Proteomic Analyses in Diverse Populations Improved Risk Prediction and Identified New Drug Targets for Type 2 Diabetes

OBJECTIVE

Integrated analyses of plasma proteomics and genetic data in prospective studies can help assess the causal relevance of proteins, improve risk prediction, and discover novel protein drug targets for type 2 diabetes (T2D).

RESEARCH DESIGN AND METHODS

We measured plasma levels of 2,923 proteins using Olink Explore among ∼2,000 randomly selected participants from China Kadoorie Biobank (CKB) without prior diabetes at baseline. Cox regression assessed associations of individual protein with incident T2D (n = 92 cases). Proteomic-based risk models were developed with discrimination, calibration, reclassification assessed using area under the curve (AUC), calibration plots, and net reclassification index (NRI), respectively. Two-sample Mendelian randomization (MR) analyses using cis-protein quantitative trait loci identified in a genome-wide association study of CKB and UK Biobank for specific proteins were conducted to assess their causal relevance for T2D, along with colocalization analyses to examine shared causal variants between proteins and T2D.

RESULTS

Overall, 33 proteins were significantly associated (false discovery rate < 0.05) with risk of incident T2D, including IGFBP1, GHR, and amylase. The addition of these 33 proteins to a conventional risk prediction model improved AUC from 0.77 (0.73-0.82) to 0.88 (0.85-0.91) and NRI by 38%, with predicted risks well calibrated with observed risks. MR analyses provided support for the causal relevance for T2D of ENTR1, LPL, and PON3, with replication of ENTR1 and LPL in Europeans using different genetic instruments. Moreover, colocalization analyses showed strong evidence (pH4 > 0.6) of shared genetic variants of LPL and PON3 with T2D.

CONCLUSIONS

Proteomic analyses in Chinese adults identified novel associations of multiple proteins with T2D with strong genetic evidence supporting their causal relevance and potential as novel drug targets for prevention and treatment of T2D.

Conventional and genetic associations of adiposity with 1463 proteins in relatively lean Chinese adults

Adiposity is associated with multiple diseases and traits, but little is known about the causal relevance and mechanisms underlying these associations. Large-scale proteomic profiling, especially when integrated with genetic data, can clarify mechanisms linking adiposity with disease outcomes. We examined the associations of adiposity with plasma levels of 1463 proteins in 3977 Chinese adults, using measured and genetically-instrumented BMI. We further used two-sample bi-directional MR analyses to assess if certain proteins influenced adiposity, along with other (e.g. enrichment) analyses to clarify possible mechanisms underlying the observed associations. Overall, the mean (SD) baseline BMI was 23.9 (3.3) kg/m2, with only 6% being obese (i.e. BMI ≥ 30 kg/m2). Measured and genetically-instrumented BMI was significantly associated at FDR < 0.05 with levels of 1096 (positive/inverse: 826/270) and 307 (positive/inverse: 270/37) proteins, respectively, with FABP4, LEP, IL1RN, LSP1, GOLM2, TNFRSF6B, and ADAMTS15 showing the strongest positive and PON3, NCAN, LEPR, IGFBP2 and MOG showing the strongest inverse genetic associations. These associations were largely linear, in adiposity-to-protein direction, and replicated (> 90%) in Europeans of UKB (mean BMI 27.4 kg/m2). Enrichment analyses of the top > 50 BMI-associated proteins demonstrated their involvement in atherosclerosis, lipid metabolism, tumour progression and inflammation. Two-sample bi-directional MR analyses using cis-pQTLs identified in CKB GWAS found eight proteins (ITIH3, LRP11, SCAMP3, NUDT5, OGN, EFEMP1, TXNDC15, PRDX6) significantly affect levels of BMI, with NUDT5 also showing bi-directional association. The findings among relatively lean Chinese adults identified novel pathways by which adiposity may increase disease risks and novel potential targets for treatment of obesity and obesity-related diseases.

