Patients

Overall, 314 patients with newly diagnosed AML were included in the analysis. The median age was 74 years (range 25–90) and 111 (35%) had antecedent myeloid neoplasm. Most patients (75%) were classified as adverse risk by ELN 2022. In total,166 (53%) were treated with HMA + VEN, and 148 (47%) were treated with HMA (127 [40%] with HMA monotherapy and 21 [7%] with HMA + non-VEN drug), Table 1. There were no significant differences in baseline characteristics between patients treated with HMA vs. HMA + VEN. When classified by ontogeny, 111 (35%) were in the TP53 group (60 with HMA + VEN, 51 with HMA); 115 (37%) in the secondary ontogeny group (68 with HMA + VEN, 47 with HMA); and 88 (28%) in the de novo ontogeny group (38 with HMA + VEN, 50 with HMA, Fig. 1 and Supplementary Fig. 1B). Decitabine was more commonly used compared with azacytidine (n = 208 [66%] vs. n = 106 [34%)], with five days of decitabine being the most common HMA regimen used (n = 166 [53%], Supplementary Table 1).

Table 1 Patient characteristics by treatment.Fig. 1: Cytogenetic and molecular plot by groups.Survival outcomesSurvival in the entire group

After a median follow-up of 29 months (95% CI 24–32), the OS was better in patients treated with HMA + VEN compared with those who received HMA without VEN (median OS 9.9 months [95% CI 7.9–13.4] vs. 7.4 months [95% CI 6.1–10.3], respectively, P = 0.018, Fig. 2A). The results were similar in a sensitivity analysis excluding patients who received HMA+ an additional non-VEN drug (n = 21, Supplementary Fig. 2). In a multivariable analysis, only TP53 ontogeny (HR 1.88 [95% CI 1.37–2.59]) and consolidation with alloHCT (HR 0.23 [95% CI 0.14–0.40]) were independently associated with OS (Supplementary Table 2).

Fig. 2: Overall survival of the entire cohort and in each ontogeny.

A All patients. B TP53 ontogeny group. C Secondary ontogeny group. D De novo ontogeny group. OS overall survival, CI confidence interval, HMA hypomethylating agents, VEN venetoclax.

Survival in each ontogeny group

To define the ontogeny-specific activity of VEN, we performed a separate analysis in each molecularly defined ontogeny group. Among patients with TP53 mutated disease, there was no difference in OS between HMA + VEN vs. HMA treatment groups (5.7 months [95% CI 4–7.6] vs. 6.1 months [95% CI 4.1–10.3], P = 0.93), Fig. 2B. Furthermore, OS was comparable in both treatment groups among patients with TP53 mutations irrespective of VAF cutoff, including 10%, 20%, and 50% (Supplementary Table 3). In the de novo group, OS was not different between HMA + VEN vs. HMA (13.2 months [95% CI 8.3–22.7] vs. 10.3 months [95% CI 6.3–13.9], P = 0.12, Fig. 2C). Conversely, in the secondary group, patients treated with HMA + VEN had significantly better OS compared to patients treated with HMA (14.1 months [9.9–20.0] vs. 6.9 months [95% CI 4.7–11.3], P = 0.0054, Fig. 2D). The results were retained in a sensitivity analysis when RUNX1 was considered as a secondary ontogeny-defining mutation (Supplementary Fig. 3).

To identify ontogeny-specific prognostic variables, we conducted Cox multivariable regression models within each ontogeny group separately. Across all groups, alloHCT was associated with improved OS (de novo: HR 0.14 [95% CI 0.03–0.58], P = 0.0067; secondary: HR 0.40 [95% CI 0.18–0.89], P = 0.026; TP53: HR 0.17 [95% CI 0.07–0.42], p < 0.001). Conversely, treatment modality (HMA + VEN vs. HMA) was associated with improved OS only in the secondary group (de novo: HR 0.67 [95% CI 0.40–1.12], P = 0.13; secondary: HR 0.59 [95% CI 0.38–0.94], P = 0.025; TP53: HR 1.19 [95% CI 0.78–1.81], P = 0.42).

Sensitivity analysis without TP53 group

Since TP53 mutations were strongly associated with worse outcomes and were highly prevalent in the cohort, their inclusion in the analysis could obscure a meaningful effect of initial treatment on survival in patients without TP53 mutations. We conducted a sensitivity analysis and regression modeling for OS in patients without TP53 mutations (n = 203). Among these patients, the median OS was 14.0 months (95% CI 10.6–17.8) with HMA + VEN vs. 8.7 months (95% CI 6.3–11.4) with HMA, P = 0.0028 (Supplementary Fig. 4). In the multivariable analysis, both alloHCT (HR 0.27 [95% CI 0.13–0.53, p < 0.001]) and treatment with HMA + VEN (vs. HMA) (HR 0.66 [95% CI 0.47–0.92], P = 0.015) were associated with improved OS (Supplementary Table 4).

