697: Precision AI Medicine Platform for Analgesic Selection and Dose Optimization in Cancer Pain

Topic: Experimental (Non-clinical) Human Studies
Subtopic: Pain Category (ICD-11), Chronic cancer related pain, Omics (Proteomics/Metabolomics/Transcriptomics/Genomics), Artificial Intelligence/Machine Learning, Molecular markers

Background and Aims: Cancer pain remains highly heterogeneous, and opioid response varies substantially between patients. Current prescribing is largely iterative and based on empirical reassessment rather than patient-specific biological prediction. "PainRx" is a proposed translational platform designed to address this gap by combining patient-derived biological profiling with artificial intelligence–enabled drug prioritization and adaptive clinical dose optimization. The aim of this project is to repurpose the logic of functional precision medicine for cancer pain by developing a workflow that can use patient-derived samples to generate a biologic response profile, map that profile to candidate analgesic drugs or combinations, and then refine treatment through individualized clinical optimization. This abstract presents the scientific rationale, platform architecture, and Phase 1 methodology components of PainRx.

Methods: Phase 1 is designed as a biological foundation study. Pre-treatment patient-derived materials, including blood and cancer-related tissue where available, will be used to generate multimodal biologic profiles relevant to opioid and nociceptive response. Planned assays include receptor-related profiling, cyclic adenosine monophosphate inhibition, beta-arrestin signaling, patch-clamp electrophysiology, and inflammatory mediator assessment. These data will be used to build a profile-to-drug mapping pipeline modeled on the proprietary platform, with the objective of prioritizing candidate analgesic drugs or combinations for a given biologic profile. A downstream clinical layer will incorporate the proprietary platform to support individualized dose and dose-ratio optimization following initial treatment selection through structured early titration and longitudinal clinical reassessment.

Results: Current project outputs include a defined translational workflow specifying candidate biospecimens, biologic profiling domains, assay modalities, and the proposed artificial intelligence pipeline linking biologic profiles to analgesic prioritization. The platform is designed to generate a clinician-facing precision analgesia report ranking candidate drugs or combinations for structured clinical validation, followed by patient-specific optimization of dose and dose ratio. Phase 1 feasibility endpoints include successful generation of multimodal biologic profiles from patient-derived materials, development of a profile prediction pipeline from raw assay data, and construction of an initial biologic profile-to-drug mapping framework.

Conclusions: PainRx proposes a precision-medicine approach to cancer pain by moving from empirical analgesic selection toward biologically informed prioritization and adaptive clinical optimization. The Optim.AI™ proprietary platform is intended to support rapid prioritization of candidate drugs or combinations from patient-derived biologic profiles, while the CURATE.AI™ proprietary platform is intended to personalize treatment after initiation using individual response data. The project will establish a coherent translational framework for evaluating whether patient-specific biologic profiling can improve analgesic decision support in cancer pain.

References: The μ-Opioid Receptor in Cancer and Its Role in Perineural Invasion: A Review. Advanced Biology. 2022;2200020. doi:10.1002/adbi.202200020.
Bidlack JM. Detection and Function of Opioid Receptors on Cells from the Immune System. Clinical and Diagnostic Laboratory Immunology. 2000;7(5):719-723. doi:10.1128/CDLI.7.5.719-723.2000.
Artificial Intelligence and Machine Learning in Cancer Pain: A Systematic Review. 2024. PMCID: PMC11534522.
Froicu EM, Onicescu OM, Creangă-Murariu I, Dascălu C, Gafton B, Afrăsânie VA, Alexa-Stratulat T, Marinca MV, Pușcașu DM, Miron L, et al. Modeling Pain Dynamics and Opioid Response in Oncology Inpatients: A Retrospective Study with Application to AI-Guided Analgesic Strategies in Colorectal Cancer. Medicina. 2025;61(10):1741. doi:10.3390/medicina61101741.
Machine Learning Predicting Acute Pain and Opioid Dose in Radiation Treated Oropharyngeal Cancer Patients. Frontiers in Pain Research. 2025. doi:10.3389/fpain.2025.1567632.
A Predictive Model for Pain Response Following Radiotherapy for Treatment of Spinal Metastases. Scientific Reports. 2021. doi:10.1038/s41598-021-92363-0.
Therapeutic Approaches to Cancer Pain. In: Holland-Frei Cancer Medicine. NCBI Bookshelf. National Center for Biotechnology Information.
Mestdagh F, Steyaert A, Lavand’homme P. Cancer Pain Management: A Narrative Review of Current Concepts, Strategies, and Techniques. Current Oncology. 2023;30(7):500. doi:10.3390/curroncol30070500. PMCID: PMC10378332.
Molecular Basis of Cancer Pain Management: An Updated Review. 2019. PMCID: PMC6780247.
Rashid MBM, Toh TB, Hooi L, Silva A, Tan PF, Teh J, et al. Optimizing Drug Combinations Against Multiple Myeloma Using a Quadratic Phenotypic Optimization Platform (QPOP). Science Translational Medicine. 2018;10(453):eaan0941. doi:10.1126/scitranslmed.aan0941.
Blasiak A, et al. Personalized Dose Selection Platform for Patients With Solid Tumors in the PRECISE CURATE.AI Feasibility Trial. npj Precision Oncology. 2025. doi:10.1038/s41698-025-00835-7.

Presenting Author: Muhammad Kumail
Submitting Author: Yacine HADJIAT

Senior Author: Yacine HADJIAT

IASP 2026 Global Year theme of Neuropathic Pain: No

• Rajan Radhakrishnan: PhD - MBRU University