Background: Two recent genome-wide association studies (GWAS) of pancreatic ductal adenocarcinoma (PDAC), conducted, respectively, in a Japanese and in a Chinese population, identified eight novel loci affecting PDAC risk.

Methods: We attempted to replicate the novel loci in a series of PDACs and healthy controls of European ancestry in the context of the newly formed PANcreatic Disease ReseArch (PANDoRA) consortium. We genotyped seven single-nucleotide polymorphisms (SNP): rs12413624, rs1547374, rs372883, rs5768709, rs6464375, rs708224, rs9502893 (one SNP identified in the Chinese GWAS is not polymorphic in Caucasians) in 1,299 PDAC cases and 2,884 controls. We also attempted stratified analysis considering the different stages of the disease and addressed the possible involvement of the selected SNPs on the survival of patients.

Results: None of the SNPs were significantly associated with PDAC risk if considering the overall population of the consortium. When stratifying for country of origin, we found that in the Polish subgroup, the G allele of rs372883 was statistically significantly associated with increased risk [OR, 6.40; 95% confidence interval (CI), 2.28–17.91]. However, the sample size of the subgroups was rather small; therefore, this result can be due to chance. None of the SNPs was associated with disease progression or survival.

Conclusions: None of the SNPs associated with PDAC risk in two Asian populations were convincingly associated with PDAC risk in individuals of European descent.

Impact: This study illustrates the importance of evaluation of PDAC risk markers across ethnic groups. Cancer Epidemiol Biomarkers Prev; 22(2); 320–3. ©2012 AACR.

Pancreatic cancer mortality rates approach incidence rates (1). Due to the lack of sufficient risk factors, finding genetic variants associated with disease risk is of utmost importance. A genome-wide association study (GWAS) performed in a Caucasian population identified 4 loci associated with pancreatic cancer risk; 2 more showed a strong association only in samples from prospective cohorts but not in retrospective case–control series (2, 3). Two recent GWAS of pancreatic ductal adenocarcinoma (PDAC), conducted in Japan (4) and China (5), identified 8 novel loci affecting PDAC risk. We attempted to replicate these loci in a series of 1,299 PDAC and 2,884 healthy controls of European ancestry in the context of the PANcreatic Disease ReseArch (PANDoRA) consortium. We also attempted stratified analysis considering the different stages of the disease and addressed the possible involvement of the single-nucleotide polymorphisms (SNP) on the survival of patients.

Characteristics of the study population were described in detail elsewhere (6). A total of 1,299 PDAC cases and 2,884 controls were used in this study. We genotyped 4 SNPs identified by Wu (5) and 3 identified by Low (4). Genotyping was conducted using the KASPar SNP genotyping system (KBiosciences; ref. 7). Genotyping for British controls has been conducted in the context of a GWAS as described before (7).

Risk analysis was conducted by logistic regression for multivariate analyses to assess the main effects of the genetic polymorphism on pancreatic cancer risk using the same inheritance model reported by Wu and Low. The most common allele in the controls was assigned as the reference category. Survival analysis was conducted using HRs and 95% confidence intervals (CI) in Cox proportional hazard models. All analyses were adjusted for age, gender, tumor–node–metastasis (TNM) stage (for survival only), and nationality. We also conducted stratified analysis for risk and survival considering the various nationalities and the different stages as different strata. All analyses were conducted with STATA software (StataCorp).

The 7 SNPs were genotyped in all cases and healthy controls. Relevant characteristics of the study population are given in Table 1. The average call rate was 95.86% (range, 93.93%–97.83%). Approximately 10% of the samples were analyzed in duplicate; the concordance rate of the genotypes was above 99%. The genotype distributions at all loci were in Hardy–Weinberg equilibrium in controls, with nonsignificant χ2 values (data not shown). The frequencies and distribution of the genotypes, the ORs and 95% CIs for the association with PDAC are shown in Table 2. Heterozygous AG carriers of rs1547374 were associated with increased risk (OR, 1.16; 95% CI, 1.00–1.35; P = 0.04), whereas we observed a trend for rs5768709 and decreased risk (Ptrend = 0.04) and in heterozygous AG carriers of rs9502893 (OR, 0.84; 95% CI, 0.71–0.98; P = 0.03). After Bonferroni correction, none of the above variants remained significant. Stratifying by nationality, we found that in the Polish subgroup, the G allele of rs372883 was significantly associated with increased risk (OR, 6.40; 95% CI, 2.28–17.91; P = 0.0004). The sample size of the subgroup was rather small; there were 7 and 21 cases with AA or AG + GG genotypes, respectively, and 43 and 80 Polish controls with either AA or AG + GG genotypes. Therefore, this result has to be taken with caution. None of the other SNPs were significantly associated with PDAC risk, even when stratifying for tumor stage, or with patient survival, considering the correction for multiple testing. Finally, we observed no statistically significant association between the SNPs and survival when stratifying for nationality or tumor stage (Table 2).

