A nomogram to predict the value of mammary adipose index (MAI) for prediction of response to neoadjuvant chemotherapy in breast cancer patients.

Abstract

e12607 Background: Breast cancer (BCa) is the most common malignancy in females on a global scale. Neoadjuvant chemotherapy (NACT) is a promising therapeutic approach for the treatment of locally advanced BCa, which result in primary tumor regression and optimal surgical planning. Cumulative evidence has identified a deleterious crosstalk between cancer cells and the surrounding mammary adipose tissue during tumor progression. Mammary adipose index (MAI) is a novel indicator defined by us, which represents the percentage of adipose tissue in breast. Therefore, this study aims to investigate the correlation between MAI and NACT efficacy of BCa by establishing an intuitive nomogram. Methods: Clinical data of 221 BCa patients who received NACT at the First Affiliated Hospital of Nanjing Medical University from January 1, 2018 to May 31, 2020 were retrospectively collected. Besides, patients were divided into major histological response and non-major histological response group in line with Miller-Payne pathology grading. The level of MAI was collected from magnetic resonance imaging of breast, and was divided into two groups according to the cut-off value. The level of MAI and clinicopathological characteristics were collected to assess the predictive roles of NACT response. Univariate analysis was performed by Chi-square test or Fisher’s exact test. Indicators with P < 0.05 were included in multivariate binary logistic regression analysis. The nomogram was constructed based on results of multivariate logistic analysis. The predictive accuracy and discriminative ability were evaluated by C-index and the calibration curve. Results: Among the 221 patients enrolled, 77 cases (34.84%) were represented by major responses of pathologic histology. In univariate analysis, the level of MAI, tumor size, node status, ER status, PR status HER-2 status and Ki-67 status were correlated with NACT. Multivariate Logistic regression analysis showed that tumor size, HER-2 status and the level of MAI were independent influencing factors for the response to NACT in BCa patients. The patients with a high level of MAI (≥0.6916) were more likely to have a worse NACT response. MAI was an independent predictor for NACT efficacy, and area under the curve of ROC, specificity and sensitivity of the predictive model were 0.805, 0.632 and 0.774 respectively. The nomogram established based on these factors showed its discriminatory ability, and the C-index for prediction was 0.881. The calibration curve indicated that the predictive ability of the nomogram was a good fit to actual observation. The difference was statistically significant (P < 0.05). Conclusions: The nomogram constructed in the present study indicated that the level of MAI could predict the efficacy of NACT in BCa patients, which might provide clinical guidance for the selection of appropriate treatment decisions.