Predictive diagnostics for I-O

In the final lecture of the session, Dr Enriqueta Felip (Vall d’Hebron University Hospital, Spain) discussed predictive biomarkers for immunotherapy in stage 4 NSCLC. The most extensively studied predictive diagnostics include PD-L1 immunohistochemistry, tumour inflammation including RNA immune signatures and tumour-infiltrating lymphocytes (TILs), tumour mutational burden (TMB), and oncogenic alterations like KRAS, STK11/LKB1, and EGFR.

PD-L1 expression is a biological continuum, so it is difficult to determine specific cut-off values. The magnitude of the benefit correlates with PD-L1 expression in the second-line treatment of NSCLC. However, some PD-L1-negative patients do respond to immunotherapy [1]. Next, a growing body of evidence suggests that TMB is predictive of immunotherapy efficacy in NSCLC. However, there are multiple relevant challenges, such as the methodology, standardisation, definition of high vs low TMB, clinical validation, and reimbursement. The ongoing phase 2/3 Blood First Assay Screening Trial (BFAST) is evaluating the predictive value of TMB, determined in plasma samples, in first-line treatment-naïve NSCLC. Also, the rapid and sensitive detection of dynamic changes in circulating tumour DNA (ctDNA) and T cell expansion could become relevant markers to guide immunotherapy for lung cancer patients. Early ctDNA clearance has been found to predict both PFS and OS [2].

1. Garon EB, et al. N Engl J Med. 2015;372:2018-28.
2. Anagnostou V, et al. Cancer Res. 2019;79:1214-1225.

Posted on 25 June 201915 March 2021