Abstract 2539: DNA content and chromatin texture measurement to predict malignant transformation in low-grade oral dysplasia
Bibliographic record
Abstract
Abstract Objectives: The key to reducing oral cancer morbidity and mortality is the early identification of oral potentially malignant lesions (OPMLs) at high risk of malignant transformation (MT). The gold standard to assess OPML risk is the grade of dysplasia. Risk prediction of low-grade dysplasias (LGDs) is challenging as only a small proportion progress. Currently no tools or biomarkers are used clinically to predict MT. Abnormal DNA content has shown to be a marker of malignancy in various sites including the oral cavity. Changes in nuclear morphology and chromatin texture have also been observed in cells with normal DNA content adjacent to cancerous fields as a result of genetic or epigenetic alterations that drive MT. The aim of this study is to use OPML lesion brushings to measure nuclear DNA content and morphometric features to identify oral LGDs at high risk of MT. Methods: Patients with oral LGDs consented into the British Columbia Oral Cancer Prediction Longitudinal Study were included in the study. Inclusion criteria included patients with mild or moderate dysplasia with lesion brushings prior to or concurrent to biopsy dates and had more than 6 months of follow up. Outcome was defined as progression to severe dysplasia or cancer. Thin monolayer slides were prepared and stained using Feulgen Thionin. Slides were scanned using the Cancer Imaging Scanner at BC Cancer using machine learning algorithms to measure DNA amount and over 120 nuclear morphometric features. Linear discriminant function (LDF) analysis of nuclear features from normal diploid cells were used to train classifiers to differentiate progressing from non-progressing lesions. Results: A total of 143 lesion brushings from 140 patients were selected. Forty-nine mild, 48 moderate, and 46 lesions with a previous history of dysplasia at the same site were included of which 16 (11%) progressed. DNA content of each nucleus was measured by a normalized scale known as DNA Index (DI) and were classified into groups: diploid (0.9 2.5). Abnormal DNA content threshold was determined using combinations of cell frequencies/percentage of each group. A lesion was considered at high-risk of MT when it showed 5 or more non-diploid cells (DI>1.2). 12 out of 16 progressors and 41 out of 116 non-progressors showed high-risk DNA content. Preliminary results from LDF show that chromatin texture features best differentiated progressing cells from non-progressing. Analysis to determine the best threshold for chromatin texture features is ongoing. Conclusion: This suggests that chromatin texture changes can be observed in cells with normal DNA content during MT. Together, they can serve as a quick, non-invasive, cost effective tool to triage high risk OPMLs to cancer care centres for monitoring and early intervention. A larger sample size is required to validate our results. Citation Format: Madhurima Datta, Denise M. Laronde, Miriam Rosin, Anita Carraro, Korbelik Jagoda, Alan Harrison, Zhaoyang Chen, Martial Guillaud. DNA content and chromatin texture measurement to predict malignant transformation in low-grade oral dysplasia [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2539.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| 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) | 0.002 | 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".