Erythropoietin-independent erythroid colonies: is the molecular basis clarified?

Comment on Garçon et al, page 1551

In this issue, Garçon and colleagues in Vainchenker's laboratory describe the molecular basis of erythropoietin-independent colonies (also referred to as endogenous erythroid colonies [EECs]) in polycythemia vera (PV).

Garçon and colleagues provide evidence that 2 signal transduction molecules (ie, STAT5 and Bcl-xL) could explain the existence of EECs. These EECs have been useful in diagnosis of those polycythemia vera patients who do not fulfill the phenotypic criteria as outlined by either the Polycythemia Vera Study Group or the World Health Organization.¹  Vainchenker's laboratory was indeed well positioned to undertake this task, since they were the first ones to focus on the role of abnormal JAK2 signaling in polycythemia vera²  and were the first to describe the JAK2 somatic mutation in polycythemia vera, JAK2V617F.³  This finding was in rapid sequence confirmed and simultaneously reported by multiple other laboratories. Since activation of JAK2 by the erythropoietin/erythropoietin receptor pathway activates multiple signal transduction mechanisms, including STAT5, MAP, PI3K, and possibly Src family kinases, the molecular mechanism of EECs remained to be determined. Garçon and colleagues report a series of experiments using inhibition of STAT5 and Bcl-xL by siRNA and overexpression of these molecules by means of retroviral transduction and conclusively demonstrate that manipulation of STAT5 and its downstream partner Bcl-xL induces the EECs.

While it is pleasing to understand the molecular mechanism of EECs, it remains to be established if erythropoietin hypersensitivity and the occasional EECs observed in other primary polycythemic disorders, such as primary familiar and congenital polycythemia and Chuvash polycythemia,⁴  have the same EEC mechanism. These disorders are not associated with activating mutations of JAK2; does the STAT5/Bcl-L pathway also mediate them? Further, not all patients with polycythemia vera and EECs have the JAK2 mutation. In addition, in rare families inheriting the predisposition for polycythemia vera, all affected polycythemia vera patients have EECs⁵  but not all of them have JAK2 mutation⁶  (this is also our experience). The finding of Garçon and colleagues needs to be extended to other primary polycythemic states. Nevertheless, in our opinion, while the assay for EECs is not necessary for most patients presenting with elevated hematocrit and for their differential diagnosis, EECs still remain useful as a diagnostic tool for some patients as well as an important research tool.

Supported by grant 1P01CA108671-O1A2 from the MPD Consortium (principal investigator, Ron Hoffman). ▪