Background: Infection with Helicobacter pylori is the leading risk factor for noncardia gastric cancer, yet its influence on prognosis of gastric cancer is largely unknown. Thus, exploring the role of Helicobacter pylori (H. pylori) in survival could lead to a greater understanding of the high mortality associated with gastric cancer.

Methods: Seropositivity to 15 H. pylori antigens was assessed using a multiplex assay in two prospective cohorts, the Shanghai Men's Health Study and the Shanghai Women's Health Study. Multivariable-adjusted Cox proportional hazards regression was used to examine the association between prediagnostic H. pylori antigen levels and gastric cancer–specific survival.

Results: Prediagnostic levels of H. pylori serum antibodies that were previously associated with gastric cancer incidence in this population were not associated with gastric cancer survival, whether assessed in a 6-antigen panel [HR = 1.29; 95% confidence interval (CI), 0.78–2.13 for men; HR = 0.93; 95% CI, 0.57–1.52 for women], focused on CagA+H. pylori (HR = 0.73; 95% CI, 0.44–1.20 forwomen; HR = 1.27; 95% CI, 0.70–2.31 for men) or on the high-risk biomarkers of dual Omp and HP 0305 seropositivity (HR = 0.97; 95% CI, 0.72–1.30 for women; HR = 1.37; 95% CI, 0.97–1.94 for men).

Conclusions: Prediagnostic H. pylori antigen levels are not associated with gastric cancer survival in East Asian populations.

Impact: Identification of additional factors associated with gastric cancer survival would further our understanding of the high mortality associated with this malignancy. Cancer Epidemiol Biomarkers Prev; 27(3); 342–4. ©2017 AACR.

Worldwide, gastric cancer is the fifth most common malignancy and the third leading cause of cancer-related death (1). More than half of the world's total incident gastric cancers occur in East Asia, the majority of which are in China. Although the risk for gastric cancer is multifactorial, the single greatest risk factor for the development of gastric cancer is infection with Helicobacter pylori (H. pylori). Despite the fact that the causal association between H. pylori infection and gastric cancer incidence is well established, associations between H. pylori and gastric cancer survival are unknown. Previously, we identified novel H. pylori blood biomarkers that are associated with gastric cancer risk (2). Although these and other biomarkers have shown promise as potential predictors of gastric cancer risk, there is no known biomarker associated with gastric cancer survival beyond clinical characteristics. Thus, we sought to determine whether H. pylori seropositivity could act as one such potential marker.

Study population

The Shanghai Men's and Women's Health Studies (SMHS and SWHS, respectively) have been described previously (3, 4). This study was approved by the Institutional Review Boards of Vanderbilt University (Nashville, TN), German Cancer Research Center (Heidelberg, Germany), and the Shanghai Cancer Institute (Shanghai, China).

H. pylori serology

Plasma samples from patients diagnosed with noncardia gastric cancer (ICDO codes C16.1-C16.6, C16.8, and C16.9) were sent to the DKFZ for analysis as described previously (2, 5, 6).

Statistical analysis

Overall survival time was calculated from date of cancer diagnosis to date of death by any cause. Patients were censored by date of last linkage to death certificates (December 31, 2014) or date of last in-person follow-up, whichever was the latest date. Analyses were conducted using Cox proportional hazards regression to calculate HRs and 95% CIs for seropositivity to individual or grouped H. pylori antigens with gastric cancer survival, adjusted for the time interval between blood collection and diagnosis, age at diagnosis, family income, and tumor–node–metastasis (TNM) stage. Stratified analyses by age at diagnosis, tumor stage, time interval between blood draw and diagnosis, and treatment by surgery were performed to determine whether the association between H. pylori and survival differed by clinical characteristics. Validation of the assumption of proportionality of the hazard was checked by creating an interaction term between predictor(s) and event time in the model and tested using the Wald χ2 test.

We sought to determine whether baseline seropositivity to select H. pylori antigens previously associated with gastric cancer incidence in East Asia (2) were also associated with gastric cancer survival. We began by examining whether an increasing number of seropositive results to 6 H. pylori antigens associated with gastric cancer risk in this population (Omp, HP0305, HyuA, HpaA, CagA, VacA; refs. 2, 7) was also associated with gastric cancer survival, comparing individuals seropositive with either 4–5 antigens or 6+ antigens to individuals seropositive to 3 or fewer antigens (Table 1). As there was no association between the quantity of antigens and relative survival, we next narrowed our focus to known individual H. pylori antigens associated with gastric cancer incidence. There were no associations between prediagnostic seropositivity to any previously identified single or combination of H. pylori antigens and gastric cancer survival. Furthermore, neither age at diagnosis, tumor stage, time interval between sample collection, and diagnosis nor history of gastric surgery modified the association between seropositivity to H. pylori antigens and gastric cancer survival (Table 2).