Adiposity and risks of vascular and non-vascular mortality among Chinese adults with type 2 diabetes: a 10-year prospective study

INTRODUCTION

Among individuals with diabetes, high adiposity has been associated with lower cardiovascular disease (CVD) mortality (the so-called 'obesity paradox' phenomenon) in Western populations, for reasons that are still not fully elucidated. Moreover, little is known about such phenomena in Chinese adults with diabetes among whom very few were obese. We aimed to assess the associations of adiposity with vascular and non-vascular mortality among individuals with diabetes, and compare these with associations among individuals without diabetes.

RESEARCH DESIGN AND METHODS

In 2004-2008, the prospective China Kadoorie Biobank recruited >512 000 adults from 10 areas in China. After ~10 years of follow-up, 3509 deaths (1431 from CVD) were recorded among 23 842 individuals with diabetes but without prior major diseases at baseline. Cox regression yielded adjusted HRs associating adiposity with mortality.

RESULTS

Among people with diabetes, body mass index (BMI) (mean 25.0 kg/m²) was positively log linearly associated with CVD incidence (n=9943; HR=1.19 (95% CI 1.15 to 1.22) per 5 kg/m²), but showed U-shaped associations with CVD and overall mortality, with lowest risk at 22.5-24.9 kg/m². At lower BMI, risk of death (n=671) within 28 days of CVD onset was particularly elevated, with an HR of 3.26 (95% CI 2.29 to 4.65) at <18.5 kg/m² relative to 22.5-24.9 kg/m², but no higher mortality risk at BMI ≥25.0 kg/m². These associations were similar in self-reported and screen-detected diabetes, and persisted after extensive attempts to address reverse causality and confounding. Among individuals without diabetes (mean BMI 23.6 kg/m²; n=23 305 deaths), there were less extreme excess mortality risks at low BMI.

CONCLUSIONS

Among relatively lean Chinese adults with diabetes, there were contrasting associations of adiposity with CVD incidence and with mortality. The high mortality risk at low and high BMI levels highlights, if causal, the importance of maintaining normal weight among people with diabetes.

Adiposity and risk of ischaemic and haemorrhagic stroke in 0·5 million Chinese men and women: a prospective cohort study

BACKGROUND

China has high stroke rates despite the population being relatively lean. Uncertainty persists about the relevance of adiposity to risk of stroke types. We aimed to assess the associations of adiposity with incidence of stroke types and effect mediation by blood pressure in Chinese men and women.

METHODS

The China Kadoorie Biobank enrolled 512 891 adults aged 30-79 years from ten areas (five urban and five rural) during 2004-08. During a median 9 years (IQR 8-10) of follow-up, 32 448 strokes (about 90% confirmed by neuroimaging) were recorded among 489 301 participants without previous cardiovascular disease. Cox regression analysis was used to produce adjusted hazard ratios (HRs) for ischaemic stroke (n=25 210) and intracerebral haemorrhage (n=5380) associated with adiposity.

FINDINGS

Mean baseline body-mass index (BMI) was 23·6 kg/m² (SD 3·2), and 331 723 (67·8%) participants had a BMI of less than 25 kg/m². Throughout the range examined (mean 17·1 kg/m² [SD 0·9] to 31·7 kg/m² [2·0]), each 5 kg/m² higher BMI was associated with 8·3 mm Hg (SE 0·04) higher systolic blood pressure. BMI was positively associated with ischaemic stroke, with an HR of 1·30 (95% CI 1·28-1·33 per 5 kg/m² higher BMI), which was generally consistent with that predicted by equivalent differences in systolic blood pressure (1·25 [1·24-1·26]). The HR for intracerebral haemorrhage (1·11 [1·07-1·16] per 5 kg/m² higher BMI) was less extreme, and much weaker than that predicted by the corresponding difference in systolic blood pressure (1·48 [1·46-1·50]). Other adiposity measures showed similar associations with stroke types. After adjustment for usual systolic blood pressure, the positive associations with ischaemic stroke were attenuated (1·05 [1·03-1·07] per 5 kg/m² higher BMI); for intracerebral haemorrhage, they were reversed (0·73 [0·70-0·77]). High adiposity (BMI >23 kg/m²) accounted for 14·7% of total stroke (16·5% of ischaemic stroke and 6·7% of intracerebral haemorrhage).

