覆盖 3 亿人 ) 目前并非所有的登记都有足够的可供上报的高质量数据 每个地方登记处提交的数据都会受到 NCCR 和国际癌症研究机构 / 国际癌症注册协会 (IARC/IACR) 的检查 质量评估包括但不限于, 形态学鉴定所占比例 (MV%), 有死亡证明的癌症病例所占比例 (DCO%), 死亡率

Transcription

1 银河医药李平祝团队 2015 中国癌症统计数据, 全文翻译 王明瑞 / 张金洋 写在前面 前两天朋友圈突然一篇影响因子 的文献刷屏了, 紧接着出现了各种中文解读文章, 各有亮点, 但我们觉得能够把原文全部翻译出来, 原滋原味的去读更有一番滋味, 因此我们银河医药团队小伙伴加班加点完成了 18 页英文文献的全文翻译, 供大家研究参考使用, 时间较短, 难免有所疏漏, 欢迎大家指正 原文 : Cancer statistics in China,2015, 发表于 CA-Cancer J Clin, 影响因子 翻译 : 银河医药团队王明瑞主译 ( 实习生 ), 张金洋 中国癌症统计数据 随着发病率和死亡率越来越高, 癌症正成为中国首要的死亡原因和一个重要的公共卫生问题 因为中国人口多 (13.7 亿人 ), 先前的国家发病率和死亡率评估都限制在上世纪 90 年代的小样本或基于特定的年份 现在通过国家中央癌症登记处的高质量数据, 作者分析了 72 个地区基于人群的癌症登记 ( ), 代表了 6.5% 的人口, 用以估计 2015 年新病例和癌症死亡的人数 趋势分析 ( ) 使用了 22 个登记处的数据 结果表明,2015 年预计有 个新癌症病例和 个癌症死亡, 肺癌的发病率和死亡率都是最高的 胃癌 食道癌和肝癌的发病率和死亡率也很高 将所有癌症的发病率和死亡率结合起来看, 农村居民的年龄标准化数据要高于城镇居民 ( 发病率 人 /10 万人 vs 人 /10 万人 ; 死亡率 149 人 /10 万人 vs 人 /10 万人 ) 将所有的癌症结合起来, 在 2000 年到 2011 年之间, 男性的发病率保持稳定 ( 每年 +0.2%;P=.1), 女性发病率明显上升 ( 每年 +2.2%;P<.05) 与此相对, 死亡率自从 2006 年之后一直在下降, 无论男性 ( 每年 -1.4%;P<.05) 还是女性 ( 每年 -1.1%;P<.05) 很多评估范围内的癌症可以通过减少癌症风险因素 提高临床护理服务的效率, 尤其是农村人口和弱势群体, 来减少发病率和死亡率 1 介绍 中国的癌症发病率和死亡率一直在上升, 从 2010 年开始已经成为主要的致死原因, 成为了中国的一个主要公共卫生问题 这个逐渐增加的压力有相当大的一部分可以归于人口的增长和老龄化以及社会人口统计的变化 尽管之前有全国发病率的评估, 但那些评估或者只能代表很小的人口 ( 小于 2%), 或者只有特定的年份 这对评估的不确定性和代表性有影响, 并且会潜在的影响癌症控制政策的制定 因为之前的中国癌症预防与控制项目 ( ) 是 10 年之前发布的, 所以一个关于中国全国与各地方癌症规模与概况的更加复杂的描绘将会提供更加清晰的优先顺序, 为制定基于癌症谱的政策和规划制定提供参考, 减轻国家的癌症负担 这篇研究评估了全国范围内的癌症发病率 死亡率和存活率 ; 几个主要癌症分区域的发病率和死亡率 ; 几个主 要癌症的时间趋势以及这个信息对中国癌症防控提供的指导 2 数据源和方法 2.1 中国癌症登记 全国肿瘤登记中心 (NCCR) 建立于 2002 年, 负责收集 评估 发布中国的癌症数据 癌症诊断会被上报给地方癌症登记, 这些数据有多个来源, 包括地方医院和社区卫生中心, 以及城镇居民基本医疗保险和新农合 自 2002 年起, 标准登记条例的实施已经大大提升了癌症数据的质量 2008 年, 卫生部通过中央财政体系实施了国家癌症登记项目 自此之后, 各地基于人群的登记数量从 2008 年的 54( 人口覆盖 1.1 亿人 ) 上升到了 2014 年的 308( 人口

2 覆盖 3 亿人 ) 目前并非所有的登记都有足够的可供上报的高质量数据 每个地方登记处提交的数据都会受到 NCCR 和国际癌症研究机构 / 国际癌症注册协会 (IARC/IACR) 的检查 质量评估包括但不限于, 形态学鉴定所占比例 (MV%), 有死亡证明的癌症病例所占比例 (DCO%), 死亡率 / 发病率比率 (M/I), 未认证癌症的比例 (UB%), 不明确或位置原发癌部位的百分比 (CPU%) 只有满足以上这些要求的数据才会在分析中使用 登记数据的具体质量分类可以在之前的出版物中找到 ( 表 1) 数据分类为 A 或 B 在这篇研究中被认为是可以接受的 提交的登记数据可供使用的比例随着年份有变动, 从 2009 年的 69.2%(104 个中 72 个可用 ), 到 2010 年的 66.2%(219 个中 145 个可用 ), 和 2011 年的 75.6%(234 个中 177 个可用 ) 我们使用了在三年中始终可用的 72 个登记处的数据 表 1: 基于人群的癌症登记点 省份 登记处 IARC 接受数据 用于时间趋势分析 用于存活率分析 北京 北京 迁西县 河北 涉县磁县 保定 山西 阳泉阳城 内蒙古 赤峰 沈阳大连 庄河 辽宁 鞍山 本溪丹东东港 黑龙江 哈尔滨 尚志 上海 上海 金坛 苏州海安 启东 海门连云港 江苏 东海灌云 淮安 盱眙 金湖 射阳 建湖 大丰

3 赣榆 ( 除赣榆外, 所有数据都用来估计 2015 年的发病率和死亡率 ) 吉林浙江安徽福建江西山东河南湖北湖南广东广西重庆四川甘肃青海新疆 扬中 泰兴 德惠 延吉 杭州 嘉兴 嘉善 海宁 上虞 仙居 肥西 马鞍山 铜陵 长乐 厦门 赣州 临朐 汶上 肥城 偃师 林州 西平 武汉 云梦 衡东 广州 四会 柳州 扶绥 重庆 成都 自贡 盐亭 景泰 武威 西宁 新源

4 2.2 癌症发病率数据 为了估计 2015 年中国新增癌症数量, 我们使用 72 个基于人群的癌症登记点的最近的数据 ( 登记的癌症病例 )( 表 1), 人口覆盖 8850 万人, 大概为全国人口的 6.5% 表 1 的更少数量的登记点 (22 个 ) 提供了 4440 万人口的覆盖率, 有着 这 12 年的高质量数据, 用于发病率的时间趋势分析 这两套数据的癌症登记点可以在表 1 中查找到 图 1:A 为 22 个登记点 ( 数据 ); B 为 72 个登记点 ( 数据 ) 我们这篇研究中不仅包括了侵袭性肿瘤, 也包括了参照国际规则定义的多原发癌 发病率数据来自 NCCR 数据库 我们使用国际疾病分类第 10 版的标准, 因为只有这版的死亡率数据是可用的 提取的变量有性别 年龄 出生日期 诊断年份 癌症位点 形态学 居住地 ( 乡村和城镇 ) 地区( 中国北部 东北 东部 中部 南部 西南 西北 ) 对于分年龄的发病率分析, 我们使用了 5 个大的年龄组 (<30,30-44,45-59,60-74, 75) 2.3 癌症死亡率数据 我们采用 72 个登记点编制的数据来估计 2015 年中国的癌症死亡人数 为了估计 之间癌症致死的趋势, 我们使用了 22 个登记点的数据来分析 这些登记点的癌症死亡数据来自地方医院 社区卫生中心 人口统计 ( 包括来自国家疾病监测系统 DSP 的数据 ) 和民政局 DSP 系统由卫生部于上世纪 80 年代建立, 例行收集由医院提供的死亡认证信息, 或者在死亡认证不可用的情况下采用按家走访的方式收集 尽管 DSP 使用具有全国代表性的位点样本, 但这仍然只覆盖了很小的一部分人口 ( 不到 1%) 2.4 癌症存活率 由于 72 个登记点没有精确的后续信息, 这篇研究中的 5 年相对存活率使用 M/I 比率来进行估计, 这是一个以前就被使用过的方法 我们只估计整体的癌症存活率, 因为使用这种方法有可能高估或者低估某一种癌症类型的存活率 我们计算 M/I 比率的前提是, 假设 2009 到 2011 到 2015 的发病率和致死率的关系没有改变, 所以我们通过年龄标准化的发病率 ( ) 来划分年龄标准化的死亡率 ( )

5 2.5 人口数据 以 5 岁年龄组和性别分组的人口数据来源于统计数据或公安人口普查 个人登记提供的数据通过各地区提供给 NCCR 这些数据来源于各地局部统计或公安局或基于人口普查数据的计算 2.6 统计学分析 我们估计了 2015 年中国的所有癌症的新增病例人数, 并根据 72 个癌症登记点的特定年龄组发病率数据 ( ) 分性别估计了 26 个特定的癌症类型的新增病例数 我们用相同的方法估计了 2015 的癌症死亡数量 对于 10 种最普遍的癌症, 这些新增患病和死亡的数字还按照城镇 / 乡村登记点以及覆盖中国的 7 个行政区域进行了细分 所有癌症的汇总数据和 6 个最普遍癌症的数据也按照 5 个大的年龄组进行了分层 (<30,30-44,45-59,60-74, 75) 从 2000 年到 2011 年发病率和死亡率的时间趋势分析是通过将连接点模型与对数转化的 年龄标准化的比率来进行计算的 为了减少这段时间内报道错误改变的可能性, 我们将所有的模型限制在最大 2 个连接点之内 趋势表达为年度变化百分比 (APC), 我们用 Z 测试来评估 APC 是否在统计学上偏离了 0 在描述趋势时, 术语 增加 或 减少 用于趋势的斜率 (APC) 统计学显著时 (P<.05) 对于非统计学分析, 我们使用术语 稳定 对于所有的分析, 我们都分性别陈述所有癌症与 10 种最常见癌症的结果 3 结果 3.1 数据质量 3 个主要的指标 (MV%, DCO%, M/I 比率 ) 对于基于人群的癌症登记 按照癌症类型的分类, 表明两种癌症登记 数据的质量都很高 ( 图 2) 由于 22 个登记点的半数都通过了 IARC 的认证, 它们的数据质量被认为高于 72 个癌症 登记点数据, 有着更高的 MV% 和更低的 DCO% 这些数据质量参数加上 UB% 和 CPU% 都呈现于表 1( 见在线信息 )

6 图 2: 两套癌症登记数据质量的三个主要测量指标 最左边的一列数字是 ICD10 的分类标准 DCO% 只表征具有 死亡认证的癌症病例百分比 M/I 为死亡率 / 发病率 ;MV% 为形态学验证的比率 预期癌症发病率 预计 2015 年将会有大约 个新增侵袭性癌症病例, 与每天 个新病例相符合 男性中最普遍的 5 种癌症依次为 : 肺和支气管癌症, 胃癌, 食道癌, 肝癌, 结直肠癌, 这些占到所有癌症病例的三分之二 女性中最普遍的 5 种癌症依次为 : 乳腺癌, 肺和支气管癌, 胃癌, 结直肠癌, 食道癌, 这些占到了所有癌症病例的 60% 单是乳腺癌就占到了所有女性癌症的 15%( 表 2) 位置 唇, 口腔, 咽部 ( 鼻咽部除外 ) 表 2: 预期 2015 年中国新增癌症患病和死亡人数 ( 千人 ) 发病率致死率 ICD-10 总男性女性总男性女性 C00- C10,C12- C 鼻咽 C 食管 C 胃 C 结直肠 C18-C 肝 C 胆囊 C23-C 胰腺 C 喉 C 肺 C33-C 其他胸部器官 C 骨 C 皮肤黑色素瘤 C 乳房 C 宫颈 C 子宫 C54-C 卵巢 C 前列腺 C 睾丸 C 肾脏 C64-C66, C 膀胱 C 脑, 中枢神经系统 C 甲状腺 C 淋巴瘤 C81-C85, C88, C90,C96 白血病 C91-C 所有其他位置和未 A_O 明确的 所有位置 ALL

7 10 种最普遍癌症的新增病例数和发病率按照城乡和居住地分组呈现于表 3 对于所有的癌症, 年龄标准化发病 率男性要高于女性 (234.9/168.7 每 人 ), 农村高于城镇 (213.6/191.5 每 人 ) 西南部有最高的癌症 发病率, 其次为北部和东北 ; 中部的发病率最低 表 3: 所有癌症的年龄标准化发病率和部分癌症分地区预计新病例 预期癌症死亡率 据估计 2015 年将会有 名中国人死于癌症, 与每天 7500 例癌症死亡相符合 男性和女性死亡率最高的癌症均为 : 肺和支气管癌 胃癌 肝癌 食道癌 结直肠癌, 占到了所有癌症死亡的四分之三 ( 表 2) 与发病率类似, 年龄标准化死亡率男性高于女性 (165.9/88.8 每 人 ), 农村高于城市 (149.0/109.5 每 人 )( 表 4) 最高的死亡率仍是西南 北部和东北, 中部最低 表 4:2015 年所有癌症的年龄标准化死亡率和部分癌症分地区死亡率

8 年分年龄段分性别的发病率和死亡率 在 60 岁之前, 肝癌是被诊断出的最普遍的癌症, 并且在男性的癌症死亡中占比最高, 其次是肺癌和胃癌, 这 也是 的发病和死亡的主要类型 ( 表 5) 对于 75 岁以上的男性, 肺癌是最广泛被诊断出的癌症, 也是最主要 的死亡原因 男性的大多数癌症新增病例和死亡都位于 岁之间 在女性中,30 岁以前被诊断出的最普遍的癌症是甲状腺癌,30-59 岁之间是乳腺癌,60 岁以后是肺癌 ( 表 5) 45 岁以下, 乳腺癌是导致死亡的最主要原因, 其次是肺癌 岁的新增和死亡病例是最多的 表 5:2015 年中国按年龄分组预期发病与死亡人数 ( 千人 ) 年预期癌症存活率 根据预测, 所有的癌症结合来看,2015 年大概 36.9% 的癌症患者能够存活 5 年以上, 女性的存活率比男性要好 (47.3%/29.3%)( 表 6) 根据诊断时的居住地得出的 5 年存活率估计有一定的潜在变化 : 农村病人的存活率比城市 更低 (30.3%/42.8%) 与前面相似, 存活率最低的是西南地区 (24.9%), 最高的是中部地区 (41.0%) 表 6:2015 年所有癌症 5 年存活率 ( 以性别和区域划分 )

