Validation of Imatinib-Therapy Failure (IMTF) Score in Children with Chronic Myeloid Leukemia in the Chronic Phase

Background About 27% children with chronic myeloid leukemia (CML) still experienced imatinib failure. However, there is no appropriate prediction model to accurately predict the outcome at diagnosis. Objective To validate the predictive performance of the imatinib-therapy failure (IMTF) score and compare the predictive discrimination of IMTF score with ELTS score in newly-diagnosed children with chronic phase CML receiving imatinib-therapy. Methods Data of children with chronic phase CML diagnosed at < 18 years receiving initial imatinib (260 to 340 mg/m2 daily) from January 2006 to December 2023 were analyzed. Diagnosis, monitoring and treatment response assessment were according to ELN recommendations. The cumulative incidence of therapy response and imatinib-therapy failure by IMTF score and ELTS score was calculated using the Fine-Gray test considering competing events defined as stopping imatinib-therapy, a switch to a 2nd or 3rd generation TKIs, a transplant or death unrelated to CML. FFS, PFS and OS were calculated by Kaplan-Meier method with the log-rank test. Result s A total of 324 children were included in this study. 196 (61%) were male. Median age at diagnosis was 13 years (IQR, 9-17 years). 298 (92%) patients achieved a CHR by 3 months. With a median follow-up of 77 months (IQR, 51-107 months) for survivor, the incidences of CCyR, MMR, MR4.0 and MR4.5 at 8 years were 96% (94, 98%), 84% (79, 88%), 60% (53, 67%), and 45% (38, 53%); the 8-year probabilities of FFS, PFS and survival were 80% (75, 84%), 92% (88, 95%), 97% (95, 99%), respectively. By IMTF score, 63 (19%), 62 (19%), 103 (32%), 67 (21%), and 29 (9%) children were classified as very low-, low-, intermediate-, high-, and very high-risk groups. Since there was no significant difference between the low- and intermediate-risk groups (p = 0.941), these groups were combined into a integrated low/intermediate risk group. In the very low-, low/intermediate-, high- and very high risk groups, the 8-year cumulative incidences of imatinib-therapy failure were 5% (0,11%), 18% (12, 24%), 30% (18, 41%) and 43% (24, 62%;p < 0.001). There were significant differences in imatinib-therapy failure rate among the very low-, low/intermediate- and high-risk groups (p < 0.001 - 0.036), but not between the high and very high-risk groups (p = 0.184). By ELTS score, the 8-year cumulative incidences of imatinib-therapy failure in the low- (n = 223, 69%), intermediate-(n = 81, 25%) and high risk group (n = 20, 6%) were 14% (9, 19%), 32% (22, 43%) and 36% (14, 58%) (p < 0.001), but no significant difference between the intermediate- and high risk group were observed (p = 0.590). The IMTF score exhibited a higher C-statistic values of 0.680 than the ELTS score (0.644). Conclusion IMTF score had better predictive discrimination for imatinib-therapy failure in children with CML-CP compared to ELTS score.