Brain metastasis is one of the most common complications of cancer, affecting between 20 to 40 per cent of all patients and leading to significant mortality and morbidity.
Stereotactic radiation therapy (SRT) is one of the main treatments for brain metastasis. This treatment delivers high-dose radiation to cancer cells while sparing surrounding healthy tissue. Despite its efficacy, up to 20 per cent of metastatic brain tumours still progress, and it can take months before patients and their doctors know if the treatment works.
Now there are some promising results from a partially funded TFRI team of Toronto-based researchers studying whether magnetic resonance imaging (MRI) can predict responses to this therapy in these patients.
“Our study demonstrated that quantitative biomarkers derived from conventional MRI in conjunction with machine learning techniques can, potentially, be used to predict outcomes for brain metastasis patients treated with stereotactic radiation therapy,” said Dr. Ali Sadeghi-Naini, a professor of biomedical engineering and computer science at York University who led the team. The work was published in Scientific Reports (December 2019).
According to Dr. Sadeghi-Naini, the biomarkers were constructed using conventional MRI data acquired from 100 patients, and were applied in conjunction with machine learning techniques for outcome prediction and risk assessment. The results indicated that heterogeneity in the surrounding regions of tumour, including the edema and tumour/lesion margins, could serve as biomarkers for treatment response.
“These regions are likely to contain malignant cells, but the number of these cells is not enough to result in an evident image contrast on standard MRI. The heterogeneity in these regions may characterize the frequency and distribution of cancerous cells and, therefore, could potentially be linked to the outcome of the treatment,” said Dr. Sadeghi-Naini.
The study followed the patients for up to five years after treatment and survival analyses showed that the cohort of the patients who were predicted by the model to have better outcomes after radiotherapy demonstrated a significantly better survival compared to those predicted to have a worse response.
“The results suggested that our predictive model can potentially have an impact on patient survival and quality of life,” said Dr. Sadeghi-Naini. “The proposed methodology can potentially facilitate early treatment adjustment on an individual patient basis that is expected to improve their outcomes.”
Quantitative MRI Biomarkers of Stereotactic Radiotherapy Outcome in Brain Metastasis
Elham Karami, Hany Soliman, Mark Ruschin, Arjun Sahgal, Sten Myrehaug, Chia-Lin Tseng, Gregory J. Czarnota, Pejman Jabehdar-Maralani, Brige Chugh, Angus Lau, Greg J. Stanisz & Ali Sadeghi-Naini
This work was supported by a The Terry Fox New Frontiers Program Project Grant in Ultrasound and MRI for Cancer Therapy
with funds from the Lotte & John Hech Memorial Foundation