P1041: LEUKOCYTOSIS SELECTS A SUBGROUP OF LOW-RISK PV PATIENTS IN WHOM PHLEBOTOMY-ALONE MAY BE INSUFFICIENT

Topic: 16. Myeloproliferative neoplasms - Clinical Background: The Low-PV randomized clinical trial provided evidence that ropeginterferon alpha-2b (ropeg) was superior to phlebotomy alone to maintain HCT on target (≤45%) in conventionally defined low-risk patients with polycythemia vera (PV), irrespective of platelet and leukocyte (WBC) counts. Each enrolled patient (n=127) was phlebotomized before starting treatment until the target hematocrit ≤45% was achieved. After 12 months of therapy, the primary endpoint was met in 81% (ropeg) and 51% (standard arm) respectively. Responders continued the assigned treatment for additional 12 months while non-responders were crossed over to the alternative treatment. Aims: To identify the minimum number and timing of phlebotomies qualifying non-responders and investigate the significance of leukocytosis in phlebotomized patients during the 24-months study. Methods: Phlebotomy-only responders vs. non-responders were compared; responders were stratified by WBC values at randomization and through 12 and 24 months. Results: (i)Compared to responders (n=32, 51%), non-responders (n=31, 49 %) were more frequently males (74% vs. 50%, p=0.048) harboring higher but not statistically different JAK2V617F allele burden (42% vs. 27%, p=0.59). After 12 months, in addition to higher HCT values (46.2% vs. 44.1%, p=0.005), non-responders exhibited higher WBC counts (15.8 x109/L vs. 9.8 x109/L, p=0.018) and number of phlebotomies per-patient (5 vs. 2, p=0.007). The greater demand of phlebotomies in non-responders was evident as early as the 6th month (3 vs. 1 phlebotomies per-patient, p=0.001), so that the prediction of treatment response given by the number of phlebotomies per-patient at 6 and 12 months was superimposable by ROC analysis (AUC: 0.767 vs. 0.764, respectively). Of note, in 6 of these patients, progressive symptomatic thrombocytosis, TIA (n=1) and splenic thrombosis (n=1) were registered. (ii)Responders to phlebotomy-only had a low phlebotomy requirement (median 2 per-patient/year) in both the first and second semesters and were stratified according to WBC counts at randomization (< or ≥ 10 x109/L): 14 (44%) presented and maintained up to 24 months a median WBC of 8x109/L vs. 18 (56%) in whom the WBC median was 13x109/L. As compared with the responders with normal WBC counts, responders with leukocytosis also had higher platelet counts (600 x109/L vs 700 x109/L, p=0.033), a trend to elevated values of LDH in 11/18 (61%) vs 4/14 (29%) (p= 0.087) and to a higher JAK2V617F variant allele frequency (VAF) (20% vs. 60%, p=0.07, Figure). Summary/Conclusion: (i)The minimum number of phlebotomies per-patient qualifying non-responders was 3 per semester; these patients can be identified in the first 6 months and may be switched to ropeg treatment earlier than 12 months. (ii)A subgroup of responders (56%), albeit with a low need for phlebotomies, presented a profile indicating a more pronounced myeloproliferation, sustained by leukocytosis, higher JAK2V617F VAF and elevated LDH, that is considered a quantitative measure of cell turnover and tumor burden. (iii)Responders with normal WBC, normal LDH and lower JAK2V617F VAF were overall 14/63 (22%) of low-risks randomized to phlebotomy-only; these patients benefited from phlebotomy-only, in spite of moderately elevated platelet number. These findings support the recent ELN guidelines suggesting to consider interferon-alpha in the subgroup of young low-risk PV patients with persistent leukocytosis.Figure. JAK2V617F VAF in responder patients stratified by median WBC 13x109/L vs 8x109/L at randomization and after 12 and 24 months. Keywords: Therapy, Polycythemia vera, Leukocytes, Myeloproliferative disorder