Optimal Methods to Characterize ADC Resistance in Solid Tumors and Identify Clinically Useful Biomarkers

International study that will evaluate the association of prespecified biomarkers with resistance to Antibody-drug conjugates (ADCs), a type of targeted cancer treatment currently used in clinical practice for treating different tumor types.

Over the past five years, antibody-drug conjugates (ADCs) have dramatically improved survival in solid and hematologic malignancies. Among 14 ADCs approved worldwide, nine are now available in Europe, and over 370 others are in clinical development. This expanding landscape indicates that ADCs could soon replace conventional chemotherapy across multiple tumor types. Given this rapid evolution, clinicians will need to select the most suitable ADC for each patient, considering tumor biology, microenvironment (TME) and patient-specific factors. Yet, despite remarkable efficacy, resistance to ADCs eventually arises. Understanding resistance mechanisms is therefore essential to guide therapeutic sequencing and optimize next-generation ADCs. ADCs are complex molecules combining an antibody, a linker, and a cytotoxic payload. Their activity depends on factors such as antigen expression, internalization, linker stability, and payload sensitivity. Resistance can result from altered vascular perfusion, antigen downregulation, defective internalization or trafficking, impaired linker cleavage, drug efflux, or payload target modifications. These multifactorial processes differ from those driving resistance to traditional chemotherapies. Existing preclinical and clinical tools (Patient-Derived Xenograft(PDX)/Cell-line-Derived Xenograft (CDX) models, standard imaging, Immunohistochemistry (IHC), genomic profiling) fail to capture this complexity or predict ADC efficacy and resistance. Furthermore, ADCs often cause significant toxicities-on-target or off-target-affecting the ocular surface, skin, lungs, and peripheral nerves. Patient factors such as age, comorbidities, and weight influence these events. Understanding the determinants of toxicity is critical to maintain quality of life and treatment adherence. The OASIS program aims to identify predictive biomarkers of ADC response and toxicity to enable personalized ADC selection and toxicity prevention. This multicenter study will integrate advanced technologies-digital pathology, liquid biopsy, and Patient-derived organoids (PDOs)-to generate comprehensive biological and clinical data. Using these datasets, a multimodal machine-learning model (OASIS Multiparametric Score) will be developed to predict both efficacy and key toxicities of ADCs. The project will prospectively include patients receiving ADCs in standard practice, with longitudinal tumor and blood sampling to investigate biomarkers of resistance and toxicity. In parallel, preclinical models derived from patient tumors will explore resistance mechanisms and screen ADC sensitivity. A retrospective cohort of patients previously treated with ADCs will first be analyzed to prioritize biomarker candidates based on published data and prior findings. From this, five binary biomarkers will be selected for the primary objective. Combined with prospective data, this retrospective work will expand the translational biobank and support the construction of the OASIS score.