Abstract 3251: Assessing risk markers for oral cancer recurrence
Bibliographic record
Abstract
Abstract Oral cancer has a poor 5-year survival rate, partly due to a high local recurrence (REC) rate. Clinical indicators such as size, color, and site have been found to be associated with progressing primary oral premalignant lesions (OPL), but have yet to be evaluated for OPL development at treated cancer sites. Similarly, recent LOH regions highly predictive of primary progression (Zhang et al., Cancer Prevention Research 2012), have not been evaluated as predictive markers of OPLs in oral cancer REC. Methods: The analysis involved 194 patients enrolled in the Oral Cancer Prevention Longitudinal (OCPL) study with early stage (Stage I & II) squamous cell carcinoma (SCC) or carcinoma-in-situ (CIS), treated with curative intent. Data was collected every 3 months post-treatment up to 2 years, then every 6 months. Demographic (age, gender, risk habits), tumor information (site, stage, treatment) and clinical data (presence of an OPL, size, color, texture and appearance) were collected. Biopsies taken during follow-up were tested for high-risk molecular patterns. REC was defined as development of a severe dysplasia, CIS or SCC within 3 cm of the primary tumor site. Results: 31(16%) patients suffered a REC at the former tumor site. Presence of an OPL in follow-up had an almost 7-fold increased risk of REC (P<0.001). 19 (10%) patients ‘always’ had an OPL, 74 (38%) had an OPL ‘sometimes’, while 101 (52%) ‘never’ had an OPL during follow-up. 74% of the ‘always’ OPL group (RR=68, P<0.001), 18% of the ‘sometimes’ OPL group (RR=5, P<0.01) and 6% of the ‘never’ OPL group developed a REC (RR=1). Patients who developed an OPL during the first year of follow-up had the highest risk of REC (RR= 6) compared to the patients who developed an OPL in the second year of follow-up. None of the OPL clinical characteristics associated with primary progression was associated with risk of REC. Of 76 patients who had a biopsy in follow-up that was available for analysis, patients with a molecular profile showing a loss at 9p (RR=4) were more likely to develop a REC than samples with 9p retention. Patients with low-grade dysplasia at the surgical margins were more apt to develop a REC (RR=2.7). Conclusion: The presence of an OPL at a former tumor site is a critical predictor of oral cancer REC regardless of its characteristics, with nearly all cases preceded in time by its appearance. The issue is whether it is reactive or a true OPL with risk. Risk of REC increases if an OPL is present at each follow-up visit and/or is present within the first year following treatment. OPLs persisting during follow-up should be biopsied. Regardless of histology, the presence of high-risk molecular patterns in these samples offers an additional opportunity to assess risk of REC. Citation Format: Denise M. Laronde, Lewei Zhang, P. Michele Williams, Bertrand Chan, Jay H. Park, Catherine F. Poh, Miriam P. Rosin. Assessing risk markers for oral cancer recurrence. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 3251. doi:10.1158/1538-7445.AM2014-3251
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| 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) | 0.004 | 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".