Logo of jnciLink to Publisher's site
J Natl Cancer Inst. Jul 7, 2010; 102(13): 959–971.
PMCID: PMC2897877

Replication of Lung Cancer Susceptibility Loci at Chromosomes 15q25, 5p15, and 6p21: A Pooled Analysis From the International Lung Cancer Consortium



Genome-wide association studies have identified three chromosomal regions at 15q25, 5p15, and 6p21 as being associated with the risk of lung cancer. To confirm these associations in independent studies and investigate heterogeneity of these associations within specific subgroups, we conducted a coordinated genotyping study within the International Lung Cancer Consortium based on independent studies that were not included in previous genome-wide association studies.


Genotype data for single-nucleotide polymorphisms at chromosomes 15q25 (rs16969968, rs8034191), 5p15 (rs2736100, rs402710), and 6p21 (rs2256543, rs4324798) from 21 case–control studies for 11 645 lung cancer case patients and 14 954 control subjects, of whom 85% were white and 15% were Asian, were pooled. Associations between the variants and the risk of lung cancer were estimated by logistic regression models. All statistical tests were two-sided.


Associations between 15q25 and the risk of lung cancer were replicated in white ever-smokers (rs16969968: odds ratio [OR] = 1.26, 95% confidence interval [CI] = 1.21 to 1.32, Ptrend = 2 × 10−26), and this association was stronger for those diagnosed at younger ages. There was no association in never-smokers or in Asians between either of the 15q25 variants and the risk of lung cancer. For the chromosome 5p15 region, we confirmed statistically significant associations in whites for both rs2736100 (OR = 1.15, 95% CI = 1.10 to 1.20, Ptrend = 1 × 10−10) and rs402710 (OR = 1.14, 95% CI = 1.09 to 1.19, Ptrend = 5 × 10−8) and identified similar associations in Asians (rs2736100: OR = 1.23, 95% CI = 1.12 to 1.35, Ptrend = 2 × 10−5; rs402710: OR = 1.15, 95% CI = 1.04 to 1.27, Ptrend = .007). The associations between the 5p15 variants and lung cancer differed by histology; odds ratios for rs2736100 were highest in adenocarcinoma and for rs402710 were highest in adenocarcinoma and squamous cell carcinomas. This pattern was observed in both ethnic groups. Neither of the two variants on chromosome 6p21 was associated with the risk of lung cancer.


In this international genetic association study of lung cancer, previous associations found in white populations were replicated and new associations were identified in Asian populations. Future genetic studies of lung cancer should include detailed stratification by histology.


Prior knowledge

Genome-wide association studies that were conducted in white populations have identified three chromosomal regions at 15q25, 5p15, and 6p21 as being associated with the risk of lung cancer. Whether genetic variants at these regions are associated with risk of lung cancer in other populations is unclear.

Study design

A coordinated genotyping study of six single-nucleotide polymorphisms at these three chromosomal regions using data from 21 independent case–control studies that included Asians studies and were not included in previous genome-wide association studies.


The 15p25 locus–risk of lung cancer association in whites was replicated, but there was no association between this locus and the risk of lung cancer in white lifetime never-smokers. There was no association between 15q25 and the risk of lung cancer in Asians. The chromosome 5p15 locus–risk of lung cancer association was replicated in whites, and a similar association was found in Asians. In both whites and Asians, the two variants in 5p15 were more strongly associated with adenocarcinoma than with other histology groups. Chromosome 6p21 was not associated with the risk of lung cancer.


Future genetic studies of lung cancer should include detailed stratification by histology.


Some of the variants at chromosome 15q25 had low minor allele frequencies in Asians. Replication of variants in Asians that were originally identified in studies of whites may not be relevant. The fact that different studies with different genotyping protocols were included could have led to heterogeneity.

From the Editors

Replication of initial genome-wide association findings is considered a gold standard for reporting genotype–phenotype associations. Three human genomic regions at chromosomes 15q25, 5p15, and 6p21 that were found to be associated with susceptibility to lung cancer in genome-wide association studies merit such replication.

The region at 15q24–25.1, which contains three nicotinic acetylcholine receptor subunit genes (CHRNA5, CHRNA3, and CHRNB4), was associated with the risk of lung cancer in three independently conducted genome-wide association studies that gave remarkably consistent results for associations between three single-nucleotide polymorphisms (SNPs) at this locus and the risk of lung cancer: The MD Anderson Cancer Center study reported an odds ratio (OR) of 1.32 (P = 10−17) for rs8034191 (1), the International Agency for Research on Cancer (IARC) study (2) reported an odds ratio of 1.30 (P = 10−20) for rs8034191 and rs16969968, and the deCODE study reported an odds ratio of 1.31 (P = 10−8) for rs1051730 (3). All of these three SNPs (rs8034191, rs16969968, and rs1051730) are in strong linkage disequilibrium.

Subsequent meta-analyses identified another putative causative region at 5p15.33 (4,5). This region contains two genes: the human telomerase reverse transcriptase gene (TERT) and cleft lip and palate transmembrane 1–like gene (CLPTM1L). The IARC (4) and the Institute of Cancer Research (ICR) and MD Anderson groups (5) reported that two different SNPs at 5p15.33 (rs402710 and rs401681, respectively), which are in strong linkage disequilibrium (D′ = 1.00, r2 = .66), are associated with the risk of lung cancer (IARC study: P = 2 × 10−7; ICR and MD Anderson groups: P = 8 × 10−9). The IARC group also identified a second SNP, rs2736100, that was associated with the risk of lung cancer (P = 4 × 10−6). A report from the deCODE group (6) provided evidence that the 5p15.33 region may be a susceptibility locus for multiple cancer types and also reported associations between risk of lung cancer and two potential susceptibility alleles.

The third region that has been implicated by genome-wide association studies in susceptibility to lung cancer is the HLA region at chromosome 6p21. Hung et al. (2) presented evidence for an association between the SNP rs4324798 at 6p21 and the risk of lung cancer (P = 4 × 10−7). Wang et al. (5) identified two other SNPs that were statistically significantly associated with risk of lung cancer and that mapped to this region: rs3117582 (P = 5 × 10−10) and rs9295740 (P = 4 × 10−7).

