Keywords
Hypomethylating agents - induction chemotherapy - performance status
Introduction
Acute myeloid leukemia (AML) is a heterogeneous disease characterized by infiltration
of the bone marrow (BM), blood, and other tissues by mutated clonal hematopoietic
progenitor cells and abnormal differentiation of hematopoietic lineages, ultimately
leading to marrow failure.[1]
[2] The incidence of AML increases with advancing age, with half of the new cases diagnosed
in adults aged ≥65 years. Approximately 80% of older AML patients will die of their
disease or its treatment with currently available antileukemic therapy because of
the adverse prognostic risk factors, such as history of myelodysplastic syndromes
(MDSs), unfavorable karyotypes, poor performance status (PS), and comorbidities associated
with aging, which can limit treatment options.[3] As a result, many older patients receive only palliative care. Median overall survival
(OS) of AML patients ≥65 years of age is only 2–8 months.
Commonly used therapeutic options to treat older patients with AML are the best supportive
care alone, standard induction chemotherapy (IC), and low-dose cytarabine arabinoside
(LDAC). Fitness for IC is decided based on factors such as age, PS, functional status,
and comorbid conditions. The National Comprehensive Cancer Network guidelines recommend
IC for fit patients with age ≥60 years. However, many older patients with AML do not
meet the fitness criteria. In elderly patients, who are not candidates for IC or who
declines IC, the preferred induction regimen is with low intensity azacitidine (AZA)
or decitabine (DAC).[4]
Epigenetic changes are heritable changes in gene expression that are not caused by
changes in the primary DNA sequence, and they affect the spatial structure of the
DNA that is coiled around histones. This spatial structure determines binding of transcription
machinery to the promoter of a gene, in order to initiate transcription.[5] Methylation and acetylation of amino acid residues in histones and methylation of
cytosine (C) bases in areas of the genome rich in CpG dinucleotides (CpG islands)
are the best-known epigenetic changes. Cancer cells exhibit genome-wide hypomethylation,
resulting in genetic instability, and CpG islands hypermethylation which modifies
gene expression.[6] Epigenetic changes are reversible, making them an attractive target for therapeutic
intervention.
Abnormal methylation plays an important role in the pathogenesis of AML. Genes such
as DNA-methyltransferase-3A (DNMT3A), Ten-Eleven- Translocation-2, and isocitrate dehydrogenase-1 and 2 are involved in
DNA methylation, and their mutated variants may help interpret the mechanisms of aberrant
DNA methylation in AML blasts.[2]
Hypomethylating agents
Five-azacytidine (AZA) and 5-aza-2′-deoxycytidine (DAC) were synthesized as analogs
of C for the treatment of AML in 1960s. They were extremely toxic at higher antineoplastic
doses and hence were phased out. Discovery of hypomethylating properties of these
drugs renewed interest in their clinical use. They act as DNMT inhibitors, leading
to global hypomethylation of C residues associated with gene expression control.[7]
In India, data regarding therapy with HMA, response to therapy and overall survival
(OS) is seldom reported. The objective of this retrospective study was to study the
response rate and survival of patients treated with HMAs in whom IC was not feasible.
Subjects And Methods
This is a retrospective and descriptive single-center study. All patients with a diagnosis
of AML who presented to the Department of Medical Oncology at our institution and
received HMAs (AZA and DAC) were enrolled. The patient’s demographic data, Eastern
Cooperative Oncology Group (ECOG) PS, comorbidities, and baseline investigations were
collected. Charlson comorbidity index (CI),[8] hematopoietic cell transplantation (HCT) CI[9] were calculated.
Treatment regimen
AZA was administered at 75 mg/m2/day for 7 days intravenous (IV) repeated every 28-day and DAC was administered at
20 mg/m2/day for 5 days, IV repeated in a 28-day at physician’s discretion. Therapy was continued
until progressive disease (PD) or toxicity in patients with partial remission (PR)
or hematologic improvement.
