Assessing Individual Risk for High-Risk Early Colorectal Neoplasm for Pre-Selection of Screening in Shanghai, China: A Population-Based Nested Case–Control Study

Figures & data

Figure 1 Flow diagram for selection of study cohort and nested case–control participants.

Table 1 Characteristics of Participants in the Training and Validation Sets

	Training Set (n = 7752)	Validation Set (n = 3324)	P value
Age (year),mean (SD)	60.28 (6.34)	60.15 (6.41)	0.29
Sex, n (%)			0.44
Female	3931 (50.71)	1712 (51.50)
Male	3821 (49.29)	1612 (48.50)
BMI (kg/m^2), n (%)	23.88 (3.03)	23.90 (2.99)	0.62
18.5 ≤ BMI < 24	2876 (50.99)	1191 (50.17)
BMI < 18.5	146 (2.59)	60 (2.53)
24 ≤ BMI < 28	2128 (37.73)	930 (39.17)
BMI ≥ 28	490 (8.69)	193 (8.13)
Family history of CRC in first-degree relatives, n (%)	258 (3.34)	114 (3.44)	0.78
Smoking^a, n (%)	2636 (34.00)	1089 (32.81)	0.23
Alcohol^b, n (%)	1694 (21.86)	721 (21.75)	0.90
Regular intake of fatty food^c, n (%)	1670 (22.05)	725 (22.42)	0.67
Regular intake of fried food^d, n (%)	522 (7.04)	231 (7.28)	0.68
Regular intake of pickled food^e, n (%)	1835 (24.30)	762 (23.70)	0.52
Regular moderate-to-intensity physical activity^f, n (%)	5678 (81.84)	1936 (81.07)	0.41
Early colorectal neoplasm, n (%)			0.05
Adenomas	1122 (14.47)	493 (14.83)
Stage 0–I colorectal cancer	170 (2.19)	61 (1.84)

Notes: ^aSmoking status was classified as smokers (current/ever) or non-smoker. ^bAlcohol drinking status was classified as drinkers (current/ever) or non-drinker. ^cRegular intake of fatty meat means the intake 3–7 days per week ^dRegular intake of fried food means the intake 3–7 days per week. ^eRegular intake of pickled food means the intake 3–7 days per week. ^fRegular moderate-to-intensity physical activity means at least 1 day per week.

Abbreviations: BMI, body mass index; SD, standard deviation; CRC, colorectal cancer.

Table 2 Baseline Characteristics of Cases of Early Colorectal Neoplasm and Controls in the Training and Validation Datasets

