The following is a summary of “Factors associated with engraftment success of patient-derived xenografts of breast cancer,” published in the March 2024 issue of Oncology by Lee et al.

Patient-derived xenograft (PDX) models represent a crucial resource for preclinical investigations of novel breast cancer therapies, offering a faithful replication of primary tumor characteristics. However, the success rate of PDX engraftment varies widely across studies. In this research endeavor, the primary objective was to discern the principal factors contributing to the successful engraftment of primary breast cancer in PDX models. By leveraging a comprehensive dataset integrating clinicopathological parameters with morphological attributes quantified through trained artificial intelligence (AI) algorithms, the researchers meticulously analyzed the determinants influencing PDX engraftment.

The multivariate logistic regression analyses revealed several noteworthy findings. Notably, a high Ki-67 labeling index (Ki-67LI), indicative of proliferative activity, emerged as a robust predictor of successful PDX engraftment (p < 0.001), alongside younger age at diagnosis (p = 0.032) and post-neoadjuvant chemotherapy (NAC) status (p = 0.006). Moreover, histologic grade, tumor size, and specific morphological features assessed by AI, including intratumoral necrosis and invasive carcinoma proportions, significantly influenced engraftment success (p < 0.05). Further stratification based on NAC status unveiled additional insights, wherein a higher Ki-67LI, lower Miller-Payne grade, and reduced proportion of intratumoral normal breast glands, as determined by AI, collectively exhibited exceptional predictive accuracy for successful PDX engraftment (area under the curve [AUC] 0.89).

In conclusion, the study underscores the pivotal role of diverse clinicopathological and morphological parameters in determining the success of PDX engraftment for primary breast cancer. Notably, high Ki-67LI, younger age, post-NAC status, higher histologic grade, larger tumor size, and specific morphological attributes identified through AI analysis collectively contribute to the predictive accuracy of PDX engraftment. These findings hold significant implications for optimizing PDX model establishment protocols and enhancing their utility in preclinical breast cancer research endeavors.

Source: breast-cancer-research.biomedcentral.com/articles/10.1186/s13058-024-01794-w