A Novel and Highly-Specific qRT-PCR System for Mutant NPM1 MRD Detection

Background

Prognosis of acute myeloid leukemia (AML) patients can be stratified based on the mutational profiles of their leukemia genome. NPM1 mutations, which are found in approximately 20% of adult AML cases, indicate favorable prognosis, and mutant NPM1 (mNPM1) mRNA levels serve as an excellent biomarker for assessing minimal residual disease (MRD) status. In fact, quantitative measurement of mNPM1 mRNA levels is highly recommended to guide therapy for NPM1-mutated AML patients in the recent ELN (European Leukemia Net) guidelines. However, lack of standardized and/or certified MRD detection systems that meet industry standards for clinical laboratory testing precludes clinical implementation in many countries worldwide, including Japan. Here, we present a novel and highly sensitive mNPM1 q-PCR assay system that enables the detection of mRNA levels of multiple mNPM1 types (types A, B and D) simultaneously via a single assay with high specificity.

Aims

The aim of this study is to develop a highly-sensitive real-time quantitative reverse transcription PCR (qRT-PCR) assay system that specifically detects mNPM1 mRNA in accordance with ELN guidelines, which recommend a sensitivity of 10^-4 to 10^-5 and a measurement range of a 4 log reduction (LR4).

Methods

We designed a fluorescence-labeled oligo probe that specifically hybridizes to all possible mNPM1 cDNA (type-A, -B and -D), and a pair of forward and reverse PCR primers flanking the probe binding sequence. To prevent hybridization of the mutant-specific probe to the wild-type (WT) NPM1 sequence, a WT-specific bridged nucleic acid (BNA) clamp was also designed. RNA was extracted from bone marrow (BM) mononuclear cells (MNC), and qRT-PCR was performed as a single assay, enabling a turn-around-time (TAT) of about 3 hours.

We first generated a series of mock samples by mixing synthetic mNPM1 RNA (types A, B and D) with RNA extracted from NPM1-WT AML BMMNCs at different dilutions ranging from a few mNPM1 copies to 10⁷ mNPM1 copies per sample. We then assessed the analytical performance of our system using a Qiagen ipsogen® NPM1 mut A, B&D MutaQuant Kit as a comparative method. We also validated our system using a total of 7 patient samples obtained from either the Kyushu Clinical sample NETwork (KCNET), a biobank at Kyushu University Hospital or ProteoGenex in accordance with the Declaration of Helsinki under an approved Kyushu university institutional research protocol (#4-142).

Results

We observed a high correlation in quantitative values of mNPM1 mRNA levels between the two methods (R²≥0.9). The detection limit, which was defined as a ratio (%) of mNPM1 copies normalized to those of ABL1, of our assay system was about 0.01%, and the sensitivity was 10^-4. Measurable range of detection spanned from a few mNPM1 copies to 10⁷ copies per assay, achieving LR4. Importantly, specificity and sensitivity of the system were also validated using clinical samples: mNPM1 mRNA signals were evident only in the NPM1-mutated AML cases with high sensitivity, but absent in WT NPM1 AML cases.

Conclusion

We have developed a novel and highly sensitive/specific qRT-PCR system for mNPM1 MRD detection in accordance with current ELN AML MRD guidelines. The assay is a cost-effective one-step qRT-PCR system, in which reverse transcription and PCR are performed on the same plate, with a short TAT and less hands-on steps. We are in the process of validating the utility and feasibility of the system in a clinical trial setting. With further validation, our assay system could serve as a standardized kit for mNPM1 MRD detection.