9 3.6 癌症发病与死亡的趋势 对于所有的癌症, 在研究的时间段内 ( ), 男性的发病率较为稳定, 女性有显著上升 (P<.05)( 图 3, 表 7) 与此相反, 两性的死亡率都有显著下降 ( 图 3, 表 8) 尽管这个趋势令人高兴, 但实际上在此期间癌症的死 亡人数增加了 ( 增加了 73.8%, 从 2000 年的 到 2011 年的 88800), 这主要是由于人口增加和老龄化 ( 图 4) 对于男性, 在 10 种最普遍的癌症中, 以时间趋势分析, 从 2000 年到 2011 年发病率增加的有 6 种,( 胰腺癌, 结直肠癌, 脑和中枢神经系统癌症, 前列腺癌, 膀胱癌, 白血病 ), 而胃癌 食道癌 肝癌则有下降 (P<.05) 肺 癌的趋势则比较稳定 ( 图 5, 表 7) 对于女性,10 种最普遍的癌症中有 6 种年龄标准化发病率显著上升 ( 结直肠癌, 肺癌, 乳腺癌, 宫颈癌, 子宫 体癌, 甲状腺癌,P<.05) 与男性相同, 胃癌 食道癌 肝癌可见下降趋势 ( 图 6, 表 7) 对于男性,10 种最普遍的癌症中,4 种的年龄标准化死亡率可见上升 ( 结直肠癌, 胰腺癌, 前列腺癌, 白血病, P<.05), 其他趋势较为稳定 ( 肺癌 膀胱癌和脑癌 )( 图 7, 表 8) 对于女性, 最普遍的 10 种癌症中 3 种的死亡率上升 ( 乳腺癌, 宫颈癌和卵巢癌 ), 结直肠癌 肺癌 子宫体癌 和甲状腺癌趋势较为稳定 ( 图 8, 表 8) 与癌症发病率相似, 胃癌 食道癌 肝癌的死亡率在两性中都有下降 ( 图 7,8) 肺癌的趋势男女都较为稳定, 这是两性最主要的癌症死亡原因 图 3: 所有癌症的发病率与死亡率变动趋势

10 表 7: 分性别的部分癌症和全部癌症的发病率趋势, 表 8: 分性别的部分癌症的所有癌症死亡率趋势,

11 图 4: 中国分性别新增病例数与死亡数趋势 图 5: 男性部分癌症发病率趋势, 图 6: 女性部分癌症的发病率趋势,

12 图 7: 男性部分癌症的死亡率趋势, 图 8: 女性部分癌症的死亡率趋势, 讨论 尽管之前已经有过对于全国癌症估计的报道, 但这些都只限于特定的年份或癌症种类, 很难进行不同癌症间的 横向比较 本研究提供了更加全面的全国范围内的癌症统计, 使用了最新 最具权威性的数据, 包含了信息与时间 趋势 癌症防控需要依赖于基于人群的发病率和死亡率数据, 以此来执行政策和评估政策的有效性 因此, 最新的全 国范围内对于癌症负担和时间趋势的分析, 对理解癌症的病因学, 和有效的预防 早期诊断和管理措施有着至关重 要的作用 这些结果也会成为未来评估中国癌症控制有效性的基线, 也将有助于区分地区间需求的优先级 中国人口占到了世界人口的五分之一, 因此这些数据将会对世界癌症负担有着重要的作用 : 大约 22% 的新增癌

13 症病例和 27% 的癌症死亡发生在中国 更重要的是, 中国的癌症谱与发达国家明显不同 中国最普遍的 4 个癌症是肺癌 胃癌 肝癌 食道癌 这几种癌症占到了中国癌症诊断的 57%, 而在美国只有 18% 同样, 中国的这几种癌症占到了全世界发病负担的 1/3 到 1/2 与此相对, 美国最普遍的癌症是肺癌 乳腺癌 前列腺癌和结直肠癌 中国最普遍的癌症生存率很低 ; 而美国的几种除了肺癌之外, 预后都非常良好, 对于前列腺癌和乳腺癌, 有相当的比例是在早期诊断扫描的时候发现的, 因而抬高了发病率 癌症发病类型的差异对于死亡率的差异有着重要的影响 我们对 2015 年的预计是基于 72 个登记点 年的数据 这些登记点只覆盖了中国人口的 6.5%, 但它们是目前可用的最好的全国范围的数据, 代表了 8550 万人口 除此之外, 与之前的研究 ( 只覆盖了人口的 2%) 相比, 本研究采用数据的人口覆盖面明显扩大, 包含了更多的西部地区, 因而对于中国整体情况就更具有代表性 另外, 大陆的全部 12 个基于人群的癌症登记点都有着高质量的数据, 完全满足 CI5 的标准 全国发病率的估计可以广泛的与之前发表的进行比较 两个中国之前发表的年度报告估计 10 年和 11 年的新增癌症病例分别为 309 万人和 337 万人 更早的一个估计是 05 年的 296 万人, 尽管使用的方法不同 我们对于 2015 年中国发病率 (429 万人 ) 的估计要明显高于 GLOBOCAN 在 2012 年作出的 340 万人的估计 这些差距主要来源于数据时间线的不同 (09-11/03-07) 代表性的不同 地理覆盖程度的不同 (72 个登记点覆盖 6.5% 的人口 /23 个登记点覆盖 3% 的人口 ) 尤其是有着更高发病率的农村居民(213.6 每 10 万人 /191.5 每 10 万人 ) 占到了 32.7%,GLOBOCAN2012 年的估计只占到了 21.5% 用来获取全国发病率的数据也是不同的, 因为 GLOBOCAN2012 通过把 23 个登记点的特定年龄 特定性别 特定位置癌症死亡率模型化转化为发病率 需要承认, 这些数据并不都满足 IARC 的质量标准, 已发表报告中的这些差异低估了中国对于提升癌症登记点覆盖率和质量的需求 与发病率相比, 我们的死亡率与 GLOBOCAN 有着更高的相似性 我们估计 2015 年中国癌症死亡为 281 万人, GLOBOCAN2012 为 246 万人 这反映了两个研究使用的死亡率数据源是相似的 :GLOBOCAN 用了 DSP(04-10), 我们使用了 72 个癌症登记点的数据 (09-11),DSP 数据就是这些数据的一部分 DSP 数据基于县, 按照地理区域划分, 被特殊设计为具有全国代表性 对于港澳的发病率和死亡率,GLOBOCAN2012 计入而我们没有计入 我们发现整体癌症的发病率地区间有显著的差异 ( 农村 / 城市, 地区间 ) 农村居民比城市居民有更高的发病率, 并且发病率在中国的 7 个行政区域都是不同的 这些地理差异可能由多种因素造成, 但农村更高的吸烟率明显是重要原因 12 种由吸烟引起的癌症占到了中国所有癌症的 75% 与我们的假设一致, 癌症发病率最高的西南地区在 2002 年的报道中有着最高的吸烟普遍性 我们发现癌症的死亡率和存活率的地理差异更大 这些差异或许可以部分解释为, 更为有限的医疗资源 更低 水平的癌症护理 农村和欠发达地区被诊断出时就已经是晚期的概率更大 这就为政府向农村和欠发达地区投入更 多资源和服务提供了理由 之间中国的癌症诊断数量有着显著的上升 很大一部分原因是中国的人口增加和老龄化 其他因素可能也有贡献, 比如不健康生活方式的流行 疾病意识的提升 诊断服务和数据完整性的提升等等 发病率增加幅度最大的是前列腺癌 宫颈癌和女性的甲状腺癌 前列腺癌增加的因素目前还不清楚 ; 这可能与逐渐应用前列腺特异性抗原扫描和活检水平的提升有关, 也有可能是由于逐渐西化的生活方式 西方的生活方式增加了中国的肥胖率, 减少了活动量, 可能会对结直肠癌和乳腺癌发病率有影响 乳腺癌发病率的上升也可能与计划生育政策有关 与发达国家发病率减少的趋势相反, 中国的宫颈癌发病率在增加 这可能也反映出了帕帕尼科拉乌试验的不足, 中国据报道只有 1/5 的女性做过帕 实验来进行宫颈癌的检查 HPV 感染率的增加, 和大陆 HPV 疫苗的不足, 表 明在可预见的未来, 中国的宫颈癌趋势将继续和国际保持差距 女性甲状腺癌的增加与其他国家相类似 ; 尽管这可能与各种新技术的使用导致的过度诊断有关, 但由于缺少疾 病阶段的信息, 也不能排除发病率确实在增加的可能性 胃癌 食道癌 肝癌的发病率和死亡率都有明显减少 尽管发病率减少了, 但人口基数的增加和老龄化仍然使 得新增病例的数量非常大 对感染的控制可能也对这个趋势有影响, 比如对导致肝癌的 HBV HCV 的控制, 对导致

