Lian Weiling, Liu Cheng, Gu Bingxin, Zhang Jianping, Lu Linjun, Pan Herong, Yao Zhifeng, Wang Mingwei, Song Shaoli, Zhang Yingjian, Yang Zhongyi
Department of Nuclear Medicine, Fudan University Shanghai Cancer Center.
Department of Oncology, Shanghai Medical College, Fudan University.
Nucl Med Commun. 2020 Mar;41(3):280-287. doi: 10.1097/MNM.0000000000001145.
To compare the predictive value of European Organization for Research and Treatment of Cancer (EORTC) criteria and PET Response Criteria in Solid Tumors (PERCIST), for the pathological response and prognosis of patients with breast cancer receiving neoadjuvant chemotherapy (NAC).
Consecutive PET/computed tomography scans in 128 operable female patients at baseline and after two courses of NAC were performed. Patients were categorized by complete metabolic response (CMR) and non-CMR groups using 2 PET criteria. CMR and non-CMR were used to predict pathological complete response (pCR) by diagnostic test evaluation, and to predict progression-free survival (PFS) using Kaplan-Meier plots and Cox proportional hazards regression.
Ninety-two patients were finally analyzed. The sensitivity, specificity, and accuracy for pCR prediction were 69.7, 76.3, and 73.9% with EORTC criteria, and 69.7, 77.9, and 75.0% with PERCIST, respectively. Peak standardized uptake value normalized to lean body mass (SULpeak), maximum standardized uptake value (SUVmax), total lesion glycolysis (TLG), and metabolic tumor volume (MTV) were pCR response with accuracy of 70.7, 60.0, 75.0, and 71.4%, respectively. CMR by the EORTC (P = 0.021) and PERCIST (P = 0.007) was significantly related to a longer PFS. The univariate and multivariate analysis suggested that CMR by PERCIST was an independent predictor of recurrence (P = 0.008).
EORTC criteria and PERCIST had early predictive value to long-term outcome, but moderate value for pCR. Furthermore, PERCIST might show more potential than the EORTC criteria and conventional PET-based parameters to predict prognosis in breast cancer patients following two cycles of neoadjuvant chemotherapy.Video abstract: see http://links.lww.com/NMC/A162.
比较欧洲癌症研究与治疗组织(EORTC)标准和实体瘤PET反应标准(PERCIST)对接受新辅助化疗(NAC)的乳腺癌患者病理反应和预后的预测价值。
对128例可手术的女性患者在基线及两个疗程NAC后进行连续的PET/计算机断层扫描。使用2种PET标准将患者分为完全代谢反应(CMR)组和非CMR组。通过诊断试验评估,用CMR和非CMR预测病理完全缓解(pCR),并使用Kaplan-Meier曲线和Cox比例风险回归预测无进展生存期(PFS)。
最终分析了92例患者。EORTC标准预测pCR的敏感性、特异性和准确性分别为69.7%、76.3%和73.9%,PERCIST标准分别为69.7%、77.9%和75.0%。以去脂体重标准化的峰值标准化摄取值(SULpeak)、最大标准化摄取值(SUVmax)、总病变糖酵解(TLG)和代谢肿瘤体积(MTV)预测pCR反应的准确性分别为70.7%、60.0%、75.0%和71.4%。EORTC标准(P = 0.021)和PERCIST标准(P = 0.007)的CMR与更长的PFS显著相关。单因素和多因素分析表明,PERCIST标准的CMR是复发的独立预测因素(P = 0.008)。
EORTC标准和PERCIST标准对长期预后有早期预测价值,但对pCR的预测价值中等。此外,在预测接受两个周期新辅助化疗的乳腺癌患者预后方面,PERCIST标准可能比EORTC标准和传统的基于PET的参数更具潜力。视频摘要:见http://links.lww.com/NMC/A162 。