Background: Survival for patients with chronic myeloid leukemia (CML) has increased dramatically since the advent of the first tyrosine kinase inhibitors (TKI) targeting the fusion gene BCR-ABL. However, survival has improved more for younger patients than for older patients. The reasons for this disparity are unclear. Here, we examine changes in survival for patients with CML by age between 2007-10 and 2011-14. In addition, we perform a subgroup analysis of patients known to have received chemotherapy.

Methods: Data were extracted from the Surveillance, Epidemiology, and End Results (SEER) database, using the SEER18 dataset. Patients with ICD-10 code C92.1 were included in the analysis. Period analysis was used to calculate 5-year relative survival. Limited data on use of chemotherapy (yes or no/unknown) is available in the SEER dataset.

Results: A total of 9209 cases of CML were found in the SEER18 database from 2007-14. Of these, 6868 were documented to have received chemotherapy. Overall 5-year relative survival increased from 69.0% in 2007-10 to 74.4% in 2011-14 (see table). Age specific analysis showed that there was little or no change for patients age 15-44, but point estimates of survival increased between 2007-10 and 2011-14 for all other age groups, although the differences were not statistically significant for individual age groups. Survival for patients age 45-54 was lower than for those aged 15-44 in 2007-10 but similar in 2011-14. However, survival continued to be lower for those aged 55 and older compared to younger patients, despite the observed increases in point estimate of survival for these patients.

When analysis was limited to those patients documented to have received chemotherapy, the pattern of survival observed, i.e. increases in survival in 2011-14 compared to 2007-10 in all groups except for age 15-44 but decreased survival with increasing age, was unchanged.

Conclusions: Survival increased for patients with CML age 45 and older between 2007-10 and 2011-14. However, there was a persistent disparity in survival for older patients. This disparity decreased only minimally when only patients known to have received chemotherapy were included, suggesting that chemotherapy usage is not the primary problem. Older patients may be treated less optimally despite receiving at least some chemotherapy, may suffer more fatal adverse events in treatment, or may have more resistant disease compared to younger patients. Further research into the optimal treatment of CML in older patients is needed to improve survival in this population.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.