Comparison of Sequential Testing and Next Generation Sequencing in advanced Lung Adenocarcinoma patients - A single centre experience.

OBJECTIVES

Molecular diagnosis determines therapeutic strategies for patients with non-small-cell lung cancer - adenocarcinoma (NSCLC-A) but depends on resources availability. We compared a sequential single-gene testing algorithm to next generation sequencing in NSCLC-A to assess differences in terms of effectiveness, costs, tissue consumption and time.

MATERIALS AND METHODS

We analyzed a retrospective cohort of advanced NSCLC-A patients treated at the Sant'Orsola-Malpighi University Hospital. The sequential testing includes a first analysis of EGFR and KRAS status with further molecular testing physician driven. The available NGS panel detects 35 hotspot mutations,19 amplifications and 23 rearrangements.

RESULTS

We included 1758 patients; 1221 characterized with the sequential algorithm between January 2014 to February 2019 and 537 with Next Generation Sequencing (NGS) until January 2020. The prevalence of EGFR, ALK and KRAS alterations was similar between the stepwise and NGS group (16.5% vs 14.3%, 6.3% vs 6.3% and 36% vs 33.5%, respectively). Differently, ROS-1 rearrangements prevalence was higher in stepwise respect to NGS group (4.7% vs 0.7%). Similarly, the stepwise group presented higher prevalence than NGS for MET amplification (11.2% vs 2.2%), MET mutations (9.0% vs 2.4%), HER2 amplification (3.3% vs 1.9%) and mutations (9.8% vs 3.0%), and BRAF mutations (4.5% vs 5.6%). Among the NGS group other mutations were found in 141 patients (26.3%) and the presence of concurrent mutations in 131 (24.4%). The stepwise algorithm presented a relevant dropout rate that increased at each step, with 11.4%, 16.4% and 49.3% respectively for ALK, ROS1 and other analysis. Sequential testing's expenditure was 1375 € per patient, vs 770 € for NGS. Moreover, NGS testing can be performed with just a 25 μm slide respect to an estimated 33.3 μm slide for sequential strategy.

CONCLUSION

NGS offered a less expensive and more reliable model of diagnosis respect to sequential one for patients affected by NSCLC-A.