RESUMO
BACKGROUND: While second-generation tyrosine kinase inhibitors (TKI) revolutionized treatment for patients with chronic myeloid leukemia (CML) who developed a suboptimal response to imatinib, many patients in developing countries are fixed to the latter due to socioeconomic barriers. Despite this scenario, scarce information is available to evaluate the clinical prognosis of these patients. METHODS: We conducted a retrospective cohort analysis to compare the overall mortality of patients with CML who developed a suboptimal response to a standard dose of imatinib and were treated with either high-dose imatinib or a second-generation TKI. We created a marginal structural model with inverse probability weighting and stabilized weights. Our primary outcome was overall survival (OS) at 150 months. Our secondary outcomes were disease-free survival (DFS) at 150 months and adverse events. RESULTS: The cohort included 148 patients, of which 32 received high-dose imatinib and 116 a second-generation TKI. No difference was found in the 150-month overall survival risk (RR: 95% CI 0.91, 0.55-1.95, P-value = .77; RD: -0.04, -0.3 to 0.21, P-value = .78) and disease-free survival (RR: 1.02, 95% CI 0.53-2.71, P-value = .96; RD: 0.01, -0.26 to 0.22, P-value = .96). There was also no difference in the incidence of adverse events in either group. CONCLUSION: Ideally, patients who develop a suboptimal response to imatinib should be switched to a second-generation TKI. If impossible, however, our findings suggest that patients treated with high-dose imatinib have a similar overall survival and disease-free survival prognosis to patients receiving a second-generation TKI.