Introduction Accurately predicting pathological complete response (pCR) after primary systemic therapy (PST) in breast cancer remains a clinical challenge. This study compares the diagnostic performance of ultrasound (US) and Fluorodeoxyglucose positron emission tomography/magnetic resonance imaging ([18F]FDG PET/MRI) in predicting breast-pCR following PST.Methods This multicenter analysis included 143 patients: 54 node-positive patients from San Raffaele Hospital, who underwent an additional [18F]FDG PET/MRI within a prospective trial (ClinicalTrials.gov ID: NCT04826211) and 89 cases retrospectively retrieved from the database of Heidelberg Hospital, irrespectively of nodal status at diagnosis. In US, breast clinical complete response (breast cCR) was defined as a tumor size of 0 mm or no sign of tumor in the breast; in [18F]FDG PET/MRI, as the absence of significant [18F]FDG uptake on PET images and contrast-enhancement on MRI. Diagnostic accuracy was assessed across PET only, MRI only, combined PET/MRI (PET or MRI positive), and concordant PET/MRI (both positive). Subgroup analyses explored associations of breast cCR with tumor molecular subtypes.Results [18F]FDG PET/MRI showed a sensitivity of 62% (95% CI 42%-81%) and a specificity of 67% (95% CI 46%-83%), whereas ultrasound showed a sensitivity of 59% (95% CI 44%-73%) and a specificity of 62% (95% CI 46%-77%) in detecting breast response to PST. No significant difference was observed between [18F]FDG PET/MRI and US in terms of sensitivity (62.9%vs59.1%, p=0.93) and specificity (66.6%vs62.2%, p=0.94) in detecting response to PST. Exploratory subgroup analysis suggested variation in sensitivity: US showed a sensitivity of 73% (95% CI 0.45-0.92) in triple-negative tumors and 50% (95% CI 16%-84%) in HER2-positive tumors, while [18F]FDG PET/MRI demonstrated a sensitivity of 77% (95% CI 46%-95%) in Luminal B cancers.Conclusion [18F]FDG PET/MRI and US showed comparable performance in diagnostic accuracy of breast pCR following PST, with varying effectiveness across tumor subtypes. Further research should focus on multimodal imaging strategies integrated with clinical and molecular data.
Diagnostic accuracy of modern imaging following neoadjuvant therapy for breast cancer: a comparative analysis of ultrasound versus [18F] FDG PET/MRI in breast response assessment / Di Micco, R., Saracoglu, S., Pfob, A., Canevari, C., Gallivanone, F., Antunovic, L., Gelardi, F., Scifo, P., Neri, I., Losio, C., Venturini, E., Della Vecchia, G., Ferrarese, G., Rotmensz, N., Zuber, V., Baleri, S., Cisternino, G., Corona, S.P., Calabretto, F., Rampa, M., et al.. - In: FRONTIERS IN ONCOLOGY. - ISSN 2234-943X. - 16:(2026). [10.3389/fonc.2026.1845216]
Diagnostic accuracy of modern imaging following neoadjuvant therapy for breast cancer: a comparative analysis of ultrasound versus [18F] FDG PET/MRI in breast response assessment
Della Vecchia G.;Ferrarese G.;Viale G.;Zambelli S.;Bianchini G.;Chiti A.;Gentilini O. D.Ultimo
2026-01-01
Abstract
Introduction Accurately predicting pathological complete response (pCR) after primary systemic therapy (PST) in breast cancer remains a clinical challenge. This study compares the diagnostic performance of ultrasound (US) and Fluorodeoxyglucose positron emission tomography/magnetic resonance imaging ([18F]FDG PET/MRI) in predicting breast-pCR following PST.Methods This multicenter analysis included 143 patients: 54 node-positive patients from San Raffaele Hospital, who underwent an additional [18F]FDG PET/MRI within a prospective trial (ClinicalTrials.gov ID: NCT04826211) and 89 cases retrospectively retrieved from the database of Heidelberg Hospital, irrespectively of nodal status at diagnosis. In US, breast clinical complete response (breast cCR) was defined as a tumor size of 0 mm or no sign of tumor in the breast; in [18F]FDG PET/MRI, as the absence of significant [18F]FDG uptake on PET images and contrast-enhancement on MRI. Diagnostic accuracy was assessed across PET only, MRI only, combined PET/MRI (PET or MRI positive), and concordant PET/MRI (both positive). Subgroup analyses explored associations of breast cCR with tumor molecular subtypes.Results [18F]FDG PET/MRI showed a sensitivity of 62% (95% CI 42%-81%) and a specificity of 67% (95% CI 46%-83%), whereas ultrasound showed a sensitivity of 59% (95% CI 44%-73%) and a specificity of 62% (95% CI 46%-77%) in detecting breast response to PST. No significant difference was observed between [18F]FDG PET/MRI and US in terms of sensitivity (62.9%vs59.1%, p=0.93) and specificity (66.6%vs62.2%, p=0.94) in detecting response to PST. Exploratory subgroup analysis suggested variation in sensitivity: US showed a sensitivity of 73% (95% CI 0.45-0.92) in triple-negative tumors and 50% (95% CI 16%-84%) in HER2-positive tumors, while [18F]FDG PET/MRI demonstrated a sensitivity of 77% (95% CI 46%-95%) in Luminal B cancers.Conclusion [18F]FDG PET/MRI and US showed comparable performance in diagnostic accuracy of breast pCR following PST, with varying effectiveness across tumor subtypes. Further research should focus on multimodal imaging strategies integrated with clinical and molecular data.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
