PURPOSE: Patients with acute myeloid leukemia with high-risk cytogenetics in first complete remission (CR1) achieve better outcomes if they undergo allogeneic hematopoietic cell transplantation (HCT) compared with consolidation chemotherapy alone. However, only approximately 40% of such patients typically proceed to HCT. METHODS: We used a prospective organized approach to rapidly identify donors to improve the allogeneic HCT rate in adults with high-risk acute myeloid leukemia in CR1. Newly diagnosed patients had cytogenetics obtained at enrollment, and those with high-risk cytogenetics underwent expedited HLA typing and were encouraged to be referred for consultation with a transplantation team with the goal of conducting an allogeneic HCT in CR1. RESULTS: Of 738 eligible patients (median age, 49 years; range, 18-60 years of age), 159 (22%) had high-risk cytogenetics and 107 of these patients (67%) achieved CR1. Seventy (65%) of the high-risk patients underwent transplantation in CR1 ( P < .001 compared with the historical rate of 40%). Median time to HCT from CR1 was 77 days (range, 20-356 days). In landmark analysis, overall survival (OS) among patients who underwent transplantation was significantly better compared with that of patients who did not undergo transplantation (2-year OS, 48% v 35%, respectively [ P = .031]). Median relapse-free survival after transplantation in the high-risk cohort who underwent transplantation in CR1 (n = 70) was 11.5 months (range, 4-47 months), and median OS after transplantation was 14 months (range, 4-44 months). CONCLUSION: Early cytogenetic testing with an organized effort to identify a suitable allogeneic HCT donor led to a CR1 transplantation rate of 65% in the high-risk group, which, in turn, led to an improvement in OS when compared with the OS of patients who did not undergo transplantation.

Abstract Measurable residual disease (MRD) by multiparametric flow cytometry (MFC) before allogeneic hematopoietic cell transplantation (HCT) identifies patients at high risk of acute myeloid leukemia (AML) relapse, often occurring early after allografting. To examine the role of MFC MRD testing to predict later relapses, we examined 935 adults with AML or myelodysplastic neoplasm/AML transplanted in first or second morphologic remission who underwent bone marrow restaging studies between day 70 and 100 after HCT and were alive and without relapse by day +100. Of 935 adults, 136 (15%) had MRD before HCT, whereas only 11 (1%) had MRD at day +70 to +100. In day +100 landmark analyses, pre-HCT and day +70 to +100 MFC MRD were both associated with relapse (both P < .001), relapse-free survival (RFS; both P < .001) overall survival (OS; both P < .001), and, for post-HCT MRD, nonrelapse mortality (P = .001) after multivariable adjustment. Importantly, although 126/136 patients (92%) with MRD before HCT tested negative for MRD at day +70 to +100, their outcomes were inferior to those without MRD before HCT and at day +70 to +100, with 3-year relapse risk of 40% vs 15% (P < .001), 3-year RFS of 50% vs 72% (P < .001), and 3-year OS of 56% vs 76% (P < .001), whereas 3-year nonrelapse mortality estimates were similar (P = .53). Thus, despite high MRD conversion rates, outcomes MRD positive/MRD negative (MRDneg) patients are inferior to those of MRDneg/MRDneg patients, suggesting all patients with pre-HCT MRD should be considered for preemptive therapies after allografting.

The current European Leukemia Net guidelines (ELN22) classify AML patients with intermediate-risk cytogenetics and MDS-associated mutations (MDSm, mutations in ASXL1, BCOR, EZH2, RUNX1, SF3B1, SRSF2, STAG2, U2AF1, ZRSR2) as adverse-risk but do not use MDSm+ to alter the risk stratification of patients who otherwise classify as favorable-risk. NPM1 mutations (NPM1+) are one of the most common alterations in AML and remain a favorable-risk biomarker for patients without co-occurring FLT3-ITD mutations or adverse-risk cytogenetics (NPM1+/ITD-/ARcyto-). However, prior studies have shown that older patients (≥65 years) with favorable-risk AML, including NPM1+/ITD-/ARcyto- genotype, have significantly higher relapse rates and decreased survival despite receiving comparable intensive chemotherapy regimens to their younger counterparts. Given that MDSm frequency increases with age in patients with NPM1+ AML, it raises the question as to whether MDSm contribute to poorer outcomes in older patients. To date, there have been differing results in retrospective studies examining the effect of MDSm on NPM1+ AML. Some cohorts demonstrate worse outcomes in those that have NPM1+/MDSm+ AML (Chan et al., Blood Adv. 2024), whereas others do not show a significant difference in NPM1+ AML with or without MDSm (Eckardt et al., Leukemia 2023; Othman et al., Blood 2024; Ruhnke et al., Blood Adv. 2025). We examined the mutational landscape in a large group of pre-treatment, diagnostic specimens from 271 patients, which included 119 patients enrolled in SWOG clinical trials (SWOG-9031, SWOG-9333, S0106, and S0112), 96 patients from the UW/FHCC Hematopoietic Repository, and 56 patients from the BeatAML cohort. All patients were treated with curative intent as per standard practice or on clinical trials and had adequate cytogenetic information and long-term follow-up data. Event-free and overall survival (EFS, OS) were estimated using the Kaplan-Meier method and compared using Cox regression models stratified by cohort. MDSm+ occurred in 17% of patients, with the most common mutations being in SRSF2 (7%) and SF3B1 (3%). To assess the