To the editor:
We have analyzed with great interest the recent correspondence by Warshawsky et al describing 2 cases of myeloproliferative neoplasms (MPNs) where the Ipsogen MutaScreen kit failed to identify JAK2V617F mutations when another variant was present.1 These variants were, however, detected with a “home-brew” melting curve analysis method.2 The authors therefore considered that the real-time quantitative polymerase chain reaction–based MutaScreen kit is inferior to other technologies and generates a potential clinical risk. We do not support these views and would like to discuss these cases within the appropriate context.
Patient 1, with polycythemia vera (PV), was homozygous for 2 mutations in codon 617 (G1849T and C1851T) and 1 in codon 618. The MutaScreen assay reported “no signal,” and the test was not interpretable.1 This cannot be considered as a false-negative result. Patient 2, with essential thrombocythemia (ET), was heterozygous for 2 mutations in codons 617 and 618, and the MutaScreen assay reported “wild-type.”1 This is a real false-negative result.
The MutaScreen kit has been designed to detect the JAK2V617F G1849T mutation with high specificity, and the probe used cannot bind to JAK2V617F in the presence of 1 or more mismatches. The assay performance has been systematically assessed,3,4 and high concordance with “home-brew” methods and reliable detection of JAK2V617F were shown.
It is well known that single nucleotide polymorphism detection methods using primer annealing can miss rare variants or mutations. Conversely, these can be detected by melting curve analysis. However this method has moderate to poor analytical sensitivity (5%-10% depending on the instrument used)2 and the likelihood of not detecting a clinically relevant mutation is high. Biologists and clinicians therefore face a classical trade-off: guaranteed identification of patients with low (1%-10%) JAK2V617F allele burden (up to 20% of ET and 5% of PV patients; C. Marzac, unpublished data, May 31, 2010) versus the ability to detect rare variants reported in approximately 0.2% of all JAK2V617F carriers.5
The absence of JAK2V617F does not mean absence of MPN. Such a conclusion would constitute inappropriate use of the MPN World Health Organization guidelines,6 which state that the presence of JAK2V617F or a similar mutation is only one of 2 possible main criteria for PV, and one of the major criteria for ET and primary myelofibrosis. They also indicate that if JAK2V617F is not detected in the presence of clinical manifestations consistent with MPN, additional tests are required to propose a final diagnosis. The MutaScreen instructions for use7 state that the absence of JAK2V617F does not exclude the presence of other mutations. Therefore, a missed JAK2V617F mutation does not create a clinical risk for patients provided they are diagnosed and managed according to the most recent standards.
In conclusion, we believe that the Mutascreen assay has the appropriate design and performance characteristics to allow accurate detection of JAK2V617F. Every molecular laboratory should select and evaluate available assays based on performance and potential limitations, their objective when using the products, the regulatory framework, and current practice guidelines.
Contribution: H.P.-S.-P. and F.H. contributed equally to writing of the article.
Conflict-of-interest disclosure: H.P.-S.-P. and F.H. are employees of Ipsogen SA.
Correspondence: Hélène Peyro-Saint-Paul, MD, Ipsogen SA, Luminy Entreprises, Case 923, 163 Ave de Luminy, 13288 Marseille cedex 9, France; e-mail:.
- © 2010 by The American Society of Hematology