Background: Tyrosine Kinase Inhibitors (TKI) dramatically improve survival in patients with chronic myeloid leukemia in chronic phase (CML-CP). However, patients may develop intolerance of or resistance to TKI therapy over years. Given multiple effective therapies available, the optimal selection of third-line therapy is not clear. The aim of this study is to evaluate the outcomes of third-line TKI therapy in patients with CML-CP. Methods: We retrospectively combined patient-level data of third-line TKI therapy in the clinical trials of PACE and OPTIC studies with patients who received third-line TKI therapy at our institution. Given equivalent outcomes between second-generation TKIs (2G-TKI), we compared the outcomes of patients between 2G-TKI and ponatinib as third-line TKI therapy. Patients with T315I mutations were excluded in our study given the known resistance to all the 2G-TKI therapies. Multiple imputations was performed for missing covariates to reduce bias. Univariate and multivariate Cox regression analysis was performed for survival to identify prognostic factors with a p-value cutoff of 0.05 from univariate to multivariate analysis. One-to-one propensity score matching was performed with a caliper of 0.200, using matching covariates with age at the time of third-line therapy, gender, race, comorbidity, prior TKI therapy, the duration of prior TKI therapy, best prior response during the front-line and second-line TKI therapy, body mass index, white blood cell count, platelet count, percentages of basophils and blasts in peripheral blood, and BCR::ABL1 levels on the international scale. Progression-free survival (PFS) was defined from the start of third-line TKI therapy to the time of loss of major cytogenetic response during the study, progression to accelerated phase or blast phase at any time, death from any cause at any time, or the date of last follow-up; overall survival (OS) was defined as the start of third-line TKI therapy to death from any cause or the date of last follow-up. The Kaplan-Meier method was used for survival with the log-rank test and a significance p-value level of 0.05. Results: We identified 354 patients who received third-line TKI therapy [204 patients (58%) from our institution; 63 patients (18%) from the PACE study; 87 (25%) patients from the OPTIC study] with a median follow-up of 46 months (Table 1). Among them as third-line TKI therapy, 173 patients received 2G-TKI [69 patients (19%), dasatinib; 64 patients (18%), nilotinib; 40 patients (11%), bosutinib], and 181 patients received ponatinib (51%). Overall, the median age at the time of third-line TKI therapy was 54 years (range, 18-87). Prior to third-line TKI therapy, 313 patients (88%) received imatinib; 156 patients (44%), dasatinib; 161 patients (45%), nilotinib; 35 patients (10%), bosutinib. The overall median BCR::ABL1 level was 16.3% on the international scale at the time of third-line TKI therapy. Patients who received ponatinib had higher rates of non-ischemic cardiovascular comorbidity, shorter total duration of prior TKI therapies, higher percentage of basophiles in peripheral blood, and higher levels of BCR::ABL1 on the international scale. Propensity score matching minimized the baseline covariates and identified 96 patients at each group with a median follow-up of 46 months after propensity score matching. There is a tendency of deeper response with ponatinib at baseline BCR::ABL1 levels of >1-10% and >0.1-1% before and after propensity score matching. Using pre-matched data, multivariate Cox regression showed older age at the time of third-line TKI therapy and higher BCR::ABL1 levels as adverse prognostic factors for survival ; the third-line ponatinib therapy was associated with favorable survival [P=0.003; hazard ratio (HR), 0.435; 95% confidence internal (CI), 0.253-0.748]. The 4-year PFS rates were 58% and 75% in the 2G-TKI and ponatinib group, respectively, before matching (P<0.001); the 4-year PFS rates were 78% and 89%, respectively, after matching (P=0.0099). The 4-year OS rates were 57% and 80% in the 2G-TKI and ponatinib group, respectively, before matching; the 4-year OS rates were 78% and 87%, respectively, after matching (Figure 1). Conclusion: Third-line ponatinib therapy is optimal selection for patients with CML-CP who failed prior 2 TKI therapies in the absence of T315I mutation. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal