Objectives and background: Early molecular response (EMR) to tyrosine kinase inhibitors (TKI) therapy defined as BCR-ABL1 transcript level ≤ 10% at 3 months is a well-established predictive factor in CML patients. Recently, several study groups presented the concept of BCR-ABL1 decline velocity or "halving time" as a new predictive factor for optimal TKI treatment response. Different reference genes were used in the quantitative RT-PCR (qRT-PCR) analysis to calculate halving time and EMR rate, potentially influencing results of these analyses. The aim of the study was to compare ABL1 and beta-glucuronidase (GUSB) as reference genes used for calculation of halving time and EMR rate in chronic phase CML patients (pts) treated with imatinib 400 mg/d in the clinical settings.

Methods: A total of 52 chronic phase CML pts with blood samples collected at baseline and at 3 months of imatinib therapy were investigated. Transcript levels of BCR-ABL1 were determined by qRT-PCR with the use of ABL1 and GUSB as control genes for each patient. Halving time for BCR-ABL1 transcripts was calculated using baseline transcript levels, transcript levels at 3 months of treatment, and number of days between these two points according to methodology reported elswhere1. Using receiver operating characteristic (ROC) analysis, the optimal halving time thresholds for achievement of EMR for each control gene were determined. Optimal thresholds along the ROC curves were calculated using the Youden index.

Results: 40 of 52 (76.9%) and 44 of 52 (84.6%) pts achieved EMR with BCR-ABL1 transcripts levels calculated with GUSB and ABL1 as control genes, respectively (p=0.55). In 4 pts there was no BCR-ABL1 transcript reduction from baseline at 3 month using GUSB as a control gene resulting with negative halving times. The GUSB based halving times of these 4 patients were imputed to the longest positive halving time (8530 days) calculated for the patient with the smallest BCR-ABL1 decline. Transcript reduction occurred in all pts when ABL1 was used as a control gene. ROC curve analysis resulted in the optimal halving time thresholds for achievement of EMR of 39 days (area under the curve [AUC ] 0.96, ;95% CI 0.9-1.0, ) and 31 days (AUC 0.95; 95% CI 0.87-1.0, ) for GUSB and ABL1 control gene, respectively.

Discussion and Conclusions: Lack of linearity in qRT-PCR analysis observed in blood samples collected in CML patients at diagnosis, with high amount of BCR-ABL1 and wild type ABL1 transcripts could affect the accuracy of early kinetics decline of BCR-ABL1 transcript calculations when ABL1 is used as a reference gene. The analysis with the use of GUSB as a control gene is not biased by this phenomenon. In this study, a non - statistically significant differences in the rate of EMR achieved at 3 months of imatinib therapy and halving time were observed. Analyses performed in our study with ABL1 and GUSB as reference genes suggest that both GUSB and ABL1 are able to identify relevant cut-offs for EMR achievement and could be used for halving time calculation.


  1. Branford S, Yeung DT, Parker WT, et al. Prognosis for patients with CML and >10% BCR-ABL1 after 3 months of imatinib depends on the rate of BCR-ABL1 decline. Blood 2014 Jul 24;124(4):511-8


Zawada:Novartis: Speakers Bureau. Czekalska:Novartis: Speakers Bureau. Sacha:Pfizer: Consultancy, Honoraria, Speakers Bureau; Incyte: Consultancy, Honoraria, Speakers Bureau; BMS: Consultancy, Honoraria, Speakers Bureau; Novartis: Consultancy, Honoraria, Speakers Bureau; Adamed: Consultancy, Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution