A new regression-based prognostic model developed by researchers at the University of North Carolina uses routine electronic health record data to estimate near-term death risks for patients with metastatic breast cancer, according to findings published in JCO Oncology Practice. Led by Dr. Emily Ray, the tool aims to help oncologists recognize high-risk periods and prompt timely end-of-life care discussions.
How the Metastatic Breast Cancer Prognostic Model Works
The newly published tool draws on information already available in standard electronic health records. According to the study, it calculates the probability of death within 30 or 90 days by analyzing laboratory results, vital signs, breast cancer subtypes, and current medications.
Unlike broad prognostic tools that combine multiple cancer types and frequently underrepresent breast cancer, this model is specifically derived from a national oncology database featuring real-world data, specifically CancerLinQ Discovery. This targeted approach allows clinicians to evaluate risks even for patients who have lived with metastatic breast cancer for extended periods.
Recognizing Subtle Clinical Signals of Decline
Recognizing when a patient enters their final days remains a persistent challenge in oncology, particularly as an increasing number of effective treatments prolong life and expand options. Physicians frequently overestimate survival timelines, and such uncertainty can delay important supportive care conversations.
“In our prognostic model, the variables most closely correlated with increased risk align with signals clinicians already recognize like worsening liver function, increased heart rate, rising needs for pain medication, and declining ability to care for oneself at home,” said Emily Ray, a medical oncologist at UNC Lineberger Comprehensive Cancer Center and associate professor in the UNC School of Medicine’s Division of Oncology.
While any clinician can spot these indicators of a worsening condition, busy schedules and emotional hurdles often cause providers to miss crucial windows to prepare patients and families, according to Ray. The model serves as a practical reinforcement of clinical intuition.
Prompting Timely Palliative Care and Care Alignment
Beyond estimating survival probabilities, the regression-based tool is designed to act as a practical nudge for clinicians. Time constraints, emotional weight, and patient reluctance often make end-of-life discussions difficult to initiate in standard practice.
“We hope that this tool will nudge oncologists to have conversations with their patients that are sometimes hard to have due to time constraints, the emotional weight, and sometimes patient reluctance to talk about these tough topics,” Ray stated. The research team notes that even when patients prefer not to discuss prognosis directly, the model can encourage providers to review treatment options, involve palliative care specialists early, and ensure that care plans match individual values and goals.
Did You Know? M. Ray and colleagues.
Frequently Asked Questions
What data does the new metastatic breast cancer model use?
The tool uses routinely collected clinical data from electronic health records, including vital signs, laboratory results, medications, and breast cancer subtype.
What does the model predict?
According to the research published in JCO Oncology Practice, the regression-based model estimates the probability of death within 30 or 90 days for patients with metastatic breast cancer.
Who led the development of this prognostic tool?
The research effort was led by Dr. Emily Ray, a medical oncologist at the UNC Lineberger Comprehensive Cancer Center and associate professor at the UNC School of Medicine.
How will the tool be used in the future?
The research team plans to conduct further studies to determine how best to integrate the model into everyday clinical practice and measure its impact on decision-making for oncologists, patients, and caregivers.