Preoperative chemoradiotherapy is the standard care for advanced rectal cancer. Yet, predicting the response remains a challenge. Our previous studies found that 53BP1 protein deletion leaded to chemoradiotherapy resistance.53BP1 deficiency played pivotal role in inducing lymphocyte mature. As the development of functional imaging, texture analysis can predict response to chemoradiotherapy. Our study aims to explore the relationship among 53BP1 protein, immune score, texture analysis and radiation sensitivity, then set a new classifier to predict response in advanced rectal cancer.
We enrolled 57 advanced rectal cancer patients who received neoadjuvant chemoradiotherapy in Wuhan Union hospital from January 2015 to January 2018. The expression of 53BP1, CD3, CD8, CD45RO was detected by immunohistochemistry. Immune score were calculated, which are based on the combination of two lymphocyte populations (CD3/CD45RO, CD3/CD8 or CD8/CD45RO) both in the center (CT) and the invasive margin (IM) of tumors. T2W MRI images studies obtained in 34 patients before treatments, texture analysis by manually delineating a region of interest. The response rate, anus preservation rate were collected. The relationship among 53BP1 protein, immune score, texture analysis parameters and radiation sensitivity were explored by t-test. A new classifier was set to predict pathological response.
The results indicated that the response rate in 53BP1 deletion group was significantly lower than that in 53BP1 high expression group (25.00% vs.84.44%, P < 0.005). "CD3+CD8" is the most reasonable immune score, the patients of 53BP1 deletion had lower immune score. The response rate in lower immune score group was lower than higher immune score group (54.54% vs.88.57%, P < 0.005). Texture analysis parameters were related with response rate. We set a new classifier to predict response in advanced rectal cancer, we randomly selected 10 patients as training group, others were selected as predict group, the accuracy is 83.33%.
A classifier of 53BP1, immune score and texture analysis of MRI can predict response in advanced rectal cancer.
Clinical trial identification
Legal entity responsible for the study
Union Hospital, Tongji Medical College, Huazhong University of Science and Technology.
All authors have declared no conflicts of interest.