A70 MOLECULAR ARCHETYPE HETEROGENEITY IN ULCERATIVE COLITIS BIOPSIES
Bibliographic record
Abstract
The assessment of patients with ulcerative colitis (UC) remains a challenge despite rapidly evolving medical therapy. Current assessment with endoscopic scoring and histopathology lacks the granularity to correlate strongly with response to therapy. Utilizing a microarray based system for colonic epithelial assessment we looked for stratification of endoscopically similar UC biopsies using molecularly heterogeneous archetypes. Molecular data from 71 UC biopsies (from 61 patients) was obtained from microarray analysis. The top 300 genes correlating with the endoscopic Mayo score (2/3 vs 0/1) were used for an unsupervised analytical method called archetypal analysis, and grouped the biopsies into three distinct clusters. Logistic regression modeling was used to compare archetype scores or cluster membership to endoscopic Mayo score and PC1 in predicting mucosal healing. A contingency table was generated to show evidence of mucosal healing within each of the archetype clusters. We then assessed a subset of biopsies from the original set of 71 (selected due to the availability of biopsies before and after therapy, all on TNF therapy) plus one IBDU case. Patients were classified as ‘responders’ (Mayo 2/3 score to Mayo 0/1 score by last obtained biopsy or scope) or ‘non-responders’ (Mayo 2/3 score that did not decrease to 0/1 by last obtained biopsy or scope). We found three unique groups of biopsies using archetypal analysis (A1: lack of inflammation, A2: inflammation and response to wounding, A3: inflammation). Logistic regression showed that the only models with statistically significant predictive value (p-value < 0.05) were those that contained archetype scores or cluster membership. Response rates differed between archetype clusters with statistical significance (Table 1), while the mayo score distribution within these clusters was not statistically different. In our subset of serial biopsies, the majority of initial biopsies were found to have an A2 archetype, moving to an A1 archetype in follow-up that mirrored response to treatment (Figure 1). This archetypal analysis suggests there is potentially important heterogeneity in UC biopsies that is not accessible by endoscopic Mayo score. Serial biopsies showed dynamic shifts in the archetype composition between biopsies. This may be a useful tool for both initailly prognosticating patients and assessing response to treatment over time with increased granularity and reliablity. UC Patient Response to Therapy Assessed by Initial Archetype Cluster All cases had a Mayo score of 2–3 on initial endoscopy. *score of 0–1 on follow up endoscopy. Pearson’s Chi-Squared = 7.222, df = 2, p = 0.027 Figure 1. Stacked and group bar charts showing archetype composition of biopsies in 10 UC patients taken over time (2–4 serial biopsies/patient). None
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) | 0.001 | 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".