impact of MDSm on outcomes of favorable-risk NPM1+ AML, we performed univariate analyses comparing OS and EFS for NPM1+ AML grouped by ELN22 favorable risk/MDSm-, ELN22 favorable/MDSm+, and ELN22 intermediate/adverse risk. Compared to NPM1+/MDSm- (n=116), favorable-risk NPM1+/MDSm+ (n=33) had worse OS with HR 2.0 [95% CI (1.1, 3.3), p=0.008] and had similar OS to those that are classified as ELN22 intermediate/adverse risk (n=122). There was no significant difference in EFS or complete remission rates between NPM1+ favorable risk with and without MDSm (p=0.36 and p=0.6, respectively). When we grouped the ELN22 favorable risk patients by age (<65 years vs. 65+ years), MDSm+ in younger patients did not show a difference for OS (HR 0.99, p=0.98). Older ELN22 favorable-risk patients (≥65) had poor OS regardless of presence of MDSm compared to younger MDSm- patients: older MDSm- (n=91) with HR 2.4, 95% CI (1.3, 4.6), p=0.005 and older MDSm+ (n=25) with HR 3.5, 95% CI (1.9, 6.3), p<0.0001. There was a trend toward worse OS in MDSm+ compared to MDSm- older patients, but this was not statistically significant (HR 1.42, p=0.33). Since most MDSm are in splicing factor genes (SF3B1, SRSF2, U2AF1, ZRSR2), we examined the effect of splicing factor mutations vs. other MDSm+ in ELN22 favorable risk and did not find a significant difference in OS (HR=1.01, p=0.97). Those with non-splicing MDSm+ had worse EFS compared to MDSm- and splicing MDSm+ (HR 4.12, p=0.0037), although no difference in OS. In this study, MDSm+ in the context of favorable-risk NPM1+ AML are associated with worse OS in a univariate analysis. However, this difference is likely due to the increased prevalence of MDSm with increasing age and overall worse outcomes of NPM1+ AML in older patients. When stratified by age, older NPM1+ AML patients had worse OS than younger NPM1+ patients overall, although presence of MDSm trended toward even worse OS in older patients. In younger patients, MDSm+ are not associated with a significant difference in OS, which is a similar finding compared to prior studies. Our findings suggest that NPM1+ AML is not necessarily “favorable-risk” in older patients, particularly with concurrent MDSm. Therefore, further understanding of the underlying biology of age in the context of NPM1+ AML is needed to improve outcomes for older patients.

Introduction: We have previously shown that age, comorbidity burden, cytogenetic risk, and geriatric health significantly impact the likelihood of receiving allo-HCT and mortality among AML patients (pts). Building on these findings, we now examine the role of socioeconomic status (SES) as determined by the areas where participants lived within the same study cohort, focusing on the receipt of allo-HCT and post-transplant mortality. Methods: Here we report results of a prospective, multi-center observational study across 13 predominantly academic centers enrolling adult AML pts. We examined the following SES factors as estimated by the zip codes in which pts lived: median household income; % of adults aged ≥25 years with < a high school education or equivalent (i.e., General Education Diploma); % of households below the poverty level; % of households receiving Supplemental Nutrition Assistance Program (SNAP)/food stamps; % of occupied housing units (rented and owner-occupied) where 30% or more of income is spent on housing; % of owner-occupied housing units with a mortgage that have either a second mortgage, home equity loan or both; % of households receiving Supplemental Security Income (SSI, disability); and % of people working for pay who drive to work alone. Fine-Gray regression models were used to evaluate factors influencing the receipt of HCT, where death without HCT was treated as a competing risk. For overall mortality after HCT, Cox regression with left truncation at HCT was employed. All SES factors were modeled as continuous linear variables, with hazard ratio (HR) presented in terms of an increase in income of $25,000 and an increase in % of other factors of 10%. Additionally, each SES model was adjusted for age, Hematopoietic Cell Transplant-specific Comorbidity Index, disease status, European LeukemiaNet risk, Karnofsky Performance Scale, Patient Health Questionnaire-9, Activities of Daily Living, 4-Meter Walk Test, and Functional Assessment of Cancer Therapy. Results: In our analysis of 695 AML pts, a multivariable analysis showed a decrease in the likelihood of HCT receipt by 32% as the percentage of residents with less than a high school education in a neighborhood increased by 10% (HR: 0.68; 95% CI, 0.55-0.84, p=0.0005). We also observed a 14% decrease in the likelihood to receive allo-HCT for each 10% increase in households receiving SNAP compared to baseline levels (HR=0.86; 95% CI, 0.74-1.00, p=0.04). On the other hand, increase in the likelihood of HCT receipt was modest (5%) as median area income increased by $25,000 compared to areas with baseline median income (HR: 1.05; 95% CI, 0.94-1.18, p=0.39). There was also a small decrease (15%) in the likelihood of HCT receipt for each 10% increase in percentage of households below the poverty level (HR=0.85; 95% CI, 0.70-1.03, p=0.10) but a larger decrease (34%) for each 10% increase in households receiving SSI (HR=0.66; 95% CI, 0.39-1.11, p=0.11). There was little evidence of an association between median area income and overall mortality after HCT (HR: 0.97; 95% CI, 0.57-1.66, p=0.92). There was a numerical increased hazard of mortality after HCT as the percentage of residents with less than a high school education