对人群每年进行低剂量胸部CT筛查可提高早期肺癌诊断率,但其假阳性率较高,常导致不必要的手术。本研究拟建立肺癌家系风险度预测模型,从中细分高危人群,从而提高筛选效能。
以经病理确诊的肺癌患者的家系作为研究人群,同时收集先证者的配偶家系作为对照家系,共收集先证者家系633例和对照家系565例。应用SPSS 17.0进行统计学分析。
先证者家系一级亲属患肿瘤的风险性为对照组家系一级亲属的1.71倍(OR=1.71,
如果一级亲属患癌个数越多,患肺癌的风险越高。根据本研究建立的风险度预测模型,风险度达普通人群10倍以上的主要为重度吸烟的吸烟人群,应加强筛查。特点为:有肺部既往疾病史的重度吸烟人群,加上男性、职业暴露和一级亲属肿瘤家族史三项中的任一项;有肺部既往疾病史或重度吸烟的人群中,有职业暴露的男性且一级亲属有不少于两位肿瘤患者。
Annual lung screening using spiral computed tomography (CT), has a high sensitivity of detecting early lung cancer (LC), but its high rates of false-positive ofen lead to unnecessary surgery. The aim of this study is to create a forecasting model of high risk individuals to lung cancer.
The pathologic diagnoses of LC in Guangdong Lung Cancer Institute were consecutively chosen as the probands. All the members of the frst-degree relatives of probands' and their spouses' were enrolled in this study. Tese pedigrees consisted of 633 probands' pedigrees and 565 spouses' pedigrees. Unless otherwise stated, analyses were performed using the SPSS 17.0 statistical sofware package.
Compared with the control, a family history of carcinoma in frst-degree relatives was signifcantly associated with LC risk (OR=1.71,
A family history of carcinoma in frst-degree relatives was signifcantly associated with increased LC risk. The more infected individuals exist in frst-degree relatives, the more risk was showed. In the forecasting model, smokers especially heavy ones whose risk were more than 10 times to the Chinese population should be receive annual screening. The population are positive at least any two conditions which including male, lung disease history, occupation expose and history of cancer in frst-degree relative.
大多数人类肿瘤和环境因素相关,但同样暴露于特定致癌物,却仅部分人群发病。另外,某些肿瘤也有明显的家族聚集现象。可见,除环境因素外,遗传背景尤其是基因的多态性差异也是重要的决定因素。来自冰岛等地的研究揭示,家族聚集可表现为不同类型肿瘤的聚集,提示存在共同的遗传因素。例如,雌激素相关基因可能和经产妇患乳腺癌与肺癌风险性增加存在交叉联系[
连续性收集2009年10月-2010年12月于广东省肺癌研究所病理学确诊的肺癌患者,并以其作为先证者,其所在家系确定为先证者家系,其一级亲属(被调查者的父母、子女及同胞)为家系成员。因肺癌先证者的子女发病率极低,与大部分年龄未到发病高峰期有关,且绝大部分子女为先证者和配偶对照组的共同子女,因此统计时将子女从一级亲属范围剔除。
对照组为肺癌先证者的配偶家系,纳入研究的配偶无肿瘤史,与肺癌家系成员之间不存在任何血缘关系。由同一调查员采用相应调查表记录对照家系成员的一般情况及家系资料,另一调查员进行复核。
由调查员对初治的肺癌患者进行面访,在先证者本人或其亲属签定知情同意书后,应用统一的调查表,对肺癌先证者及其配偶进行调查。调查表内容包括性别、年龄分组、吸烟指数、肺部既往疾病史、居住环境、职业接触、一级亲属肿瘤家族史和亲属的情况等。由最了解情况者作为问讯对象,提高可靠性。为了减少回忆偏倚,我们尽可能增加了样本量,同时又让调查对象对一些不确定情况通过电话咨询的方式进行证实。
用EpiData 3.1软件建立数据库,应用SPSS 17.0对先证者及其家系资料和对照资料进行统计学分析。分类资料的比较采用卡方检验,如不满足卡方检验条件者采用
先证者及对照组的籍贯属于中国东部沿海的25个省或直辖市,广东省内患者来自广东省21个地级市。两者家系在年龄、地区、居住环境和一级亲属人数的比较,无统计学差异(
临床流行病学资料比较
Comparison of epidemiology data
Characteristic | Case ( | Control ( | |
a: Missing values were not included in percentage calculations. | |||
Age cohort | 0.070 | ||
50 years | 125 (19.7) | 136 (24.1) | |
≥50 years | 508 (80.3) | 429 (75.9) | |
Provincea | 0.550 | ||
Guangdong Province | 461 (72.8) | 402 (71.3) | |
Non-Guangdong Province | 172 (27.2) | 162 (28.7) | |
