Pegylated interferon alfa-2a for polycythemia vera or essential thrombocythemia resistant or intolerant to hydroxyurea

Abstract

Prior studies have reported high response rates with recombinant interferon-α (rIFN-α) therapy in patients with essential thrombocythemia (ET) and polycythemia vera (PV). To further define the role of rIFN-α, we investigated the outcomes of pegylated-rIFN-α2a (PEG) therapy in ET and PV patients previously treated with hydroxyurea (HU). The Myeloproliferative Disorders Research Consortium (MPD-RC)-111 study was an investigator-initiated, international, multicenter, phase 2 trial evaluating the ability of PEG therapy to induce complete (CR) and partial (PR) hematologic responses in patients with high-risk ET or PV who were either refractory or intolerant to HU. The study included 65 patients with ET and 50 patients with PV. The overall response rates (ORRs; CR/PR) at 12 months were 69.2% (43.1% and 26.2%) in ET patients and 60% (22% and 38%) in PV patients. CR rates were higher in CALR-mutated ET patients (56.5% vs 28.0%; P = .01), compared with those in subjects lacking a CALR mutation. The median absolute reduction in JAK2V617F variant allele fraction was -6% (range, -84% to 47%) in patients achieving a CR vs +4% (range, -18% to 56%) in patients with PR or nonresponse (NR). Therapy was associated with a significant rate of adverse events (AEs); most were manageable, and PEG discontinuation related to AEs occurred in only 13.9% of subjects. We conclude that PEG is an effective therapy for patients with ET or PV who were previously refractory and/or intolerant of HU. This trial was registered at www.clinicaltrials.gov as #NCT01259856.

Graphical abstract

Figure 1.

Genomic profile of somatic mutations and copy number variants (CNVs) in baseline samples, as detected by NGS data. Each column represents a patient (n = 110) and each row represents a gene that is mutated in at least 2 patients (n = 22). In the bottom of the figure, each row represents a CNV event (n = 10). The bar plot (top) indicates the number of somatic mutation(s) per patient, colored according to the type of the alteration. The bar plot on the right shows the number of somatic mutations per gene (top) or CNV event (bottom). The frequency of mutations in the cohort is listed on the left border of the figure. The clinical response to treatment is displayed in the annotation bar at the lower axis of the figure according to the legend. The second annotation bar at the bottom panel denotes the classification of each patient according to the molecular classification criteria.

Figure 2.

Response assessment. Overall response rate at 12 months.

Figure 3.

Changes in mutant JAK2 (p.V617F) VAF over time. (A) VAF in evaluable samples at baseline, month 12, and end of study. The added P-values highlight a significant difference in median JAK2V617F VAF between CR and NR groups in baseline samples (Wilcoxon test; P = 9.51e-03). (B) JAK2V617F VAF waterfall plot. The y-axis indicates the absolute change of JAK2V617F VAF from baseline to last sample. Each bar represents a patient, highlighted by the best clinical response (as described in the color key). Analysis of the number of patients with increased and decreased VAF between groups indicates that the chance of achieving CR is 15 times higher for patients with a reduced JAK2V617F allele burden (Fisher’s exact test; P = 9.071e-06; OR, 15).

Figure 4.

Mean changes from baseline during treatment. Negative changes indicate improvement. *MPN-SAF TSS and EORTC QLQ-C30 GHS/QoL transformed to a 0 to 10 scale, where 10 represents the worst outcome, for consistency with other displayed items.