New pharmacodynamic parameters linked with ibrutinib responses in chronic lymphocytic leukemia: Prospective study in real-world patients and mathematical modeling

Background One of the first clinical observations of ibrutinib activity in the treatment of chronic lymphocytic leukemia (CLL) is a rapid decline in lymph nodes size. This phenomenon is accompanied by an hyperlymphocytosis, either transient or prolonged, which is associated with distinct clinical responses and thus has an impact on long-term outcomes. Understanding which factors determine distinct disease courses upon ibrutinib treatment remains a scientific challenge. Methods and findings From 2016 to 2021, we conducted a longitudinal and observational study in 2 cohorts of patients with chronic lymphocytic leukemia (CLL) (cohort 1, n = 41; cohort 2, n = 81). These cohorts reflect the well-known clinical features of CLL patients, such as Male/Female sex ratio of 2/1, a median age of 70 years at diagnosis, and include patients in first-line therapy (27%) or relapsed/refractory patients (73%). Blood cell counts were followed for each patient during 2 years of ibrutinib treatment. In addition, immunophenotyping and whole-body magnetic resonance imaging (MRI) were assessed in patients from cohort 1. These data were integrated in a newly built mathematical model, inspired by previous mathematical works on CLL treatment and combining dynamical and statistical models, leading to the identification of biological mechanisms associated with the 2 types of clinical responses. This multidisciplinary approach allowed to identify baseline parameters that dictated lymphocytes kinetics upon ibrutinib treatment. Indeed, ibrutinib-induced lymphocytosis defined 2 CLL patient subgroups, transient hyperlymphocytosis (tHL) or prolonged hyperlymphocytosis (pHL), that can be discriminated, before the treatment, by absolute counts of CD4+ T lymphocytes (p = 0.026) and regulatory CD4 T cells (p = 0.007), programmed cell death protein 1 PD1 (p = 0.022) and CD69 (p = 0.03) expression on B leukemic cells, CD19/CD5high/CXCR4low level (p = 0.04), and lymph node cellularity. We also pinpointed that the group of patients identified by the transient hyperlymphocytosis has lower duration response and a poor clinical outcome. The mathematical approach led to the reproduction of patient-specific dynamics and the estimation of associated patient-specific biological parameters, and highlighted that the differences between the 2 groups were mainly due to the production of leukemic B cells in lymph node compartments, and to a lesser extent to T lymphocytes and leukemic B cell egress into bloodstream. Access to additional data, especially longitudinal MRI data, could strengthen the conclusions regarding leukemic B cell dynamics in lymph nodes and the relevance of 2 distinct groups of patients. Conclusions Altogether, our multidisciplinary study provides a better understanding of ibrutinib response and highlights new pharmacodynamic parameters before and along ibrutinib treatment. Since our results highlight a reduced duration response and outcome in patients with transient hyperlymphocytosis, our approach provides support for managing ibrutinib therapy after 3 months of treatment. Author summary Why was this study done? Treatment of chronic lymphocytic leukemia (CLL) was revolutionized by the introduction of targeted therapies such as ibrutinib. Previous studies showed that lymphocyte kinetics under ibrutinib treatment are related to clinical outcome, but some potential mechanisms underlying these responses have not been fully explored. This study was done to decipher key biological and pharmacodynamic factors involved in ibrutinib response. What did the researchers do and find? Relying on a cohort of patients treated with ibrutinib, we combined clinical parameters, biological monitoring, whole-body tissue imaging, and mathematical modeling to decipher key biological and pharmacodynamic factors underlying the 2 distinct groups of ibrutinib response, transient or prolonged hyperlymphocytosis. We elaborated a mathematical model, based on previous ones of CLL and combining both dynamical and statistical approaches, that characterizes inter-patient variability and estimates patient-specific biological parameters associated with the 2 groups of response. We identified clinical factors (age, absolute lymphocyte counts) and biological factors, such as absolute counts of CD4+ T lymphocytes and regulatory CD4 T cells, programmed cell death protein 1 and CD69 expression on B leukemic cells, CD19/CD5high/CXCR4low level in blood, and lymph node cellularity, before and during ibrutinib therapy. These biomarkers are predictive of ibrutinib response. The mathematical approach confirmed the identified clinical factors, by highlighting that the differences between the 2 types of response to ibrutinib treatment were mainly due to the production of leukemic B cells in lymph node compartments, and to a lesser extent to T lymphocytes and leukemic B cell egress into bloodstream. We found that the group of patients identified by the transient hyperlymphocytosis has lower duration response and a poor clinical outcome. What do these findings means? Estimation of patient-specific biological parameters in the mathematical models highlights that only few biological processes are group-specific when CLL patients are treated with ibrutinib. Biomarkers defined in this study can be easily monitored in CLL patients before and during ibrutinib treatment. These results offer potential support for optimizing the management of ibrutinib treatment, especially in the first months of therapy. Access to additional data, especially longitudinal magnetic resonance imaging (MRI) data, could strengthen the conclusions regarding leukemic B cell dynamics in lymph nodes.