Estimating prognostic relevant cutoff values for a multiplex PCR detecting BCR::ABL1 in chronic myeloid leukemia patients on tyrosine kinase inhibitor therapy in resource-limited settings

The prognosis of chronic myeloid leukemia (CML) on tyrosine kinase inhibitor (TKI) treatment is based on the quantification of BCR::ABL1 fusion gene transcript copy number, harmonized by an international scale (IS) based on TaqMan-based real-time quantitative PCR (qRT-PCR). In Ethiopia, as in most low- and middle-income countries (LMICs), access to standard diagnostic, follow-up, and prognostic tools is very limited, and it has been challenging to strictly follow international guidelines. This seriously compromises clinical outcome, despite the availability of TKIs through the Glivec International Patient Assistance Program (GIPAP). Multiplex PCR (mpx-PCR), conventionally regarded as a screening tool, offers a potential solution to this problem. A total of 219 samples from confirmed CML patients were assayed. In reference to qRT-PCR, the AUC of ROC curve for mpx-PCR was 0.983 (95% CI: 0.957 to 0.997). At the optimum cut-off value, equivalent to BCR::ABL1 (IS) transcript copy number of 0.6%, the specificity and sensitivity were 93% and 95%, respectively, with 94% accuracy. Albeit the sensitivity and accuracy of mpx-PCR decrease below the optimum cutoff of 0.6% (IS), the specificity at 0.1% (IS) was 100%, making it an attractive means to rule-out relapse and drug non-adherence at later stages of treatment, which is particularly an issue in a low income setting. We conclude that the relative simplicity and low cost of mpx-PCR and prognostic relevant cutoff values (0.1-0.6% IS) should allow its use in peripheral clinics and thus maximize the positive impact of TKIs made available through GIPAP in most LMICs.

Automated summary

In low- and middle-income countries, including Ethiopia, access to standard diagnostic and prognostic tools for chronic myeloid leukemia (CML) is limited. This study found that multiplex PCR (mpx-PCR) can serve as a screening tool with high sensitivity and specificity, helping to rule out relapse and non-adherence to medications, thus reducing treatment risks.