14 胃癌的幽门螺杆菌的控制 通过对婴儿注射疫苗来进行 HBV 早期预防已经取得了明显的成效 :0-19 岁的肝癌死亡率 15 年内下降了 95% 尽管 HBV 疫苗的成功对于预防儿童肝癌取得了明显的成效, 但也许不能解释对于全年龄段的影响 还有其他很多的重要因素, 比如受到黄曲霉毒素污染的玉米的减少和饮用水的净化 计划生育政策减少了家庭内部的儿童间 HBV 的传染, 更加规范的注射操作减少了医院内 HBV 和 HCV 的传染, 这些因素可能也会对整体肝癌发病率有影响 4.1 对中国癌症预防的提示 根据估计, 大约 60% 的癌症可以通过减少风险因素的暴露来避免 中国可避免癌症死亡的最大贡献因素是慢性 感染, 这大约占到癌症死亡的 29%, 主要是胃癌 肝癌和宫颈癌 吸烟导致了中国 23% 到 25% 的癌症死亡 ;2010 年超过半数的中国成年男性是吸烟者, 青少年和青年的吸烟率仍在上升 即使现在的吸烟比率保持稳定, 预计本世纪前十年每年吸烟相关的 100 万死亡到 2030 年将会加倍 吸烟相关疾病在开始吸烟 年之后才会开始显著, 不论控烟项目如何, 中国未来几十年的癌症负担将持续加重 尽管当前中国的吸烟状况不容乐观, 但国家正进行立法行动, 比如 15 年 6 月北京的控烟条例 如果能够在全国推行, 如果烟草产业能够从政府的烟草控制活动中分离开来, 这些改变将为中国的下一代大大减少烟草相关的癌症 经济增长 城市化和生活方式的西化导致了环境污染的加重 户外空气污染被认为是全世界最烂的, 用煤炭等化石燃料进行取暖和做饭导致室内空气污染, 土壤和饮用水污染, 这些都意味着环境污染中有很多致癌物质 尽管经过测量的环境污染的归因危险度很低, 但很多高发病率 高死亡率的 癌症村 为这两者之间的联系提供了强有力的证据 中国正在为减少环境污染作出努力 ; 然而立法和实施之间的差距仍然非常大 环境污染对健康的影响仍将持续好几十年, 尤其是面对着非常原始的生存环境的农村地区 4.2 对于中国早期诊断和管理的提示 尽管预防措施对于减少长期癌症负担有着重要的作用, 但这些措施无法在近期内见效 因此, 加强早期诊断和提升医疗服务将是快速缓解中国癌症负担的重要措施 尤其是, 地区间的重大差异证明了确保公平的诊断时间 癌症护理可用性和医疗服务质量的重要性 解决这个问题的一个巨大困难就是中国巨大的人口基数和地理的多样性 即便是按照目前的乳腺癌监测扩张速率, 仍需要 40 年来为目标年龄组的每一个女性进行一次检测 另外, 相比高收入国家更加年轻的诊断患病年龄中位数也为我国提出一些建议, 中国的资源应当集中于提高意识和检测乳房肿块时的早期诊断 尽管有这些地理和人口的障碍, 对于食道癌的内窥镜扫描项目正在扩张 另外, 新一代基于高风险 HPV 的扫描测试正逐步应用于中低收入的农村地区 由于 Ⅰ 期肺癌的手术治疗已经正式可以显著提升存活率, 因此使用低剂量的 CT 更早地探测肺癌不仅可以降低现有的死亡率, 也可以间接提升公共健康预防和控烟运动的有效性 由于中国很多医院继续使用 X 光来诊断肺癌, 增加医学检测容量, 尤其是在农村地区, 仍然有着很高的优先级 为了解决地理多样性和医疗资源分配不均 ( 城市有 30% 的人口和 70% 的医疗资源 ), 中国已经实施了癌症护理超级中心策略, 集中了很多癌症专家, 处理癌症病例的密度非常大 然而, 拿掉得到最优治疗的地理和财政屏障仍然有着最高的优先级, 因为农村人口和弱势群体不仅短缺医生, 得到医疗服务的距离也更远 另外, 尽管基本医疗保险几乎实现了全覆盖, 但它几乎不为癌症治疗负担哪怕部分费用, 这就意味着病人要么自费治疗, 要么放弃治疗 任何试图提升早期诊断和治疗的动机都需要考虑中国的独特传统和文化信仰 很多人得了癌症就听天由命, 不远谈论治疗和预后, 因为无论怎样, 得了癌症就会死 更好的理解这些思想对于实施正确的项目和提升医患间的信任非常重要 与此同时, 传统中医已经在中国的卫生体系里面存在了几千年, 与中国的文化 政策 历史都相关 因此, 或许可以将癌症的护理和治疗与中医研究中心整合起来

15 为了更好地量化早期诊断与治疗对于发病率和死亡率的影响, 我们还需要疾病所处阶段和接受治疗的数据 既 然这种数据在目前的中国癌症登记体系中是没有的, 这就需要具有足够大 有代表性 基于人群的特殊的调查研究 4.3 限制 尽管这篇研究中的数据覆盖人口是之前研究的两倍, 但这仍少于中国人口的十分之一 仍然有许多未知水平的不确定因素 尽管我们在控制数据质量方面已经付出了很大的努力, 但数据质量中仍然有许多变数 M/I 比率被用作近五年相对存活率的近似值, 这种解释可能是有问题的, 因为死亡率和发病率可能是涉及到完全不同的人群 这使得它更容易受发病率的影响, 因此为更容易死亡的癌症提供了更加精确的估计 然而,17 个登记点的未发布数据表明, 所有癌症的 M/I 比率只比计算得到的 5 年相对存活率高了 1.4% 最终, 对于一个 14 亿人的国家来说, 要保证分子面临的风险和分母相同实在是一个不小的挑战, 尤其是考虑到在大城市医疗机构里面治疗的病例和来自农村的移民病例 患病案例的地理信息是基于永久居住地而不是治疗地 另外, 通过城居和新农合得到的外出务工者 ( 占人口的 9%) 的癌症诊断, 都是基于他们的户口登记得到的 5 结论 为了制定一个恰当的癌症控制计划, 拥有一套细节的 有代表性的 精确的 基于人群的数据是非常关键的 这些评估和癌症登记的努力都是为了达成这一目标所进行的重要步骤 尽管这些全国估算中仍然有不精确的地方, 但这都是基于可用的最优数据来进行的发病率和死亡率的估计 这可能为中国未来的癌症防控提供可供比较的基线和评价标准, 并帮助发现最需要援助的地区 根据国际经验, 当需求更加明确 有更详实的证据支持时, 政府和其他卫生服务提供者将会更有动力提供帮助 根据这篇研究的数据, 中国正面临, 并且未来将继续面临极大的癌症压力, 因此需要政府和各非政府组织的共同努力 关键区域可能是总体水平上临床癌症护理水平的提升, 通过有目标的政策改革和投资来提升农村地区的医疗服务水平, 为弱势群体提供医疗服务 癌症的初级预防项目, 比如控烟和缓和西式生活方式的不良影响, 提高早期诊断的有效性和覆盖率, 这些对于逆转中国癌症的流行趋势至关重要 保证现有的空气和水污染控制法律得到有效实行仍是当务之急 考虑到中国对世界癌症负担的重要性, 特别是 4 种主要癌症 ( 肺癌 肝癌 胃癌 食道癌 ), 我们必须采取适当的策略和政策来减少这些可预防的癌症 ( 通过减少烟草的流行和与癌症相关的感染 ), 这将对中国和世界的癌症负担有着重要的影响

