A new study from the Netherlands shows that biomarker testing to predict treatment response and disease progression in non-small cell lung cancer (NSCLC) varied widely in both cost and the tests ordered. Analyzing variations in biomarker testing in patients with NSCLC, researchers discovered how it was delivered in a real-world setting in a single patient cohort. Their analysis includes a comprehensive view of molecular and genetic test results, the techniques used, costs, and turnaround time.
“While there is a lot of interest in the use of real-world data to analyze treatment variation and outcomes, this study demonstrates the importance of identifying variation in the use of molecular tests as the gateway to most cancer treatments,” explains lead investigator Professor Maarten IJzerman, University of Twente, Enschede, the Netherlands; and the University of Melbourne Centre for Cancer Research in Melbourne, Australia.
The continued introduction of biomarkers and innovative testing methods—like whole-genome sequencing and the evaluation of circulating tumor DNA using liquid biopsies—makes complex diagnosis in patients with stage IV NSCLC likely to be even more complex.
In this analysis, biomarker testing records of 102 stage IV NSCLC patients were reviewed for differences in the types of tests ordered, when, and how timely results were provided.
The range of testing—and their timing and costs—varied substantially. Ninety-nine unique biomarker-test combinations were found in 102 patients; almost none underwent the same tests in the same order. Targeted gene panels for EGFR, ALK, BRAF V600E, and ROS1 to predict response to treatment were most frequently conducted. Next was IHC analysis for programmed cell death protein ligand 1 (PD-L1) to predict response to immunotherapy.
Typically, the most common biomarkers were assessed within the first tests, and emerging biomarkers were tested further down the test sequence. The number of tests per patient also showed substantial variation. Tests were completed at different times in each patient, and in most patients more than one test was completed.
The mean cost per patient of biomarker testing was $2,258.52. At the cost of current biomarker testing, the authors found that replacing current testing with whole-genome sequencing would have led to cost-savings in only two patients (2%).
For response prediction and for selecting the optimal treatment, biomarker testing needs to be completed before treatment initiation. Given the sequential nature of biomarker testing, whether additional tests are conducted is partly dependent on the results of previous tests. As a result, the turnaround time of biomarker testing directly influences the time at which a patient can be started on treatment.
This study revealed that multiple biomarker tests for NSCLC can result in unnecessary delays in treatment. The substantial variation of biomarker tests in patients suggests the possibility of inequities in access to the tests and subsequently in access to targeted therapy or immunotherapy. For example, multiple biomarker tests, such as targeted gene panels and IHC analysis, can result in unnecessary delays if there is an unplanned cascade of tests along the pathway.
While next-generation sequencing, and particularly WGS, can increase efficiency by substituting all other tests used for biomarker testing, WGS is more expensive compared to other biomarker tests. But, the downstream value that more comprehensive molecular diagnostics provides by improving the treatment decision is potentially much higher than a reduction in the costs of testing.
“Perhaps the value of WGS should be seen in light of the reduction in the complexity of the diagnostic pathway, as it is unlikely that the cost of WGS will be competitive,” the authors write. “The value of reducing the complexity of the diagnostic pathway is an aspect of the value that WGS may provide but has not yet been explored in detail. It could be an exciting avenue for future research.”
