Extramedullary disease in AML: A high-risk feature beyond ELN 2022 classification
Bibliographic record
Abstract
Abstract Background Extramedullary disease (EMD) in acute myeloid leukemia (AML) involves leukemic infiltration outside the bone marrow and is observed in up to 25% of patients. Despite its frequency, EMD is not currently incorporated into the European LeukemiaNet (ELN) 2022 risk stratification system. Recent studies have demonstrated molecular discordance between leukemic cells in the bone marrow and those found at extramedullary (EM) sites, raising concerns that molecular profiling based solely on bone marrow samples may not adequately guide clinical management of patients with EMD. Objective To characterize the clinical features, treatment outcomes, and prognostic implications of EMD in a large, retrospective cohort of AML patients treated over two decades at Princess Margaret Cancer Centre. Methods We conducted a retrospective analysis of 617 adult AML patients, categorizing them into non-EMD (n=371) and EMD groups (n=246). The EMD group was further subdivided into isolated EM involvement (n=53) and concurrent marrow and EM involvement (n=193). We compared complete remission (CR) rates, relapse rates, rates of allogeneic hematopoietic stem cell transplantation (allo-HSCT), overall survival (OS), and event-free survival (EFS). Propensity score matching (PSM) was used to adjust for differences in age, ELN risk category, induction regimen, and allo-HSCT. Multivariate analyses (MVA) were performed to identify independent prognostic factors. Results Among the 246 patients with EMD, 53 (21.5%) had isolated EMD, while 193 (78.5%) presented with concurrent bone marrow involvement. CR rates following induction therapy were similarly high in both groups (96% in isolated EMD vs. 89% in combined involvement; p=0.18). Relapse rates (32% vs. 34%; p=0.77) and rates of allogeneic HSCT (40% vs. 32%; p=0.32) were also comparable, indicating similar outcomes between isolated EMD and concurrent marrow involvement with EMD.Prior to PSM, comparing non-EMD from 2015-2018 (n=371) patients to EMD patients (n=246), EMD patients were significantly younger (p=0.04) and had more intermediate-risk disease (60% vs. 19%, p<0.01) and NPM1 mutations (48% vs. 25%, p<0.01). They received intensive induction therapy more frequently (89% vs. 74%, p<0.01) but underwent allo-HSCT less often (34% vs. 43%, p=0.02). OS and EFS were significantly inferior in the EMD group (median OS: 14.2 vs. 53.2 months; median EFS: 10.6 vs. 43.7 months; both p<0.0001). After PSM (168 matched pairs), EM remained associated with inferior OS (median 14.2 vs. 64.1 months; and EFS :9.5 vs. 55.9 months; both p<0.0001).MVA confirmed EMD as an independent adverse prognostic factor for OS (HR 1.87, 95% CI 1.2–2.91, p=0.005) and EFS (HR 1.93, 95% CI 1.36–2.75, p=0.0002). Among EMD patients, allo-HSCT was associated with improved EFS (HR 0.49, 95% CI 0.24–0.99, p=0.048), although OS benefit did not reach significance (HR 0.60, p=0.20).The most common EM involvement sites were the skin (39%) and central nervous system (CNS; 15%). Neither specific EM site involvement, nor mutations frequently identified in EMD patients (NPM1, TET2, ASXL1, DNMT3A) were significantly associated with differences in OS or EFS.Notably, within the ELN 2022 favorable-risk subgroup, EMD presence was associated with higher relapse rates (31% vs. 14%, p=0.059) and trends toward inferior median OS (54.7 months vs. not reached, p=0.08) and EFS (40.6 months vs. not reached, p=0.05), highlighting its prognostic significance even in traditionally lower-risk AML patients. Conclusions This large, retrospective study emphasizes the negative prognostic impact of EMD in AML, independent of ELN risk stratification, even among patients classified as favorable-risk. Although allo-HSCT may offer some improvement in disease control, it does not fully overcome the poor prognosis linked to EMD. Additionally, these results underscore the importance of site-specific molecular profiling in guiding therapeutic decisions. EMD warrants inclusion in future AML risk stratification models and prospective therapeutic trials.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".