Givinostat in polycythemia vera: Long-term safety and efficacy

Polycythemia vera (PV) is a BCR-ABL1-negative subtype of myeloproliferative neoplasms (MPN), characterized by excessive erythroid progenitor and mature cell production. In the majority of cases, this is a result of a mutation in the Janus kinase 2 gene (JAK2 V617F). Givinostat, a histone-deacetylase inhibitor, selectively targets JAK2 V617F cell growth to reduce hematopoietic cell proliferation.

Three open-label 24-week phase I/II studies have reported that givinostat, either as monotherapy or in combination with hydroxyurea, gave a high response rate (50–80%) and had a good safety profile in patients with PV.¹ Those who received clinical benefit from givinostat in these studies were able to enter a compassionate use program. Alessandro Rambaldi and colleagues conducted a long-term study (NCT01761968) of patients in the program, and here we present a summary of their 4-year findings, which were published in Blood Cancer Journal.¹

Study design

Eligible patients were ≥18 years old, JAK2 V617F-positive with a diagnosis of MPN according to the World Health Organization criteria, and had tolerated previous givinostat treatment and achieved a clinical benefit at the end of one of the three open-label core studies and/or a compassionate use program. The core studies were:

• A pilot study in 29 patients with JAK2 V617F-positive MPN who were refractory to hydroxyurea, or were young and required cytoreductive therapy; patients received givinostat monotherapy at 50–150 mg/day² (NCT00606307)
• A phase II study of 44 patients with JAK2 V617F-positive PV who were non-responders to hydroxyurea and received givinostat at 50–150mg/day alongside hydroxyurea at the maximum tolerated dose³ (NCT00928707)
• A dose-determination study of givinostat in 48 patients with JAK2 V617F-positive PV⁴ (NCT01901432)

Primary objectives of Rambaldi et al.’s study were determination of long-term safety and tolerability, and assessment of efficacy (complete response [CR] and partial response [PR] rate).

Patients received givinostat with or without concomitant hydroxyurea at the last tolerated dose and regimen as in the core study or compassionate use program. Dose adjustments to optimize individual responses were permitted.

Patients

In total, 78 patients with MPN across 15 centers in France, Germany, Italy, and the UK entered the compassionate use program; 53 of these subsequently enrolled in the long-term study, in addition to one patient who had not participated in any of the core studies. The majority of patients had PV (n = 51) and, at the cut-off date of December 31, 2018, 50 of these had received ≥1 dose of givinostat. Of these, 15 were treated with concomitant hydroxyurea. Baseline characteristics of the patients with PV are shown in Table 1. Median exposure to givinostat was 2.8 years (range, 3 months – 11 years). At data cut-off, 31 patients (62%) were ongoing in the study.

Table 1. Baseline characteristics for patients with PV treated with givinostat^*

Key findings

Safety

During the long-term study, 96% of patients experienced at least one adverse event (AE); however, 89.7% of these events were Grade <3. Overall, 64% of patients (n = 32) experienced treatment-related AEs (TRAEs) of any grade, the most common of which were blood and lymphatic system disorders (24%) and gastrointestinal disorders (32%). Five patients reported Grade 3 TRAEs, including thrombocytopenia, diarrhea, asthenia, QTc (corrected QT interval) prolongation, and hypertension (n = 1 each), and there were no Grade 4 or 5 events. Treatment-related QTc prolongation was noted in three patients, one case of which was Grade 3 and resulted in treatment discontinuation. All patients who received concomitant hydroxyurea experienced at least one AE, including Grade 3 cases of thrombocytopenia and asthenia.

Efficacy

Over 80% patients had a CR or PR maintained for the duration of follow-up (Figure 1). Four patients (8%) were defined as non-responders on study entry, but they were enrolled as givinostat provided them with benefit from disease-related symptoms and/or hematological parameters. Interestingly, overall response was higher in patients treated with givinostat monotherapy (97%) compared to those who received concomitant hydroxyurea (80%). Moreover, a reduction in mean JAK2 V617F allele burden was observed at the majority of patients’ annual visits.

Figure 1. Responses to givinostat during follow-up¹ 

AE, adverse event; CR, complete response; NR, non-responder; ORR, overall response rate; PR, partial response.

Conclusion

The first interim analysis of this long-term study found that givinostat demonstrated a good safety and efficacy profile over 4 years, supporting its long-term use in patients with PV. Limitations of the study included its ongoing nature; thus patients were at different stages of treatment at the cut-off with full 4- and 5-year data unavailable for all patients. In addition, only patients who had previously achieved clinical benefit with givinostat were included, with no comparator arms.

One of the study authors, MPN Hub Steering Committee member Alessandro M. Vannucchi, discussed the findings below.