Metabolic Radiomics for Pretreatment 18F-FDG PET/CT to Characterize Locally Advanced Breast Cancer: Histopathologic Characteristics, Response to Neoadjuvant Chemotherapy, and Prognosis

Abstract

Radiomics has been spotlighted as imaging biomarker for estimation of intratumoral heterogeneity (ITH) which is regarded as the main reason for resistance to tumor treatment. Although a number of studies has shown clinical evidences that separate measurement of metabolic ITH by texture features (TFs) on ¹⁸F-fluorodeoxyglucose positron emission tomography/computed tomography (¹⁸F-FDG PET/CT) has prognostic ability in various tumors, there has been no consensus regarding the best parameter representing ITH. Besides, it is yet uncertain that TFs are useful for estimation of histopathologic markers, prediction of response to neoadjuvant chemotherapy (NAC), or prognostic ability in breast cancer. To depart from the traditional approach, we evaluated the clinical usefulness of integrated metabolic radiomics using unsupervised clustering with 109 TFs measured from pretreatment ¹⁸F-FDG PET/CT scans of 73 patients with locally advanced breast cancer (LABC) underwent NAC before surgery. Our study shows that metabolic radiomics patterns of LABC are associated with Ki67 expression, achievement of pathologic complete response after NAC, and risk of recurrence. Integrated metabolic radiomics has potential for clinically relevant pretreatment biomarker with predictive and prognostic ability for personalized management in LABC.

Figure 1

Inclusion and exclusion criteria for the study population. Seventy-three patients were involved in radiomics analysis. Sixty-six patients were involved in survival analysis. Abbreviations: NAC, neoadjuvant chemotherapy; TCs, tumor clusters.

Figure 2

Correlogram after multiple comparison correction (P < 0.0005). Correlogram shows close association of TFs to each other. Correlation coefficients are expressed by color scale from red to blue. Representative TFs are marked on the correlogram. Notes: c2, NL_Homogeneity^GLCM; c14, ZP^GLSZM; c16, Entropy^TFCCM; c24, MTV; c45, HILZE^GLSZM; c50, Skewness; c64, SUV_max; c66, NL_Dissimilarity^GLCM; c91, TLG; c104, CV; c105, NL_Entropy^GLCM. Abbreviations: TFs, texture features; NL, normalized; GLCM, gray level co-occurrence matrix; ZP, zone percentage; GLSZM, gray level size zone matrix; TFCCM, texture feature coding co-occurrence matrix; HILZE, high-intensity large-zone emphasis.

Figure 3

Unsupervised radiomics heat map with 109 texture features. Three individual tumor clusters (TCs) with distinctive metabolic radiomics patterns were identified after unsupervised clustering. Notes: Row, cases; column, texture features or clinical information; green circle, positive or high expression; red circle, negative or low expression; h2, ZP^GLSZM; h10, NL_Homogeneity^GLCM; h26, NL_Dissimilarity^GLCM; h29, SUV_max; h61, TLG; h70, CV; h71, NL_Entropy^GLCM; h91, HILZE^GLSZM; h98, MTV. Abbreviations: TNBC, triple negative breast cancer; HR, hormone receptor; ER, estrogen receptor; PgR, progesterone receptor; human epidermal growth receptor 2, HER2; ZP, zone percentage; GLSZM, gray level size zone matrix; NL, normalized; GLCM, gray level co-occurrence matrix; SUV_max, maximum of standardized uptake value; HILZE, high-intensity large-zone emphasis; MTV, metabolic tumor volume.

Figure 4

Cox regression analysis with DFS. The TC I has a hazard ratio of 19.755 (P = 0.010) for recurrence (a). Stage III had a trend of poor prognosis compared with stage II; however, this was not statistically significant (P > 0.05) (b). A pCR had a trend of a favorable prognosis compared with non-pCRs but this was not statistically significant (P > 0.05) (c). Abbreviations: DFS, disease free survival; TCs, tumor clusters; AJCC, American Joint Committee on Cancer; NAC, neoadjuvant chemotherapy; pCR, pathologic complete response.