Immune checkpoint inhibitors (ICIs) have obtained durable responses in many cancers, reviving a great hope for cancer patients. However, immunotherapies are limited at the moment to a minority of patients because of potential toxicity. Immune checkpoint cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) plays an important role in hindering anticancer immunity. Small chemical inhibitors generally do not degrade their targets. The generation of such molecules could bring new knowledge on the immune system response in anticancer and improve current therapies. To achieve this purpose in this work with our artificial intelligence (AI)-assisted we virtually screened among 10 million compounds to identify 72 top scores of primary hit compounds targeting an innovative binding site of CTLA-4. We have narrowed 72 compounds to several top hits with four biochemical, cellular and two in vivo prophylactic or tumor-established models. We determined the binding affinity between purified CTLA-4 protein and the compounds by determining the delta thermal shift and SRM. We tested these compounds in disrupting the binding activity between human CTLA-4 and CD80/CD86 in a cell-free system using an AlphaLISA assay (Amplified Luminescent Proximity Homogeneous Assay). We tested the potency of these compounds in altering the interaction of CTLA-4 with its ligands in a reporter cell system with effector Jurkat cells carrying a luciferase gene and antigen presenting Raji cells. Compounds binding to CTLA-4 blocked the interaction of CTLA-4 with its ligands and resulted in increased luminescent signal. We determined whether these and additional hits affect the expression and/or membrane distribution of CTLA-4 using fluorescence-activated cell sorting (FACS). In conclusion, compounds did not degrade CTLA-4 on the membrane, while acting as inhibitors; they were able to prevent cancer prophylatically and inhibited tumor growth in vivo. This application has three innovation elements. First, we will use small compounds instead of antibodies to inhibit CTLA-4 signaling for cancer therapy. Second, this AI-assisted drug design and development will represent a breakthrough in deep learning and biomedical advances. Third, the compounds can work as inhibitors and not degraders potentially preventing toxicity, while maintaining their tumor inhibitory ability.

Progress on novel and more powerful CTLA-4 inhibitors for the treatment of cancer

ACS Fall 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.