Community-Based Gastric Cancer Screening Coupled With a National Colorectal Cancer Screening Program: Baseline Results

Abstract

Gastric cancer (GC) remains a pressing international health problem, with more than 1 million new cases annually.1Bray F. et al.CA Cancer J Clin. 2018; 68: 394-424Crossref PubMed Scopus (39142) Google Scholar Because Helicobacter pylori is an initiator, a screen-and-treat strategy provides a possible solution.2Graham D.Y. Gastroenterology. 2015; 148: 719-731Abstract Full Text Full Text PDF PubMed Scopus (292) Google Scholar,3Sugano K. et al.Gut. 2015; 64: 1353-1367Crossref PubMed Scopus (738) Google Scholar However, such a policy has rarely been realized.4Chiang T.H. et al.Gut. 2021; 70: 243-250Google Scholar Obstacles include difficulties in integrating policy into the national health care priorities and a lack of infrastructure to support these services. Colorectal cancer (CRC) screening using the fecal immunochemical testing (FIT) has been used globally.1Bray F. et al.CA Cancer J Clin. 2018; 68: 394-424Crossref PubMed Scopus (39142) Google Scholar H pylori stool antigen (HPSA) testing can be done in conjunction with FIT. A randomized controlled trial was conducted by inviting a population at risk to screen for both GC and CRC with fecal tests. In this report, we show how a nationwide FIT screening program can be used as a platform for initiating a community-based GC screening and report the preliminary results upon the completion of baseline screening. The Taiwanese government has already offered asymptomatic adults aged 50–69 years a biennial service screening for CRC with FIT.5Lee Y.C. Hsu C.Y. et al.Int J Epidemiol. 2019; 48: 538-548Crossref PubMed Scopus (16) Google Scholar We used this platform to offer the screening and eradication of H pylori. The methodology is provided in the Supplementary Materials. Eligible individuals were randomized 1:1 into the HPSA plus FIT arm (also called the added-on arm) or the FIT-only arm (each with 120,000 individuals) before invitation. Those declining the invitation after 3 attempts at contact were considered nonparticipants. Participants testing positive via HPSA and FIT were referred for eradication treatment and colonoscopy, respectively. The long-term outcomes would include the incidence rates of GC and CRC. During 2014–2017, 63,508 individuals were invited to the HPSA plus FIT arm and 88,995 to the FIT-only arm; the corresponding numbers of participants (rate) were 31,497 (49.6%) and 31,777 (35.7%), respectively (Supplementary Figure 1). The administration of an extra HPSA test increased the participation rate by 13.9% (95% confidence interval [CI], 13.4–14.4). Interaction occurred between the uptake of screening and participant characteristics. Compared with FIT-only participants, those in the add-on arm were more likely to attend the first screening and to be smokers, alcohol drinkers, or betel nut chewers; to use antihypertension or nonsteroidal anti-inflammatory drugs; and to have first-degree relatives with H pylori infection, peptic ulcer, or GC (P < .001) (Supplementary Table 1). The positivity rate for HPSA was 38.5%. Of 12,142 participants with positive results, 8809 (72.5%) were referred for eradiation medication, with 8664 (98.4%) completing treatment. For the FIT screening, the positivity rate (7.1% vs 7.0%, P = .58) and colonoscopic referral rate (76.1% vs 79.8%, P = .32) were similar between the 2 arms. The eradication rates of H pylori were 91.9% (95% CI, 91.1–92.7) and 97.6% (95% CI, 97.1–98.0) according to the intention-to-treat and per-protocol analyses, respectively. The reinfection rate was estimated as 3.0 (95% CI, 0.5–16.0) per 1000 person-years. The primary resistance rates to amoxicillin, clarithromycin, metronidazole, levofloxacin, and tetracycline were 2.0%, 14.3%, 24.4%, 23.8%, and 4.5%, respectively. Among participants who received upper endoscopy, 33.1% (1937/5858) were diagnosed with gastric ulcer, duodenal ulcer, gastric adenoma, or GC. Risk factors included smoking (adjusted odds