Daily contact | 0.374 | ||
No | 527 (83.3) | 481 (85.1) | |
Yes | 106 (16.7) | 84 (14.9) |
由
病例组和对照组年龄分布图
The distribution of age in case arm and control arm
由
先证者一级亲属患癌风险
Risk of cancer in first-degree relatives of proband
Factor | Case/Control | Crude OR (95%CI) | Adjusted ORa (95%CI) |
a: ORs calculated by conditional logistic regression adjusted for sex, lung disease history, smoking status, living environment and occupation expose. | |||
Proband’s family history | |||
No | 432/438 | 1.00 | 1.00 |
Yes | 201/127 | 1.60 (1.24-2.08) | 1.71 (1.28-2.28) |
N of families with | |||
0 | 432/438 | 1.00 | 1.00 |
1 | 149/111 | 1.36 (1.03-1.80) | 1.55 (1.14-2.12) |
2+cancers | 52/16 | 3.30 (1.85-5.86) | 2.65 (1.42-4.94) |
先证者的一级亲属患癌风险性(以年龄分层)
Risk of cancer in first-degree relatives of Proband stratified by age
Age cohort | Factors | Case/Control | Crude OR (95%CI) | Adjusted ORa (95%CI) | |
a: ORs calculated by conditional logistic regression adjusted for sex, lung disease history, smoking status, living environment and occupation expose. | |||||
< 50 years | Proband’s family history | No | 86/112 | 1.00 | 1.00 |
Yes | 39/24 | 2.12 (1.18-3.78) | 2.01 (1.08-3.74) | ||
N of families with | 0 | 86/112 | 1.00 | 1.00 | |
1 | 32/23 | 1.81 (0.99-3.32) | 1.71 (0.90-3.28) | ||
2+cancers | 7/1 | 9.12 (1.10-75.50) | 8.85 (1.01-77.80) | ||
≥50 years | Proband’s family history | No | 346/326 | 1.00 | 1.00 |
Yes | 162/103 | 1.48 (1.11-1.98) | 1.64 (1.18-2.27) | ||
N of families with | 0 | 346/326 | 1.00 | 1.00 | |
1 | 117/88 | 1.25 (0.91-1.72) | 1.52 (1.07-2.17) | ||
2+cancers | 45/15 | 2.83 (1.55-5.17) | 2.21 (1.14-4.28) |
为控制混杂因素,提高预测准确率,建立以性别(女、男)[
预测模型的二分类
A binary
Variable | B | SE | Wald | OR (95%CI) | |
Sex | 0.464 | 0.178 | 6.78 | 0.009 | 1.59 (1.12-2.26) |
Smoking status | 62.69 | < 0.001 | |||
Light smoker | 0.135 | 0.241 | 0.31 | 0.576 | 1.14 (0.71-1.84) |
Heavy smoker | 1.545 | 0.211 | 53.74 | < 0.001 | 4.69 (3.10-7.09) |
Lung disease history | 1.689 | 0.322 | 27.49 | < 0.001 | 5.41 (2.88-10.18) |
Occupation expose | 0.470 | 0.172 | 7.44 | 0.006 | 1.60 (1.14-2.24) |
N of families with | 15.02 | 0.001 | |||
1 | 0.436 | 0.158 | 7.63 | 0.006 | 1.55 (1.14-2.11) |
2+cancers | 0.957 | 0.318 | 9.06 | 0.003 | 2.60 (1.40-4.86) |
Constant a′ | -0.867 | 0.102 | 72.03 | < 0.001 | 0.420 |
根据
① 模型用于预测:由于病例对照研究的非条件
其中
研究对象的肺癌发病概率预测及相对风险度
Prediction of lung cancer morbidity and relative risk in the study objects
Row | Sex | Lung disease | Occupation | Light | Heavy | Infected | Infected | Constant | Probability | RRa |
0.464 | 1.689 | 0.47 | 0.135 | 1.545 | 0.436 | 0.957 | -0.867 | |||
a: prediction of study object with lung cancer morbidity compared to Chinese population. | ||||||||||
1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0.000, 186 | 0.38 | |
2 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0.000, 213 | 0.43 | |
