Whole-slide image analysis identifies a high content of Hodgkin Reed-Sternberg cells and a low content of T lymphocytes in tumor microenvironment as predictors of adverse outcome in patients with classic Hodgkin lymphoma treated with ABVD.

Abstract

Classic Hodgkin lymphoma (cHL) constitutes the most frequent lymphoma in young adults. Its histopathology is unique as a scattered tumor population, termed Hodgkin Reed-Sternberg (HRS) cells is diluted in a prominent tumor microenvironment (TME) composed of T lymphocytes, B lymphocytes, macrophages, neutrophils, eosinophils and histiocytes. Traditionally, the identification of prognostic biomarkers in the cHL TME has required visual inspection and manual counting by pathologists. The advent of whole-slide imaging (WSI) and digital image analysis methods could significantly contribute to improve this essential objective in cHL research, as a 10-20% of patients are still refractory or relapsed after conventional chemotherapy. In this work, we have digitized a total of 255 diagnostic cHL slides and quantified the proportion of HRS cells (CD30), B cells (CD20) and T cells (CD3) by digital image analysis. Data obtained where then correlated with the overall survival (OS) and progression free survival (PFS) of cHL patients. Quantification of HRS cells, B cells and T cells reflects the biological heterogeneity of the different cHL histological subtypes analyzed. A percentage of 2.00% of HRS cells statistically significantly discriminated between patients achieving a complete metabolic response (CMR) and refractory or relapsed (R/R) patients both for the OS (P=0.001) and PFS (P=0.005). Furthermore, patients with a percentage of T cells below the 26.70% in the TME showed a statistically significantly shorter OS (P=0.019) and PFS (P=0.041) in comparison with patients above this threshold. A subgroup of patients with a low content of T cells and high content of HRS cells exhibited a special aggressive clinical course. Currently, there is the need to implement quantitative and easy scalable methods to enhance clinical translation, as the cHL TME plays a central role in the clinical course of the disease. The results of this study could contribute to the identification of prognostic biomarkers specifically looking at the cHL TME and their inclusion in future clinical trials.