Does extreme thrombocytosis in patients with low-risk essential thrombocythemia increase the risk of thrombotic events and benefit from specific therapy? A retrospective analysis

Essential thrombocythemia (ET) is a chronic, clinically heterogeneous myeloproliferative neoplasm characterized by an elevated platelet count (≥450 × 10⁹/L) with a favorable prognosis compared with polycythemia vera and primary myelofibrosis. However, 22% of patients with low-risk ET present with extreme thrombocytosis (ExT; platelet count ≥1,000 × 10⁹/L), with an assumed risk for more thrombotic complications. In addition, despite reported data for patients with low-risk ET and ExT suggesting similar rates of subsequent thrombotic events in the presence or absence of prophylactic cytoreductive therapy (24% vs 13%, respectively),¹ there remains a lack of evidence to support the optimal treatment of ExT in patients with otherwise low-risk ET.

Tefferi and colleagues recently published a retrospective analysis on vascular events and treatment strategies in patients with low-risk ET and ExT in the American Journal of Hematology, comparing thrombotic events in patients with and without ExT and the impact of cytoreductive and antiplatelet therapy on such events.²

The MPN Hub has previously published a summary of the updates on the diagnosis, risk stratification, and management of polycythemia vera and ET, which can be found here.

Study design

• Retrospective data analysis of 433 patients with low-risk ET (n = 183 with ExT and n = 250 without ExT) derived from the Mayo Clinic ET database.
  • Low-risk ET was defined by the absence of thrombosis history and age ≤60 years.
• The primary outcome was thrombosis-free survival (TFS).

Results

Baseline characteristics

• The baseline characteristics were comparable in both groups for gender distribution, moderate anemia, splenomegaly, abnormal bone marrow karyotype, and microvascular symptoms.
• Patients with ExT were more likely to be younger, exhibit a CALR mutation, have a higher leukocyte count, and present with leukocytosis compared with patients without ExT. In addition, patients with ExT demonstrated a lower incidence of cardiovascular risk factors compared with patients without ExT (Table 1).
• The median follow-up for all patients was 15.3 years (range, 1–47 years) and was longer in patients with ExT compared with those without ExT (16.1 vs 13.2 years, respectively; p = 0.01).

Table 1. Baseline characteristics for patients with and without ExT*

CV, cardiovascular; ExT, extreme thrombocytosis. *Adapted from Tefferi et al.2 †Values in bold are statistically significant. ‡Data available for 99 patients with ExT and 167 patients without ExT.
Baseline characteristic	Patients with ExT (n = 183)	Patients without ExT (n = 250)	p value†
Median age, years (range)	40.1 (17.6–59.2)	46.5 (18.2–59.5)	<0.01
Female, n (%)	128 (70.0)	169 (67.6)	0.6
Median hemoglobin, g/dL (range)	13.5 (8.4–17.5)	13.8 (8.5–17.7)	0.03
Hemoglobin <10 g/dL, n (%)	3 (1.8)	3 (1.3)	0.6
Median leucocyte count, × 109/L (range)	9 (3.5–28.1)	7.9 (3.5–39.0)	<0.01
Leucocyte count ≥11 × 109/L, n (%)	39/161 (24.2)	37/242 (15.3)	0.03
CV risk factor, n (%)               Diabetes mellitus               Hypertension               Smoking	44/137 (32.1) 7/138 (5.1) 30/139 (21.6) 16/133 (12.0)	89/212 (41.9) 7/215 (3.3) 62/215 (28.8) 32/211 (15.2)	0.06 0.39 0.13 0.41
Palpable splenomegaly, n (%)	42/179 (23.5)	45/248 (18.2)	0.18
Driver mutation status, n (%)‡               JAK2V617F               CALR               MPL               Triple negative	34 (34.3) 49 (49.5) 2 (2.0) 14 (14.1)	98 (58.7) 46 (27.5) 1 (0.6) 22 (13.2)	<0.01
Median follow-up, years (range)	16.1 (1.2–46.9)	13.2 (1.3–40.6)	0.01

Thrombotic events

• A total of 81 thrombotic events occurred during the follow-up period, with similar rates for arterial and venous thrombotic events between patients with or without ExT (Table 2). None of the thrombotic events showed any significant differences; they were similar between patients presenting with or without ExT (p = 0.32). The results were also similar when arterial (p = 0.29) and venous (p = 0.9) TFS were analyzed separately.
• The similarity in TFS in both groups was independent of variables such as age (p = 0.04), JAK2/CALR mutation status, leukocyte count ≥11 × 10⁹/L, and cardiovascular risk factors.
• Patients with ExT were more likely to present with a lower International Prognostic Score of Thrombosis for Essential Thrombocythemia (IPSET) score (p < 0.01).