3 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0.000, 288 | 0.58 | |
4 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0.000, 296 | 0.60 | |
5 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0.000, 298 | 0.60 | |
6 | 0 | 0 | 0 | 1 | 0 | 1 | 0 | 0.000, 330 | 0.66 | |
7 | 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0.000, 339 | 0.68 | |
8 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0.000, 341 | 0.69 | |
9 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 0.000, 474 | 0.95 | |
10 | 0 | 0 | 0 | 0 | 0 | 0 | 1 | 0.000, 485 | 0.98 | |
11 | 0 | 0 | 1 | 1 | 0 | 1 | 0 | 0.000, 528 | 1.06 | |
12 | 1 | 0 | 0 | 1 | 0 | 1 | 0 | 0.000, 525 | 1.06 | |
13 | 1 | 0 | 1 | 1 | 0 | 0 | 0 | 0.000, 543 | 1.09 | |
14 | 0 | 0 | 0 | 1 | 0 | 0 | 1 | 0.000, 556 | 1.12 | |
15 | 1 | 0 | 1 | 1 | 0 | 1 | 0 | 0.000, 840 | 1.69 | |
16 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0.000, 874 | 1.76 | |
17 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0.000, 884 | 1.78 | |
18 | 0 | 0 | 1 | 1 | 0 | 0 | 1 | 0.000, 889 | 1.79 | |
19 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0.001, 009 | 2.03 | |
20 | 0 | 1 | 0 | 1 | 0 | 0 | 0 | 0.001, 155 | 2.32 | |
21 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | 0.001, 351 | 2.72 | |
22 | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0.001, 390 | 2.80 | |
23 | 0 | 0 | 1 | 0 | 1 | 0 | 0 | 0.001, 398 | 2.81 | |
24 | 1 | 0 | 1 | 1 | 0 | 0 | 1 | 0.001, 414 | 2.84 | |
25 | 1 | 1 | 0 | 0 | 0 | 0 | 0 | 0.001, 605 | 3.23 | |
26 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 0.001, 615 | 3.25 | |
27 | 0 | 1 | 0 | 1 | 0 | 1 | 0 | 0.001, 786 | 3.59 | |
28 | 1 | 1 | 0 | 1 | 0 | 0 | 0 | 0.001, 837 | 3.70 | |
29 | 0 | 1 | 1 | 1 | 0 | 0 | 0 | 0.001, 848 | 3.72 | |
30 | 1 | 0 | 0 | 0 | 1 | 1 | 0 | 0.002, 149 | 4.32 | |
31 | 0 | 0 | 1 | 0 | 1 | 1 | 0 | 0.002, 162 | 4.35 | |
32 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0.002, 224 | 4.47 | |
33 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 0.002, 275 | 4.58 | |
34 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 0.002, 568 | 5.17 | |
35 | 1 | 1 | 0 | 1 | 0 | 1 | 0 | 0.002, 841 | 5.72 | |
36 | 0 | 1 | 1 | 1 | 0 | 1 | 0 | 0.002, 858 | 5.75 | |
37 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0.002, 939 | 5.91 | |
38 | 0 | 1 | 0 | 1 | 0 | 0 | 1 | 0.003, 008 | 6.05 | |
39 | 1 | 0 | 1 | 0 | 1 | 1 | 0 | 0.003, 439 | 6.92 | |
40 | 1 | 0 | 0 | 0 | 1 | 0 | 1 | 0.003, 619 | 7.28 | |
41 | 0 | 0 | 1 | 0 | 1 | 0 | 1 | 0.003, 641 | 7.33 | |
42 | 1 | 1 | 1 | 1 | 0 | 1 | 0 | 0.004, 546 | 9.15 | |
43 | 0 | 1 | 0 | 0 | 1 | 0 | 0 | 0.004, 731 | 9.52 | |
44 | 1 | 1 | 0 | 1 | 0 | 0 | 1 | 0.004, 784 | 9.62 | |
45 | 0 | 1 | 1 | 1 | 0 | 0 | 1 | 0.004, 812 | 9.68 | |
46 | 1 | 0 | 1 | 0 | 1 | 0 | 1 | 0.005, 790 | 11.65 | |