Table 1.

Characteristics of patients and healthy controls included in this study

CzechGermanItalianEnglishPolishGreek
NationalityCasesControlsCasesControlsCasesControlsCasesControlsCasesControlsCasesControls
N 228 305 672 1,579 159 462 98 280 54 163 88 95 
Type of controls  Blood donors and hospitalized individuals  Blood donors  Blood donors and hospitalized individuals  General population  Blood donors  Hospitalized individuals 
Gender 
 Female 82 132 297 775 66 196 36 112 35 87 40 46 
 Male 146 173 375 804 91 263 61 168 19 75 48 49 
 Missing 0 0 0 0 2 3 1 0 0 1 0 0 
Age at diagnosis/recruitment 
Mean (SD) 62.13 (10.46) 59.6 (12.25) 64.08 (9.83) 56.56 (9.7) 69.72 (9.94) 55.15 (10.96) 64.5 (10.15) 66.81 (9.23) 60.5 (9.27) 66.83 (10.34) 64.11 (10.11) 47.51 (15.66) 
Median (25%–75%) 63 (57–70) 59.5 (57–68) 65 (58–71) 59 (54–63) 71 (63–77) 58 (47–63) 64 (58–72) 67 (60–74) 61 (53–67) 68 (62–73) 66 (58–72) 45 (35–53) 
Survival days 
Mean (SD) 289 (243)  490 (406)  355 (227)  410 (442)  327 (176)  358 (275)  
Median (25%–75%) 252.5 (100–305)  353 (207–667)  319 (180–461)  294 (160–487)  382 (206–458)  300 (180–420)  
Deaths 150  474  97  85  21  78  
CzechGermanItalianEnglishPolishGreek
NationalityCasesControlsCasesControlsCasesControlsCasesControlsCasesControlsCasesControls
N 228 305 672 1,579 159 462 98 280 54 163 88 95 
Type of controls  Blood donors and hospitalized individuals  Blood donors  Blood donors and hospitalized individuals  General population  Blood donors  Hospitalized individuals 
Gender 
 Female 82 132 297 775 66 196 36 112 35 87 40 46 
 Male 146 173 375 804 91 263 61 168 19 75 48 49 
 Missing 0 0 0 0 2 3 1 0 0 1 0 0 
Age at diagnosis/recruitment 
Mean (SD) 62.13 (10.46) 59.6 (12.25) 64.08 (9.83) 56.56 (9.7) 69.72 (9.94) 55.15 (10.96) 64.5 (10.15) 66.81 (9.23) 60.5 (9.27) 66.83 (10.34) 64.11 (10.11) 47.51 (15.66) 
Median (25%–75%) 63 (57–70) 59.5 (57–68) 65 (58–71) 59 (54–63) 71 (63–77) 58 (47–63) 64 (58–72) 67 (60–74) 61 (53–67) 68 (62–73) 66 (58–72) 45 (35–53) 
Survival days 
Mean (SD) 289 (243)  490 (406)  355 (227)  410 (442)  327 (176)  358 (275)  
Median (25%–75%) 252.5 (100–305)  353 (207–667)  319 (180–461)  294 (160–487)  382 (206–458)  300 (180–420)  
Deaths 150  474  97  85  21  78  
View Large
Table 2.