16 CA CANCER J CLIN 2016;66: Cancer Statistics in China, 2015 Wanqing Chen, PhD, MD 1 ; Rongshou Zheng, MPH 2 ; Peter D. Baade, PhD 3 ; Siwei Zhang, BMedSc 4 ; Hongmei Zeng, PhD, MD 5 ; Freddie Bray, PhD 6 ; Ahmedin Jemal, DVM, PhD 7 ; Xue Qin Yu, PhD, MPH 8,9 ; Jie He, MD 10 With increasing incidence and mortality, cancer is the leading cause of death in China and is a major public health problem. Because of China s massive population (1.37 billion), previous national incidence and mortality estimates have been limited to small samples of the population using data from the 1990s or based on a specific year. With high-quality data from an additional number of population-based registries now available through the National Central Cancer Registry of China, the authors analyzed data from 72 local, population-based cancer registries ( ), representing 6.5% of the population, to estimate the number of new cases and cancer deaths for Data from 22 registries were used for trend analyses ( ). The results indicated that an estimated 4292,000 new cancer cases and 2814,000 cancer deaths would occur in China in 2015, with lung cancer being the most common incident cancer and the leading cause of cancer death. Stomach, esophageal, and liver cancers were also commonly diagnosed and were identified as leading causes of cancer death. Residents of rural areas had significantly higher age-standardized (Segi population) incidence and mortality rates for all cancers combined than urban residents (213.6 per 100,000 vs per 100,000 for incidence; per 100,000 vs per 100,000 for mortality, respectively). For all cancers combined, the incidence rates were stable during 2000 through 2011 for males (10.2% per year; P 5.1), whereas they increased significantly (12.2% per year; P <.05) among females. In contrast, the mortality rates since 2006 have decreased significantly for both males (21.4% per year; P <.05) and females (21.1% per year; P <.05). Many of the estimated cancer cases and deaths can be prevented through reducing the prevalence of risk factors, while increasing the effectiveness of clinical care delivery, particularly for those living in rural areas and in disadvantaged populations. CA Cancer J Clin 2016;66: VC 2016 American Cancer Society. Keywords: cancer, China, health disparities, incidence, mortality, survival, trends Introduction Cancer incidence and mortality have been increasing in China, making cancer the leading cause of death since 2010 and a major public health problem in the country. 1 Much of the rising burden is attributable to population growth and ageing and to sociodemographic changes. Although previous estimates of the national incidence rates have been reported, they either represented a small sample of the Chinese population (<2%) 2 or were based on data from a specific year. 3,4 This has a bearing on the uncertainty of the estimates and their degree of national representativeness and, thus, would potentially limit the evidence available to develop appropriate policies for effective cancer control. Because the previous Program of Cancer Prevention and Control in China ( ) 5 was released more than 10 years ago, a more complete picture of the national and regional scale and profile in China would provide greater clarity in prioritizing and developing specific policies and programs across the spectrum of cancer control aimed at reducing the burden and suffering from the disease at the national level. Additional supporting information may be found in the online version of this article. Correction added on 3 February 2016, after first online publication: the definition of ASR has been corrected in the footnote of Table 3. 1 Deputy Director, National Office for Cancer Prevention and Control, National Cancer Center, Beijing, China; 2 Associate Researcher, National Office for Cancer Prevention and Control, National Cancer Center, Beijing, China; 3 Senior Research Fellow, Cancer Council Queensland, Brisbane, Queensland, Australia; 4 Associate Professor, National Office for Cancer Prevention and Control, National Cancer Center, Beijing, China; 5 Associate Professor, National Office for Cancer Prevention and Control, National Cancer Center, Beijing, China; 6 Head, Section of Cancer Surveillance, International Agency for Research on Cancer, Lyon, France; 7 Vice President, Surveillance and Health Services Research Program, American Cancer Society, Atlanta, GA; 8 Research Fellow, Cancer Council New South Wales, Sydney, New South Wales, Australia; 9 Adjunct Lecturer, Sydney School of Public Health, University of Sydney, Sydney, New South Wales, Australia; 10 Director, National Cancer Center, Beijing, China. The last 2 authors contributed equally to this article. Corresponding author: Jie He, MD, Director, National Cancer Center, No. 17 Pan-jia-yuan South Lane, Chaoyang District, Beijing, China; hejie@cicams.ac.cn We thank the Bureau of Disease Control, National Health and Family Planing Commission and Cancer Institute & Hospital, Chinese Academy of Medical Sciences for their support to this study. We would like to express our gratitude to all staff of the contributing cancer registries who have made a great contribution to the study, especially on data collection, supplements, auditing, and cancer registration database management. DISCLOSURES: This study is supported by a National Program Grant to the Cancer Registry from the National Health and Family Planning Commission of China and by a Program Grant in Fundamental Research from the Ministry of Science and Technology (no. 2014FY121100). The authors report no conflicts of interest. doi: /caac Available online at cacancerjournal.com VOLUME 66 _ NUMBER 2 _ MARCH/APRIL

17 Cancer Statistics in China, 2015 This study reports the most recent cancer incidence, mortality, and survival estimates nationally; the most recent incidence and mortality patterns for several major cancers by geographic area in China; and temporal trends for some major cancers as well as the implications of this information for cancer control in China. Data Sources and Methods Cancer Registration in China The National Central Cancer Registry of China (NCCR), established in 2002, is responsible for the collection, evaluation, and publication of cancer data in China. Cancer diagnoses are reported to local cancer registries from multiple sources, including local hospitals and community health centers as well as the Urban Resident Basic Medical Insurance program and the New Rural Cooperative Medical Scheme. Since 2002, the implementation of standard registration practices has seen a marked improvement in the quality of cancer registration in China. In 2008, the National Program of Cancer Registries was launched by the Ministry of Health of China through a central financing mechanism. Since then, the number of local population-based registries in China has increased from 54 in 2008 (population coverage of 110 million) to 308 (population coverage of 300 million) in Not all of these registries currently have sufficiently high data quality for reporting purposes. The quality of submitted data for each local registry was checked and evaluated by the NCCR based on the Guidelines for Chinese Cancer Registration 6 and International Agency for Research on Cancer/International Association of Cancer Registries (IARC/IACR) data-quality criteria. 7 The assessments of quality measures include, but are not limited to, the proportion of morphologic verification (MV%), the percentage of cancer cases identified with death certification only (DCO%), the mortality(m) to incidence (I) ratio (M/I), the percentage of uncertified cancer (UB%), and the percentage of cancer with undefined or unknown primary site (CPU%). Only data from those local registries that consistently met appropriate levels of quality were included in these analyses. Detailed quality categories of the registry data can be found in a previous publication (Table 1). 8 Data classified as category A or B were deemed acceptable for inclusion in this study. The proportions of cancer registries that submitted data sets and were accepted for inclusion varied by year of submission, from 69.2% (72 of 104 registries) in 2009, 9 to 66.2% (145 of 219 registries) in 2010, 4 and 75.6% (177 of 234 registries) in We included data from the 72 cancer registries that were available for all 3 years. Cancer Incidence Data To estimate the numbers of new cancers in China in 2015, we used the most recent data (cancer cases registered during ) from 72 local population-based cancer registries TABLE 1. List of Population-Based Cancer Registries in China Used for Incidence/Mortality Estimates, Temporal Trends, or Survival Estimates PROVINCE REGISTRY PROVINCE REGISTRY Beijing Beijing*,, Jilin Dehui Hebei Qianxi Yanji Shexian Zhejiang Hangzhou Cixian*,, Jiaxing*, Baoding Jiashan*,, Shanxi Yangquan Haining*,, Yangcheng* Shangyu Inner Mongolia Chifeng Xianju Liaoning Shenyang Anhui Feixi Dalian, Maanshan Zhuanghe Tongling An shan Fujian Changle, Benxi Xiamen Dandong Jiangxi Zhanggong District, Ganzhou Donggang Shandong Linqu Heilongjiang Daoli District, Harbin Wenshang Nangang District, Feicheng, Harbin*, Shangzhi Henan Yanshi Shanghai Shanghai*, Linzhou, Jiangsu Jintan Xiping Suzhou Hubei Wuhan, Haian Yunmeng Qidong*,, Hunan Hengdong Haimen Guangdong Guangzhou Lianyungang Sihui, Donghai Zhongshan*,, Guanyun Guangxi Liuzhou Huaian District, Fusui Huai an Huaiyin District, Huai an Chongqing Jiulongpo District, Chongqing Xuyi Sichuan Qingyang District, Chengdu Jinhu Ziliujing District, Zigong Sheyang Yanting* Jianhu Gansu Jingtai Dafeng Liangzhou District, Wuwei Ganyu, Qinghai Xining Yangzhong Xinjiang Xinyuan Taixing *These are registries from which data were accepted by the International Agency for Research on Cancer for the most recent publication of Cancer Incidence in Five Continents (2014). These data were used for temporal trends analyses. These data were used for survival analyses. All registry data were used to estimate incidence and mortality in 2015 except data from Ganyu. (Table 1), providing a population coverage of about 85.5 million people, about 6.5% of the national population. A much smaller number of registries (n 5 22) (Table 1), which provided a population coverage of 44.4 million, had data of sufficient quality over the 12-year period ( ) for inclusion in incidence trend analyses. The locations of these 2 sets of cancer registries can be found in Figure 1. We included invasive tumors only in this study, and multiple primary cancers were defined using the international rules 116 CA: A Cancer Journal for Clinicians

