Abstract 5858: Prognostic and predictive role of T cell infiltration in stage III colon cancer (CC) treated with celecoxib: CALGB/SWOG 80702
Bibliographic record
Abstract
Observational studies have demonstrated that aspirin and/or PTSG2 inhibitor (e.g., celecoxib) use, before or after colon cancer (CC) diagnosis, is associated with a lower risk of recurrence. However, a randomized trial of adjuvant celecoxib therapy in stage III CC did not show improved survival compared to placebo treatment. We therefore hypothesized that immune microenvironment features may identify patient subgroups with stronger survival benefits from anti-inflammatory celecoxib treatment. Drawing from an NCI-Intergroup trial (Cancer and Leukemia Group B [now Alliance for Clinical Trials in Oncology]/SWOG 80702) for patients with stage III resected CC that compared 3 versus 6 months of adjuvant fluorouracil, leucovorin, and oxaliplatin ± 3 years of celecoxib, we conducted tissue-based tumor microenvironment profiling on 1,098 resected tumors using a custom 9-plex, T-cell multiplex immunofluorescence assay coupled with digital image analysis and supervised machine learning. The Kaplan-Meier method was used to describe disease-free survival, based on T cell density, and log-rank testing was performed. Cox proportional hazards models were used to examine unadjusted and adjusted associations between T cell density and disease-free survival (DFS). Two-sided P values ≤0.05 were considered statistically significant. We found that higher CD3+ T cell density was associated with longer DFS with an adjusted hazard ratio (HR) of 0.69 (95% confidence interval [CI] 0.52-0.91) for patients with densities in the highest tertile as compared the lowest tertile (P=0.01). T cell subset analysis showed that stromal regulatory T helper cells (HR 0.42 comparing highest to lowest tertile, CI 0.31-0.58, P<0.001) and stromal memory cytotoxic T cells (HR 0.55 comparing highest to lowest tertile, CI 0.42-0.74, P<0.001) were most strongly associated with longer DFS. Compared to the placebo group, adjuvant celecoxib use associated with improved DFS for patients with low but not high stromal CD3+ T cell density (adjusted HRs of 0.45 [CI 0.31-0.65] for lowest tertile and 1.13 [CI 0.73-1.74] for highest tertile; Pinteraction=0.01). This association was also seen when the analysis was restricted to microsatellite stable tumors (HR 0.50 [CI 0.32-0.80] for lowest density tertile and 1.50 [CI 0.87-2.59] for highest density tertile; Pinteraction=0.01). T cell subset analysis showed that DFS benefit for celecoxib use most strongly associated with stromal memory helper T cell density (HR 0.38 [CI 0.26-0.56] for lowest density tertile and 1.26 [CI 0.80-1.97] for highest density tertile; Pinteraction<0.001). Together, these results show that while higher T cell infiltration is associated with improved DFS, lower T cell infiltration is associated with improved DFS for patients treated with celecoxib. This unexpected finding may inform immunomodulatory treatment strategies for CC. Citation Format: Yasutoshi Takashima, Chao Ma, Qian Shi, Andressa Dias Costa, Tyler Twombly, Juha P. Väyrynen, Melissa Zhao, Ardaman Shergill, Pankaj Kumar, Felix Couture, Philip Kuebler, Smitha Krishnamurthi, Benjamin Tan, Eileen M. O'Reilly, Anthony F. Shields, Shuji Ogino, Charles S. Fuchs, Jeffrey A. Meyerhardt, Jonathan A. Nowak. Prognostic and predictive role of T cell infiltration in stage III colon cancer (CC) treated with celecoxib: CALGB/SWOG 80702 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5858.
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 | 0.000 | 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 | 0.000 |
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 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".