in a neighborhood increased by 10%, but the increase was modest (HR: 1.14; 95% CI, 0.66-1.98, p=0.64), little evidence of an increase or decrease in mortality as the percentage of households below the poverty level increased by 10% (HR=0.98; 95% CI, 0.61-1.57, p=0.92), and numerical (but modest) increases in mortality as the percentage of households receiving SNAP increased by 10% (HR=1.10; 95% CI, 0.77-1.57, p=0.60), and as the percentage of households receiving SSI increased by 10% (HR=1.23; 95% CI, 0.73-1.29, p=0.40). Conclusion: These findings underscore potential impact of socioeconomic disparities on receipt of HCT in AML pts. Lower educational attainment could be a barrier to receiving HCT. This suggests the primary issue is access to HCT, as SES does not seem to definitively affect post-transplant outcomes among those who receive HCT. This highlights the need for targeted interventions to improve access to HCT for pts from lower socioeconomic backgrounds. Efforts should focus on addressing financial barriers, improving health literacy, and enhancing support systems to ensure equitable access to life-saving treatments. These are some of the key focuses of the ACCESS initiative by the ASTCT and NMDP (Auletta, TCT 2022).

Introduction: It is increasingly recognized that racial and socioeconomic disparities play an important prognostic role and contribute to differences in outcomes in adults receiving chemotherapy for acute myeloid leukemia (AML). Structural racism has also been implicated in limiting access to allogeneic hematopoietic cell transplantation (HCT). However, it is unclear to what degree racial and socioeconomic disparities impact post-HCT outcomes of patients with AML in contemporary practice. To address this uncertainty, we investigated the impact of socioeconomic, racial, and non-biologic disparities on outcomes in a large cohort of adults with AML undergoing allogeneic HCT while in first or second morphologic remission. Methods: We identified 1,004 adults with AML (2022 ICC criteria) who underwent allogeneic HCT between 5/2006 and 10/2021 at a single large university-affiliated cancer center. Data was collected on each patient's individual socioeconomic status (SES), including occupation, insurance type, and education attainment. Distance traveled, marital status, household size, country of birth, and religious affiliation were also recorded. Each occupation was classified by the International Socio-Economic Index (ISEI), a standard index of occupational status. Area-level SES was analyzed through the Area Deprivation Index (ADI), a validated measure of socioeconomic disadvantage which corresponds to U.S. census tract measurements. Collected HCT variables include the AML treatment-related mortality score (TRM), pre-HCT flow cytometric MRD status, HLA donor status, cytogenetic risk, conditioning regimen intensity, and incidence of severe acute and chronic graft-vs-host disease. Measured outcomes include non-relapse mortality (NRM), overall survival (OS), relapse, and relapse free survival (RFS). Cause-specific hazard models were used for univariate and multivariable regression analysis and C-statistics were calculated to quantify each model's ability to predict outcomes. Results: In univariate analysis, none of the socioeconomic or non-biologic factors were significantly associated with NRM, and all had poor predictive performance, with C-statistics not exceeding 0.53 for: marital status, larger household size (i.e., patients with children), distance traveled, ISEI, ADI, insurance status, educational attainment, or country of birth. TRM score (P<0.001), non-myeloablative conditioning (P<0.001), reduced intensity conditioning (P<0.001), CR2 (P=0.017) and marital status (P=0.014) were significantly associated with NRM in a multivariable model, but predictive performance of this model was only fair (C=0.69; Table 1). Marital status was only associated with NRM after multivariable adjustment but not on univariate evaluation, possibly because married patients were, on average, older (P<0.001), more likely to be male (P<0.001), and have higher occupational status/ISEI (P<0.001) and TRM scores (P<0.001), with age and sex as likely confounders (Table 2). African American patients (n=14) were found to have worse NRM in both univariate (P=0.036) and multivariate (P=0.024) models, consistent with previously identified racial/ethnic disparities in AML cohorts. Similar results were found for SES and non-biologic variables in regards to OS, relapse, and RFS. Notable exceptions after multivariable adjustment include: ADI was significantly associated with OS (P=0.036) and RFS (P=0.044) and marital status was not significantly associated with OS or relapse. Additional analyses demonstrated that ISEI and ADI were not associated with increased TRM or HLA donor status. Finally, distance traveled decreased in HCT recipients post-2013 (P<0.003), though the number of AML-related transplants remained unchanged. Conclusion: In this retrospective, single center analysis, socioeconomic and non-biologic factors had very limited impact on outcomes following allografting for adults with AML. While reassuring, patients able to undergo allogeneic HCT are highly selected (e.g. need for caregiver, insurance, etc.) possibly accounting for this finding, as could integrated, multidisciplinary cancer care networks and longitudinal care coordination post-HCT. Further research is needed to investigate disparities among HCT-eligible patients and improve access to HCT with a focus on reducing racial/ethnic disparities.

