Correlation of 18F-FDG PET/CT with pathological features and survival in primary breast cancer. Issue 8 (August 2017)

• Record Type: Journal Article
• Authors: Has Şimşek, Duygu; Şanli, Yasemin; Külle, Cemil B.; Karanlik, Hasan; Kiliç, Berkay; Kuyumcu, Serkan; Önder, Semen; Özmen, Vahit
• Is Part Of: Nuclear medicine communications. Volume 38:Issue 8(2017:Aug.)
• Publication Date: 2017-08
• Subjects: breast cancer -- 18F-FDG PET/CT -- prognostic value; Nuclear medicine -- Periodicals; 616.07575
• DOI: 10.1097/MNM.0000000000000694
• Languages: English
• ISSNs: 0143-3636

Abstract

Abstract : Objective: The aim of this study was to determine the correlation between the primary tumor (PT) maximum standardized uptake value (SUVmax ) and breast cancer prognostic factors, overall survival, and relapse-free survival on the basis of histopathological and molecular characteristics. Patients and methods: In this retrospective study, 436 female patients with breast cancer were evaluated following a pretreatment 18 F-FDG PET/CT scan. The PT SUVmax and histopathological/molecular characteristics were determined from primary tumor tissues and analyzed using the Mann–Whitney U and Kruskal–Wallis tests. Results: The median SUVmax of 436 PT was 10.1 (1.7–72). The PT SUVmax values were higher in ER− versus ER+ ( P =0.001), PR− versus PR+ ( P =0.001), Her2+ versus Her2− ( P =0.01), Ki-67% of at least 20 versus Ki-67% of less than 20 ( P <0.001), histological grade 3 versus grade 1–2 ( P <0.001), nuclear pleomorphism score 3 versus score 1–2 ( P <0.001), and mitotic score 3 versus score 1–2 patients ( P <0.001). The lowest SUVmax levels were observed in the LumA group and the highest SUVmax levels were observed in the Her2 group ( P <0.001). LumA patients with PR values greater than 20% had lower PT SUVmax values than the patients with PR values of 20% or less ( P =0.023). The PT SUVmax was higher in patients with recurrence ( P =0.03) and died related to disease ( P <0.001) independent of time. Conclusion: The PT SUVmax showed a significant correlation with most of the prognostic factors and histopathological subtypes as a noninvasive tool. It is also usable in the prediction of tumor-related deaths or relapse independent of time. Our results could guide future studies to provide new histopathologic subtype definitions on the basis of new PR criteria.