Response rates
Criteria developed by the International Working Group (IWG) were used to define response
rates such as complete remission (CR), CR with incomplete hematologic recovery (CRi), PR, and stable disease, PD.[10]
[11]
Transfusion independence
Transfusion independence was defined as a transfusion-free period of 3 months after
treatment assignment. Transfusion dependence at baseline was defined as two or more
transfusions per month within 90 days before the assignment.[12]
Statistical methods
All the numerical characteristics such as age and duration of disease were discussed
through the descriptive statistics in terms of mean and standard deviation or median
and interquartile range. Qualitative variables were described as percentages. OS was
calculated from the date of treatment initiation to the date of death. The analysis
was performed using IBM Corp. Released 2012. IBM SPSS Statistics for Windows, Version
21.0. (Armonk, NY: IBM Corp.).
Results
Twenty-three patients received HMAs as treatment for AML (12 patients received DAC
and 11 patients received AZA). The patient characteristics are summarized in [Table 1]. The median age was 62 years, ranging from 34 to 78 years. Patients included in
this analysis were deemed unfit for IC in view of advanced age, comorbidities, poor
PS, or financial constraints. There were 11 male patients and 12 female patients.
ECOG PS was 1 in 4 patients and ≥2 in 19 patients. HCT-CI was ≤2 in 16 patients and
≥3 in 7 patients. Charlson CI was ≤5 in 19 patients and ≥6 in 4 patients. One patient
received prior therapy with LDAC and two patients received IC. Three patients had
antecedent MDS.
Table 1
Patient characteristics
|
Demographics
|
Patients, n (%)
|
|
ECOG – Eastern Cooperative Oncology Group; PS – Performance status; DM – Diabetes
mellitus; HTN – Hypertension; HCT – Hematopoietic cell transplantation; CI – Comorbidity
index; LDAC – Low-dose cytarabine arabinoside; IC – Induction chemotherapy; MDS –
Myelodysplastic syndrome; WBC – White blood cell; Hb – Hemoglobin; BM – Bone marrow
|
|
Sex
|
|
|
Male
|
11 (47.8)
|
|
Female
|
12 (52.2)
|
|
Median age (years) (range)
|
62 (34-78)
|
|
ECOG PS
|
|
|
0
|
0
|
|
1
|
4 (17.4)
|
|
2
|
8 (34.7)
|
|
3
|
10 (43.4)
|
|
4
|
1 (4.3)
|
|
Comorbidities
|
|
|
Type II DM
|
9 (39.1)
|
|
HTN
|
7 (30.4)
|
|
Renal disease
|
4 (17.4)
|
|
Infections
|
6 (26.1)
|
|
HCT-CI
|
|
|
0
|
6 (26.1)
|
|
1-2
|
10 (43.4)
|
|
≥3
|
7 (30.5)
|
|
Charlson CI
|
|
|
0-3
|
4 (17.4)
|
|
4-5
|
15 (65.2)
|
|
≥6
|
4 (17.4)
|
|
Previous therapy
|
|
|
Nil
|
20 (86.9)
|
|
LDAC
|
1 (4.3)
|
|
IC
|
2 (08.7)
|
|
Antecedent MDS
|
3 (13)
|
|
Hb
|
|
|
Median
|
7.8 (6.7-11.4)
|
|
<10
|
21 (91.3)
|
|
≥10
|
2 (8.7)
|
|
WBC
|
|
|
Median × 103/μL
|
6.2 (0.68-159)
|
|
<5000
|
11 (47.8)
|
|
5000-50,000
|
7 (30.4)
|
|
>50,000
|
5 (21.8)
|
|
Platelet count
|
|
|
Median × 105/μL
|
0.42 (0.02-3.74)
|
|
<50,000
|
13 (56.5)
|
|
≥50,000
|
10 (43.5)
|
|
BM blasts percentage
|
|
|
Median
|
59 (8-95)
|
|
<30
|
4 (17.4)
|
|
30-50
|
5 (21.8)
|
|
>50
|
13 (56.5)
|
|
Missing
|
1 (04.3)
|
|
Cytogenetics
|
|
|
Missing
|
6 (26)
|
|
Favorable
|
2 (8.7)
|
|
Intermediate
|
7 (30.4)
|
|
Adverse
|
8 (34.8)
|
|
Molecular features
|
|
|
Demographics
|
Patients,
n
(%)
|
|
FLT3-ITD
|
2 (8.7)
|
|
NPM1/FLT3-ITD
|
1 (4.3)
|
|
Negative
|
5 (21.7)
|
|
Missing
|
15 (65.3)
|
Median hemoglobin (Hb) was 7.8 g/dl (range 6.7–11.4). The median white blood cell
(WBC) count was 6.2 × 103/μL (range 0.68–159). Median platelet count was 0.42 × 105/μL (0.02–3.74). Four patients (18%) had BM blasts <30% and nineteen patients (82%)
had > 30%, respectively, and median BM blast count was 59% (range 8–95). The 2017
European Leukemia Net risk stratification of AML by genetics was used to risk stratify
patients. Cytogenetics were available in 17 patients (74%). Seven patients (30.4%)
had intermediate-risk and eight patients (34.8%) had poor-risk cytogenetics. Molecular
studies were available only in eight patients (34.8%). Two patients (8.7%) had mutated
FLT3-ITD and one patient (4.3%) had mutated nucleophosmin with FLT3-ITD. Five patients
had no detected mutations. Median dose/day was 100 mg (range 100–149) for AZA and
32 mg (range 25–40) for DAC.
