乳癌病人使用tamoxifen發生子宮相關病變及婦科監測之研究

Abstract

研究背景：Tamoxifen 為 ER/PR(+) 乳癌病人的抗荷爾蒙用藥。有研究顯示使用 tamoxifen 從 5 年延長至 10 年，可以更有效地降低乳癌的復發率及死亡率。治療指引因此建議將 tamoxifen 的服藥期延長至 10 年，但罹患子宮體癌的機率也相對提高。雖曾有研究以台灣健保資料庫分析，發現年齡及 tamoxifen 之服藥期與罹患子宮體癌風險有關，但仍需更多研究探討延長 tamoxifen 之服藥期（超過5年）是否有較高機率罹患子宮體癌。另外有些研究以子宮內膜厚度作為子宮體癌的替代終點，仍存有爭議，因此需更多研究評估使用 tamoxifen 乳癌病人，所接受之例行性婦科監測。 研究目的： 1.分析國內病人使用tamoxifen後，子宮體癌的發生率，並分析相關風險因子，如：年齡、延長tamoxifen服藥期等。 2.分析國內病人使用tamoxifen後，子宮體癌的相關診療資料，如：組織類型等。 3.分析國內病人使用tamoxifen後，接受子宮內膜搔刮術的發生率及其病理結果。 研究方法：以臺大醫院的相關資料，進行回溯性世代研究。在臺大醫院癌症登記檔中，確診為原發性乳癌的女性病人，經納入以及排除標準篩選後，共 9,668 人納入，依照所接受之乳癌輔助性荷爾蒙治療，分為只使用過 tamoxifen（以 Tamoxifen-only 簡稱，共 2612 人，27.0%），皆曾使用過 tamoxifen 和 aromatase inhibitor（以 Both 簡稱，共 1754 人，18.1%），只使用過 aromatase inhibitor（以 AI-only 簡稱，共 1536 人，15.9%），未使用過乳癌輔助性荷爾蒙治療（以 Non-user 簡稱，共 3766 人，39.0%），並分析子宮體癌的發生率及相關風險因子。 研究結果：在 9,668 位乳癌病人之中，後續被診斷為子宮體癌共有40人，其中 Tamoxifen-only 組的發生率最高（19/2612=0.7%），相較於 AI-only 組（4/1536=0.3%）或 Non-user 組（11/3766=0.3%）。就子宮體癌組織類型而言，Tamoxifen-only組最多為癌瘤（carcinoma）（84.2%），包含子宮內膜腺癌（endometrioid adenocarcinoma）（78.9%），亮細胞腺癌（clear cell adenocarcinoma）（5.3%）；另一類則為肉瘤（sarcoma）（15.8%）包含惡性肉瘤（carcinosarcoma）（10.5%）和基質惡性肉瘤（endometrial stromal sarcoma）（5.3%），其中並沒有一般子宮體癌之肉瘤族群（sarcoma）較常發生的平滑肌肉瘤（leiomyosarcoma）。由 KM 存活曲線顯示，Tamoxifen-only 組相較於其他三組，有顯著較低的無子宮體癌存活率（p=0.038）。 由單變項比例風險迴歸分析結果得知，在整體族群中，BMI≥25 kg/m2 、乳小葉原位癌（LCIS）和乳管原位癌（DCIS），皆顯著增加子宮體癌發生風險。由多變項比例風險迴歸分析結果得知，Tamoxifen-only組，相較於Non-user組，有較高的子宮體癌發生風險，且達統計顯著。另外，乳癌診斷年齡50 歲以上及BMI≥25 kg/m2 皆顯著增加子宮體癌發生風險。另將 tamoxifen 開始使用至追蹤結束之間隔時間，以每年做次族群之多變項比例風險迴歸分析，並同時放入年齡≥50歲與 BMI≥25 kg/m2作校正。迴歸分析的結果顯示tamoxifen 在開始使用後4年之內，每一年分層之Tamoxifen only組，相較於Non-user 組皆有顯著增加子宮體癌的風險。 於乳癌診斷後接受子宮內膜搔刮術共 507 次，其中最多為 Tamoxifen-only 組（354/507=69.8%），相較於 Both 組（16.4%），Non-user 組（11.0%），以及AI-only 組（2.8%）。病理結果顯示為子宮體癌或內膜增生，最多為 Tamoxifen-only 組（21/354=5.9%），大部分在接受子宮搔刮術前，沒有異常子宮出血（237/354=66.9%），但仍有一定比例之病理結果顯示為子宮體癌或內膜增生(9/354=2.6%)。 結論：由多變項比例風險迴歸分析結果得知，Tamoxifen-only 組、乳癌診斷年齡 50 歲以上以及 BMI≥25 kg/m2皆顯著增加子宮體癌發生風險。就子宮體癌組織類型而言，Tamoxifen-only 組雖最多為癌瘤（carcinoma），但另一類肉瘤（sarcoma）較一般子宮體癌族群發生比例高（15.8% 相較於 2~5%）。此外，於乳癌診斷後，接受子宮內膜搔刮術者，Tamoxifen-only 組大部分在接受子宮搔刮術前，沒有異常子宮出血，但仍有一定比例的病理結果顯示為子宮體癌或內膜增生。因此病人在開始使用 tamoxifen，尤其具上述風險因子時，應每年定期接受婦科檢查。