We aimed to replicate these findings in a large sample size dataset because there is still no consensus about the relative impact with respect to risk of lung cancer of the chromosome 15q25 locus on smoking behavior vs a direct lung carcinogenic effect. In addition, the newly identified susceptibility loci on 5p15 and 6p21 require further investigation in a larger sample size and in different ethnic groups. It is also important to evaluate effect modification by sex, age at cancer diagnosis, and family history, as well as by histological classification.

The International Lung Cancer Consortium (ILCCO) was established in 2004 with the aim of sharing comparable data from ongoing case–control and cohort studies of lung cancer. The overall objectives of the consortium are to share the data to increase statistical power, especially for subgroup analyses, reduce duplication of research efforts, replicate novel findings, and realize substantial cost savings through large collaborative efforts. Details of how the consortium was established have been published previously (7). With the aim of replicating association findings concerning these variants and the risk of lung cancer with sufficient statistical power for analysis of specific subgroups, we invited the principle investigators of all case–control studies from ILCCO to conduct genotyping of their lung cancer case patients and control subjects of European and Asian ancestry for two variants at the 15q25 locus (rs8034191 and rs16969968), two variants at 5p15 (rs402710 and rs2736100), and two variants at 6p21 (rs4324798 and rs2256543, the latter of which was the second most statistically significant SNP on chromosome 6 from the IARC genome-wide association study). For studies that were conducted in Asian populations, we selected three additional variants in the 15q25 region for genotyping (rs12914385, rs1317286, and rs931794) and the variants in 6p21 were not genotyped because of their low prevalence in these populations (according to the HapMap genome browser, www.hapmap.org).

Materials and Methods

Study Population

Twenty-one of the 52 case–control studies from the ILCCO participated in this pooled analysis. Of these studies, nine were conducted in North America, eight in Europe, and four in Asia. The study designs are briefly outlined in Table 1, and some of them have been described in more detail previously (6,819). The studies are referred to here either by the study location or the name of the coordinating institution.

Table 1
Summary of the participating studies from the International Lung Cancer Consortium*

The Singapore study included only women; the MD Anderson, Norwegian, and French studies included only ever-smokers. All studies had detailed information on histology that was based on International Classification of Diseases codes or pathology reports. All studies included incident cases of lung cancer. In most of the studies, the control subjects were frequency matched to the case patients on age and sex; some studies also matched on ethnicity (Hawaii and Canadian studies), place of residence (UCLA study), or smoking status (Norway and MD Anderson studies). Written informed consent was obtained from all study subjects, and the investigations were approved by the institutional review boards at each study center. Only individuals who reported white or Asian ethnicity were included in this analysis of 11 645 lung cancer case patients and 14 954 control subjects, of whom 85% were white and 15% were Asian (Table 1).

Genotyping and Quality Control

Genotyping from genomic DNA isolated from blood sample or saliva (the extraction technique for each center is available upon request) was performed locally at the participating centers using TaqMan probes (Applied Biosystems, Foster City, CA) (probe and primer sequences are provided in Supplementary Table 1 [available online]) that were supplied by IARC with the following exceptions: Two studies (Toronto and France) used genotyping data that were obtained from HumanHap300 BeadChips (Illumina, San Diego, CA), the German multicenter and Saarland studies used the iPLEX assay (Sequenom, San Diego, CA), and two studies (Spain and the Netherlands) performed genotyping with the use of the Centaurus platform (Nanogen, San Diego, CA). All genotyping assays were performed according to the manufacturer’s protocol. The quality of the Centaurus assays was evaluated by genotyping each assay in the HapMap CEU sample, which comprises Utah residents with ancestry from Northern and Western Europe (www.hapmap.org), and comparing the results with the HapMap publicly released data. Assays with a mismatch rate greater than 1.5% were not included in the statistical analysis. Standardized quality control procedures were applied in all centers that used TaqMan or iPLEX assays: Each center genotyped a generic series of 90 standard DNA samples (from either the HapMap CEU sample or a generic series from IARC) in their local genotyping facility. The genotype concordance across studies was subsequently computed for each genotyping assay. When more than one discrepancy between the genotypes obtained from the local genotyping technique and the HapMap publicly available genotypes or the IARC generic series genotypes for a variant was found in a study, that study was excluded from the analysis of that variant (Supplementary Table 2, available online). The average genotype completion rate per SNP varied from 97.1% to 99.6% in the pooled data, and all genotype completion rates per study were greater than 90% for each variant.

We used a χ2 test with 1 df to verify that the allele distributions for each SNP were in Hardy–Weinberg equilibrium within each study and separately among white control subjects and Asian control subjects. A Bonferroni correction for multiple tests was applied for the Hardy–Weinberg equilibrium test and gave a P value of .0005 as the cutoff for statistical significance (based on approximately 100 independent tests carried out). No deviation from Hardy–Weinberg equilibrium was observed (Supplementary Table 3, available online).

Statistical Analysis

We used unconditional logistic regression to estimate odds ratios and 95% confidence intervals (CIs). The heterozygous and homozygous carriers of the risk allele were each compared with the homozygous carriers of the nonrisk allele. Odds ratios per allele or P values for trend were calculated by assuming a log-additive genetic model with 1 df. Pooled odd ratios were calculated using individual-level data. Information on demographic variables (age, sex, ethnicity), tobacco exposures, family history of cancer, and histology classification (for case patients) was available. Mean numbers of cigarettes smoked per day were derived from analysis of variance and are adjusted for age, sex, study, and case–control status when appropriate. Ethnicity was self-declared, and only subjects who declared themselves to be white or Asian were included in the analysis. Whites and Asians were analyzed separately. Models were adjusted for potential confounders, including age, sex, study center, and, where appropriate, cumulative tobacco consumption (expressed as pack-years). To evaluate effect modification, we conducted analyses stratified by smoking status (never, former, current), smoking quantity (by 10-pack-year categories), sex, age at lung cancer diagnosis (by 10-year age groups), or family history of cancer among first-degree relatives. We also analyzed associations between genetic variants and the risk of lung cancer by major histological subtypes (squamous cell carcinoma, adenocarcinoma, small-cell carcinoma, and large-cell carcinoma). Heterogeneity of odds ratios across the studies and across the stratification groups was assessed by using the Cochran Q test.

All analyses were conducted with SAS software (version 9.1; SAS Institute, Cary, NC). All statistical tests were two-sided, and statistical significance required a P value of .05 or less.