INTERPRETATION

In Chinese adults, adiposity was strongly positively associated with ischaemic stroke, chiefly through its effect on blood pressure. For intracerebral haemorrhage, leanness, either per se or through some other factor (or factors), might increase risk, offsetting the protective effects of lower blood pressure.

FUNDING

UK Wellcome Trust, UK Medical Research Council, British Heart Foundation, Cancer Research UK, Kadoorie Charitable Foundation, Chinese Ministry of Science and Technology, Chinese National Natural Science Foundation.

Associations of General and Central Adiposity With Incident Diabetes in Chinese Men and Women

OBJECTIVE

We assess associations of general and central adiposity in middle age and of young adulthood adiposity with incident diabetes in adult Chinese and estimate the associated population burden of diabetes.

RESEARCH DESIGN AND METHODS

The prospective China Kadoorie Biobank enrolled 512,891 adults 30-79 years of age from 10 localities across China during 2004-2008. During 9.2 years of follow-up, 13,416 cases of diabetes were recorded among 482,589 participants without diabetes at baseline. Cox regression yielded adjusted hazard ratios (HRs) for incident diabetes associated with measures of general (e.g., BMI and BMI at 25 years) and central (e.g., waist circumference [WC]) adiposity.

RESULTS

The mean (SD) BMI was 23.6 kg/m² (3.4 kg/m²), and 3.8% had a BMI ≥30 kg/m². Throughout the range examined (19-32 kg/m²), BMI showed a positive log-linear relationship with diabetes, with adjusted HRs per SD higher usual BMI greater in men (1.98; 95% CI 1.93-2.04) than in women (1.77; 1.73-1.81) (P for heterogeneity <0.001). For WC, HRs per SD were 2.13 (95% CI 2.07-2.19) in men and 1.91 (1.87-1.95) in women (P for heterogeneity <0.001). Mutual adjustment attenuated these associations, especially those of BMI. BMI at age 25 years was weakly positively associated with diabetes (men HR 1.09 [95% CI 1.05-1.12]; women 1.04 [1.02-1.07] per SD), which was reversed after adjustment for baseline BMI. In China, the increase in adiposity accounted for ∼50% of the increase in diabetes burden since 1980.

CONCLUSIONS

Among relatively lean Chinese adults, higher adiposity-general and central-was strongly positively associated with the risk of incident diabetes. The predicted continuing increase in adiposity in China foreshadows escalating rates of diabetes.

Smoking, alcohol, and diet in relation to risk of pancreatic cancer in China: a prospective study of 0.5 million people

In China, the incidence of pancreatic cancer (PC) has increased in recent decades. However, little is known about the relevance to PC risk of lifestyle and behavioral factors such as smoking, alcohol drinking, and diet. The China Kadoorie Biobank prospective study recruited 512,891 adults (210,222 men, 302,669 women) aged 30–79 (mean 52) years from 10 diverse areas during 2004–08. During ~9 years of follow‐up, 688 incident cases of PC were recorded among those who had no prior history of cancer at baseline. Cox regression yielded adjusted hazard ratios (HR) for PC associated with smoking, alcohol and selected dietary factors. Overall, 74% of men were ever‐regular smokers and 33% of men drank at least weekly, compared with only 3% and 2% of women, respectively. Among men, current regular smoking was associated with an adjusted HR of 1.25 (95% CI 1.08–1.44) for PC, with greater excess risk in urban than rural areas (1.46 [1.19–1.79] vs 1.04 [0.86–1.26]). Heavy, but not light to moderate, alcohol drinking (i.e. ≥420 g/week) was associated with significant excess risk (1.69 [1.21–2.37]), again more extreme in urban than rural areas (1.93 [1.29–2.87] vs 1.35 [0.74–2.48]). Overall, regular consumption of certain foodstuffs was associated with PC risk, with adjusted daily vs never/rare consumption HRs of 0.66 (0.56–0.79) for fresh fruit and 1.16 (1.01–1.33) for red meat. In China, smoking and heavy alcohol drinking were independent risk factors for PC in men. Lower fresh fruit and higher red meat consumption were also associated with higher risk of PC.

