AUTHORS
Amer M. Zeidan, MD, Sophia Zhao, PhD2, Dina Christensen3, Mie Yang4, Xiaoyan Wang5, Maximilian Stahl, MD6, Fang Sun, MD7
AFFILIATIONS
1 Yale University School of Medicine, New Haven, United States
2 Taiho Oncology, Princeton, United States
3 Taiho Oncology, Princeton, United States
4 NouStarX, Westport, United States
5 NouStarX, Westport, United States
6 Yale University School of Medicine, New Haven, United States
7 Taiho Oncology, Princeton, United States
ABSTRACT
Background. Hypomethylating agents (HMAs) remain the standard of care for patients with intermediate- to high-risk myelodysplastic syndromes (MDS), chronic myelomonocytic leukemia (CMML), and low-blast acute myeloid leukemia (AML), where TP53 mutation confers poorer prognosis. With the approval of oral decitabine-cedazuridine, both oral and parenteral HMAs are now available. However, their efficacy/effectiveness and safety have not been comprehensively reviewed and synthesized. We conducted a systematic literature review (SLR) and meta-analysis of studies evaluating oral decitabine-cedazuridine (Oral DEC-C), parenteral azacitidine (Par-AZA), and parenteral decitabine (Par-DEC) in intermediate- to high-risk MDS, CMML, and low-blast AML.
Methods. Following PRISMA guidelines, MEDLINE, Embase, and the Cochrane Library were searched for studies published between 2015 and 2025, together with conference proceedings from 2023–2025. Study selection followed predefined PICO criteria. Random-effects single-arm meta-analyses were performed for efficacy/effectiveness outcomes—median overall survival (OS) and overall response rate (ORR)—and grade ≥3 safety outcomes (thrombocytopenia, neutropenia, and infection). Median OS was pooled by inverse-variance meta-analysis on the log scale. ORR and safety outcomes were pooled as proportions. Heterogeneity was assessed using Higgins’ I2. Sources of heterogeneity for OS were explored by random-effects meta-regression (candidate moderators: log sample size, risk category, publication year, study design), restricted to groups with sufficient studies (Par-AZA, Par-DEC). All pooled estimates represent single-arm summaries rather than head-to-head comparisons.
Results. The systematic review identified 276 eligible studies. Of these, 63 studies (71 treatment cohorts) evaluating HMA monotherapy were included in the OS meta-analysis: Par-AZA (52 cohorts, ~13,000 patients), Par-DEC (15 cohorts; ~1,700 patients), and Oral DEC-C (4 cohorts; ~620 patients). The ORR and grade ≥3 safety analyses comprised 86 and 22-30 cohorts, respectively. Pooled median OS was numerically longer for Oral DEC-C [24.4 months; 95% CI 19.0–31.3; I²=83%; k=4], followed by Par-DEC [17.1 months; 95% CI 15.5–18.9; I²=53%; k=15] and Par-AZA [16.3 months; 95% CI 14.7–18.2; I²=91%; k=52]. Among patients with TP53-mutated disease, the mean of study-level median OS was 9.2 months (observed range 5.9–15.6; k=8) for Par-AZA and 11.9 months (9.3–14.0; k=3) for Par-DEC, while one Oral DEC-C study reported median OS of 22.1 months in single-hit and 11.5 months in multi-hit TP53 disease.
Pooled ORR was 60.9% [95% CI 54–67%; I²=0%; k=2] for Oral DEC-C, compared with 55.5% [95% CI 50–61%; I²=81%; k=29] for Par-DEC and 49.1% [95% CI 45–53%; I²=85%; k=55] for Par-AZA. Pooled grade ≥3 thrombocytopenia was 38.6% [95% CI 29.1–48.6; I²=94%; k=20] for Par-AZA, 46.2% [95% CI 24.5–68.6; I²=96%; k=7] for Par-DEC, and 47.4% [95% CI 31.8–63.2; I²=83%; k=3] for Oral DEC-C; grade ≥3 neutropenia was 44.9% [95% CI 35.3–54.8; I²=94%; k=20] for Par-AZA, 61.3% [95% CI 36.6–83.3; I²=97%; k=6] for Par-DEC, and 45.6% [95% CI 30.2–61.3; I²=83%; k=3] for Oral DEC-C; and grade ≥3 infection was 19.4% [95% CI 9.0–32.4; I²=97%; k=15] for Par-AZA, 25.6% [95% CI 14.3–38.9; I²=87%; k=4] for Par-DEC, and 10.2% [95% CI 4.3–18.1; I²=66%; k=3] for Oral DEC-C. In meta-regression of median OS, log (sample size) was the only significant predictor (β=−0.0878 for Par-AZA, p=0.001; β=−0.195 for Par-DEC, p<0.001), explaining R²=24.8% (Par-AZA) and 69.5% (Par-DEC) of between-study variance. Risk category, publication year, and study design were not significant (all p>0.3).
Conclusions.
In this SLR and meta-analysis, Oral DEC-C, Par-AZA, and Par-DEC showed broadly comparable single-arm efficacy/effectiveness and grade ≥3 safety, with numerically longer OS for Oral DEC-C from a small evidence base. Between-study heterogeneity was substantial and partly attributable to study size, underscoring that these indirect, single-arm comparisons should be interpreted cautiously. Prospective head-to-head trials are warranted.
DISCLOSURES
Amer M. Zeidan, MD: AbbVie, Akesobio, Agios, Amgen, Astellas, BioCryst, BioThera, BeiOne, Boehringer Ingelheim, Celgene/BMS, Chiesi/Cornerstone, Daiichi-Sankyo, Dr.Reddy’s, Epizyme, Faron, Fibrogen, GSK, Glycomimetics, Genentech, Gilead, Geron, Innocare, Janssen, Jasper, Karyopharm, Kyowa Kirin, Keros, Kura, Novartis, Notable, Oncoverity, Orum, Otsuka, Pfizer, Puretech/Gallop Technology, Regeneron, Rigel, Seattle Genetics, Shattuck Labs, Schrodinger, Sumitomo, Syros, Syndax, Systimmune, Sillajen, Servier, Takeda, Treadwell, Taiho, Vincerx, and Zentalis - Consultancy (Includes expert testimony); AbbVie, Akesobio, Agios, Amgen, Astellas, BioCryst, BioThera, BeiOne, Boehringer Ingelheim, Celgene/BMS, Chiesi/Cornerstone, Daiichi-Sankyo, Dr.Reddy’s, Epizyme, Faron, Fibrogen, GSK, Glycomimetics, Genentech, Gilead, Geron, Innocare, Janssen, Jasper, Karyopharm, Kyowa Kirin, Keros, Kura, Novartis, Notable, Oncoverity, Orum, Otsuka, Pfizer, Puretech/Gallop Technology, Regeneron, Rigel, Seattle Genetics, Shattuck Labs, Schrodinger, Sumitomo, Syros, Syndax, Systimmune, Sillajen, Servier, Takeda, Treadwell, Taiho, Vincerx, and Zentalis - Consultancy (Includes expert testimony)