Zella 201: A Biomarker-Guided Phase II Study of Alvocidib Followed By Cytarabine and Mitoxantrone in MCL-1 Dependent Relapsed/Refractory Acute Myeloid Leukemia (AML)

Background

Multiple studies have shown the clinical activity of alvocidib followed by cytarabine and mitoxantrone in newly diagnosed and relapsed/refractory (R/R) AML. Alvocidib's anti-leukemic pharmacologic activity appears to be predominantly due to the inhibition of transcriptional regulator, CDK9, resulting in suppression of CDK9-regulated genes, such as the BCL-2 family member, MCL-1. Pre-treatment bone marrow samples from newly diagnosed AML patients revealed an increased sensitivity to alvocidib in those with MCL-1 dependence of ≥40% as measured by a BH3 profiling biomarker assay (J Clin Oncol 33, 2015 suppl; 7062). Thus, we hypothesized that alvocidib, followed by cytarabine and mitoxantrone, may be preferentially active in those with MCL-1 dependence (≥ 40%). Here, the findings from stage 1 of the Zella 201 trial in which this biomarker assay is used to select for patients with MCL-1 dependence, are reported.

Aims

To evaluate the efficacy and safety of alvocidib, in combination with cytarabine and mitoxantrone, in MCL-1 dependent R/R AML patients.

Methods

The key eligibility criteria were: ages 18-65 years; refractory to 1-2 cycles of induction therapy, or in first relapse AML with complete remission (CR) duration ≤ 2 years; ≥ 40% myeloblast MCL-1 dependency determined by BH3 profiling; ECOG PS 0-2; and no major organ dysfunction. Patients who received prior allogeneic stem cell transplant (alloSCT) were eligible, if it was greater than two months after SCT and there was no active GVHD. Treatment consisted of alvocidib 30 mg/m² as a 30-minute IV bolus followed by 60 mg/m² over 4 hours on Days 1-3, cytarabine 667 mg/m²/day by continuous IV infusion days 6-8, and mitoxantrone 40 mg/m² IV on day 9 starting 12 hours after completing cytarabine. Up to 3 additional cycles of the same regimen (with or without mitoxantrone) were permittedin responders. The primary endpoint was the rate of CR+CR with incomplete recovery (CRi). Stage I was determined to be positive if ≥13 CRs were seenin the first 23 evaluable patients. Key secondary endpoints were overall survival, event-freesurvival, the combinedresponse rate and safety assessed by adverse events and laboratory results.

Results

A total of 163 patients were screened, of which 47 (29%) were determined to be MCL-1 dependent. Of these, 25 patients were enrolledin Stage 1 (Table 1), with 21 evaluable for response. Median MCL-1 dependence score was 55% (range: 41-98%). Of the 21 evaluable patients, 11 (52%) were refractory to frontline therapy (resistant disease or CR < 90d). The overall CR/CRi rate in evaluable patients was 62% (13/21) meeting the primary endpoint of stage 1. Seven out of 11 (64%) patients with primary refractory disease achieved a CR and five of these patients proceeded to an alloSCT. Overall, 10 patients received a post-study alloSCT. The most common NCI CTCAE ≥Grade 3treatment-emergent nonhematologic AEs noted in >1 patient in the safety population (n=25) were tumor lysis syndrome (20% Grade 3, 8% Grade 4); diarrhea (24% Grade 3); increased AST (12% Grade 3, 8% Grade 4), sepsis (16% Grade 5, 4% Grade 4); and peripheral edema, (8% Grade 3). To date, overall 30- and 60-day mortality rates were 16% and 20%, respectively, due to sepsis (n=4), and mitral valve rupture (n=1).

Conclusion

Our findings indicate that alvocidib given beforecytarabine and mitoxantrone in MCL-1-dependent AML has clinical activity, particularly in those refractory to frontline therapy. Given these findings, stage 2 of the Zella 201 trial has been initiated,randomizing patients to alvocidib, cytarabine, and mitoxantrone versus cytarabine and mitoxantrone alone in MCL-1 dependent R/R AML. Furthermore, a Phase Ib study of alvocidib followed by 7+3 induction in newly diagnosed AML (Zella 101) is being conducted.