The myeloproliferative condition essential thrombocythemia (ET) is characterized by a persistent thrombocytosis in the absence of a recognizable cause. Approximately 50% of patients carry the acquired mutation JAK2 V617F, but it is unclear whether this is the initiating event in the overproduction of platelets or a secondary process arising in a situation where the thrombopoietic drive is already increased. In vitro studies have shown that mutant-positive erythroid cells have a proliferative advantage compared to wild-type (WT) cells. However, in JAK2 V617F-positive ET patients, the mutation is only found in a proportion of neutrophils, with the mutant level remaining stable over many years, and the JAK2 WT neutrophils are polyclonal by X-chromosome inactivation analysis. This may not be true of platelets as expansion of the mutant-positive cells may be restricted to the megakaryocytic lineage. We therefore quantified the mutant level in neutrophils and platelets purified from 10 JAK2 V617F-positive ET patients prior to the initiation of cytoreductive therapy using PCR with a fluorescently-labeled reverse primer and a mismatch forward primer that allowed discrimination between WT and JAK2 V617F alleles following AflIII digestion. There was no significant difference in the mutant levels determined using neutrophil DNA and RNA (median 15% [range, 11%–27%] and 21% [12%–31%] respectively). Mutant levels in platelet RNA were significantly higher than those in neutrophil RNA (median 27% [range, 20%–39%] versus 21% [12%–31%] respectively; P = 0.002), but still indicated that the JAK2 mutant was present in only a subpopulation of platelets. Assuming that all cells were heterozygous for the mutation, the data indicate that a median of only 54% (range 40%–78%) of the platelets were mutant-positive. We then calculated the absolute number of JAK2 WT and mutant-positive platelets for each patient from the quantified proportion of mutant alleles in platelet RNA and the total platelet count at the time of testing. The absolute number of JAK2 mutant-positive platelets in the patients varied between 263 and 798 × 109/L, and strongly correlated with the percentage of JAK2 mutant alleles (r2 = 0.81, P = 0.0004). The WT platelet count varied between 225 and 426 × 109/L, and there was a significant negative correlation between the absolute number of WT and mutant-positive platelets (r2 = 0.70, P = 0.002). However, when the absolute number of WT platelets was plotted against the total platelet count, there was no relationship between them (r2 = 0.003, P = 0.87). These data suggest that the negative feedback from the total platelet mass on normal (JAK2 WT) thrombopoiesis was incomplete; in no case was the WT platelet count below the lower limit of normal. This may relate to the observation that in ET, levels of thrombopoietin, the lineage-specific cytokine which drives platelet production, are often normal or even increased, unlike the situation in polycythemia vera where erythropoietin levels are reduced and normal red cell production is suppressed. Furthermore, extrapolation of the data from the WT platelet counts and JAK2 V617F mutant levels raises the possibility that when the mutation was first acquired and mutant levels were very low, the WT platelet count was at the upper limit of normality or elevated. This suggests that the JAK2 V617F mutation could have arisen on a background of increased thrombopoiesis, and was not the initiating event in the development of the disorder.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author