PET/CT radiomics in breast cancer: Mind the step
Graphical abstract
Introduction
An estimated risk or probability of the presence of a specific disease or condition (diagnostic setting) or of a specific future outcome (prognostic setting) provided by a prediction model determines treatment and follow-up decisions. However, studies developing or validating multivariable prediction models based on a radiomics approach have been found to be of insufficient quality, since information on all aspects of model development has been incomplete [1]. Such limitations ultimately lead to prediction models that cannot be implemented in clinical practice.
In this context, however, biomarkers of tumor aggressiveness and potential resistance to medical treatments would be of fundamental importance in order to inform treatment decisions. Novel approaches are necessary to avoid potentially toxic and ineffective chemotherapies, to target aggressive cancer foci, and to develop patient-tailored therapeutic approaches.
Imaging, which allows non-invasive, repeatable, whole-body assessment, has the potential to provide image-derived biomarkers. Numerous parameters (features) can be extracted by applying mathematical methods describing the relationships between the intensity of the pixels or voxels and their position within the region of interest. Identification of correlations between these potential biomarkers and relevant outcomes represents the essence of radiomics. This approach has been evaluated on different imaging modalities and appears promising in breast cancer (BC) [2]. BC is the second most common cancer in the global population and the most frequent cancer occurring in women worldwide. Although overall survival rates are improving globally thanks to better screening programs and early stage diagnosis, it still represents the fifth most common cause of female death [3]. Diagnosis and tumor biology characterization, staging, treatment response, and outcome prediction represent the unmet needs in BC patient management. The aim of the present review was to assess the current status of positron emission tomography/computed tomography (PET/CT) radiomics research in BC, and in particular to analyze the strengths and weaknesses of the published papers in order to identify challenges and suggest possible solutions and future research directions.
Section snippets
Search, eligibility criteria, and study selection
Combinations of the terms “breast” AND “radiomic” AND “PET”, “breast” AND “radiomics” AND ”PET“, ”breast“ AND “texture” AND ”PET“, and ”breast“ AND “textural” AND ”PET“ were used for the literature search within the PubMed/MEDLINE database. No start date limit was used, and the search was extended until 8 July 2019. We applied the following inclusion criteria: (a) articles on breast cancer and (b) articles on texture analysis derived from PET/CT. We excluded (a) articles not in English language;
Diagnosis
Currently, definitive diagnosis of suspicious breast nodules relies on core-needle biopsy or surgery. A commonly encountered diagnostic dilemma is nodule classification as benign vs malignant, especially with respect to those lesions that are classified as probably benign (BI-RADS 3) at imaging [4]. Vogl et al. [5] investigated the potential of textural features (TFs) derived from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), diffusion weighted imaging (DWI), and 18
Discussion
Despite encouraging results, the included studies are far from providing definitive conclusions. Radiomics investigations include clinical, technical, and analysis steps, starting with identification of the clinical need, hypothesis generation, and study design, and proceeding to image acquisition and reconstruction, segmentation of the region of interest, processing of the image, actual computation of the texture parameters, collection of other relevant data (clinical parameters, outcome,
Conclusions
We reviewed up-to-date literature on the role of PET radiomics in breast cancer, focusing on methodological aspects of the radiomics workflow and possible pitfalls. Variations in acquisition, reconstruction, segmentation, and radiomics processing were found among studies. Therefore, most of the current evidence on the clinical role is at feasibility level. The shortcomings discussed in the review can be mitigated when corrective strategies are applied. Such strategies lie mainly in the use of
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgement
MK PhD scholarship was funded by the Italian Association for Research on Cancer (AIRC) grant number IG-2016-18585.
Funding
This research did not receive any grant from funding agencies in the public, commercial, or not-for-profit sectors.
Ethical statement
In view of the design of the present study (i.e., review), ethical approval was considered unnecessary.