Associations between selected SNPs and PDAC risk and patients survival

Risk analysisSurvival analysis
SNPGenotypesCasesaControlsaOR (95% CI)bPvaluePtrendTestSubjectsaDeathsaHR (95% CI)cPvalue
rs12413624 TT 392 882   0.7528 Per allele 1,091 887 0.95 (0.86–1.06) 0.38 
 AT 616 1393 1.02 (0.87–1.19) 0.832  TT vs. AT   0.99 (0.83–1.19) 0.91 
 AA 269 585 1.05 (0.86–1.28) 0.626  TT vs. AA   0.90 (0.72–1.12) 0.35 
 AT + AA   1.03 (0.88–1.19) 0.727  TT vs. (AT + AA)   0.96 (0.81–1.14) 0.63 
rs1547374 AA 526 1,279   0.1864 Per allele 1,094 895 0.94 (0.83–1.06) 0.29 
 AG 611 1,265 1.16 (1.00–1.35) 0.043  AA vs. AG   0.89 (0.75–1.05) 0.17 
 GG 144 335 1.05 (0.83–1.32) 0.685  AA vs. GG   0.92 (0.71–1.21) 0.56 
 AG + GG   1.14 (0.99–1.31) 0.067  AA vs. (AG + GG)   0.90 (0.76–1.05) 0.18 
rs372883 AA 344 775   0.952 Per allele 1,059 860 0.90 (0.80–1.00) 0.06 
 AG 613 1,446 0.97 (0.82–1.14) 0.7  AA vs. AG   0.90 (0.74–1.10) 0.31 
 GG 288 650 0.98 (0.80–1.19) 0.83  AA vs. GG   0.81 (0.65–1.01) 0.06 
 AG + GG   0.97 (0.83–1.14) 0.715  AA vs. (AG + GG)   0.87 (0.72–1.04) 0.13 
rs5768709 AA 510 951   0.0378 Per allele 1,083 883 0.97 (0.86–1.09) 0.57 
 AG 580 1,221 0.92 (0.79–1.07) 0.277  AA vs. AG   0.98 (0.82–1.16) 0.79 
 GG 180 408 0.86 (0.69–1.07) 0.166  AA vs. GG   0.93 (0.73–1.19) 0.56 
 AG + GG   0.90 (0.78–1.04) 0.17  AA vs. (AG + GG)   0.96 (0.82–1.14) 0.67 
rs6464375 GG 1124 2,522   0.3589 Per allele 1,111 904 1.26 (1.00–1.58) 0.05 
 AG 169 353 1.13 (0.92–1.38) 0.258  GG vs. AG   1.22 (0.96–1.55) 0.10 
 AA 6 9 1.83 (0.62–5.36) 0.271  GG vs. AA   3.26 (0.80–13.21) 0.10 
 AG + AA   1.14 (0.93–1.40) 0.199  GG vs. (AG + AA)   1.24 (0.98–1.57) 0.07 
rs708224 GG 449 919   0.7476 Per allele 1,104 900 1.11 (0.99–1.23) 0.08 
 AG 598 1,378 0.90 (0.77–1.05) 0.189  GG vs. AG   1.09 (0.91–1.30) 0.37 
 AA 246 501 1.07 (0.87–1.30) 0.524  GG vs. AA   1.23 (0.98–1.53) 0.07 
 AG + AA   0.94 (0.82–1.09) 0.441  GG vs. (AG + AA)   1.12 (0.95–1.33) 0.17 
rs9502893 AA 452 899   0.5069 Per allele 1,084 880 1.00 (0.90–1.11) 0.98 
 AG 555 1,338 0.84 (0.71–0.98) 0.027  AA vs. AG   0.98 (0.79–1.21) 0.85 
 GG 265 536 0.98 (0.81–1.19) 0.861  AA vs. GG   1.07 (0.90–1.26) 0.44 
 AG + GG   0.88 (0.76–1.02) 0.087  AA vs. (AG + GG)   1.13 (0.94–1.35) 0.20 
Risk analysisSurvival analysis
SNPGenotypesCasesaControlsaOR (95% CI)bPvaluePtrendTestSubjectsaDeathsaHR (95% CI)cPvalue
rs12413624 TT 392 882   0.7528 Per allele 1,091 887 0.95 (0.86–1.06) 0.38 
 AT 616 1393 1.02 (0.87–1.19) 0.832  TT vs. AT   0.99 (0.83–1.19) 0.91 
 AA 269 585 1.05 (0.86–1.28) 0.626  TT vs. AA   0.90 (0.72–1.12) 0.35 
 AT + AA   1.03 (0.88–1.19) 0.727  TT vs. (AT + AA)   0.96 (0.81–1.14) 0.63 
rs1547374 AA 526 1,279   0.1864 Per allele 1,094 895 0.94 (0.83–1.06) 0.29 
 AG 611 1,265 1.16 (1.00–1.35) 0.043  AA vs. AG   0.89 (0.75–1.05) 0.17 
 GG 144 335 1.05 (0.83–1.32) 0.685  AA vs. GG   0.92 (0.71–1.21) 0.56 
 AG + GG   1.14 (0.99–1.31) 0.067  AA vs. (AG + GG)   0.90 (0.76–1.05) 0.18 
rs372883 AA 344 775   0.952 Per allele 1,059 860 0.90 (0.80–1.00) 0.06 
 AG 613 1,446 0.97 (0.82–1.14) 0.7  AA vs. AG   0.90 (0.74–1.10) 0.31 
 GG 288 650 0.98 (0.80–1.19) 0.83  AA vs. GG   0.81 (0.65–1.01) 0.06 
 AG + GG   0.97 (0.83–1.14) 0.715  AA vs. (AG + GG)   0.87 (0.72–1.04) 0.13 
rs5768709 AA 510 951   0.0378 Per allele 1,083 883 0.97 (0.86–1.09) 0.57 
 AG 580 1,221 0.92 (0.79–1.07) 0.277  AA vs. AG   0.98 (0.82–1.16) 0.79 
 GG 180 408 0.86 (0.69–1.07) 0.166  AA vs. GG   0.93 (0.73–1.19) 0.56 
 AG + GG   0.90 (0.78–1.04) 0.17  AA vs. (AG + GG)   0.96 (0.82–1.14) 0.67 
rs6464375 GG 1124 2,522   0.3589 Per allele 1,111 904 1.26 (1.00–1.58) 0.05 
 AG 169 353 1.13 (0.92–1.38) 0.258  GG vs. AG   1.22 (0.96–1.55) 0.10 
 AA 6 9 1.83 (0.62–5.36) 0.271  GG vs. AA   3.26 (0.80–13.21) 0.10 
 AG + AA   1.14 (0.93–1.40) 0.199  GG vs. (AG + AA)   1.24 (0.98–1.57) 0.07 
rs708224 GG 449 919   0.7476 Per allele 1,104 900 1.11 (0.99–1.23) 0.08 
 AG 598 1,378 0.90 (0.77–1.05) 0.189  GG vs. AG   1.09 (0.91–1.30) 0.37 
 AA 246 501 1.07 (0.87–1.30) 0.524  GG vs. AA   1.23 (0.98–1.53) 0.07 
 AG + AA   0.94 (0.82–1.09) 0.441  GG vs. (AG + AA)   1.12 (0.95–1.33) 0.17 
rs9502893 AA 452 899   0.5069 Per allele 1,084 880 1.00 (0.90–1.11) 0.98 
 AG 555 1,338 0.84 (0.71–0.98) 0.027  AA vs. AG   0.98 (0.79–1.21) 0.85 
 GG 265 536 0.98 (0.81–1.19) 0.861  AA vs. GG   1.07 (0.90–1.26) 0.44 
 AG + GG   0.88 (0.76–1.02) 0.087  AA vs. (AG + GG)   1.13 (0.94–1.35) 0.20 