	Training Data		P value	Validation Data		P value
	Cases (n = 1292)	Controls (n = 6460)		Cases (n = 554)	Controls (n = 2770)
Age,year,mean (SD)	60.90 (5.99)	60.16 (6.40)	0.00	61.1 (5.99)	59.95 (6.47)	0.00
Sex, n(%)			0.00			0.00
Male	799 (61.84)	3022 (46.78)		345 (62.27)	1267 (45.74)
Female	493 (38.16)	3438 (53.20)		209 (37.73)	1503 (54.26)
BMI, kg/m,^2 mean (SD)	24.19 (3.09)	23.81 (3.02)	0.00	24.20 (3.19)	23.84 (2.95)	0.02
18.5 ≤ BMI < 24	485 (47.41)	2391 (51.79)		204 (48.11)	987 (50.62)
BMI < 18.5	24 (2.35)	122 (2.64)		6 (1.42)	54 (2.77)
24 ≤ BMI < 28	397 (38.81)	1731 (37.49)		175 (41.27)	755 (38.72)
BMI ≥ 28	117 (11.44)	373 (8.08)		39 (9.20)	154 (7.90)
Family history, n (%)			0.00			0.00
Yes	89 (6.92)	169 (2.62)		51 (9.24)	63 (2.28)
No	1197 (93.08)	6270 (97.38)		501 (90.76)	2698 (97.72)
Smoking, n (%)			0.00			0.00
No	708 (54.80)	4408 (68.24)		310 (55.96)	1920 (69.44)
Current/ever	574 (45.20)	2052 (31.76)		244 (44.04)	845 (30.56)
Alcohol, n (%)			0.00			0.00
No	893 (69.12)	5161 (79.94)		384 (69.44)	2210 (80.01)
Current/ever	399 (30.88)	1295 (20.06)		169 (30.56)	552 (19.99)
Intake of fatty food^a, n (%)			0.05			0.96
Occasional	964 (75.85)	4941 (78.38)		417 (77.51)	2091 (77.59)
Regular	307 (24.15)	1363 (21.62)		121 (22.49)	604 (22.41)
Intake of fried food^a, n (%)			0.02			0.31
Occasional	1135 (91.38)	5754 (93.27)		486 (93.82)	2455 (92.50)
Regular	107 (8.62)	415 (6.73)		32 (6.18)	199 (7.50)
Intake of pickled food^a, n (%)			0.01			0.07
Occasional	926 (72.91)	4790 (76.26)		392 (73.27)	2061 (76.90)
Regular	344 (27.09)	1491 (23.74)		143 (26.73)	619 (23.10)
Moderate-to-intensity physical activity^b, n (%)			0.00			0.18
Never	225 (21.99)	806(17.31)		90 (21.23)	362 (18.43)
Regular	798 (78.01)	3849(82.69)		334 (78.77)	1602 (81.57)

Notes: ^aThe dietary factors including the intake of fatty meat, fried food, and pickled food were categorized as occasional (<3 days per week) or regular (3–7 days per week). ^bModerate-to-intensity physical activity was classified as never or regular (at least 1 day per week).

Abbreviations: SD, standard deviation; BMI, body mass index.

Table 3 Univariate and Multivariate Predictors of Early Colorectal Neoplasm in the Training Set

	Unadjusted		Adjusted
	OR (95% CI)	P value	β Coefficient	OR (95% CI)	P value
Age (year)		0.00			0.00
50-	1		0.00	1
60-	1.32 (1.16, 1.49)	0.00	0.37	1.45 (1.25, 1.69)	0.00
70-	1.15 (0.93, 1.42)	0.21	0.31	1.37 (1.07, 1.75)	0.01
Sex
Male	1.84 (1.63, 2.08)	0.00	0.30	1.35 (1.09, 1.67)	0.01
Female	1		0.00	1
BMI (kg/m^2)		0.00			0.01
18.5 ≤ BMI < 24	1		0.00	1
BMI <18.5	0.97 (0.62, 1.52)	0.89	0.09	1.10 (0.69, 1.76)	0.68
24 ≤ BMI < 28	1.13 (0.98, 1.31)	0.10	0.12	1.13 (0.97, 1.31)	0.13
BMI ≥ 28	1.55 (1.23, 1.95)	0.00	0.41	1.51 (1.18, 1.92)	0.00
Family history					0.00
No	1		0.00	1
Yes	2.76 (2.12, 3.59)	0.00	1.02	2.78 (2.08, 3.71)
Smoking					0.02
No	1		0.00	1
Current/ever	1.77 (1.57, 2.00)	0.00	0.25	1.28 (1.04, 1.58)
Alcohol					0.03
No	1		0.00	1
Current/ever	1.78 (1.56, 2.03)	0.00	0.20	1.23 (1.02, 1.48)
Intake of fatty food^a					0.65
Occasional	1		0.00	1
Regular	1.15 (1.00, 1.33)	0.04	−0.04	0.96 (0.79, 1.15)
Intake of fried food^a					0.12
Occasional	1		0.00	1
Regular	1.30 (1.05, 1.63)	0.02	0.22	1.25 (0.94, 1.65)
Intake of pickled food^a					0.27
Occasional	1		0.00	1
Regular	1.19 (1.04, 1.37)	0.01	0.09	1.10 (0.93, 1.31)
Moderate-to-intensity physical activity^b					0.01
Never	1		0.00	1
Regular	0.74 (0.63, 0.88)	0.00	− 0.24	0.79 (0.66, 0.94)

Notes: ^aThe dietary factors including intake of fatty meat, fried food and pickled food were categorized as occasional (<3 days per week) or regular (3–7 days per week). ^bModerate-to-intensity physical activity was classified as never or regular (at least 1 day per week).

