First-in-human study of an implantable microdevice for personalized drug testing in colorectal liver metastases.

TPS278 Background: Colorectal liver metastases (CRLM) are the leading cause of death in metastatic colorectal cancer. Standard perioperative chemotherapy has not improved overall survival, underscoring the urgent need for novel therapeutic strategies. Implantable microdevices (IMDs) enable in vivo, patient-specific drug sensitivity testing using microdoses of multiple agents, providing rapid pharmacodynamic readouts without systemic toxicity. IMDs have been evaluated in other solid tumors but never in CRLM. This investigator–initiated clinical trial aims to establish the safety and feasibility of IMD implantation and retrieval during hepatectomy for CRLM, and to generate individualized drug sensitivity profiles. Methods: This single–institution, first–in–human pilot study (NCT registered but # pending) will enroll 10 adults with CRLM ≥2 cm in an accessible location undergoing curative-intent hepatectomy. Up to four IMDs, each containing 22 reservoirs, will be placed percutaneously 3 days prior to metastasectomy. Reservoirs will be loaded with cytotoxic and immunomodulatory agents formulated in polyethylene glycol (PEG), including 5-fluorouracil (5-FU), FOLFOXFIRI (5-FU, oxaliplatin, leucovorin, irinotecan), bevacizumab, Fc–enhanced multifunctional anti–CTLA–4 antibody (botensilimab, AGEN1181), anti–PD–1 antibody (balstilimab), and anti–CD137 antibody (AGEN2373). PEG–only reservoirs serve as controls. Doxorubicin autofluorescence will allow orientation and confirm local drug diffusion. All drugs are delivered at microdose levels (∼1/100,000th systemic dose) with diffusion limited to a radius from each reservoir of ∼600–800μm, minimizing systemic exposure. IMDs will be retrieved en bloc with the surgical specimen and analyzed for histopathology, immunofluorescence, spatial transcriptomics, and proteomics. The primary endpoint is safety of IMD placement and retrieval, assessed by CTCAE v5.0. Secondary endpoints include feasibility, defined as histopathologic analysis of ≥50% of reservoirs, and creation of individualized drug sensitivity profiles. Exploratory endpoints include immune-modulatory effects and transcriptomic and proteomic pathway profiling. This trial is currently open to enrollment but has not yet enrolled its first patient. Conclusions: This is the first–in–human study of IMDs in CRLM, establishing a novel in vivo platform for rapid, patient-specific therapeutic profiling. If successful, IMDs could redefine therapeutic development in CRLM by enabling real-time, patient-specific drug testing to guide precision treatment.