The Atlas of Blood Cancer Genomes (ABCG) Project: A Comprehensive Molecular Characterization of Leukemias and Lymphomas
Bibliographic record
Abstract
Abstract Introduction Blood cancers are collectively common and strikingly heterogeneous diseases both clinically and molecularly. According to the WHO taxonomy, there are over 100 distinct myeloid and lymphoid neoplasms. Genomic profiling of blood cancers has been applied in a somewhat ad hoc fashion using diverse sequencing approaches including the use of targeted panels, whole exome sequencing, whole genome sequencing, RNA sequencing, etc. The lack of data uniformity has made it difficult to comprehensively understand the clinical and molecular spectrum within and across diseases. Systematic genomic approaches can address the central challenges in the diagnosis and treatment of blood cancers. For the diagnosis of blood cancers, the incorporation of genomics could greatly enhance the accuracy and speed of clinical diagnostics. Genomics could also inform their pathology classification. However, these applications must be preceded by a clear understanding of the particular genetic aberrations and expression profiles that unite and distinguish different leukemias and lymphomas. Therapeutic development can also be aided by genomic approaches through identification of new targets and establishing the relevance of existing targets and treatments. Targeted therapies including those directed at specific surface markers (e.g. CD19, CD30 and CD123) or molecular targets (e.g. BCR-ABL fusions, IDH1 mutations and EZH2 mutations) are rarely restricted to a single disease, with most occurring in multiple blood cancers. A systematic understanding of the presence or overlap of these targets within or across blood cancers would significantly expand the therapeutic possibilities and better enable the use of existing therapies in both common and rare cancers. However, such therapeutic possibilities need to be established through a rigorous, data-driven approach. We initiated the Atlas of Blood Cancers Genomes (ABCG) project to systematically elucidate the molecular basis of all leukemias and lymphomas by building upon advances in genomic technologies, our capabilities for data analysis and economies of scale. Using a uniform approach to systematically profile all blood cancers through DNA and RNA sequencing at the whole exome/whole transcriptome level, we aim to link genomic events with clinical outcomes, disease categories and subcategories, thereby providing a complete molecular blueprint of blood cancers. Methods/Results The ABCG project consists of collaborators from 25 institutions around the world who have collectively contributed samples from 10,481 patients comprising every type of blood cancer in the current WHO classification. The samples include thousands of myeloid leukemias and mature B cell lymphomas, hundreds of Hodgkin lymphoma and plasma cell myeloma, as well as every rare type of hematologic malignancy (along with case-matched normal tissue). All cases were de-identified and their associated pathology and detailed clinical information entered into a purpose-built web-based system that included disease-specific data templates. All cases were subjected to centralized pathology review and clinical data review by experienced hematopathologists and oncologists. All 10,481cases are being sequenced at the DNA and RNA level, and are being profiled to define the genetic alterations and expression changes that are characteristic of each disease. Analysis will include translocations, copy number alterations, and viral status. These molecular features will be examined in conjunction with genetic events, pathologic factors, and the clinical features. We have already generated results for ALK-negative anaplastic large B cell lymphoma and primary mediastinal B cell lymphomas (N=210). These data demonstrate novel subgroup and molecular discoveries that are enabled by integrative DNA and RNA sequencing analysis and the examination of molecular features across different diseases as well as within individual entities. In addition, other disease entities and the collective data will be presented in the meeting. Conclusion The ABCG project will comprehensively study the genetic and clinicopathological features of all blood cancers using systematic genomic approaches. We anticipate our data, approaches and results will serve as a lasting resource for the molecular classification and therapeutic development for leukemias and lymphomas. Disclosures McKinney: Novartis: Research Funding; Nordic Nanovector: Research Funding; Molecular Templates: Consultancy, Research Funding; Kite/Gilead: Honoraria, Speakers Bureau; Incyte: Research Funding; Genetech: Consultancy, Honoraria, Research Funding; Epizyme: Consultancy; Celgene: Consultancy, Research Funding; BTG: Consultancy; Beigene: Research Funding; ADC Therapeutics: Consultancy, Speakers Bureau; Pharmacyclics: Consultancy; Verastem: Consultancy. Behdad: Lilly: Speakers Bureau; Roche/Foundation Medicine: Speakers Bureau; Thermo Fisher: Speakers Bureau.
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 | 0.003 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.007 | 0.010 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.006 | 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".