Latent class analysis of conventional ultrasound features: a novel approach to predicting non-response to neoadjuvant chemotherapy in breast cancer

1

The study analyzed 509 breast cancer patients to assess ultrasound features' association with non-response to neoadjuvant chemotherapy.
2

Latent class analysis (LCA) outperformed other clustering methods in identifying ultrasound patterns linked to pathological non-response.
3

Two distinct ultrasound patterns were identified, with Pattern B showing a 2.51-fold increased likelihood of non-response compared to Pattern A.
4

In high-risk groups, Pattern B was associated with a non-response risk increase of 12.84% to 21.34%, while non-high-risk groups showed a 1.12% to 3.45% increase.
5

The findings suggest that ultrasound-derived patterns could help identify breast cancer patients unlikely to benefit from neoadjuvant chemotherapy.

Frontiers In Oncology

by Mei-Lian Zhang, Yi Tang, Cheng-Hua Kong, Li Sun, Xing-Qing Li, Yu Zhou, Zhi-Kui Chen
May 7, 2026

1