47 | 0 | 1 | 0 | 0 | 1 | 1 | 0 | 0.007, 317 | 14.72 | |
48 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 0.007, 525 | 15.14 | |
49 | 0 | 1 | 1 | 0 | 1 | 0 | 0 | 0.007, 570 | 15.23 | |
50 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 0.007, 654 | 15.40 | |
51 | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 0.011, 637 | 23.41 | |
52 | 0 | 1 | 1 | 0 | 1 | 1 | 0 | 0.011, 707 | 23.56 | |
53 | 1 | 1 | 1 | 0 | 1 | 0 | 0 | 0.012, 039 | 24.22 | |
54 | 0 | 1 | 0 | 0 | 1 | 0 | 1 | 0.012, 320 | 24.79 | |
55 | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 0.018, 619 | 37.46 | |
56 | 1 | 1 | 0 | 0 | 1 | 0 | 1 | 0.019, 593 | 39.42 | |
57 | 0 | 1 | 1 | 0 | 1 | 0 | 1 | 0.019, 711 | 39.66 | |
58 | 1 | 1 | 1 | 0 | 1 | 0 | 1 | 0.031, 349 | 63.08 |
由
② 模型用于判别以验证正确率(
预测模型效果检验
Classification tablea of forecasting model
Observed group | Predicted group | Percentage correct | |
Control | Case | ||
a: The cut value is 0.500. | |||
Control | 440 | 125 | 77.9 |
Case | 232 | 401 | 63.3 |
Total | 672 | 526 | 70.2 |
③ 风险度为普通人群10倍以上的群体预测正确率(
风险度为普通人群十倍以上的群体预测情况
Prediction of people whose degree of risk are more than ten times to the Chinese population
Probability | RR | Case/control (Forecast) | Case/control (Fact) | ||
46 | 0.005, 790 | 11.65 | 20 | 20/0 | 17/3 |
47 | 0.007, 317 | 14.72 | 0 | 0 | 0 |
48 | 0.007, 525 | 15.14 | 0 | 0 | 0 |
49 | 0.007, 570 | 15.23 | 2 | 2/0 | 2/0 |
50 | 0.007, 654 | 15.40 | 0 | 0 | 0 |
51 | 0.011, 637 | 23.41 | 0 | 0 | 0 |
52 | 0.011, 707 | 23.56 | 0 | 0 | 0 |
53 | 0.012, 039 | 24.22 | 26 | 26/0 | 22/4 |
54 | 0.012, 320 | 24.79 | 0 | 0 | 0 |
55 | 0.018, 619 | 37.46 | 8 | 8/0 | 8/0 |
56 | 0.019, 593 | 39.42 | 0 | 0 | 0 |
57 | 0.019, 711 | 39.66 | 0 | 0 | 0 |
58 | 0.031, 349 | 63.08 | 3 | 3/0 | 3/0 |
预测模型中风险度为普通人群十倍以上的群体效果检测
Classification tablea of people whose degree of risk are more than ten times to the Chinese population in the forecasting model
Observed group | Predicted group | Percentage correct | |
Control | Case | ||
a: The cut value is 0.500. | |||
Control | 0 | 7 | 0 |
Case | 0 | 52 | 100 |
Total | 0 | 59 | 88.1 |
吸烟是目前肺癌最重要的危险因素。其它的危险因素包括二手烟、生活接触、职业暴露、HPV等病毒感染、空气污染和结核等。基因易感性在年轻肺癌中起到尤其重要的意义。
肿瘤二次打击学说主要适用于有遗传倾向的肿瘤,如视网膜母细胞瘤等,对应的临床特点为:早年发病、病灶双侧或多发和家族聚集倾向,其中家族史是最明显的临床特征。Xu等[
本研究为单位时间内连续收集病例的大样本量病例对照研究,地域分布均衡,具有人群普遍性。Ziogas等[
美国肺癌遗传流行病学联盟2004年首次定位了和肺癌家系关联的区域——染色体6q23-25,并发现随着家系中癌症成员的增加,易感基因与6号染色体上的遗传标记的连锁相关性也增强[
日本有一项大规模的前瞻性队列研究JPHC研究[
早期肺癌筛查国际行动计划(International Early Lung Cancer Action Program, I-ELCAP)此前的研究数据[
本文针对吸烟指数、性别、年龄分组、一级亲属患癌个数、肺部既往疾病史、生活接触史和职业接触史,建立回归模型,赋值后得到各亚组肺癌发病概率与人群相比的风险度在0.38-63.08的结论,准确率为70.2%。这可能是因为在低风险的群体,暴露因素和疾病联系的强度不大从而影响了预测效率。而在联系强度为很强、风险度为普通人群10倍以上的群体,应用本模型的预测准确率为88.1%。特点为:有肺部既往疾病史的重度吸烟人群,加上男性、职业暴露和一级亲属肿瘤家族史三项中的任一项;有肺部既往疾病史或重度吸烟的人群中,有职业暴露的男性且一级亲属有不少于两位肿瘤患者。因此,建议风险度为人群10倍以上的高危人群可每年进行低剂量CT筛查,可望提高筛查效能。但因病例组和对照组为配偶关系,生活环境基本相同,所以生活接触史未保留在模型中,应用时应结合本因素综合考虑。