Development of a predictive population survival model according to the cytogenetic response rate for patients with chronic myeloid leukemia in the chronic phase

Abstract

This study sets out to investigate whether the proportion of patients with chronic myeloid leukemia (CML) in the chronic phase who achieve a major cytogenetic response (MCR) can be used as the basis for estimating long-term survival through the use of modeling. Data from seven randomized controlled trials of drugs to treat patients with CML in the chronic phase were used to explore the association between MCR and survival by way of regression analysis. The estimated weighted odds ratio for the survival of those who achieved an MCR when compared with those who did not was 7 (95% CI 5 – 11) at 2 years and 5 (95% CI 3 – 8) at 4 years. Four long-term survival models were subsequently constructed. All models were found to be robust to variations in the data included. Model D was favored using the ‘Ockham's razor’ principle; it suggests that the median survival may be increased by 1.8 years for every 25 percentage point increase in MCR rate. The results support the use of the proportion of patients with CML in the chronic phase with an MCR to estimate overall long-term survival.

Keywords:

• Predictive model
• CML
• cytogenetic response
• survival
• drug evaluation