18 CA CANCER J CLIN 2016;66: FIGURE 1. Maps of the 2 Sets of Contributing Cancer Registries and Geographic Regions in China. Dots indicate the locations of the cancer registries. (A) This is a map for 22 cancer registries (data from 2000 to 2011). (B) This is a map for 72 cancer registries (data from 2009 to 2011). for multiple primary cancers. 10 Incidence data were extracted from the NCCR database. Although cancer site information is available through codes from both the International Classification of Diseases for Oncology, 3rd revision (ICD-O-3), and the International Classification of Diseases, 10th revision (ICD-10), we have reported incidence data using the ICD-10 classification for consistency, with mortality data that were only available in the ICD-10 classification. The variables extracted were sex, age, date of birth, year of diagnosis, cancer site, morphology, residence (urban and rural areas), and region (North China, Northeast, East China, Central China, South China, Southwest, Northwest China). For age-specific VOLUME 66 _ NUMBER 2 _ MARCH/APRIL

19 Cancer Statistics in China, 2015 incidenceanalysis,5broadagegroupswereused(younger than 30 years, years, years, years, and 75 years or older). Cancer Mortality Data Mortality data compiled by 72 local cancer registries were used to estimate the numbers of cancer deaths in China in To estimate trends in mortality because of cancer between 2000 and 2011, mortality data were obtained from the same 22 local registries that were used in the incidence trend analyses. These registries compile data on cancer deaths from local hospitals, community health centers, vital statistics (including data from the national Disease Surveillance Points [DSP] system), and the Civil Administration Bureau. 11 The DSP system, which was established by the Ministry of Health in the early 1980s, routinely collects information on deaths based on the death certificate provided by hospitals or obtained from the next of kin by a household visit if a death certificate was unavailable. 11 While the DSP uses a nationally representative sample of sites, these cover only a very small (1%) proportion of the population. 12,13 Cancer Survival In the absence of precise follow-up information from the 72 cancer registries, estimates of 5-year relative survival in China for 2015 are presented in this report using the complement of the cancer M/I ratio (1-M/I) from these registries, a method that has been used previously. 14,15 We present these survival estimates only for all cancers combined, because survival from certain cancer types may have led to an overestimation or underestimation using this proxy measure. 14 We calculated the M/I ratio for all cancers combined by assuming that the ratio between incidence and mortality has not changed between 2009 to 2011 and 2015, so we divided the age-standardized mortality rate ( ) by the age-standardized incidence rate ( ). Population Data National population data by 5-year age group and sex were obtained from statistics or public security census (data.stats. gov.cn/; accessed April 20, 2014). 8 Individual registries provided population data in the respective areas to the NCCR. These data were sourced from local Statistical or Public Security Bureaus or from calculations based on census data. Statistical Analysis We estimated the numbers of new cases in China in 2015 for all cancers combined and for 26 individual cancer types by sex by applying age-specific incidence rates from 72 cancer registries ( ) to the projected age-specific population in China in We estimated the numbers of cancer deaths in China in 2015 using the same method. For the 10 most common cancers, these estimated numbers of new cases and cancer deaths were further stratified by urban/rural registries and by 7 administrative regions that cover China. Agespecific numbers of new cases and cancer deaths by 5 broad age groups (younger than 30 years, years, years, years, and 75 years or older) are also presented for all cancers combined and for the 6 most common cancers for Temporal trends in incidence and mortality rates from 2000 to 2011 (22 registries) were examined by fitting joinpoint models 16,17 to the log-transformed, age-standardized rates (per 100,000 population), standardized according to the world standard population. 18 To reduce the possibility of reporting spurious changes in trends over the period, all models were restricted to a maximum of 2 joinpoints (3 line segments). Trends were expressed as an annual percentage change (APC), and the Z test was used to assess whether the APC was statistically different from zero. In describing trends, the terms increase or decrease were used when the slope (APC) of the trend was statistically significant (P <.05). For nonstatistically significant trends, the term stable was used. For all those analyses, we present the results for all cancers combined and for the 10 most common cancers stratified by sex. Results Data Quality The 3 main measures (MV%, DCO%, and M/I ratio) of data quality for population-based cancer registries, stratified by cancer type, show that overall data quality is reasonably good for both sets of cancer registries (Fig. 2). Because half of the 22 cancer registries are certified by the IARC, their data quality was considerably higher than when considering data from the combined 72 cancer registries, indicated by higher MV% and lower DCO%. Values of these dataquality measures plus UB% (the percentage of uncertified cancer) and CPU% (the percentage of cancer with undefined or unknown primary site) are presented Supporting Table 1 (see online supporting information). Expected Cancer Incidence in 2015 It is predicted that there will be about 4292,000 newly diagnosed invasive cancer cases in 2015 in China, corresponding to almost 12,000 new cancer diagnoses on average each day. The 5 most commonly diagnosed cancers among men, in descending order, are: cancers of the lung and bronchus, stomach, esophagus, liver, and colorectum, accounting for about two-thirds of all cancer cases. The corresponding cancers among women are breast, lung and bronchus, stomach, colorectum, and esophagus, accounting for nearly 60% of all cases. Breast cancer alone is expected to account for 15% of all new cancers in women (Table 2). 118 CA: A Cancer Journal for Clinicians

