JAK2V617F: better diagnostic tool than marrow?

In patients with Budd-Chiari syndrome and with portal vein thrombosis, Kiladjian et al observed that JAK2V617F positivity is indicative of the diagnosis of an underlying Ph¹-negative myeloproliferative disorder, that is, polycythemia vera or essential thrombocytosis.

JAK2V617F mutation is present in over 90% of patients with polycythemia vera (PV) but only in 50% to 70% of those with essential thrombocytosis (ET), and thus is not useful for the diagnosis of ET. It is also not a good indicator of prothrombotic risk for patients with these myeloproliferative disorders (MPDs). On the other hand, in splanchnic vein thrombosis (SVT), including Budd-Chiari syndrome (BCS) and portal vein thrombosis (PVT), Ph¹-negative MPDs are the major underlying predisposing factors. However, it is often difficult to diagnose MPDs in patients with SVTs due to concurrent hypersplenism, gastrointestinal bleeding, and hemodilution. In this issue of Blood, Kiladjian and colleagues present a retrospective multicenter study of 241 patients, 104 with BCS and 137 with PVT, using diagnostic methods including bone marrow biopsy, endogenous erythroid colony formation (EEC), and red cell mass. These were compared with molecular detection of JAK2V617F, JAK2 exon 12 mutation, and MPL mutations (W515L and W515K) for the diagnosis of an underlying MPD.

JAK2V617F was found in 39% of the 241 patients, and abnormal marrow histology was noted in 30%. No patients in this series had JAK2 exon 12 or MPL mutations. The combination of both JAK2V617F and abnormal marrow histology was noted in 44% of patients. JAK2V617F was present in nearly all those patients with both positive marrow histology and abnormal EEC results, while 25% of BCS patients had JAK2V617F but no abnormal marrow histology or EEC. In other words, in these 25% of BCS patients, the MPD diagnosis could not have been made by marrow histology or EEC. This study confirms the findings of an earlier retrospective study with fewer patients,¹  but additionally shows that combined positive results for JAK2V617F and marrow histology provide a high degree of diagnostic accuracy. Both studies indicate that JAK2V617F is a better diagnostic tool for MPDs in BCS and PVT than the traditional hematologic methods of bone marrow histology, endogenous erythroid colony formation (EEC), and red cell mass.

Abnormal EEC without JAK2V617F was observed in 23% of patients, and 58% had discordant marrow histology and EEC. One may speculate that abnormal EEC represented some occult and yet-unrecognized subset of MPDs.

As both BCS and PVT are serious complications, a good prognostic test is always welcome. In this study, the presence of JAK2V617F was able to discriminate those BCS patients with the worst Child-Pugh scores, Clichy prognostic indices, and Rotterdam BCS scores. This is not the case with PVT, nor did presence of PVT affect long-term survival.

One weakness in the Kiladjian study is the tendency to combine BCS and PVT as one syndrome of SVT. BCS is often a more serious condition. Although the diagnostic value for MPDs using the JAK2V617F mutation applies to both conditions, there are differences between BCS and PVT shown in this study. Also, ET and PV were not analyzed separately. Although JAK2V617F does not predict thrombotic risk in PV, some published studies suggest a positive correlation between JAK2V617F and thromboembolic events in ET.