Therapy-Resistant Residual Cancer Cell Dependencies Mapped

Why It Matters

Targeting persister cells could dramatically reduce residual disease and tumor relapse, addressing a major obstacle in precision oncology. The platform’s scalable, data‑rich approach enables faster validation of combination strategies, potentially shortening the path to clinical approval.

Key Takeaways

• Robotic ResMap screens 10,000 experiments across 384‑well plates
• Nine drug candidates showed consistent anti‑persister activity
• Persister cells share vulnerabilities across EGFR and KRAS lung models
• Platform integrates automation, machine learning, and persistence metric
• Dataset will support community‑wide combination‑therapy design

Pulse Analysis

Persister cells—rare, drug‑tolerant subpopulations that survive initial cancer therapy—have long frustrated oncologists because they seed relapse while evading conventional drug screens. Their scarcity and phenotypic plasticity make them difficult to isolate, limiting insight into the pathways that sustain them. As a result, despite a decade of research, no persister‑directed therapy has reached the clinic, leaving a critical gap in achieving durable responses for oncogene‑driven cancers.

UCSF’s ResMap platform tackles this challenge by marrying robotics with advanced analytics. Thousands of micro‑tumors are arrayed in 384‑well plates, where a robotic arm deposits sequential doses of standard therapy followed by candidate persister drugs. Integrated machine‑learning algorithms normalize imaging data and calculate a persistence‑specific metric, enabling rapid, reproducible assessment of drug efficacy across multiple genetic backgrounds and oxygen levels. In a single study, the system evaluated 94 compounds, confirming nine that consistently weakened persisters in both EGFR‑mutant and KRAS‑mutant lung models.

The implications extend beyond a single dataset. By revealing shared survival pathways, ResMap provides a blueprint for rational combination regimens that pair oncogene inhibitors with persister‑targeted agents, potentially preventing residual disease before it manifests clinically. The open‑access nature of the platform invites broader research collaboration, accelerating validation of new targets across tumor types. As the system scales to additional cancers and treatment contexts, it could reshape drug discovery pipelines, shortening timelines from bench to bedside and improving long‑term outcomes for patients facing aggressive malignancies.