Young adulthood and adulthood adiposity in relation to incidence of pancreatic cancer: a prospective study of 0.5 million Chinese adults and a meta-analysis

BACKGROUND

Adult adiposity is positively associated with pancreatic cancer in Western populations. Little is known, however, about the association in China where many have lower body mass index (BMI) or about the relevance of young adulthood adiposity for pancreatic cancer in both Western and East Asian populations.

METHODS

The China Kadoorie Biobank (CKB) recruited 512 891 adults aged 30-79 years during 2004-2008, recording 595 incident cases of pancreatic cancer during 8-year follow-up. Cox regression yielded adjusted HRs for pancreatic cancer associated with self-reported young adulthood (mean ~25 years) BMI and with measured adulthood (mean ~52 years) BMI and other adiposity measures (eg, waist circumference (WC)). These were further meta-analysed with published prospective studies.

RESULTS

Overall, the mean BMI (SD) was 21.9 (2.6) at age 25 years and 23.7 (3.3) kg/m² at age 52 years. Young adulthood BMI was strongly positively associated with pancreatic cancer in CKB (adjusted HR=1.36, 95% CI 1.16 to 1.61, per 5 kg/m² higher BMI) and in meta-analysis of CKB and four other studies (1.18, 1.12 to 1.24). In CKB, there was also a positive association of pancreatic cancer with adulthood BMI (1.11, 0.97 to 1.27, per 5 kg/m²), similar in magnitude to that in meta-analyses of East Asian studies using measured BMI (n=2; 1.08, 0.99 to 1.19) and of Western studies (n=25; 1.10, 1.06 to 1.12). Likewise, meta-analysis of four studies, including CKB, showed a positive association of adulthood WC with pancreatic cancer (1.10, 1.06 to 1.14, per 10 cm).

CONCLUSIONS

In both East Asian and Western populations, adiposity was positively associated with risk of pancreatic cancer, with a somewhat stronger association for young than late-life adiposity.

Diabetes, plasma glucose and incidence of pancreatic cancer: A prospective study of 0.5 million Chinese adults and a meta-analysis of 22 cohort studies

Diabetes is associated with an increased risk of pancreatic cancer (PC) in Western populations. Uncertainty remains, however, about the relevance of plasma glucose for PC among people without diabetes and about the associations of diabetes and high blood glucose with PC in China where the increase in diabetes prevalence has been very recent. The prospective China Kadoorie Biobank (CKB) study recruited 512,000 adults aged 30‐79 years from 10 diverse areas of China during 2004‐2008, recording 595 PC cases during 8 years of follow‐up. Cox regression yielded adjusted hazard ratios (HRs) for PC associated with diabetes (previously diagnosed or screen‐detected) and, among those without previously diagnosed diabetes, with levels of random plasma glucose (RPG). These were further meta‐analysed with 22 published prospective studies. Overall 5.8% of CKB participants had diabetes at baseline. Diabetes was associated with almost twofold increased risk of PC (adjusted HR = 1.87, 95% CI 1.48‐2.37), with excess risk higher in those with longer duration since diagnosis (p for trend = 0.01). Among those without previously diagnosed diabetes, each 1 mmol/L higher usual RPG was associated with a HR of 1.12 (1.04‐1.21). In meta‐analysis of CKB and 22 other studies, previously diagnosed diabetes was associated with a 52% excess risk (1.52, 1.43‐1.63). Among those without diabetes, each 1 mmol/L higher blood glucose was associated with a 15% (1.15, 1.09‐1.21) excess risk. In Chinese and non‐Chinese populations, diabetes and higher blood glucose levels among those without diabetes are associated with an increased risk of PC.

