Long-Term Outcomes of Radiotherapy for Primary Pediatric Brain Tumors: Statistics and Strategies of a Single Institute in Taiwan

原發兒童腦瘤接受放射治療之長期追蹤結果：台灣單一醫療機構之統計及治療策略探討

10.6316/TRO/201017(2)85

楊凱琳(Kai-Lin Yang)；黃品逸(Pin-I Huang)；黃棣棟(Tai-Tong Wong)；張開屏(Kai-Ping Chang)；王令瑋(Ling-Wei Wang)；蕭正英(Cheng-Ying Shiau)；劉裕明(Yu-Ming Liu)；藍耿立(Keng-Li Lan)；胡育文(Yu-Wen Hu)；賴宜君(I-Chun Lai)；陳一瑋(Yi-Wei Chen)；顏上惠(Sang-Hue Yen)

原發兒童腦瘤 ； 放射治療 ； 存活率 ； 電子登錄系統 ； Primary pediatric brain tumors ； Radiotherapy ； Survival ； Electronic registry system

放射治療與腫瘤學

17卷2期（2010 / 06 / 01）

85 - 99

英文

目的：在台灣，兒童腦瘤的長期追蹤報告仍為少見，我們發展了一套電子登錄系統(ERS)，並以之建立了一個兒童腦瘤資料庫，其中包含了自西元1975年1月起至2004年5月止的986位小於18歲的兒童腦瘤病歷，本研究的目的是分析其中曾接受放射治療的371個病人存活分析狀況。 材料與方法：在本院原始癌症登記資料庫中，共記錄529個原發兒童腦瘤病例曾於1975年1月至2004年5月接受過放射治療，我們透過調閱原始放射治療之紙本資料並記錄於新建之電子登錄系統(ERS)中，總共有371位具完整資料(70.1%)。依照不同的兒童腦瘤分類，統計其男女比、放射治療後的死亡病例數、接受放射治療時的年齡、放射治療劑量、和各疾病接受放射治療之比率。並分析數種疾病之總存活率(overall survival)及部分疾病之無惡化存活率(progression-free survival)或放射治療後無失敗存活率(post-radiotherapy failure-free survival)。結果：在371個曾接受放射治療的病人中，髓母細胞瘤(medulloblastoma)所佔人數最高(n=93)，其次依序是生殖性細胞胚芽瘤(germinoma; n=59)，星狀細胞瘤(astrocytoma, WHO grade Ⅱ; n=36)，再生不良性星狀細胞瘤(anaplastic astrocytoma, WHO grade Ⅲ; n=36)和腦幹腫瘤(brain stem tumor, n=29)。髓母細胞瘤、非典型畸胎／類橫紋肌細胞瘤(atypical teratoid/rhabdoid tumor)、星狀細胞瘤、再生不良性星狀細胞瘤、膠質母細胞瘤(glioblastoma)和生殖性細胞胚芽瘤等疾病接受放射治療的比率分別為70.9%、90.4%、48.65%、80.0%、41.0%和78.6%。所有疾病接受放射治療之後的總死亡率為43.93%。髓母細胞瘤、星狀細胞瘤、再生不良性星狀細胞瘤、和生殖性細胞胚芽瘤等疾病接受放射治療後的5年總存活率(overall survival)分別為81.73%、71.49%、25.34%和88.94%。髓母細胞瘤的5年無惡化存活率(progression-free survival)是66.46%。非典型畸胎／類橫紋肌細胞瘤的中間總存活率(median overall survival)和中間放射治療後無失敗存活率(median post-radiotherapy failure-free survival)分別為17個月和11個月。另外，有7位患者(1.89%)於放射治療結束後一段時間（2.8至24年；中位數19.7年），罹患可能與先前放射治療有關之顱內第二種腫瘤。 結論：電子登錄系統(ERS)可有效便利協助過往罕見兒童腦瘤疾病之資料保存及相關統計的分析進行，本研究報告之比率統計和存活分析皆與過往文獻中刊載的數據相當，此兒童腦瘤資料庫值得繼續運作，期望將提供相當多寶貴的資料給未來的研究之進行。

Purpose In Taiwan, data regarding the long-term outcomes of radiotherapy for pediatric brain tumors are still rare. We developed an electronic registry system (ERS) that can help to establish a database, which consisted of 986 cases of primary brain tumors in children younger than 18 years of age in Taipei Veterans General Hospital from January 1975 to May 2004. In this study, we focused on the long-term outcomes of 371 available patients treated by radiotherapy. Materials and Methods: In our original cancer registry data, there were 529 pediatric brain tumor patients who underwent radiotherapy between January 1975 and May 2004. However, after reviewing the cases, only 371 cases (70.1%) were fully recorded without missing data. Profiles of radiotherapy for these patients were input into our newly established ERS. The demographic data were analyzed, including classification of tumors, gender distribution, mortality case numbers after radiotherapy, age at treatment, dose profile of radiotherapy, and proportion of patients receiving radiotherapy. For some diseases of interest, overall survival, progression-free survival and postradiotherapy failure-free survival were also analyzed by the Kaplan-Meier method. Results: Of the 371 patients receiving radiotherapy, treatment for medulloblastomas is the most common (n=93), followed by germinomas (n=59), astrocytomas (n=36), anaplastic astrocytomas (n=36) and brain stem tumors (n=29). The proportions of patients receiving radiotherapy for medulloblastomas, atypical teratoid/rhabdoid tumors, astrocytomas (grade Ⅱ), anaplastic astrocytomas, glioblastomas and germinomas were 70.9%, 90.4%, 48.65%, 80.0%, 41.0% and 78.6%, respectively. The pooled mortality rate of all diseases after radiotherapy was 43.93%. The five-year overall survival rates of medulloblastomas, astrocytomas (grade Ⅱ), anaplastic astrocytomas and germinomas were 81.73%, 71.49%, 25.34% and 88.94%, respectively. The 5-year progression-free survival of medulloblastomas was 66.46%. Additionally, seven patients (1.89%) suffered from radiation-induced secondary tumors after completion of radiotherapy for a period of time (range, 2.8 to 24 years; median, 19.7 years). Conclusion: An electronic registry system can help to facilitate the statistical analysis for rare pediatric brain tumor data. Our demographic and survival analyses were mostly comparable with those of previous studies. The database of pediatric brain tumors is worthy continuing to provide materials for further studies.