Table 1.

Noncardia gastric cancer survival by seroprevalence to the individual H. pylori antigens in the SWHS and SMHS

Female (n = 270)Male (n = 285)
SerostatusNo. of deaths, n (%)HRa (95% CI)No. of deaths, n (%)HRa (95% CI)
H. pylori Serostatus 
 0–3 Antigens 63 (23.3) 1.00 (Ref) 42 (14.7) 1.00 (Ref) 
 4–5 Antigens 171 (63.4) 0.84 (0.60–1.19) 178 (62.5) 1.40 (0.91–2.14) 
 6+ Antigens 36 (13.3) 0.93 (0.57–1.52) 65 (22.8) 1.29 (0.78–2.13) 
 Ptrend  0.63  0.37 
H. pylori and CagA 
 H. pylori− and/or CagA− 24 (8.9) 1.00 (Ref) 18 (6.3) 1.00 (Ref) 
 H. pylori+ and CagA+ 246 (91.1) 0.73 (0.44–1.20) 267 (93.7) 1.27 (0.70–2.31) 
Omp and HP 0305 
 Omp− and/or HP 0305− 102 (37.8) 1.00 (Ref) 68 (23.9) 1 (Ref) 
 Omp+ and HP 0305+ 168 (62.2) 0.97 (0.72–1.30) 217 (76.1) 1.37 (0.97–1.94) 
Female (n = 270)Male (n = 285)
SerostatusNo. of deaths, n (%)HRa (95% CI)No. of deaths, n (%)HRa (95% CI)
H. pylori Serostatus 
 0–3 Antigens 63 (23.3) 1.00 (Ref) 42 (14.7) 1.00 (Ref) 
 4–5 Antigens 171 (63.4) 0.84 (0.60–1.19) 178 (62.5) 1.40 (0.91–2.14) 
 6+ Antigens 36 (13.3) 0.93 (0.57–1.52) 65 (22.8) 1.29 (0.78–2.13) 
 Ptrend  0.63  0.37 
H. pylori and CagA 
 H. pylori− and/or CagA− 24 (8.9) 1.00 (Ref) 18 (6.3) 1.00 (Ref) 
 H. pylori+ and CagA+ 246 (91.1) 0.73 (0.44–1.20) 267 (93.7) 1.27 (0.70–2.31) 
Omp and HP 0305 
 Omp− and/or HP 0305− 102 (37.8) 1.00 (Ref) 68 (23.9) 1 (Ref) 
 Omp+ and HP 0305+ 168 (62.2) 0.97 (0.72–1.30) 217 (76.1) 1.37 (0.97–1.94) 

Abbreviation: CagA, cytotoxin associated gene A.

aMultivariable-adjusted HRs adjusted for time interval between sample collection to diagnosis, age at diagnosis, family income (yuan per capita per year), education, surgery, and TNM stage were estimated by the proportional hazards model.

View Large
Table 2.

