PET/CT radiomics in breast cancer: promising tool for prediction of pathological response to neoadjuvant chemotherapy
- 236 Downloads
To assess the role of radiomics parameters in predicting pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) in patients with locally advanced breast cancer.
Seventy-nine patients who had undergone pretreatment staging 18F-FDG PET/CT and treatment with NAC between January 2010 and January 2018 were included in the study. Primary lesions on PET images were delineated, and extraction of first-, second-, and higher-order imaging features was performed using LIFEx software. The relationship between these parameters and pCR to NAC was analyzed by multiple logistic regression models.
Nineteen patients (24%) had pCR to NAC. Different models were generated on complete information and imputed datasets, using univariable and multivariable logistic regression and least absolute shrinkage and selection operator (lasso) regression. All models could predict pCR to NAC, with area under the curve values ranging from 0.70 to 0.73. All models agreed that tumor molecular subtype is the primary predictor of the primary endpoint.
Our models predicted that patients with subtype 2 and subtype 3 (HER2+ and triple negative, respectively) are more likely to have a pCR to NAC than those with subtype 1 (luminal). The association between PET imaging features and pCR suggested that PET imaging features could be considered as potential predictors of pCR in locally advanced breast cancer patients.
KeywordsBreast cancer Neoadjuvant chemotherapy Radiomics Advanced features 18F-FDG PET/CT Treatment response prediction
We would like to acknowledge Dr. Katia Marzo and Mrs. Elena Bissolotti for database preparation. The contribution of Drs. Elisa Agostinetto, Jelena Jandric, and Giulia Vatteroni to data collection is greatly acknowledged. We thank also Dr. Bethania Fernandes for histological analyses. We also acknowledge Drs. Alberto Testori and Valentina Errico for patients enrollment. We thank Mr. Lorenzo Leonardi for technical support and image acquisition.
The manuscript has been seen and approved by all authors, whose individual contributions were as follows:
Conception and design: LA, RDS, MS
Patient management and referral: RDS, ASagona, CT, RT, ASantoro
Acquisition of data: LA, MK, RDS
Image analysis: LA, MS, MK, LC
Statistical analysis: LA, RZ, RDS
Interpretation of data: LA, RDS, ASagona, MK, RZ
Drafting the article: LA, RDS, MS
Final approval of the revised manuscript: LA, RDS, LC, ASagona, CT, RT, ASantoro, MK, AC, RZ, MS.
Raw data are available on specific request to the corresponding author.
Compliance with ethical standards
Conflict of interest
A. Chiti received speaker honoraria from General Electric, Blue Earth Diagnostics, and Sirtex Medical, acted as scientific advisor for Blue Earth Diagnostics and Advanced Accelerator Applications, and benefited from an unconditional grant from Sanofi to Humanitas University. A. Santoro received speaker honoraria from Mundipharma, Takeda, Glaxo, Celgene, Roche, Teva, Arqule, Amgen, Eisai, BMS, Bayer, MSD, Astra Zeneca, Gilead, Sandoz, Servier, and Novartis, acted as scientific advisor for Gilead, Pfizer, Eisai, Servier, MSD, Bayer, Eli Lilly, Amgen, Marck Serono, Takeda, Ariad, Italfarmaco, and Celgene. L. Cozzi acts as Scientific Advisor to Varian Medical Systems and is Clinical Research Scientist at Humanitas Cancer Center.
All honoraria and grants are outside the scope of the submitted work.
All other authors have no conflict of interest.
The study was approved by the Local Ethics Committee (authorization number 1591). Specific informed consent was not required according to Local Ethics Committee rules for retrospective study design.
- 13.Kajáry K, Tőkés T, Dank M, Kulka J, Szakáll S, Lengyel Z. Correlation of the value of 18F-FDG uptake, described by SUVmax, SUVavg, metabolic tumour volume and total lesion glycolysis, to clinicopathological prognostic factors and biological subtypes in breast cancer. Nucl Med Commun. 2015;36:28–37.CrossRefGoogle Scholar
- 15.Lee SS, Bae SK, Park YS, Park JS, Kim TH, Yoon HK, et al. Correlation of molecular subtypes of invasive ductal carcinoma of breast with glucose metabolism in FDG PET/CT: based on the recommendations of the St. Gallen Consensus Meeting 2013. Nucl Med Mol Imaging (2010). 2017;51:79–85.CrossRefGoogle Scholar
- 16.Garcia Vicente AM, Soriano Castrejón A, Amo-Salas M, Lopez Fidalgo JF, Muñoz Sanchez MM, Alvarez Cabellos R, et al. Glycolytic activity in breast cancer using 18F-FDG PET/CT as prognostic predictor: a molecular phenotype approach. Rev Esp Med Nucl Imagen Mol. 2016;35:152–8.Google Scholar
- 23.Yoon H-J, Kim Y, Chung J, Kim BS. Predicting neo-adjuvant chemotherapy response and progression-free survival of locally advanced breast cancer using textural features of intratumoral heterogeneity on F-18 FDG PET/CT and diffusion-weighted MR imaging. Breast J. 2018:1–8. [Epub ahead of print]Google Scholar
- 24.Azad GK, Cousin F, Siddique M, Taylor B, Goh V, Cook GJR. Does measurement of first-order and heterogeneity parameters improve response assessment of bone metastases in breast cancer compared to SUVmax in [18F]fluoride and [18F]FDG PET? Mol Imaging Biol. 2018. [Epub ahead of print]. https://doi.org/10.1007/s11307-018-1262-3.
- 25.Gong C, Ma G, Hu X, Zhang Y, Wang Z, Zhang J, et al. Pretreatment 18F-FDG uptake heterogeneity predicts treatment outcome of first-line chemotherapy in patients with metastatic triple-negative breast cancer. Oncologist. 2018;23(10):1144–52. https://doi.org/10.1634/theoncologist.2018-0001.
- 26.Zwanenburg A, Leger S, Vallières M, Löck S, for the Image Biomarker Standardisation Initiative (IBSI). Image biomarker standardisation initiative — feature definitions 2016. https://arxiv.org/abs/1612.07003.
- 28.Lemarignier C, Martineau A, Teixeira L, Vercellino L, Espié M, Merlet P, et al. Correlation between tumour characteristics, SUV measurements, metabolic tumour volume, TLG and textural features assessed with 18F-FDG PET in a large cohort of oestrogen receptor-positive breast cancer patients. Eur J Nucl Med Mol Imaging. 2017;44:1145–54.CrossRefGoogle Scholar
- 29.Garcia-Vicente AM, Pérez-Beteta J, Amo-Salas M, Molina D, Jimenez-Londoño GA, Soriano-Castrejón AM, et al. Predictive and prognostic potential of volume-based metabolic variables obtained by a baseline 18F-FDG PET/CT in breast cancer with neoadjuvant chemotherapy indication. Rev Esp Med Nucl Imagen Mol. 2018;37:73–9.Google Scholar
- 31.Arboretti R, Salmaso L. Model performance analysis and model validation in logistic regression. Statistica. 2003;63:375–96.Google Scholar
- 33.Van Buuren S, Oudshoorn K. Flexible multivariate imputation by MICE TNO prevention and health. Leiden: TNO Prevention and Health; 1999.Google Scholar
- 35.Groheux D, Biard L, Lehmann-Che J, Teixeira L, Bouhidel FA, Poirot B, et al. Tumor metabolism assessed by FDG-PET/CT and tumor proliferation assessed by genomic grade index to predict response to neoadjuvant chemotherapy in triple negative breast cancer. Eur J Nucl Med Mol Imaging. 2018;45:1279–88.CrossRefGoogle Scholar
- 37.Airola A, Pahikkala T. A comparison of AUC estimators in small-sample studies. JMLR - Work Mach Learn Syst Biol. 2009; 8:3–13.Google Scholar