ABSTRACT
Molecular monitoring of BCR-ABL1 transcripts is a critical prognostic indicator of treatment response in chronic myeloid leukemia (CML). Quantification of BCR-ABL1 transcripts using ABL1 or GUSB as control genes on the early molecular response (MR) to frontline nilotinib was studied using data from 60 patients with chronic-phase CML from the Evaluating Nilotinib Efficacy and Safety in Clinical Trials as First-Line Treatment (ENEST1st) substudy. Effects of BCR-ABL1/ABL1 and BCR-ABL1/GUSB ratios at early time points as independent variables on subsequent MR were determined by logistic regression analyses and predictive cut-off values determined by receiver operating curve analyses. From day 45, concordance was found for both control genes' early transcript kinetics and ability to predict subsequent deep MR at 18 months. From baseline to 3 months, transcripts descended linearly with both control genes. Use of ABL1 allowed for an earlier prediction (2 months) of subsequent MR than with GUSB (3 months), with cut-off values of 1.5% and 0.19%, respectively. The dynamic determination of BCR-ABL1 transcripts using either internal control gene is valid and predictive of subsequent MR. The use of GUSB to predict an earlier and more accurate response than ABL1 is not supported in the results. Accurate early indicators of MR are essential to identify patients likely to have inferior outcomes who may benefit from treatment with an alternative tyrosine kinase inhibitor.
