Impact of the 2022 Who Classification on Chronic Myeloid Leukemia

The latest edition of the 2022 World Health Organization (WHO) classification for chronic myeloid leukemia (CML) omitted the definition of accelerated phase (AP). In contrast, the International Consensus Classification (ICC) 2022 retained the definition of a three-phase disease: chronic phase (CP), accelerated phase (AP), and blast crisis (BC). The justificative of WHO is that the prognosis of AP has dramatically improved with tyrosine kinase inhibitors (TKI), and high-risk features at diagnosis associated with CP progression and resistance to TKIs are more relevant factors. However, the omission of AP is still a matter of discussion since it may affect the initial dose of tyrosine kinase inhibitors and treatment reimbursement.

Objectives: To evaluate in real practice the impact of the updated 2022 WHO classification on the prognosis of CML patients treated with imatinib in the first line.

Methods: This is a retrospective, single-center, observational study conducted in a university hospital. Between January 2015 and December 2023, we collected data from newly diagnosed CML patients and compared the clinical outcomes using the WHO 2017 and the new WHO and ICC 2022 classifications. Clinical and laboratory data were collected from patients’ charts. Patients were treated with imatinib 400-600 mg/day. We excluded cases with insufficient data to classify disease phase before starting the treatment. Overall survival (OS) and progression-free survival (PFS) were calculated from the start of imatinib until the date of death or last follow-up. The cut-off date for this analysis was May 2024.

Results: We evaluated 139 patients, with median age of 53 years (17-83); 51% were female. At diagnosis: median white blood cells (WBC) were 1172 x10⁹/L (5.4-588.0), basophils 3.2% (0-18), blasts 2% (0-60), platelets 387 x10⁹/L (71-3.862), hemoglobin 11.4 g/dL (3.5-16.6), bone marrow blasts 1% (0-63) and bone marrow basophils 1.5% (0-45%). Sokal risk was low (28%), intermediate (43%), and high (29%). Two patients died before starting imatinib and 137 started imatinib in a median time of 10 days. According to the 2017 WHO and ICC 2022, 90% were in the chronic phase, 6% in the accelerated phase, and 4% in the blast crisis; and according to 2022 WHO, 96% were in the chronic phase and 4% in the blast crisis. There was a significant change in the 2022 WHO classification compared to the 2017 version (P<0.0001%). Eight of 139 (6%) patients changed from AP to CP. Five out of eight patients switched to a second-generation TKI because of failure, and two required allogeneic hematopoietic stem cell transplantation; one died from the procedure. One patient progressed to BC, one patient lost follow-up, and four are still alive. In the total group, there were 22 deaths, 15 (68%) related to CML. OS and PFS, according to Sokal, were 100%, 87%, and 56%, and 100%, 85%, and 56% for low, intermediate, and high risk (both P<0.0001). According to 2017 WHO/ICC 2022, OS was 86% and 70% at 60 and 96 months for CP, 0% for AP, and 25% for BC (P<0.0001); according to 2022 WHO was 85% and 69% at 60 and 96 months for CP and 25% for BC, respectively (P<0.00001). 2017 WHO/ICC 2022, PFS at 60 and 96 months were 85% and 69% for CP and 25% for BC (P<0.0001); according to 2022 WHO, 83% and 68% at 60 and 96 months in CP and 25% in BC (P<0.0001). In the multivariate analysis, the independent factors for OS were basophils, 2022 WHO phase, and age at TKI start (P=0,013;<0.0001; 0.019), and for PFS were 2022 WHO (BC) and age at diagnosis (P<0.0001 and 0.010).

Conclusion: The omission of AP in the 2022 WHO classification did not significantly change the prognosis of our cohort regarding OS and PFS. Age, higher basophil count, and advanced disease at diagnosis remain relevant factors associated with worse outcomes. However, the potential challenges that the omission of AP may pose for individual patients in Brazil and for the clinical hematologist or oncologist should not be underestimated, as higher doses of TKI are only reimbursed for the “old” definition of AP. This underscores the need for further discussion in this area.