Skip to main content

Rapidly declining trend of signet ring cell cancer of the stomach may parallel the infection rate of Helicobacter pylori



Studies indicate that gastric cancer (GC) incidence has decreased, whereas signet ring cell carcinoma (SRC) incidence has increased. However, recent trends in GC incidence are unclear. We used our hospital cancer registry to evaluate the changes in the incidence of GC, SRC, and non-SRC (NSRC) over time in comparison to changes in the H. pylori infection rates over time.


We identified 2532 patients with GC enrolled in our registry between January 2007 and December 2018 and statistically analyzed SRC and NSRC incidence. The H. pylori infection rate in patients with SRC was determined by serum anti-H. pylori antibody testing, urea breath test, biopsy specimen culture, and immunohistochemical analysis (IHC) of gastric tissue. Additionally, genomic detection of H. pylori was performed in SRCs by extracting DNA from formalin-fixed paraffin-embedded gastric tissue and targeting 16S ribosomal RNA of H. pylori.


Overall, 211 patients had SRC (8.3%). Compared with patients with NSRC, those with SRC were younger (P <  0.001) and more likely to be female (P <  0.001). Time series analysis using an autoregressive integrated moving average model revealed a significant decrease in SRC (P <  0.001) incidence; NSRC incidence showed no decline. There was no difference in H. pylori infection prevalence between the SRC and NSRC groups. IHC and genomic methods detected H. pylori in 30 of 37 (81.1%) SRCs.


Reduction in H. pylori infection prevalence may be associated with the decrease in the incidence of SRC, which was higher than that of NSRC.

Peer Review reports


Gastric cancer (GC) is the fifth most frequently diagnosed cancer and the third leading cause of cancer-related death worldwide [1]. Histologically, GC is classified into intestinal and diffuse types [2]. The intestinal type is related to corpus-dominant gastritis with intestinal metaplasia, whereas the diffuse type usually originates from superficial pangastritis without atrophy [3]. The intestinal type is often associated with Helicobacter pylori (H. pylori) infection, while the diffuse type is more often associated with genetic abnormalities [4].

Signet ring cell (SRC) carcinoma is a form of adenocarcinoma whose histologic diagnosis is based on microscopic characteristics defined by the World Health Organization (WHO) [1]. SRC is classified as diffuse type and non-SRC (NSRC) is mostly classified as intestinal type according to Lauren’s classification [5]. SRC and NSRC are considered distinct biologic entities originating from different pathways of carcinogenesis [5, 6].

Epidemiologically, the worldwide decline in GC incidence has mainly been associated with a decrease in intestinal-type GC in western countries [7]. By contrast, the incidence of diffuse-type GC, particularly SRC, is reported to have increased [7, 8]. SRC represented 28–70% of GC in western countries [8,9,10]. Moreover, SRC incidence has significantly increased by 6.5% per year on average, representing an approximately 10-fold increase from 1973 to 2000 [8]. However, the trends in the incidence of SRC after 2000 have not been fully examined.

GC rates in Japan are one of the highest in the world, presumably because of the high H. pylori infection rate [11]. However, the prevalence of H. pylori infection has drastically decreased, i.e., from 80 to 90% in older generations born before around 1950 to < 10% currently in individuals aged < 20 years [12]. According to a previous study using Joinpoint regression analysis, the prevalence of H. pylori infection in subjects born between 1927 and 1949 decreased from 48.6 to 43.5%, with a decline of 0.2% per year. Subsequently, a rapid decline in the prevalence of H. pylori infection in those born between 1949 (43.5%) and 1961 (22.7%) was found, with a decline of 1.7% per year. Another decrease was observed between 1961 (22.7%) and 1988 (6.3%), with a decline of 0.6% per year. The drastic decline in the prevalence of H. pylori infection by birth year can be explained by the change in sanitary conditions during childhood, when H. pylori infection is predominantly acquired [13].

Treatment for H. pylori infection was developed in 2000; however, governmental health insurance plan coverage of this treatment was limited to patients with peptic ulcers. In February 2013, the indications for treatment were expanded to include chronic gastritis. Subsequently, it was estimated that the number of patients with successful H. pylori eradication drastically increased after 2013, doubling to > 1,300,000 from approximately 650,000 per year between 2001 and 2012 [14].

