PT - JOURNAL ARTICLE AU - Tiffany Chan AU - Edward O'Neill AU - Bart Cornelissen TI - High-throughput screening and bioinformatics analysis of 2,000 <sup>177</sup>Lu-PSMA and Radiotherapy + drug combinations DP - 2021 May 01 TA - Journal of Nuclear Medicine PG - 94--94 VI - 62 IP - supplement 1 4099 - http://jnm.snmjournals.org/content/62/supplement_1/94.short 4100 - http://jnm.snmjournals.org/content/62/supplement_1/94.full SO - J Nucl Med2021 May 01; 62 AB - 94Introduction: Despite the promising results obtained thus far with 177Lu-PSMA for the treatment of advanced metastatic prostate cancer, there remains room for improvement, particularly in patients with poorly defined and heterogeneous 177Lu-PSMA uptake. To improve the treatment outcomes of 177Lu-PSMA therapy, there has been rising interest in the use of combination therapies. While multiple 177Lu-PSMA combination therapies are currently being investigated in preclinical and clinical trials, the majority of these are based on established radiosensitisers of external beam radiotherapy (EBRT) despite the known differences in the radiobiology of EBRT and targeted radionuclide therapies. Here, we combined 2,000 drugs with 177Lu-PSMA and EBRT in a high-throughput clonogenic screen and developed a bioinformatics approach based on network theory to further our insight into the radiosensitisation mechanisms involved. Methods: A high-throughput clonogenic screen was conducted in which 2,000 drugs were combined with equitoxic doses of 177Lu-PSMA or EBRT therapy in 22rv1 human prostate carcinoma cells. Drugs investigated in our screen include 1,600 FDA-approved drugs of varying drug classes and 400 oncology-focused drugs. Combination indices for all combinations were calculated from the drug dose-response curves obtained. To probe the combinations further, a network analysis approach was developed. Briefly, the targets of the tested drugs and their downstream protein-protein interactions were extracted from DrugBank and STRING databases respectively and integrated to build networks using an in-house MATLAB script. Networks were visualised on Cytoscape, clustered using Markov clustering and assigned to specific biological pathways. Results: Our screen identified a number of synergistic 177Lu-PSMA + drug combinations that were previously unknown. Clear differences in combination indices were observed between several 177Lu-PSMA + drug versus EBRT + drug combinations, suggesting different radiosensitisation mechanisms at play (Figure 1a). To further our understanding of the biological pathways involved in 177Lu-PSMA and EBRT radiosensitisation, we developed a methodology to build, visualise and analyse drug-target networks (Figure 1b). Through network analysis, we have been able to begin identifying molecular pathways that are likely to lead to effective radiosensitisation when combined with 177Lu-PSMA, as well as those that lead to antagonistic interactions, allowing for more informed design of combination therapies. Conclusions: Our high-throughput screen identified several novel synergistic 177Lu-PSMA combination therapies and highlighted that effective radiosensitisers of 177Lu-PSMA are not always effective radiosensitisers of EBRT and vice versa. Through the application of network analysis, we have been able to further our understanding of 177Lu-PSMA and EBRT radiosensitisation mechanisms. Acknowledgements This work was supported by funding from Prostate Cancer Research, Cancer Research UK and the UK Medical Research Council. Figure 1: a) Combination indices of 2,000 drugs combined with 177Lu-PSMA and EBRT determined using a high-throughput clonogenic screen. b) Drug-target networks for the investigated combinations were created (left) and clustered to identify implicated biological pathways (right). Nodes and edges in red represent antagonistic interactions, those in green represent synergistic interactions and those in yellow represent additive interactions.