California Initiative to Advance Precision Medicine Integrated Machine-Learning Platform to Inform Precision Therapy in Breast Cancer Patients

This project connnects a community of researchers, clinicians, patients and community advocates, who join forces as Project CELSUS to apply precision medicine genomic profiling to find new treatments and create a more precise and rational system to match TNBC tumors and existing cancer treatments. The project draws its name from Aulus Cornelius Celsus, an early pioneer of evidenced-based medicine in the 1st century A.D.

Since launching on August 14, 2019, the Project CELSUS team has made important strides towards developing project infrastructure and bringing together stakeholders from 14 separate institutions.

The Project CELSUS team is building on its 2020 accomplishments by improving their models, refining their system for testing the effectiveness of medications, validating their results in additional models, and strengthening engagement with partnering community hospitals. The work will support the ultimate goal of optimizing treatments for Hispanic women with TNBC.

Analyzing Existing Data

As an early action, Project CELSUS collected and studied information about clinical breast tumors from several large, publicly available datasets including The Cancer Genome Atlas, Cancer Cell Line Encyclopedia, and Gene Expression Omnibus. The UCSD team developed a prototype of a database that combines information from these datasets with additional clinical data previously collected by CELSUS team members. Project CELSUS has collaborated with UC San Diego Health Information Services and Amazon Web Services to develop a secure, HIPAA-compliant, cloud-based data repository to store and track patient samples.

Disease Model for TNBC

Project CELSUS also developed a disease model for TNBC that could help oncologists predict how a tumor will react to medication, or a combination of medications. The team used the data from their prototype database to design a model system classifying TNBC tumors into subtypes by identifying patterns within tumors' molecular signatures.

The researchers identified molecular signatures associated with a subset of particularly aggressive TNBC tumors and observed a potentially significant association between a patient's ethnicity (in this case African American and Hispanic), their Body Mass Index, and tumor aggressiveness. The researchers will continue to refine this classification system.

Of note, the team's model suggests that combining two specific medications–both already FDA-approved–could be highly effective against a type of TNBC tumor that is known to be especially aggressive and resistant to other treatments.

Community Clinics

Project CELSUS has made progress toward its goal of strengthening the research capabilities of the community clinics on the collaborative team. The team members from the community clinics experience challenges when participating in research, mostly because the time they spend attending project meetings is time that can't be spent seeing patients. The Project CELSUS team has been creating ways for the community investigators to maximize their share of research funds, train staff, register research protocols, share patient data, and send samples to third-party vendors for sequencing. In addition, formal interviews in Spanish and English with community clinicians and Latina patients with breast cancer are being conducted across San Diego and Imperial counties to learn about their experiences.