20 CA CANCER J CLIN 2016;66: FIGURE 2. Three Major Measures of Data Quality by Cancer Types for the 2 Sets of Cancer Registries in China. Codes in the far left column are from the International Classification of Diseases, 10th Revision. DCO% indicates the percentage of cancer cases identified with death certification only; M/I, morality-to-incidence ratio; MV%, proportion of morphological verification. The estimated incidence rates for all cancers combined and the number of new cases for the most common 10 cancer types by urban versus rural status and region of residence are presented in Table 3. For all cancers combined, the age-standardized incidence rates per 100,000 population per year are higher in men than in women (234.9 vs per 100,000) and are higher in rural areas than in urban areas (213.6 vs per 100,000). Southwest China has the highest cancer incidence rates, followed by North China and Northwest China; and Central China has the lowest incidence rate. Expected Cancer Mortality in 2015 It is estimated that about 2814,000 Chinese will die from cancer in 2015, corresponding to over 7500 cancer deaths on average per day. The 5 leading causes of cancer death among both men and women are cancers of the lung and bronchus, stomach, liver, esophagus, and colorectum, accounting for about three-quarters of all cancer deaths (Table 2). Similar to the incidence rates, the agestandardized mortality rate for all cancers combined is substantially higher in men than in women (165.9 vs 88.8 per 100,000) and in rural areas than in urban areas (149.0 vs VOLUME 66 _ NUMBER 2 _ MARCH/APRIL

21 Cancer Statistics in China, 2015 TABLE 2. Estimated New Cancer Cases and Deaths (Thousands) by Sex: China, 2015* INCIDENCE MORTALITY SITE ICD-10 TOTAL MALE FEMALE TOTAL MALE FEMALE Lip, oral cavity, & pharynx (except nasopharynx) C00-C10, C12-C Nasopharynx C Esophagus C Stomach C Colorectum C18-C Liver C Gallbladder C23-C Pancreas C Larynx C Lung C33-C Other thoracic organs C37-C Bone C40-C Melanoma of the skin C Breast C Cervix C Uterus C54-C Ovary C Prostate C Testis C Kidney C64-C66, C Bladder C Brain, CNS C70-C Thyroid C Lymphoma C81-C85, C88, C90, C Leukemia C91-C All other sites and unspecified A_O All sites ALL CNS, central nervous system; ICD-10, International Classification of Diseases, 10th revision. *The total number of cases projected for 2015 are based on the average incidence rates for the most recent 3 years (2009 to 2011) of data from 72 population-based cancer registries per 100,000) (Table 4). Likewise, the highest cancer mortality rates were found in Southwest China, followed by North China and Northwest China, with Central China exhibiting the lowest rate. Age-Specific Incidence and Mortality of Selected Cancers by Sex in 2015 Before the age of 60 years, liver cancer is the most commonly diagnosed cancer and the leading cause of cancer death in men, followed by lung and stomach cancer, which are the dominant types of cancer for both cases and deaths in the group ages 60 to 74 years (Table 5). Lung cancer is the most commonly diagnosed cancer and the leading cause of cancer death in men aged 75 years or older. Most new cancer cases and cancer deaths in men occur in the age range from 60 to 74 years. Among women, thyroid cancer is the most commonly diagnosed cancer before the age of 30 years, followed by breast cancer at ages 30 to 59 years, and lung cancer in women aged 60 years or older (Table 5). Breast cancer is the leading cause of cancer death in women younger than 45 years, followed by lung cancer. The largest proportion of new cancer cases and deaths among women are diagnosed among those between ages 60 and 74 years. Expected Cancer Survival in 2015 It is predicted that, for all cancers combined, 36.9% of cancer patients in China will survive at least 5 years after diagnosis around 2015, with women having much better survival than men (47.3% vs 29.3%) (Table 6). There is substantial variation in the 5-year survival estimate according to residence at the time of diagnosis: rural patients have much lower survival than their city counterparts (30.3% vs 42.8%). Likewise, the lowest survival rates were found in Southwest China (24.9%), with Central China showing the highest rate (41.0%). Trends in Cancer Incidence and Mortality For all cancers combined, the age-standardized incidence rates were stable over the study period ( ) for males, while significant upward trends were observed for females (P <.05) (Fig. 3, Table 7). In contrast, the agestandardized mortality rates decreased significantly for both males and females (Fig. 3, Table 8). Despite this favorable trend, however, the number of cancer deaths substantially increased (73.8% increase) during the corresponding period (from 51,090 in 2000 to 88,800 in 2011) because of the aging and growth of the population (Fig. 4). 120 CA: A Cancer Journal for Clinicians

22 CA CANCER J CLIN 2016;66: TABLE 3. Age-Standardized (Segi Standard Population) Incidence Rates for All Cancers Combined and Estimated New Cases (Thousands) for Selected Cancers in China, 2015, by Geographic Location ESTIMATED NEW CASES (THOUSANDS) AREAS SEX ASR* ALL CANCERS LUNG ESOPHAGUS STOMACH COLORECTUM LIVER BREAST CERVIX THYROID BRAIN PANCREAS All areas Total Male Female Urban areas Total Male Female Rural areas Total Male Female North China Total Male Female Northeast Total Male Female East China Total Male Female Central China Total Male Female South China Total Male Female Southwest Total Male Female Northwest Total Male Female ASR, age-standardized rate. *Age-standardized incidence rates for all cancers are based on the Segi standard population. Among the 10 most common cancers considered in the temporal trend analyses for men, incidence rates from 2000 to 2011 increased for 6 cancer types (pancreas, colorectum, brain and central nervous system, prostate, bladder, and leukemia), whereas the rates decreased for cancers of the stomach, esophagus, and liver (P <.05). A stable trend was observed for cancer of the lung (Fig. 5, Table 7). For women, 6 of the 10 most common cancers had a significant upward trend in age-standardized incidence rates (cancers of the colorectum, lung, breast, cervix, uterine corpus, and thyroid; P <.05). As with men, a downward trend was seen for cancers of the stomach, esophagus, and liver (Fig. 6, Table 7). An upward trend in age-standardized mortality rates was observed for 4 of the 10 most common cancers in men (colorectum, pancreas, prostate, and leukemia; P <.05), whereas stable trends were seen for other cancer types (cancers of the lung, bladder, and brain) (Fig. 7, Table 8). In women, an increasing trend in mortality was observed for 3 of the 10 most common cancers (breast, cervix, and ovary), with trends stable for colorectum,lung,uterine,andthyroidcancers(fig.8,table8). Similar to the trends in cancer incidence rates, declining trends in age-standardized mortality rates were observed for cancers of the stomach, esophagus, and liver in both sexes (Figs. (7 and 8)). Stable trends were observed in both men and women for lung cancer, which was the leading cause of cancer mortality for both men and women. Discussion Although national estimates of cancer for China have been previously reported, these are limited to only a snapshot of the patterns by cancer site during a single year 3,4,12,19 or are reported for specific cancers, making comparisons of trends across cancer types difficult. This study provided more comprehensive nationwide cancer statistics in China using the latest and most representative data and including information on temporal trends. Cancer prevention and control rely on population-based incidence and mortality data as an incentive both to act and to assess the effectiveness of current interventions and policies. VOLUME 66 _ NUMBER 2 _ MARCH/APRIL

