Establishing contemporary benchmarks for Acute Myeloid Leukemia outcomes and measurable residual disease: Real-world data utilization from the ELN-david MRD international working group
Bibliographic record
Abstract
Abstract Background The treatment landscape of acute myeloid leukemia (AML) is rapidly evolving due to advances in novel targeted agents, lower-intensity treatments, and allogeneic stem cell transplantation. Simultaneously, technologic advances are allowing increasingly sensitive detection of measurable residual disease (MRD) to guide treatment decision-making. However, our knowledge of AML patient characteristics, treatments, clinical outcomes and utility of multiparameter flow cytometry and molecular MRD techniques is primarily based on data from clinical trials. There is a need for real-world data to establish AML outcomes and MRD testing in routine practice. Thus, ELN-DAVID, an international European LeukemiaNet (ELN) working group focused on the assessment and validation of MRD in AML, launched the BENCHMARK initiative to establish a regularly updated platform for real-world clinical data and MRD assessment practices in ELN participating centers. Methods Centers provided de-identified, aggregated data from 100 or more unselected, consecutively seen patients with AML starting backwards from December 2022. Centers could include patients in clinical trials, if allowed by the clinical trial protocol. Descriptive data on patient characteristics, treatment strategies, response (morphologic and MRD) and survival were combined for collective analysis. As pooled survival analysis was not possible, the range of the median survival between centers is reported. European LeukemiaNet risk classification is defined as either the 2017 or 2022 version, depending on what the center used and entered. Results To date, data were provided for 1457 patients from 14 international centers treated between 2016 and 2022. Of these patients, 801 (55%) were male and a total of 974 (68%) patients received intensive chemotherapy, 323 (23%) non-intensive treatment, and 136 (9%) supportive care only. Across all treatment groups, most patients were not treated on clinical trials (86%), targeted (gemtuzumab-ozogamicin, FLT3 or IDH inhibitors) treatment was given to 36%, and the majority underwent allogeneic stem cell transplantation (64%). In patients treated with intensive chemotherapy the ELN risk group was 33% favorable, 29% intermediate, and 38% adverse. Of the non-intensively treated patients, most received venetoclax-based therapy (62%). Response evaluations showed complete remission (CR), or CR with incomplete count recovery (CRi) or CR with incomplete platelet recovery (CRp) in 80% of patients that were treated with intensive chemotherapy and in 54% of patients that received non-intensive treatment. Median follow-up time ranges from 4 months to 51 months and the median overall survival in each of the intensively treated ELN risk categories ranges from 16 months to not reached (NR), 8 months to NR, and 5.5 months to NR for favorable, intermediate and adverse risk patients, respectively. MRD data were available for 598 (61%) intensively treated patients after 2 cycles of treatment, 360 (37%) at the end of treatment and 295 (30%) during follow-up. Flow cytometry was the most frequently used MRD technology across all time points (cycle 2: 76%; end of treatment: 64%; follow-up: 62%), followed by quantitative PCR (cycle 2: 40%; end of treatment: 48%; follow-up: 43%). Next-generation sequencing was performed in 8% of patients after two cycles, 12% at end of treatment and 16% during follow-up. Among non-intensively treated patients, 107 (33%) had MRD data available, the majority (73%) was analyzed by flow cytometry. Conclusions The survival data from BENCHMARK-2025, reflecting international practices, are comparable to those expected from recent AML clinical trials. Most centers performed MRD assessment using flow cytometry for patients treated with intensive chemotherapy. MRD was infrequently performed in non-intensively treated patients. A standardized database has been developed to allow future analyses of individual patient data. It is anticipated that data will be updated regularly, giving the opportunity to construct synthetic control cohorts, allowing longitudinal correlation between evolving MRD practices and AML outcomes, and serving as benchmark for contemporary experience.
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 | 0.066 | 0.072 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.006 | 0.009 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.005 | 0.002 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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".