Noncardia gastric cancer survival by H. pylori infection in the SWHS and SMHS

Multivariate HRs
H. pylori and CagAa positiveOmp and HP 0305b positiveNo. of seropositive results to six H. pylori proteinsc
HR (95% CI)HR (95% CI)4–5 vs. 0–36 vs. 0–3
Variablen, noncardia gastric cancer–specific deaths/total casesHR (95% CI)HR (95% CI)
Age at diagnosis, years 
 <68 179/277 1.03 (0.58–1.83) 1.11 (0.80–1.54) 1.26 (0.85–1.86) 1.09 (0.66–1.79) 
 ≥68 203/278 0.71 (0.43–1.19) 1.17 (0.85–1.60) 0.80 (0.55–1.17) 0.98 (0.61–1.58) 
Tumor stage 
 I/II 110/242 1.24 (0.64–2.43) 1.40 (0.90–2.18) 1.44 (0.84–2.44) 1.39 (0.72–2.69) 
 III 70/82 0.49 (0.18–1.32) 1.32 (0.77–2.27) 1.04 (0.51–2.13) 0.94 (0.40–2.18) 
 IV 89/93 0.77 (0.31–1.95) 0.83 (0.51–1.33) 0.97 (0.54–1.73) 1.00 (0.49–2.06) 
 Unknown 113/138 0.93 (0.47–1.82) 1.03 (0.70–1.51) 0.76 (0.47–1.23) 0.79 (0.40–1.59) 
Time interval between sample collection to diagnosis, years 
 <2 75/97 1.32 (0.61–2.83) 0.84 (0.50–1.40) 1.38 (0.75–2.53) 0.86 (0.40–1.85) 
 ≥2 307/458 0.75 (0.49–1.16) 1.18 (0.92–1.52) 0.98 (0.72–1.33) 1.10 (0.75–1.62) 
Surgery 
 No 114/120 0.73 (0.35–1.52) 0.97 (0.65–1.47) 0.70 (0.42–1.18) 1.09 (0.59–2.03) 
 Yes 268/435 0.97 (0.62–1.52) 1.13 (0.86–1.48) 1.22 (0.88–1.69) 1.01 (0.67–1.51) 
Multivariate HRs
H. pylori and CagAa positiveOmp and HP 0305b positiveNo. of seropositive results to six H. pylori proteinsc
HR (95% CI)HR (95% CI)4–5 vs. 0–36 vs. 0–3
Variablen, noncardia gastric cancer–specific deaths/total casesHR (95% CI)HR (95% CI)
Age at diagnosis, years 
 <68 179/277 1.03 (0.58–1.83) 1.11 (0.80–1.54) 1.26 (0.85–1.86) 1.09 (0.66–1.79) 
 ≥68 203/278 0.71 (0.43–1.19) 1.17 (0.85–1.60) 0.80 (0.55–1.17) 0.98 (0.61–1.58) 
Tumor stage 
 I/II 110/242 1.24 (0.64–2.43) 1.40 (0.90–2.18) 1.44 (0.84–2.44) 1.39 (0.72–2.69) 
 III 70/82 0.49 (0.18–1.32) 1.32 (0.77–2.27) 1.04 (0.51–2.13) 0.94 (0.40–2.18) 
 IV 89/93 0.77 (0.31–1.95) 0.83 (0.51–1.33) 0.97 (0.54–1.73) 1.00 (0.49–2.06) 
 Unknown 113/138 0.93 (0.47–1.82) 1.03 (0.70–1.51) 0.76 (0.47–1.23) 0.79 (0.40–1.59) 
Time interval between sample collection to diagnosis, years 
 <2 75/97 1.32 (0.61–2.83) 0.84 (0.50–1.40) 1.38 (0.75–2.53) 0.86 (0.40–1.85) 
 ≥2 307/458 0.75 (0.49–1.16) 1.18 (0.92–1.52) 0.98 (0.72–1.33) 1.10 (0.75–1.62) 
Surgery 
 No 114/120 0.73 (0.35–1.52) 0.97 (0.65–1.47) 0.70 (0.42–1.18) 1.09 (0.59–2.03) 
 Yes 268/435 0.97 (0.62–1.52) 1.13 (0.86–1.48) 1.22 (0.88–1.69) 1.01 (0.67–1.51) 

Abbreviation: CagA, cytotoxin associated gene A.

aAdjusted for time interval between sample collection to diagnosis, age at diagnosis, sex, family income (yuan per capita per year), education, surgery, and TNM stage (excluding the variable used for strata analysis).

bAdjusted for time interval between sample collection to diagnosis, age at diagnosis, sex, family income (yuan per capita per year), education, surgery, and TNM stage (excluding the variable used for strata analysis).

cAdjusted for time interval between sample collection to diagnosis, age at diagnosis, sex, family income (yuan per capita per year), education, surgery, and TNM stage (excluding the variable used for strata analysis).

View Large

Our previous work has shown that antibody levels to select H. pylori antigens could be useful biomarkers for predicting gastric cancer risk among East Asian populations, including those from the SMHS and SWHS. Therefore, we sought to examine whether the seropositivity to a quantity of antigens or select high cancer-risk antigens were associated with relative gastric cancer survival. We found that neither the high cancer-risk antigens (including CagA, HP0305, and Omp) nor any of the other 15 tested antigens assessed prior to cancer diagnosis were useful in predicting gastric cancer survival within this population.

The strength of our study is that we have accessed an exceedingly high gastric cancer-risk population to test our hypothesis that prediagnostic seropositivity to H. pylori antigens associated with incidence may also be associated with survival. In our cohort, we have a strong follow-up rate and a wealth of both baseline demographic and lifestyle information that have helped validate other studies of demographic factors associated with survival conducted within this region (8). A limitation of our study is that the levels of H. pylori antibodies and other characteristics examined were obtained at the time of enrollment, well prior to cancer onset, which may be less relevant for prognosis than status at cancer diagnosis.

To conclude, although prediagnostic H. pylori antigen levels are associated with gastric cancer incidence in this population, they are not associated with gastric cancer survival.

No potential conflicts of interest were disclosed.

