COMMITMENT

Background

Breast cancer is the second leading cause of cancer-related deaths in women. There are 3 subtypes of breast cancer: hormone receptor-positive (HR+), HER2-positive (HER2+), and triple-negative breast cancer (TNBC). Each of these subtypes has a different prognosis, with their own treatment options. TNBC is generally the most aggressive subtype of breast cancer, with patients often undergoing surgery to remove the tumor, followed by chemotherapy. However, a recent study showed that 50% of TNBC patients who did not undergo systemic chemotherapy were still recurrence-free after 10 years. Conversely, for patients with HR+ breast cancer without additional risk factors, the typical treatment of hormonal therapy may not be sufficient. Of all HR+ breast cancers that are deemed low risk and are therefore not treated with chemotherapy, 23% of patients still develop a recurrence within 10 years. Finally, neoadjuvant chemotherapy is increasingly prescribed to breast cancer patients to reduce tumor size before surgery. However, a substantial group of patients does not respond to treatment while still suffering from its side effects, underlining the need for clinical biomarkers that can predict treatment outcomes.

Aim

The aim of COMMITMENT (COMputational pathology for IMproved Treatment decision Making for brEast caNcer paTients) is to improve treatment decision-making for patients with breast cancer by developing new biomarkers with the use of artificial intelligence to identify at-risk patients. The project has 3 main goals:

Identify TNBC patients with a low baseline risk of recurrence who may safely forgo chemotherapy (treatment de-escalation);
Identify HR+/HER2- patients at an increased risk of recurrence who should receive chemotherapy (treatment escalation);
Distinguish between patients who will versus will not respond to neoadjuvant chemotherapy.

Within this multidisciplinary project, pathologists and AI researchers from various international hospitals work together to create new biomarkers for breast cancer treatment. COMMITMENT will improve existing AI models to facilitate automated scoring of tumor tissue in order to provide a fast, consistent method of predicting disease prognosis and treatment response. The developed algorithms will be validated and deployed to clinical practice, and with the help of our experienced industry partners at Aiosyn, we will work towards certification and potential commercialization of the algorithms developed within COMMITMENT.