Drug development challenges in polycythemia vera

Ellen Kelly Ritchie

In this issue of Blood, Mascarenhas et al report on the safety and efficacy of oral idasanutlin in high-risk patients with polycythemia vera [PV]) with the goal of targeting hematopoietic stem cells and progenitor cells (HSC/HPCs) in this indolent malignancy.1

PV is a myeloproliferative neoplasm (MPN) characterized by expansion of the red blood cell mass. The majority of patients have a point mutation in the JAK2 gene in exon 14 or 12. Initial treatment of this disease aims to decrease the proliferation of blood cells and the risk of embolic and cardiovascular events. Therapies most often used first include phlebotomy or hydroxyurea. These therapies accomplish the goal of lowering blood counts but do not treat the underlying disease or protect against progression to myelofibrosis or the development of acute leukemia.

Interferon-α2a, an alternative therapy may not protect against progression either, but it can induce molecular remissions that are durable.2 The clinical activity of interferon-α2a may be partially attributable to the upregulation of TP53 activity.3 JAK2V617F mutations increasing MDM2 protein translation by upregulating the La antigen which in turn alters p53 responses to DNA damage.4 In the presence of elevated MDM2 protein levels, TP53 messenger RNA levels are lower as compared with normal CD34+ cells. MDM2 and MDM4 can each independently bind to p53 and block its transcriptional activity, and together have ubiquitin ligase activity.5,6 In an earlier publication, Mascarenhas and collaborators showed that MDM2 is upregulated in PV 34+ stem cell progenitors and that nutlins (a class of drugs that inhibit MDM2 activity) are capable of depleting mutated PV HSC/HPCs.7 MDM inhibition results in upregulation of TP53 which may allow for curative therapy.

JAK2V617F+ patients were treated with idasanutlin at 2 dose levels on this phase 1 expansion trial to evaluate the safety and efficacy of this drug. Patients were treated for 5 days out of each cycle with the study drug. For patients who achieved a response to treatment, subsequent cycles were not given until prespecified hematologic parameters were met. Those patients who were tolerant of the drug but did not achieve a response with single-agent idasanutlin after 6 cycles were eligible to receive combination therapy with idasanutlin combined with interferon-α2a.

The overall response rate was 58% (7/12) for single-agent idasanutlin and 50% (2/4) for the combination arm with interferon-α2a. The median duration of response was 16.8 months. Scores on the MPN Symptom Assessment Form were improved with a maximum total symptom score (TSS) reduction of 81.5%. Idasanutlin therapy was associated with a 43% mean reduction in the JAK2V617F variant allele frequency in all patients except one, who had a p53 mutation. Toxicities of the drug were mainly nonhematologic. Five patients experienced grade 3 adverse events, mainly of the gastrointestinal tract during days 3-6 of drug administration. Eighty-three percent of patients experienced grade 1 or 2 nausea and needed a 3-drug antiemetic regimen consisting of ondansetron, decadron, and lorezepam to control symptoms during the 5 days of study drug administration. Four patients withdrew from the study, and 3 patients were taken off study by the investigator.

This study illustrates some important points about clinical trials in patients with PV. Given that PV is a rare disease with prevalence in 2003 of 22 per 100 000, the number of patients eligible for PV clinical trials is small.8 Combined with the fact that only 5% of adults in the United States participate in clinical trials, there are few potential participants, and, given the small numbers enrolled, it is difficult to fully assess safety and efficacy of drugs. There are few resources to increase clinical trial participation in the United States or to encourage those with rare diseases to enroll in higher numbers. To make progress, there needs to be a funded national network to connect numerous study sites across the country and build consensus among sites to look at individual drugs for phase 1 to 2 testing.

The authors consider this agent promising as responses were seen in the majority of patients with only grade 1 to 2 nausea responding to treatment with a 3-drug antiemetic regimen during the 5 days of treatment. Although physicians often dismiss grade 1 to 2 toxicities as “tolerable” in a drug used to treat a neoplasm, they may not be acceptable to a patient population that may need to take the drug for many years. In this study, even though there were excellent hematologic responses and improvements in TSS on the MPN-SAF, the main reason patients discontinued the drug was because of the gastrointestinal toxicity. Even low-grade toxicity can contribute to decreased day-to-day quality of life, and the tradeoff for patients may not be worth a better long-term outcome. With long-term administration, even grade 1 to 2 toxicities might not be tolerable in PV patients.

The rationale for treating PV patients with an MDM2 inhibitor was strong. Although the overall response rate was impressive, the median duration of response was disappointing at only 16.8 months. Why was the therapy not more durable, and how can it be improved upon? It may be that the patients did not have sufficient elevation of MDM2 to have an optimal response to treatment. Additionally, although idasanutlin may cause cell-cycle arrest, there may be quiescent cells that linger and proliferate leading to new mutations after the drug has been stopped. Additional P53 regulators may block the response to MDM2 antagonists, such as MDM4 or NPM-1 or resistance may be caused by additional mutations acquired during idasanutlin treatment that cause resistance to the drug. In this study, patients who were unresponsive to single-agent therapy could receive combination therapy with interferon-α2a. In the 4 patients treated with combination therapy, there was a 50% response rate. Idasanutlin may therefore need to be used in combination with other drugs to prevent resistance and increase effectiveness, and prolong duration of response.9 Although interferon-α2a is a logical partner for idasanutlin, there are other possible combinations. Nonetheless, the problem with combination therapies in PV is the additive toxicities of using 2 drugs. The difficulty of treating this particular population and the low tolerance for even grade 1 to 2 toxicities will make combination therapy a challenge.


  • Conflict-of-interest disclosure: E.K.R. receives research funding from Jazz Pharmaceuticals; consultation fees from Agios, Celgene, Tolero, Novartis, Genentech, and Abbvie; and speaker fees from Incyte Pharmaceuticals.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.