Wednesday , October 27 2021

Artificial intelligence predicts the risk of metastatic relapse in patients with breast cancer



[ad_1]

The RACE AI study conducted by Gustave Roussy and the startup Owkin, as part of the AI ​​for Health challenge organized by the Ile-de-France region in 2019, was presented as a paper presented at the ESMO (European Society of Medical Oncology). This study demonstrates that thanks to in-depth learning analysis applied to digitized pathology slides, artificial intelligence can classify patients with breast cancer located between high risk and low risk of metastatic relapse in the next five years. This AI could become an aid to therapeutic decision-making and prevent unnecessary chemotherapy and its impact on the personal, professional, and social lives of low-risk women. This is one of the first proof of concept that illustrates the power of an AI model to identify parameters associated with relapse that the human brain could not detect.

With 59,000 new cases a year, breast cancer ranks first among cancers in women, clearly ahead of lung cancer and colorectal cancer. It is also the cancer that causes the highest number of deaths in women, with 14% of female cancer deaths in 2018. 80% of breast cancers are said to be hormone-sensitive or hormone-dependent. But these cancers are extremely heterogeneous and approximately 20% of patients relapse with distant metastases.

RACE AI is a retrospective study that was conducted in a cohort of 1,400 patients managed at Gustave-Roussy between 2005 and 2013 for hormone-sensitive localized breast cancer (HR +, HER2-). These women were treated with surgery, radiation therapy, hormone therapy, and sometimes chemotherapy to reduce the risk of relapse at a distance.

Chemotherapy is not given routinely because not all women will benefit from it because of a naturally favorable prognosis. The choice of doctor is based on clinical-pathological criteria (age of the patient, size and aggressiveness of the tumor, invasion of lymph nodes, etc.) and the decision to administer or not adjuvant chemotherapy varies between cancer centers. There are currently genomic signatures to help identify women who benefit from chemotherapy, but they are not recommended by the French National Health Authority and are not reimbursed by the French national health insurance (although they are included in the RIHN reimbursement list), which makes access and use heterogeneous in France.

Gustave Roussy and Owkin have taken on the challenge of proposing a new, simple, economical and easy-to-use method in all oncology centers as a therapeutic tool for decision-making. Ultimately, the goal is to direct patients identified as high-risk toward new innovative therapies and avoid unnecessary chemotherapy for low-risk patients.

In the RACE AI study, Owkin data scientists, guided by Gustave Roussy’s research physicians, developed an AI model capable of reliably assessing the risk of relapse with an AUC of 81%. to help the doctor determine the benefit / risk balance of chemotherapy. This calculation is based on the patient’s clinical data combined with the analysis of stained and digitized histological slides of the tumor. These slides, used daily in pathology departments by pathologists, contain very rich and crucial information for the treatment of cancer. There is no need to develop a new technique or equip a specific technical platform. The only essential equipment is a slide scanner, which is common equipment in laboratories. Like an office scanner that scans text, this scanner scans the morphological information present on the slide.

The results of this first study by the Owkin and Gustave Roussy teams open strong perspectives and the following steps include prospective validation of the model in an independent cohort of patients treated outside of Gustave Roussy. If the results are confirmed, providing reliable information to physicians, this artificial intelligence tool will prove to be a valuable aid for therapeutic decisions.

[ad_2]
Source link