ratio [aOR], 1.39; 95% CI, 1.28–1.52; P < .001), use of nonsteroidal anti-inflammatory drugs (aOR, 1.34; 95% CI, 1.19–1.51; P < .001), and use of antiplatelets or anticoagulants (aOR, 1.52; 95% CI, 1.25–1.84; P < .001). The colonoscopic adenoma detection rate in the add-on arm was higher than that in the FIT-only arm (31.5 vs 25.3 per 1000; adjusted rate ratio (RR), 1.31; 95% CI, 1.23–1.39; P < .001) (Table 1). In the HPSA plus FIT arm, H pylori carriers had a higher adenoma detection rate than noncarriers (adjusted RR, 1.15; 95% CI, 1.03–1.28; P = .01).Table 1Rates of the GC and Colorectal Neoplasms Between 2 Groups According to the Invited Persons and ParticipantsFIT + HPSAFIT-onlyGastric and Colorectal NeoplasmsnRate, per 1000nRate, per 1000Crude RR (95% CI)Adjusted RRaRRs were adjusted for age, sex, and township clustering. (95% CI)P ValueInvited persons63,50888,995GC180.3350.40.72 (0.41–1.27)0.74 (0.41–1.32).31CRC1382.21992.20.99 (0.79–1.22)1.00 (0.81–1.24).98Participants31,49731,777GC80.390.30.90 (0.35–2.32)0.91 (0.44–1.89).80CRC571.8541.71.06 (0.73–1.54)0.96 (0.66–1.40).84Colorectal adenoma99131.580525.31.24 (1.13–1.36)1.31 (1.23–1.39)<.001Advanced adenoma bAdvanced adenoma was defined as an adenoma measuring 10 mm or more in diameter or having a villous component or high-grade dysplasia.1655.21233.91.35 (1.07–1.71)1.54 (1.29–1.84)<.001Nonadvanced adenoma82626.268221.51.22 (1.11–1.35)1.26 (1.16–1.36)<.001a RRs were adjusted for age, sex, and township clustering.b Advanced adenoma was defined as an adenoma measuring 10 mm or more in diameter or having a villous component or high-grade dysplasia. Open table in a new tab The mean (median) follow-up periods for all invited individuals and participants were 3.00 (3.66) and 2.55 (3.13) years until the end of 2017, yielding 458,183 and 161,402 person-years, respectively. The incidence rate of GC was slightly lower in the add-on arm for both the invited (adjusted RR, 0.74; 95% CI, 0.41–1.32; P = .31) and participants (adjusted RR, 0.91; 95% CI, 0.44–1.89; P = .80) (Table 1). In both arms, the majority of participants diagnosed with GC were identified as having advanced-stage disease (approximately 80%). The cancer stage distribution was similar between participants and nonparticipants. The incidence rates of CRC were similar between the 2 groups in both the invited individuals (adjusted RR, 1.00; 95% CI: 0.81–1.24, P = .98) and participants (adjusted RR, 0.96; 95% CI, 0.66–1.40; P = .84). However, participants in both groups were more likely to be diagnosed with early-stage CRC (approximately 70%) than nonparticipants (approximately 37%). We showed that a population-based FIT screening program for CRC is a useful platform for GC screening and that the addition of HPSA increased participation in screening. HPSA testing may also improve GC outcomes, but the last point requires a longer follow-up interval to determine. The high proportion of early-stage CRC in both groups supported the effectiveness of FIT. Providing an additional HPSA test not only increased participation but also motivated a greater proportion of unscreened individuals and those with risky behaviors (eg, smoking and alcohol drinking) to participate. Because such individuals were more likely to have colorectal neoplasms, the colonoscopic adenoma detection rate in the HPSA plus FIT group was higher than that in the FIT-only group. Our study also showed an increased adenoma detection rate in H pylori carriers; mechanisms may involve changes in immune response, intestinal microbiome, and gastrin production.6Butt J. Gastroenterology. 2019; 156: 175-186Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar Although approximately one third of H pylori carriers have significant intragastric lesions, the detection of early-stage GC was similarly low for those with HPSA plus FIT, those with FIT only, or the nonparticipants, which did not justify the use of upper endoscopy. Priorities must be set by considering personal risk factors or by using additional tests.7Dinis-Ribeiro M. et al.Endoscopy. 