Among the white subjects, 57% were male, whereas among the Asian subjects, a slight majority was female (50% of the case subjects and 58% of the control subjects) (Table 2). We observed a higher prevalence of never-smoking lung cancer case patients among Asians (40%) than among whites (10%). This difference is mainly because of the Singapore study, which included only women, of whom 79% were never-smokers.

Table 2
Distribution of selected demographic variables by ethnic group*

Table 3 summarizes the pooled estimates of the main effects for each variant. In whites, both of the variants at 15q25 were strongly associated with the risk of lung cancer and exhibited similar odds ratios in heterozygotes, in homozygotes, and per allele. The strongest association of the two variants at 15q25 was for rs16969968 (OR = 1.26, 95% CI = 1.21 to 1.32, Ptrend = 2 × 10−26). We also noted associations between the two variants located on chromosome 5p15 and the risk of lung cancer (rs2736100: OR = 1.15, 95% CI = 1.10 to 1.20, Ptrend = 1 × 10−10; rs402710: OR = 1.14, 95% CI = 1.09 to 1.19, Ptrend = 5 × 10−8). Among the two variants at 6p21, we observed a statistically significant association between the wild-type allele of rs4324798 and the risk of lung cancer among homozygotes (OR = 1.39, 95% CI = 1.04 to 1.87); however, in the log-additive model, neither variant on this chromosome was associated with the risk of lung cancer. In Asians, the minor allele frequencies of rs16969968 and rs8034191 on chromosome 15q25 were lower than 5% and no association with the risk of lung cancer was observed. None of the other variants selected from this region was associated with risk of lung cancer in this ethnic group. However, for chromosome 5p15, there were statistically significant associations between rs2736100 (OR = 1.23, 95% CI = 1.12 to 1.35, Ptrend = 2 × 10−5) and rs402710 (OR = 1.15, 95% CI = 1.04 to 1.27, Ptrend = .007) and the risk of lung cancer. No statistically significant heterogeneity by study was observed. The study-specific odds ratios are presented in Supplementary Figures 1 and 2 (available online).

Table 3
Summary estimates of the main effects of the selected variants in whites and Asians*

We conducted stratified analyses of the chromosome 15q25 variants in whites (Figure 1). Because linkage disequilibrium between rs16969968 and rs8034191 was high (D′ = .95, r2 = .88), we reported the results only for rs16969968 (results for rs8034191 were similar and are not shown). Among never-smokers, there was no association between rs16969968 and the risk of lung cancer (OR = 1.02, 95% CI = 0.91 to 1.14). Among ever-smokers, this association was statistically significantly stronger in current smokers (OR = 1.39, 95% CI = 1.29 to 1.50) than in former smokers (OR = 1.27, 95% CI = 1.18 to 1.37, Pheterogeneity = 1 × 10−5). We also noted a higher odds ratio in subjects younger than 50 years compared with older subjects (Pheterogeneity = 9 × 10−4). The mean age at lung cancer diagnosis was statistically significantly higher among homozygous carriers of the common allele than among homozygous carriers of the rare allele (62.8 vs 60.7 years, difference = 2.1 years, 95% CI = 1.2 to 3.3 years). There were no statistically significant differences in the odds ratio estimates by histology, pack-years of cumulative tobacco consumption in ever-smokers, or sex. Among subjects with no missing data for pack-years and smoking status, the overall odds ratio adjusted for these two variables was slightly lower than the unadjusted odds ratio (adjusted OR = 1.25, 95% CI = 1.18 to 1.32; unadjusted OR = 1.33, 95% CI = 1.26 to 1.40). A stratified analysis of rs16969968 by family history of lung cancer among first-degree relatives revealed no heterogeneity between subjects with and without a family history of cancer (data not shown).

Figure 1
Stratified analysis of the association between rs16969968 and the risk of lung cancer in whites. Except for the odds ratios (ORs) for heterozygous and homozygous effect, odds ratios and 95% confidence intervals (CIs) were derived from the per-allele model. ...

We also investigated the association between rs16969968 and the risk of lung cancer in the context of smoking intensity in whites and observed a gene dosage effect with the mean number of cigarettes smoked per day (Table 4). Overall, the mean number of cigarettes smoked per day was 20.74 (95% CI = 20.36 to 21.12) among homozygous carriers of the common allele and 23.48 (95% CI = 22.92 to 24.04) among homozygous carriers of the risk allele. Similar trends were observed in case patients and control subjects when analyzed separately.

Table 4
Association between rs16969968 and smoking intensity expressed in cigarettes smoked per day in whites*

Figure 2 shows the stratified estimates for rs2736100 and rs402710 at chromosome 5p15. We observed statistically significant heterogeneity by histology for rs2736100 for both whites (P < .001) and Asians (P = .01), and the risks of adenocarcinomas and large-cell carcinomas were higher than the risks of squamous and small-cell carcinomas. We also observed heterogeneity by histology for rs402710, with stronger associations for adenocarcinomas and large-cell and squamous cell carcinomas than for small-cell carcinomas; however, this heterogeneity was statistically significant only in whites (Pheterogeneity = .03). We also observed a sex difference for rs2736100, with a stronger association in women than in men (Pheterogeneity = .02 for whites and Pheterogeneity = .03 for Asians).

Figure 2
Stratified analysis of associations between rs2736100 and rs402710 and the risk of lung cancer in whites and Asians. Except for the odds ratios (ORs) for heterozygous and homozygous effect, odds ratios and 95% confidence intervals (CIs) were derived from ...

Because a higher proportion of adenocarcinomas is usually more frequent in women than in men (in this study, adenocarcinomas were diagnosed in 21% of the female case patients vs 15% of the male case patients), we stratified the analysis of rs2736100 by histology and by sex. For both men and women, the association between the risk of lung cancer and rs2736100 remained stronger for adenocarcinomas than for other histologies (data not shown). Conversely, when the analysis of rs2736100 was restricted to adenocarcinomas, no heterogeneity by sex was observed (data not shown).

We also compared patients with a family history of lung cancer with patients without a family history of lung cancer and observed a statistically significant association between having a family history of lung cancer and carrying the rare allele of rs2736100 (OR = 1.16, 95% CI = 1.03 to 1.32, Ptrend = .02). Likewise, the risk of lung cancer associated with the variant genotype was higher among subjects with a family history (OR = 1.31, 95% CI = 1.16 to 1.48) than among those without a family history (OR = 1.14, 95% CI = 1.06 to 1.23); however, the difference was not statistically significant (Pheterogeneity = .06). No evidence of heterogeneity by family history was observed for rs402710, and no heterogeneity by age group was observed for either chromosome 5p15 variant (data not shown). These results did not change after adjustment for smoking intensity (pack-years) or smoking status.

