Azacitidine plus pembrolizumab safe, feasible for certain patients with AML
Click Here to Manage Email Alerts
ORLANDO — Azacitidine in combination with pembrolizumab appeared to be a safe and feasible regimen for patients with relapsed or refractory acute myeloid leukemia, as well as newly diagnosed older patients with AML, according to phase 2 study results presented at ASH Annual Meeting and Exposition.
“We hypothesized that epigenetic reprogramming of [interleukin-2] and PD-1 promoter by azacitidine would result in inflammatory, interferon beta-driven responses, which potentiate the activity of checkpoint inhibition in different tumor mouse models,” Ivana Gojo, MD, co-director of the leukemia drug development program and associate professor of oncology at Johns Hopkins University, said during the presentation.
Azacitidine has demonstrated an ability to upregulate pathways related to immunity and immune evasion in tumor cells, such as PD-L1. Because of this, Gojo and colleagues studied whether azacitidine in combination with pembrolizumab (Keytruda, Merck), a PD-1 inhibitor, could safely induce responses after at least two cycles of therapy among patients with relapsed or refractory AML and newly diagnosed older patients with AML.
The researchers analyzed two cohorts of patients.
Cohort one included 37 patients (median age, 65 years; range, 19-83) who failed a median two (range, 1-6) prior therapies for AML, including intensive chemotherapy (81%), hypomethylating agents (57%) and allogeneic stem cell transplant (30%). Researchers assigned the first six patients in the run-in phase to 75 mg/m2 azacitidine on days 1 to 7 and 200 mg pembrolizumab beginning on day 8 and every 3 weeks thereafter. Patients repeated azacitidine cycles every 4 weeks.
No patients in this group experienced dose-limiting toxicities after a minimum of three cycles. After establishing the safety of the dosing schedule, researchers began enrolling patients for the second cohort.
Cohort two included 22 patients aged 65 years or older (median age, 75 years; range, 67-83) with newly diagnosed AML who were not candidates for or unwilling to receive intensive chemotherapy.
Patients in both cohorts had adequate organ function and no autoimmune processes requiring systemic immunosuppression.
Cohort one results
Researchers presented results of 29 patients in cohort one who completed at least two cycles of treatment.
After median follow-up of 14.9 months, two (7%) achieved complete remission with hematologic improvement, two (7%) achieved complete remission without hematologic improvement, one (4%) achieved partial remission, four (14%) experienced hematologic improvement and seven (24%) had stable disease for at least 6 cycles.
Median number of cycles to response was four (range, 2-6).
Researchers observed mortality rates of 8% at 4 weeks, experienced only by patients with rapidly progressive disease, and 13% at 8 weeks.
Median OS was 10.8 months for the entire group, 13.9 months for those who responded and had stable disease, and 17.9 months for those who achieved complete response, regardless of hematologic recovery, or partial response.
Median EFS was 6 months for the entire group, 8.7 months for those who responded and had stable disease and 2.6 months for all others (P < .0001). Median DFS for those who achieved complete response was 8.5 months, regardless of hematologic recovery.
Patients remained on study for a median 4.4 months and received a median four (range, 1-16) cycles of treatment.
Cohort two results
After median follow-up of 19 months among 17 evaluable patients in cohort two, six (35%) achieved complete response with hematologic recovery, two (12%) achieved complete response without hematologic recovery, two (12%) achieved partial response, two (12%) had hematologic improvement and four (24%) had stable disease for at least six cycles.
Patients responded after a median two (range, 2-15) cycles of therapy.
Researchers observed mortality rates of 5% at 4 weeks, including one patient who received azacitidine on 1 day only, and 9% at 8 weeks.
Median OS was 13.1 months for the whole group, 13.4 months for those who responded and had stable disease, and not reached for those who achieved complete response, regardless of hematologic recovery, or partial response.
Median EFS was 9.6 months for the whole group, 13.4 months in those who responded and had stable disease, and 2.1 months for all others (P < .0001). Median DFS for those who achieved complete response was 16.6 months, regardless of hematologic recovery.
Patients remained on study for a median 6 months and received a median five cycles of treatment (range, 1-21).
Toxicity profile
Common grade 2 or higher immune-related adverse events among patients in both cohorts included skin rash (n = 7, including four who underwent prior allogeneic stem cell transplant), hepatitis (n = 5, including one with grade 2 liver graft-versus-host-disease), pneumonitis (n = 5), hypothyroidism (n = 4), arthralgia (n = 2) and fever. Ten patients received systemic steroids, including three with exacerbation of GVHD,
One patient with progressive disease and grade 3 pneumonitis/hepatitis died of multiorgan failure. Two additional patients who had progressive disease and grade 3 and grade 4 hepatitis were referred to hospice.
Median time to onset of adverse events was two (range, 1-9) treatment cycles.
“The azacitidine and pembrolizumab combination is safe, feasible and well-tolerated for both relapsed and refractory AML and newly diagnosed patients,” Gojo said. “Ongoing correlative studies are exploring potential biomarkers of response.” – by John DeRosier
Reference:
Gojo I, et al. Abstract 832. Presented at: ASH Annual Meeting and Exposition; Dec. 7-10, 2019; Orlando.
Disclosures: Gojo reports advisory/consultant roles with and honoraria from AbbVie, Amgen and Jazz Pharmaceuticals, and research funding from Amgen, Amphivena, Juno Therapeutics and Merck. Please see the abstract for all other authors’ relevant financial disclosures.
Perspective
Back to TopJames K. McCloskey, MD
The relapsed/refractory cohort showed a much less impressive response rate among this set of patients with AML. After the initial cohort accrued, the researchers studied the drug in the front-line setting, where they saw a much more impressive response rate.
This response is impressive but not surprising, and may have some correlation with T-cell function and T-cell exhaustion.
We know with more treatment over time we can exhaust T cells and T cell-deplete patients, so that explains the difference. We will continue to see a relationship between T-cell function and cytokines.
The heavy lifting will have to be done by the patients’ immune systems. If they have a functional and attacking immune system, it’s reasonable to conceive that those patients will be the ones to respond.
As for pembrolizumab, I don’t know if we have enough data yet to determine whether it will be more effective than nivolumab (Opdivo, Bristol-Myers Squibb), but that’s something we will keep watching for.
Hackensack University Medical Center
Click Here to Manage Email Alerts