Abbreviations: BMI, body mass index; OR, odds ratio; 95% CI, 95% confidence interval..

Figure 2 ROC curves for early colorectal neoplasm risk predictions. The yellow solid line is reference line. The solid line with blue is for risk model with only age and family history of first-degree relatives with CRC. The green line is for risk model with age, family history of first-degree relatives with CRC and lifestyle risk factors. The horizontal line corresponds to 1-specificity. The vertical line corresponds to sensitivity.

Table 4 Sensitivity and Specificity of Early Colorectal Neoplasm Risk-Based Model at Different Predicted Risk Cutoff Values

Predicted Risk Cutoff Values (%)	Sensitivity (%)	Specificity (%)	Youden’s Index (%)
10	95.27	8.73	4.00
15	75.21	39.97	15.17
17	63.99	53.82	17.81
21	43.62	73.66	17.28
25	26.23	86.19	12.43
30	21.20	90.00	7.56
35	6.48	95.62	4.54
40	4.42	98.05	3.45

Table 5 Summary of Risk Prediction Models for CRC and ACN

Author (Year)	Country	Outcome	Risk Factors	Data Source (Sample Size)	AUC
Betes 2003^Citation29	Spain	ACN	Age, gender, BMI	Hospital-based (2210)	0.67
Cai 2012^Citation20	China	ACN+CRC	Age, gender, smoking, diabetes mellitus, green vegetables, pickled food, fried food, white meat	Hospital-based (7541)	0.74
Chen 2014^Citation31	China	ACN	Age, gender, history of coronary heart disease, egg intake, defaecation frequency	Hospital-based (905)	0.75
Li 2018^Citation27	China	CRC	Age, sex, education level, occupations, diarrhea, constipation, colon mucosa and bleeding, gallbladder disease, a stressful life event, family history and a qualitative FIT	Population based (891,199) considers the history of clinical symptoms and signs of bowel disease together with the FIT results	0.84
Kaminski 2014^Citation32	Poland	ACN	Age, gender, BMI, smoking, number and age affected of first degree relatives with CRC	Cross-sectional analysis (35,918)	0.64
Stegeman 2014^Citation33	Netherlands	ACN	Age, smoking, first degree relative with CRC, FIT, calcium intake	Population-based (6600)	0.76
Tao 2014^Citation22	Germany	ACN	Age, gender, smoking, first-degree relative with CRC, alcohol, previous polyp, red meat consumption, NSAIDS, previous colonoscopy	Population-based (7891)	0.67
Tao 2014^Citation22	Germany	CRC	Age, gender, smoking, first-degree relative with CRC, alcohol, previous polyp, red meat consumption, NSAIDS, previous colonoscopy	Population-based (7891)	0.71
Yeoh 2011^Citation30	Asia	ACN	Age, gender, smoking, first degree relative with CRC	Hospital-based (2752)	0.66
Driver 2007^Citation16	USA	CRC	Age, BMI, history of smoking, weekly or daily alcohol use	Population-based (21,581 male)	0.7
Ma 2008^Citation17	Japan	CRC	Age, BMI, smoking, physical activity, alcohol	Population-based (28,115 male)	0.7
Taylor 2011^Citation18	USA	CRC	Age, first, second and third degree relatives with CRC	Population-based (431,153)	0.67
Yarnall 2013^Citation19	UK	CRC	BMI, smoking, alcohol, fibre intake, red meat intake, physical activity, 14 SNPs	META	0.63
Auge 2014^Citation34	Spain	ACN	Age, sex and categorized FIT results	Retrospective (3109)	0.67
Cooper 2017^Citation25	UK	CRC+ ACN	Age, sex, first time invite, previous non-responder, previous responder and quantitative FIT result	Population based (40,930)	0.68
Aniwan 2015^Citation23	Thailand	CRC	Age, sex, family history and smoking status as APCS risk score and a qualitative FIT	Hospital-based (948)	0.85
Aniwan 2015^Citation23	Thailand	ACN	Age, sex, family history and smoking status as APCS risk score and a qualitative FIT	Hospital-based (948)	0.67
Dunlop 2013^Citation21		CRC	Family history of CRC and 10 common genetic variants	Population-based (42,103)	0.56
Hsu 2015^Citation24	USA and Germany	CRC	Family history of CRC and 27 common genetic variants, endoscopy history	Population-based (12,000)	0.59 for men; 0.56 for women
Ibáñez-Sanz 2017^Citation26	Spain	CRC	Family history of CRC,alcohol consumption, obesity, physical activity, red meat and vegetable consumption, and nonsteroidal anti-inflammatory drug use, 21 SNPs	Population-based (4080)	0.63
Joen 2018^Citation28	USA	CRC	Family history of CRC, 19 lifestyle and environmental factors, 63 CRC-associated SNPs	Population-based (20,338)	0.63 for men; 0.62 for women