Author contributions
MS and MK conceptualized the study; MS and GN participated in data collection and selection; MS, LCozzi, GN, and LCavinato performed data analysis; MK and MS supervised data selection; LA, MS, LCozzi, GN, LCavinato, and MK drafted the paper; AC supervised the activities; all the authors read, commented on, and approved the manuscript.
References (51)
- et al.
Radiomics: Extracting more information from medical images using advanced feature analysis
Eur. J. Cancer
(2012) - et al.
Histological types of breast cancer: how special are they?
Mol. Oncol.
(2010) - et al.
on behalf of the ESMO Guidelines Committee, Primary breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up
Ann. Oncol.
(2015) Prospective versus retrospective studies
Ophthalmology
(1992)Sample size estimation in diagnostic test studies of biomedical informatics
J. Biomed. Inform.
(2014)- et al.
4th ESO–ESMO international consensus guidelines for advanced breast cancer (ABC 4)†
Ann. Oncol.
(2018) - et al.
Beyond imaging: The promise of radiomics
Phys. Medica.
(2017) - et al.
Radiomics and artificial intelligence for biomarker and prediction model development in oncology
Comput. Struct. Biotechnol. J.
(2019) - et al.
Repeatability and reproducibility of radiomic features: a systematic review
Int. J. Radiat. Oncol.
(2018) - et al.
Quality of science and reporting of radiomics in oncologic studies: room for improvement according to radiomics quality score and TRIPOD statement
Eur. Radiol.
(2019)
Global cancer statistics GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries
CA Cancer J. Clin.
BI-RADS 3: Current and future use of probably benign
Curr. Radiol. Rep.
Automatic segmentation and classification of breast lesions through identification of informative multiparametric PET/MRI features
Eur. Radiol. Exp.
International Agency for Research on Cancer
Ability of 18 F-FDG PET/CT Radiomic Features to Distinguish Breast Carcinoma from Breast Lymphoma
Contrast Media Mol. Imaging
Highlights from the 14(th) St Gallen International Breast Cancer Conference 2015 in Vienna: Dealing with classification, prognostication, and prediction refinement to personalize the treatment of patients with early breast cancer
Ecancermedicalscience
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
Intratumoral metabolic heterogeneity predicts invasive components in breast ductal carcinoma in situ
Eur. Radiol.
Do clinical, histological or immunohistochemical primary tumour characteristics translate into different 18F-FDG PET/CT volumetric and heterogeneity features in stage II/III breast cancer?
Eur. J. Nucl. Med. Mol. Imaging
Exploration of PET and MRI radiomic features for decoding breast cancer phenotypes and prognosis
NPJ Breast Cancer
Metabolic radiomics for pretreatment 18F-FDG PET/CT to characterize locally advanced breast cancer: histopathologic characteristics, response to neoadjuvant chemotherapy, and prognosis
Sci. Rep.
[18F]FDG PET/CT features for the molecular characterization of primary breast tumors
Eur. J. Nucl. Med. Mol. Imaging
Texture analysis of high-resolution dedicated breast 18 F-FDG PET images correlates with immunohistochemical factors and subtype of breast cancer
Eur. J. Nucl. Med. Mol. Imaging
Comparison of the volumetric and radiomics findings of 18F-FDG PET/CT images with immunohistochemical prognostic factors in local/locally advanced breast cancer
Nucl. Med. Commun.
Intratumoral heterogeneity in 18F-FDG PET/CT by textural analysis in breast cancer as a predictive and prognostic subrogate
Ann. Nucl. Med.
Cited by (40)
AI-Enhanced PET and MR Imaging for Patients with Breast Cancer
2023, PET ClinicsExplainable domain transfer of distant supervised cancer subtyping model via imaging-based rules extraction
2023, Artificial Intelligence in Medicine
- 1
ORCID 0000-0003-2214-6492.
- 2
ORCID 0000-0001-8207-1256.
- 3
ORCID 0000-0003-1832-1083.
- 4
ORCID 0000-0002-5806-1856.
- 5
ORCID 0000-0002-3923-1151.