PB1908 THE SIGNIFICANCE OF THE INTERNATIONAL PROGNOSTIC INDEX FOR CHRONIC LYMPHOCYTIC LEUKEMIA (CLL‐IPI) IN SERBIAN CLL PATIENTS

Abstract

Background:The international prognostic index for chronic lymphocytic leukemia (CLL‐IPI) has been recently established as a powerful tool for risk stratification of CLL patients with regard to both overall survival (OS) and time to first treatment (TTFT).Aims:The aim of this study was to examine the applicability of CLL‐IPI to Serbian CLL patients.Methods:Our cohort consisted of 82 unselected CLL patients diagnosed and managed within the Clinic of Hematology, Clinical Center of Serbia, Belgrade, Serbia during the previous 15 years. The data regarding initial clinical and laboratory features were collected retrospectively from patient medical records. The analyses of cytogenetics (done by fluorescent in situ hybridization (FISH)), mutational status of immunoglobulin heavy variable (IGHV) genes and TP53 gene were performed from samples collected prospectively in the period between diagnosis and the first treatment. The data regarding age, stage (Rai and Binet), β2‐microglobulin (β2m), IGHV mutational status and FISH were available for all 82 patients, while TP53 status was known for 38 patients (46%). All the patients were assigned with CLL‐IPI and stratified according to The international CLL‐IPI working group. TTFT and OS analyses were performed using Kaplan‐Meier and Cox regression methods.Results:The median age at diagnosis was 59 (range 38‐81). Male to female ratio was 2.4. Fifty percent of patients were diagnosed with Binet A stage, 32% with Binet B, and 18% with Binet C. Median levels of serum β2m and lactate dehydrogenase (LDH) at diagnosis were 3.12 mg/L (range 0.2‐11.5 mg/L) and 374 IU/L (range 80‐930 IU/L), respectively. Regarding IGHV genes, 38 patients (41.5%) were mutated, while the rest were unmutated cases. Normal karyotype, del(13q), trisomy(12), del(11q), and del(17p) were detected by FISH in 32.9%, 43.9%, 7.3%, 13.4%, and 2.4%, respectively. TP53 was mutated in 4/38 patients. According to the CLL‐IPI, 18 patients (22%) were assigned with low risk, 33 (40.2%) intermediate risk, 28 (34.1%) high risk, and 3 (3.7%) with very high risk. After a median of 55 months (range 0‐169 months) patients were treated as follows: 51 (81%) with fludarabine‐based therapy (71% FC and 29% FCR), 7 (8.5%) with chlorambucil, and 5 (6.1%) with cyclophosphamide, vincristine, and prednisone (COP). Patients experienced significantly different median TTFT across risk groups: for low‐risk patients, median TTFT was not reached, for intermediate risk was 47 months, for high risk 31 months, and for very high risk 1 month (Log‐rank test across groups p < 0.001) (Figure 1A). Cox regression analysis confirmed that the risk of treatment commencement increased significantly with every extra score point (HR 1.542; 95%CI 1.337‐1.777; p < 0.001). Median OS in our cohort was 9 years (108 months) (range 4‐191 months). Median OS was not reached in the low‐risk group, while in intermediate, high, and very high‐risk groups were 94 months, 78 months, and 17 months, respectively (Log‐rank test across groups p < 0.001) (Figure 1B). Cox regression analysis for OS confirmed these findings (HR 1.542; 95%CI 1.283‐1.810; p < 0.001). In the group of Binet A patients, those in low risk according to CLL‐IPI exhibited significantly longer TTFT compared to intermediate and high‐risk patients (Log‐rank test across groups p = 0.022) (Figure 1C).Summary/Conclusion:The CLL‐IPI is highly applicable in Serbian CLL patients in predicting both TTFT and OS and represents a good discriminator of patients in early stage regarding disease aggressiveness.image