Chapter 7 238 TIL infiltration in relatively small biopsy samples, which is a promising development considering the challenges associated with biopsy sampling in NSCLC. In our study, we identified temporal heterogeneity, a phenomenon previously noted in PD-L1 IHC assessments18,19. Our results suggest that samples collected immediately before the initiation of PD-1 blockade monotherapy provided more accurate predictions of clinical benefit. It is worth noting that we excluded substantially older samples from our analysis, as archival samples might no longer accurately represent the tumor’s immunological status. No difference in predictive accuracy was observed between samples collected from primary or metastatic sites. In chapter 2 we further highlight that lesion-specific response better correlated with the frequency of PD-1T TILs compared to the overall response assessment by RECIST 1.1 criteria. Specifically, only a minority of the PD-1T high lesions progressed compared to the PD-1T low lesions. This observation may contribute to the reduced specificity and positive predictive value (PPV) of our biomarker, given that a significant number of tumors from patients with PD were PD-1T high. Previous work by Osorio and colleagues examined the dynamics of individual metastases following PD-1 blockade monotherapy and demonstrated that progression tended to occur heterogeneously across different metastatic sites22. Therefore, the PD-1T TIL status may reflect the local, and less likely the systemic anti‑tumor immunity. Hence, we expect that the frequency of these tumor-reactive TILs may vary among different metastatic lesions. While it may not always be clinically feasible to obtain fresh tumor tissue or biopsy multiple lesions within a patient, clinicians should recognize the limitations of biomarkers. Biomarkers can serve as valuable diagnostic tools in shared decisionmaking for therapeutic strategies. There is potential for further refinement of PD-1T TILs as a biomarker by integrating it into multimodal predictors that comprehensively capture the tumor-immune microenvironment and other tumorrelated factors. This aspect is explored in chapter 3 of this thesis. Additionally, a more in-depth characterization of the tumor reactive TIL subset holds promise for the development of novel gene signature biomarkers, a subject that is investigated in chapter 4 of this thesis. Chapter 3; The additive value of biomarker combinations The interaction between tumors and the immune system constitutes a complex spatiotemporal process, involving various stimulating and inhibitory factors acting at different stages of the cancer immunity cycle23. PD-(L)1 blockade therapy relies on promoting a dynamic anti-tumor immune response, and is not limited to targeting
RkJQdWJsaXNoZXIy MTk4NDMw