In Pursuit of Surrogate Markers for Treatment-Free Remission in Patients with Chronic Myeloid Leukemia from Argentina Stop Trial

INTRODUCTION: Long-term survival of patients with chronic myeloid leukemia (CML) has significantly improved since the introduction of BCR-ABL1 tyrosine kinase inhibitors (TKIs). Several considerations about the side effects, risks and cost associated with the lifetime treatment, have led patients and physicians to explore the possibility of TKI discontinuation after achievement of a sustained deep molecular response (DMR), so-called treatment-free remission (TFR). Several clinical trials show that approximately half of patients who achieve a sustained DMR during TKI treatment maintain molecular remission after suspension of TKIs. There is currently no biomarker that reliably predicts TFR in CML, mainly due to different study designs that have generated inconsistent data. Thus, further investigations are needed to identify factors that consistently favor achievement of TFR. With the aim of developing a biomarker for TFR prediction we analyzed the phenotype of Natural Killer (NK) cells and their relation to successful TKI cessation.

METHODS: This analysis was conducted as a substudy of the Argentina Stop Trial. Altogether, 50 consecutive chronic phase CML patients who participated in the clinical trial were recruited from 7 Argentinian centers. Peripheral blood samples were collected before stopping TKI treatment, at month 3, 12 and at any time when MR^3.0 was lost. Freshly isolated mononuclear cells from 46 patients were immunophenotyped by staining with CD3, CD16, CD25, CD56, CD57, CD158, NKp30, NKp44, NKp46, NKG2A, NKG2C, NKG2D and PD-1 antibodies and NK cells subpopulations were analyzed by flow cytometry (BD FACS Canto™II). Molecular recurrence-free survival was estimated by the Kaplan-Meier method and compared within groups by the log-rank test. The cutoffs of the numerical variables were optimized according to the log-rank test. Quantitative variables were dichotomized according to receiver operating characteristics (ROC) curves in order to describe sensibility and specificity. Multivariate analysis was performed through Cox proportional hazards model. Main results are provided with hazard ratio (HR) at 6 months and 95% confidence intervals (95% CI).

RESULTS: At the time of discontinuation the median proportion of NK cells (CD3^-CD56⁺) among lymphocytes was significantly increased in patients compared with controls (15% vs 9%, P = 0.0016). A significant difference between molecular relapsed vs no-relapsed patients was observed when optimal cutoff (0.43) for CD56^bright low and high was determined (at 6 months 74% vs 100% respectively, log rank test, p=0.023). At this time of follow up, no significant difference was observed for CD56^dim NK cells. Phenotypic markers for adaptive-like NK cells were analyzed, however, no significant differences were observed between the non-relapsing and relapsing groups. Nevertheless, molecular non-relapsing patients had significantly higher frequencies of PD-1⁺ NK cells as compared with molecular relapsing patients (at 6 months 85% vs 64%, Log Rank test, P=0.009). Based on the ROC and Youden Index analysis, at 6 months the 1.2 cutoff shows an 80% specificity and 50% sensitivity. Moreover, after multivariable Cox proportional analysis, including age, time of treatment, deep molecular response time, Sokal risk, NK cells and PD-1⁺ NK cells, the last subpopulation was identified as an independent prognostic factor for molecular-relapse-free survival (Hazard ratio = 3.63; 95% CI 1.3 - 10.1; P=0.014).

CONCLUSIONS: The clinical impact of NK cells in patients who have discontinued TKIs is controversial. The effects of TKIs against immune cells, including NK cell subsets, could depend on the type of TKIs; this aspect is particularly relevant in Argentinian treated patients real world, since many different copies of TKIs are routinely used in the clinical setting. Our study is the first, to our knowledge, to report a significant increase in PD-1 expression in NK cells at TKI cessation in patients who do not relapse. Accordingly to recent reports, PD-1 expression is more abundant on NK cells with an activated and more responsive phenotype and does not mark NK cells with an exhausted phenotype. To fully understand how PD-1 on NK cells modulates immune responses we are planning to carry out functional studies. In the future, we are also planning a comprehensive study of immune suppressors, including regulatory T cells and myeloid-derived suppressor cells.