Deep learning-based quantification of eosinophils and lymphocytes shows complementary prognostic effects in colorectal cancer patients
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
The immune microenvironment of colorectal cancer is a major component of the disease and influences not only tumor progression and patient outcome but also therapy response. Expanding on existing studies which have explored the prognostic value of the adaptive immune response with lymphocytes, our study integrates innate immune cells, specifically eosinophils, in a combined analysis. To evaluate the prognostic significance of eosinophils within the context of lymphocyte infiltration, we analyzed a large collective of 1625 colorectal cancer cases from four different centers. For this purpose, we develop an automatic deep learning pipeline for quantification of these immune cells directly from hematoxylin and eosin-stained whole slide images. Our analysis shows eosinophils in the tumor front (EosF) as independent prognostic factor (HR = 0.70, 95%CI = 0.55 - 0.90, p = 0.005), particularly also in microsatellite instability (MSI) cases (HR = 0.32, 95%CI = 0.14 - 0.74, p = 0.008). Moreover, EosF and intraepithelial lymphocytes (IELs) counts are statistically independent and provide additive prognostic information (EosF: HR = 0.71, 95%CI = 0.55 - 0.90, p = 0.005, IELs HR = 0.59, 95%CI = 0.35 - 0.99, p = 0.047). Our study demonstrates that eosinophils are an independent prognostic factor, which can be automatically quantified, underscoring its high potential for translation to a diagnostic biomarker. Moreover, our work could pave the way towards an integrated immune score directly from hematoxylin and eosin-stained sections.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| 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) | 0.000 | 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 it