Abstract
Polycythemia vera (PV) arises due to a somatic mutation(s) of a single hematopoietic stem cell leading to clonal hematopoiesis. Greater than 80% of PV patients carry a somatic mutation in JAK2 (V617F). Growing evidence suggests that increased frequency of the JAK2V617F allele may have a prognostic impact on certain clinical aspects of PV, and, possibly, in other myeloproliferative disorders associated with this mutation. We have developed a novel approach to primer design for Real-Time quantitative allele-specific PCR. Allelic discrimination is enhanced by the combined synergistic effects of an artificial mismatch introduced in the −1 position, starting from the 3′ end of the primer, and the use of a locked nucleic acid (LNA) modified nucleoside placed at the −2 position. We provide evidence that the −2 LNA assists in stabilizing the 3′ end, while the −1 mismatch provides specificity but not stability. The difference in cycle number between the two allele-specific reactions is used to calculate the relative allele frequencies. We demonstrate the robustness, sensitivity and reproducibility of our design. The proportion of mutant JAK2 allele determined by pyrosequencing and kinetic allele-specific PCR was highly concordant with an average allele frequency deviation of 2.6%. Repeated determination of allelic ratios in multiple patient samples was highly reproducible with a standard deviation of 1.5%. We have also determined that the design and assay is highly sensitive; as little as 0.1% mutant allele in 40–50 ng of genomic DNA can be detected. We further tested the applicability of this technique to the analysis of individual BFU-E colonies in order to address the question whether the JAK2V617F is the disease initiating mutation. Less than 10% of a single isolated BFU-E colony, originating from a single progenitor, is sufficient for determination of allele frequency. The remainder of the colony may be used for other analyses. A proportion of 0 or 50 or 100 percent JAK2 mutant allele is expected from each individual BFU-E colony, which was indeed observed. However, when we tested granulocytes from PV females, wherein the granulocytes were found to be clonal by the X-chromosome transcriptionally based clonality assay, we found 3 females <50 (27.5 ±11) and 7 females >50 (75 ±10.5)percent mutant JAK2 allele frequencies. This result suggests the presence of a heterogeneous population of cells with differing genotypes regarding the JAK2 mutant allele, and is further supported by our genotyping results with individual BFU-E colonies as described above. Our PV data suggest that the JAK2V617F may not be the PV initiating mutation. This novel primer design is simple, does not require tedious optimization of reaction conditions, and can be applied to any kinetic PCR platform for reliable and sensitive determination of allele frequencies. Potential applications are varied, such as, quantitative determination of mosaicism, proportion of fetal cells in maternal circulation, detection of minimal residual disease associated with known somatic mutation (such as reduction of malignant cells by chemotherapy or reappearance of resistant clone), rapid monitoring of efficacy of new drugs in both “in vitro” systems as well as clinical trials, and many others that require quantitation of allele frequencies.
Distribution-based measures of tumor heterogeneity are sensitive to mutation calling and lack strong clinical predictive power