Counterpoint: There is a best duration of deep molecular response for treatment-free remission, but it is patient-specific, and that is the challenge

There is considerable clinical and scientific interest in identifying reliable predictors of treatment-free remission in chronic myeloid leukaemia. Most predictors have been identified from non-randomized clinical trials or retrospective cohorts that could be subject to bias. The validity of predictive factors, such as duration of treatment or of deep molecular response, has been questioned. We briefly review the relevant data and the potential for bias, arguing that the risk of bias may be overstated, and that accumulating data strongly suggest that depth and duration of molecular response are critical factors to enable us to predict the probability of treatment-free remission.