Gammon M D, Wolff M S, Neugut A I, Terry M B, Britton J A, Greenebaum E, Hibshoosh H, Levin B, Wang Q, Santella R M
Division of Epidemiology, Columbia University School of Public Health, New York, New York 10032, USA.
Cancer Epidemiol Biomarkers Prev. 1996 Jun;5(6):467-71.
Small studies have examined, with conflicting results, whether breast cancer risk is increased among women exposed to high levels of chlorinated hydrocarbons, as measured in breast fat tissue or peripheral blood collected prior to treatment (pretreatment blood). For a population-based, case-control study, collection of pretreatment blood is a labor-intensive effort. An alternative is to collect blood from cases at interview, as is done for controls, after breast cancer treatment has commenced (posttreatment blood). It is unknown whether treatment affects blood levels of the organochlorines 1,1-dichloro-2,2-bis(p-chlorophenyl)ethylene (DDE) or polychlorinated biphenyls (PCBs). The purpose of this study was to determine whether pretreatment versus posttreatment blood samples yielded significantly different estimates of cumulative exposure to DDE and PCBs among newly diagnosed breast cancer patients. Two-ml blood samples were collected prior to and after treatment for breast cancer from 22 nonfasting women, ages 45-87 years, newly diagnosed with invasive disease. Treatment was defined as major surgery (mastectomy or node removal), radiation, hormones (tamoxifen), or chemotherapy. Pretreatment and posttreatment blood samples were assayed for DDE and PCBs in blinded, matched pairs. The reported concentrations (volume basis) were adjusted for estimated total plasma lipids. For DDE, mean differences in unadjusted [0.99 ng/ml; 95% confidence interval (CI), -0.36 to 2.34 ng/ml] and lipid-adjusted (0.05 microgram/g lipid; 95% CI, -0.04 to 0.13 microgram/g lipid) levels were small. For PCBs, the unadjusted (0.68 ng/ml; 95% CI, 0.05 to 1.30 ng/ml) and adjusted (0.070 microgram/g lipid; 95% CI, -0.009 to 0.149 microgram/g lipid) mean differences were of borderline statistical significance. The mean percent change in lipid-adjusted organochlorine levels did not vary substantially between treatment groups, except for those patients receiving chemotherapy [n = 5; 15.8% (DDE), 29.4% (PCBs)]. Adjusted mean differences also increased with increasing time between the pretreatment and posttreatment blood draws. In multiple regression models that included treatment, age, race, stage, and time between blood draws, only chemotherapy appeared to predict the percent change in adjusted pretreatment and posttreatment levels of DDE or PCBs (P = 0.10 and 0.06, respectively). Posttreatment blood samples drawn within 3 months of pretreatment samples, with the exception of those drawn after the commencement of chemotherapy, provide similar measures of DDE body burden levels among breast cancer cases. The use of blood samples collected after treatment, rather than before treatment, for characterizing PCB levels may lead to misclassification of exposure.
一些小型研究探讨了接触高浓度氯代烃的女性患乳腺癌风险是否增加，这些氯代烃是在治疗前采集的乳房脂肪组织或外周血（治疗前血液）中检测到的，结果相互矛盾。对于一项基于人群的病例对照研究而言，采集治疗前血液是一项耗费人力的工作。另一种方法是在乳腺癌治疗开始后（治疗后血液），像对对照组那样，在访谈时从病例中采集血液。尚不清楚治疗是否会影响有机氯1,1-二氯-2,2-双（对氯苯基）乙烯（DDE）或多氯联苯（PCBs）的血液水平。本研究的目的是确定治疗前与治疗后血液样本对新诊断乳腺癌患者中DDE和PCBs累积暴露的估计是否有显著差异。从22名年龄在45 - 87岁、新诊断为浸润性疾病的非空腹女性中，在乳腺癌治疗前后采集了2毫升血液样本。治疗定义为大手术（乳房切除术或淋巴结切除术）、放疗、激素（他莫昔芬）或化疗。对治疗前和治疗后血液样本进行双盲、配对检测DDE和PCBs。报告的浓度（基于体积）针对估计的总血浆脂质进行了调整。对于DDE，未调整的平均差异[0.99纳克/毫升；95%置信区间（CI），-0.36至2.34纳克/毫升]和脂质调整后的差异（0.05微克/克脂质；95% CI，-0.04至0.13微克/克脂质）较小。对于PCBs，未调整的（0.68纳克/毫升；95% CI，0.05至1.30纳克/毫升）和调整后的（0.070微克/克脂质；95% CI，-0.009至0.149微克/克脂质）平均差异具有临界统计学意义。脂质调整后的有机氯水平的平均百分比变化在各治疗组之间差异不大，接受化疗的患者除外[n = 5；15.8%（DDE），29.4%（PCBs）]。调整后的平均差异也随着治疗前和治疗后采血时间间隔的增加而增大。在包含治疗、年龄、种族、分期和采血时间间隔的多元回归模型中，只有化疗似乎能预测调整后的治疗前和治疗后DDE或PCBs水平的百分比变化（分别为P = 0.10和0.06）。除了化疗开始后采集的样本外，在治疗前样本采集后3个月内采集的治疗后血液样本，能提供乳腺癌病例中DDE体内负荷水平的类似测量值。使用治疗后而非治疗前采集的血液样本用于表征PCBs水平可能会导致暴露的错误分类。
