Abstract 1417: Development of a miRNA-based prediction tool to discriminate cutaneous blastic plasmacytoid dendritic cell neoplasm from cutaneous myeloid sarcoma
Bibliographic record
Abstract
Abstract Background Blastic plasmacytoid dendritic cell neoplasm (BPDCN) and myeloid sarcoma (MS) are two extremely rare and aggressive hematological diseases. Both malignancies most commonly arise with skin lesions with or without extramedullary organ involvement before leukemic dissemination. Given its rarity and less known biology, in some cases with defective/ambiguous phenotype distinguish between these two entities may be challenging for pathologists and clinicians.1 Can the machine learning predictive model help to solve this diagnostic question? Here we performed the first study of microRNA (miRNA) profiling of BPDCN and MS in order to 1) Discover new molecular features selectively driving BPDCN respect to MS 2) Develop a machine learning based-prediction tool useful for discriminating BPDCN and MS. Methods We performed miRNA profiling (NanoString Technologies) of cutaneous biopsies of 16 BPDCN and 23 MS cases. Using Supervised Analysis, we identified 49 miRNAs differentially expressed that were randomly validated by qRT-PCR and next interrogated by functional enrichment analysis. Finally, a machine learning algorithm based on Linear Discriminant Analysis2 was applied to identify candidate miRNAs able to discriminate BPDCN from MS cases. Results In line with the overlapping clinical features of BPDCN and MS, the molecular profiling of these two diseases was proved to be extremely similar. Unsupervised Analysis well demonstrated that the miRNA profiles of the two malignancies are closely related and indeed, BPDCN and MS cases cluster together. When a Supervised Analysis was applied, we identified a set of 49 miRNAs differentially expressed in BPDCN respect to MS, 25 down- and 24 up-regulated. Of relevance, down-regulated miRNAs were predicted to be markedly involved in the apoptosis regulation of BPDCN. Machine learning predictive model identified a set of 12 miRNAs (5 up and 7 down) able to discriminate cutaneous BPDCN from MS. Conclusion This is the first miRNA profiling study in BPDCN and MS that showed how strongly these two diseases overlap at molecular level. Despite their similarity, BPDCN cases displayed a set of miRNAs significantly down-regulated when compared to MS, with possible dysregulation of cell death pathway. Of practical interest, we designed a machine learning predictive model based on the expression of 12 miRNAs, which alone may be applied to distinguish between the two hematological diseases. This tool, if validated in a larger set of cases, may help to differentiate BPDCN and MS in cases with defective/ambiguous phenotype. References 1. Weltgesundheitsorganisation. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Revised 4th edition. (Swerdlow SH, Campo E, Harris NL, et al., eds.). Lyon: International Agency for Research on Cancer; 2017. 2. Laginestra MA, Piccaluga PP, Fuligni F, et al. Pathogenetic and diagnostic significance of microRNA deregulation in peripheral T-cell lymphoma not otherwise specified. Blood Cancer J. 2014;4:259. doi:10.1038/bcj.2014. Citation Format: Maria Rosaria Sapienza, Fabio Fuligni, Federica Melle, Valentina Tabanelli, Valentina Indio, Alessandro Pileri, Lorenzo Cerroni, Francesco Bacci, Giovanna Motta, Maria Antonella Laginestra, Saveria Mazzara, Luciano Cascione, Alessandro Laganà, Claudio Agostinelli, Manuela Ferracin, Elena Sabattini, Carlo Croce, Stefano Pileri. Development of a miRNA-based prediction tool to discriminate cutaneous blastic plasmacytoid dendritic cell neoplasm from cutaneous myeloid sarcoma [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1417.
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) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.002 |
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".