Chromosome analysis is essential for diagnosing genetic disorders. For hematologic malignancies, identification of somatic clonal aberrations by karyotype analysis remains the standard of care. However, karyotyping is costly and time-consuming because of the largely manual process and the expertise required in identifying and annotating aberrations. Efforts to automate karyotype analysis to date fell short in aberration detection. Using a training set of ~10k patient specimens and ~50k karyograms from over 5 years from the Fred Hutchinson Cancer Center, we created a labeled set of images representing individual chromosomes. These individual chromosomes were used to train and assess deep learning models for classifying the 24 human chromosomes and identifying chromosomal aberrations. The top-accuracy models utilized the recently introduced Topological Vision Transformers (TopViTs) with 2-level-block-Toeplitz masking, to incorporate structural inductive bias. TopViT outperformed CNN (Inception) models with >99.3% accuracy for chromosome identification, and exhibited accuracies >99% for aberration detection in most aberrations. Notably, we were able to show high-quality performance even in "few shot" learning scenarios. Incorporating the definition of clonality substantially improved both precision and recall (sensitivity). When applied to "zero shot" scenarios, the model captured aberrations without training, with perfect precision at >50% recall. Together these results show that modern deep learning models can approach expert-level performance for chromosome aberration detection. To our knowledge, this is the first study demonstrating the downstream effectiveness of TopViTs. These results open up exciting opportunities for not only expediting patient results but providing a scalable technology for early screening of low-abundance chromosomal lesions.

Background: C10603/RATFIY was the first AML trial to show the benefit of adding a targeted agent to intensive chemotherapy for a specific genetically determined subset (Stone R et al, NEJM 2017). The addition of the multi-kinase inhibitor midostaurin (M) to chemotherapy in previously untreated adults with FLT3-mutant AML resulted in superior event-free (EFS) and overall survival (OS) compared to placebo (P) and led to the approval of this agent in combination therapy. Subsequently, additional targeted drugs including gilteritinib for relapsed/refractory FLT3-mutant AML and quizartinib plus chemotherapy in untreated adults with AML with FLT3-ITD disease have gained approval. Long-term follow-up after reporting of the primary endpoint, especially if met, is important in oncology. Objective: We report the EFS and OS data from C10603 using 10 years of follow up to determine the persistence of a midostaurin benefit and to assess the nature of late relapses or toxicity. Methods: C10603 enrolled 717 pts (360 on the M and 357 on the P arm; median age 47.8 years (range 18-61), 398 women (55.5%) of whom 51.7% were randomized to M and 59.4% to P (p=0.04), and 89% white) from 2011-2015 with previously untreated AML who had either a FLT3-TKD or ITD mutation with allelic ratio of >0.05 to receive daunorubicin/cytarabine (3+7) induction (one reinduction permitted) followed by up to 4 consolidation cycles with cytarabine 3 g/m2 every 12h on days 1, 3 and 5. M (50 mg bid) or P was given orally on days 8-22 of each chemotherapy cycle as well as daily during one year of single-agent maintenance. Allogeneic transplantation (alloHCT) was employed at investigator's discretion; study drug was not continued post-HCT. Pts stayed on their randomized arm and were not re-randomized prior to maintenance. Pts were followed for relapse and survival after the end of protocol therapy, regardless of subsequent therapy, for 10 years. Log-rank tests, Kaplan-Meier methods, and Cox proportional hazard models were used to evaluate the treatment