The researchers’ computer model estimated likely time-to-treatment, cost and yield outcomes for newly diagnosed patients with mNSCLC who underwent gene testing for programmed death ligand 1 (PD-L1) status and EGFR, ALK, ROS1, BRAF, MET, HER2, RET and NTRK1 mutation status if testing was performed sequentially as single-gene tests (with or without KRAS testing) or using hotspot gene panels, or with upfront NGS. The model estimated costs and time-to-test differences between NGS and both sequential single-gene testing (with or without KRAS testing) and hotspot gene panels, for testing costs and re-biopsy costs — but not treatment costs — at private commercial-insurer and government Medicaid/Medicare insurance reimbursement rates.1
The computer model involved several assumptions and inputs based on expert opinions that need to be validated using empirical data, the authors acknowledged.
“They make several assumptions, but I think the paper was interesting,” said Mark Gerstein, PhD, Albert L Williams Professor of Biomedical Informatics, Yale School of Medicine and the Yale Program in Computational Biology & Bioinformatics, New Haven, Connecticut. “I tend to agree with their thrust that NGS, doing it all once and properly upfront, is probably a better thing to do than these one-after-one smaller gene tests.”
The price of NGS has fallen rapidly, Dr Gerstein noted.
“More [important] than that, I think people are just getting very used to doing NGS,” he said. “There’s much infrastructure currently being created for dealing with next-gen sequencing, so it will surely end up being a de-facto standard. And it will, consequently, wind up being used in many diverse contexts.”
As NGS test costs and data storage costs drop, it will eventually make sense in clinical as well as research settings to sequence patients’ entire genomes, Dr Gerstein predicted. He likened it to a very accurate whole-body imaging data set that can be used to answer an immediate clinical question, and then archived for future questions.
“I think it’s going to eventually take over and everyone will have their genome done in all of these different contexts,” he said.