Figure 1.
Quantification bias in the assessment of BCR-ABL1 transcript levels by using ABL1 as control gene.
Localization of primers (arrows) and probes (rectangles) used for the relative quantification of the M-BCR-ABL1 transcript according to the EAC protocol [16], [17] are represented. Boxes represented the exon number. Red signs indicate primers and probe relative to the fusion transcript, blue signs to the control gene ABL1. The ABL1 primers localization explains both amplification of ABL1 and BCR-ABL1 during the Q-RT-PCR reaction. ENF = forward primer, ENPr = TaqMan reverse probe, ENR = reverse primer.
Table 1.
Patient Characteristics (N = 84).
Figure 2.
The use of GUS as control gene would need the introduction of a conversion factor.
The conversion from BCR-ABL1/ABL1 to BCR-ABL1/GUS values would need the introduction of an additional conversion factor to define the major thresholds used in the clinics for the MRD monitoring, as exemplified for the MMR threshold. (a) Graph A plots the (BCR-ABL1/ABL1)IS % value of each patient close to the MMR threshold. The mean of 42 samples was 0.092%, 2 s.d. range 0.038–0.147%. (b) Graph B plots the BCR-ABL1/GUS % value of the same samples. The mean was 0.036% and 2 s.d. range 0.010–0.062%.
Figure 3.
Box plot of the bias between (BCR-ABL1/ABL1)IS and BCR-ABL1/GUS measurement in the 3 groups of samples.
The bias (leading to conversion factor calculation) was compared between 3 groups of samples: diagnostic samples, samples with 6–14% and 0.1% transcript levels. It is significantly higher in the group of diagnostic samples (*** represents p-value<0.001; ns, non-significant).
Table 2.
Summary of the results obtained in the 3 groups of samples.
Figure 4.
Mean copy number of both control genes for the 3 groups of samples.
GUS and ABL1 copy number is significantly higher at diagnosis (*** represents p-value<0.001; ns, non-significant).
Figure 5.
The copy number of GUS and BCR-ABL1 are highly correlated.
This supports the hypothesis that GUS may be overexpressed in leukemic cells compared to normal cells.
Table 3.
Prediction of MMR achievement at 12 months with the use of different EMR markers.
Figure 6.
BCR-ABL1/ABL1IS transcript evolution of patients according to their classification with the 3 EMR markers.
Three markers were used to assess early molecular response (EMR) with the use of BCR-ABL1/ABL1IS ratio: transcript level at 3 months, halving time and log reduction. Patients are classified as “high-risk” or “low-risk” according to the cut-offs described for each marker (see table 3) and their transcript level evolution at 6 and 12 months is reported. A: In most cases, all three markers were consistent to classify these patients as high risk (left part) or low risk (right part). B: In some discordant cases one of the markers showed a predictive value that differed from the other two. In this latter case, we took into account both concordant markers to classify the patients.