摘要
目的探讨非小细胞肺癌(NSCLC)组织学类型和分化程度对其^18F—脱氧葡萄糖(FDG)PET/CT显像标准摄取值(SUV)的影响。方法260例NSCLC患者,按组织学类型不同分为腺癌、鳞癌、腺鳞癌和其他类型4组,按组织分化不同分为Ⅰ(高分化)、Ⅱ(中分化)、Ⅲ(低分化+未分化)3个级别,回顾其治疗前的^18F—FDGPET/CT显像结果,测量肺内原发灶最大SUV(SUVmax)。用多元回归分析方法提取SUVmax的影响因素,采用协方差分析修正病灶大小(取最大长径)对SUVmax的影响后,比较不同组织学类型和分化程度NSCLC病灶SUVmax的差异。结果260例患者共260个原发NSCLC病灶,腺癌161个(10个细支气管肺泡癌归于Ⅰ级腺癌,Ⅰ级15个、Ⅱ级88个、Ⅲ级58个),鳞癌74个(Ⅰ级6个、Ⅱ级39个、Ⅲ级29个),腺鳞癌15个(Ⅱ级7个、Ⅲ级8个),其他类型10个。只有病灶大小(F=87.046,P〈0.001)、分化程度(F=87.604,P〈0.001)和组织学类型(F=66.663,P〈0.001)被纳入SUVmax多元回归方程,标准化回归系数分别为0.436(t=8.910,P〈0.001),0.391(t=8.322,P〈0.001),0.190(t=3.885,P=0.0001)。修正病灶大小对SUVmax的影响后,SUVmax由小到大依次为腺癌Ⅰ级〈腺癌Ⅱ级〈鳞癌Ⅰ级〈腺鳞癌Ⅱ级〈鳞癌Ⅱ级〈其他类型NSCLC〈腺癌Ⅲ级〈腺鳞癌Ⅲ级〈鳞癌Ⅲ级,其均数及95%可信区间分别为3.3(2.1~4.5)、6.0(5.5~6.6)、6.1(4.2~8.0)、6.6(4.8~8.4)、7.8(7.0~8.6)、8.1(6.6~9.6)、8.3(7.6~8.9)、8.7(7.0—10.4)、8.9(8.0~9.8)。分化较好的Ⅰ、Ⅱ级腺癌SUVmax均分别低于Ⅰ、Ⅱ级鳞癌,差异有统计学意义(q值分别为-2.786,-1.776,P值分别为0.017,〈0.001),分化较差的Ⅲ级腺癌、鳞癌以及腺鳞癌间SUVmax差异均无统计学意义(q值分别为-0.593,-0.422,0.171,P值分别为0.288,0.642,0.860);随细胞分化变差,腺癌、鳞癌及腺鳞癌组病灶SUVmax升高,腺癌Ⅰ、Ⅱ、Ⅲ级间(g值分别为-2.720,-4.943,-2.223,P均〈0.001)以及鳞癌Ⅰ、Ⅲ级间(g=-2.751,P=0.012)SUVmax差异均有统计学意义。结论NSCLC病灶的组织学类型和分化程度都可以影响其^18F-FDG摄取,分化程度对^18F—FDG摄取的影响意义更大。
Objective To determine the effect of histotype and histodifferenfiation on the maximum standardized uptake value ( SUVmax ) of non-small cell lung cancer ( NSCLC ) ^18F—fluorodeoxyglueose (FDG) PET/CT imaging. Methods Two hundred and sixty patients with NSCLC underwent ^18F—FDG PET/CT imaging. They were classified according to ( 1 ) histotype : as adenoeareinoma ( AC), squamous cell carcinoma (SQC) , adenosquamous carcinoma (ASC) and other type carcinoma (OTC) , and (2) histodifferentiation: as grade Ⅰ (well-differentiated), grade Ⅱ (moderate-differentiated) and grade Ⅲ (poor-differentiated). The SUVmax and size (long diameter) of the primary lesions were measured. Multivariate regression analysis was used to analyze the relationship between the SUVmax and variable factors including histotype, histodifferentiation, lesion size, age, sex, body height, body weight, body mass index (BMI) , blood glucose level, dose, and rate of dose. Results Two hundred and sixty patients had 260 primary NSCLC tumors. There were 161 AC (15 grade Ⅰ , 88 grade Ⅱ , 58 grade Ⅲ), 74 SQC (6 grade Ⅰ , 39 grade Ⅱ , 29 grade Ⅲ) , 15 ASC (7 grade Ⅱ , 8 grade Ⅲ) and OTC (8 large cell, 2 carcinosarcoma). Only lesion size ( F = 87. 046, P 〈 0.001 ), histodifferentiation ( F = 87. 604, P 〈 0.001 ) and histotype ( F = 66. 663, P 〈 0.00l ) were included for multivariate regression analysis with SUVmax. After adjustment for lesion size, the SUVmax (mean and 95% confidence interval) in ascending order was AC Ⅰ: 3.3(2.1-4.5), ACⅡ: 6.0(5.5-6.6), SQC 1: 6.1(4.2-8.0), ASC Ⅱ: 6.6(4.8-8.4), SQC Ⅱ: 7.8(7.0-8.6), OTC:8.1(6.6-9.6), AC Ⅲ: 8.3(7.6-8.9), ASC Ⅲ: 8.7(7.0-10.4), and SQC Ⅲ: 8.9(8.0-9.8). The SUVmax of AC Ⅰ was significantly lower than that of SQC Ⅰ (q = - 2.786, P =0.017) , same for AC Ⅱ and SQC Ⅱ (q = - 1. 776, P 〈0. 001 ), but no statistically significant differences were found among AC Ⅲ, ASC Ⅲ and SQC Ⅲ (q= -0.593, -0.422, 0. 171, P=0.288, 0.642, 0.860, respectively). For the same histotype lesions, the difference of SUVmax among AC Ⅰ, Ⅱand Ⅲwas statistically significant (q = -2.720, -4.943, -2.223, all P〈0.001), as also for SQC Ⅰ and Ⅲ (q = -2.751, P=0.012). Conclusion Histotype and histodifferentiation are significant correlative factors for ^18F-FDG uptake of NSCLC, with histodifferentiation being the factor with greater impact.
出处
《中华核医学杂志》
CAS
CSCD
北大核心
2010年第1期5-9,共5页
Chinese Journal of Nuclear Medicine