2012; 44: 74-94Crossref PubMed Scopus (505) Google Scholar Image-enhanced endoscopy may improve the detection of early-stage GC, with histologic assessment to guide surveillance for patients with premalignant gastric lesions.8Gupta S. et al.Gastroenterology. 2020; 158: 693-702Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar This report used baseline data; long-term follow-up is needed to elucidate outcomes such as the preventive effect on GC incidence, identification of persons who benefit from endoscopic surveillance, changes in antibiotic resistance, and impact of H pylori eradication on CRC incidence. Nonetheless, baseline results provide strong evidence to support the applicability of this new strategy. The authors thank the following members of the Taiwan Community-Based Integrated Screening Group: Sam Li-Sheng Chen, Amy Ming-Fang Yen, Jean Ching-Yuan Fann, Sherry Yueh-Hsia Chiu, Chen-Yang Hsu, Shu-Lin Chuang, Kun-Ching Chou, Wei-Wen Su, Shih-Tien Chen, Chao-Sheng Liao, Yu-Min Lin, Hung-Chuen Chang, Tsung-Hui Hu, Yi-Jen Fang, and Chang-Chuan Chan. The authors also thank the following clinics and hospitals for providing care to the participants in this study: Chang Hong Medicine Clinic, Chang Yu Medicine Clinic, Chen Yen Hsueh Medicine Clinic, Chi Hsu Lien Clinic, Chia Ho Medicine Clinic, Chia Lin Clinic, Ching Hung Clinic, Huang Gung Chin Medicine and Pediatrics Clinic, Feng-An Group Clinic, Lu Tian En Clinic, Shih-Ren Clinic, Wang Chang Hao Medicine Clinic, Wang Jian Lung Medicine Clinic, Wu Shun Yu Medicine Clinic, Ye Teng Hsin Medicine Clinic, Chang-Hua Hospital, Chang Bing Show Chwan Memorial Hospital, Changhua Show Chwan Memorial Hospital, Changhua Christian Hospital, Changhua Christian Hospital–Erlin Branch, Changhua Christian Hospital–Lukang Branch, Han Ming Hospital, Yuan Rung Hospital, and Yuan Sheng Hospital. Yi-Chia Lee, MD, PhD (Conceptualization: Lead; Data curation: Lead; Formal analysis: Lead; Funding acquisition: Equal; Investigation: Equal; Methodology: Equal; Supervision: Lead; Writing – original draft: Lead; Writing – review & editing: Lead); Tsung-Hsien Chiang, MD (Conceptualization: Equal; Data curation: Equal; Investigation: Supporting); Han-Mo Chiu, MD (Conceptualization: Equal; Data curation: Equal; Funding acquisition: Equal; Resources: Equal); Ming-Shiang Wu, MD (Conceptualization: Equal; Project administration: Equal; Resources: Equal; Supervision: Equal); Yen-Po Yeh, MD (Conceptualization: Lead; Data curation: Lead; Funding acquisition: Lead; Investigation: Lead; Methodology: Lead; Project administration: Lead; Resources: Lead; Supervision: Lead); Tony Hsiu-Hsi Chen, PhD (Conceptualization: Lead; Data curation: Lead; Formal analysis: Lead; Project administration: Lead; Resources: Lead; Supervision: Lead; Writing – review & editing: Equal). This appendix has been provided by the authors to give readers additional information about their methods. To clarify the effect of additional HPSA on the uptake of FIT screening, we applied the Zelen design.1Zelen M. A new design for randomized clinical trials.N Engl J Med. 1979; 300: 1242-1245Crossref PubMed Scopus (789) Google Scholar The flow diagram is shown in Supplementary Figure 1. Using the population registry list provided by the Health Promotion Administration, we randomized the eligible individuals into 2 arms (HPSA + FIT and FIT only) before invitation. Our purpose was to reduce the psychosocial influences on their choice to participate in the FIT-only group in the presence of an alternative method for screening (HPSA + FIT). Among 22 cities and counties in Taiwan, we selected Changhua County for this study because this population was representative of the average burden of GC and CRC. With a population of 1.3 million, the age-standardized incidence rates of GC were 10.6 and 6.6 per 100,000 person-years for males and females, respectively. For CRC, the rates were 51.9 and 36.9 per 100,000 person-years for males and females, respectively.2National Cancer Registry. Available at: