Karlijn Hummelink

Chapter 1 16 Outline of this thesis In spite of the impressive results observed with checkpoint inhibitor-based immunotherapies in a small subset of patients with advanced NSCLC, robust predictive biomarkers are still lacking. Notably, there are no predictive biomarkers capable of accurately identifying patients who do not derive clinical benefit to ICB treatment. In this thesis, we examined direct effectors of the anti-tumor immune response as potential biomarkers to predict response and non-response to PD-1 blockade monotherapy in advanced NSCLC. Additionally, we explored alternative bio-sources for biomarker assessment to avoid the need for invasive and complicated biopsy procedures, particularly in cases where tumor tissue is not available. In Chapter 2 we examine the accuracy of a tumor-reactive TIL population, known as PD‑1T TILs, as predictive biomarker in a cohort of 120 advanced stage NSCLC patients treated with PD-1 blockade. The frequency of PD-1T TILs was quantified using digital image analysis. Additional exploratory analyses addressed the impact of lesionspecific responses, tissue sample properties, and the combination of PD-1T TILs with other biomarkers on their predictive value. Chapter 3 describes a study that investigates whether the predictive performance of biomarkers can be improved by combining them in pairs. The assessed biomarkers included both well-established ones, such as PD-L1, CD8/CD3 TILs and TIS, as well as recently developed biomarkers like PD-1T TILs, CD20+ B cells and TLS. All these biomarkers are known for their pivotal roles in the anti-tumor immune response upon PD‑1 blockade monotherapy. In Chapter 4 we investigate whether a tumor’s PD-1T TIL status can be translated into an mRNA signature using the Nanostring nCounter platform. As digital quantification of PD‑1T TILs requires a substantial user interaction, a PD-1T mRNA signature, developed on a robust clinical grade platform, will facilitates its implementation in a clinical setting. This study develops and validates a PD-1T mRNA signature using gene expression data from 100 advanced NSCLC patients treated with PD-1 blockade from two independent cohorts. In Chapter 5 we highlight the use of liquid biopsies as an alternative bio-source, obviating the need for tissue biopsies. This study develops and validates a serumderived protein signature designed to predict durable clinical benefit in 289 advanced stage NSCLC patients treated with PD-1 blockade.

RkJQdWJsaXNoZXIy MTk4NDMw