Impact of Comorbidities on Survival of Chronic Myeloid Leukemia Patients Treated with Tyrosine Kinase Inhibitors

Tyrosine kinase inhibitors treatment changed dramatically the survival of CML patients, approaching to the survival of the normal population. Comorbidities may also impact treatment results. Aim: to investigate the impact of comorbidities in the survival of CML patients treated with TKI Methods: We collected clinical and laboratory data from CML patients treated in a single center from medical records from January 2001 to December 2018. Comorbidities were collected at diagnosis before treatment and Charlson comorbidity index (CCI) was calculated. Because of CML diagnosis, the lowest possible score was 2. Patients were treated with imatinib 400-600 mg and one was treated with dasatinib in first line. Management followed the European Leukemia Net recommendations. OS probabilities were calculated using Kaplan-Meier method. OS was defined as the time between diagnosis and death of any cause, independent of treatment. Cox models were estimated for the multivariate analysis. All calculations were performed with version 24.0 SPSS software. Results: We included 273/310 patients in this analysis; 37 pts were excluded because of previous treated in other centers. Patients characteristics: 57.3% males, median age at diagnosis was 49 (14-86); 263 (96%) chronic phase and 11 (4%) accelerated phase; Sokal low risk 36.8%, intermediate 31.8% high risk 31.4%; EUTOS low risk (81.7%) and high risk (18.3%); median follow-up time was 92 months (1-223). Most frequent comorbidities: hypertension (27.2%); diabetes (7.5%). Smoking at diagnosis (15.3%). CCI group categories were: 2 (n=131), 3 (n=52), 4 (n=46), 5 or more (n=45). Probabilities of OS at 10 years for patients with CCI 2, 3 to 4 and ≥ 5 to were 81%, (95% confidence interval [CI]: 73-89%), 77% (95%CI: 63-100%), 69% (95%CI:53-85%), and 62% (95% CI: 38-86%), respectively. Probabilities of OS according to Sokal score was 86% (low risk) vs. 73% (intermediate) vs. 68% (high risk), P=0.002 and for EUTOS score 79% (low risk) vs. 61% (low risk), P=0.004. OS was lower in patients with hypertension (63% vs. 80%), P=0.018. In a multivariate analysis including CCI Sokal and EUTOS Score, smoking, diabetes, hypertension, the most powerful predictive factors for death were high and intermediate Sokal score and hypertension ((Wald test, P =0.003 and 0.049, respectively). There were 58 deaths, 22 not related to CML and 2 causes unknown. Conclusions: Comorbidities represents an important cause for mortality in CML patients treated in real life, besides known disease risk factors as Sokal and EUTOS scores. CCI high scores and hypertension were related to a lower OS. Figure 1 Disclosures Delamain: Novartis: Honoraria. Pagnano:Abbvie: Consultancy; Sandoz: Consultancy; Pint Pharma: Consultancy.