代謝症候群與大腸直腸腺腫與腺癌 之流行病學與預防

Abstract

背景 以往大腸直腸癌的預防以次段預防篩檢發現腺腫加以切除，進而阻斷大腸直腸癌自然病史為主，近年來許多研究顯示胰島素阻抗與大腸直腸癌風險相關。而胰島素阻抗是代謝症候群的主軸，因此釐清代謝症候群與大腸直腸癌或腺腫之風險相關，及量化其如何影響自然病史中哪個步驟將有助於進行大腸直腸癌防治策略及代謝症候群介入對死亡率及發生率降低之成本效益分析。 研究目的 (一)探討代謝症候群或其各個組成因子與大腸直腸腺腫風險之相關，及其性別差異。(二)探討代謝症候群影響大腸直腸癌癌化過程，包括：1)小腺腫之發生 2)小腺腫轉移至大腺腫 3) 大腺腫轉移至腺癌症前期 4)腺癌症前期轉移至腺癌期。釐清其作用於自然病史哪個階段，及其性別差異。(三)進行大腸直腸癌次段預防篩檢策略(每兩年糞便潛血篩檢)或結合代謝症候群防治策略(運動介入或藥物介入)組合之成本效益分析。 材料與方法 根據「基隆市社區闔家歡篩檢」1999~2004年間(KCIS)中40歲以上參與糞便潛血篩檢個案(包括重複篩檢個案) 第一次篩檢個案共43100位及參加第二次以上重複篩檢個案共25180位作為研究樣本。利用以族群篩檢為基礎之觀察性前瞻性世代研究 ( Observational prospective cohort study) 分析代謝症候群或其組成因子如何影響大腸直腸腺腫之發生。利用馬可夫模式估計分析代謝症候群對大腸直腸癌自然病史影響。利用蒙地卡羅模擬針對是否介入代謝症候群及其他防治策略組合進行成本效益分析。 結果 代謝症候群為大腸直腸腺腫之危險因子，在調整相關變項後男性勝算比1.42(95%CI, 1.05-1.92)具統計相關，女性勝算比1.31(95%CI, 0.85-2.02)則未達統計顯著。就代謝症候群各別組成而言，男性危險因子為高三酸油酯勝算比1.67(95%CI, 1.22-2.30)，女性則為腰圍異常勝算比1.61(95%CI, 1.04-2.51)及空腹血糖偏高勝算比1.64(95%CI, 1.01-2.64)。 若將非經腺腫之大腸直腸癌病史(De novo pathway)癌化機轉加以考慮進行馬可夫五階段自然病史估計顯示大腸直腸小腺腫(≦1公分)年發生率以男性為高(0.0026)，此發生率受代謝症候群的影響，女性相對危險性2.63 (95%CI, 1.88~3.69)較男性1.93 (95%CI, 1.46~2.55)大，且代謝症候群對於小腺腫之後轉變為大腺腫至腺癌症前期至臨床期的自然病史過程影響皆未達統計相關。但是腺腫後自然病史的演進卻是女性較快。 以降低死亡率為主的成本效益分析發現，以無篩檢介入為指標、單獨FOBT篩檢、FOBT篩檢合併運動介入、FOBT篩檢合併藥物(Metformin)及運動介入其效益分別為降低大腸直腸癌死亡率16%, 21%, 22%。 以增加成本效益指標(ICER) 的成本效益分析發現，以機率性成本效益分析來看，單獨FOBT相對於無任何介入者，FOBT合併運動介入相對於無任何介入，FOBT合併藥物及運動介入策略相對於無任何介入，平均救活一人年需要花費約＄16963.3727 (5861.99~25035.79)＄451370.546(87060.15-1478313.57), ＄341047.83 (94068.36-952026.42)。 以降低發生率為主的成本效益分析發現，相對於無任何篩檢及介入組，單獨FOBT篩檢、FOBT合併運動介入、FOBT合併運動及藥物介入其發生率降低分別15%, 20%, 20%。 以減少大腸直腸癌個案為主的成本效益分析發現，以機率性成本效益分析來看相對於單獨大腸直腸癌篩檢組，FOBT合併運動介入、FOBT合併運動及藥物介入平均每避免一例個案需要花費為＄3284.36 (1372.14-4774.53)、＄68315.26 (13069.19-218644.52)、＄52016.45(13687.7-140681.43)。 結論 本研究證實代謝症候群為大腸直腸腺腫之危險因子，在調整相關變項之下男性具統計相關，女性則未達統計顯著。代謝症候群不同組成對大腸直腸腺腫危險性分析，發現男性影響因子為高三酸甘油酯，女性為異常腰圍及早期血糖異常。五階段大腸直腸癌自然病史分析在考量非經腺腫之大腸直腸癌病史(De novo pathway)之量性估計下釐清代謝症候群影響小腺腫的發生，大腸直腸腺腫發生率男性較高，腺腫之後自然病史的演進卻以女性為快。成本效益分析由初段預防透過代謝症候群介入加上次段預防篩檢，要很高的花費才具成本效益，但若將代謝症候群介入可能減少之慢性病甚至其他癌症之死亡一併考量，可能具成本效果。未來需要進行更多相關此方向的研究。

