Comparative efficacy and safety of tyrosine kinase inhibitors for chronic myeloid leukaemia: A systematic review and network meta-analysis.

BACKGROUND: The pharmacotherapy of chronic myeloid leukaemia (CML) is mainly based on tyrosine kinase inhibitors (TKIs). The aim of this study was to compare the efficacy and safety of all TKIs in CML patients. METHODS: We conducted a systematic review with network meta-analysis (NMA) of randomised controlled trials (RCTs), including imatinib, nilotinib, dasatinib, bosutinib, radotinib and ponatinib. Searches were performed in PubMed, Scopus, Web of Science and SciELo (March 2018). The NMAs were built for six outcomes at 12 months: complete cytogenetic response (CCyR), major cytogenetic response (MCyR), deep molecular response, major molecular response (MMR), complete haematologic response and incidence of serious adverse events. We conducted rank order and surface under the cumulative ranking curve (SUCRA) analyses. RESULTS: Thirteen RCTs were included (n = 5079 patients). Statistical differences were observed for some comparisons in all outcomes. Imatinib 400 mg was considered the safest drug (SUCRA values of 10.3%) but presented low efficacy. Overall, nilotinib 600 mg was superior to the other TKI in efficacy (SUCRA values of 61.1% for CCyR, 81.0% for MMR, 90.0% for MCyR); however, no data on its safety profile at 12 months were reported. INTERPRETATION: Our results suggest that nilotinib should be upgraded to first-line therapy for CML, although further cost-effectiveness analyses, including the new TKI (i.e., ponatinib, radotinib), are needed.

Tyrosine kinase inhibitors in patients with chronic myelogeneous leukemia: defining the role of social risk factors and non-adherence to treatment.

OBJECTIVE: To assess the role of social risk factors on adherence to tyrosine kinase inhibitors (TKI) therapy in chronic myeloid leukemia (CML) patients. METHODS: This is a retrospective study and eligible patients were adults with CML on TKI treatment. Cases of no adherence to treatment were confirmed during pharmacists' consultation (patient-reported adherence). Baseline characteristics between groups were compared between cases and controls groups. Risk factors identified in bivariate analysis (p<0.2) were included in multivariate model. A qualitative investigation assessed whether such predictors of non-adherence had causal relationship. RESULTS: Of 151 patients with CML consulted by pharmacists, 21% had adherence problems. Despite patients with secondary school (p=0.03), most of investigated social risk factors did not differ between groups. However, by using a qualitative approach, patients' level of education could not explain low adherence rates behavior. CONCLUSIONS: Social determinants of health, herein investigated, were unlikely to predict adherence to treatment. Regression techniques may lead to untrue statements, so future researches should consider investigating the causes, not only the statistical estimates.