Effectiveness of Artificial Intelligence Models in Predicting Lung Cancer Recurrence: A Gene Biomarker-Driven Review
Bibliographic record
Abstract
BACKGROUND/OBJECTIVES: Lung cancer recurrence, particularly in NSCLC, remains a major challenge, with 30-70% of patients relapsing post-treatment. Traditional predictors like TNM staging and histopathology fail to account for tumor heterogeneity and immune dynamics. This review evaluates AI models integrating gene biomarkers (TP53, KRAS, FOXP3, PD-L1, and CD8) to enhance the recurrence prediction and improve the personalized risk stratification. METHODS: Following the PRISMA guidelines, we systematically reviewed AI-driven recurrence prediction models for lung cancer, focusing on genomic biomarkers. Studies were selected based on predefined criteria, emphasizing AI/ML approaches integrating gene expression, radiomics, and clinical data. Data extraction covered the study design, AI algorithms (e.g., neural networks, SVM, and gradient boosting), performance metrics (AUC and sensitivity), and clinical applicability. Two reviewers independently screened and assessed studies to ensure accuracy and minimize bias. RESULTS: A literature analysis of 18 studies (2019-2024) from 14 countries, covering 4861 NSCLC and small cell lung cancer patients, showed that AI models outperformed conventional methods. AI achieved AUCs of 0.73-0.92 compared to 0.61 for TNM staging. Multi-modal approaches integrating gene expression (PDIA3 and MYH11), radiomics, and clinical data improved accuracy, with SVM-based models reaching a 92% AUC. Key predictors included immune-related signatures (e.g., tumor-infiltrating NK cells and PD-L1 expression) and pathway alterations (NF-κB and JAK-STAT). However, small cohorts (41-1348 patients), data heterogeneity, and limited external validation remained challenges. CONCLUSIONS: AI-driven models hold potential for recurrence prediction and guiding adjuvant therapies in high-risk NSCLC patients. Expanding multi-institutional datasets, standardizing validation, and improving clinical integration are crucial for real-world adoption. Optimizing biomarker panels and using AI trustworthily and ethically could enhance precision oncology, enabling early, tailored interventions to reduce mortality.
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 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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.003 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".