Elevated ROS Levels in Response to CD20-Targeting Enhance Senescence Entry after Immunochemotherapy in Human B-Cell Lymphoma

Introduction: Besides triggering effector cascades of the immune system, CD20-directed immunotherapy induces programmed death of malignant B-cells. Therapy-mediated senescence is a long-term growth arrest of metabolically active tumor cells with prognostic relevance in preclinical lymphoma models, especially under conditions with impaired integrity of apoptotic pathways. Several therapeutic antibodies have been implied to modulate senescence levels in human tumor cells. However, the contribution of senescence to outcome to rituximab-based immunochemotherapy in B-cell lymphoma remains to be elucidated. Methods: We applied CD20-directed immunotherapy or the anti-lymphoma drug adriamycin (ADR, a well-known inducer of cellular senescence) to B-cell lines and primary EBV-transformed B-lymphoblastoid cells. Growth characteristics, senescence-associated β-galactosidase (SA-β-Gal) reactivity, selected components of the senescence-associated secretory phenotype (SASP) as well as intracellular reactive oxygen species (ROS) were analyzed. Senescence-related gene sets were probed for enrichment in a publically available sequential data set derived from patients with chronic lymphocytic B-cell leukemia (CLL) undergoing their first cycle of rituximab-based immunochemotherapy. Results: Major features of cellular senescence could be detected in response to CD20-directed immunotherapy as well as ADR in B-cell lines and primary EBV-transformed B-lymphoblastoid cells in vitro. Moreover, additional rituximab-based treatment increased levels of SA-β-Gal-reactive cells as well as secretion of the SASP components interleukin-6 and 8 and further reduced cell fractions in S-phase incorporating 5-bromodeoxyuridine when compared to application of ADR as a single agent. Elevated ROS levels were detected as a response to CD20-directed therapy and rituximab-mediated senescence was entirely abrogated upon prior administration of the ROS scavenger N-acetylcysteine. Moreover, we found senescent-related gene sets that had been inferred from relevant publications or were defined based on our own Chip-on-Chip analyses in human fibroblasts undergoing oncogene-induced senescence to be selectively enriched for in post-treatment samples from CLL patients later achieving partial or complete remission, but not with stable or progressive disease. Discussion: Our results show ROS-mediated lymphoma cell senescence as a hitherto unrecognized consequence of rituximab-based immunotherapy and argue for a contribution of enhanced senescence levels after combined immunochemotherapy to the clinical superiority of treatment regimens that include rituximab. Moreover, sequential gene set enrichment analyses hint towards an association of therapy-induced senescence and favorable outcome to immunochemotherapy in CLL. Our data warrant further investigation of cellular senescence as an effector program of antineoplastic treatment in human lymphoma.