https://cris.hpa.gov.tw/pagepub/Home.aspx. Access on August 6, 2020.Google Scholar Of individuals aged 50–69 years, about 260,000 residents were eligible for the biennial FIT. During 2012–2013, a pilot program established the standards for invitation, testing, referral, and treatment.3Sugano K. et al.Gut. 2015; 64: 1353-1367Crossref PubMed Scopus (738) Google Scholar,4Chiang T.H. et al.Gut. 2021; 70: 243-250Google Scholar The screening process is shown online.5Changhua’s innovative two-in-one H. pylori screening.https://www.youtube.com/watch?v=ihMCia7fTYE&feature=youtu.beGoogle Scholar Individuals with any of the following criteria were excluded from the trial: (1) history of total gastrectomy, (2) pregnant women, (3) severe concurrent diseases or malignancy, and (4) having participated in the pilot program. The study protocol was approved by Ethics Committee of National Taiwan University Hospital, and written informed consent was obtained from participants (201205030RIB). The ClinicalTrials.gov registration identifier is NCT01741363. Starting in 2014, eligible persons who had been randomized into 2 arms were invited by the staff from 27 public health units in the 26 townships. Each unit was required to enroll approximately 8000 participants each year for each arm over a 4-year period according to the population size and capacity. Those who chose to participate received screening with HPSA (Easy 1-Step Test, Firstep Bioresearch, Inc) and/or FIT (OC Sensor, Eiken Chemical Co). The HPSA test was a qualitative test, with a sensitivity and specificity of 88% and 99%, respectively, for detecting H pylori.6Lee Y.C. Tseng P.H. Liou J.M. et al.Performance of a one-step fecal sample-based test for diagnosis of Helicobacter pylori infection in primary care and mass screening settings.J Formos Med Assoc. 2014; 113: 899-907Crossref PubMed Scopus (20) Google Scholar The FIT was a quantitative test with a cutoff of 20 mg hemoglobin/g feces; the sensitivity and specificity for detecting colorectal cancer were 79% and 94%, respectively.7Lee J.K. Liles E.G. Bent S. Levin T.R. Corley D.A. Accuracy of fecal immunochemical tests for colorectal cancer: systematic review and meta-analysis.Ann Intern Med. 2014; 160: 171Crossref PubMed Google Scholar In addition to the fecal testing, a structured questionnaire was used to collect data on social habits, medication use, and family history. Fecal samples were returned to the public health units, testing was performed at certified laboratories, and the results were reported to the participants by mail and/or telephone. Those who tested positive for HPSA or FIT were referred for eradication treatment or colonoscopic examination, respectively. Individuals with positive HPSA results were referred to 15 clinics and 9 hospitals to receive a 10-day sequential therapy consisting of esomeprazole 40 mg once daily and amoxicillin 1 g twice daily for days 1–5, followed by esomeprazole 40 mg once daily, clarithromycin 500 mg, and metronidazole 500 mg twice daily for days 6–10. Posttreatment H pylori status was determined by using HPSA approximately 6–8 weeks after treatment ended. Those for whom the initial treatment failed were retreated with a 10-day triple therapy of 40 mg esomeprazole once a day, 1 g amoxicillin twice a day, and 500 mg levofloxacin once a day. During the treatment course, telephone contact (3 times) was maintained to report compliance and any adverse events. Because our eligible population was middle aged or older, irreversible damage could occur.8Altayar O. Davitkov P. et al.Technical review on gastric intestinal metaplasia-epidemiology and risk factors.Gastroenterology. 2020; 158: 732-744Abstract Full Text Full Text PDF PubMed Scopus (24) Google Scholar In addition to the eradication therapy, upper endoscopy was considered at the discretion of physicians based on the symptoms, history, physical examination, and patient’s willingness after face-to-face counseling. The endoscopic procedure was performed in the usual fashion without a specific biopsy protocol. Surveillance was designed following the national guidelines.4Sheu B.S. Wu M.S. Chiu C.T. et al.Consensus on the clinical management, screening-to-treat, and surveillance of Helicobacter pylori infection to improve gastric cancer control on a nationwide scale.Helicobacter. 