Final Results From a Phase II Trial of Triapine® Plus Fludarabine for Adults with Aggressive Myeloproliferative Disorders

Abstract 1755

Patients with aggressive myeloproliferative neoplasms (MPN) including MPN in accelerating phase, advanced stage chronic myelomonocytic leukemia (CMML), chronic myelogenous leukemia (CML) in accelerated phase or blast crisis, and other MPN transformed into acute myelogenous leukemia (sAML) have dire outcomes with conventional therapies including bone marrow transplant (BMT). One approach for malignant cell control is the ribonucleotide reductase (RR) inhibitor hydroxyurea (Hu). Triapine® is a novel RR inhibitor that binds to the M2 subunit of RR with 100 – 1000 fold greater potency than Hu and selectively depletes intracellular dATP. On this basis, we previously designed a phase I trial of Triapine® (Tri) and fludarabine (Flu) for patients with refractory acute leukemia and aggressive MPN. The overall response rate (ORR) was 21%, with the majority of responses occurring in patients with MPN including patients with sAML out of MPN. We have now completed a phase II trial of this novel combination in 37 patients (19 male, 18 female) where median age was 64 (35–75) and 9/37 (24%) were over 70 years old. Diagnoses included CMML (8), accelerating MPN (3), sAML out of CMML (7) or MPN (15), and imatinib resistant CML in blast crisis (4). The duration of underlying CMML or MPN ranged from 1 month to 19 years. Twenty three (59%) had previous exposure to Hu and 15 (41%) had received 1–3 prior induction regimens including full multi-drug AML induction regimens. Seven patients had prior allogeneic BMT. Poor risk cytogenetics were present in 16 cases (43%). Four patients had splenectomy (3) or splenic radiation (1) prior to enrollment.

Triapine® was administered at 105 mg/m² followed at 4 hours by fludarabine, 30 mg/m² daily × 5 days. A total of 99 cycles (median 2.5 per patient, range 1–6) were administered, with > 4 cycles given to 9 patients. Four patients died of multi-organ failure after less than 4 weeks of treatment; three of these four presented with hyperleukocytosis (50 – 260K/mm³) and leukostasis. Drug-related toxicities included grade 3 acidosis (13/99 cycles), hypoxemia without dyspnea (6/99), and transient methemoglobinemia. In patients receiving > 4 cycles, chemotherapy was delivered every 33 days (range 20 – 61). Nine patients (24%) achieved complete remission lasting between 2 – 26 months (mean = 8 months, median = 5 months), and an additional three patients (8%) achieved partial remission, giving an ORR of 32%. An additional 8 patients (21%) had hematological improvement on trial. Three patients underwent BMT in CR and one received splenectomy after cytoreduction.

We calculated hazard ratios evaluating multiple host and disease clinical and molecular features. Positive trends were observed for achieving CR, male sex, MPN or CMML diagnosis, JAK2 mutation. Negative trends were observed for CML-BC diagnosis, age over 60, prior splenectomy. Patients with JAK2 mutations received significantly more cycles of therapy compared to wild type (WT) individuals (3.6 vs. 2.1, p=0.02). The proportion of individuals alive at 6 months with JAK2 mutation vs. WT was 0.71 vs. 0.45. To determine if JAK2 mutations bore a relationship to in vitro drug sensitivity, we performed drug sensitivity assays of HEL cells that contain the JAK2 mutation, compared to controls. There were no differences in individual component sensitivities, testing for combination synergy is in progress. The hazard ratios for patients with WBC >50,000 and proliferative disease were no different, arguing against cell cycle effects as the sole mechanism of cytotoxicity of the combination. Finally, SNP-A genomic analysis of amplification, deletion, and uniparental disomy identified acquired molecular abnormalities in the majority of the cases tested, but no specific loci correlated with either positive or negative response.

In summary, the novel combination of Triapine® plus fludarabine, modeled for biochemical synergy, demonstrates clinical responses in poor prognosis aggressive MPDs, including those transformed to AML. JAK2 mutational status may select for responders. Selected patients may benefit from this therapy as a bridge to bone marrow transplant or splenectomy. Our final report solidifies our preliminary results (McDevitt et al., 2008, ASH) and suggests that additional studies looking at this type of combination in addition to molecularly targeted JAK2 inhibitors may improve CR rate and duration.