结直肠癌某亚型与肥胖和活动不足相关

《癌症研究》3月1日发表的一项针对861例结直肠癌患者的前瞻性研究数据分析表明，肥胖和缺乏身体活动人群罹患一种特定分子亚型结直肠癌的几率较大(2013;73:1600-10)。

波士顿Dana-Farber癌症研究所的Teppei Morikawa博士及其同事报告称，体重指数(BMI)每增加5 kg/m²，生物标志物β-连环蛋白(CTNNB1)阴性结直肠癌发病几率显著增加34%，但与CTNNB1阳性结直肠癌发病几率无关。

身体活动水平以代谢当量(MET)分值表示，定义为休闲时间活动(散步、特殊锻炼、运动或其他剧烈活动)代谢率除以静息代谢率，每项活动的MET分值相加即为每周总MET小时(MET-h/w)。每增加10 MET-h/w，CTNNB1阴性结直肠癌风险显著降低7%，但CTNNB1阳性结直肠癌风险未见明显变化。

换言之，肥胖程度越高和身体活动越少与CTNNB1阴性结直肠癌风险显著增加相关，但与CTNNB1阳性结直肠癌风险无关，这表明人体能量平衡和代谢可影响不同亚型结直肠癌的特殊致癌过程。

研究者指出，既往研究表明WNT信号通路活化在结直肠癌发生过程中起着关键作用，而CTNBB1是WNT通路中的主要介质，且有研究提示WNT-CTNNB1在脂肪形成、肥胖、糖代谢和代谢性疾病发病过程中具有信号传导作用。本研究结果表明，能量代谢对致癌作用的影响方式较少依赖于WNT/CTNNB1活化。

结肠癌通常作为单一疾病进行治疗，大多数以人口为基础的研究也没有考虑肿瘤的异质性。研究者认为，进一步开展分子病理学的流行病学研究，或将有助于确认究竟哪些人群易患CTNNB1阴性结直肠癌，CTNNB1能否作为治疗靶点，以及制定不同人群的预防策略。

研究者利用护士健康研究(NHS) 109,046例女性受试者和健康专业人员随访研究(HPFUS) 47,684例男性受试者的随访资料，确认了1986~2004年确诊且具有核CTNNB1表达状态资料的861例结直肠癌患者。他们查阅了受试者病历和病理报告，并通过国家死亡索引(NDI)确认死于未报告癌症的情况。研究者对受试者肿瘤标本行免疫组化确认CTNNB1表达状况，采用盲法解读所有病例CTNNB1结果。

研究者应用Cox比例风险模型进行风险分析，并校正了已知与结直肠癌风险有关的其他因素，包括年龄，叶酸、维生素D和钙摄入量，总热量摄入量，红肉消费量，吸烟状况，30岁以前年吸烟包数，酒精摄入量，多维素或阿司匹林应用，乙状结肠镜或结肠镜检查，结直肠癌家族史，绝经以及绝经后激素治疗情况。

结果显示，无论男性或女性亚组，还是所有患者， BMI较高均与CTNNB1阴性结直肠癌风险增加相关。与BMI为18.5~22.9 kg/m²的患者相比，BMI为27.5~29.9 kg/m²的患者罹患CTNNB1阴性结直肠癌风险增加77%，BMI≥30 kg/m²的患者风险增加84%。

对活动水平的分析研究纳入了具有活动资料且可确定CTNNB1状态的767例患者。结果显示，男性或女性亚组活动水平似乎均与CTNNB1阴性结直肠癌较低风险相关，但男性亚组的相关性未达到统计学显著水平。男女亚组合并人群的相关性分析显示，身体活动水平增加可降低CTNNB1阴性结直肠癌风险7%，虽然降低幅度较小但仍具有统计学意义。

综合BMI和活动水平两个因素的探索性分析显示， 与BMI较低(＜25 kg/m²)且活动水平较高(≥9 MET-h/w)的患者相比，BMI 较高(≥25 kg/m²)且活动水平较低(＜9 MET-h/w) 的患者CTNNB1阴性结直肠癌风险增加82%，但BMI和活动水平与CTNNB1阳性结直肠癌无关。

Morikawa博士报告无利益冲突。

爱思唯尔版权所有  未经授权请勿转载

By: SHERRY BOSCHERT, Internal Medicine News Digital Network

A specific molecular subtype of colorectal cancer is more likely in people with greater obesity and physical inactivity, analyses of prospective data on 861 cancer patients found.

