Increased TACE (Tumor necrosis factor-alpha±-converting enzyme; ADAM17) Activity Associates with Decreased CD62L Expression, Increased Soluble CD62L Plasma Levels and Predicts Molecular Response to Nilotinib Therapy in Patients with Early Chronic Phase Chronic Myelogenous Leukemia (CML-CP): Results from an ENEST1st Substudy

Background and Aim: Tyrosine kinase inhibitors (TKI) imatinib and dasatinib modulate immune responses in vitro and in vivo. Immunological surveillance in the MRD-situation might be of particular relevance for long-term control or even elimination of CML-repopulating stem cells. Moreover, baseline immunological characteristics may be associated with response to TKI therapy. Little is known about potential immune-modulatory effects of nilotinib in vivo. The ENEST1st study (NCT01061177) evaluated the role of first-line nilotinib therapy in CML-CP. The primary endpoint was the MR4 rate at 18 months. A comprehensive immunological monitoring program within this ENEST1st substudy characterized baseline and therapy-induced immunological variables to correlate them with biological disease characteritics and clinical response parameters.

Methods: Peripheral blood was taken prior to treatment initiation and after 6 and12 months (mo) from 52 patients. Samples were analyzed by nine colour flow cytometry employing six panels of optimized antibodies to determine various leukocyte populations (e.g. T cell subpopulations including Treg and NKT cells, NK cells, B cells, monocytes, MDSC, dendritic cell subsets). Plasma concentrations of soluble CD62L (sCD62L) and TACE (tumor necrosis factor-α-converting enzyme; ADAM17, CD156b), the metalloproteinase inducing proteolytic cleavage of CD62L from the cell surface, were either measured by ELISA or (in case of the enzymatic activity of TACE) using a fluorogenic assay. Changes in immune cell parameters were correlated to biological disease features and clinical endpoints.

Results: The most striking finding of this study is the drastic loss of the lymph-node homing marker CD62L on immune cells (T cell subsets and granulocytes) at baseline (basCD62L), which increased back to normal levels during nilotinib therapy. The proportion of basCD62L⁺ cells among both CD4⁺ and CD8⁺ T cell subsets significantly correlated with Sokal score (both as continous and categorial variable, i.e. high vs. low/int). Low basCD62L expression levels on both T cell subsets correlate with increased spleen size, higher BM and PB blast and WBC counts as well as it correlates to higher BCR-ABL copy numbers at almost all time points during treatment. Similarly, lower basCD62L on either CD4⁺ or CD8⁺ T cells is linked to a longer duration to reach the respective molecular endpoint. Patients reaching MR⁴ at 18 months (primary study endpoint) had significantly higher levels of basCD62L on both CD4⁺ (p=0.02) and CD8⁺ (p=0.008) T cells. Consequently, MR⁴ at 18 months was attained in a significantly higher percentage of patients in the basCD62^hi compared to the CD62^lo patients (63% vs. 13.0%). Vice versa, patients who reached MR⁴ at 18 months had significantly higher proportions of basCD62L expressing cells among both CD4⁺ and CD8⁺ T cells. Moreover, as depicted by a cumulative response rate, patients with high proportions of basCD62L^hi T cells, achieved MMR and MR⁴ significantly earlier and in a higher proportion throughout the observation period. A detailed characterization of other T cell differentiation marker (CD45RA, CD45R0, CD28, CD27, and CD95) did not reveal significant baseline T cell subset alterations as explanation for altered CD62L expression. In contrast to low basCD62L surface expression levels, its shed form sCD62L is significantly increased at diagnosis but subsequently drops back during nilotinib therapy. Similar to surface CD62L expression, also sCD62L associates with biological disease features and molecular response to nilotinib. Finally, low CD62L surface expression was associated with elevated sCD62L levels and increased proteolytic activity but not total amount of TACE.

Conclusion: At baseline, increased proteolytic activity of TACE sheds CD62L from the immune cell surface. During nilotinib therapy, TACE activity gets normalized leading to re-expression of CD62L on T cells and vice versa a drop of sCD62L. Low baseline T cell expression levels of CD62L and increased sCD62L levels correlate to a more aggressive CML phenotype and are linked to inferior molecular response to nilotinib in early CML-CP. Larger prospective studies including also other TKIs are needed to confirm the prognostic relevance of sCD62L/CD62L expression as response-prediction marker, as this marker is easy to measure by ELISA in plasma samples or flow-cytometry.