A Phase II Study of Vorinostat (MK-0683) in Patients with Polycythemia Vera and Essential Thrombocythemia

Recent findings lend strong support to the contention that histone deacetylase inhibition may be an important epigenetic therapy in the treatment of patients with myeloproliferative neoplasms and emphasize the need to characterize the efficacy and safety of this novel class of cytoreductive agents.

Aims: This study was a non-randomized, open-label phase II multicenter study with sixty-three patients (21 essential thrombocythaemia (ET), 42 polycythemia vera (PV)) included from 15 centers. The primary objective was to investigate, if vorinostat as monotherapy in patients with PV and ET was followed by a decline in clonal myeloproliferation as assessed by conventional disease activity parameters. Secondary objectives included assessment of adverse effects during treatment; changes in bone marrow morphology before and after treatment with vorinostat and to investigate whether treatment with vorinostat influences the JAK2 mutant allele burden as assessed by quantitative PCR. (qPCR)

Thirty-one patients (49%) were followed to the end of the intervention period. Eighty-one percent of these had a partial (n=20) – or complete (n=5) hematological response according to ELN criteria. Response rates were found to be independent of JAK-status in ET patients. (P=0.83). A tendency towards less favorable responses was observed among patients with > 1 previous therapies albeit statistically insignificant (P=0.11). For all but two patients, a clinicohematological response was not followed by a histological remission.

The prevalence of splenomegaly was lowered from 48% to 24% (P=0.02). A statistically significant reduction of JAK2 mutant allele burden was observed (P=0.006). No JAK2 positive patients experienced a complete molecular response defined as undetectable JAK2V617F by qPCR. No significant correlation between severity of tumor allele burden at inclusion and a more pronounced molecular response to vorinostat was found using a Spearman correlation analysis (rho = 0.16, p = 0.3).

The most commonly reported adverse effects (AEs) during the intervention period among patients who completed the protocol were fatigue and gastrointestinal (anorexia, nausea, vomiting, diarrhea, dryness of the mouth). Gastrointestinal symptoms were generally manageable. Seventy percent of included patients experienced hair loss. Seventeen percent experienced renal toxicity (3 pts. grade I, 1 pt. grade II) and 1 pt. (4%) liver toxicity of unknown grade, which resolved after withdrawal of vorinostat. Forty-three percent of included patients required at least one dose reduction due to AE′s. Forty patients (63 % of patients) discontinued study drug before end of study period due to adverse events (65%), unknown (17.5%), withdrawal of consent (7.5%), no response (2.5%), or progression to acute leukemia (7.5%). Response rates for patients who discontinued therapy, however, showed that approximately 50% of patients who discontinued experienced a clinicohematological response providing evidence that toxicity issues were of concern rather than lack of clinical effect.

Vorinostat is effective in patients with ET and PV normalizing elevated leukocyte and platelet counts as well as providing a statistically significant reduction of the JAK2 mutant allele burden and splenomegaly. However, vorinostat was associated with a high dropout rate. Considering the heterogeneity and complexity of oncogenic events, involving both deregulated tyrosine kinase activity as well as epigenetic deregulation in MPN-pathogenesis combination therapy (eg. JAK1-2 inhibitors, DNA-hypomethylating agents, interferon-alpha2) may be more efficacious than single agent therapy. This strategy might also allow for dose reduction of single agents and accordingly a decrease in toxicity, which was the major limiting factor for patients adhering to treatment in the present study.

No relevant conflicts of interest to declare.