Early prediction of neoadjuvant chemotherapy efficacy among patients with triple-negative breast cancer using an ultrasound-based radiomics nomogram

Abstract

Purpose: To develop and validate a radiomics nomogram based on early ultrasound (US) imaging for predicting pathologic complete response (pCR) in patients with triple-negative breast cancer (TNBC) receiving neoadjuvant chemotherapy (NAC).

Methods: This retrospective study included 328 patients with TNBC treated between September 2019 and January 2024, divided into a training cohort (n = 230) and a validation cohort (n = 98). Clinicopathologic data, US features before NAC, tumor volume reduction (TVR) after two cycles of NAC, and radiomics features were collected. Multiple logistic regression was applied to identify the potential predictors of pCR. The efficacy of the nomogram was evaluated through the receiver operating characteristic, calibration, and decision curve analyses. The study was approved by the ethics committee on February 28, 2024, with approval number 2023-SR-799, and the requirement for informed consent was waived.

Results: Twelve features were selected to construct the radiomics signature (RS). The nomogram, incorporating tumor histologic grade, TVR, and RS, yielded an area under the curve of 0.856 [95% confidence interval (CI), 0.807-0.905] in the training cohort and 0.836 (95% CI, 0.749-0.923) in the validation cohort, outperforming both the clinico-ultrasonic and RS models. The calibration and decision curves confirmed the nomogram's excellent calibration and clinical utility.

Conclusion: The nomogram, which includes US characteristics, clinical variables, and radiomics features, exhibited satisfactory performance in predicting NAC efficacy in patients with TNBC.

Clinical significance: The US-based radiomics nomogram, incorporating histologic grade, TVR, and RS, shows preliminary clinical application potential for predicting NAC efficacy in patients with TNBC.

Keywords: Triple-negative breast cancer; neoadjuvant chemotherapy; pathologic complete response; radiomics; ultrasonography.

Figure 1

Flowchart showing the study exclusion criteria. TNBC, triple-negative breast cancer; NAC, neoadjuvant chemotherapy; US, ultrasound.

Figure 2

(a) Radiomics workflow and (b) study flowchart. US, ultrasound; ROI, region of interest.

Figure 3

Grey-scale US images. (a, b) The three largest tumor dimensions in the transverse and longitudinal images before NAC. (c, d) The three largest tumor dimensions in the transverse and longitudinal images after two cycles of NAC. The patient completed neoadjuvant systemic therapy, and surgical pathology confirmed a pCR. (e, f) The three largest tumor dimensions in the transverse and longitudinal images before NAC. (g, h) The three largest tumor dimensions in the transverse and longitudinal images after two cycles of NAC. The patient completed neoadjuvant systemic therapy, and surgical pathology confirmed a non-pCR. US, ultrasound; NAC, neoadjuvant chemotherapy; pCR, pathologic complete response.

Figure 4

(a) Heatmap of selected radiomics features from the training cohort (GLSZM, grey level size zone matrix; NGTDM, neighboring gray tone difference matrix). (b) Comparison of Radscores between the pCR and non-pCR groups in the training and validation cohorts. (c) Features selected using LASSO regression. pCR, pathologic complete response; LASSO, least absolute shrinkage and selection operator.

Figure 5

Development of the US-based radiomics nomogram for predicting pCR in TNBC. The nomogram integrates histologic grade, TVR, and RS. US, ultrasound; TNBC, triple-negative breast cancer; TVR, tumor volume reduction; RS, radiomics signature.

Figure 6

(a, b) ROC comparisons of the clinico-ultrasonic model, RS model, and radiomics nomogram in the training and validation cohorts. (c, d) Calibration curves of the radiomics nomogram in the training and validation cohorts. (e) Decision curve of the RS model, clinico-ultrasonic model, and radiomics nomogram. ROC, receiver operating characteristic; RS, radiomics signature; pCR, pathologic complete response.