Conception and design: T. Wang, Y.-T. Gao, B.-T. Ji, M. Pawlita, M. Epplein

Development of methodology: T. Wang, M. Pawlita, T. Waterboer, M. Epplein

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): Y.-B. Xiang, Y.-T. Gao, M. Pawlita, T. Waterboer, W. Zheng, X.-O. Shu

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M.G. Varga, T. Wang, H. Cai, M. Pawlita, X.-O. Shu, M. Epplein

Writing, review, and/or revision of the manuscript: M.G. Varga, T. Wang, H. Cai, Y.-B. Xiang, Y.-T. Gao, B.-T. Ji, T. Waterboer, W. Zheng, X.-O. Shu, M. Epplein

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): Y.-B. Xiang, Y.-T. Gao, B.-T. Ji, M. Epplein

Study supervision: M. Pawlita, W. Zheng, M. Epplein

Other (principal investigator of the parent study): W. Zheng

Other (designed, lead, and obtained funding for the parent studies): X.-O. Shu

This work was supported by NCI R01CA174853 (to M. Epplein), the Molecular and Genetic Epidemiology of Cancer Fellowship R25CA160056 (to X.-O. Shu, principal investigator; M.G. Varga, fellow), the Cancer Control Education Program 2T32CA057726-26 (to K. Ribisl, principal investigator; to M.G. Varga, fellow), R37 CA070867, UM1 CA182910 (to W. Zheng), R01 CA082729, and UM1 CA 173640 (to X.-O. Shu).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Ferlay
 
J
,
Soerjomataram
 
I
,
Dikshit
 
R
,
Eser
 
S
,
Mathers
 
C
,
Rebelo
 
M
, et al 
Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012
.
Int J Cancer
 
2015
;
136
:
E359
–
86
.
Google Scholar
Crossref
Search ADS
PubMed
 
2.
Cai
 
H
,
Ye
 
F
,
Michel
 
A
,
Murphy
 
G
,
Sasazuki
 
S
,
Taylor
 
PR
, et al 
Helicobacter pylori blood biomarker for gastric cancer risk in East Asia
.
Int J Epidemiol
 
2016
;
45
:
774
–
81
.
Google Scholar
Crossref
Search ADS
PubMed
 
3.
Yang
 
G
,
Xiang
 
YB
,
Zheng
 
W
,
Xu
 
WH
,
Zhang
 
X
,
Li
 
HL
, et al 
Body weight and weight change in relation to blood pressure in normotensive men
.
J Hum Hypertens
 
2007
;
21
:
45
–
52
.
Google Scholar
Crossref
Search ADS
PubMed
 
4.
Zhang
 
X
,
Shu
 
XO
,
Gao
 
YT
,
Yang
 
G
,
Li
 
Q
,
Li
 
H
, et al 
Soy food consumption is associated with lower risk of coronary heart disease in Chinese women
.
J Nutr
 
2003
;
133
:
2874
–
8
.
Google Scholar
Crossref
Search ADS
PubMed
 
5.
Michel
 
A
,
Waterboer
 
T
,
Kist
 
M
,
Pawlita
 
M
. 
Helicobacter pylori multiplex serology
.
Helicobacter
 
2009
;
14
:
525
–
35
.
Google Scholar
Crossref
Search ADS
PubMed
 
6.
Shakeri
 
R
,
Malekzadeh
 
R
,
Nasrollahzadeh
 
D
,
Pawlita
 
M
,
Murphy
 
G
,
Islami
 
F
, et al 
Multiplex H. pylori serology and risk of gastric cardia and noncardia adenocarcinomas
.
Cancer Res
 
2015
;
75
:
4876
–
83
.
Google Scholar
Crossref
Search ADS
PubMed
 
7.
Epplein
 
M
,
Signorello
 
LB
,
Zheng
 
W
,
Peek
 
RM
 Jr.,
Michel
 
A
,
Williams
 
SM
, et al 
Race, African ancestry, and Helicobacter pylori infection in a low-income United States population
.
Cancer Epidemiol Biomarkers Prev
 
2011
;
20
:
826
–
34
.
Google Scholar
Crossref
Search ADS
PubMed
 
8.
Zheng
 
L
,
Wu
 
C
,
Xi
 
P
,
Zhu
 
M
,
Zhang
 
L
,
Chen
 
S
, et al 
The survival and the long-term trends of patients with gastric cancer in Shanghai, China
.
BMC Cancer
 
2014
;
14
:
300
.
Google Scholar
Crossref
Search ADS
PubMed
 
©2017 American Association for Cancer Research.
2017
American Association for Cancer Research.