Abstract 7433: Predicting overall survival (OS) differences using longitudinal AI-driven imaging-based prognostication (IPRO) in patients with advanced non-small cell lung cancer (aNSCLC)
Bibliographic record
Abstract
Abstract Background: Early prediction of overall survival benefit associated with cancer treatments remains a key challenge in clinical trials and patient care. Standardized measures such as the Response Evaluation Criteria in Solid Tumours (RECIST) v1.1 are used in clinical trials to define tumour response and progression, but are subject to inter- and intra-observer variability, and are infrequently used in day to day clinical practice. Imaging-based prognostication delta (IPRO-Δ) is an artificial intelligence system trained on longitudinal changes in rich features extracted from serial real-world imaging data, automatically predicting survival from pairs of baseline (BL) and follow up (FU) thoracic computed tomography (CT) scans. Methods: IPRO-Δ was evaluated in an external real-world dataset of 194 aNSCLC patients treated with first line systemic therapy at 17 cancer centers, with pretreatment BL and FU (8±4 weeks after treatment initiation) scans. IPRO-Δ generated survival predictions at week 8 for each patient. Kaplan-Meier analyses and hazard ratios (HR) evaluated the association of IPRO-Δ quartiles with OS, with the highest quartile representing progressive disease, the lowest treatment response, and the middle stable disease. Results: The cohort included 19 stage IIIB (9.8%) and 175 stage IV (90.2%) aNSCLC patients, with 94 females (48.5%) and a median age of 64 years. Most tumours were adenocarcinoma (n = 147, 75.8%), followed by squamous cell (n = 18, 9.3%), large cell (n = 5, 2.6%) and other histologies (n = 24, 12.4%). Table 1 shows survival stratification amongst IPRO-Δ quartiles for the overall population and the subgroup of EGFR+ patients. Conclusion: IPRO-Δ successfully predicted survival differences amongst aNSCLC patients using BL and week-8 CT scans. Future work will compare IPRO-Δ to RECIST v1.1, potentially leading to IPRO-Δ's use as a surrogate endpoint in aNSCLC trials. Citation Format: Mohammed A. Alvi, John Riskas, Shahid A. Haider, Vignesh Sivan, Oleksandra Samorodova, Jay Hennesy, Duoaud Shah, Felix Baldauf-Lenschen, Marina Salluzzi, Ronald Bridges, Omar F. Khan. Predicting overall survival (OS) differences using longitudinal AI-driven imaging-based prognostication (IPRO) in patients with advanced non-small cell lung cancer (aNSCLC) [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 7433.
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 | 0.000 |
| 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) | 0.001 | 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".