Clinical Report: Response from Authors on AI-Driven Radiomics in NSCLC

Background

The integration of AI-driven radiomics into clinical practice is being explored for its ability to predict treatment responses in NSCLC, which is important for managing neoadjuvant immunotherapy. Understanding the performance of AI models compared to traditional criteria is necessary for patient outcomes and treatment strategies.

Data Highlights

No numerical data presented in the source material.

Key Findings

• The total of 72 test assessments in the meta-analysis refers to assessments rather than unique participants.
• AI demonstrated greater sensitivity for predicting pathological complete response (pCR) compared to conventional criteria, which had greater specificity.
• In a direct comparison, Deng et al. reported a sensitivity of 100.0% and specificity of 94.1% for their AI model, compared to lower values for RECIST 1.1 and PERCIST.
• The pooled analysis included only validation cohorts, limiting the generalizability of findings to an intention-to-treat population.
• Future studies should clarify whether pathological response pertains to the primary tumor, nodes, or both, and include nonsurgical patients.
• Implementation of AI models requires multicenter validation and standardized protocols.

Clinical Implications

Clinicians should consider the sensitivity-specificity trade-off when utilizing AI-driven radiomics for predicting treatment responses in NSCLC.

Conclusion

The authors note that AI-driven radiomics has shown diagnostic performance in predicting pathological responses in NSCLC.

Related Resources & Content

1. Kan W, Pei H, Zhang M, J Med Internet Res, 2026 -- Same-patient, same–time point evidence for radiomics benchmarking
2. Jiang Z, Xu Y, Jia S, Liu H, J Med Internet Res, 2026 -- Radiomics-based AI for predicting neoadjuvant immunochemotherapy pathological response in non-small cell lung cancer: systematic review and meta-analysis
3. Deng Y, Zhang X, Hu F, Lan X, Eur J Nucl Med Mol Imaging, 2025 -- Quantitative 18F-FDG PET/CT model for predicting pathological complete response to neoadjuvant immunochemotherapy in NSCLC
4. Forde PM, Spicer J, Lu S, et al., N Engl J Med, 2022 -- Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer
5. The ASCO Post — Can Artificial Intelligence Predict Treatment Response and Outcomes in Breast Cancer?
6. ASCO Publications — Cancer patients' messages about radiology/pathology reports: Insights for AI.
7. ASCO Publications — Artificial intelligence to accurately identify breast cancer patients with a pathologic complete response for omission of surgery after neoadjuvant systemic therapy: An international multicenter analysis.
8. Overall Survival with Neoadjuvant Nivolumab plus Chemotherapy in Lung Cancer
9. Radiomics-Based AI for Predicting Neoadjuvant Immunochemotherapy Pathological Response in NSCLC
10. IASLC MULTIDISCIPLINARY RECOMMENDATIONS FOR PATHOLOGIC ASSESSMENT OF LUNG CANCER RESECTION SPECIMENS FOLLOWING NEOADJUVANT THERAPY - PMC