Although the physical distance between rs2736100 and rs402710 on chromosome 5 is approximately 34 kb, the linkage disequilibrium between these two variants is low (whites: r2 = .03, D′ = .23; Asians: r2 = .04, D′ = .38). We therefore examined the independent associations of rs402710 and rs2736100 with the risk of lung cancer by adjusting the effect of each variant for the other and found that the association remained statistically significant for both variants (data not shown). Moreover, we calculated the association between rs2736100 (C allele) and the risk of lung cancer among those who were homozygous for the common allele of rs402710 (GG genotype) and found an allelic odds ratio of 1.15 (95% CI = 1.08 to 1.22) in whites and 1.14 (95% CI = 1.00 to 1.30) in Asians. Conversely, the allelic odds ratio for rs402710 (G allele) among homozygous carriers of rs2736100 (AA genotype) was 1.09 (95% CI = 1.00 to 1.19) in whites and 1.15 (95% CI = 0.98 to 1.36) in Asians.

When we summed the number of risk alleles for the rs2736100 and rs402710 genotypes, we found a statistically significant odds ratio per risk allele (whites: OR = 1.12, 95% CI = 1.09 to 1.16; Asians: OR = 1.15, 95% CI = 1.08 to 1.23) (Table 5).

Table 5
Association between the risk of lung cancer and combined genotypes of rs402710 and rs2736100*

We also analyzed the risk of lung cancer associated with the combined genotypes of rs16969968, rs2736100, and rs402710 in whites (Table 6). The odds ratio of lung cancer for homozygous carriers of the three risk variants compared with individuals with no risk alleles was 2.64 (95% CI = 1.86 to 3.74, P = 4 × 10−8).

Table 6
Association between the risk of lung cancer and the number of risk alleles combining genotypes of rs402710, rs2736100, and rs16969968 in whites*


We replicated the results of previous genome-wide association studies for associations between the 15p25 locus, which includes the α5α3β4 family of nicotinic receptor genes, and the risk of lung cancer in whites and obtained an odds ratio of similar magnitude to that previously reported. We also confirmed previous reports (1,21) of no association between this locus and the risk of lung cancer in white lifetime never-smokers. We also observed no association between 15q25 and the risk of lung cancer in Asians. For the chromosome 5p15 region, we confirmed the statistically significant association with risk of lung cancer in whites reported previously and now report an association of similar magnitude in Asians. We also noted a stronger association between the two variants in 5p15 and adenocarcinoma vs other histology groups for both whites and Asians. We did not replicate the association between chromosome 6p21 and the risk of lung cancer that was reported by Hung et al. (2).

This replication study in two distinct ethnic groups represents, to our knowledge, the largest international effort in lung cancer based on independent studies that were not included in previous genome-wide association studies. In addition to replicating associations from the genome-wide association studies in whites, we expanded our analysis to Asian populations because we hypothesized that the different genetic architecture and linkage disequilibrium structure of Asians might elucidate associations with the putative causal variants. The sample size allowed us to analyze individual-level data and ensured that we had adequate statistical power for stratified analyses of associations between variants and the risk of lung cancer based on histology, age at lung cancer diagnosis, smoking status, smoking quantity, family history of lung cancer, and ethnicity.

There is unequivocal evidence that the 15q25 locus is associated with smoking status and nicotine dependence in whites. Saccone et al. (20) identified this region in a candidate gene study that compared nicotine-dependent smokers with nondependent smokers who were categorized according to measures derived from the Fagerström Test of Nicotine Dependence. Subsequently, Berrettini et al. (21) identified the same genetic association between chromosome 15q25 and smoking intensity in a comparison of heavy vs light smokers. Finally, in a genome-wide association study on smoking quantity and nicotine dependence, Thorgeirsson et al. (3) found a statistically significant association between the number of cigarettes smoked per day and the 15q25 locus. Nicotine dependence was also statistically significantly associated with the same genetic markers at 15q25. These findings were subsequently confirmed by other groups that used correlated clinical characteristics, such as smoking quantity and heavy vs light smoking groups, in different populations, including community-based populations and alcohol-dependent subjects (22,23). In a genome-wide association study, Caporaso et al. (24) identified multiple SNPs that were statistically significantly associated with the number of cigarettes consumed per day at a P value less than .001. They also combined their 15q25 results with data from the three published lung cancer genome-wide association studies and found that rs1051730 was highly statistically significantly associated with the number of cigarettes smoked per day (P = 5 × 10−32), as was rs8034191 (P = 2 × 10−29).

Le Marchand et al. (25) also reported that smokers who carried either the rs1051730 or the rs16969968 variant had higher internal doses of nicotine (nicotine equivalents) and 4-(methylnitrosamino)-I-(3-pyridyl)-1-butanone (a tobacco-specific carcinogen) per cigarette smoked compared with smokers who did not carry either variant, indicating that carriers of these variants not only smoke more cigarettes but also smoke more intensely, extracting a greater amount of nicotine and carcinogens per cigarette, compared with noncarriers. rs16969968 causes an amino acid substitution in the neuronal acetylcholine receptor subunit alpha-5. In vitro studies by Bierut et al. (26) have shown that in carriers of the risk allele, the α4β2α5 receptors exhibit a lower response to an agonist. In a sample of 1050 nicotine-dependent case patients and 879 non–nicotine dependent control subjects, Saccone et al. (27) reported two distinct loci in CHRNA5CHRNA3CHRNB4 gene cluster to be associated with nicotine dependence.

In this study, we report different results in whites and in Asians for the selected variants at chromosome 15q25. In whites, we identified a statistically significant gene dosage effect with the highest reported number of cigarettes smoked per day in both case and control subjects who were carriers of the rs16969968 homozygous mutant genotype. We found no difference in the association between this variant and the risk of lung cancer by histological subtype, sex, or number of pack-years smoked, although current smokers exhibited a slightly higher risk of lung cancer compared with former smokers. However, we did note that the highest overall risk between rs16969968 and the risk of lung cancer was among patients who were diagnosed before the age of 50 years. This finding confirm previous observations from the MD Anderson genome-wide association study that risk estimates for subjects who carried the variant genotype were higher for younger patients and that carriers of the variant genotype exhibited earlier age at lung cancer diagnosis than noncarriers (22). This inverse trend with age may argue for a direct role of this region in lung carcinogenesis. However, in this study, among the 922 white patients who had never smoked, there was no evidence of any association between rs16968869 and the risk of lung cancer, suggesting that active smoking is a necessary cofactor for lung carcinogenesis.