effect in EFS and OS, as well as the alloHCT effect. All p values are two-sided. Results: We have completed the 10-year FU on this trial and data collection has concluded. The median EFS was 8.2 months (95% CI, 5.5-11.4) on the M arm compared with 3.0 months for P (95% CI, 1.9-6.0) with a HR of 0.79 (95% CI 0.67-0.94, p=0.0067). Among the 420 CR pts who achieved complete remission (CR) by the protocol specified time of 60 days, 253 have had an event (death in 69 and relapse in 184); these events were relatively equal on each arm. OS, the co-primary endpoint, remained marginally superior for those randomized to M compared to P; the 10-year OS estimate was 43.7% (95% CI, 38.7-49.3%) vs 38.6% (95% CI, 33.6-44.4%) (p= 0.0485). The hazard ratios for OS favored randomization to M in each key biological subgroup (FLT3-ITD low or high allelic ratio or FLT3-TKD), but smaller subsets precluded statistical significance. There was a 10-year OS benefit in favor of M in men (p=0.005), but not women (p=0.9). A total of 414 transplants were performed: only 5 in the latter half of the follow-up period. Transplantation in CR1 was highly beneficial overall (10-year OS, 56% vs 35.8% without transplantation, p=0.001). Among those transplanted in CR1, randomization to M was marginally better than P (61.5% vs 49%, p=.064). After censoring for alloHCT in CR1, there was an OS trend in favor of M (51.8% v 48.0%, p=0.102). In CR1 patients who were randomized to maintenance therapy, M reduced the cumulative incidence of relapse in ELN 2017 favorable and intermediate risk pts, but not in those with adverse risk disease (HR 0.71, 0.47, and 1.01, respectively); maintenance M had no effect on OS. Only 3 pts relapsed after 5 years (all randomized to M). 25 pts randomized to M vs 13 P pts died after 5 years of follow-up. 5 pts on each arm died of a new primary cancer whereas 17 M and 5 P pts were coded as having died from treatment-related causes. None of the 3 pts who relapsed after 4 years had spliceosome mutations at diagnosis; whereas 18% (21/117) patients pts who had relapsed earlier had spliceosome mutations at diagnosis (p=0.419). Conclusions: The EFS benefit of randomization to midostaurin vs placebo when added to chemotherapy was maintained over time, although the benefit for OS was diminished, likely due in part to aging. Patient and disease factors differed between early vs late relapses, which could suggest the use of alternative therapies during distinct therapeutic stages.

Lack of HLA-matched related/unrelated donor remains a barrier to allogeneic hematopoietic cell transplantation (HCT) for adult acute myeloid leukemia (AML), with ongoing uncertainty about optimal donor type if more than one alternative donor is available. To assess the relationship between donor type, pre-HCT measurable residual disease (MRD), and post-HCT outcomes, we retrospectively analyzed 1265 myelodysplastic neoplasm (MDS)/AML and AML patients allografted in first or second remission with an HLA-matched sibling (MSD) or unrelated donor (MUD), HLA-mismatched unrelated donor (MMD), an HLA-haploidentical donor, or umbilical cord blood (UCB) at a single institution. Relapse risk was non-significantly higher after HLA-haploidentical and lower after UCB HCT. Non-relapse mortality (NRM) was significantly higher in patients undergoing MMD HCT, HLA-haploidentical HCT, and UCB, translating into significantly lower relapse-free survival (RFS) and overall survival for MMD and HLA-haploidentical HCT. There was a significant interaction between conditioning intensity and post-HCT outcomes for UCB HCT with better RFS for UCB HCT after MAC but higher NRM after non-MAC. In patients with pre-HCT MRD receiving MAC, relapse risk was significantly lower and RFS higher in those who underwent UCB HCT in comparison to MSD/MUD. Together, UCB HCT is a valuable alternative for MAC HCT, particularly in patients with pre-HCT MRD.