What's new?

Diabetes is associated with increased risk of pancreatic cancer (PC) in Western populations, but uncertainty remains about the relevance of plasma glucose among people without diabetes. In this first prospective study investigating the association of diabetes with PC risk in China, where the increase in diabetes prevalence has been very recent, diabetes was associated with an almost twofold increased risk of PC, with random plasma glucose being positively associated with PC risk among participants without prior diagnosis of diabetes. In meta‐analysis, the results were consistent with previous evidence in Western populations, highlighting diabetes as a potential aetiological factor of PC.

Association Between Diabetes and Cause-Specific Mortality in Rural and Urban Areas of China

IMPORTANCE

In China, diabetes prevalence has increased substantially in recent decades, but there are no reliable estimates of the excess mortality currently associated with diabetes.

OBJECTIVES

To assess the proportional excess mortality associated with diabetes and estimate the diabetes-related absolute excess mortality in rural and urban areas of China.

DESIGN, SETTING, AND PARTICIPANTS

A 7-year nationwide prospective study of 512 869 adults aged 30 to 79 years from 10 (5 rural and 5 urban) regions in China, who were recruited between June 2004 and July 2008 and were followed up until January 2014.

EXPOSURES

Diabetes (previously diagnosed or detected by screening) recorded at baseline.

MAIN OUTCOMES AND MEASURES

All-cause and cause-specific mortality, collected through established death registries. Cox regression was used to estimate adjusted mortality rate ratio (RR) comparing individuals with diabetes vs those without diabetes at baseline.

RESULTS

Among the 512 869 participants, the mean (SD) age was 51.5 (10.7) years, 59% (n = 302 618) were women, and 5.9% (n = 30 280) had diabetes (4.1% in rural areas, 8.1% in urban areas, 5.8% of men, 6.1% of women, 3.1% had been previously diagnosed, and 2.8% were detected by screening). During 3.64 million person-years of follow-up, there were 24 909 deaths, including 3384 among individuals with diabetes. Compared with adults without diabetes, individuals with diabetes had a significantly increased risk of all-cause mortality (1373 vs 646 deaths per 100 000; adjusted RR, 2.00 [95% CI, 1.93-2.08]), which was higher in rural areas than in urban areas (rural RR, 2.17 [95% CI, 2.07-2.29]; urban RR, 1.83 [95% CI, 1.73-1.94]). Presence of diabetes was associated with increased mortality from ischemic heart disease (3287 deaths; RR, 2.40 [95% CI, 2.19-2.63]), stroke (4444 deaths; RR, 1.98 [95% CI, 1.81-2.17]), chronic liver disease (481 deaths; RR, 2.32 [95% CI, 1.76-3.06]), infections (425 deaths; RR, 2.29 [95% CI, 1.76-2.99]), and cancer of the liver (1325 deaths; RR, 1.54 [95% CI, 1.28-1.86]), pancreas (357 deaths; RR, 1.84 [95% CI, 1.35-2.51]), female breast (217 deaths; RR, 1.84 [95% CI, 1.24-2.74]), and female reproductive system (210 deaths; RR, 1.81 [95% CI, 1.20-2.74]). For chronic kidney disease (365 deaths), the RR was higher in rural areas (18.69 [95% CI, 14.22-24.57]) than in urban areas (6.83 [95% CI, 4.73-9.88]). Among those with diabetes, 10% of all deaths (16% rural; 4% urban) were due to definite or probable diabetic ketoacidosis or coma (408 deaths).

