Experts Sound the Alarm on Lack of Comprehensive Genomic Profiling in NSCLC

The 3-fold effect of test cancellations, increased turnaround time, and insufficient tissue for testing has experts calling for a revamping of the order in which genomic profiling tests are ordered for patients who have non–small cell lung cancer (NSCLC). Comprehensive genomic profiling (CGP; DNA and RNA sequencing), and not single-gene testing (SGT), should come first, they emphasize in Oncology and Therapy.

“Retrospective real-world electronic health record studies confirm there is wide variation in molecular testing patterns for NSCLC in clinical practice,” they wrote. “The efficacy of using a sequential testing approach—initial SGT followed by CGP—has not been described.” In their analysis, they added, when SGT came first, more patients failed subsequent CGP or had delayed results because the sampling tissue was depleted.

For their analysis, they used Labcorp data to contact oncologists who had ordered SGT for guideline-recommended biomarkers between May and December 2022. Each oncologist was offered CGP instead of SGT or following a negative result for SGT.

A 30% drop in complete DNA and RNA sequencing was seen when CGP followed SGT | Image credit: Connect world - stock.adobe.com

Overall, 80 community practice sites were represented, and they accounted for 561 patients who had advanced and metastatic NSCLC for whom CGP was ordered. Twenty-seven percent of the patients had record of a negative SGT in the 6 months before their CGP.

Of the SGTs ordered, ALK rearrangements, EGFR mutations, and PD-L1 expression were ordered most often. The least common tests were for MET amplification and RET arrangements. The Labcorp data also show that of SGT ordered for patients, results typically filtered in over several days even though all tests may have been ordered on the same day.

Regarding just SGT, when several separate SGT orders were placed, 84% of tests included ALK, EGFR, and PD-L1; 50% of SGT order combinations included KRAS; 28% included ALK, EGFR, ROS1, and PD-L1; 17% included ALK, BRAF, EGFR, KRAS, RET, ROS1, and PD-L1; and 16% included ALK, BRAF, EGFR, KRAS, ROS1, and PD-L1. These top 5 order group combinations accounted for more than 70% of all SGT orders, and they had a median turnaround time of 8 to 26 days.

SGT was ordered for a scant 3% of patients according to guideline-indicated markers, and the longest turnaround time (32.5 days) was for SGT that covered ALK, EGFR, MET, ROS1, and PD-L1.

When outcomes related to CGP were analyzed, the investigators saw that geographic region, turnaround time, and tissue type and age were associated with variation in CGP order strategy. Oncologists were less likely to order SGT first in geographic areas with higher total CGP test volume, which the investigators defined as 100 or more orders during the study period. Turnaround time for CGP was also 3 days longer (16 vs 13 days) in patients who had prior SGT, and fewer of these orders were carried out compared with patients with no history of SGT. Patient age, sex, and tumor histology did not have significant effects on CGP orders.

When the same tissue specimen was used for CGP after SGT, there was a 30% drop in complete DNA and RNA sequencing. Further, when all CGP test orders were considered, including cancellations and failures, fewer variants were detected among patients who had previous negative SGT vs positive (31% vs 43%; P = .008). These patients also had lower rates of guideline-recommended variant detection.

“Our findings have significant implications for patients with NSCLC,” the study authors wrote. “Undertesting and incomplete testing for oncogenic driver mutations can lead not only to underutilization of frontline targeted therapy, but also to improper use of upfront immune checkpoint inhibitors in patients with actionable genomic variants who can derive clinically meaningful benefits from biomarker-driven targeted therapy.”

They believe that among patients who have advanced and metastatic NSCLC, SGT is an antiquated method for biomarker detection, “regardless of when it is performed,” and that CGP should be performed on every NSCLC tumor.

Reference

Nesline MK, Subbiah V, Previs RA, et al. The impact of prior single-gene testing on comprehensive genomic profiling results for patients with non-small cell lung cancer. Oncol Ther. Published online March 19, 2024. doi:10.1007/s40487-024-00270-x