Response to therapy
Only eight patients received three or more cycles of HMAs. Fifteen patients received
<3 cycles because of early death in 11 patients, PD in 4 patients. Initial response
(including CR/CRi/PR according to IWG) was evaluated after three cycles. Among patients
who received three cycles of therapy, seven (87.5%) patients had response to therapy,
(two [25%] had CR, four [50%] had CRi, one [12.5%] had PR) and one (12.5%) had stable
disease [Table 2]. One patient who achieved CR after three cycles of AZA therapy received IC with
high-dose cytarabine (HIDAC). Post-HIDAC, he had PD and was restarted on AZA. He had
PR with AZA and received a total of 9 cycles after which he had PD. One patient achieved
CRi after 3 cycles of DAC therapy and remained in CRi till 8 cycles. He developed
cytopenia and had 8% BM blasts after, continued to receive DAC therapy for 18 cycles
at the time of this publication and remained in PR.
Table 2
Response at initial evaluation and best response
|
Response
|
After 3 cycles (%)
|
Best response (%)
|
|
aPatients who received <3 cycles of therapy. CR – Complete remission; CRi – CR with
incomplete hematologic recovery; PR – Partial remission; SD – Stable disease
|
|
Overall response
|
8 (34.7)
|
8 (34.7)
|
|
CR
|
2 (8.69)
|
3 (13)
|
|
Cri
|
4 (17.4)
|
4 (17.4)
|
|
PR
|
1 (4.35)
|
0
|
|
SD
|
1 (4.35)
|
1 (4.35)
|
|
Othersa
|
15 (65.2)
|
15 (65.2)
|
Median progression-free survival (PFS) was 8 ± 2.39 weeks and median OS was 10 ± 5.29
weeks [Figure 1]. Among patients, who received at least three cycles of therapy, the median PFS was
26 ± 2.82 weeks and median OS was 30 ± 5.66 weeks [Figure 2].
Figure 1: Median progression-free survival and overall survival among all patients
Figure 2: Median progression-free survival and overall survival in patients who received
median three cycles of therapy
Among eight patients who responded to therapy, eight patients (100%) were <65 years
of age, five patients (62.5%) had PS <3, six patients (75%) had HCT-CI <3, seven patients
(87.5%) had Charlson CI <6, six patients (75%) had WBC counts <50,000/cumm, six patients
(75%) had blasts more than 30%, and six patients (75%) had intermediate- or poor-risk
cytogenetics, cytogenetics was not available in the other two patients.
In univariate analysis [Table 3], one of the 10 variables was found to have association with response to HMA therapy.
Patients with the age <65 years had good response to therapy. Cytogenetics, PS, blasts
percentage, WBC counts, Hb, platelet count, sex, and HCT-CI had no significant impact
on response to therapy.
