Analytical Assessment of an IVD Multiplex BCR-ABL1 Monitoring Assay Directly Aligned to the WHO Primary Standards and Reporting on Two Continuous Scales (%IS and MR)

Introduction: Quantification of BCR-ABL1 e13a2/e14a2 fusion transcripts (major breakpoint) of t(9;22) assesses tumor burden in CML. The International Scale (IS) was established to standardize the reporting of these transcripts against a common baseline. As newer TKI therapies create more significant clinical responses, an assay should confidently detect molecular responses of ≥4.5 logs below baseline (0.0032%IS or MR4.5). Various reporting formats have evolved over time, creating a non-contiguous language of monitoring: baseline, 10%IS, 1%IS, MMR, MR4, MR4.5, CMR. We describe the analytical assessment via method comparison of (1) a laboratory developed test using commercial BCR-ABL1 RUO reagents that require conversion to %IS values, and (2) an IVD multiplex assay system directly reporting continuous MR values and %IS values via automated software analysis.

Methods: We assessed 76 specimens using the currently validated LDT that uses BCR/ABL1 Quant reagents (RUO, Asuragen) that target Major (e13a2 and e14a2) and minor (e1a2) transcripts: RT performed on the ABI 9700 and qPCR on the ABI 7500 using SDS v2 software (RUO) followed by Excel-based determination of BCR-ABL1:ABL1 ratios. These LDT ratios were then converted to %IS values using a conversion factor generated using separate ARQ IS Calibrator Panels (Asuragen) (RUO). The same specimens were assessed with the QuantideX® qPCR BCR-ABL IS Kit (IVD, Asuragen) that targets Major (e13a2 and e14a2) transcripts. Both RT and qPCR steps were performed on the ABI 7500 Fast Dx using SDS v1.4 software (IVD). Both BCR-ABL1 assays contained Armored RNA Quant (ARQ) technology in blends of nuclease-resistant BCR-ABL1 and ABL1 RNA transcripts. In the IVD kit, a single 4-point standard curve using ARQ blends mimics the WHO Primary BCR-ABL1 reference materials and accounts for the relative batch run-specific efficiency of the RT step by generating cDNA in parallel to clinical samples. Automated software analysis of SDS files directly generated %IS and MR values, and included an algorithm to ensure sufficient ABL1 detection in negative samples.

Results: The correlation of MR values between methods was excellent, with a Pearson R correlation coefficient of ≥0.95. Within duplicates, the IVD assay resulted in a Pearson R of 0.996 across the dynamic range (MR0.2 to MR4.5), supporting singleton testing as claimed in the Instruction for Use. The bias between methods was uniform (slope near zero) and negligible (mean difference of 0.1 MR unit across the range). This is further supported by a mean difference of ≤3-fold between %IS values of the two methods, with 90% (28/31) of specimens within the reportable range being within 5-fold. The concordance for MMR (MR3 or 0.1%IS) was 91%. Concordance of positivity was 96% (65/68), apparently due to the higher analytical sensitivity of the IVD test for specimens up to MR4.2 (LOD of IVD is MR4.7 or 0.002%IS). Despite deep analytical sensitivity, the IVD system maintains analytical specificity where minor breakpoint BCR-ABL1 e1a2 specimens up to 74% ratio were true positive in the RUO assay and true negative in the IVD assay. Both tests showed excellent linearity using a commercially available cell line-based dilution series (Invivoscribe, RUO).

Conclusions: The QuantideX qPCR BCR-ABL IS Kit improved workflow with streamlined reagent formulation, multiplex assay format, and automated software analysis. It facilitated assessment on the IS without conversion through kit-integrated ARQ materials traceable to the WHO Primary, and generated continuous value results sufficient for studies in deep molecular responses.