Jun 27, 2022

Researchers at U of T improve personal risk measurements for non-invasive breast cancer

Research: Artificial Intelligence in healthcare, Research: Cancer, Research: Genetics, Genomics & Proteomics, Research: Molecular & cell biology, Our people
Dr. Sharon Nofech-Mozes

Ductal carcinoma in situ (DCIS) is a non-invasive form of Breast Cancer, occurring in the milk ducts, and accounts for around one in five of all screen-detected Breast Cancer cases. As non-invasive cancer, malignant cells are not spreading to other tissues, or metastasizing, and by itself the condition is not fatal. However, depending on the treatment, some women will develop invasive cancer over time - around 10% of women treated by breast conserving surgery alone will have an invasive breast cancer within 10 years of follow up.  

Up until now, doctors have been unable to tell which women are at higher risk of developing invasive cancer following a diagnosis of DCIS, so all patients have been treated the same. Historically, treatment involved a mastectomy - removing the whole breast, but in the last 30-40 years, akin to invasive cancer, the treatment is typically a partial breast resection conserving the breast (or lumpectomy) where only the area around the tumour is removed, followed by radiation therapy.

Dr. Sharon Nofech-Mozes is a Professor in the Department of Laboratory Medicine and Pathobiology in the Temerty Faculty of Medicine and staff pathologist at Sunnybrook Health Sciences Centre where she is a part of the breast pathology-gynecological pathology service.

“It is obvious we are overtreating patients with non-invasive cancer by removing the entire breast or by adding radiation therapy to all patients following partial breast resection,” explains Dr. Nofech-Mozes. “There were several randomized trials comparing the outcome of patients treated with lumpectomy with and without radiation therapy which showed that universally radiation therapy reduced the risk of invasive recurrence by 50%. However, clinical and histopathological factors are unable to identify women who won’t benefit from the addition of radiation therapy.” 

To find out more, a team of researchers based in Sunnybrook Health Sciences Centre, led by Radiation Oncologist Dr. Eileen Rakovitch and Dr. Nofech-Mozes as the lead pathologist, identified The DCIS Ontario Cohort, the largest population-based cohort with an associated tumour bank in the world. The team has spent 10 years identifying 6,000 women diagnosed with DCIS from 78 institutions across Ontario, of whom 3,500 were treated by breast conserving surgery, to learn more about the disease and what effect treatments were having. They conducted a comprehensive pathology review and selected representative archival tissue for molecular studies. “When we started the study, I was an early career pathologist with four young children so it was very challenging to juggle between the clinical service, scholarly activity and daily chores, but it was also interesting to travel all over the province,” says Dr. Nofech-Mozes.

The study gave researchers an opportunity to investigate different aspects of DCIS in terms of prognostic and predictive markers that led to several findings. One was age – they found that women younger than 50 are at a higher risk of recurrence following a diagnosis of DCIS, as were women with multifocal vs unifocal disease (how it is spread across the breast). Multiple research teams world-wide have made an extensive effort to improve our ability to assess individual patient’s risk and response to treatment in invasive and non-invasive breast cancer. Several multigene assays became available to improve outcome prediction in invasive Breast Cancer. 

With archival tissue from the Ontario DCIS cohort analyzed, they were able to validate a 12-gene assay that can, in combination with other factors, improve individual prediction of the risk of recurrence. The clinical validation has led to the 12-gene assay being commercially available internationally.

“This personal risk estimate helps women and their health providers make much more informed choices about their care,” explains Dr. Nofech-Mozes, “Women whose risk of recurrence is very low are likely to prefer no additional treatment following breast conserving surgery but women with high risk are likely to accept a recommendation for radiation therapy. Women whose risk remains high, even with radiation therapy, are more likely to consider a mastectomy.”

“This is a great improvement compared to the lack of options prior to the study. However, this prognostic assay will only inform on the basic risk of recurrence. A predictive assay is designed to determine the benefit from treatment, and this is our future goal,” says Dr. Nofech-Mozes. The team is currently interrogating archival tissue from the Ontario DCIS cohort through modern molecular techniques and bioinformatics, as they search for a signature that can predict the benefit from radiation therapy. 

The original work gathering the cases in this cohort started in 2004 and the team is planning to build a new cohort for future studies. Technologic advancement in breast imaging and radiation therapy means that imaging today is much more detailed and radiation therapy has improved. “Our historic cohort has the advantage of prolonged follow up period but has the weakness of relying on old imaging techniques. By expanding and modernising the cohort, we will be able to validate our findings in a more up to date group of patients.”

They are about to start a clinical trial prospectively evaluating the outcomes in women diagnosed with low-risk DCIS as defined by the 12-gene assay and clinico-pathological features. “If we can improve the way we select patients for therapy, there are benefits to the hospitals in terms of resources, but most importantly, we can ensure patients get the very best, personalised care for their condition. The days of one treatment fits all are behind us,” says Dr. Nofech-Mozes. 

Dr. Sharon Nofech-Mozes was promoted to full Professor this year in recognition of her research and other academic achievements.