Mutations in exon 12 of <i>JAK2</i> are mainly found in JAK2 V617F‐negative polycythaemia vera patients

Abstract

Following the identification in 2005 of the recurrent V617F mutation in exon 14 of JAK2 in Myeloproliferative Disorder (MPD) patients, many studies have confirmed the frequency of this mutation in distinct MPD subclasses. The presence of the JAK2 V617F mutation has proved even more useful to distinguish, in patients with true polycythaemia, those with Polycythaemia Vera (PV, 90%JAK2 V617F positive) from those with Idiopathic Erythrocytosis (IE, V617F-negative) (Percy et al, 2006; unpublished observations). The JAK2 V617F mutation is also present in about half of the patients with splanchnic vein thrombosis (Patel et al, 2006), suggesting that occurrence of thrombosis in these patients is caused by an undiagnosed MPD. However, there still exists patients in whom the diagnosis of MPD is highly suspected but the absence of the JAK2 V617F mutation does not allow correct classification. Novel recurrent mutations clustered in a highly conserved region in exon 12 of JAK2 have been described in patients with erythrocytosis (Scott et al, 2007). We took advantage of our cohort of JAK2 V617F-negative patients and analyzed three distinct groups: two groups of patients with increased red cell mass, either with (PV) or without (IE) sufficient criteria for a diagnosis of PV, and one group of patients with splanchnic vein thrombosis. All patients had been previously found negative for the JAK2 V617F mutation using allele-specific real-time polymerase chain reaction (AS-PCR) with a sensitivity of 1–2% as previously described (Kiladjian et al, 2006). Identification of JAK2 exon 12 mutations was performed on DNA extracted from peripheral blood using the AS-PCR assays previously described (Scott et al, 2007). Serum erythropoietin (Epo) levels were measured using a radioimmunoassay (DiaSorin, France) (Schlageter et al, 1990) and Endogenous Erythroid Colonies (EECs) analysed on semi solid media (StemCell Technologies, Vancouver, BC, Canada). Among 47 patients with an increased red cell mass (>25% excess), 26 fulfilled the Polycythaemia Vera Study Group (PVSG) or World Health Organization (WHO) criteria for PV at diagnosis, while 21 were classified as having IE on the basis of a raised red cell mass, with no identifiable cause of secondary erythrocytosis and absence of PV according to PVSG or WHO criteria. Eight cases were found to carry an exon 12 mutation; all of them fulfilled the WHO or PVSG criteria for PV. As shown in Table I, most of these patients were characterized by a young age at diagnosis, a trend that was also observed in the first report of exon 12 mutations (Scott et al, 2007). One patient (Patient 2 in Table I), diagnosed in childhood, was the youngest patient reported to date with an exon 12 mutation at diagnosis. All mutation-positive patients tested (7/7) had a low serum Epo level, a feature frequently reported in this group of patients (5/8 (Scott et al, 2007), 2/2 (Butcher et al, 2007), 5/5 (Pardanani et al, 2007) and 16/17 (Pietra et al, 2007) respectively). Similarly, 4/4 patients tested had EECs, a constant feature of patients with exon 12 mutations (Butcher et al, 2007; Percy et al, 2007; Scott et al, 2007). Interestingly, in previous reports, exon 12 JAK2 mutated patients had either normal levels of other blood cells (Pardanani et al, 2007; Pietra et al, 2007) or occasionally increased levels (Butcher et al, 2007; Scott et al, 2007). In our study, increased white blood cell or platelet counts were noted in four out of the eight cases (Table I). The presence of exon 12 mutations is therefore not restricted to patients with an isolated increase in red blood cells, and an increase in other additional blood cell lineages may be observed. In 7/8 exon 12 mutated cases, the N542-E543del allele was the mutation detected (Table I). This was also the most frequent mutant reported in the other published series: 4/10 (Scott et al, 2007), 2/5 (Pardanani et al, 2007) and 10/17 (Pietra et al, 2007) patients. Therefore, 57% (23 of 40 cases, including our present data) of patients with exon 12 mutations have been reported to carry the N542-E543del mutant. None of the 21 JAK2 V617F-negative patients with increased red cell mass that were diagnosed with IE (unpublished observations) carried a mutation (see Table II for clinical and biological characteristics). In two recent reports, 2/44 and 8/58 patients classified as IE were found to carry exon 12 mutations, six out of the 10 positive patients having the N542-E543del mutant (Percy et al, 2007; Williams et al, 2007). Finally, we analysed DNA from 23 patients with splanchnic venous thrombosis, eight with Budd-Chiari syndrome and 15 with portal venous thrombosis. No exon 12 mutations were detected in these patients (sex: eight females and 14 males; median age: 53 years, range 33–70; median hematocrit: 40% range 27–50). Thus, JAK2 exon 12 mutations were detected in 31% of our JAK2 V617F negative patients with increased red cell mass and PV criteria, with a preponderance of the N542-E543del mutant. This quite low percentage suggests that other genetic alterations may cause PV, but could also be explained by the recent description of new JAK2 exon 12 mutants (Butcher et al, 2007; Pietra et al, 2007) that could be overlooked by the AS-PCR assays we used. One important concern for a diagnostic laboratory will be the choice of a method for detecting these mutations, as the complexity of the several mutations reported does not allow the development of simple high throughput assays as used for the JAK2 V617F mutant. In addition, using a sequencing approach, Butcher et al (2007) detected JAK2 exon 12 mutations only on isolated endogenous erythroid burst-forming units, suggesting that the mutated allele burden could be too small to allow the identification of these mutations by the relatively insensitive sequencing method (sensitivity, approximately 20%), as frequently observed for the JAK2 V617F mutation. Furthermore, in our study, AS-PCR identified two mutated patients that had previously been negative using a sequencing approach (Patients 1 and 3). In conclusion, our data and recently published literature show that exon 12 mutations are very useful markers to identify MPD patients. Their frequency, unlike the JAK2 V617F mutation, is low, suggesting that the search for JAK2 exon 12 mutations should be restricted to patients who fulfil PV diagnostic criteria, have low serum Epo levels and EECs and do not carry the JAK2 V617F mutation. We are very grateful to Marie-Laurence Menot and Nadine Bonnin for technical assistance and sample management and Rose Ann Padua for correcting the manuscript.