In Asians, we observed no association between the five variants at 15q25 and the risk of lung cancer or the number of cigarettes smoked per day. The lower minor allele frequency of rs16969968 in Asians compared with whites (0.03 vs 0.3; Table 3) and the high proportion of Asian never-smokers (40% of the case patients and 60% of the control subjects) may partially explain these negative findings. We also lack any a priori evidence that the genetic markers at 15q25 that we chose to study are relevant in Asians. However, other studies in Asian populations have reported lung cancer susceptibility loci at chromosome 15q25. For example, a Japanese case–control study (28) reported an association between rs16969968 and the risk of lung cancer (OR = 2.2, 95% CI = 1.5 to 3.4, P = 1.5 × 10−4), and similar associations were observed among never-smokers (OR = 2.4, 95% CI = 1.2 to 4.7, P = .013) and ever-smokers (OR = 2.2, 95% CI = 1.1 to 4.1, P = .016). Although the minor allele frequency of rs16969968 in the Japanese study population was, as expected, very low (0.015), the proportion of never-smokers among the case patients was lower than that in our study (21% vs 40%). Another large study conducted in China (29) identified statistically significant associations between four novel SNPs in the 15q25 region and age at lung cancer diagnosis and smoking behavior, whereas none of these associations was reported for the 15q25 SNPs that were previously reported in whites. The associations between the four novel SNPs and the risk of lung cancer were substantially stronger in younger age at diagnosis patients, similar to our findings in white populations, and were consistent among never-smokers and smokers (29). The latter observation argues strongly for a role of the 15q25 locus in lung cancer that is independent of smoking behavior in Asians.

This replication analysis provided conclusive evidence for associations between rs2736100 and rs402710 at chromosome 5p15 and susceptibility to lung cancer in both whites and Asians. These associations appeared to be independent because restricting the analysis to homozygous carriers of common alleles of one variant did not alter the association of the other variant. This finding may also suggest the existence of an unknown variant that is in linkage disequilibrium with rs2736100 and rs402710 that captures the effect of both rs2736100 and rs402710. In contrast with the chromosome 15q25 findings, both SNPs at chromosome 5p15 were associated with statistically significant increased risks of lung cancer in never-smokers as well as in ever-smokers, and there were no patterns of association by age at diagnosis or duration of smoking for either ethnic group.

These results were in accordance with those reported by previous genome-wide association studies in a white population. McKay et al. (4) reported odds ratios of 1.14 for rs2736100 and 1.18 for rs402710, whereas Wang et al. (5) reported an odds ratio of 1.14 for rs401681 (in strong linkage disequilibrium with rs402710). Neither of these studies reported heterogeneity by histology, smoking status, age at diagnosis, or sex. The magnitudes of these associations are consistent with our findings. However, associations between these variants and the risk of lung cancer differed by histology, and this finding was consistent in both ethnic groups. In particular, we identified an increased risk for adenocarcinomas for both variants in both whites and Asians and an absence of any risk for small-cell carcinomas. Squamous and large-cell carcinomas gave intermediate results. Another study conducted in Iceland (6) found associations between rs401681, located in the CLPTM1L gene, and several smoking-related cancers, including lung cancer (P = 7 × 10−8), as well as cancers of the bladder, prostate, skin, and cervix. They also noted that rs2736098, which is located in the TERT gene at 5p15, showed a stronger association with lung cancer (P = 3 × 10−5), bladder and prostate cancers, and basal cell carcinomas.

In a case–control study conducted in a Chinese population, Jin et al. (30) reported that rs2736100 was associated with an increased risk of non–small cell lung carcinomas, with odds ratios of 1.26 (95% CI = 1.05 to 1.51) and 1.31 (95% CI = 1.04 to 1.66) for one and two copies of the variant C allele, respectively. They noted that the association was more prominent among women (Pheterogeneity = .044), nonsmokers (Pheterogeneity = .054), and patients with adenocarcinoma (Pheterogeneity = .058). Our data suggested that the association between rs2736100 and the risk of lung cancer was strongest for adenocarcinoma and that the difference between males and females was at least partially explained by the higher proportion of adenocarcinoma among women.

rs2736100 and rs402710 were also previously found to be associated with other diseases. rs2736100 was found to be associated with glioma susceptibility in two recent genome-wide association studies (31,32) conducted in whites. Another genome-wide association study conducted in a Japanese population found that this SNP was associated with idiopathic pulmonary fibrosis (33).

TERT encodes the enzyme telomerase reverse transcriptase, which is the catalytic subunit of telomerase that adds telomeric repeat sequences onto chromosome ends (34). High expression of telomerase is commonly observed in lung cancer, which suggests that TERT may have a critical role in lung tumorigenesis (3537). An association between short telomeres and an increased risk of cancer has been reported for several types of cancers, including basal cell carcinoma, cancers of the lung, head and neck, bladder, kidney, esophagus, and breast, as well as lymphoma (6,3842). Rafnar et al. (6) reported that rs401681 and rs2736098 were associated with shorter telomere length in older healthy women but not in younger healthy women and suggested that these variants may lead to an increase in the gradual shortening of telomeres over time. Zhu et al. (43) reported that TERT gene amplification is more commonly seen in adenocarcinoma than in squamous cell carcinoma and that overexpression of TERT mRNA is correlated with TERT gene amplification in adenocarcinoma but not in squamous cell carcinoma. Zhu et al. hypothesized that overexpression of TERT mRNA in adenocarcinomas is largely due to TERT amplification, whereas in other lung tumor types, it is mainly controlled by epigenetic factors. Our data are in accordance with these findings.

Concerning the variants at chromosome 6p21, the associations with rs2256543 and rs4324798 were not replicated in whites in this study. It should be noted that for this replication analysis, we selected the two most statistically significant variants from the IARC genome-wide association study (4). The borderline statistically significant associations we reported may indicate that other SNPs in this region may be better candidates than the ones we selected. For example, Wang et al. (5) reported associations between two 6p21 variants—rs3117582 in BAT3 and rs3131379 in MSH5—that are located approximately 3 Mb away from the 6p21 variants analyzed in this study. These two SNPs are highly correlated (r2 = .99), and the genes in which they reside are strong candidates for lung cancer susceptibility loci: BAT3 is implicated in apoptosis (44), and MSH5 is involved in DNA mismatch repair (45). Further investigation of this region is warranted.