23 Cancer Statistics in China, 2015 TABLE 4. Age-Standardized (Segi Standard Population) Mortality Rates for All Cancers Combined and Estimated Deaths (Thousands) for Selected Cancers in China, 2015, by Geographic Location ESTIMATED DEATHS (THOUSANDS) AREAS SEX ASR* ALL CANCERS LUNG ESOPHAGUS STOMACH COLORECTUM LIVER BREAST CERVIX THYROID BRAIN PANCREAS All areas Total Male Female Urban areas Total Male Female Rural areas Total Male Female North China Total Male Female Northeast Total Male Female East China Total Male Female Central China Total Male Female South China Total Male Female Southwest Total Male Female Northwest Total Male Female ASR, age-standardized mortality rate. *ASRs for all cancers are based on the Segi standard population. Thus, the updated nationwide estimates of cancer burden and time trends presented here are critical to understanding the etiology of cancer and the effectiveness of prevention, early detection, and management of cancer in China. These results will also serve as a baseline for future assessment of the overall effectiveness of the cancer control effort in China and will provide insights into the areas of greatest need for prioritization. Because of China s large population size, approximately one-fifth of the world population, these Chinese data contribute significantly to the global burden of cancer: almost 22% of global new cancer cases and close to 27% of global cancer deaths occur in China. 23 More importantly, the cancer profile in China is markedly different from those of developed countries. The 4 most common cancers diagnosed in China were lung, stomach, liver, and esophageal cancer. These cancers account for 57% of cancers diagnosed in China, compared with 18% in the United States. 24 Also, these cancers diagnosed in China comprise between onethird and one-half of the global incidence burden from lung, stomach, liver, and esophageal cancers. 23,25 In comparison, the most common cancers diagnosed in the United States are cancers of the lung, breast, prostate, and colorectum. 26 The most common cancers in China are those associated with rather poor survival; whereas those in the United States, with the exception of lung cancer, are dominated by cancers with a good to excellent prognosis, and, for prostate and breast cancers, the incidence may be inflated by diagnostic activities linked to early detection and screening. 24,27 This difference in cancer type distribution contributes significantly to the higher overall cancer mortality rate in China. Our cancer estimates for 2015 were based on data from 72 Chinese population-based cancer registries capturing cancer diagnoses from 2009 to These registries cover only about 6.5% of the Chinese national population, but they remain the best-available nationwide data for cancer incidence, representing a base population of 85.5 million people. Moreover, the data used in this study have an enlarged population coverage compared with previous 122 CA: A Cancer Journal for Clinicians

24 CA CANCER J CLIN 2016;66: TABLE 5. Estimated New Cancer Cases and Deaths (Thousands) for Selected Cancers by Age Groups: China, 2015 SITE < ALL Male (thousands) Incidence Prostate Colorectum Esophagus Liver Stomach Lung All sites Mortality Pancreas Colorectum Esophagus Liver Stomach Lung All sites Female (thousands) Incidence Thyroid Cervix Colorectum Stomach Lung Breast All sites Mortality Breast Colorectum Liver Esophagus Stomach Lung All sites AGE, y studies (with coverage less than 2% of the population), 2,28 including more registries in the western regions of China and are thus more representative of the general population in China. In addition, all 12 population-based cancer registries in mainland China with high-quality data that fulfilled the Cancer Incidence in Five Continents (CI5) inclusion criteria for the present CI5 volume X were included in the analysis. 29 These national incidence estimates for China are broadly comparable to those published previously. 3,4,12 The 2 most recently published annual reports in China found that the estimated number of new cancer cases were 3.09 million and 3.37 million for 2010 and 2011, respectively. 3,4 An earlier estimate was 2.96 million for 2005, although a different method was used. 12 Our estimate of cancer incidence for 2015 (4.29 million cases) in China is considerably higher than that reported by the GLOBOCAN 2012 initiative of 3.40 million. 23,25 Reasons for these discrepancies may include differences in data timelines ( vs ) and representativeness and geographic coverage (72 cancer registries covering 6.5% of the national population vs 23 cancer registries covering 3.0% of the population). In particular, rural residents, who have a higher incidence rate than urban residents (213.6 per 100,000 vs per 100,000), account for 32.7% the population in our estimates compared with 21.5% in the 2012 GLOBOCAN estimates. The methods used to obtain the national incidence estimates are also different, because GLOBOCAN 2012 converts national mortality estimates for 2012 to incidence by modeling the agespecific, sex-specific, and site-specific M/I ratios from the 23 Chinese cancer registries. While acknowledging that not all of these cancer registries met the IARC quality standards, 15,30 these discrepancies in published estimates underscore the need for further improvements in the coverage and quality of registries in China to provide more accurate statistics on the cancer burden in the country. In contrast to incidence, there was greater consistency in our mortality estimates and those in GLOBOCAN. We estimated that there would be 2.81 million cancer deaths in 2015, and the corresponding figure was 2.46 million in GLOBOCAN This may reflect that the mortality data used by both studies were obtained from a similar source: DSP data ( ) were used for GLOBO- CAN 2012, while we used mortality data collected from VOLUME 66 _ NUMBER 2 _ MARCH/APRIL

25 Cancer Statistics in China, 2015 TABLE 6. AREAS Expected 5-Year Survival for All Cancers Combined by Sex and Geographic Area: China, 2015 SEX ASR INCIDENCE* ASR DEATHS* 1-(M/I) All areas Total Male Female Urban areas Total Male Female Rural areas Total Male Female North China Total Male Female Northeast Total Male Female East China Total Male Female Central China Total Male Female South China Total Male Female Southwest Total Male Female Northwest Total Male Female (M/I), complement to the mortality (M) to incidence (I) ratio; ASR, agestandardized mortality rate. *Age-standardized rates for all cancers are based on the Segi standard population. 72 cancer registry areas ( ) for which DSP data were part of the whole set of data on cancer deaths. The DSP data were based on counties and stratified by geographic regions, with sampling further stratified by urban or rural location and per capita gross domestic product, and the DSP system was specifically designed to be nationally representative. 19 For both incidence and mortality estimates, data from Hong Kong and Macao were also used in the estimates from GLOBOCAN 2012 but were excluded from our analyses. We found significant differences in cancer incidence rates for all cancers combined by place of residence (rural vs urban and between regions) in China. Rural residents have higher incidence compared with their urban counterparts, and incidence rates varied substantially across the 7 administrative regions. It is likely that many factors contribute to this geographic differential, but the higher smoking prevalence in rural populations compared with those in urban areas 31,32 likely plays a dominating role. The 12 cancers formally established as being caused by smoking 33 account for about 75% of all cancers combined in China. Consistent with this hypothesis, the area with the highest observed cancer incidence rate, Southwest China, was also reported to have the cities with the highest smoking prevalence in We found even greater geographic variations in cancer mortality and the survival proxies across China. It is likely that at least part of these geographic disparities could be explained by the more limited medical resources, lower levels of cancer care, and a larger proportion of patients diagnosed with cancer at a late stage in rural and underdeveloped areas in China. 27 Recognizing that differences in other factors, such as competing causes of death and comorbidities, may also contribute, the results reported here, together with those reported previously, 27,35 provide a strong justification for providing more government-funded health resources and services for cancer control in rural and underdeveloped areas in China to reduce these apparent inequalities. There has been a marked increase in the numbers of cancers diagnosed in China between 2000 and Much of this is explained by the aging and growth of the population. Other factors that may have contributed to the increase in the burden of cancer include increases in the prevalence of unhealthy behaviors or cancer-related lifestyle and improvements in disease awareness, detection services, and data completeness. The largest increase in incidence was seen for cancers of the prostate, cervix, and thyroid for women. The factors driving the increase in prostate cancer are not entirely understood; however, they may include gradual implementation of prostate-specific antigen screening and improved biopsy techniques 36 or the impact of an increasingly westernized lifestyle. 37,38 Westernized lifestyle, particularly increases in the prevalence of obesity and physical inactivity in recent decades in China, is likely to have had an impact on the observed rise in colorectal and breast cancer incidence. 39,40 For breast cancer, the increasing trend may also reflect changes in reproductive behavior in China in recent decades because of the one-child policy implemented since the 1970s. 41 In contrast to the decreasing incidence trends in developed countries, a substantial increase in cervical cancer incidence was seen in China. This may reflect the inadequacy of Papanicolaou (Pap) test screening in China, because only one-fifth of Chinese women reported having ever had a Pap test for cervical cancer screening. 42,43 The increasing prevalence of human papillomavirus (HPV) infection, especially in younger women, 44,45 and the lack of HPV vaccines in mainland China due to the absence of formal drug approvals, 39 suggest that the disparity in cervical cancer incidence trends between China and international countries may continue for the foreseeable future. The dramatic rise in thyroid cancer among women is consistent with that observed in other countries ; and, while it may reflect overdiagnosis through increased use of new imaging technologies (ultrasound, computed 124 CA: A Cancer Journal for Clinicians