Context The decision on whether patients with acute myeloid leukemia (AML) should proceed with allogeneic hematopoietic cell transplantation (allo-HCT) involves complex considerations. Aim To examine how patient health factors and clinician assessments influence this decision. Methods Our prospective, multicenter observational study included 695 adults with AML, conducted across 13 predominantly academic centers. Data on demographics, cytogenetic risks, treatment responses, and geriatric assessment were collected at enrollment and at intervals of 1, 3, 6, 9, 12, 18, and 24 months. Competing risk Cox regression analysis was used to evaluate factors influencing mortality and HCT receipt. Results By 9 months, 43% of participants had undergone HCT. In participants who did not undergo HCT, multivariable analysis identified factors associated with increased risk of mortality, including high HCTcomorbidity index scores (HR, 2.11; P<0.0001); age >50 years, particularly for those ≥70 years who had a higher risk (HR, 2.71; P<0.0001); intermediate- (HR, 2.43; P=0.0003) and adverse-risk (HR, 4.3; P<0.0001) ELN cytogenetics; R/R disease at enrollment (HR, 2.04; P<0.0001); low-intensity chemotherapy regimens (HR, 1.42; P=0.04); dependence in activities of daily living <14 (HR, 1.59; P=0.005); and depression measured by PHQ-9 (HR, 1.56; P=0.009). Among those who survived long enough to potentially receive allo-HCT, factors associated with not receiving HCT were age ≥70 years (HR, 1.56; P=0.009), low ELN risk (HR, 0.28; P<0.0001), low-intensity induction (HR, 0.56; P=0.02), poor Karnofsky performance status (≤70) (HR, 0.49; P=0.0005), and relapse after achieving CR1 (HR, 0.41; P=0.001). Among patients aged ≥60 years, impaired cognition using Blessed Orientation-Memory-Concentration test (HR, 0.45; P=0.007) and impaired hearing (HR, 0.71; P=0.009) independently reduced chances of receiving HCT. Patients who were more likely to receive HCT had high-risk MDS (HR, 2.43; P<0.0001), had R/R disease when first seen at the center (HR, 2.43; P<0.0001) versus newly diagnosed AML, or were in CR1 after induction (HR, 4.59; P<0.0001) versus CR2 or later. Conclusions Multiple factors influence HCT receipt and mortality of patients with AML. A significant age-related disparity against those older than 70 years or with aging-associated vulnerabilities were noted. Multiple modifiable other patient-related factors limit chances to receive HCT. Future efforts will address roles of social determinants in likelihood of HCT recipients and novel trials to enhance patient readiness for HCT.

NPM1 mutations (NPM1+) are found in about 30% of adult AML. While NPM1+ are generally associated with favorable risk, prior studies have shown that NPM1+ older patients, defined as ≥ 65 years, have significantly worse outcomes. The updated European Leukemia Net 2022 (ELN22) now classifies NPM1+ AML patients with FLT3-ITD (ITD+) and MDS-associated mutations (MDSm+) as adverse-risk, while NPM1+/ITD-/MDSm+ patients remain favorable-risk unless they have an adverse-risk karyotype. A recent report (Chan et al., Blood Adv 2024) found MDSm+ conferred worse OS only in patients ≥ 60 years; however another report (Othman et al., Blood 2024) did not find MDSm+ were significantly associated with adverse outcomes in a larger cohort of NPM1+ patients. To further assess this question, we examined the mutational landscape in a large cohort of pre-treatment, diagnostic specimens from 456 patients enrolled in SWOG clinical trials (SWOG-9031, SWOG-9333, S0106, and S0112) and 126 patients from the UW/FHCC Hematopoietic Repository (Total N=582). Patients were treated with curative intent as per standard practice or clinical trials. Out of the 582 patients, 191 were NPM1+ (33%), of which 148 patients had cytogenetic and long-term outcome data. Kaplan-Meier and cumulative incidence estimates were estimated. C-statistics and AUC for Cox, cause-specific and logistic regression models for overall survival (OS), event-free survival (EFS), time to relapse (TTR, calculated in n=106 who achieved complete remission [CR], death in remission analyzed as a competing risk) and CR were calculated for age, ELN17, and ELN22, as well as ELN17 and ELN22 adjusting for age. ELN17 favorable-risk patients reclassified as adverse-risk (N=5 of 108) demonstrated significantly shorter EFS (HR=2.50, p=0.05) and CR rate (OR=0.20, p=0.024), although with similar OS (HR 1.30, p=0.66) and TTR (HR=1.20, p=0.86). ELN17 favorable-risk patients reclassified as ELN22 intermediate-risk (N=23 of 108) did not show a significant difference in OS (HR 1.47, p=0.20), EFS (HR 1.35, p=0.12), CR rate (OR 0.75, p=0.46), or TTR (HR=1.64, p=0.11) compared to ELN22 favorable-risk. For ELN17 intermediate-risk (N=40), 5 patients were reclassified as ELN22 adverse-risk (12.5%) and had significantly shorter