aNumbers may not add up to 100% of subjects due to genotyping failure. All samples that did not give a reliable result in the first round of genotyping were resubmitted to up to 2 additional rounds of genotyping. Data points that were still not filled after this procedure were left blank.

bAll analyses were adjusted for age, gender, and nationality. Significant associations (P < 0.05) are reported in bold.

cAll analyses were adjusted for age, gender, TNM stage, and nationality.

View Large

We previously replicated the majority of loci identified by the first GWAS on pancreatic cancer in a subset of the cases and controls used in the present study (7). GWAS on PDAC risk in the Japanese (4) and Chinese population (5) yielded 3 new loci on chromosomes 6p25.3 (rs9502893, upstream of FOXQ1), 12p11.21 (rs708224, in the second intron of BICD1), and 7q36.2 (rs6464375, in the first intron of DPP6) and 5 novel susceptibility loci at chromosomes 21q21.3 (rs372883, in the BACH1 gene), 5p13.1 (rs2255280, in the DAB2 gene), 21q22.3 (rs1547374, upstream of TFF1 gene), 22q13.32 (rs5768709), and 10q26.11 (rs12413624). The last 2 SNPs are not located in the immediate vicinity of any gene. We attempted to validate 7 of the novel hits in an independent cohort of different ethnicity. We did not analyze rs2255280 because it is monomorphic in Caucasians (http://www.ncbi.nlm.nih.gov/projects/SNP/snp_ref.cgi?rs=2255280). Such validation is necessary because attempts to generalize genetic associations across ethnicities have had mixed results. Also, the incidence of pancreatic cancer is substantially different among populations of distinct ancestry (5), possibly reflecting differences in genetic susceptibility. We had more than 95% statistical power to detect the reported associations; nevertheless, we observed none. Our results highlight the genetic differences across human populations and illustrate the importance of evaluating PDAC risk markers across ethnic groups.

No potential conflicts of interest were disclosed.

