Introduction - Targeted therapy of chronic myeloid leukemia (CML) using tyrosine kinase inhibitors (TKI) very effectively suppresses the growth of leukemic clone to a clinically safe level of the disease. Therefore sensitive monitoring of molecular response (MR) based on standardized measurements of BCR-ABL1 transcript levels plays an important role for prediction either of optimal response or progression. Thus, its sensitive and accurate detection is very important for the individualized therapy, especially for a TKI stopping treatment management in patients achieving long-term deep MR. While the transcript levels may not exactly correspond to the number of leukemic cells, BCR-ABL1 quantification at the DNA level using patient-specific assays may bring more accurate information about residual disease. Moreover, BCR-ABL1 DNA quantification may contribute to a stratification of patients for whom the cessation of TKI therapy may be safe.

Objectives - This work is focused on the comparison of BCR-ABL1 quantity at DNA and mRNA level in CML patients after TKI initiation until the MR achievement, and after TKI discontinuation within the EURO-SKI clinical trial (Europe Stops TKI in CML), respectively, and on determination of differences on BCR-ABL1 positivity/negativity of samples analyzed. Additionally, the BCR-ABL1 DNA data obtained from the measurements by qPCR and droplet digital PCR (ddPCR) were compared.

Methods - Our study included analyses of 232 samples of 13 patients with CML in chronic phase (CP), who achieved deep MR during TKI treatment. Four of 13 patients (total of 85 samples analyzed) discontinued the treatment within the EURO-SKI study. The levels of BCR-ABL1 mRNA were determined by standardized RT-qPCR method and DNA levels were detected by qPCR with developed patient-specific qPCR assays after BCR-ABL1 genomic fusions characterization. The ddPCR method was performed on 72 samples using QX-200 Droplet Digital PCR System (Bio-Rad). The determined ratios of BCR-ABL1/GUSB (mRNA) and BCR-ABL1/albumin (DNA level) of follow-up samples of every patient were related to the diagnostic sample (considered as 100% level) for the most appropriate comparison of mRNA and DNA levels.

Results - Standardized RT-qPCR and patient-specific qPCRs were able to detect the mRNA and DNA BCR-ABL1 levels, respectively, reduced by more than 5 logs from the levels at the time of diagnosis. We observed a significant correlation between the mRNA and DNA levels by comparing 148 paired BCR-ABL1 positive results (correlation coefficient r2=0.9055; P˂0.0001). When comparing the frequency of the BCR-ABL1 negative/positive results at the DNA vs. mRNA levels, BCR-ABL1 positivity at DNA level was found in 21 samples which were BCR-ABL1-negative at mRNA level. BCR-ABL1 mRNA was found either negative or positive at MR5.0 in 3/4 EURO-SKI patients within the period between MR achievement until the TKI discontinuation (19 samples analyzed). The level of BCR-ABL1 DNA was continuously positive in this period before the TKI cessation. The molecular relapse (i.e. MMR loss) was observed in all those 3 patients several months after the therapy cessation. The levels of BCR-ABL1 DNA detected by ddPCR method correlated with DNA qPCR results (r2=0.9684; P˂0.0001) and the obtained values were not significantly different.

Conclusion - The levels of the BCR-ABL1 mRNA and DNA in peripheral blood (PB) correlated significantly, thus both consistently reflecting the course of CP-CML at the molecular level. Although much more patients with TKI cessation need to be tested for the presence of BCR-ABL1 DNA, our data suggest that quantification at the DNA level appears to be (at least in some cases) more sensitive method for detection of residual leukemic cells in PB in comparison to the mRNA quantification. The BCR-ABL1 DNA measurement may have a potential prognostic significance in deciding for eventual withdrawal of TKI therapy, which is currently being investigated within the clinical trials. Digital PCR enables absolute quantification without a need of patient-specific calibration curves and with the potentially higher sensitivity and accuracy for BCR-ABL1 DNA quantification.

This work was supported by the project 15-31540A of the Czech Health Research Council, GAUK 554214 and EURO-SKI Research Consortium.

Disclosures

Klamova:Novartis: Consultancy, Honoraria, Research Funding; Bristol Myers-Squibb: Consultancy, Honoraria, Research Funding. Saussele:ARIAD: Honoraria; Pfizer: Honoraria, Other: Travel grant; BMS: Honoraria, Other: Travel grant, Research Funding; Novartis Pharma: Honoraria, Other: Travel grant, Research Funding. Mahon:Pfizer: Consultancy, Research Funding; BMS: Consultancy, Research Funding; Ariad: Consultancy, Research Funding; Novartis: Consultancy, Research Funding. Machova Polakova:Bristol Myers-Squibb: Consultancy, Honoraria, Research Funding; Novartis: Consultancy, Honoraria, Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution