ABSTRACT
Patients with newly diagnosed acute myeloid leukemia (ND‐AML) derive variable survival benefit from venetoclax + hypomethylating agent (Ven‐HMA) therapy. The primary objective in the current study was to develop genetic risk models that are predictive of survival and are applicable at the time of diagnosis and after establishing treatment response. Among 400 ND‐AML patients treated with Ven‐HMA at the Mayo Clinic, 247 (62%) achieved complete remission with (CR) or without (CRi) count recovery. Multivariable analysis–derived hazard ratios (HR), including 1.8 for European LeukemiaNet (ELN) adverse karyotype, 4.7 for KMT2Ar, 1.7 for TP53 MUT, 2.6 for KRAS MUT, and 2.1 for IDH2 WT were applied to develop an HR‐weighted risk model: low, intermediate, and high; respective median survival censored for allogeneic stem cell transplant (ASCT) (3‐year survival) were “not reached” (67%), 19.1 (33%), and 7.1 months (0%). In patients achieving CR/CRi, adverse karyotype, KMT2Ar, KRAS MUT, IDH2 WT predicted inferior survival, allowing for a complementary response‐stratified risk model. The model was externally validated and was shown to be superior to the ELN 2024 risk model (AIC 179 vs. 195 and AUC 0.77 vs. 0.69). Survival was inferior with failure to achieve CR/CRi or not receiving ASCT; 3‐year survival for high‐risk with or without ASCT was 42% versus 0% (p < 0.01); intermediate 72% versus 43% (p = 0.06); and low‐risk 88% versus 78% (p = 0.53). The Mayo genetic risk models offer pre‐treatment and response‐based prognostic tools for ND‐AML treated with Ven‐HMA. The current study underscores the prognostically indispensable role of achieving CR/CRi and ASCT.
Keywords: karyotype, mutations, remission, survival, venetoclax
The Mayo genetic risk models offer pre‐treatment and response‐based prognostic tools for newly diagnosed AML treated with Ven‐HMA.
1. Introduction
Venetoclax + hypomethylating agent (Ven‐HMA) combination therapy has shown favorable response rates and overall survival (OS) in unfit patients with newly diagnosed acute myeloid leukemia (ND‐AML) [1, 2, 3]. However, there remains substantial heterogeneity in survival outcomes [4, 5]. Achievement of complete remission with (CR) or without (CRi) count recovery has long been recognized to have independent association with improved survival in AML while refractory/relapsed disease is associated with significantly shorter survival. In the seminal phase 3 VIALE‐A study [6], Ven‐azacitidine was compared to azacitidine‐placebo, in ND‐AML, and resulted in CR/CRi of 66% with a median OS of 14.7 months. The study also showed significantly longer OS in patients without versus with measurable residual disease (median 34.2 vs. 18.7 months) [6, 7]. Since then [6], several retrospective studies have looked into clinical and genetic markers that are predictive of Ven‐HMA treatment response and post‐Ven‐HMA overall and relapse‐free survival (RFS) [4, 8, 9, 10]. The most recent publication, in this regard, highlighted a 4‐gene molecular signature for Ven‐HMA response and survival, based on the presence or absence of TP53, KRAS, NRAS, and FLT3‐ITD mutations [11]. Other studies have suggested different sets of genetic predictors of outcome, including IDH2 MUT, NPM1 MUT, DDX41 MUT, and adverse karyotype [4, 10, 12]. In the current study, we took advantage of a recently updated and expanded institutional database, in order to develop enhanced genetic risk models for treatment response and survival following Ven‐HMA therapy in ND‐AML. Two independent external patient cohorts were accessed for model validation.
2. Methods
Patients were retrospectively recruited from Mayo Clinic, USA (MN, AZ, FL), after Institutional Review Board approval and based on documentation of ND‐AML and treatment with at least one cycle of Ven‐HMA. All patients were treated outside clinical trials between November 2018 and May 2024 with follow‐up updated in July 2024. Cytogenetic and molecular studies (48‐gene panel) were performed through conventional karyotyping and next‐generation sequencing (NGS), respectively. KMT2A (also known as MLL) rearrangements (KMT2Ar) were confirmed by fluorescence in situ hybridization (FISH). Patients received either azacitidine 75 mg/m2 intravenously or subcutaneously days 1–7 or decitabine 20 mg/m2 intravenous days 1–5 plus Ven (median 200 mg; range 50–400 mg) by mouth daily for 7–28 days during the first cycle. Diagnosis, cytogenetic risk stratification, and response assessments were conducted according to the ELN 2022 criteria [13]. In the majority of cases, response was assessed after completion of one or two cycles, based on treating physician discretion. MRD was quantified using multiparameter flow cytometry with a minimum sensitivity of 0.01% and measured once CR or CRi with < 5% bone marrow blasts was documented. Relapse was defined by the emergence of ≥ 5% bone marrow or peripheral blood blasts, in patients with CR/CRi. RFS was calculated from the time of remission to relapse or last follow‐up/death and OS from the time of Ven‐HMA initiation to last‐follow up/death. During survival analysis, patients receiving allogeneic stem cell transplant (ASCT) were censored at the time of transplantation. Cox proportional hazard regression model was used to evaluate covariate associations with OS and RFS and the Kaplan–Meier method used to estimate RFS and OS. Risk models were developed using hazard ratio (HR)‐based risk point allocation and predictive accuracy was compared using Akaike Information Criterion (AIC) and area under the ROC curve (AUC). Statistical analyses were conducted using JMP Pro 18.0.0 software (SAS Institute, Cary, NC, USA).
3. Results
3.1. Patient Characteristics and Treatment Details
A total of 400 adult patients with ND‐AML were considered: median age 73 years (range: 19–98); 64% males; 95% white; 60% de novo; 22% secondary (post‐myelodysplastic syndrome [MDS], or myelodysplastic/myeloproliferative neoplasm [MDS/MPN]); and 18% therapy‐related. Treatment included a median of 4 cycles (range: 1–65) of Ven with decitabine (n = 265, 66%) or azacitidine (n = 148, 37%). Prior HMA exposure was documented in 30 (8%) patients. Cycle 1 Ven duration was 28 days (n = 237, 59%), 21 days (n = 59, 15%), 14 days (n = 63, 16%), or 7 days (n = 9, 2%). Key clinical and laboratory findings at time of treatment initiation are described in Table 1. ELN cytogenetic risk was evaluable in 398 cases and included favorable 2% (n = 7), intermediate 60% (n = 240), or adverse 38% (n = 151). 102 (26%) and 7 (2%) of cases harbored complex or monosomal karyotype and KMT2Ar, respectively. Recurrently mutated genes included TP53 in 26% (101/394, 90% multi‐hit), TET2 19% (74/387), RUNX1 19% (76/392), SRSF2 18% (70/387), ASXL1 18% (70/387), DNMT3A 15% (57/392), NPM1 12% (49/394), IDH2 12% (48/393, 9% R140Q, 3% R172K), FLT3‐ITD 10% (40/397), NRAS 9% (34/392), BCOR 8% (32/387), STAG2 8% (22/376), IDH1 7% (26/393), U2AF1 6% (22/372), SF3B1 5% (19/376), KRAS 4% (15/392), and DDX41 in 4% (14/376) of informative cases; 36% of DDX41 MUT were confirmed as germline.
TABLE 1.
Clinical characteristics at time of treatment with venetoclax and hypomethylating agent for 400 patients with newly diagnosed acute myeloid leukemia stratified by achievement of complete response with (CR) or without (CRi) count recovery.
| Variables | All patients N = 400 | Patients in CR/CRi N = 247 (62%) | Patients not in CR/CRi N = 153 (38%) | Univariate p‐value |
|---|---|---|---|---|
| Age in years, median (range) | 73 (19–98) | 73 (37–91) | 73 (19–98) | 0.85 |
| Male, n (%) | 256 (64) | 152 (59) | 104 (41) | 0.19 |
| Race, n (%) | 393 | 245 | 148 | 0.68 |
| White | 375 (95) | 235 (63) | 140 (37) | |
| Black | 9 (2) | 6 (67) | 3 (33) | |
| Asian | 3 (1) | 1 (33) | 2 (67) | |
| Other | 6 (2) | 3 (50) | 3 (50) | |
| AML type, n (%) | ||||
| De novo | 239 (60) | 157 (66) | 82 (34) | 0.07 |
| Secondary | 87 (22) | 45 (52) | 42 (48) | 0.03 |
| Therapy‐related | 74 (18) | 45 (62) | 29 (38) | (Secondary vs. others) |
| Prior HMA, n (%) | 30 (8) | 5 (17) | 25 (83) | < 0.01 |
| Hemoglobin, g/dl, median (range) | 8.5 (2.7–15.3) | 8.7 (2.7–15.3) | 8.2 (5.1–13) | 0.05 |
| Leukocyte count × 109/L, median (range) | 3.5 (0.4–243.8) | 3.09 (0.4–200) | 4.21 (0.5–243.8) | 0.91 |
| Platelet count × 109/L, median (range) | 53 (5–601) | 62 (7–601) | 40 (5–473) | 0.02 |
| Circulating blasts %, median (range) | 11 (0–93) | 11 (0–93) | 12 (0–92) | 0.37 |
| Bone marrow blasts %, median (range) | 43 (1–97) | 43 (1–95) | 43 (5–97) | 0.89 |
| ELN 2022 cytogenetic risk stratification, n (%) | 398 | 246 | 152 | |
| Favorable | 7 (2) | 5 (71) | 2 (29) | |
| Intermediate | 240 (60) | 169 (70) | 71 (30) | < 0.01 |
| Adverse | 151 (38) | 72 (48) | 79 (52) | |
| KMT2A rearrangement, a n (%) | 7/398 (2) | 3/246 (43) | 4/152 (57) | 0.31 |
| Mutations on NGS, evaluable, n (%) | ||||
| TP53 (394) | 101 (26) | 46 (46) | 55 (54) | < 0.01 |
| RUNX1 (392) | 76 (19) | 37 (49) | 39 (51) | < 0.01 |
| TET2 (387) | 74 (19) | 50 (68) | 24 (32) | 0.23 |
| SRSF2 (387) | 70 (18) | 50 (71) | 20 (29) | 0.06 |
| ASXL1 (387) | 70 (18) | 43 (61) | 27 (39) | 0.99 |
| DNMT3A (392) | 57 (15) | 44 (77) | 13 (23) | < 0.01 |
| NPM1 (394) | 49 (12) | 41 (84) | 8 (16) | < 0.01 |
| IDH2 (393) | 48 (12) | 37 (77) | 11 (23) | 0.02 |
| IDH1 (393) | 26 (7) | 20 (77) | 6 (23) | 0.09 |
| FLT3‐ITD (397) | 39 (10) | 16 (41) | 23 (59) | < 0.01 |
| NRAS (392) | 34 (9) | 22 (65) | 12 (35) | 0.71 |
| BCOR (387) | 32 (8) | 16 (50) | 16 (50) | 0.17 |
| U2AF1 (372) | 22 (6) | 12 (55) | 10 (45) | 0.47 |
| CEBPA (394) | 22 (6) | 18 (82) | 4 (18) | 0.04 |
| CEBPA bZIP (393) | 12 (3) | 10 (83) | 2 (17) | 0.10 |
| STAG2 (376) | 22 (8) | 17 (77) | 5 (23) | 0.11 |
| SF3B1 (376) | 19 (5) | 10 (53) | 9 (47) | 0.41 |
| KRAS (392) | 15 (4) | 8 (53) | 7 (47) | 0.49 |
| DDX41 (376) | 14 (4) | 13 (93) | 1 (7) | < 0.01 |
| PTPN11 (376) | 12 (3) | 7 (58) | 5 (42) | 0.81 |
| EZH2 (387) | 11 (3) | 4 (36) | 7 (64) | 0.09 |
| WT1 (387) | 11 (3) | 4 (36) | 7 (64) | 0.09 |
| SETBP1 (387) | 11 (3) | 5 (45) | 6 (55) | 0.27 |
| PHF6 (376) | 10 (3) | 6 (60) | 4 (40) | 0.91 |
| JAK2 (376) | 9 (2) | 7 (78) | 2 (22) | 0.30 |
| CBL (375) | 7 (2) | 4 (57) | 3 (43) | 0.79 |
| HMA used, n (%) | ||||
| Azacitidine | 144 (36) | 88 (62) | 56 (38) | 0.84 |
| Decitabine | 256 (64) | 159 (62) | 97 (38) | |
| Final dose of Venetoclax, mg, median (range) | 200 (50–400) | 200 (50–400) | 200 (50–400) | 0.29 |
| Allogeneic transplant, n (%) | 65 (16) | 60 (92) | 5 (8) | < 0.01 |
Abbreviations: ELN, European Leukemia Net; HMA, hypomethylating agent; NGS, next generation sequencing.
KMT2A‐PTD were not assessed.
3.2. Clinical and Genetic Predictors of Response
Overall, 153 (38%) patients achieved CR, and 94 (24%) CRi, resulting in CR or CRi in 247 (62%) of patients. CR/CRi rates were lower in patients with prior HMA exposure (17% vs. 65%; p < 0.01). MRD by multiparameter flow cytometry was undetected in 117 (70%) of 166 informative cases. Median time to CR/CRi was 1.3 months (range: 1–9) and median remission duration 6 months (range: 1–31). Table 1 compares the clinical and genetic profile of responders versus non‐responders. CR/CRi rates were inferior in patients with secondary AML (52% vs. 65%; p = 0.03). Considering genetic variables only, higher CR/CRi rates were seen with NPM1 MUT (86% vs. 59%; p < 0.01), IDH2 MUT (77% vs. 60%; p = 0.02; IDH2 MUT 172 K vs. 140Q) (91% vs. 73%; p = 0.18), DDX41 MUT (93% vs. 61%; p = 0.01), or DNMT3A MUT (77% vs. 59%; p < 0.01); SRSF2 MUT (71% vs. 59%; p = 0.06) or IDH1 MUT (77% vs. 61%; p = 0.09) displayed borderline significance. CR/CRi rates were lower in the presence of TP53 MUT (45% vs. 67%; p < 0.01), FLT3‐ITDMUT (41% vs. 64%; p = 0.01), or RUNX1 MUT (49% vs. 65%; p < 0.01) or ELN 2022 adverse karyotype (48% vs. 71%; p < 0.01). Of note, CR/CRi rates were not influenced by the presence of KRAS MUT (53% vs. 62%; p = 0.49) or NRAS MUT (65% vs. 61%; p = 0.71) mutations, or KMT2Ar (43% vs. 62%; p = 0.31). In a separate analysis which excluded patients with prior HMA exposure, results were unchanged, except secondary AML did not impact response rates (CR/CRi; 65% vs. 66%; p = 0.87).
Table S1 lists predictors of treatment response based on univariate and multivariable analyses, inclusive of (i) mutations alone, (ii) mutations and karyotype, and (iii) mutations, karyotype, and clinical variables. In multivariable analysis that included both clinical and genetic variables, response rates were higher in the presence of NPM1 MUT (p < 0.01; OR: 0.41) and lower in the presence of secondary AML (p = 0.03; OR: 1.8), adverse karyotype (p < 0.01; OR: 2.3), TP53 MUT (p = 0.04; OR: 1.9), FLT3‐ITD MUT (p < 0.01; OR: 4.8), or RUNX1 MUT (p < 0.01; OR: 2.4). In multivariate analysis of karyotype and mutations, independent predictors of superior response included NPM1 MUT (p < 0.03; OR: 0.53) and inferior response adverse karyotype (p < 0.01; OR: 2.3), TP53 MUT (p = 0.04; OR: 1.9), FLT3‐ITD MUT (p < 0.01; OR: 4.8), or RUNX1 MUT (p < 0.01; OR: 2.4). The presence of TP53 MUT did not appear to influence response rates among patients with either ELN defined non‐adverse (CR/CRi; 57% vs. 71% in TP53 MUT vs. TP53 WT; p = 0.45) or adverse karyotype (CR/CRi, 45% vs. 53% in TP53 MUT vs. TP53 WT; p = 0.34). Similar findings were obtained when complex or monosomal karyotype was considered.
In multivariable analysis limited to mutations, NPM1 MUT, IDH2 MUT, and DDX41 MUT were identified as positive and TP53 MUT, FLT3‐ITD MUT, and RUNX1 MUT as negative predictors of treatment response. CR/CRi rate was highest at 87% in patients harboring one or more favorable mutations (NPM1, IDH2, DDX41) and no unfavorable mutation (TP53, FLT3‐ITD, RUNX1). CR/CRi rate was lowest at 44% in patients with at least one unfavorable mutation and no favorable mutation (p < 0.01). CR/CRi rates were similar in patients with neither favorable nor unfavorable mutations versus those harboring favorable with unfavorable mutations (73% vs. 63%; p > 0.1) (Figure 1). Application of the VIALE‐A 4‐gene signature for Ven‐HMA response yielded CR/CRi rates of 46% in the presence of TP53 MUT (lower‐benefit group), 52% in TP53 WT and presence of either FLT3‐ITD MUT or KRAS/NRAS MUT (intermediate group) and 73% in the absence of TP53 MUT, KRAS/NRAS MUT, or FLT3‐ITD MUT (higher‐benefit group) (Figure 1); in particular, CR/CRi rates were found to be similar for intermediate vs. lower‐benefit groups (52% vs. 46%; p = 0.39). Among NPM1 MUT and IDH2 MUT cases, presence of FLT3‐ITD MUT and K/NRAS MUT did not appear to significantly influence response rates; among 48 patients with IDH2 MUT, FLT3‐ITD MUT was present in 5 (10%) while 2 (4%) harbored K/NRAS MUT; CR/CRi rate was 60% versus 79% (presence vs. absence of FLT3‐ITD MUT p > 0.1) and 100% versus 76% (presence vs. absence of K/NRAS MUT) (p > 0.1). Among 49 NPM1 MUT patients, 11 (22%) harbored FLT3‐ITD MUT (CR/CRi; 73% vs. 87%; p = 0.29), and 13 (27%) K/NRAS MUT (CR/CRi; 85% vs. 83%; p = 0.88). Of note, DDX41 mutated patients did not harbor either FLT3‐ITD MUT or K/NRAS MUT.
FIGURE 1.
Genetic signature for response in newly diagnosed acute myeloid leukemia treated with venetoclax and hypomethylating agent. [Color figure can be viewed at wileyonlinelibrary.com]
CR rates in comparison to CRi were lower in FLT3‐ITD (38% vs. 63%; p = 0.04), SRSF2 (46% vs. 54%; p = 0.016), and ASXL1 mutated cases (42% vs. 58%; p < 0.01). A trend for lower CR versus CRi rates was observed in IDH2 (49% vs. 51%; p = 0.09), KRAS (33% vs. 68%; p = 0.16), and RUNX1 mutated (49% vs. 51%; p = 0.09); on the other hand, CR rates were significantly higher in DDX41 mutated (92% vs. 8%; p < 0.01), and non‐significantly higher CR versus CRi rates were observed in TP53 mutated cases (69% vs. 30%; p = 0.19).
3.3. Clinical and Genetic Predictors of Relapse
Relapse was documented in total of 99 (40%) patients who achieved CR (39%) or CRi (41%) after a median remission duration of 6 months (1–31). Patients that relapsed within the first‐year versus those that did not were more likely to be males (75% vs. 55%), MRD positive (44% vs. 24%), and harbor TP53 MUT (23% vs. 16%). Univariate analysis for RFS identified male gender (12 vs. 29 months; p = 0.02), secondary AML (9 vs. 18 months; p < 0.01), adverse karyotype (9 vs. 22 months; p < 0.01), TP53 MUT (6 vs. 18 months, p < 0.01), and MRD positive remission (13 months vs. not reached; p < 0.01) as risk factors for relapse; CRi versus CR were borderline significant (15 vs. 17 months; p = 0.09); on the other hand, IDH2 MUT (not reached vs. 15 months; p < 0.01) were associated with a lower risk of relapse.
Multivariable analysis inclusive of genetic variables and MRD status confirmed TP53 MUT (HR: 2.4) and MRD positive remission (HR: 2.0) as unfavorable predictors of relapse (Table 2). Subsequently, in a three‐tiered relapse prediction model, RFS was significantly inferior in the presence of two risk factors (TP53 MUT, MRD positive remission) (n = 11, median 4 months), vs. one (n = 57, median 16 months) versus none of the risk factors (n = 96, median not reached) (p < 0.01); 1‐year cumulative incidence of relapse was 84%, 45%, and 28% in the presence of two, one and none of the risk factors, respectively (Figure S1).
TABLE 2.
Predictors of overall survival and relapse‐free survival in 400 patients with newly diagnosed acute myeloid leukemia receiving venetoclax plus hypomethylating agent therapy.
| Variables | Overall survival transplant‐censored | Relapse‐free survival Relapse, N = 166 | ||||
|---|---|---|---|---|---|---|
| Univariate p‐value HR (95% CI) | Multivariate with pre‐treatment variables clinical + genetics p‐value HR (95% CI) | Multivariate with genetic variables only p‐value HR (95% CI) | Multivariate with Response p‐value HR (95% CI) | Univariate p‐value HR (95% CI) | Multivariate with genetic variables p‐value HR (95% CI) | |
| Age |
0.83 |
0.17 | ||||
| Gender |
< 0.01 1.6 (1.2–2.2) Male vs. female |
< 0.01 2.1 (1.5–2.9) |
< 0.01 2.2 (1.6–2.9) Male vs. female |
0.02 1.7 (1.1–2.6) Male vs. female |
||
| Secondary AML |
< 0.01 1.5 (1.2–1.9) Presence vs. Absence |
< 0.01 1.6 (1.2–2.2) |
0.05 |
< 0.01 1.8 (1.2–2.9) Presence vs. Absence |
||
| ELN 2022 adverse karyotype |
< 0.01 2.6 (2.1–3.4) Presence vs. Absence |
< 0.01 1.9 (1.3–2.8) |
< 0.01 1.8 (1.2–2.6) |
< 0.01 1.8 (1.3–2.7) |
< 0.01 2.3 (1.5–3.5) Presence vs. Absence |
0.31 |
|
KMT2A rearrangement |
< 0.01 5.5 (2.4–12.5) |
< 0.01 7.9 (3.3–19.1) |
< 0.01 4.7 (2.0–11.1) |
< 0.01 7.8 (3.3–18.6) |
0.15 | |
| TP53 mutation |
< 0.01 2.6 (1.9–3.5) Presence vs. Absence |
< 0.01 1.9 (1.3–2.8) |
< 0.01 1.7 (1.2–2.6) |
0.07 |
< 0.01 2.4 (1.5–2.9) Presence vs. Absence |
< 0.01 2.4 (1.3–4.4) |
| IDH2 mutation |
< 0.01 2.6 (1.5–4.4) Absence vs. Presence |
0.01 2.0 (1.1–3.4) |
< 0.01 2.1 (1.2–3.6) |
< 0.01 2.4 (1.4–4.1) |
< 0.01 2.6 (1.2–5.7) Absence vs. Presence |
0.10 |
| IDH1 mutation |
0.01 2.1 (1.1–4.1) Absence vs. Presence |
0.19 |
0.08 |
0.66 | 0.68 | |
| NPM1 mutation |
0.06 1.5 (0.98–2.3) Absence vs. Presence |
0.86 | 0.13 | |||
| DNMT3A mutation |
0.01 1.6 (1.1–2.4) Absence vs. Presence |
0.48 | 0.21 | 0.45 |
0.51 |
|
| FLT3‐ITD mutation |
0.63 |
0.31 | ||||
|
RUNX1 mutation |
0.73 | 0.62 | ||||
| DDX41 mutation |
0.02 3.2 (1.2–8.7) Absence vs. Presence |
0.07 |
0.08 |
0.09 |
0.83 | |
|
KRAS mutation |
0.05 1.7 (1.0–3.2) Presence vs. Absence |
0.02 2.1 (1.2–3.9) |
< 0.01 2.2 (1.2–4.0) |
0.15 | 0.81 | |
|
NRAS mutation |
0.46 | 0.25 | ||||
|
CR/CRi |
< 0.01 5.4 (3.5–6.8) Absence vs. Presence |
< 0.01 5.2 (3.9–6.9) |
||||
| MRD positive vs. negative | 0.31 |
< 0.01 2.1 (1.3–3.6) Presence vs. Absence |
< 0.01 2.0 (1.2–3.4) |
|||
Abbreviations: CR, complete remission; CRi, CR with incomplete count recovery; ELN, European Leukemia Net; MRD, measurable residual disease.
50 of 99 (65%) patients received salvage therapy which included cladribine‐cytarabine‐Ven (n = 19), intensive induction chemotherapy (n = 9), FLT3 inhibitors (n = 5), Ven‐HMA‐FLT3 inhibitor (n = 1), IDH1/2 inhibitors (n = 5), glasdegib‐cytarabine (n = 4), gemtuzumab (n = 3), lenalidomide (n = 1), or investigational therapies (n = 3).
3.4. Risk Factor Analysis for Overall Survival and Development of a New Genetic Risk Model
After a median follow‐up of 10.5 months (range 0.5–66), 248 death (62%; including 62 within 3 months of treatment initiation) and 65 ASCT (16%; including 60 in CR/CRi) were documented. Median OS for the cohort was 12.6 months; 30‐day and 60‐day mortality rates were 4% and 11%, respectively. Table 2 outlines clinical and genetic variables found to affect transplant‐censored OS in univariate and multivariate analyses. Univariate analysis for transplant‐censored OS disclosed superior survival in presence of CR/CRi (median 21 vs. 3.6 months; HR: 5.4), IDH2 MUT (not reached vs. 11 months; HR: 2.6), IDH1 MUT (not reached vs. 11.9 months; HR: 2.1), DNMT3A MUT (18.2 vs. 10.9 months; HR: 1.6), or DDX41 MUT (25.8 vs. 11.9 months; HR: 3.2). By contrast, male gender (9.7 vs. 19.2 months; HR: 1.6), secondary AML (9.7 vs. 14.1 months; HR: 1.5), adverse karyotype (7.4 vs. 18.6 months; HR: 2.6), KMT2Ar (2.5 vs. 12.9 months; HR: 5.5), TP53 MUT (6.2 vs. 16.6 months; HR: 2.6), or KRAS MUT (5.7 vs. 12.7 months; HR: 1.7) were associated with inferior survival (Table 2). Among IDH2 mutated patients, survival was superior in patients with IDH2 MUT R172K versus R140Q mutations (not reached vs. 19.2 months; p = 0.02). OS was similar in patients achieving CR versus CRi (23 vs. 19 months; p = 0.44). In a separate analysis which excluded patients with prior HMA exposure, results were unchanged, except secondary AML did not impact OS (p = 0.21).
Multivariable analysis of clinical and genetic variables confirmed independent prognostic significance for male gender, secondary AML, adverse karyotype, KMT2Ar, TP53 MUT, KRAS MUT and IDH2 WT, with respective HRs (95% CI) of 2.1 (1.5–2.9), 1.6 (1.2–2.2), 1.9 (1.3–2.8), 7.9 (3.3–19.1), 1.9 (1.3–2.8), 2.2 (1.2–4.0), and 2.0 (1.1–3.4). Multivariable analysis limited to genetic variables only identified four independent risk factors for OS: HR (95% CI) were 1.8 (1.2–2.6) for adverse karyotype, 4.7 (2.0–11.1) for KMT2Ar, 1.7 (1.2–2.6) for TP53 MUT, 2.6 (1.4–4.7) for KRAS MUT, and 2.1 (1.2–3.6) for IDH2 WT. HR‐weighted scoring led to respective assignment of one adverse point each for ELN adverse karyotype, TP53 MUT, KRAS MUT, IDH2 WT, and two adverse points for KMT2Ar, which resulted in a three‐tiered risk stratification: low‐risk (0 points, n = 40), intermediate (1 point, n = 186), and high‐risk (≥ 2 points, n = 165). Figure 2 displays survival data using the Mayo genetic risk model with respective median survival (3‐year survival rate) of median not reached (67%), 19.1 months (30%), and 7.1 months (0%) (p‐values < 0.01 for all comparisons). Among seven patients with KMT2Ar, one patient was upgraded from intermediate to high‐risk and six from high to very‐high risk with median OS of 2.2 months for very‐high risk patients (≥ 4 points, n = 6) (Figure S2).
FIGURE 2.
Mayo genetic risk model versus European leukemia net (ELN) 2024 model in newly diagnosed acute myeloid leukemia treated with venetoclax and hypomethylating agent. [Color figure can be viewed at wileyonlinelibrary.com]
3.5. Overall Survival Analysis in Treatment Responders and Development of a Response‐Stratified Survival Model
Achievement of CR/CRi was the strongest predictor of survival with median (3‐year survival rate) of 3.6 versus 21 months (4% vs. 43%) and HR 5.4 in the absence versus presence of CR/CRi (Figure 3). Accordingly, a separate analysis of risk factors for survival was performed in treatment responders (Table S2). As expected, survival was superior in MRD negative versus positive cases (39 vs. 23 months; p = 0.09). In addition, in univariate analysis, IDH2 MUT was associated with superior OS (median not reached vs. 20.5 months). Male gender (median 17.6 vs. 29.2 months), TP53 MUT (median 11.9 vs. 24.3 months), KRAS MUT (median 11.4 vs. 22 months), adverse karyotype (median 12.6 vs. 29.4 months), and KMT2Ar (median 5.2 vs. 22 months) were associated with inferior OS. Multivariable analysis of genetic variables resulted in HRs (95% CI) of 2.3 (1.5–3.6) for adverse karyotype, 7.5 (1.7–32.7) for KMT2Ar, 3.4 (1.4–8.5) for KRAS MUT, and 2.8 (1.2–6.5) for IDH2 WT; TP53 MUT was no longer significant (p = 0.14). Accordingly, a response‐stratified Mayo genetic risk model was generated with assignment of one point each to adverse karyotype, KRAS MUT and IDH2 WT and two adverse points for KMT2Ar: high risk ≥ 2 points (n = 75; median 11.9 months; 3‐year survival, 0%); intermediate risk = 1 point (n = 136; median 24.3 months; 3‐year survival, 44%); and low‐risk = 0 points (n = 32; median not reached; 3‐year survival, 80%; p < 0.01) (Figure 4). 3‐year survival for high‐risk with or without ASCT was 42% versus 0% (p < 0.01); intermediate 72% versus 43% (p = 0.06); and low‐risk 88% versus 78% (p = 0.53).
FIGURE 3.
Overall survival in patients with newly diagnosed acute myeloid leukemia treated with venetoclax and hypomethylating agent, stratified by response. [Color figure can be viewed at wileyonlinelibrary.com]
FIGURE 4.
Mayo genetic risk model in newly diagnosed acute myeloid leukemia treated with venetoclax and hypomethylating agent and achieving complete remission with (CR) or without (CRi) count recovery. [Color figure can be viewed at wileyonlinelibrary.com]
3.6. Comparison of the Mayo Prognostic Model and ELN 2024 Genetic Risk Model
Figure 2 illustrates performance comparisons between the Mayo and ELN 2024 genetic risk models. Based on the ELN 2024 genetic risk model, low‐risk (TP53 WT, K/NRAS WT, FLT3‐ITD WT), intermediate (FLT3‐ITD MUT or K/NRAS MUT with TP53 WT) and high‐risk (TP53 MUT) groups had respective median survival (3‐year survival rate) of 19.2 (34%), 12.9 months (21%), and 6.2 months (0%); intermediate versus low risk (p = 0.01, HR: 1.5) and high versus intermediate (p < 0.01, HR: 1.9). On the other hand, the Mayo genetic risk model provided a more pronounced distinction between high versus intermediate (7.1 vs. 19.1 months; p < 0.01, HR 2.7) and intermediate versus low‐risk categories (19.1 months vs. not reached; p = 0.02; HR 2.2). In addition, comparison of AUC/AICc estimates for the Mayo and ELN 2024 models considering 3‐year survival probability demonstrated superior performance of the Mayo model (AUC 0.77, AICc 179) versus ELN 2024 model (AUC 0.69, AICc 195).
3.7. Validation of the Mayo Genetic Risk Model
The Mayo genetic risk model was validated by using an MD Anderson Cancer Center cohort (MDACC) of 117 ND‐AML patients (median age 73; years, 61% males) receiving Ven‐HMA; 53 (45%) of patients harbored ELN 2022 adverse karyotype; 3 (3%) with KMT2Ar; mutations involved TP53 (35%), NPM1 (22%), IDH2 (15%), RUNX1 (12%), IDH1 (10%), FLT3‐ITD (3%), DDX41 (3%). In this cohort, CR/CRi was achieved in 88 (75%) of patients (Table S4). Application of the Mayo genetic risk model to the MDACC cohort yielded median OS of 43.7 (0 points, n = 13), 19.9 months (1 point, n = 45), and 6.6 months (≥ 2 points, n = 59) (p < 0.01). In addition, implementation of the response‐stratified Mayo risk model which incorporated adverse karyotype, KMT2Ar, KRAS MUT, and IDH2 WT also disclosed similar results (Figure S3).
In a second external cohort of 117 ND‐AML patients (median age 74 years, 60% males) treated with Ven‐HMA at the University of Chicago (UOC), 45 of 113 (40%) of informative cases harbored adverse karyotype; 2 (2%) with KMT2Ar; mutations involved TP53 (30%), RUNX1 (22%), IDH1/2 (11%), NPM1 (10%), FLT3‐ITD (9%), DDX41 (3%). CR/CRi was documented in 61 (52%) of patients. In multivariable analysis which included relevant clinical and genetic variables, CR/CRi was the only independent predictor of OS (p < 0.01, HR: 5.6) (Table S5).
4. Discussion
Ven‐HMA is currently the standard‐of‐care for patients with ND‐AML who are elderly and/or unfit for intensive induction chemotherapy [3, 14]. AML is a genetically heterogeneous disease with highly variable patient outcomes following Ven‐HMA therapy [4, 5, 10]. Accordingly, prognostic information including estimation of the likelihood of response, disease relapse, and chances of survival are instrumental in the decision‐making process. While previous studies have investigated the prognostic relevance of genetic factors in the context of Ven‐HMA therapy [8, 11], none had accounted for response to treatment. The current study defines four distinct molecular signatures of treatment response to Ven‐HMA: (i) NPM1 MUT, IDH2 MUT, or DDX41 MUT with TP53 WT, RUNX1 WT, and FLT3‐ITD WT (CR/CRi, 87%), (ii) NPM1 WT, IDH2 WT, DDX41 WT, TP53 WT, RUNX1 WT, FLT3‐ITD WT (CR/CRi, 73%), (iii) NPM1 MUT, IDH2 MUT, or DDX41 MUT with TP53 MUT, RUNX1 MUT, or FLT3‐ITD MUT(CR/CRi, 63%), and (iv) TP53 MUT, RUNX1 MUT, or FLT3‐ITD MUT with NPM1 WT, IDH2 WT, or DDX41 WT (CR/CRi, 44%). By contrast, in a pooled analysis of 279 patients treated with Ven‐azacitidine in the phase 3 VIALE‐A (NCT02993523) and phase 1b study (NCT02203773), CR/CRi rates were 77.2% in TP53 WT, K/NRAS WT, and FLT3‐ITD WT (higher‐benefit group), 59.2% in the presence of FLT3‐ITD MUT or K/NRAS MUT with TP53 WT (intermediate‐benefit group) and 47.6% in the presence of TP53 MUT (lower‐benefit subgroup) [11]. When applied to the Mayo cohort, the VIALE‐A molecular subgroups resulted in respective CR/CRi rates of 73%, 52% and 46%, for higher, intermediate, and lower‐benefit groups, with marginal distinction between intermediate and lower‐benefit groups in terms of response. Interestingly, in the VIALE‐A molecular prognostic risk score (mPRS), NPM1 MUT and IDH2 MUT, which have previously shown superior response to Ven‐HMA [4, 10, 15], were not found to impact response in the context of TP53 MUT, K/NRAS MUT, FLT3‐ITD MUT, DDX41 MUT which has also been associated with favorable response to Ven‐HMA [12, 16], was not analyzed in the VIALE‐A mPRS.
TP53 mutations are known to cluster with adverse karyotype, which was the case in 93% of TP53 MUT patients in the current study; however, TP53 MUT did not appear to influence response rates, independent of karyotype. On the other hand, presence of adverse or complex/monosomal karyotype was independently associated with inferior response (CR/CRi 48% vs. 71% with or without adverse karyotype, and 50% versus 67% with or without complex/monosomal karyotype; p < 0.01). In a separate analysis of the VIALE‐A study, including patients with poor‐risk cytogenetics plus TP53 MUT, Ven‐azacitidine improved remission rates but not duration of remission or OS compared with azacitidine alone; moreover, remission and OS rates were higher in patients with poor‐risk cytogenetics and TP53 WT [17].
The vulnerability of IDH2 mutations to Ven‐HMA therapy makes a case for its upfront use; in a multicenter study of 151 patients ≥ 60 years with IDH1/2 mutated AML (90 IDH2 mutated), receiving Ven‐HMA or intensive chemotherapy, CR/CRi (67% vs. 67%) and overall survival rates (2 year survival; 49% vs. 38%) were similar after adjusting for baseline patient characteristics [18]. Unlike the VIALE‐A mPRS, the unfavorable prognosis typically associated with RAS mutations was limited to KRAS as opposed to NRAS mutations.
Genetic risk factors which influenced survival in the current study included ELN‐defined adverse karyotype, KMT2Ar, TP53 MUT, KRAS MUT, and IDH2 WT, which were accordingly incorporated into a pre‐treatment genetic risk model. Importantly, our study illustrates the prognostic value of CR/CRi which consistently overshadowed pre‐treatment risk variables in predicting survival in all three study cohorts in the current study (Mayo, MDACC, and UOC) and therefore allowed for a novel response‐stratified risk model inclusive of ELN‐defined adverse karyotype, KMT2Ar, KRAS MUT, and IDH2 WT. It is to be noted that the inclusion of response status in survival analysis resulted in the loss of prognostic impact from TP53 MUT. Both pre‐treatment and response‐stratified genetic models were validated by an external cohort and are amenable to further refinements with additional MRD information.
It is important to recognize the key differences between the Mayo prognostic model and the VIALE‐A mPRS. Unlike the case in our study, presence of IDH2 MUT was not found to be independently prognostic in the ELN model and was reportedly identified in all three risk categories; median OS for IDH2 MUT patients was 36.9 months versus 12.2 months in the higher and intermediate‐benefit groups, respectively [11]. Furthermore, in multivariable analysis of clinical and genetic variables in the VIALE‐A cohort, only genetic variables (TP53 MUT, KRAS MUT, NRAS MUT, and FLT3‐ITD MUT) were significant, whereas age ≥ 75 years, male gender, secondary AML, baseline blast counts, and ECOG performance status were not [11]. In the current study, the prognostic impact of adverse karyotype was independent of mutations; additionally, clinical variables of independent prognostic relevance included male gender and secondary AML. The association of male gender with inferior survival has been recently reported in older intensively treated patients with AML [19]. The above discrepancies between our findings and the VIALE‐A mPRS might have stemmed from differences in patient population with inclusion of cases with prior HMA exposure; additionally, all survival analyses in the current study accounted for ASCT.
Similar to the observation in the current study, a recent report by the BEAT‐AML investigators which included 595 patients with ND‐AML treated with a number of lower‐intensity therapies, including Ven‐HMA, identified IDH2 MUT as an independent favorable prognostic factor, and KRAS MUT, MLL2 (KMT2D)MUT, and TP53 MUT as unfavorable [20]. Based on these findings, a BEAT‐AML 2024 risk model was proposed with favorable, intermediate, and adverse‐risk groups with 2‐year OS: 48% versus 33% versus 11%, respectively, p < 0.01. Subsequently, this model was applied to patients receiving Ven‐HMA; risk group allocation included favorable (n = 49), intermediate (n = 83), and adverse‐risk (n = 70), with a significant difference between favorable and intermediate‐risk (p = 0.03); however, difference between intermediate and adverse risk groups was found to be marginal (OS and p‐value not provided) [20]. It is worthy to note that clinical variables, karyotype and ASCT, did not appear to be factored into the survival analysis [20].
The current study confirms that AML with KMT2Ar has adverse outcomes. In an MDACC study including 172 adult patients with KMT2Ar AML compared to 522 age‐matched AML patients with normal karyotype treated with high or low‐intensity chemotherapy regimens, the former had significantly inferior OS (median OS of 0.9 years vs. 2.1 years and 5‐year OS of 20% vs. 34%; p < 0.01), which was improved with ASCT (median 10.4 years and 5‐year OS 52%) [21].
Limitations of the current study include heterogeneous duration of venetoclax use and variability in timing of response assessment. The above‐discussed information underlines the challenges in prognostic factor assessments for Ven‐HMA‐treated ND‐AML. Methodological descriptions in some published studies were not always clear and established prognostic factors were not always considered in the statistical analysis, precluding comparison of study results. Nonetheless, the current study demonstrated superior model performance by the Mayo genetic risk model, compared to the recently published ELN 2024 risk model and also provides a novel response‐based risk stratification. The overarching value of such a model is consistent with established knowledge regarding the value of CR/CRi in predicting survival outcomes in ND‐AML treated with both intensive and less‐intensive therapies [10, 22, 23]. Going forward, incorporation of MRD information is likely to further enhance prognostic assessments [24, 25].
Author Contributions
N.G., A.T. designed the study, collected data, performed analysis and wrote the paper. A.E., F.A., I.M.J., MA collected data. K.M., A.A.‐K., H.B.A., K.H.B., M.E., A.M., A.M., A.N.S., M.H., M.R.L., W.H., M.S., M.M.P., A.P., T.B., H.M., J.A., J.P., L.S., N.K., C.A. contributed patients. S.Y., A.S., E.D., A.A.P. collected data and contributed patients. A.B. collected data, performed analysis and contributed patients. J.S., A.B., C.D., T.K. contributed patients.
Ethics Statement
IRB approval obtained.
Consent
Waived due to minimal risk research.
Conflicts of Interest
M.R.L: Research support from Abbvie, Astellas, Amgen, Actinium, Pluristem, Sanofi. Speaker's Bureau for Amgen, Beigene. Data Safety Monitoring Committee for Biosight. N.G.: Advisory Board for DISC medicine and Agios. M.S.: Research funding to the institution from Astellas, Celgene, and Marker Therapeutics. M.M.P.: Member of the board of directors or advisory committees of Stemline Therapeutics and received research funding from Kura Oncology. A.A.P.: Research funding from Pfizer, Kronos Bio, Sumitomo; honoraria from AbbVie, Sobi, Bristol Myers Squibb. ED: Honoraria from AbbVie. C.D.: Consultant/Advisory Boards: Abbvie, AstraZeneca, Astellas, BMS, Genentech, GenMab, GSK, Notable Labs, Rigel, Ryvu, Schrodinger, Servier. C.D.: is supported by the LLS Scholar in Clinical Research Award. T.K.: Grant support: BMS, Abbvie, Amgen, Ascentage Pharma Group, Astellas Pharma, DrenBio, Astex, AstraZeneca, BMS, Celgene, Incyte, Cellenkos, Cyclacel, Delta‐Fly pharma, Genentech, Genfleet, Glycomimetics, Iterion, Janssen, Jazz Pharmaceuticals, Pfizer, Pulmotect, Regeneron, SELLAS. Consulting fees: AbbVie, Agios, Daiichi Sankyo, Genentech, Genzyme, Jazz Pharmaceuticals, Liberum, Novartis, Pfizer, PinotBio, Pukmotect, Sanofi‐Aventis, Servier. Payment or honoraria: AbbVie, Agios, Daiichi Sankyo, DAVA Oncology, Delta‐Fly, DrenBio, Genentech, Genfleet, Genzyme, Jazz Pharmaceuticals, Liberum, Novartis, Pfizer, Rigel, Sanofi‐Aventis, SELLAS, Servier. The remaining authors declare no competing financial interests.
Supporting information
Data S1: Supporting Information.