Conception and design: D. Campa, C. Rizzato, J.P. Neoptolemos, M.W. Büchler, F. Canzian

Development of methodology: J.P. Neoptolemos, M.W. Büchler

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A.S. Bauer, G. Capurso, E. Costello, R. Talar-Wojnarowska, K. Jamroziak, R. Pezzilli, M. Gazouli, K.-T. Khaw, T.J. Key, F. Bambi, B. Mohelnikova-Duchonova, A. Heller, S. Landi, L. Vodickova, G. Theodoropoulos, P. Bugert, P. Vodicka, J.D. Hoheisel, G. Delle Fave, J.P. Neoptolemos, P. Soucek, M.W. Büchler, N. Giese

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): D. Campa, C. Rizzato, J. Werner, G. Capurso, G. Delle Fave, P. Soucek, M.W. Büchler

Writing, review, and/or revision of the manuscript: D. Campa, C. Rizzato, J. Werner, G. Capurso, E. Costello, M. Gazouli, K.-T. Khaw, T.J. Key, L. Vodickova, P. Bugert, P. Vodicka, J.D. Hoheisel, G. Delle Fave, J.P. Neoptolemos, P. Soucek, M.W. Büchler, F. Canzian

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): C. Rizzato, A.S. Bauer, J. Werner, R. Talar-Wojnarowska, M. Gazouli, K.-T. Khaw, B. Mohelnikova-Duchonova, A. Heller, S. Landi, L. Vodickova, P. Bugert, P. Vodicka, J.D. Hoheisel, J.P. Neoptolemos, M.W. Büchler, N. Giese

Study supervision: M.W. Büchler, F. Canzian

1.
Ferlay
 
J
,
Parkin
 
DM
,
Steliarova-Foucher
 
E
. 
Estimates of cancer incidence and mortality in Europe in 2008
.
Eur J Cancer
 
2010
;
46
:
765
–
81
.
Google Scholar
Crossref
Search ADS
PubMed
 
2.
Amundadottir
 
L
,
Kraft
 
P
,
Stolzenberg-Solomon
 
RZ
,
Fuchs
 
CS
,
Petersen
 
GM
,
Arslan
 
AA
, et al 
Genome-wide association study identifies variants in the ABO locus associated with susceptibility to pancreatic cancer
.
Nat Genet
 
2009
;
41
:
986
–
90
.
Google Scholar
Crossref
Search ADS
PubMed
 
3.
Petersen
 
GM
,
Amundadottir
 
L
,
Fuchs
 
CS
,
Kraft
 
P
,
Stolzenberg-Solomon
 
RZ
,
Jacobs
 
KB
, et al 
A genome-wide association study identifies pancreatic cancer susceptibility loci on chromosomes 13q22.1, 1q32.1 and 5p15.33
.
Nat Genet
 
2010
;
42
:
224
–
8
.
Google Scholar
Crossref
Search ADS
PubMed
 
4.
Low
 
SK
,
Kuchiba
 
A
,
Zembutsu
 
H
,
Saito
 
A
,
Takahashi
 
A
,
Kubo
 
M
, et al 
Genome-wide association study of pancreatic cancer in Japanese population
.
PLoS One
 
2010
;
5
:
e11824
.
Google Scholar
Crossref
Search ADS
PubMed
 
5.
Wu
 
C
,
Miao
 
X
,
Huang
 
L
,
Che
 
X
,
Jiang
 
G
,
Yu
 
D
, et al 
Genome-wide association study identifies five loci associated with susceptibility to pancreatic cancer in Chinese populations
.
Nat Genet
 
2012
;
44
:
62
–
6
.
Google Scholar
Crossref
Search ADS
 
6.
Campa
 
D
,
Rizzato
 
C
,
Capurso
 
G
,
Giese
 
N
,
Funel
 
N
,
Greenhalf
 
W
, et al 
Genetic susceptibility to pancreatic cancer and its functional characterisation: the PANcreatic Disease ReseArch (PANDoRA) consortium
.
Dig Liver Dis
. 
2012 Nov 30
.
[Epub ahead of print]
.
Google Scholar
 
7.
Rizzato
 
C
,
Campa
 
D
,
Giese
 
N
,
Werner
 
J
,
Rachakonda
 
PS
,
Kumar
 
R
, et al 
Pancreatic cancer susceptibility loci and their role in survival
.
PLoS One
 
2011
;
6
:
e27921
.
Google Scholar
Crossref
Search ADS
PubMed
 
©2012 American Association for Cancer Research.
2012