Extracellular vesicle-based biomarker assay for the detection of early-stage ovarian cancer.
Bibliographic record
Abstract
5542 Background: Detection of cancer with improved discrimination compared to current blood tests could be achieved using an approach that assesses extracellular vesicles (EVs). This approach should have high sensitivity (se) because of EVs abundance in blood and high specificity (sp) by assaying EVs with multiple cancer-related protein and glycosylation epitopes (PGEs) co-localized on their surfaces. We are developing a platform technology that detects multiple cancer-related PGEs co-localized on the same EV using immunoaffinity-capture and proximity-ligation qPCR. In this study, we compare the performance of this technology vs plasma CA125 for correctly categorizing early-stage high-grade serous ovarian cancer (HGSOC) vs healthy/benign ovarian tumors (OT). Methods: We evaluated our EV-based platform technology using 7 PGE combinations to discriminate HGSOC from benign adnexal masses. We first derived a prediction model on a retrospectively collected cohort of 42 HGSOC and 26 benign OT samples from 2 commercial vendors and 24 healthy controls (HC) using a machine-learning algorithm. We validated this model on an independent cohort [89 HGSOC: Stage I (17), II (35), III (37); 192 benign OT] from university-associated biobanks and 124 HC. We also assessed the assay’s performance in plasma from 87 women with off-target cancers and 42 women with inflammatory conditions from commercial vendors. For each sample, we also measured CA125 levels using a commercial ELISA. Results: The prediction model distinguishes HGSOC from benign and HC with an AUC of 0.965 (95% CI 0.93-0.99), with 89.9% (0.82-0.95) se at 98% sp. For stage I/II HGSOC, the model achieves an AUC of 0.942 (0.9-0.99), with 84.6% (0.72-0.93) se at 98% sp. By comparison, CA125 achieves an AUC of 0.875 (0.81-0.94) and 44.2% (0.3-0.59) se at 98% sp. Direct comparison of CA125 and our model shows a significant difference at 98% sp for both all and stage I/II HGSOC (McNemar p-val < 0.001). When comparing HGSOC to HC, there is no significant difference between our model and CA125 (p-val = 1.0). There is a significant difference when comparing patients with all stage and stage I/II HGSOC to patients with benign OT (p-val < 0.001). Our assay had 1 false positive and CA125 had 3 false positives out of 42 inflammatory cases. Conclusions: These preliminary data suggest our platform technology for detecting PGEs co-localized on individual EVs may detect all stages of HGSOC from plasma with high se at a very high sp. Our assay may improve on CA125 by distinguishing stage I/II HGSOC from benign OT and could have clinical utility for both early detection and surgical referral recommendation for benign and malignant OT. While the diverse cohorts in this study may present challenges in interpretation, the reproducibility in an independent cohort is encouraging and supports further investigation using cases and controls from well-defined cohort studies.
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 | 0.001 | 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".