Photoimmunotherapy Retains Its Anti-Tumor Efficacy with Increasing Stromal Content in Heterotypic Pancreatic Cancer Spheroids.

Pancreatic ductal adenocarcinoma (PDAC) is an aggressive disease characterized by increased levels of desmoplasia that contribute to reduced drug delivery and poor treatment outcomes. In PDAC, the stromal content can account for up to 90% of the total tumor volume. The complex interplay between stromal components, including pancreatic cancer-associated fibroblasts (PCAFs), and PDAC cells in the tumor microenvironment has a significant impact on the prognoses and thus needs to be recapitulated in vitro when evaluating various treatment strategies. This study is a systematic evaluation of photodynamic therapy (PDT) in 3D heterotypic coculture models of PDAC with varying ratios of patient-derived PCAFs that simulate heterogeneous PDAC tumors with increasing stromal content. The efficacy of antibody-targeted PDT (photoimmunotherapy; PIT) using cetuximab (a clinically approved anti-EGFR antibody) photoimmunoconjugates (PICs) of a benzoporphyrin derivative (BPD) is contrasted with that of liposomal BPD (Visudyne), which is currently in clinical trials for PDT of PDAC. We demonstrate that both Visudyne-PDT and PIT were effective in heterotypic PDAC 3D spheroids with a low stromal content. However, as the stromal content increases above 50% in the 3D spheroids, the efficacy of Visudyne-PDT is reduced by up to 10-fold, while PIT retains its efficacy. PIT was found to be 10-, 19-, and 14-fold more phototoxic in spheroids with 50, 75, and 90% PCAFs, respectively, as compared to Visudyne-PDT. This marked difference in efficacy is attributed to the ability of PICs to penetrate and distribute homogeneously within spheroids with a higher stromal content and the mechanistically different modes of action of the two formulations. This study thus demonstrates how the stromal content in PDAC spheroids directly impacts their responsiveness to PDT and proposes PIT to be a highly suited treatment option for desmoplastic tumors with particularly high degrees of stromal content.

3D molecular generative framework for interaction-guided drug design.

Deep generative modeling has a strong potential to accelerate drug design. However, existing generative models often face challenges in generalization due to limited data, leading to less innovative designs with often unfavorable interactions for unseen target proteins. To address these issues, we propose an interaction-aware 3D molecular generative framework that enables interaction-guided drug design inside target binding pockets. By leveraging universal patterns of protein-ligand interactions as prior knowledge, our model can achieve high generalizability with limited experimental data. Its performance has been comprehensively assessed by analyzing generated ligands for unseen targets in terms of binding pose stability, affinity, geometric patterns, diversity, and novelty. Moreover, the effective design of potential mutant-selective inhibitors demonstrates the applicability of our approach to structure-based drug design.

PIGNet: a physics-informed deep learning model toward generalized drug-target interaction predictions.

Recently, deep neural network (DNN)-based drug-target interaction (DTI) models were highlighted for their high accuracy with affordable computational costs. Yet, the models' insufficient generalization remains a challenging problem in the practice of in silico drug discovery. We propose two key strategies to enhance generalization in the DTI model. The first is to predict the atom-atom pairwise interactions via physics-informed equations parameterized with neural networks and provides the total binding affinity of a protein-ligand complex as their sum. We further improved the model generalization by augmenting a broader range of binding poses and ligands to training data. We validated our model, PIGNet, in the comparative assessment of scoring functions (CASF) 2016, demonstrating the outperforming docking and screening powers than previous methods. Our physics-informing strategy also enables the interpretation of predicted affinities by visualizing the contribution of ligand substructures, providing insights for further ligand optimization.