For every 5-kg/m2 increase in body mass index, the likelihood that a biomarker called CTNNB1 would be absent in colorectal cancer tumors increased significantly by 34%. Increasing BMI was not associated with CTNNB1-positive tumors, however, Dr. TeppeiMorikawa and his associates reported.

Activity levels were reported as a metabolic equivalent task (MET) score calculated as the metabolic rate for leisure time activities such as walking, specific exercises, sports, or other vigorous activities divided by the resting metabolic rate. MET scores for individual activities were added together as total MET hours per week (MET-h/week). For every 10 MET-h/week increase in activity, the risk of CTNNB1-negative cancer decreased significantly by 7% but the chances of CTNNB1-positive colorectal cancer did not change significantly.

In other words, greater obesity and physical inactivity were associated with significantly higher risk for CTNNB1-negative but not CTNNB1-positive colorectal cancer. This suggests that a person’s energy balance and metabolism affect a specific pathway of carcinogenesis that is different from that of some other colorectal cancer subtypes, reported Dr. Morikawa of the Dana-Farber Cancer Institute, Boston.

The findings were published in the March 1 issue of Cancer Research (2013;73:1600-10).

Previous studies have pointed to the activation of a network of proteins known as the WNT signaling pathway as a critical player in colorectal carcinogenesis. Beta-catenin(CTNBB1)  is a major mediator of the WNT pathway, and previous studies have implicated WNT-CTNNB1 signaling in adipogenesis, obesity, glucose metabolism, and metabolic disease. The current findings suggest that energy metabolism affects carcinogenesis in ways that are less dependent on WNT/CTNNB1 activation, Dr. Morikawa reported.

Colon cancer typically is treated as a single disease, and most population-based studies do not consider tumor heterogeneity. Additional epidemiologic studies of the molecular pathology might help identify people who are susceptible to CTNNB1-negative colorectal tumors, determine if CTNNB1 could be a target of treatment, or illuminate prevention strategies for subsets of people, the investigators suggested.

They tapped data on 109,046 women from the Nurses’ Health Study and 47,684 men in the Health Professionals Follow-Up Study to analyze the records of 861 patients who developed rectal and colon cancers between 1986 and 2004 and who had tissue immunohistochemistry data on nuclear CTNNB1 expression.

The current investigators reviewed participants’ medical records and pathology reports and identified deaths from unreported cancers through the National Death Index. They obtained paraffin-embedded tissue blocks from hospitals where participants had undergone tumor resection to confirm colorectal cancer and to conduct immunohistochemistry for CTNNB1 expression. Dr. Morikawa was blinded to other study data when he interpreted the CTNNB1 results in all cases.

The association with higher BMI was seen in subgroups of men or women and in the combined cohort. Compared with patients who had BMIs of 18.5-22.9 kg/m2 , BMIs of 27.5-29.9 kg/m2 were associated with a 77% increased likelihood of CTNNB1-negative colorectal cancer, and BMIs of 30 kg/m2 or greater were associated with an 84% increased risk of CTNNB1-negative cancer.

The analysis of activity levels included 767 patients with activity data whose CTNNB1 status could be determined. Activity level seemed to be associated with lower risk of CTNNB1-negative cancer in subgroups based on sex, but this did not reach statistical significance in men. Combining the two subgroups produced the small but statistically significant 7% lower risk for CTNNB1-negative cancer with increasing levels of physical activity.

An exploratory analysis that combined BMI and activity level suggested that patients with high BMI (25 kg/m2 or greater) and low activity (less than 9 MET-h/week) were 82% more likely to have CTNNB1-negative colorectal cancer compared with patients with low BMI (less than 25 kg/m2) and high activity level (9 or more MET-h/week). BMI and activity level combined were not associated with CTNNB1-positive cancer.

The Cox proportional hazards model used for risk analysis controlled for the effects of other factors that have been associated with colorectal cancer risk, including age; intakes of folate, vitamin D, and calcium; total caloric intake; red meat consumption; smoking status; pack-years of smoking before age 30 years; alcohol intake; use of multivitamins or aspirin; having had a sigmoidoscopy or colonoscopy; family history of colorectal cancer; menopausal status; and use of postmenopausal hormone therapy.

Dr. Morikawa reported having no financial disclosures