Phase II trial comparing timed sequential cytosine arabinoside with or without MK-8776 in R/R AML

In the October 2017 issue of Leukemia Research, Jonathan A. Webster from The Sidney Kimmel Comprehensive Cancer Center at John Hopkins, Baltimore, MD, et al., reported results from their phase II randomized study (NCT01870596) which compared cytosine arabinoside (AraC) plus MK8876 with AraC alone in patients with Relapsed or Refractory (R/R) Acute Myeloid Leukemia (AML).

AraC exerts its cytotoxic effects on leukemic cells via incorporation into DNA. However, this incorporation can lead to the activation of Checkpoint Kinase 1 (Chk1), which can diminish Ara-C induced cytotoxicity in leukemia cells. MK8876 is a potent and selective Chk1 inhibitor that has been shown to sensitize AML cells to Ara-C in vitro.   

In this phase II study, 32 R/R AML patients with a median age of 60 years (range, 29–72 years) who were enrolled between June 2013 and September 2014 at multiple institutions were randomized into two arms, Arm A (n = 14) or B (n = 18). Patients in Arm A were administered AraC (2 g/m² over 72h intravenous [IV] continuous infusion beginning on Day 1 and Day 10) with MK-8776 (100 mg over 30 min by IV infusion beginning 24 ± 4 h and 48 ± 4 h after the start of each AraC infusion on Days 2, 3, 11, and 12). Patients in Arm B were administered AraC alone. The AraC dosing strategy used in this study was based on Timed Sequential Therapy (TST). TST is a therapeutic strategy that exploits the leukemia cell cycle kinetics by administering a second drug in close approximation with the leukemia proliferation induced by the first drug.

The key findings of the study were:

• Safety

  • Grade 3/4 Treatment-related Adverse Events (TAEs) in Arm A and B, respectively
    • Febrile neutropenia; 64% vs 61%
    • Infectious complications; 50% vs 50%
    • Sepsis; 21% vs 17%
  • Two treatment related deaths occurred in Arm B

• Outcomes

  • Complete Responses (CR/CRi) in Arm A and B; 36% (5/14) vs 44% (8/18)
  • Partial Response (PR) in Arm A and B; 7% vs 6%

• Survival

  • Median Overall Survival (OS) in patients in Arm A and B; 5.91 vs 4.48 months, HR = 1.18, P = 0.69
  • Median Event Free Survival (EFS) in responders in Arm A and B; 5.82 vs 9.74 months, HR = 0.72, P = 0.63
• There was an increase in _Ƴ-H2AX (measures DNA damage) expression on viable circulating leukemic blasts obtained from patients in Arm A (n = 8) directly after MK-8776; 16.9% ± 6.1% (prior to MK-8776) vs 36.4% ± 6.8 (1 h after MK-8776) positive cells, P = 0.016

In summary, “the combination of AraC and MK-8776 was well tolerated but did not improve CR/CRi rates or OS when compared to AraC alone “in spite of transiently augmenting DNA damage in vivo”.

The authors noted that “accrual to this phase II study was stopped due to the termination of the clinical development of MK-8776 and its eventual expiration” which they suggested limited the assessment of efficacy of the AraC and MK-8776 combination regimen.

Abstract

PURPOSE:

Cytosine arabinoside (AraC) remains the backbone of most treatment regimens for acute myeloid leukemia (AML). Incorporation of AraC into DNA activates checkpoint kinase 1 (Chk1), leading to cell-cycle arrest and diminished AraC cytotoxicity, which can be reversed by the selective Chk1 inhibitor MK-8776. Building on a Phase I trial, we conducted a phase II trial comparing timed sequential AraC with or without MK-8776.

METHODS:

Patients with relapsed or primary refractory AML were randomized 1:1 to receive either AraC with MK-8776 (Arm A); or AraC alone (Arm B).

RESULTS:

32 patients were treated: 14 assigned to Arm A and 18 to Arm B. There were 5 (36%) complete responses (CR/CRi) and 1 (7%) partial response (PR) in Arm A, and 8 (44%) CR/CRis and 1 (6%) PR in Arm B. Median survival did not differ significantly between the two groups (5.9months in Arm A vs. 4.5 months in Arm B). MK-8776 led to a robust increase in DNA damage in circulating leukemic blasts as measured by increased γ-H2AX (16.9%±6.1% prior and 36.4%±6.8% at one hour after MK-8776 infusion, p=0.016).

CONCLUSION:

Response rates and survival were similar between the two groups in spite of evidence that MK-8776 augmented DNA damage in circulating leukemic blasts. Better than expected results in the control arm using timed sequential AraC and truncated patient enrollment may have limited the ability to detect clinical benefit from the combination.