This study has several limitations. First, we selected variants that were found to be associated with lung cancer in genome-wide association studies that were conducted in whites. Replication of these variants in Asians may not be relevant. As we reported in this study, the variants we selected at chromosome 15q25 (rs16969968 and rs8034191) had a very low minor allele frequency in Asians. Three other variants at 15q25 were therefore selected for this population, albeit with a lack of a priori evidence, and were also not replicated. Second, many different studies with different genotyping protocols were included in this study, which could have lead to heterogeneity. However, we implemented stringent interlaboratory quality control procedures in all centers and found no evidence of any such heterogeneity by study.

In conclusion, this analysis exemplifies the timely and cost-effective contributions that international consortia can provide to genome-wide association replication studies. Our observations of heterogeneity by histology of associations between variants at 5p15 and the risk of lung cancer are particularly notable and indicate that further study of the role of this locus in lung cancer development is warranted. Future lung cancer genome-wide association studies should routinely include histology-specific analyses.

Supplementary Data

Supplementary data can be found at http://www.jnci.oxfordjournals.org/.


US National Institutes of Health (NIH); National Cancer Institute (NCI) (R03 CA133939-01 and R01 CA092039).

The individual studies were funded by the following sources: MD Anderson study: NCI (CA55769, CA127219, CA121197, and CA133996); German multicenter study: Lung Cancer in the Young study was partly funded by the Deutsche Forschungsgemeinschaft (DFG; BI 576/2-1 and BI 576/2-2), the genome-wide study was funded by the Helmholtz Association, Germany; the Heidelberg lung cancer study was supported by the Deutsche Krebshilfe; the European Prospective Investigation into Cancer and Nutrition-Heidelberg study was funded by “Europe Against Cancer” Programme of the European Commission (Santé et protection des consommateurs [SANCO]); German Cancer Aid; German Cancer Research Center; German Federal Ministry of Education and Research; Mayo Clinic study: NCI (CA77118, CA80127, and CA84354); University of California, San Francisco study: National Institute of Environmental Health Sciences (ES06717); Aichi study: Scientific Research grant from the Ministry of Education, Science, Sports, Culture and Technology of Japan (17015052) and grant for the Third-Term Comprehensive 10-Year Strategy for Cancer Control from the Ministry of Health, Labor and Welfare of Japan (H20-002); Singapore study: National Medical Research Council, Singapore (NMRC/1075/2006); Norwegian study: The Norwegian Cancer Society; UCLA study: NIH (DA11386, CA90833, CA09142, and ES011667) and the Alper Research Center for Environmental Genomics of the UCLA's Jonsson Comprehensive Cancer Center; New England Lung Cancer study: National Center for Research Resources, a component of the NIH (P20RR018787); Israel study: United States–Israel Binational Science Foundation (BSF) (2003159); Penn State study: NCI (PO1 CA68384, K07 CA104231, and K99CA131477) and PA-DOH 4100038714 and PA-DOH 4100038715 from the Pennsylvania Department of Health; Saarland study: Baden-Württemberg Ministry of Research, Science and Arts; Cologne study: this study was funded by the Helmholtz Association, Germany, through VH.VI-143 and by the Monika Kutzner Stiftung; University of Hawaii Study: NCI (CA55874 and CA85997); Toronto study: Canadian Cancer Society (20214).

Supplementary Material

[Supplementary Data]


K. Stefansson owns stock in deCODE Genetics, which produces genetic risk assessment tests for common diseases and traits. Neither deCODE Genetics nor the study sponsors played a role in the design of the study, analysis or interpretation of the data, the writing of the manuscript, or the decision to submit the manuscript for publication.


1. Amos CI, Wu X, Broderick P, et al. Genome-wide association scan of tag SNPs identifies a susceptibility locus for lung cancer at 15q25.1. Nat Genet. 2008;40(5):616–622. [PMC free article] [PubMed]
2. Hung RJ, McKay JD, Gaborieau V, et al. A susceptibility locus for lung cancer maps to nicotinic acetylcholine receptor subunit genes on 15q25. Nature. 2008;452(7187):633–637. [PubMed]
3. Thorgeirsson TE, Geller F, Sulem P, et al. A variant associated with nicotine dependence, lung cancer and peripheral arterial disease. Nature. 2008;452(7187):638–642. [PubMed]
4. McKay JD, Hung RJ, Gaborieau V, et al. Lung cancer susceptibility locus at 5p15.33. Nat Genet. 2008;40(12):1404–1406. [PMC free article] [PubMed]
5. Wang Y, Broderick P, Webb E, et al. Common 5p15.33 and 6p21.33 variants influence lung cancer risk. Nat Genet. 2008;40(12):1407–1409. [PMC free article] [PubMed]
6. Rafnar T, Sulem P, Stacey SN, et al. Sequence variants at the TERT-CLPTM1L locus associate with many cancer types. Nat Genet. 2009;41(2):221–227. [PubMed]
7. Hung RJ, Christiani DC, Risch A, et al. International Lung Cancer Consortium: pooled analysis of sequence variants in DNA repair and cell cycle pathways. Cancer Epidemiol Biomarkers Prev. 2008;17(11):3081–3089. [PMC free article] [PubMed]
8. Heck J, Andrew A, Onega T, et al. Lung cancer in a US population with low to moderate arsenic exposure. Environ Health Perspect. 2009;117(11):1718–1723. [PMC free article] [PubMed]
9. Gallagher CJ, Muscat JE, Hicks AN, et al. The UDP-glucuronosyltransferase 2B17 gene deletion polymorphism: sex-specific association with urinary 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol glucuronidation phenotype and risk for lung cancer. Cancer Epidemiol Biomarkers Prev. 2007;16(4):823–828. [PubMed]
10. Cui Y, Morgenstern H, Greenland S, et al. Polymorphism of Xeroderma Pigmentosum group G and the risk of lung cancer and squamous cell carcinomas of the oropharynx, larynx and esophagus. Int J Cancer. 2006;118(3):714–720. [PubMed]
11. Wrensch MR, Miike R, Sison JD, et al. CYP1A1 variants and smoking-related lung cancer in San Francisco Bay area Latinos and African Americans. Int J Cancer. 2005;113(1):141–147. [PubMed]
12. Zienolddiny S, Campa D, Lind H, et al. Polymorphisms of DNA repair genes and risk of non-small cell lung cancer. Carcinogenesis. 2006;27(3):560–567. [PubMed]
13. Sauter W, Rosenberger A, Beckmann L, et al. Matrix metalloproteinase 1 (MMP1) is associated with early-onset lung cancer. Cancer Epidemiol Biomarkers Prev. 2008;17(5):1127–1135. [PubMed]
14. Wichmann HE, Gieger C, Illig T. KORA-gen—resource for population genetics, controls and a broad spectrum of disease phenotypes. Gesundheitswesen. 2005;67(suppl 1):S26–S30. [PubMed]
15. Wichmann HE, Rosario AS, Heid IM, Kreuzer M, Heinrich J, Kreienbrock L. Increased lung cancer risk due to residential radon in a pooled and extended analysis of studies in Germany. Health Phys. 2005;88(1):71–79. [PubMed]
16. Boeing H, Wahrendorf J, Becker N. EPIC-Germany—a source for studies into diet and risk of chronic diseases. European Investigation into Cancer and Nutrition. Ann Nutr Metab. 1999;43(4):195–204. [PubMed]
17. Breitling LP, Raum E, Muller H, Rothenbacher D, Brenner H. Synergism between smoking and alcohol consumption with respect to serum gamma-glutamyltransferase. Hepatology. 2009;49(3):802–808. [PubMed]
18. Kim JH, Kim H, Lee KY, et al. Genetic polymorphisms of ataxia telangiectasia mutated affect lung cancer risk. Hum Mol Genet. 2006;15(7):1181–1186. [PubMed]
19. Suzuki T, Matsuo K, Hiraki A, et al. Impact of one-carbon metabolism-related gene polymorphisms on risk of lung cancer in Japan: a case control study. Carcinogenesis. 2007;28(8):1718–1725. [PubMed]
20. Saccone SF, Hinrichs AL, Saccone NL, et al. Cholinergic nicotinic receptor genes implicated in a nicotine dependence association study targeting 348 candidate genes with 3713 SNPs. Hum Mol Genet. 2007;16(1):36–49. [PMC free article] [PubMed]
21. Berrettini W, Yuan X, Tozzi F, et al. Alpha-5/alpha-3 nicotinic receptor subunit alleles increase risk for heavy smoking. Mol Psychiatry. 2008;13(4):368–373. [PMC free article] [PubMed]
22. Spitz MR, Amos CI, Dong Q, Lin J, Wu X. The CHRNA5-A3 region on chromosome 15q24-25.1 is a risk factor both for nicotine dependence and for lung cancer. J Natl Cancer Inst. 2008;100(21):1552–1556. [PMC free article] [PubMed]
23. Weiss RB, Baker TB, Cannon DS, et al. A candidate gene approach identifies the CHRNA5-A3-B4 region as a risk factor for age-dependent nicotine addiction. PLoS Genet. 2008 4(7):e1000125. [PMC free article] [PubMed]
24. Caporaso N, Gu F, Chatterjee N, et al. Genome-wide and candidate gene association study of cigarette smoking behaviors. PLoS ONE. 2009 4(2):e4653. [PMC free article] [PubMed]
25. Le Marchand ML, Derby KS, Murphy SE, et al. Smokers with the CHRNA lung cancer-associated variants are exposed to higher levels of nicotine equivalents and a carcinogenic tobacco-specific nitrosamine. Cancer Res. 2008;68(22):9137–9140. [PMC free article] [PubMed]
26. Bierut LJ, Madden PA, Breslau N, et al. Novel genes identified in a high-density genome wide association study for nicotine. Hum Mol Genet. 2007;16(1):24–35. [PMC free article] [PubMed]
27. Saccone NL, Saccone SF, Hinrichs AL, et al. Multiple distinct risk loci for nicotine dependence identified by dense coverage of the complete family of nicotinic receptor subunit (CHRN) genes. Am J Med Genet B Neuropsychiatr Genet. 2009;150B(4):453–466. [PMC free article] [PubMed]
28. Shiraishi K, Kohno T, Kunitoh H, et al. Contribution of nicotine acetylcholine receptor polymorphisms to lung cancer risk in a smoking-independent manner in the Japanese. Carcinogenesis. 2009;30(1):65–70. [PubMed]
29. Wu C, Hu Z, Yu D, et al. Genetic variants on chromosome 15q25 associated with lung cancer risk in Chinese populations. Cancer Res. 2009;69(12):5065–5072. [PubMed]
30. Jin G, Xu L, Shu Y, et al. Common genetic variants on 5p15.33 contribute to risk of lung adenocarcinoma in a Chinese population. Carcinogenesis. 2009;30(6):987–990. [PubMed]
31. Shete S, Hosking FJ, Robertson LB, et al. Genome-wide association study identifies five susceptibility loci for glioma. Nat Genet. 2009;41(8):899–904. [PubMed]
32. Wrensch M, Jenkins RB, Chang JS, et al. Variants in the CDKN2B and RTEL1 regions are associated with high-grade glioma susceptibility. Nat Genet. 2009;41(8):905–908. [PMC free article] [PubMed]
33. Mushiroda T, Wattanapokayakit S, Takahashi A, et al. A genome-wide association study identifies an association of a common variant in TERT with susceptibility to idiopathic pulmonary fibrosis. J Med Genet. 2008;45(10):654–656. [PubMed]
34. Janknecht R. On the road to immortality: hTERT upregulation in cancer cells. FEBS Lett. 2004;564(1–2):9–13. [PubMed]
35. Lantuejoul S, Salon C, Soria JC, Brambilla E. Telomerase expression in lung preneoplasia and neoplasia. Int J Cancer. 2007;120(9):1835–1841. [PubMed]
36. Mavrogiannou E, Strati A, Stathopoulou A, Tsaroucha EG, Kaklamanis L, Lianidou ES. Real-time RT-PCR quantification of human telomerase reverse transcriptase splice variants in tumor cell lines and non-small cell lung cancer. Clin Chem. 2007;53(1):53–61. [PubMed]
37. Wu TC, Lin P, Hsu CP, et al. Loss of telomerase activity may be a potential favorable prognostic marker in lung carcinomas. Lung Cancer. 2003;41(2):163–169. [PubMed]
38. Han J, Qureshi AA, Prescott J, et al. A prospective study of telomere length and the risk of skin cancer. J Invest Dermatol. 2009;129(2):415–421. [PMC free article] [PubMed]
39. Risques RA, Vaughan TL, Li X, et al. Leukocyte telomere length predicts cancer risk in Barrett's esophagus. Cancer Epidemiol Biomarkers Prev. 2007;16(12):2649–2655. [PubMed]
40. Shen J, Terry MB, Gurvich I, Liao Y, Senie RT, Santella RM. Short telomere length and breast cancer risk: a study in sister sets. Cancer Res. 2007;67(11):5538–5544. [PubMed]
41. Widmann TA, Herrmann M, Taha N, Konig J, Pfreundschuh M. Short telomeres in aggressive non-Hodgkin's lymphoma as a risk factor in lymphomagenesis. Exp Hematol. 2007;35(6):939–946. [PubMed]
42. Wu X, Amos CI, Zhu Y, et al. Telomere dysfunction: a potential cancer predisposition factor. J Natl Cancer Inst. 2003;95(16):1211–1218. [PubMed]
43. Zhu CQ, Cutz JC, Liu N, et al. Amplification of telomerase (hTERT) gene is a poor prognostic marker in non-small-cell lung cancer. Br J Cancer. 2006;94(10):1452–1459. [PMC free article] [PubMed]
44. Sasaki T, Gan EC, Wakeham A, Kornbluth S, Mak TW, Okada H. HLA-B-associated transcript 3 (Bat3)/Scythe is essential for p300-mediated acetylation of p53. Genes Dev. 2007;21(7):848–861. [PMC free article] [PubMed]
45. Her C, Zhao N, Wu X, Tompkins JD. MutS homologues hMSH4 and hMSH5: diverse functional implications in humans. Front Biosci. 2007;12:905–911. [PubMed]

Articles from JNCI Journal of the National Cancer Institute are provided here courtesy of Oxford University Press
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

  • Genome-wide association analysis identifies new lung cancer susceptibility loci in never-smoking women in Asia[Nature genetics. 2012]
    Lan Q, Hsiung CA, Matsuo K, Hong YC, Seow A, Wang Z, Hosgood HD III, Chen K, Wang JC, Chatterjee N, Hu W, Wong MP, Zheng W, Caporaso N, Park JY, Chen CJ, Kim YH, Kim YT, Landi MT, Shen H, Lawrence C, Burdett L, Yeager M, Yuenger J, Jacobs KB, Chang IS, Mitsudomi T, Kim HN, Chang GC, Bassig BA, Tucker M, Wei F, Yin Z, Wu C, An SJ, Qian B, Lee VH, Lu D, Liu J, Jeon HS, Hsiao CF, Sung JS, Kim JH, Gao YT, Tsai YH, Jung YJ, Guo H, Hu Z, Hutchinson A, Wang WC, Klein R, Chung CC, Oh IJ, Chen KY, Berndt SI, He X, Wu W, Chang J, Zhang XC, Huang MS, Zheng H, Wang J, Zhao X, Li Y, Choi JE, Su WC, Park KH, Sung SW, Shu XO, Chen YM, Liu L, Kang CH, Hu L, Chen CH, Pao W, Kim YC, Yang TY, Xu J, Guan P, Tan W, Su J, Wang CL, Li H, Sihoe AD, Zhao Z, Chen Y, Choi YY, Hung JY, Kim JS, Yoon HI, Cai Q, Lin CC, Park IK, Xu P, Dong J, Kim C, He Q, Perng RP, Kohno T, Kweon SS, Chen CY, Vermeulen R, Wu J, Lim WY, Chen KC, Chow WH, Ji BT, Chan JK, Chu M, Li1 YJ, Yokota J, Li J, Chen H, Xiang YB, Yu CJ, Kunitoh H, Wu G, Jin L, Lo YL, Shiraishi K, Chen YH, Lin HC, Wu T, Wu YL, Yang PC, Zhou B, Shin MH, Fraumeni JF Jr, Lin D, Chanock SJ, Rothman N. Nature genetics. 2012 Dec; 44(12)1330-1335
  • Genetic Architectures of Psychiatric Disorders: The Emerging Picture and Its Implications[Nature reviews. Genetics. ]
    Sullivan PF, Daly MJ, O’Donovan M. Nature reviews. Genetics. 13(8)537-551
  • Pleiotropic Associations of Risk Variants Identified for Other Cancers With Lung Cancer Risk: The PAGE and TRICL Consortia[JNCI Journal of the National Cancer Institu...]
    Park SL, Fesinmeyer MD, Timofeeva M, Caberto CP, Kocarnik JM, Han Y, Love SA, Young A, Dumitrescu L, Lin Y, Goodloe R, Wilkens LR, Hindorff L, Fowke JH, Carty C, Buyske S, Schumacher FR, Butler A, Dilks H, Deelman E, Cote ML, Chen W, Pande M, Christiani DC, Field JK, Bickebӧller H, Risch A, Heinrich J, Brennan P, Wang Y, Eisen T, Houlston RS, Thun M, Albanes D, Caporaso N, Peters U, North KE, Heiss G, Crawford DC, Bush WS, Haiman CA, Landi MT, Hung RJ, Kooperberg C, Amos CI, Le Marchand L, Cheng I. JNCI Journal of the National Cancer Institute. 106(4)dju061
  • Heritability of Radiation Response in Lung Cancer Families[Genes. ]
    Rosenberger A, Rössler U, Hornhardt S, Sauter W, Bickeböller H, Wichmann HE, Gomolka M. Genes. 3(2)248-260
  • The SNP rs402710 in 5p15.33 Is Associated with Lung Cancer Risk: A Replication Study in Chinese Population and a Meta-Analysis[PLoS ONE. ]
    Lu X, Ke J, Luo X, Zhu Y, Zou L, Li H, Zhu B, Xiong Z, Chen W, Deng L, Lou J, Wang X, Zhang Y, Wang Z, Miao X, Cheng L. PLoS ONE. 8(10)e76252
See all...


  • EST
    Published EST sequences
  • Gene
    Gene links
  • GEO Profiles
    GEO Profiles
    Related GEO records
  • HomoloGene
    HomoloGene links
  • PubMed
    PubMed citations for these articles
  • SNP
    PMC to SNP links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...