Table 3
Univariate analysis for response to therapy
|
|
Response
|
|
|
OR (95% confidence interval)
|
P
|
|
ECOG – Eastern Cooperative Oncology Group; PS – Performance status; HCT – Hematopoietic
cell transplantation; CI – Comorbidity index; WBC – White blood cell; Hb – Hemoglobin;
OR – Odds ratio
|
|
Age (<65 years)
|
0.53 (0.33-0.85)
|
0.021
|
|
Sex (male/female)
|
0.87 (0.15-4.87)
|
0.87
|
|
ECOG PS (0-1/2+)
|
0.22 (0.03-1.49)
|
0.11
|
|
HCT-CI (>2)
|
3.2 (0.46-22.16)
|
0.26
|
|
Charlson CI (>6)
|
0.73 (0.54-0.99)
|
0.10
|
|
Hb (>10 g/dl)
|
0.87 (0.71-1.05)
|
0.28
|
|
WBC (≥15,000)
|
0.8 (0.53-2.69)
|
0.65
|
|
Platelet count
|
0.68 (0.12-3.96)
|
0.67
|
|
Blasts (>30%)
|
1.06 (0.64-1.71)
|
0.78
|
|
Cytogenetics
|
0.83 (0.64-1.07)
|
0.33
|
None of the variables had a significant impact on response to therapy in the multivariate
analysis. Transfusion independence was noted in three patients. Two patients were
transfusion independent after two cycles of therapy and one patient after four cycles
of therapy.
Discussion
The management of AML in India remains a challenge. In a study conducted in CMC, Vellore,
by Philip et al.[13] 271 (71.31%) of 380 newly diagnosed AML patients did not receive standard of care.
The main reasons for not receiving standard of care were financial constraints, poor
PS and comorbidities. Treatment in AML patients with comorbidities, poor PS, advanced
age, and financial constraints remain a challenge. In India, all patients do not receive
standard of care with IC due to various reasons as discussed earlier.[13] Treatment with HMAs is a reasonable alternative in such individuals as HMAs are
well tolerated with less adverse events when compared with IC. The efficacy of HMAs
in the treatment of AML is well established based on many clinical trials [Table 4].
Table 4
Acute myeloid leukemia trials and subset analysis of acute myeloid leukemia (20%-30%
bone marrow blasts) in myelodysplastic syndrome trials_
|
Reference
|
AML type
|
Median age (years)
|
Drug dose and schedule
|
Patients, n
|
CR/ PR (%)
|
ORRa (%)
|
Median OS (months)
|
|
aORR, including CR/PR, CRi, and/or hematological improvements; b13 patients received study drug after crossover; cCR only; dPR criteria used in this study, included patients with persistent cytopenia; eIf >30% marrow blasts and WBC <15 G/L. BM – Bone marrow; BSC – Best supportive care;
DAC – Decitabine; LDAC – Low-dose cytarabine arabinoside; ICT – Intensive chemotherapy; AML – Acute myeloid leukemia; CR – Complete remission;
PR – Partial remission; AZA – Azacitidine; IV – Intravenous; ORR – Overall response
rate; OS – Overall survival; WBC – White blood cell; CRi – CR with incomplete hematologic
recovery; SQ – Subcutaneous
|
|
Silverman 2006[16]
|
|
|
|
|
|
|
|
|
CALGB 8421
|
AML (20%-30%
|
65
|
AZA, 75 mg/m2/day, for 7 days, IV
|
25
|
4 (12)
|
12 (48)
|
–
|
|
CALGB 9221
|
BM blasts)
|
69
|
Two randomization arms
|
|
|
|
|
|
|
|
|
AZA, 75 mg/m2/day for 7 days, SQ
|
27
|
2 (7)
|
10 (37)
|
19.3
|
|
|
|
|
BSC
|
25b
|
0 (0)
|
2 (8)b
|
12.9
|
|
Fenaux 2010[15]
|
AML (20%-30% BM blasts)
|
70
|
Two randomization arms AZA, 75 mg/m2/day for 7 days, SQ
|
55
|
10 (18)c
|
|
24.5
|
|
|
|
|
BSC, LDAC, or ICT
|
58
|
9 (16)c
|
|
16
|
|
Lubbert 2012[20]
|
WHO AML
|
72
|
DAC 15 mg/m2/8 h for 3 days, IV
|
227
|
59 (26)d
|
-
|
5.5
|
|
Kantarjian 2012[21]
|
WHO AML
|
73
|
DAC 20 mg/m2/day, for 5 days, IV
|
242
|
44 (18)
|
73 (30)
|
7.7
|
|
|
|
|
BSC or LDAC
|
243
|
27 (11)
|
34 (14)
|
5
|
|
Dombret 2015[27]
|
WHO AMLe
|
75
|
AZA, 75 mg/m2/day for 7 days, SQ
|
231
|
50 (22)
|
70 (30)
|
10.4
|
|
|
|
|
BSC, LDAC, or ICT
|
247
|
57 (23)
|
65 (26)
|
6.5
|
Azacitidine
Cancer and Leukemia Group B (CALGB) cooperative group evaluated the efficacy of HMAs
in AML/MDS patients. Therapy with AZA in MDS resulted in response rates ranging from
30% to 60%, with documented improved survival.[14]
[15]
[16] In the phase 3 AZA-001 trial, older patients with 20%–30% BM blasts treated with
AZA had prolonged OS compared with conventional care regimens (CCRs).[15] In the Austrian Azacitidine Registry[17]
[18] and French compassionate use program,[19] patients with AML treated with AZA had a median OS of approximately 9–10 months.