CONCLUSIONS AND RELEVANCE

Among adults in China, diabetes was associated with increased mortality from a range of cardiovascular and noncardiovascular diseases. Although diabetes was more common in urban areas, it was associated with greater excess mortality in rural areas.

Evaluation of type 2 diabetes genetic risk variants in Chinese adults: findings from 93,000 individuals from the China Kadoorie Biobank

AIMS/HYPOTHESIS

Genome-wide association studies (GWAS) have discovered many risk variants for type 2 diabetes. However, estimates of the contributions of risk variants to type 2 diabetes predisposition are often based on highly selected case-control samples, and reliable estimates of population-level effect sizes are missing, especially in non-European populations.

METHODS

The individual and cumulative effects of 59 established type 2 diabetes risk loci were measured in a population-based China Kadoorie Biobank (CKB) study of 93,000 Chinese adults, including >7,100 diabetes cases.

RESULTS

Association signals were directionally consistent between CKB and the original discovery GWAS: of 56 variants passing quality control, 48 showed the same direction of effect (binomial test, p = 2.3 × 10(-8)). We observed a consistent overall trend towards lower risk variant effect sizes in CKB than in case-control samples of GWAS meta-analyses (mean 19-22% decrease in log odds, p ≤ 0.0048), likely to reflect correction of both 'winner's curse' and spectrum bias effects. The association with risk of diabetes of a genetic risk score, based on lead variants at 25 loci considered to act through beta cell function, demonstrated significant interactions with several measures of adiposity (BMI, waist circumference [WC], WHR and percentage body fat [PBF]; all p interaction < 1 × 10(-4)), with a greater effect being observed in leaner adults.

CONCLUSIONS/INTERPRETATION

Our study provides further evidence of shared genetic architecture for type 2 diabetes between Europeans and East Asians. It also indicates that even very large GWAS meta-analyses may be vulnerable to substantial inflation of effect size estimates, compared with those observed in large-scale population-based cohort studies.

ACCESS TO RESEARCH MATERIALS

Details of how to access China Kadoorie Biobank data and details of the data release schedule are available from www.ckbiobank.org/site/Data+Access .

Contrasting male and female trends in tobacco-attributed mortality in China: evidence from successive nationwide prospective cohort studies

BACKGROUND

Chinese men now smoke more than a third of the world's cigarettes, following a large increase in urban then rural usage. Conversely, Chinese women now smoke far less than in previous generations. We assess the oppositely changing effects of tobacco on male and female mortality.

METHODS

Two nationwide prospective studies 15 years apart recruited 220,000 men in about 1991 at ages 40-79 years (first study) and 210,000 men and 300,000 women in about 2006 at ages 35-74 years (second study), with follow-up during 1991-99 (mid-year 1995) and 2006-14 (mid-year 2010), respectively. Cox regression yielded sex-specific adjusted mortality rate ratios (RRs) comparing smokers (including any who had stopped because of illness, but not the other ex-smokers, who are described as having stopped by choice) versus never-smokers.

