Click here to

Session: Radiomics for Outcome Modeling [Return to Session]

Predictive Value of Pre-Operative MRI Radiomics for Treatment Response in Invasive Breast Cancer

A Rincon1*, L Young2, Y Yang3, F Yang4, (1) Department Of Biomedical Engineering, University Of Miami, Coral Gables, FL, (2) Department Of Radiation Oncology, University of Washington, Seattle, WA, (3) Department of Radiation Oncology, The First Affiliated Hospital of University of Science and Technology of China, Hefei, China, (4) (1) Department Of Radiation Oncology, University Of Miami, Miami, FL

Presentations

SU-H400-IePD-F5-4 (Sunday, 7/10/2022) 4:00 PM - 4:30 PM [Eastern Time (GMT-4)]

Exhibit Hall | Forum 5

Purpose: MRI is commonly used in the pre-operative evaluation of women with breast cancer to determine the extent of the tumor and assist in surgical planning. Radiomics analysis of these images may provide the potential for non-invasive early treatment response prediction that could serve as a trigger for treatment adaption. The purpose of this study was to assess the discrimination power of pre-operative MRI radiomics features between breast cancer patients who responded to treatment and those who were non-responsive.

Methods: The current study employed patient data (n=259) from the Duke Breast Cancer MRI collection, for which imaging and clinical data are publicly available from The Cancer Imaging Archive (TCIA). All patients had T1-weighted and dynamic contrast enhanced MRI (DCE-MRI). Imaging features were extracted from the delineated tumor and fibroglandular tissue volumes. This study utilized ten gray-level co-occurrence matrix (GLCM) features, and the endpoint of the study was tumor response. The radiomics features for patients who responded to treatment versus those who were non-responsive were analyzed and compared using the Wilcoxon signed-rank test and Pearson’s correlation coefficient. P-values of 0.05 or less, after multiple test correction using the method of Bonferroni–Holm, were considered statistically significant.

Results: Wilcoxon signed-rank tests revealed two features, sum_average_tissue_T1 and sum_entropy_tissue_T1, statistically different between the responders and non-responders with a p-value of 0.005 and 0.004, respectively. Both these features lie in the fibroglandular tissue enhancement texture category. In addition, the two identified features demonstrated a moderately strong correlation with Person’s correlation coefficient of 0.86.

Conclusion: Pre-operative MRI radiomics features demonstrated discriminative ability in identifying patients with invasive breast cancer who were non-responsive to treatment and may have a role to play in guiding potential treatment adaptation strategies for this patient population.

Keywords

Not Applicable / None Entered.

Taxonomy

Not Applicable / None Entered.

Contact Email

Share: