INTRODUCTION -The prognosis of acute myeloid leukemia (AML) in elderly population is inherently poor due to multiple factors including: frailty, medical co-morbidities and high treatment related morbidity with traditional induction chemotherapy and presence of adverse risk cytogenetics in 17-20% of patients. For patients who are not candidates for intensive chemotherapy, outcomes in terms of complete response/complete response with incomplete hematologic response (CR/CRi) and median overall survival (mOS) have been evaluated using hypomethylating agents(HMA) including azacytidine (AZA)(CR/CRi=18-27.8%, mOS=10.4-24.5 months) and decitabine (CR/CRi= 18-47%, mOS= 7-7.8 months). Recently, addition of venetoclax to HMA has shown to further improve CR/CRi (75%) and mOS (17.5 m). We conducted a systematic review of literature to identify recent studies that were published between 2015-2020 with an aim is to evaluate newer combination drug regimens (CDR) involving hypomethylating agents (AZA and decitabine) in treating elderly patients with newly diagnosed AML.

METHODS- A comprehensive literature search was conducted in PubMed, Embase, and Cochrane databases. We included phase I/II studies that used CDR with HMA in elderly patients with newly diagnosed AML.

RESULTS- Initial database search lead to 1120 studies. After exclusion (duplicates, case reports/series, relapsed/refractory AML) final analysis included 12 studies (n=655).

Seven phase I/II with CDR using decitabine (n=500) were included in our review. Drugs used in combination included gemcitabine ozagamicin (n=40, CR/CRi =45%, mOS= 7 m), cladribine and low dose cytarabine (n=118, CR/CRi= 68%, mOS =13.8 m), vadastuximab talirine (n=53, CR/CRi=70%, mOS=11.3 m) and selinexor (n=5, CR/Cri= 80%) . In addition, three studies compared outcomes of CDR involving decitabine with all trans retinoic acid (ATRA)(n=93, ORR=21.9% vs 13.5%, p=0.06; mOS= 8.2 m vs 5.1 m, p=0.006) or talacotuzumab (CR/CRi= 15% vs 11%, p=0.44; mOS= 5.36 m vs 7.26 m, p=0.78) or bortezomib (CR/CRi= 39% vs 38%, p=0.91, mOS=9.3 m vs 8.9,p=0.18) to decitabine alone.

In addition, cladribine and low dose cytarabine with decitabine in patients with adverse cytogenetics showed a decent CR/CRi of 50%, mOS= 10.5 months. In TP 53 positive mutations, CR/CRi was 40% and mOS was 5.4 months.

Five phase I/II studies using CDR with AZA (n=155) were included in our review. Drugs included midostaurin (n= 88 , CR/CRi=25-29% mOS= 6-8 m) and pracinostat (n= 50, CR/CRi =46% mOS=19.1 m), In addition, two studies compared outcomes using CDR involving AZA with panobinostat (n=22, CR= 22.4 vs 30.8%, OS at 1 year = 60% vs 70%) or entinostat (n=18, ORR= 0% vs 16.6%, mOS=6 m vs 13 m) to AZA alone. None of the two comparative CDR studies using AZA showed superior outcome compared to AZA alone.

CONCLUSIONS - Novel drug combinations involving decitabine including cladribine and low dose cytarabine, vadastuximab talarine showed superior CR/CRi and mOS compared to decitabine alone used in historic studies. In addition, CR/CRi were similar to HMA and venetoclax combination. In small number of patients, CDR with selinexor also showed superior CR/CRi compared to decitabine alone. CDR involving decitabine and ATRA showed superior mOS in direct comparison with decitabine alone. In addition, cladribine, low dose cytarabine and decitabine has shown promising outcomes in patients with adverse cytogenetics.

Among CDR involving AZA only Proctinostat showed superior CR/CRi and mOS compared to historic studies with AZA alone. A recent phase III study though involving this CDR in comparison with decitabine was terminated early due to lack of efficacy on preliminary analysis.

The above listed efficacious CDR involving decitabine in phase I/II studies need to be evaluated in large, randomized trials to assess for definitive benefit. If proven efficacious in larger studies, they could serve as additional first line CDR options in addition to HMA and venetoclax in treating elderly patients with newly diagnosed AML.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.