S170: MYELODYSPLASTIC NEOPLASMS (MDS) CLASSIFICATION FROM WHO 2017 TO WHO 2022 AND ICC 2022): AN EXPANDED ANALYSIS OF 7017 PATIENTS ON BEHALF OF THE INTERNATIONAL CONSORTIUM FOR MDS (ICMDS)
Bibliographic record
Abstract
Topic: 10. Myelodysplastic syndromes - Clinical Background: In 2022 two new classifications for myeloid neoplasms were published: the World Health Organization (WHO) and International Consensus Classification (ICC). Aims: To validate and compare these 2 classifications in large international cohort of MDS patients (pts) and provide data-driven evidence for future harmonization. Methods: We analyzed data separately in 2 large annotated MDS datasets from the USA (Moffitt Cancer Center:MCC) and Europe (GenoMed4all:GM) under the auspices of icMDS. All pts with molecular data were reclassified by WHO and ICC criteria (n=7017 pts). Results: MDS-SF3B1 group demonstrated favorable outcome. In MCC cohort (n= 294, 13% by WHO and n=277,12% by ICC), median leukemia free survival (LFS) was 100.6 months (mo) and 109.4 mo, median overall survival (OS) was 101.8 mo and 111.6 mo by WHO and ICC respectively. Similarly in GM cohort (n=654, 13.9% by WHO, n=594, 12.6% by ICC) LFS was 102.2 mo and 101.9 mo, OS was 104.9 mo and 101.9 mo by WHO and ICC respectively. Pts with MDS-del5q had good prognosis. In MCC cohort (n=107, 5% by WHO; n=108, 5% by ICC) LFS was 65 mo and OS 75.6 mo. In GM cohort (n=219, 4.6% by WHO; 223, 4.7% by ICC) OS was 81.1 mo. TP53-mutated (mTP53) MDS pts had the shortest survival. In MCC, the WHO category of MDS with bi-allelic TP53 (MDS-biTP53; n=214; 10%) had LFS of 10 mo and OS of 13.2 mo. Similarly, the ICC mTP53 groups MDS-mTP53 (n=194; 9%) and MDS/AML-mTP53 (n= 115; 5%) had a LFS of 11.5 and 6.4 mo, OS of 14.2 and 11 mo, P< 0.01. In GM cohort, WHO MDS-biTP53 (n=443, 9.4%) had a LFS of 13.4 mo and 14 mo OS. By ICC, MDS-mTP53 (n=290; 6.2%) and MDS/AML-mTP53 (n= 146; 3.1%) had LFS of 16.3 and 9.7 mo, OS of 17.6 and10 mo, respectively, P<0.01. Pts with MDS-RS SF3B1 wild type (MDS-RS WT) had similar outcome to MDS low blasts (LB). In MCC Cohort, MDS-RS WT (n= 78, 4%) had a LFS (50.5 vs. 46.2 mo, P = 0.72) and OS (54.3 vs. 56 mo; P = 0.99) compared to LB (n=704,31%). In GM cohort, both MDS-RS WT (n= 126, 2.9%) and LB (n= 1612, 34.3%) had similar LFS (50.3 vs 52.3 mo, P=0.79) and OS (58.2 vs 59.5 mo, P=0.82). In ICC, MDS with multilineage dysplasia (MDS-MLD) had worse outcome compared to single lineage dysplasia (SLD). In MCC cohort, MDS-MLD (n= 610, 28%) had shorter LFS (41.4 vs. 74.2 mo, P< 0.01) and OS (49.5 vs. 79.4 mo, P< 0.01) than SLD (n=247,11%). In GM cohort, MDS-MLD (n=1300, 27.6%) had shorter LFS (40.9 vs 70.7 mo, P<0.01,) and OS (56.2 vs 71.9 mo, P<0.01) compared to SLD (n=519,11%). The WHO hypoplastic MDS (n=94; 4%) had similar OS to the LB. Increased myeloblasts correlated with worse outcome. In MCC cohort, WHO MDS-IB1 had a shorter LFS (20.5 vs. 47.4 mo, P< 0.01) and OS (28.1 vs. 55.3 mo, P< 0.01) than LB. Pts with MDS-IB2 had shorter LFS than IB1 (11 vs. 20.5 mo, P< 0.01) but no difference in OS (23.7 vs. 28.1 mo; P= 0.79). In GM cohort, MDS-IB1 had a shorter LFS (24.6 vs 50.3 mo, P<0.01) and OS (33.5vs. 59.3mo, P< 0.01) compared to LB. MDS-IB2 also had shorter LFS (16.5vs. 24.6 mo, P< 0.01) and OS (22.2 vs. 33.5 mo; P<0.01) than IB1. The WHO MDS with fibrosis had worse OS than MDS-IB (18.9 vs. 26.1 mo; p<0.01). Summary/Conclusion: We assessed the WHO and ICC MDS classifications in an international dataset. Genetically defined entities (SF3B1, del5q, and bi-TP53) are clearly unique. Survival for MDS-RS WT pts was similar to MDS-LB. MDS-MLD pts had shorter OS than SLD. Pts with increased myeloblasts have worse outcome, however, a blast cut-off needs to be refined. Ongoing analyses seek to generate a hierarchical harmonized classification using advanced statistical methods of inference.Keywords: WHO classification, ICC, MDS
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".