Ultrasound radiomics features to identify patients with triple-negative breast cancer

• Verfasst von: Cai, Lie [VerfasserIn]
• Verfasst von: Sidey-Gibbons, Chris [VerfasserIn]
• Verfasst von: Nees, Juliane [VerfasserIn]
• Verfasst von: Riedel, Fabian [VerfasserIn]
• Verfasst von: Schäfgen, Benedikt [VerfasserIn]
• Verfasst von: Togawa, Riku [VerfasserIn]
• Verfasst von: Killinger, Kristina [VerfasserIn]
• Verfasst von: Heil, Jörg [VerfasserIn]
• Verfasst von: Pfob, André [VerfasserIn]
• Verfasst von: Golatta, Michael [VerfasserIn]
• Titel: Ultrasound radiomics features to identify patients with triple-negative breast cancer
• Titelzusatz: a retrospective, single-center study
• Verf.angabe: Lie Cai, Chris Sidey-Gibbons, PhD, Juliane Nees, MD, Fabian Riedel, MD, Benedikt Schaefgen, MD,Riku Togawa, MD, Kristina Killinger, MD, Joerg Heil, MD, André Pfob, MD, Michael Golatta, MD
• Ausgabe: Online version of record before inclusion in an issue
• E-Jahr: 2023
• Jahr: 09 December 2023
• Umfang: 12 S.
• Illustrationen: Illustrationen
• Fussnoten: Gesehen am 29.01.2024
• Titel Quelle: Enthalten in: Journal of ultrasound in medicine
• Ort Quelle: Hoboken, NJ : Wiley, 1982
• Jahr Quelle: 2023
• Band/Heft Quelle: (2023), Seite 1-12
• ISSN Quelle: 1550-9613
• DOI: doi:10.1002/jum.16377
• Datenträger: Online-Ressource
• Sprache: eng
• Sach-SW: breast cancer
• Sach-SW: diagnostic imaging
• Sach-SW: machine learning
• Sach-SW: neoadjuvant systemic treatment
• Sach-SW: triple-negative breast cancer
• K10plus-PPN: 1879409445

Abstract

Objectives Patients with triple-negative breast cancer (TNBC) exhibit a fast tumor growth rate and poor survival outcomes. In this study, we aimed to develop and compare intelligent algorithms using ultrasound radiomics features in addition to clinical variables to identify patients with TNBC prior to histopathologic diagnosis. Methods We used single-center, retrospective data of patients who underwent ultrasound before histopathologic verification and subsequent neoadjuvant systemic treatment (NAST). We developed a logistic regression with an elastic net penalty algorithm using pretreatment ultrasound radiomics features in addition to patient and tumor variables to identify patients with TNBC. Findings were compared to the histopathologic evaluation of the biopsy specimen. The main outcome measure was the area under the curve (AUC). Results We included 1161 patients, 813 in the development set and 348 in the validation set. Median age was 50.1 years and 24.4% (283 of 1161) had TNBC. The integrative model using radiomics and clinical information showed significantly better performance in identifying TNBC compared to the radiomics model (AUC: 0.71, 95% confidence interval [CI]: 0.65-0.76 versus 0.64, 95% CI: 0.57-0.71, P = .004). The five most important variables were cN status, shape surface volume ratio (SA:V), gray level co-occurrence matrix (GLCM) correlation, gray level dependence matrix (GLDM) dependence nonuniformity normalized, and age. Patients with TNBC were more often categorized as BI-RADS 4 than BI-RADS 5 compared to non-TNBC patients (P = .002). Conclusion A machine learning algorithm showed promising potential to identify patients with TNBC using ultrasound radiomics features and clinical information prior to histopathologic evaluation.