Abbreviations: CRC, colorectal cancer; ACN, advanced colorectal neoplasm; FIT, faecal immunochemical test; NSAIDS, nonsteroidal anti-inflammatory drug use; BMI, body mass index; SNP, single nucleotide polymorphism; APCS, Asia-Pacific Colorectal Screening Scoring System.

Betés M, Muñoz-Navas MA, Duque JM, et al. Use of colonoscopy as a primary screening test for colorectal cancer in average risk people. Am J Gastroenterol. 2003;98(12):2648–2654. doi:10.1111/j.1572-0241.2003.08771.x14687811

 PubMed Web of Science ®Google Scholar

Cai QC, Yu ED, Xiao Y, et al. Derivation and validation of a prediction rule for estimating advanced colorectal neoplasm risk in average-risk Chinese. Am J Epidemiol. 2012;175(6):584–593. doi:10.1093/aje/kwr33722328705

 PubMed Web of Science ®Google Scholar

Chen G, Mao B, Pan Q, Liu Q, Xu X, Ning Y. Prediction rule for estimating advanced colorectal neoplasm risk in average-risk populations in southern Jiangsu Province. Chin J Cancer Res. 2014;26(1):4–11. doi:10.3978/j.issn.1000-9604.2014.02.0324653621

 PubMed Web of Science ®Google Scholar

Li W, Zhao LZ, Ma DW, et al. Predicting the risk for colorectal cancer with personal characteristics and fecal immunochemical test. Medicine (Baltimore). 2018;97:e0529. doi:10.1097/MD.000000000001052929718843

 PubMed Web of Science ®Google Scholar

Kaminski MF, Polkowski M, Kraszewska E, Rupinski M, Butruk E, Regula J. A score to estimate the likelihood of detecting advanced colorectal neoplasia at colonoscopy. Gut. 2014;63(7):1112–1119. doi:10.1136/gutjnl-2013-30496524385598

 PubMed Web of Science ®Google Scholar

Stegeman I, de Wijkerslooth TR, Stoop EM, et al. Combining risk factors with faecal immunochemical test outcome for selecting CRC screenees for colonoscopy. Gut. 2014;63(3):466–471. doi:10.1136/gutjnl-2013-30501323964098

 PubMed Web of Science ®Google Scholar

Tao S, Hoffmeister M, Brenner H. Development and validation of a scoring system to identify individuals at high risk for advanced colorectal neoplasms who should undergo colonoscopy screening. Clin Gastroenterol Hepatol. 2014;12(3):478–485. doi:10.1016/j.cgh.2013.08.042.24022090

