Predicting response to immunotherapy in lung cancer: an early HTA of predictive tests.

Objectives: Predictive biomarkers can identify patients who are more likely to respond to immunotherapy, which can guide treatment decisions. The objective of this study was to assess the potential value of predictive biomarkers in advanced NSCLC patients to guide the development of cost-effective biomarkers in this field.

Methods: A decision analytical model was constructed to compare theoretical new strategies with biomarkers to the current standard of care. The analysis was performed for three different patient groups based on PD-L1 status. Differences in health outcomes (QALYs) and costs were assessed between the current practice and these biomarker strategies.

Results: Omitting immunotherapy in NSCLC patients with a PD-L1 score < 1 percent or between 1 and 49 percent, and a negative biomarker test, could potentially reduce healthcare costs significantly a small loss in QALYs. In these groups, a biomarker test is potentially cost-effective as the incremental cost-effectiveness ratio largely exceeds a willingness-to-accept threshold of €80,000 saved per QALY lost. For patients with a PD-L1 score > 50 percent, a considerable QALY gain can potentially be realized by adding chemotherapy to patients with a negative biomarker test. However, this comes at a significant increase in costs and appears not to be cost-effective.

Conclusions: In general, predictive biomarkers seem to have the potential to increase the cost-effectiveness of treatment with immunotherapy in patients with advanced NSCLC. Optimal positioning of a biomarker depends on the weighing between health impact and costs.