. Wild type KRAS droplet counts were very high in all platelet samples indicating that KRAS transcripts were highly abundant in platelets. Conclusions In conclusion, the fraction of tumour-derived RNA transcripts in plasma and platelets appears to be very low or undetectable. Our data indicates that plasma and platelet-derived RNA is not a good source to identify tumour cell specific genomic aberrations. Legal entity responsible for the study University of Groningen, University Medical Center Groningen. Funding Dutch Cancer Society. Disclosure All authors have declared no conflicts of interest.