Toward Functional Precision Oncology to Predict, Prevent, and Treat Early Metastatic Recurrence of Triple-Negative Breast Cancer
Abstract
Unfortunately, most promising new therapies for breast cancer fail to show benefit when tested in clinical trials. Although mortality for breast cancer is steadily decreasing by about 2% every year, more than 42,000 people still die from this disease each year in the U.S. alone. We still have a lot of work to do to eliminate mortality from breast cancer. In part, the disappointing failure of new therapies can be attributed to using model systems for research that do not perfectly represent real breast cancers in living individuals. For example, the gold standard for preclinical testing is showing the effect of the potential drug on breast cancer cell lines that have been cultured in the laboratory for decades – out of the body, in isolation, and in artificial tissue culture conditions where many changes occur in the cancer cells. On the other hand, it is very difficult to generate ideal model systems that fully recapitulate the disease, and no model is perfect. Moreover, we now appreciate that every tumor is genetically and functionally different, so testing new therapies on a few cancer cell lines does not represent effects that will be seen in a large patient population. Several years ago, we initiated a large, collaborative translational research project to create new models for breast cancer. We developed techniques to implant a small fragment of patient’s tumors (from the breast or a metastatic site) into the mammary glands of mice. These patient-derived xenograft (PDX) models are now considered to be the most accurate models for human breast cancer. In our first study of 42 patient tumors that were implanted into mice, we serendipitously found a correlation between successful PDX engraftment and poor outcome for patients. Thus, with funding from a 2013 DOD BCRP Breakthrough Award, our group launched a clinical study to determine if this finding was correct, in a trial of 80 patients. Final results from this study are not yet published because of the long follow-up time for each patient; however, our interim analysis strongly shows that we can use PDX engraftment to assess which newly diagnosed Stage I-III triple-negative breast cancer (TNBC) or hormone receptor (HR)-low/HER2-negative patients will experience early metastatic recurrence. In this study, patients whose tumors grew as a PDX have a remarkably high risk of early recurrence (more than 30-fold), while those that couldn’t grow have very low rates of early recurrence. We have also developed new ways to tailor therapy on a personalized level using the patient’s own tumor samples grown in the mice as PDX or in the lab in three-dimensional cultures called organoids. Here, we plan to enroll an additional 80 patients in a new trial and test the hypothesis that we can not only predict early metastatic recurrence for these patients, but also use the models to personalize therapy once the tumor recurs. This will be done by actually testing approved therapies for their ability to kill that patient’s tumor cells, and then informing the physician of the results. This will be for breast cancer patients with TNBC or HR-low/HER2- breast cancer, because we will have time during this grant period to observe the recurrences and test the personalized therapies (other breast cancer subtypes often don’t recur for years after diagnosis and initial therapy). We hope to use these personalized tests to predict the best, yet least toxic, therapy for individual patients, rather than taking the current trial-and-error approach with all available therapies. While personalized modeling and therapy selection is not feasible for every breast cancer patient, this might be crucial for certain high-risk breast tumors. Thus, we are also proposing to study the tumors from the 160 patients in our two trials (80 patients each) to identify the key features of the aggressive breast cancers that can grow as PDX, without having to actually grow them. Instead, we could simply develop a test f
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Dec 28, 2022
- Source ID
- W81XWH2210124
Entities
People
- Christos Vaklavas
Organizations
- United States Army
- University of Utah