In 2006, the Japanese government began to encourage hospitals to create cancer registries. In 2013 the government changed the law to mandate these registries, and enforcement began in 2016. Our hospital voluntarily initiated a cancer registry in 2006, and since then we have accumulated information on all malignant neoplasms seen at our hospital. From 2007 to 2018, our registry enrolled 22,674 patients with various cancer types. Using the registry data, we evaluated the changes in the incidence of GC, SRC, and NSRC over time in comparison to changes in H. pylori infection rates over time using time series analysis.



Our hospital covers the central part of Kofu City (population 400,000) in the highland area of Mt. Fuji, which is 100 km west of Tokyo. Compared with the people in the metropolitan area of Tokyo, the people in our area tend to stay here for life, making it easier to obtain follow-up studies on the patients. Approval for this retrospective review study was obtained from the Institutional Review Board at Yamanashi Central Hospital. The requirement for written informed consent was waived by the institutional review board. Since clinical data and previously collected samples were used in this retrospective study, the use of an opt-out consent method was approved by the institutional review board.

Pathological confirmation of SRC and NSRC

Starting in January 2007, we registered all patients with histologically proven GC. By the end of 2018, 2532 patients had been enrolled, and all of them underwent surgery, endoscopic submucosal dissection, and/or biopsy for diagnosis and treatment. All SRC and NSRC cases were confirmed by pathological diagnosis. SRC was described according to the WHO classification, i.e., poorly cohesive tumor cells with prominent cytoplasmic mucin and a crescent-shaped eccentrically placed nucleus [15].

Clinical features

Age, sex, and clinical data (including dates of diagnosis, histopathological diagnosis, the Union for International Cancer Control TNM classification, types of treatment, and dates of death) of patients were all documented in the databases of the in-hospital cancer registry. With the help of the Japanese government and the recently established law, we obtained the mortality data of all 2532 patients. Background gastric mucosal atrophy was classified into four grades of severity (i.e., normal, mild, moderate, and marked) on the basis of the updated Sydney system by an expert pathologist (T. O.) using formalin-fixed and paraffin-embedded (FFPE) tissue [16].

Detection of H. pylori infection

Serological testing

Blood samples of patients with GC were collected and used for serology testing; anti-H. pylori antibodies were detected by latex agglutination turbidimetry (BML, Tokyo, Japan). Patients with an anti-H. pylori antibody titer > 10 U/mL were classified as H. pylori antibody-positive.

Urea breath test (UBT).

The patients ingested 13C-labeled urea (100 mg), which was converted to 13CO2 by the urease enzyme produced by H. pylori if the bacteria were present in the stomach (BML). The released 13CO2 diffused into the blood and was released from the lungs. The expired air was collected 20 min after 13C-labeled urea ingestion to measure the 13C/12C ratio. A positive result of the UBT was defined as a difference between baseline and test samples of > 2.5‰.

Bacterial culture

Gastric mucosa specimens were collected from patients by endoscopic biopsy. The processed specimens were smeared in Helicobacter agar media (Nissui Pharmaceutical, Tokyo, Japan) and horse blood agar media (Kyokuto Pharmaceutical Industrial, Tokyo, Japan) and cultivated at 37 °C in a microaerophilic environment.

Immunohistochemical detection

Immunohistochemical (IHC) detection was performed using 3-μm-thick serial sections of FFPE tissue from resections or biopsies. The sections were deparaffinized, and antigen activation was performed by heat treatment in ethylenediaminetetraacetic acid solution at pH 8.0. Protein expression was evaluated on the 3-μm-thick FFPE sections with rabbit polyclonal anti-H. pylori antibodies (diluted 1:10; Institute of Immunology, Tokyo, Japan) using the Ventana BenchMark ULTRA (Roche, Tucson, Arizona) [17, 18].

Genomic detection of H. pylori 16S ribosomal RNA (rRNA)

For detection of H. pylori, we obtained FFPE tissues from patients with SRC. DNA was extracted from 10-μm-thick sections of FFPE using FormaPure DNA (Beckman Coulter, USA) with magnetic beads according to the manufacturer’s instructions. DNA concentration was determined using the Nano Drop 2000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA). Primers were designed to amplify the H. pylori 16S rRNA region and generate 110 bp polymerase chain reaction (PCR) products. Primer sequences were: forward (5′-.CTGGAGAGACTAAGCCCTCC-3′); reverse (5′-ATTACTGACGCTGATTGTGC-3′) [