 PubMed Web of Science ®Google Scholar

Yeoh KG, Ho KY, Chiu HM, et al. The Asia-Pacific colorectal screening score a validated tool that stratifies risk for colorectal advanced neoplasia in asymptomatic Asian subjects. Gut. 2011;60(9):1236–1241. doi:10.1136/gut.2010.22116821402615

 PubMed Web of Science ®Google Scholar

Driver JA, Gaziano JM, Gelber RP, et al. Development of a risk score for colorectal cancer in men. Am J Med. 2007;120(3):257–263. doi:10.1016/j.amjmed.2006.05.055.17349449

 PubMed Web of Science ®Google Scholar

Ma E, Sasazuki S, Iwasaki M, Sawada N, Inoue M, Tsugane S. Japan Public Health Center-based Prospective Study Group. 10-Year risk of colorectal cancer: development and validation of a prediction model in middle-aged Japanese men. Cancer Epidemiol. 2010;34(5):534–541. doi:10.1016/j.canep.2010.04.02120554262

 PubMed Web of Science ®Google Scholar

Taylor DP, Stoddard GJ, Burt RW, et al. How well does family history predict who will get colorectal cancer? Implications for cancer screening and counseling. Genet Med. 2011;13(5):385–391. doi:10.1097/GIM.0b013e318206438421270638

 PubMed Web of Science ®Google Scholar

Yarnall JM, Crouch DJM, Lewis CM. Incorporating non-genetic risk factors and behavioural modifications into risk prediction models for colorectal cancer. Cancer Epidemiol. 2013;37:324–329. doi:10.1016/j.canep.2012.12.00823375517

 PubMed Web of Science ®Google Scholar

Auge JM, Pellise M, Escudero JM, et al. Risk stratification for advanced colorectal neoplasia according to fecal hemoglobin concentration in a colorectal cancer screening program. Gastroenterology. 2014;147(3):628–636.e1. doi:10.1053/j.gastro.2014.06.00824937264

 PubMed Web of Science ®Google Scholar

Cooper JA, Parsons N, Stinton C, et al. Risk-adjusted colorectal cancer screening using the FIT and routine screening data: development of a risk prediction model. Br J Cancer. 2018;118(2):285–293. doi:10.1038/bjc.2017.37529096402

 PubMed Web of Science ®Google Scholar

Aniwan S, Rerknimitr R, Kongkam P, et al. A combination of clinical risk stratification and fecal immunochemical test results to prioritize colonoscopy screening in asymptomatic participants. Gastrointest Endosc. 2015;81(3):719–727. doi:10.1016/j.gie.2014.11.03525708760

 PubMed Web of Science ®Google Scholar

Dunlop MG, Tenesa A, Farrington SM, et al. Cumulative impact of common genetic variants and other risk factors on colorectal cancer risk in 42,103 individuals. Gut. 2013;62(6):871–881. doi:10.1136/gutjnl-2011-30053722490517

 PubMed Web of Science ®Google Scholar

Hsu L, Jeon J, Brenner H, et al. A model to determine colorectal cancer risk using common genetic susceptibility loci. Gastroenterology. 2015;148(7):1330–1339.e14. doi:10.1053/j.gastro.2015.02.01025683114

 PubMed Web of Science ®Google Scholar

Ibáñez -sanz G, Diez-Villanueva A, Alonso MH, et al. Risk model for colorectal cancer in Spanish population using environmental and genetic factors: results from the MCC-Spain study. Sci Rep. 2017;7:43263. doi:10.1038/srep43263.28233817

 PubMed Web of Science ®Google Scholar

Jeon J, Du MM, Schoen RE, et al. Determining risk of colorectal cancer and starting age of screening based on lifestyle, environmental, and genetic factors. Gastroenterology. 2018;154(8):2152–2164. doi:10.1053/j.gastro.2018.02.02129458155

 PubMed Web of Science ®Google Scholar