Background:

In myeloproliferative neoplasms, especially essential thrombocythemia (ET), platelet (Plt) counts and von Willebrand factor (vWF):Ristocetin cofactor (RCo) levels have been reported to be inversely correlated. However, there have been no reports of the comparison of high vs. low values of JAK2 allele burden and necessary of Plt counts reduction to achieve vWF:RCo levels ≥50%, which is required for major surgery. We investigated the correlation between vWF level and Plt counts for JAK2V617F mutation positive (JAK2V617F+) polycythemia vera (PV) and ET (allele burdens ≥50% and <50%) and calreticulin mutation-positive (CALR+) ET.

Method:

We recruited patients with PV and ET who were diagnosed per the 2008 World Health Organization criteria at 3 hospitals since 2011 to 2018. All patients were analyzed at Juntendo University for JAK2 (V617F, exson12), CALR, and MPL mutations. We collected data of JAK2V617F+ or CALR+ patients. Triple-negative mutations of ET were excluded in this study because it is difficult to differentiate between ET and secondary thrombocytosis.

We analyzed the correlation between Plt counts and vWF:RCo levels for three mutation groups mutations (JAK2V617F allele burden ≥50%, <50%, CALR+) using Spearman's rank correlation test when blood samples were obtained for the first time from patients after enrolling in this study. In addition, bleeding risk was compared in three mutation groups using Fisher's exact test in patients who were controlled with relatively low Plt counts <600×10⁹/L.

Results:

We collected 146 PV and ET patients. Among 54 PV patients, 50 had JAK2V617F+, 3 had JAK2 exon 12, and 1 had a triple-negative mutation(s) in PV patients. Among 92 ET patients, 53 had JAK2V617F+, 35 had CALR+ (22 del52, 12 ins5 and 1 del34), and 4 had MPL mutations in ET patients. vWF:RCo levels were weakly inversely correlated with Plt counts in patients with JAK2V617F+ PV and ET from whom blood samples were obtained for the first time after enrolling in this study (Ps=-0.531 and Ps=-0.439, respectively). In contrast, vWF:RCo levels and Plt counts in patients with CALR+ ET showed a strong inverse correlation (Ps=-0.762). Interestingly, 3/50 PV and 4/53 ET patients with JAK2V617F mutations showed vWF:RCo levels >150% (normal vWF:RCo levels range: 50-150%). Furthermore, 3/50 PV and 1/53 ET patients with Plt counts <600×10⁹/L showed vWF: RCo levels <50%. However, none of the patients with CALR+ ET had vWF:RCo levels >150% or <50%.

During the entire observation period, 8/50 and 7/53 patients with JAK2V617F+ PV and ET, respectively, and 1/35 patients with CALR+ ET had vWF:RCo levels >150%. One case of JAK2V617F+ ET (allele burden <50%) complicated with deep vein thrombosis; blood sampling data revealed vWF:RCo levels of 200% and a Plt count of 269×10⁹/L. The patient exhibited only age (86 years) as a risk factor for thrombosis and did not have other risk factors (diabetes mellitus, smoking history, hypertension, hyperlipidemia, thrombosis history).

When Plt counts were controlled to <600×10⁹/L, all patients with JAK2V617F+ PV and ET and CALR+ ET had vWF:RCo levels >30%. However, 13/86 patients did not achieve 50%, which is the standard for safely performing the major surgical procedures prescribed in most guidelines. In patients whose Plt counts controlled to <600×10⁹/L, vWF:RCo levels <50% was more frequently seen in patients with JAK2V617F allele burden ≥50% than those those with JAK2V617F allele burden <50% (10/32 vs 2/35, p = 0.00956) or CALR mutations (10/32 vs 1/19, p = 0.0373).

Conclusion:

For patients with CALR+ ET, we propose that vWF:RCo levels can be predicted based on Plt counts; however, it is difficult to predict vWF:RCo levels by Plt counts in JAK2V617F+ PV and ET patients because the inverse correlation between vWF:RCo levels and Plt counts was weak.

Overactivation of vWF:RCo levels was often observed in JAK2+ PV and ET. Moreover, in 1 case, the patient exhibited only age as the risk factor for the complication of thrombosis. Because of these findings, overactivation of vWF:RCo levels may be one of the reasons for thrombotic events in JAK2V617F+ PV and ET. In addition, it was also found that cases of JAK2V617F+ allele burden ≥50% in PV and ET should be careful for hemorrhage in major surgery, even if Plt counts were controlled <600×10⁹/L.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution