Christina Curtis, PhD, MSc, recipient of the 2024 American Association for Cancer Research (AACR) Outstanding Investigator Award for Breast Cancer Research, was recognized for her contributions to understanding molecular determinants of breast cancer and for her work in developing prognostic and predictive biomarkers.

Dr. Curtis, the RZ Cao Professor of Medicine, Genetics, and Biomedical Data Science and Director of Artificial Intelligence and Cancer Genomics at Stanford University School of Medicine, delivered a lecture about her work on Wednesday, December 11, during the 2024 San Antonio Breast Cancer Symposium®. The session recording will be available on demand for registered 2024 SABCS® participants until March 31, 2025.

Since 2008, this annual award — supported by the Breast Cancer Research Foundation — has honored investigators under the age of 50 for their novel and significant work that has had or may have a far-reaching impact on the etiology, diagnosis, treatment, or prevention of breast cancer.

During her presentation, The integrative breast cancer subtypes: From relapse prediction to interception, Dr. Curtis focused on how discoveries of fundamental mechanistic processes driving cancer evolution across the continuum — from pre-invasive lesions to recurrent and relapsed treatment-resistant disease — can be translated to patient populations and are reflected in molecular subtypes.

Dr. Curtis noted that while there are numerous challenges in precision oncology, there are also abundant opportunities.

Human epidermal growth factor 2 (HER2) gene amplification has been successful as both a biomarker and a therapeutic target in breast cancer, Dr. Curtis noted. She and her colleagues questioned whether it would be possible to replicate the success of HER2 by identifying other oncogenic drivers affected by copy number modifications.  

In earlier research, Dr. Curtis and colleagues developed a molecular map of breast cancer that identified “integrative clusters” based on both genomic and transcriptomic data from 1000 tumors within the METABRIC cohort. These integrative clusters were grouped into four distinct breast cancer molecular subtypes — typical-risk estrogen receptor-positive (ER+), triple-negative, high-risk ER+, and HER2+ disease. Over the course of two decades, the integrative clusters were found to be superior at predicting late relapse compared to predictions that were based on clinical variables.

Dr. Curtis’ ongoing research of these integrative cluster-mediated subgroups has helped identify characteristic genomic amplifications of oncogenes.

The data indicated that HER2 is not unique; rather, it serves as an exemplar of amplified oncogenes that define breast cancer subtypes. Furthermore, the oncogene amplicons identified by Dr. Curtis and her team were associated with intrinsic resistance to endocrine therapy.

These amplicons were enriched in certain patients, such as in young women with high-risk ER+ disease. Notably, endocrine therapy is not as effective in this patient population.

To address this issue, Dr. Curtis and collaborators have designed the Targeting Estrogen Receptor-Positive Selected Integrative Clusters at High Risk of Relapse (TERPSICHORE) study. TERPSICHORE is an umbrella “window-of-opportunity” study assessing whether adding a therapy targeting integrative subtype-specific genomic changes identified pre-operatively can improve outcomes over standard endocrine therapy in patients with high-risk ER+ breast cancer.

“To learn even more, we realized that we had to go back and build new functional models of these high-risk-of-relapse groups, so we’ve established a biobank of patient-derived organoids where we can start to ask about the mode of action, response to particular agents,” Dr. Curtis said. “We can correlate this with patient molecular data through spatial molecular profiling, and we can now start to learn how we might better optimize therapy, why did these clinical trials fail, which patients benefited, what can we learn about new targets?”

Dr. Cutis rounded out her presentation by describing her group’s latest studies, which focus on the mutational processes that drive cancer progression across various disease stages — from ductal carcinoma in situ to metastatic disease — and molecular subtypes, including typical-risk ER+, triple-negative, high-risk ER+, and HER2+ breast cancer. This work has highlighted the role of germline variants in epitope presentation and immune surveillance of nascent tumors.

Dr. Curtis described the dynamic immunoediting process as a “war of clones between the tumor and the immune system,” wherein high germline epitope burden can mediate immune surveillance and clearance of tumor cells during initial stages of carcinogenesis but proliferation of tumor cells with low germline epitope burden combined with immune suppression can lead to immune escape in aggressive tumors.

Deciphering the early events that drive tumorigenesis and their trajectory in tumor evolution across the cancer continuum has been a central theme in her research career.

“We have an opportunity to leverage the fact that these early alterations set up the tumor, and they establish contingencies,” Dr. Curtis concluded, “and that gives us a real opportunity to think about predictive models. I would argue that we can, in fact, learn the evolutionary rulebook of disease and we can exploit it.”