Division of Surgical Oncology, Department of Surgery, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya, Aichi, 466-8550, Japan.
Department of Radiology, Nagoya University Graduate School of Medicine, Nagoya, Japan.
Int J Clin Oncol. 2023 Mar;28(3):482-490. doi: 10.1007/s10147-022-02287-6. Epub 2022 Dec 30.
Survival in patients with retroperitoneal liposarcoma (RPLS) depends on the surgical management of the dedifferentiated foci. The present study investigated the diagnostic yield of contrast-enhanced CT, F-fluorodeoxyglucose positron emission tomography (PET), and diffusion-weighted MRI in terms of dedifferentiated foci within the RPLS.
Patients treated with primary or recurrent RPLS who underwent the above imaging between January 2010 and December 2021 were retrospectively reviewed. The diagnostic accuracy of the three modalities for histologic subtype of dedifferentiated liposarcoma (DDLS) and French Federation of Cancer Center (FNCLCC) grade 2/3 were compared using receiver operating characteristic curves and areas under the curves (AUCs).
The cohort involved 32 patients with 53 tumors; 30 of which exhibited DDLS and 31 of which did FNCLCC grades 2/3. The optimal thresholds for predicting DDLS were mean CT value of 31 Hounsfield Unit (HU) (AUC = 0.880, 95% CI 0.775-0.984; p < 0.001), maximum standardized uptake value (SUVmax) of 2.9 (AUC = 0.865 95% CI 0.792-0.980; p < 0.001), while MRI failed to differentiate DDLS. The cutoff values for distinguishing FNCLCC grades 1 and 2/3 were a mean CT value of 24 HU (AUC = 0.858, 95% CI 0.731-0.985; p < 0.001) and SUVmax of 2.9 (AUC = 0.885, 95% CI 0.792-0.978; p < 0.001). MRI had no sufficient power to separate these grades.
Contrast-enhanced CT and PET were useful for predicting DDLS and FNCLCC grade 2/3, while MRI was inferior to these two modalities.
腹膜后脂肪肉瘤(RPLS)患者的生存取决于去分化灶的手术治疗。本研究探讨了增强 CT、F-氟代脱氧葡萄糖正电子发射断层扫描(PET)和弥散加权 MRI 对 RPLS 内去分化灶的诊断效能。
回顾性分析 2010 年 1 月至 2021 年 12 月期间接受上述影像学检查的原发性或复发性 RPLS 患者。使用受试者工作特征曲线和曲线下面积(AUC)比较三种方法对去分化脂肪肉瘤(DDLS)的组织学亚型和法国癌症中心联合会(FNCLCC)分级 2/3 的诊断准确性。
该队列包括 32 例患者的 53 个肿瘤;其中 30 个为 DDLS,31 个为 FNCLCC 分级 2/3。预测 DDLS 的最佳阈值为平均 CT 值 31 亨氏单位(HU)(AUC=0.880,95%CI 0.775-0.984;p<0.001),最大标准化摄取值(SUVmax)2.9(AUC=0.865,95%CI 0.792-0.980;p<0.001),而 MRI 无法区分 DDLS。区分 FNCLCC 分级 1 和 2/3 的截断值为平均 CT 值 24 HU(AUC=0.858,95%CI 0.731-0.985;p<0.001)和 SUVmax 2.9(AUC=0.885,95%CI 0.792-0.978;p<0.001)。MRI 没有足够的能力来区分这些分级。
增强 CT 和 PET 对预测 DDLS 和 FNCLCC 分级 2/3 有用,而 MRI 逊于这两种方法。