P1034: from snapshot to continuum: cumulative time in response as an emerging proxy for thrombotic risk in polycythemia vera

Topic: 16. Myeloproliferative neoplasms - Clinical Background: According to recent real-world evidence, patients with high-risk polycythemia vera (PV) receiving standard care spend <30% of time in response for blood count targets recommended by European LeukemiaNet (ELN) (Carpenter et al 2022; eJHaem). Individual responses fluctuated considerably; a novel observation not detectable by conventional evaluation at predefined timepoints. Ropeginterferon alfa-2b induces higher response rates than standard treatment (hydroxyurea [HU]/best available treatment [BAT]) after long-term therapy, but whether this corresponds with cumulative time in response is unknown. Therefore, final PROUD-PV/CONTINUATION-PV data were evaluated regarding time in peripheral blood count remission. Aims: To assess the cumulative proportion of time in response for blood counts according to ELN targets in patients treated with ropeginterferon alfa-2b vs HU/ BAT over a period of ≥6 years. Methods: Patients with PV (WHO 2008 criteria) who were HU-naïve/pre-treated and gave informed consent were randomized 1:1 to ropeginterferon alpha-2b or control (HU) for one year in PROUD-PV (EudraCT: 2012-005259-18). Dosing was optimized per protocol in both arms. In the extension CONTINUATION-PV (2014-001357-17), patients in the control arm could switch from HU to BAT. Hematology was evaluated at intervals of 2-12 weeks for 5 years, and 6-monthly in the final year. Complete hematologic response (CHR) was defined per modified ELN criteria (hematocrit <45% with no phlebotomy for ≥3 months, PLT<400x109/L and WBC<10x109/L). Cumulative time in CHR and in response for individual parameters were assessed in the CONTINUATION-PV full analysis set based on the last response at each assessment, up to study completion/discontinuation. All safety data were analyzed. Results: The full analysis set comprised 169 patients: 95 in the ropeginterferon alfa-2b arm and 74 in the control arm. Individual patient data for time spent in CHR over ≥6 years demonstrate more consistent blood count responses among patients treated with ropeginterferon alfa-2b than in the standard treatment arm, in which 88% received HU as of the last assessment (Figure 1). Despite a more gradual onset of response in the ropeginterferon alfa-2b arm, on average, patients spent 60.9% of time in CHR vs. 41.2% for patients in the control arm (p=0.0437). These findings align with results of log-binomial modelling of longitudinal blood count assessments over the entire 6 years, indicating a higher CHR rate for ropeginterferon alfa-2b vs. standard treatment (RR: 1.27 [95% C.I. 1.05 to 1.53; p=0.0152]). Differences were also observed for individual parameters, with significantly more time spent with WBC counts on target in the ropeginterferon alfa-2b arm vs. control (mean: 93.7% vs. 80.9% p=0.0249). The respective mean proportion of time in response was 65.1% vs 57.0% (p=0.4501) for hematocrit and 87.7% vs 75.6% (p=0.0875) for PLT counts. Including PROUD-PV (N=254), 5 thromboembolic events occurred in the ropeginterferon alfa-2b arm; CHR was recorded at the preceding visit in only 1 case. On standard treatment, 7 thromboembolic events occurred, 3/7 preceded by CHR. Statistical correlation with CHR was precluded by the low incidence of events. Summary/Conclusion: These results underscore the difficulty of maintaining PV patients within peripheral blood count targets during standard treatment and indicate that more consistent control of these parameters can be achieved with ropeginterferon alfa-2b. Thromboembolic events were rare during ropeginterferon alfa-2b therapy and largely occurred during periods of non-response, in agreement with current response criteria.Keywords: Myeloproliferative disorder, Interferon alpha, Polycythemia vera