Survival in patients treated with HMA + VEN

In patients treated with HMA + VEN the OS was similar between de novo and secondary groups (median OS 13.2 months [95% CI 8.3–22.7] vs. 14.1 months [95% CI 9.9–20.0], respectively, P = 0.92); and each group had better OS compared to TP53 group (median 5.7 months [95% CI 4.0–7.6], comparison vs. de novo P = 0.007; comparison vs. secondary p < 0.001), Fig. 3. In the multivariable analysis of patients treated with HMA + VEN, the comparable OS between secondary and de novo groups was retained (HR 1.07 [95% CI 0.63–1.84, P = 0.79]), as well as the worse survival in the TP53 group (compared with de novo, HR 2.57 [95% CI 1.53–4.33, P < 0.001]) and better survival with alloHCT as a time-varying covariate (HR0.22 [95% CI 0.11–0.46], P < 0.001, Table 2).

Fig. 3: Overall survival by ontogeny in patients treated with HMA + VEN.

CI confidence interval, HMA hypomethylating agents, VEN venetoclax.

Table 2 Univariable and multivariable Cox regression overall OS of patients treated with HMA + VEN.Response and transplant rates

To determine whether the distinct survival patterns in each ontogeny were associated with other measures of clinical response, we evaluated cCR and alloHCT rates in the entire cohort and within each ontogeny group separately. Overall, the cCR rate was better in patients treated with HMA + VEN than in those who received HMA (49% vs. 28%, P = 0.001). When analyzing molecularly defined ontogeny groups separately, the cCR rates were higher in patients treated with HMA + VEN vs. HMA in the de novo group (54% vs. 29%, P = 0.034]) and secondary group (61% vs.18%, p < 0.001), but not different in patients with TP53 mutations (33% vs. 37%, P = 0.82, Table 3).

Table 3 Best response and alloHCT by ontogeny.

As long-term survival among patients with high-risk AML subtypes depends on alloHCT [12], we evaluated whether HMA + VEN may be associated with higher alloHCT rates in the entire cohort and in each ontogeny group. Overall, 41 (13%) patients were consolidated with alloHCT, with patients treated with HMA + VEN more frequently transplanted (17%, n = 29) than those treated with HMA (8%, n = 12), P = 0.018. However, when analyzed within each ontogeny group, the rates of alloHCT were similar within the TP53 group (15% [n = 9] vs. 8% [n = 4], P = 0.38) and in the de novo group (11% [n = 4] vs.10% [n = 5], p > 0.99). Conversely, the rates were higher among patients in the secondary group treated with HMA + VEN (24%, n = 16) vs. HMA (6%, n = 3), P = 0.02.

Subgroup analysis of secondary ontogeny

Mutations that affect RNA splicing (SF3B1, SRSF2, U2AF1, ZRSR2) have been reported to be associated with improved survival in patients treated with VEN-based regimens[13]. To determine whether the aforementioned response and survival benefit with HMA + VEN vs. HMA in the secondary ontogeny group is attributable solely to splicing abnormalities in these mutations, we compared patients with (n = 86) vs. without (n = 29) these mutations within the secondary group. First, splicing vs. non-splicing subgroups were comparable in age, prior therapy, or myeloid malignancy rates and ELN2022 risk criteria (Supplementary Table 5). The splicing vs. non-splicing subgroup had higher rates of NPM1 (16% vs. 0%, respectively P = 0.02) and lower rates of ASXL1 (42% vs. 79%, respectively, P < 0.001). The cCR rates between splicing and non-splicing mutation subgroups were comparable among patients treated with HMA (22 vs. 0%, P = 0.3) or HMA + VEN (61% VS. 60%, P > 0.99). Similarly, OS was comparable between splicing and non-splicing subgroups treated with HMA (median OS 7.8 months [95% CI 4.2–12.1] vs. 6.5 months [95% CI 1.9-11.3], P = 0.78) or with HMA + VEN (median OS 14.9 months [95% CI 10.6–20.1] vs. 10.8 months [95% CI 6.1-NR], P = 0.92), Fig. 4.

Fig. 4: Overall survival in secondary ontogeny by treatment and splicing mutations.

CI confidence interval, HMA hypomethylating agents, VEN venetoclax.