Chronic Myelogenous Leukemia in the Era of Tyrosine Kinase Inhibitors: Is Sokal Score Still a Valid Predictor of Prognosis?

In the era when the Sokal score was developed (1984), Tyrosine kinase inhibitors (TKI's) were not available for treatment of Chronic Myelogenous Leukemia (CML). With the treatments available then, the 2-year survival in the low risk group was 90% and in the intermediate and high-risk groups was 65%. Since TKI's were approved for treatment of CML in 2001, overall estimated survival at 5 years has improved to 89%.

To evaluate risk stratification by Sokal score as an indicator of prognosis in patients with CML in the present era of tyrosine kinase inhibitors.

Sokal Score is a prognostic score used to risk stratify patients as low-risk, intermediate-risk and high-risk based on diagnosis based on the spleen size, platelet count, age and blast count. Seventy-eight (78) subjects diagnosed with CML and treated at John H. Stroger Hospital of Cook County between 2000 and 2011 were retrospectively analyzed after institutional review board approval. Ten (10) subjects with blast crisis at diagnosis were excluded from this study. Effect of race, sex, age, white cell count, amount of reticulin fibrosis in bone marrow biopsy, total peripheral blood eosinophil and basophil counts, Lactate Dehydrogenase level, spleen and liver size at diagnosis and, risk stratification by Sokal score on overall survival was evaluated using standard statistical analysis.

In our population, 27/68 (39.7%) were more than 45 years old, 30/68 (44.11%) were African American, 20/68 (29.4%) were Hispanic, male to female ratio was 2.4:1. WBC>100,000 was observed in 25/68 (51.4%), 57/68 (83.8%) had LDH>200IU/L, 50/68 (73.5%) had palpable splenomegaly >5cm at the time of diagnosis and 67/68 (99%) of subjects had reticulin fibrosis in bone marrow. Of these, 35/67 (52.2%) had mild, 12/67 (17.9%) had moderate and 20/67 (29.8%) had severe fibrosis. According to Sokal score risk stratification, 22/68 (32.3%) were low risk, 40/68 (58.8%) were intermediate risk and 6/68 (8.8%) were high risk at diagnosis. Five year overall survival (OS) in our population was 90%. On univariate analysis, risk stratification by Sokal score at diagnosis was statistically significant for Overall Survival. Subjects with low risk score had better OS as compared to intermediate and high-risk subjects (P=0.007, HR 11.596, 95%CI 1.946–69.103). Similarly, subjects with intermediate score had better OS compared to high-risk group (P=0.007, HR 0.140, 95%CI 0.020–0.539).

Sokal Score is still a valid predictor of prognosis in this new age of tyrosine kinase inhibitors and can still be used for risk stratification at diagnosis for chronic phase CML patients.

No relevant conflicts of interest to declare.