Background The mainstay for preventing colorectal cancer (CRC) is the secondary prevention to block its natural history. As insulin resistance may induce metabolic syndrome (MS), the natural history of CRC influenced by MS, and the cost-effectiveness analysis can not be overemphasized. Research Purpose The aims of the study are to: (1) Investigate the association between MS or each component and adenoma or CRC stratified by gender. (2) Quantify the effect of MS on the disease natural history of CRC. (3) Do cost-effectiveness analysis of the primary prevention for MS by intensive lifestyle intervention or prophylactic medication, and secondary prevention by fecal occult blood test (FOBT) biennially. Materials and Methods Data sources are based on the Keelung Community-based Integrated Screening (KCIS) between1999 and 2004, the study subjects aged 40 years or older with the uptake of FOBT included the prevalent screen with 43100 subjects and the subsequent screens with 25180 subjects. The observational prospective cohort study from KCIS is designed to investigate the association between MS or its each component and CRC. Markov model was used to estimate the parameters pertaining to natural history of CRC. Decision tree analysis was applied to perform cost-effectiveness analysis by four preventive strategies of CRC including strategy (A) no screening (B) FOBT biennial (C) FOBT biennial + intensive lifestyle intervention (D) FOBT biennial + intensive lifestyle intervention + metformin intervention.

Results MS was positively associated with colorectal adenoma in male but not in female. The odds ratio is 1.42(95%CI, 1.05~1.92) for male and 1.31(95%CI, 0.85~2.02) for female. By each components of MS, the odds ratios are 1.67(95%CI, 1.22~2.30) for hypertriglycemia in male, 1.61(95%CI, 1.04~2.51) for abnormal waist circumference and 1.64(95%CI, 1.01~2.64) for abnormal fasting blood sugar in female. The incidence rate of small adenoma was 0.0026 for male after taking De novo pathway into account under the context of five-state model. Relative risks of small adenoma incidence are 2.63(95%CI, 1.88~3.69) for MS versus non-MS in female and 1.93(95%CI, 1.46~2.55) in male. We do not find the effect of MS either on small adenoma, large adenoma, or the progression from preclinical CRC to CRC. Comparing three preventive strategies against no screening (A), the CRC mortality reduction rates are 16% for (B), 21% for (C) and 22% for (D). Incremental cost ratios per life saved are ＄16963.37 (5861.99~25035.79) for (B)＄451370.55 (87060.15-1478313.57) for (C) and＄341047.83 (94068.36-952026.42) for (D)。Reducing the CRC incidence rates are 15% for (B), 20% for (C) and 20% for (D) compared with (A). The ICER to avert one additional CRC case are ＄3284.36 (1372.14-4774.53)for(B),＄68315.26 (13069.19-218644.52)for (C)and ＄52016.45(13687.7-140681.43)for (D)。 Conclusion Adenoma plays a role in occurrence of CRC in male but not in female. By each component of MS, risk factors are hypertriglycemia for male, abnormal waist circumference and fasting blood sugar for female. Using the five-state Markov models of CRC, the effect of MS on the disease natural history of CRC is remarkably manifested in occurrence of small adenoma but not on from small adenoma, large adenoma to preclinical CRC or CRC. The cost seems too high for adding the primary prevention to the existing secondary prevention of CRC using biennial FOBT. Economic appraisal may take into account other chronic diseases or neoplasm affected by MS. Further studies for this viewpoint are needed.