Work

Evaluation of lesion and neoadjuvant treatment response using low-keV dual-energy CT in breast cancer

Year: 2026

Type: article

Abstract: Exploring the capacity of detecting breast cancer lesion on different monoenergetic images (MEIs) and predicting NAT response using quantitative parameters derived from DECT. This retrospective study enrolled patients with invasive breast cancer who received NAT followed by surgery. Meanwhile, all participants underwent chest DECT and breast MRI before and after NAT. The first end was to assess different MEIs for detecting breast lesion using objective and subjective measurements. The second end was to establish model for predicting pCR using quantitative parameters, and compare the model with visual MRI-rCR. The Kaplan-Meier method was used to analyze prognosis. 199 women with breast cancer were included (median age, 50 years). 52 (26.13%) achieved pCR and 147 (73.87%) were non-pCR after NAT. Low keV MEI had optimal performance to characterize lesion, especially on 40 keV MEI at venous phase (CNR: 6.55, SNR: 7.23, Likert score: 5; all p < 0.001). The multivariable regression showed HER2 positivity (OR, 0.155; 95% CI: 0.061–0.393; p < 0.001) and CT value on 40 keV MEI at venous phase (OR, 1.067; 95% CI: 1.020–1.117; p = 0.005) were independently associated with pCR. The established combined model yielded an AUC of 0.900 (95% CI: 0.856–0.944) when assessed on the full study cohort, beyond the MRI-rCR (AUC, 0.900 vs. 0.782; p = 0.003). The model also showed a nominally significant association with disease-free survival (DFS) on the whole dataset (log-rank, p = 0.046). MEI with low keV significantly improved visualization of breast cancer. Quantitative parameters derived from 40 keV MEI demonstrated superior value for assessing NAT response and prognosis compared to MRI-rCR. Dual-energy CT has been used for diagnosis of breast lesion. However, the predictive performance to evaluate neoadjuvant treatment response remains unexplored in breast cancer. Monoenergetic images (MEIs) with low keV has optimal performance for visualization of breast cancer and its quantitative parameters can identify NAT response and prognosis. Low keV MEI effectively visualizes breast cancer as a complementary examination. Its quantitative parameters at 40 keV MEI demonstrated superior value for assessing NAT response and prognosis compared to MRI-rCR so as to assist clinical strategies. Not applicable.

Source: Cancer Imaging

DOI: 10.1186/s40644-026-01116-7