Infrequent occurrence of MPL exon 10 mutations in polycythemia vera and post‐polycythemia vera myelofibrosis

Polycythemia vera (PV) is molecularly well characterized with 96% and 3% of patients exhibiting JAK2V617F and JAK2 exon 12 mutations, respectively [1, 2]. MPL exon 10 mutations have not been identified in PV; among the Philadelphia-negative classic myeloproliferative neoplasms (MPN), MPL mutations appear to be restricted to essential thrombocythemia (ET) or myelofibrosis (primary or post-ET) with overall mutation frequency of 4 and 8%, respectively [3-6]. Although JAK2V617F and MPL mutations have been identified concurrently in ET and primary myelofibrosis (PMF), such occurrences are relatively infrequent, and this combination has not been described in PV [3-6]. We screened 168 patients, 149 with PV and 19 with post-PV MF, for presence of JAK2 and MPL mutations; 162 patients (96.4%) and two patients (1.2%) harbored JAK2 and MPL mutations, respectively. Both MPL mutated patients, one each with PV and post-PV MF, were also JAK2V617F-positive. Thus, PV patients may infrequently harbor MPL mutations suggesting the disease is molecularly more complex than previously appreciated. It reinforces the view that currently identified MPN-relevant mutations are neither mutually exclusive nor are they disease specific. We screened a total of 168 patients for presence of JAK2V617F and MPL exon 10 mutations; of these, 149 patients had PV and 19 patients post-PV MF. Overall 162 patients (96.4%) harbored JAK2 mutations; of these, 160 patients exhibited JAK2V617F and two patients JAK2 exon 12 mutations (both with PV). JAK2 mutation frequency in PV and post-PV MF was 95.9 and 100%, respectively. Two patients (1.2%), one each with PV and post-PV MF, harbored a MPL exon 10 mutation; thus, MPL mutation frequency was < 1% in PV and 5% in post-PV MF in this cohort. The PV patient harbored the MPLW515R allele and the post-PV MF patient MPLW515K (Fig. 1). Both patients with MPL mutations harbored JAK2V617F concurrently. DNA sequence traces showing concurrent presence of JAK2V617F and MPL exon 10 mutations in a patient with (a) PV and (b) post-PV myelofibrosis (post-PV MF). The former patient (PV) harbored MPLW515R and the latter (post-PV MF) MPLW515K. The arrows indicate the nucleotide substitution for missense mutations. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.] Clinical and laboratory information for the two MPL mutated patients is as follows: Patient #1 was a 45-year-old male who diagnosed with PV approximately a year prior to his initial presentation at our institution. A review of his laboratory records showed peak hemoglobin (Hgb) value of 19.4 g/dL. The serum erythropoietin level was undetectable or demonstrated to be abnormally low on several occasions. Phlebotomies were instituted after the diagnosis was established; however, no cytoreductive therapy was initiated. At the time of Mayo referral, the laboratory values were as follows: Hgb 15.2 g/dL, hematocrit 47.1%, red blood cell mean corpuscular volume 75.9 fl, serum ferritin level 6 μg/L, and serum erythropoietin level 1.9 mU/mL (normal range 4-24). Bone marrow examination showed morphologic findings consistent with treated PV (absent iron stores); reticulin stain showed normal reticulin fibers and cytogenetic studies revealed a normal karyotype. Analysis of DNA collected at the time of referral showed presence of JAK2V617F and the MPLW515R mutation in the heterozygous state (Fig. 1, top panel). Patient #2 was a 54-year-old female with a diagnosis of blastic transformation of post-PV MF; at the time of referral to our institution she had re-achieved chronic-phase disease after having received an allogeneic stem cell graft from her HLA-identical sibling brother. PV had been diagnosed 22 years prior; subsequent to the interim fibrotic transformation of PV, she had received multiple therapies including hydroxyurea, erythropoiesis stimulating agents, androgens, and had also undergone splenectomy. At referral, the patient was pancytopenic and bone marrow examination showed myelofibrosis with approximately half the metaphases exhibiting recipient-derived complex cytogenetic abnormalities. Analysis of DNA collected at the time of referral showed presence of JAK2V617F and the MPLW515K mutation in the heterozygous state (Fig. 1, bottom panel). The identification of MPL mutations in PV and post-PV MF was somewhat unexpected given previous reports of the putative anti-erythropoietic effect of such mutations. In contrast to JAK2V617F-mutated patients, we and others have failed to obtain erythropoietin-independent erythroid colonies in vitro in studies of PMF or ET patients [3, 7, 8]. Furthermore, in ET, patients harboring MPL mutations exhibit significantly lower hemoglobin values, higher serum erythropoietin levels, and reduced total and erythroid bone marrow cellularity, as compared to MPL wild-type/JAK2V617F positive patients [3, 6]. Also, some [4] but not other studies (Pardanani et al., manuscript submitted) in PMF patients have identified an association between MPL mutations and more severe anemia as well as increased dependency for red cell concentrate transfusions. Although this study clearly demonstrates that JAK2V617F and MPL mutations can coexist in PV/post-PV MF, the cumulative published data to date suggest that such occurrences are likely to be rare. Likewise, in ET the mutation concurrence rate is likely ≤ 1% [3, 6]. In contrast, in PMF, the published mutation concurrence rate has varied widely from 0% (Pardanani et al., submitted) to up to 38% [4, 5], likely reflecting the smaller patient numbers relative to ET and a more heterogeneous patient population. The number of MPL-mutated PV/post-PV MF patients is too small to draw any conclusions regarding possible clinical or laboratory associations or impact on clinical outcome; the current observations do however further demonstrate the lack of disease specificity of MPN-relevant mutations and call into question the presumed phenotype-modifying properties of such mutations. Given that the MPL mutations in the aforementioned PV/post-PV MF cases were easily identified by direct DNA sequencing and given the observed mutant: wild-type allele peak heights on DNA chromatograms (Fig. 1), the current observations cannot be easily explained by presence of a minor MPL mutant clone on a background of a dominant JAK2V617F harboring clone. Further, at least in the case of the aforementioned PV case, the MPLW515R mutation was demonstrable relatively early during the course of the disease and in the absence of any cytoreductive or potentially mutagenic therapy being administered. A similar early acquisition of MPL mutations has been previously demonstrated in PMF and ET [3, 9, 10]. Although these observations alone do not provide conclusive evidence for the pathogenetic relevance of MPL mutations in PV, it does not exclude this possibility either. The current data do not clarify whether each mutation targets an independent stem cell clone (i.e., biclonal disease) or whether both mutations arise in the same stem cell clone with sequential mutation acquisition (i.e., clonal evolution) [11]. Unfortunately, unavailability of viably preserved cells from the two PV/post-PV MF patients precluded genotyping of individual hematopoietic colonies grown in vitro that may have clarified this issue. This study was approved by the Mayo Clinic institutional review board. Patients provided informed written consent for study sample collection as well as permission for its use in research. Inclusion to this study required availability of archived peripheral blood and/or bone marrow sample collected at the time of diagnosis or first referral. The diagnoses of PV and post-PV MF were according to the World Health Organization criteria [12]. Screening of patients for MPL exon 10 genetic variations was performed using the High Resolution Melting (HRM) Dye assay using the LightCycler® 480 Instrument (Roche-Applied-Science, Mannheim, Germany). For each 20-μL reaction, 15–30 ng of DNA template was added to a reaction mix of: 1.5-mM MgCl2, 500 nM of both forward and reverse primer (MPL-forward 5′-TAGCCTGGATCTTCTCCTTGGTG-3′; MPL-reverse 5′-GCGGTACCTG TAGTGTGCAG-3′), ddH20, and 1X of enzyme mix (HRM Master Kit, Roche-Applied-Science, Mannheim, Germany). Each reaction was first amplified by real-time PCR using the following cycling parameters: 95°C for 10 min followed by 45 cycles of 95°C for 10 s, 60°C for 15 s, and 72°C for 30 s. After amplification, each target sequence was subjected to an increasing temperature and data from the intercalated HRM Melting Dye signal resulting from dissociation of double-stranded DNA to single-stranded DNA was plotted. Analysis of melting curves was executed using Gene Scanning Software (Roche-Applied-Science, Mannheim, Germany) and was normalized to wild-type sequence. Any sample sequence that deviated from the normal plot was Sanger sequenced for confirmation of presence of MPL genetic variation. Sequencing was performed bidirectionally using two ABI PRISM™ 3730xl DNA analyzers (96-capillary). JAK2V617F mutation screening was performed as previously described [13]. A Pardanani, TL Lasho, and A Tefferi designed the study, contributed patient samples, analyzed the data, and wrote the paper. TL Lasho and C Finke performed the mutational analysis. Animesh Pardanani*, Terra L. Lasho*, Christy M. Finke*, Ayalew Tefferi*, * Division of Hematology, Department of Medicine, Mayo Clinic Rochester, Minnesota.