FINDINGS

Two-thirds of the men smoked; there was little dependence of male smoking prevalence on age, but many smokers had not smoked cigarettes throughout adult life. Comparing men born before and since 1950, in the older generation, the age at which smoking had started was later and, particularly in rural areas, lifelong exclusive cigarette use was less common than in the younger generation. Comparing male mortality RRs in the first study (mid-year 1995) versus those in the second study (mid-year 2010), the proportional excess risk among smokers (RR-1) approximately doubled over this 15-year period (urban: RR 1·32 [95% CI 1·24-1·41] vs 1·65 [1·53-1·79]; rural: RR 1·13 [1·09-1·17] vs 1·22 [1·16-1·29]), as did the smoking-attributed fraction of deaths at ages 40-79 years (urban: 17% vs 26%; rural: 9% vs 14%). In the second study, urban male smokers who had started before age 20 years (which is now typical among both urban and rural young men) had twice the never-smoker mortality rate (RR 1·98, 1·79-2·19, approaching Western RRs), with substantial excess mortality from chronic obstructive pulmonary disease (COPD RR 9·09, 5·11-16·15), lung cancer (RR 3·78, 2·78-5·14), and ischaemic stroke or ischaemic heart disease (combined RR 2·03, 1·66-2·47). Ex-smokers who had stopped by choice (only 3% of ever-smokers in 1991, but 9% in 2006) had little smoking-attributed risk more than 10 years after stopping. Among Chinese women, however, there has been a tenfold intergenerational reduction in smoking uptake rates. In the second study, among women born in the 1930s, 1940s, 1950s, and since 1960 the proportions who had smoked were, respectively, 10%, 5%, 2%, and 1% (3097/30,943, 3265/62,246, 2339/97,344, and 1068/111,933). The smoker versus non-smoker RR of 1·51 (1·40-1·63) for all female mortality at ages 40-79 years accounted for 5%, 3%, 1%, and <1%, respectively, of all the female deaths in these four successive birth cohorts. In 2010, smoking caused about 1 million (840,000 male, 130,000 female) deaths in China.

INTERPRETATION

Smoking will cause about 20% of all adult male deaths in China during the 2010s. The tobacco-attributed proportion is increasing in men, but low, and decreasing, in women. Although overall adult mortality rates are falling, as the adult population of China grows and the proportion of male deaths due to smoking increases, the annual number of deaths in China that are caused by tobacco will rise from about 1 million in 2010 to 2 million in 2030 and 3 million in 2050, unless there is widespread cessation.

FUNDING

Wellcome Trust, MRC, BHF, CR-UK, Kadoorie Charitable Foundation, Chinese MoST and NSFC.

Emerging tobacco-related cancer risks in China: A nationwide, prospective study of 0.5 million adults

BACKGROUND

In China, cigarette consumption has increased substantially since the 1980s, almost exclusively in men. This study was aimed at assessing the emerging cancer risks.

METHODS

A nationwide, prospective study recruited 210,259 men and 302,632 women aged 30 to 79 years from 10 areas of China from 2004 to 2008; approximately 18,000 incident cancers were recorded during 7 years of follow-up. Cox regression yielded adjusted risk ratios (RRs) comparing smokers (including those who had stopped because of illness but not those who had stopped by choice) with never-smokers.

RESULTS

Among men, 68% were smokers; their overall cancer risk was significantly increased (RR, 1.44; 95% confidence interval [CI], 1.37-1.53), and it was greater in urban (RR, 1.55; 95% CI, 1.41-1.70) than in rural areas (RR, 1.39; 95% CI, 1.30-1.49). This excess accounted for 23% of all cancers between the ages of 40 and 79 years, with significantly elevated risks of lung cancer (RR, 2.51; 95% CI, 2.18-2.90), liver cancer (RR, 1.32; 95% CI, 1.12-1.54), stomach cancer (RR, 1.34; 95% CI, 1.16-1.55), esophageal cancer (RR, 1.47; 95% CI, 1.24-1.73), and an aggregate of 5 other minor sites (RR, 1.52; 95% CI, 1.25-1.86). For lung cancer, the RRs were much greater for nonadenocarcinoma (RR, 5.83; 95% CI, 5.02-6.77) than for adenocarcinoma (RR, 1.78; 95% CI, 1.36-2.34). Among exsmokers (6.7%) who had stopped by choice, there was little excess cancer risk approximately 15 years after quitting. Among the few female smokers (3%), the overall cancer risk was also significantly increased (RR, 1.42; 95% CI, 1.28-1.57). Smoking was estimated to cause approximately 435,000 new cancers per year in China (approximately 360,000 in men and approximately 75,000 in women).

CONCLUSIONS

In China, smoking now causes a quarter of all adult male cancers. High male uptake rates before the age of 20 years and nearly universal use of cigarettes foreshadow substantial tobacco-attributed risks in China unless there is widespread cessation.