EFS (HR 2.89, p=0.037) compared to ELN22 intermediate risk, although there was not a significant difference in CR rate (OR 0.31, p=0.23) or OS (HR 0.81, p=0.67). We next compared the prognostic significance of age relative to ELN risk classification. Age was associated with higher C-statistics (OS=0.65, TTR=0.57 and EFS=0.60) and AUC (CR=0.64) compared to both ELN17 (OS=0.56, TTR=0.54, EFS=0.54, CR=0.54) or ELN22 (OS=0.57, TTR=0.57, EFS=0.57, CR=0.58). The combination of age and ELN22 resulted in the highest C-statistics for OS, TTR, and EFS (OS=0.68, TTR=0.60, EFS=0.64). To further assess the impact of MDSm on risk-stratification of NPM1+ AML, we performed multivariable analyses adjusting for FLT3-ITD, MDSm, and age in patients without adverse-risk disease. These analyses showed MDSm did not have a significant association with OS when controlling for age or FLT3-ITD. We also compared ELN22 favorable-risk patients with and without MDSm (MDSm+ N=18; MDSm- N=61). Compared with the MDSm-, there was a trend toward an inferior OS (HR 1.85, p=0.054) for MDSm+ patients but not EFS (HR 1.38, p=0.3) or TTR (HR 1.2, p=0.64). When stratified by age (<65, N=53 vs. ≥65, N=26), MDSm+ in younger patients did not show a difference for OS (HR 0.96, p=0.96), EFS (HR 0.63, p=0.53) or TTR (HR 0.4, p=0.37). Older ELN22 favorable-risk patients (≥65) had similarly poor OS (MDSm- HR 3.33 and MDSm+ 3.27, respectively, p=0.001 for both) and EFS (MDSm- HR 3.11 and MDSm+ 2.39, p=0.001 and 0.01) compared to younger MDSm- patients. In our study, the ELN22 guidelines incrementally improve risk stratification for EFS in NPM1+ patients. However, age remains the most prognostic factor, and patients ≥ 65 years have similar adverse clinical outcomes, whether classified as favorable risk by ELN17, ELN22, or incorporating MDSm. Moreover, MDSm do not confer a worse prognosis in NPM1+ patients when accounting for age or FLT3-ITD status. These findings differ from a recent report by Chan et al. and are in keeping with the report by Othman et al. Therefore, further understanding of the underlying biology of age and its effect on NPM1+ AML is needed to improve outcomes for older patients.

Background: Measurable residual disease (MRD), assessed by multiparameter flow cytometry (MFC) before allogeneic hematopoietic cell transplantation (HCT), is strongly and independently associated with poor post-HCT outcomes in adults with acute myeloid leukemia (AML). On the other hand, the relative prognostic value of MFC-based MRD testing is unknown in patients with myelodysplastic neoplasm (MDS)/AML, a disease entity newly defined by the 2022 International Consensus Classification (ICC) that encompasses cases of MDS with 10-19% blasts and recognizes the diagnostic continuum between MDS and AML. Methods: To assess the relationship between pre-HCT, disease type, and post-HCT outcomes, we retrospectively analyzed 1,262 adults ≥18 years with MDS/AML (n=150) or AML (n=1,112), based on the ICC 2022 criteria, who received a first allograft while in first or second morphologic remission at a single institution between 4/2006 and 3/2023. Results: The MDS/AML and AML cohorts differed significantly regarding patient and disease-specific characteristics, including age at HCT (60 vs. 56 years in MDS/AML and AML patients, respectively, P<0.001), HCT-CI score ( P<0.001), adverse cytogenetic risk according to ELN 2022 criteria (39% vs. 26%, P<0.001), intensive therapy prior to allogeneic HCT (69% vs. 97%, P<0.001), time from last remission (76 vs. 98 days, P<0.001), first remission status (97% vs. 77%, P<0.001), pre-HCT MRD by MFC (35% vs. 19%, P<0.001), and peripheral blood stem cells as graft source (89% vs. 80%, P=0.015). There were no differences in donor source or conditioning intensity. With a median follow-up of 5.11 years (interquartile range [IQR]: 2.38-9.66) in survivors, there were 384 relapses, 590 deaths, and 258 non-relapse mortality (NRM) events contributing to the estimates for relapse, NRM, relapse-free survival (RFS), and overall survival (OS). At three years, NRM (21% [14-28%] for MDS/AML vs. 16% [14-19%] for AML, P=0.19), relapse (29% [22-36%] vs. 29% [27-32%], P=0.94), RFS (50% [42-59%] vs. 54% [51-58%], P=0.25), and OS (52% [44-61%] vs. 60% [57-63%], P=0.09) were similar between the two groups. After multivariable adjustment, relapse rates for patients with MDS/AML vs. those with AML were similar (hazard ratio [HR]=0.77 [0.50-1.17], P=0.2), as were NRM (HR=1.41 [0.91-2.18), P=0.12), RFS (HR=0.99 [0.73-1.35], P>0.9) and OS (HR=1.17 [0.85-1.61], P=0.3). Importantly, a significant interaction was found between pre-HCT MRD status and disease type (MDS/AML vs. AML) for relapse ( P=0.009), RFS ( P=0.013), and OS ( P=0.048). The interaction models indicated that the HRs for the association between pre-HCT MRD and post-HCT outcomes