2017; 22: e12368Crossref Scopus (38) Google Scholar The presence of antibiotic-resistant H pylori was determined in the National Taiwan University Hospital with the minimum inhibitory concentration test. We hypothesized that H pylori eradication could reduce the GC incidence rate by approximately 50%, as shown in the Shandong trial.9Li W.Q. Zhang J.Y. Ma J.L. et al.Effects of Helicobacter pylori treatment and vitamin and garlic supplementation on gastric cancer incidence and mortality: follow-up of a randomized intervention trial.BMJ. 2019; 366: l5016Crossref PubMed Scopus (64) Google Scholar Results from the pilot program showed an H pylori prevalence rate of approximately 38% and a referral rate of approximately 70% after a positive HPSA result.3IARC Helicobacter pylori Working Group (2014). Available at: https://publications.iarc.fr/Book-And-Report-Series/Iarc-Working-Group-Reports/-Em-Helicobacter-Pylori-Em-Eradication-As-A-Strategy-For-Preventing-Gastric-Cancer-2014. Accessed on August 6, 2020.Google Scholar Given a power of 90% and a .05 1-sided type 1 error in rejecting the null hypothesis, at least 30,000 participants for each group with a follow-up period of 10 years were required to demonstrate a significant risk reduction in receiving HPSA plus FIT vs FIT only. For baseline characteristics, categorical data were expressed as percentages and compared by using the chi-square test; continuous data were expressed as mean (standard deviation) and compared by using the Student t test. We first evaluated the eradication rate of H pylori. All patients who received treatment were included in the intention-to-treat analysis. Those who did not take at least 80% of treatment medications or whose posttreatment H pylori status was unknown, were excluded from the per-protocol analysis. Second, in patients who received upper endoscopy, the presence of gastroduodenal lesions was evaluated, and the results were expressed as aORs with 95% CIs. The colonoscopic adenoma detection rate was defined as the number with adenomas divided by the number of participants. Third, the outcomes of GC and CRC were ascertained based on the databases of the National Cancer Registry, which is characterized by its high coverage (>99%; hospitals in Taiwan are instructed to report all cases of cancer) and high accuracy (ie, the percentage of death certificate–only cases was <1%).2National Cancer Registry. Available at: https://cris.hpa.gov.tw/pagepub/Home.aspx. Access on August 6, 2020.Google Scholar Histopathology was classified according to the criteria of the World Health Organization. We evaluated the outcomes for both the invited persons and participants. Between-group differences were reported as adjusted RRs with 95% CIs, adjusted for age, sex, and township clustering. All statistical analyses were performed by using SAS, version 9.4 (SAS Institute). All P values were 2 sided, and P values of <.05 were considered All authors had to the study data and and approved the Supplementary Table of the Invited Persons and + FIT = only = mean sex, were defined as those who the invitation, received the fecal test, and returned the stool mean sex, screening to FIT, habits, alcohol alcohol and betel nut were defined as having of at least twice per betel nut alcohol and betel nut were defined as having of at least twice per medication was defined as medication having been in 3 anti-inflammatory or of first-degree H pylori Gastric results, of of to pylori for pylori treatment to colonoscopy, for colonoscopy, Participants were defined as those who the invitation, received the fecal test, and returned the stool alcohol and betel nut were defined as having of at least twice per medication was defined as medication having been in 3 Open table in a new tab

Topics & Concepts