Table 2. Thrombotic events and disease transformation in patients with and without ExT*

ExT, extreme thrombocytosis; IPSET, International Prognostic Score of Thrombosis for Essential Thrombocythemia. *Adapted from Tefferi et al.2 †Values in bold are statistically significant. ‡Major arterial thrombosis included myocardial infarction, angina, cerebrovascular accidents, transient ischemic attack, peripheral arterial thrombosis, and aortic thrombosis. §Major venous thrombosis included deep venous thrombosis, pulmonary embolism, portal/mesenteric/hepatic vein thrombosis, and cerebral sinus thrombosis. ‖Major hemorrhage included bleeding events requiring red cell transfusions.
Event/classification, n (%)	Patients with ExT (n = 183)	Patients without ExT (n = 250)	p value†
Major thrombosis after diagnosis               Arterial thrombosis‡               Venous thrombosis	33/183 (18) 21/183 (12) 15/183 (8)	48/250 (19) 34/250 (14) 17/250 (7)	0.76 0.51 0.58
Major hemorrhage‖ before or after diagnosis	21/145 (15)	21/218 (10)	0.16
Microvascular symptoms	28/145 (19)	53/215 (25)	0.23
IPSET risk group               Low               Intermediate               High	86/117 (74) 20/117 (17) 11/117 (10)	92/184 (50) 57/184 (31) 35/184 (19)	<0.01
Fibrotic transformation	41/183 (22)	39/250 (16)	0.07
Leukemic transformation	11/183 (6)	10/250 (4)	0.34
Death	65/183 (36)	53/250 (21)	<0.01

Influence of specific therapies initiated at diagnosis on thrombotic events in patients with ExT

TFS was significantly better in the presence of aspirin therapy (p = 0.03) among low-risk patients with ExT. A total of 12 events were reported in 78 patients with ExT on initial aspirin therapy compared with 12 events in 35 patients who were not on initial aspirin therapy. Age and aspirin therapy were significant in a multivariate analysis investigating the impact of different variables on TFS (p = 0.02 and p = 0.03, respectively).

Using or not using cytoreductive therapy did not show any difference in TFS among 129 low-risk patients with ExT (p = 0.9). In addition, 8/38 patients on initial cytoreductive therapy developed thrombotic events compared with 18/91 patients who did not receive it.

The data on specific therapy are documented in Table 3, which compared patients with or without ExT depending on treatment modality. As can be seen, the rate of thrombotic events was higher in patients with ExT who received initial cytoreductive therapy compared with patients without ExT, while the reverse was true for initial aspirin therapy. In addition, among patients who experienced arterial events, 61% had received cytoreductive therapy and 33% had received aspirin. In patients who experienced venous events, 39% had received cytoreductive therapy and 34% had received aspirin.

Table 3. The effect of specific therapies on thrombotic events in patients with and without ExT*

ExT; extreme thrombocytosis. *Adapted from Tefferi et al.2 †Values in bold are statistically significant.
Initial treatment	Thrombotic event, n (%)		p value†
	Patients with ExT	Patients without ExT
None	13/137 (10)	27/209 (13)	0.32
Aspirin alone	21/101 (21)	96/191 (50)	<0.01
Aspirin with cytoreductive agent	45/101 (45)	48/191 (25)	<0.01
Cytoreductive agents	91/129 (71)	79/202 (39)	<0.01
Hydroxyurea	60/129 (47)	39/202 (19)	<0.01
Anagrelide	29/129 (23)	37/202 (18)	0.36
Interferon	2/129 (2)	0 (0)	0.05

Conclusion

This retrospective analysis suggests that patients with low-risk ET and ExT are not at a significantly increased risk of thrombotic events compared with patients with ET but without ExT. Furthermore, the analysis showed that there is an association between ExT and a younger age, higher leukocyte count, CALR driver mutations, and shortened overall survival. Interestingly, patients with ExT who develop thrombosis tend to be classified in the low IPSET-thrombosis risk group.

When analyzing the treatment modalities for thrombosis prevention, fewer patients with ExT on initial aspirin therapy developed thrombotic events compared with patients with ExT on cytoreductive drugs. This analysis challenges the benefit of using cytoreductive therapy in reducing thrombotic risk in patients with low-risk ET and ExT and suggests that aspirin therapy should be considered first as prevention for thrombotic events. However, the findings from this retrospective study should be interpreted with caution and further prospective controlled studies are needed to evaluate the benefit of antiplatelet therapy in improving TFS for patients with low-risk ET and ExT.