| 摘要: | 背景：近年的研究發現罹患婦科癌症的女性有較高的後續風險得到另一個原發癌症，然而對於後續衍生大腸癌的風險，尤其是罹患子宮頸癌後衍生大腸癌仍有爭議。此外，治療效應對於後續發生大腸癌的相關，包括放射線治療以及化學治療仍未確定。因此，本研究針對婦科癌症的女性患者進行後續發生大腸癌以及發生大腸息肉(癌前病變)的關係進行探討，並對放射線治療以及化學治療的效應進行進一步分析，希望藉由本研究為婦癌存活患者提議適當的大腸癌篩檢計畫。
方法：本研究根據台灣健保資料庫的全人口住院檔為基礎進行回溯性世代設計的研究。研究世代建置自1998年至2009年止，對子宮頸癌、子宮內膜癌以及卵巢癌等婦科案例進行追蹤分析到2010年初，評估後續的大腸癌以及大腸息肉和接受切除術的發生率，並進一步探討接受放射線治療以及化學治療的相關。利用Cox正比例涉險模組(Cox proportional hazards regression model)計算出婦癌患者發生大腸癌以及須接受切除的大腸息肉危害比值(hazard ratio, HR)，並利用Kaplan–Meier方法繪存活曲線，推估三種婦癌世代以及對照組間之累積發生百分率，探討後續發生大腸癌以及大腸息肉程度差異。
結果：研究期間有41452位至少活一年的三種婦癌女性，後續發生大腸癌的風險均明顯增加。子宮頸癌、子宮內膜癌以及卵巢癌婦女的大腸癌發生率分別為1.37, 2.20和1.76每10,000 人年；整體發生大腸癌的校正危害比值 (adjusted hazard ratio, aHR) 相較一般族群為高，和子宮頸癌、子宮內膜癌以及卵巢癌相關之大腸癌危害比依序為1.20 (95% CI: 1.03-1.40); 2.20 (95% CI: 1.77-2.90); 2.09 (95% CI: 1.59-2.76)。各自大腸癌發生的年齡別危害比，和子宮頸癌有關的，以40-49歲最高 (aHR = 1.67, 95% CI: 1.13-2.48)，其次50-64歲(aHR = 1.32, 95% CI: 1.02-1.70)，子宮內膜癌和卵巢癌的最高大腸癌發生危害比則為30-39歲婦女，其aHR分別為6.18 (95% CI: 2.11-18.1) 和 6.37 (95% CI: 2.71–15.0), 其次是年齡在40-49 歲的婦女其aHR 3.46 (95% CI: 2.05-5.84) 和2.83 (95% CI: 1.59–5.01)。比較接受不同治療方式的影響，三種未接受治療婦癌女性均有較高的大腸癌風險，而子宮內膜癌接受化療患者有最高的風險發大腸癌(HR: 3.39, 95% CI: 1.69-6.80)。以Cox模式逐年分析發生大腸癌危害比顯示子宮內膜癌及卵巢癌患者的世代在前三年的追蹤期間有較明顯的增加大腸癌的發生，而子宮頸癌患者的大腸癌則較多發生在罹病後4到7年(P<0.0001)。但子宮內膜癌在10年後的大腸癌HR又增加到5.57(95% CI: 1.97-15.7)。
結論：罹患婦科癌症患者有較高風險後續發生大腸癌，發生年齡較年輕，癌症較多發生在早期追蹤期間，接受過化療的子宮內膜癌患者尤其高。
Background: Recent studies have shown that gynecologic malignancy survivors are at higher risk of subsequent primary cancers. However, studies on the subsequent colon cancer risk have revealed controversial results, especially in the association with cervical cancer. The effect of radiotherapy (RT) and chemotherapy (CT) on the occurrence of subsequent colon cancer remains unclear. This study evaluated the relationship between the gynecologic cancers and subsequent colon cancer and colon polyps (precancerous tumors), with the consideration of the treatment effects of radiotherapy and chemotherapy. We sought to propose an optimal colon screening program for these survivors from this study.
Method: We performed a population-based retrospective cohort study using data obtained from National Health Research Institutes for the period from 1998 to 2009 to established a gynecological cancer cohort consisting of patients with cervical cancer (N = 25370), endometrial cancer (N = 8149) and ovarian cancer (N = 7933). A comparison cohort (N = 165808) without cancer was randomly selected frequency matched by age and disease date. We calculated incidence rates of subsequent colon cancer and subsequent polypectomy for both cohorts after one-year survival. Gynecologic cancer patients were stratified into groups by radiotherapy and chemotherapy for measuring the risk of colon cancer and polypectomy by the end of 2010. The Cox proportional hazards regression model was used to estimate the hazard ratio (HR) of colorectal cancer and polypectomy associated with the gynecologic cancers. We also used the Kaplan–Meier survival analysis to estimate the proportions of subjects with these three studied cancers developing subsequent colon cancers or polypectomy during the follow-up period in both cohorts.
Results: Overall, 41452 women with one-year survivors of these gynecologic cancers were included for data analysis. The incidence rates of colorectal cancer in those with cervical, endometrial and ovarian cancers were 1.26-, 2.20-, and 1.61-fold higher than that in the general population (1.09 per 10,000 person-years). Overall, the corresponding adjusted HRs for developing colorectal cancers were 1.20 (95% CI: 1.03-1.40), 2.26 (95% CI: 1.77-2.90), and 2.09 (95% CI: 1.59-2.76), respectively. The incidences of polypectomy were higher in women with these gynecologic cancers than general women, but not significant. The respective cancer-to-reference colorectal cancer incidence rate ratio was age dependent. In women with cervical cancer, the highest adjusted HR of colorectal cancer was 1.67 (95% CI 1.13-2.48) found for those aged 40-49 years old, followed by those aged 50-64 years old (adjusted HR 1.32, 95% CI 1.02-1.70). In women with endometrial and ovarian cancers 30-39 years of ages had the highest risk of colorectal cancer with adjusted HRs of 6.18 (95% CI: 2.11-18.1) and 6.37 (95% CI: 2.71–15.0), respectively, followed by those women aged 40-49 years, with adjusted HRs of 3.46 (95% CI: 2.05-5.84) and 2.83 (95% CI: 1.59–5.01). Among cancer patients receiving different therapeutic modalities, endometrial cancer patients receiving chemotherapy alone had the highest adjusted HR of 3.39 (95% CI: 1.69–6.80). Colorectal cancer appeared to be higher in the first three years for women after being diagnosed with endometrial and ovarian cancers (p <0.0001). The increased incidence of colorectal cancer appeared during 4th-7th years after women with cervical cancer diagnosed (p <0.0001).
Conclusion: Patients with gynecologic cancers are at an increased risk of developing colorectal cancers, the relative hazard is higher in the younger gynecologic cancers patients. Women receiving chemotherapy are also at higher risk, especially for those with endometrial cancer. |
