PD-L1 Molecular Imaging-Based Evaluation of Cancer Immunotherapy Efficacy: Technological Advances and Clinical Translation Challenges

Abstract

The precision of cancer immunotherapy urgently requires real-time, non-invasive efficacy evaluation tools. Imaging molecular technology, through the combination of molecular probes and AI algorithms, provides a new direction for visualizing the immune microenvironment and predicting treatment response. Currently, PET/SPECT imaging based on PD-L1 specific probes, MRI targeted nano-contrast agents, and optical real-time tracking technology have achieved dynamic monitoring of immune checkpoints. AI-driven radiomics models further mine prognostic markers from multi-modal data, promoting personalized treatment decisions. However, existing technologies still face core challenges such as lack of standardization, insufficient interpretability of algorithms, and barriers to clinical translation. This article systematically analyzes the principles, advantages, and limitations of imaging molecular technology, revealing the necessity of multi-modal image fusion and AI-coordinated optimization: PET-MRI combined imaging can break through the sensitivity and resolution limitations of a single modality, federated learning algorithms can effectively solve multi-center data heterogeneity, and the development of portable equipment is expected to lower the technical threshold. Studies have shown that these technologies can construct a "imaging-treatment-feedback" closed-loop system, significantly improving the accuracy of efficacy prediction, but its clinical promotion is still limited by insufficient sample size and long-term safety controversies. Future research needs to focus on establishing standardized evaluation systems, developing interpretable AI models, and exploring the linkage mechanism between gene editing and molecular imaging to promote the comprehensive transformation of cancer immunotherapy from the laboratory to the clinic.