AZA was approved for use in AML in 2004 in the USA based on the CALG 9221 trial.
Decitabine
Various dosing schedules of DAC had been studied in AML patients. The European organization
for Treatment of Cancer cooperative group trial in MDS/AML patients using a DAC at
15 mg/m2 dose every 8 h 3 days schedule resulted in complete and PR rates of 26% with no difference
between patients with or without adverse cytogenetics. The median OS was 5.5 months,
1-year survival and 2-year survival rates were 28% and 13%, respectively.[20] DACO-016 study compared the efficacy and safety of DAC (20 mg/m2/day for 5 days every 4 weeks) versus treatment choice in 485 patients IC ineligible
patients.[21] Planned primary analysis of this trial did not show a significant improvement of
OS (median OS 7.7 months vs. 5.0 months), follow-up analysis was in favor of DAC.
DAC was approved for the treatment of AML in Europe based on the data from this study.
In our retrospective study of 23 patients with AML, who were treated with HMAs, a
response rate of 34.7% (including CR/Cri/PR) was observed. In a study of AZA in untreated
AML by Thépot et al., the best response rate was 33%.[19] Our results are consistent with those reported in other studies.[21]
[22] AZA dose was capped at 100 mg/day in patients who had financial constraints. The
median time to best response was 136 days.
All the patients who had response to therapy were younger than 65 years of age. In
our study, younger patients were offered HMA in view of ineligibility for IC regimens
due to various reasons as stated earlier. Age <65 years had a significant impact on
the response to therapy. Patients with higher comorbidity indices (HCT-CI >2, Charlson
CI > 5) did not have good response to therapy in our study, indicating that comorbidities
have a significant impact on response and survival in AML patients treated with HMAs.
Hb level, platelet counts, and WBC did not have an impact on response to therapy in
our trial. However, in patients who presented with leukopenia had a nonsignificant
trend toward worse outcomes (odds ratio - 1.42, 95% confidence interval 0.51–3.91).
AZA therapy prolonged OS compared with CCRs in older patients with 20%–30% BM blasts
in the phase 3 AZA-001 trial.[15] The same results could not be established in our study as the patient population
with blasts <30% is very small (17%).
IC and LDAC provide no OS benefit in older patients with AML and poor cytogenetics,
and in such patients, HMA therapy provides better outcomes.[23]
[24]
[25] In our study, patients cytogenetic risk had no impact on response to therapy. This
can be attributed to the nonavailability of risk stratification in all patients and
small sample size.
The median PFS and OS observed in our study are 6.06 ± 0.65 months and 7 ± 1.32 months,
respectively, for patients who received a minimum of three cycles of therapy. In a
multicenter DAC phase II trial in 227 older AML patients, CR/PR rate was 26% and median
OS was 5.5 months.[26] Our results are comparable with other studies.[17]
[19]
[22]
[27] When survival rates are evaluated in the entire study group, the median PFS and
OS were 1.86 ± 0.55 months and 2.33 ± 1.23 months’ respectively. The decrease in the
survival rates can be attributed to early deaths in 15 (65%) patients. The incidence
of adverse events could not be evaluated in the study group due to lack of documentation,
as the study being a retrospective analysis.
Conclusion
AZA and DAC provide an important additional treatment option in newly diagnosed AML
patients who are older, with poor PS, higher comorbidity indices and who refuse IC.