Background: Tamoxifen is an anti-hormonal drug which is commonly used for hormone receptor-positive breast cancer patients. Two randomized controlled trials showed that the extended use of tamoxifen for 10 years, compared to 5 years, significantly reduced the recurrence and mortality of breast cancer. However, the risk of endometrial cancer increased. Few publications examine whether the extended use of tamoxifen (>5 years) have a higher risk of tamoxifen-related uterine events. Since the increase in endometrium thickness is still controversial as a surrogate measure of uterine cancer, studies are needed to evaluate feasible routine gynecologic surveillance in breast cancer patients with tamoxifen regimen. Study objectives: 1.To analyze the rates of uterine cancer in patients treated with tamoxifen and to identify the risk factors of tamoxifen-related uterine cancer such as age, extended use of tamoxifen, etc. 2.To evaluate tamoxifen-related uterine cancer characteristics, such as histology, etc. 3.To evaluate the event count and pathological diagnosis in patients proceeding dilatation and curettage (D and C) after tamoxifen use. Methods: This is a retrospective cohort study enrolling female patients diagnosed with breast cancer from the National Taiwan University Hospital cancer register. A total of 9,668 patients who fulfilled inclusion and exclusion criteria were grouped by the treatment with tamoxifen-only (N=2612, 27%), both tamoxifen and aromatase inhibitor (N=1754, 18.1%), aromatase inhibitor-only (N=1536, 18.1%), and non-user (N=3766, 39%) for analyses of incidence rates and risk factors of uterine cancer. Results: In the 9668 breast cancer patients, 40 patients were subsequently diagnosed with uterine cancer. The tamoxifen-only group had higher incidence rate (19/2612, 0.7%) than AI-only or non-user group (4/1536, 0.3% or 11/3766, 0.3%). Uterine cancer in tamoxifen-only group was mostly carcinoma (84.2%), including endometrioid adenocarcinoma (78.9%) and clear cell adenocarcinoma (5.3%), subtyped by histologic classification. Sarcoma was found 15.8%, including carcinosarcoma (10.5%) and endometrial stromal sarcoma (5.3%). None was leiomyosarcoma which had been mostly found in all cause uterine sarcoma. The Kaplan-Meier survival curve showed that tamoxifen-only group had the lowest uterine cancer-free probability among the four groups (p=0.038). BMI≥25 kg/m2, lobular carcinoma in situ (LCIS), ductal carcinoma in situ (DCIS) breast cancer were significant risk factors of uterine cancer in the total population using a univariate cox regression model. Furthermore, the multivariate cox regression model indicated significantly higher risk of uterine cancer in the tamoxifen-only group than the non-user group. Age above 50 years and BMI≥25 kg/m2 also increased the risk of uterine cancer significantly. When patients were stratified with time from tamoxifen treatment to the end of follow-up in one-year intervals, the multivariate cox regression model, adjusted by age above 50 years and BMI above 25 kg/m2, indicated that the uterine cancer risk was higher in the tamoxifen-only group than the non-user group within 4 years of tamoxifen use. The dilatation and curettage (D and C) events were counted 507 in all patients after breast cancer diagnosis. The tamoxifen-only group contributed to the total event mostly (354/507, 69.8%), compared to the both group (16.4%), non-user group (11.0%) and AI-only group (2.8%). The tamoxifen-only group had the highest event count of either uterine cancer or endometrial hyperplasia (21/354, 5.9%) among the four groups. Two third of the tamoxifen-only patients had no abnormal uterine bleeding before D and C (237/354, 66.9%) with positive rate of uterine cancer or endometrial hyperplasia as much as 2.6% (9/354). Conclusion: The tamoxifen-only group vs non-user group, age above 50 years and BMI≥25 kg/m2 increased the risk for uterine cancer significantly. Most uterine cancer in tamoxifen-only group was histologically classified as carcinoma while the sarcoma rate was higher than all cause uterine cancer population (15.8% vs 2-3%). Positive rate of uterine cancer or endometrial hyperplasia was as much as 2.6% in the tamoxifen-only patients without abnormal uterine bleeding before D and C. It indicates that routine gynecological check is an important follow-up in breast cancer patients with tamoxifen treatment.