Funding: The authors received no specific funding for this work.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. DiNardo C. D., Pratz K. W., Letai A., et al., “Safety and Preliminary Efficacy of Venetoclax With Decitabine or Azacitidine in Elderly Patients With Previously Untreated Acute Myeloid Leukaemia: A Non‐randomised, Open‐Label, Phase 1b Study,” Lancet Oncology 19, no. 2 (2018): 216–228. [DOI] [PubMed] [Google Scholar]
- 2. DiNardo C. D., Pratz K., Pullarkat V., et al., “Venetoclax Combined With Decitabine or Azacitidine in Treatment‐Naive, Elderly Patients With Acute Myeloid Leukemia,” Blood 133, no. 1 (2019): 7–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. DiNardo C. D., Jonas B. A., Pullarkat V., et al., “Azacitidine and Venetoclax in Previously Untreated Acute Myeloid Leukemia,” New England Journal of Medicine 383, no. 7 (2020): 617–629. [DOI] [PubMed] [Google Scholar]
- 4. DiNardo C. D., Tiong I. S., Quaglieri A., et al., “Molecular Patterns of Response and Treatment Failure After Frontline Venetoclax Combinations in Older Patients With AML,” Blood 135, no. 11 (2020): 791–803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Gangat N., Johnson I., McCullough K., et al., “Molecular Predictors of Response to Venetoclax Plus Hypomethylating Agent in Treatment‐Naive Acute Myeloid Leukemia,” Haematologica 107, no. 10 (2022): 2501–2505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Pratz K. W., Jonas B. A., Pullarkat V., et al., “Long‐Term Follow‐Up of VIALE‐A: Venetoclax and Azacitidine in Chemotherapy‐Ineligible Untreated Acute Myeloid Leukemia,” American Journal of Hematology 99, no. 4 (2024): 615–624. [DOI] [PubMed] [Google Scholar]
- 7. Pratz K. W., Jonas B. A., Pullarkat V., et al., “Measurable Residual Disease Response and Prognosis in Treatment‐Naïve Acute Myeloid Leukemia With Venetoclax and Azacitidine,” Journal of Clinical Oncology 40, no. 8 (2022): 855–865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Bataller A., Bazinet A., DiNardo C. D., et al., “Prognostic Risk Signature in Patients With Acute Myeloid Leukemia Treated With Hypomethylating Agents and Venetoclax,” Blood Advances 8, no. 4 (2024): 927–935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Gangat N., Johnson I., McCullough K., et al., “Molecular Predictors of Response to Venetoclax Plus Hypomethylating Agent in Treatment‐Naïve Acute Myeloid Leukemia,” Haematologica 107, no. 10 (2022): 2501–2505. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Gangat N., Karrar O., Iftikhar M., et al., “Venetoclax and Hypomethylating Agent Combination Therapy in Newly Diagnosed Acute Myeloid Leukemia: Genotype Signatures for Response and Survival Among 301 Consecutive Patients,” American Journal of Hematology 99, no. 2 (2024): 193–202. [DOI] [PubMed] [Google Scholar]
- 11. Döhner H., Pratz K. W., DiNardo C. D., et al., “Genetic Risk Stratification and Outcomes Among Treatment‐Naive Patients With AML Treated With Venetoclax and Azacitidine,” Blood 144 (2024): 2211–2222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Bataller A., Loghavi S., Gerstein Y., et al., “Characteristics and Clinical Outcomes of Patients With Myeloid Malignancies and DDX41 Variants,” American Journal of Hematology 98, no. 11 (2023): 1780–1790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Döhner H., Wei A. H., Appelbaum F. R., et al., “Diagnosis and Management of AML in Adults: 2022 Recommendations From an International Expert Panel on Behalf of the ELN,” Blood 140, no. 12 (2022): 1345–1377. [DOI] [PubMed] [Google Scholar]
- 14. Pratz K. W., Panayiotidis P., Recher C., et al., “Venetoclax Combinations Delay the Time to Deterioration of HRQoL in Unfit Patients With Acute Myeloid Leukemia,” Blood Cancer Journal 12, no. 4 (2022): 71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Pollyea D. A., DiNardo C. D., Arellano M. L., et al., “Impact of Venetoclax and Azacitidine in Treatment‐Naive Patients With Acute Myeloid Leukemia and IDH1/2 Mutations,” Clinical Cancer Research 28 (2022): 2753–2761. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Nanaa A., He R., Foran J. M., et al., “Venetoclax Plus Hypomethylating Agents in DDX41‐Mutated Acute Myeloid Leukaemia and Myelodysplastic Syndrome: Mayo Clinic Series on 12 Patients,” British Journal of Haematology 204, no. 1 (2024): 171–176. [DOI] [PubMed] [Google Scholar]
- 17. Pollyea D. A., Pratz K. W., Wei A. H., et al., “Outcomes in Patients With Poor‐Risk Cytogenetics With or Without TP53 Mutations Treated With Venetoclax and Azacitidine,” Clinical Cancer Research 28, no. 24 (2022): 5272–5279. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Bewersdorf J. P., Shimony S., Shallis R. M., et al., “Combination Therapy With Hypomethylating Agents and Venetoclax Versus Intensive Induction Chemotherapy in IDH1‐ Or IDH2‐Mutant Newly Diagnosed Acute Myeloid Leukemia—A Multicenter Cohort Study,” American Journal of Hematology 99, no. 8 (2024): 1640–1643. [DOI] [PubMed] [Google Scholar]
- 19. Versluis J., Metzner M., Wang A., et al., “Risk Stratification in Older Intensively Treated Patients With AML,” Journal of Clinical Oncology 43, no. 34 (2024): 4084–4094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Hoff F. W., Blum W., Huang Y., et al., “Beat‐AML 2024 ELN‐Refined Risk Stratification for Older Adults With Newly Diagnosed AML Given Lower‐Intensity Therapy,” Blood Advances 8 (2024): 5297–5305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Issa G. C., Zarka J., Sasaki K., et al., “Predictors of Outcomes in Adults With Acute Myeloid Leukemia and KMT2A Rearrangements,” Blood Cancer Journal 11, no. 9 (2021): 162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Tefferi A., Gangat N., Al‐Kali A., et al., “A Dynamic 3‐Factor Survival Model for Acute Myeloid Leukemia That Accounts for Response to Induction Chemotherapy,” American Journal of Hematology 97, no. 9 (2022): 1127–1134. [DOI] [PubMed] [Google Scholar]
- 23. Norsworthy K. J., Gao X., Ko C.‐W., et al., “Response Rate, Event‐Free Survival, and Overall Survival in Newly Diagnosed Acute Myeloid Leukemia: US Food and Drug Administration Trial‐Level and Patient‐Level Analyses,” Journal of Clinical Oncology 40, no. 8 (2022): 847–854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Hourigan C. S., Dillon L. W., Gui G., et al., “Pre‐MEASURE: Multicenter Evaluation of the Prognostic Significance of Measurable Residual Disease Testing Prior to Allogeneic Transplantation for Adult Patients With AML in First Remission,” Journal of Clinical Oncology 40, no. S16 (2022): 7006. [Google Scholar]
- 25. Heuser M., Freeman S. D., Ossenkoppele G. J., et al., “2021 Update on MRD in Acute Myeloid Leukemia: A Consensus Document From the European LeukemiaNet MRD Working Party,” Blood 138, no. 26 (2021): 2753–2767. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: Supporting Information.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.