were lower in patients with MDS/AML (for relapse: HR=1.73 [0.96-3.11] in MDS/AML vs. HR=3.56 [2.90-4.38], in AML; for RFS: HR=1.60 [1.03-2.48] vs. HR=2.67 [2.25-3.16]; for OS: HR=1.53 [0.97-2.39] vs. HR=2.34 [1.96-2.78]). Pre-HCT MRD was associated with a higher risk of relapse in both MDS/AML and AML patients but to a lesser extent in the MDS/AML patients. In these patients, relapse at three years was 37% [24-51%] for those with pre-HCT MRD vs. 24% [16-33%] for those without ( P=0.068), translating into lower RFS (39% [28-56%] vs. 56% [47-67%], P=0.038) and OS (41% [29-58%] vs. 58% [48-70%], P=0.066) in MDS/AML patients with pre-HCT MRD. For comparison, in patients with AML, relapse at three years was 22% [20-25%] for those without MRD vs. 56% [50-62%] for those with ( P<0.001), translating into lower RFS (27% [22-31%] vs. 61% [58-64%], P<0.001) and OS (35% [29-42%] vs. 65% [62-68%], P<0.001) in those with pre-HCT MRD. In line with these findings, C-statistics showed that pre-HCT MRD was not as predictive in MDS/AML patients in comparison to AML patients (for relapse: 0.57 in MDS/AML patients vs. 0.62 in AML patients; for RFS: 0.57 vs. 0.59; for OS: 0.56 vs. 0.58). Conclusion: Our data suggest that, overall, patients with MDS/AML or AML in first or second morphologic remission have similar outcomes following allogeneic HCT. However, MFC-based pre-HCT MRD testing is prognostically substantially less informative for MDS/AML than AML patients, possibly because of increased rates of misclassification. These findings support the rationale to examine the prognostic value of molecular MRD testing in patients with MDS/AML. Figure. Cumulative incidence of relapse stratified (A) by disease type at diagnosis and (B) by disease type at diagnosis and pre-HCT MRD.

The European LeukemiaNet (ELN) genetic risk classifications were developed based on data from younger adults receiving intensive chemotherapy. Emerging analyses from patients receiving less-intensive therapies prompted a proposal for an ELN genetic risk classification specifically for this patient population.

Second allogeneic hematopoietic cell transplantation (HCT2) is potentially curative for adults with acute myeloid leukemia (AML) or myelodysplastic neoplasm (MDS)/AML experiencing relapse after a first allograft (HCT1), but prognostic factors for outcomes are poorly characterized. To provide a detailed analysis of HCT2 outcomes and associated prognostic factors in a large single-center cohort, with a focus on identifying predictors of relapse and nonrelapse mortality (NRM), we studied adults >= 18 years who underwent HCT2 at a single institution between April 2006 and June 2022 for relapsed AML (n = 73) or MDS/AML (n = 8). With a median follow-up among survivors of 74.0 (range: 10.4 to 187.3) months, there were 30 relapses and 57 deaths, of which 29 were NRM events, contributing to the estimates for relapse, overall survival (OS), relapse-free survival (RFS), and NRM. Three-year estimates for relapse, RFS, and OS were 37% (95% confidence interval: 27% to 48%), 32% (23% to 44%), and 35% (26% to 47%). The rate of NRM at 100 days and 18 months was 20% (12% to 29%) and 28% (19% to 39%). Outcomes differed markedly across patient subsets and were substantially worse for patients who underwent HCT2 with active disease (ie, morphologic evidence of bone marrow and/or extramedullary disease), for patients who relapsed <= 6 months after HCT1, and for patients with higher HCT-specific Comorbidity Index (HCT-CI) or treatment-related mortality (TRM) scores. After multivariable adjustment, active disease was associated with a higher risk of relapse (hazard ratio [HR] = 3.19, P = .006) and shorter RFS (HR = 2.41, P = .008) as well as OS (HR = 2.17, P = .027) compared to transplant in morphologic remission without multiparameter flow cytometric evidence of measurable residual disease. Similarly, a relapse-free interval <= 6 months after the first allograft was associated with higher risk of relapse (HR = 5.86, P < .001) and shorter RFS (HR = 2.86; P = .001) and OS (HR = 2.45, P = .003). Additionally, a high HCT-CI score was associated with increased NRM (HR = 4.30, P = .035), and shorter RFS (HR = 3.87, P = .003) and OS (HR = 3.74, P = .006). Likewise, higher TRM scores were associated with increased risk of relapse (HR = 2.27; P = .024) and NRM (HR = 2.01, P = .001), and inferior RFS (HR = 1.90 P = .001) and OS (HR = 1.88, P = .001). A significant subset of patients with AML or MDS/AML relapse after HCT1 are alive and leukemia-free 3 years after undergoing HCT2. Our study identifies active leukemia at the time of HCT2 and early relapse after HCT1 as major adverse prognostic factors, highlighting patient subsets in particular need of novel therapeutic approaches, and supports the use of the HCT-CI and TRM scores for outcome prognostication.