Findings from the first colorectal cancer screening among 103 542 individuals in Vietnam with systematic review of colorectal cancer screening programs in Asia-Pacific region

Abstract

Background

Colorectal cancer is a leading cancer incidence and cause of death worldwide and in Vietnam. Although screening is considered an effective measure to prevent and control colorectal cancer, there is no such effort in Vietnam.

Methods

Between 01 January 2018 and 31 October 2019, a population-based colorectal cancer screening program was conducted in Hanoi, Vietnam. A health advocacy campaign and follow-up phone calls were used to enroll residents aged ≥40 years old to complete an immunochemical-fecal occult blood testing. Positive immunochemical-fecal occult blood testing was followed by a colonoscopy. We also conducted a systematic review of the colorectal cancer screening programs in the Asia-Pacific region that used similar approach by searching Ovid Medline and PubMed databases.

Results

During study period, 103 542 individuals among 672 742 eligible residents attended the screening of whom 81.5% participants finished immunochemical-fecal occult blood testing test and the positive rate was 6.1%. The coverage rate for immunochemical-fecal occult blood testing test was 11.9%. Among 2278 individuals who underwent colonoscopy, 3.5% were histologically diagnosed with cancer, 17.8% with advanced adenomas, and 23.1% with non-advanced adenomas. Males had significantly higher detection rate of advanced adenomas, cancer or ≥ two polyps/tumor than females (P < 0.0001). The systematic review showed that in two-step modality (i.e. immunochemical-fecal occult blood testing/fecal immunochemical test and colonoscopy), the test positive was from 4.1 to 10.6%. Once colonoscopy was performed subsequently, the rate of cancer among positive participants was from 1.7 to 16.4% and that of advanced adenomas was from 7.1 to 23.1%.

Conclusion

We showed that the two-step modality is a promising strategy for colorectal cancer screening in Vietnam that might apply to similar settings with limited resources

This report described the first-ever population-based colorectal cancer screening program in 103 542 individuals in Vietnam, using immunochemical-fecal occult blood testing test and subsequent colonoscopy among those with positive test results of immunochemical-fecal occult blood testing.

Introduction

Colorectal cancer (CRC) is the third most common cancer in men and the second in women, and the second leading cause of cancer death worldwide. In 2018, there were over 1.8 million new cases and over 880 000 deaths due to CRC worldwide (1). In the general US population, 1 in 18 individuals will develop CRC over their lifetime, and 40% will die within 5 years of diagnosis, mainly due to diagnosis at a late stage (2,3).

In Vietnam, CRC ranked fifth among most common cancers in incidence and mortality in both sexes and the incidence has been increasing (4). In 2018, the age-standardized incidence rate of CRC in Vietnam was 13.4 per 100 000 population, and the age-standardized mortality rate due to CRC was 7.0 per 100 000 population (4). The colon cancer incidence in Ho Chi Minh City showed an increase from 7.5 per 100 000 population during 1995–1998 to 11.1 per 100 000 population during 2008–2012 in men and from 5.3 per 100 000 population during 1995–1998 period to 7.2 per 100 000 population during 2008–2012 period in women (5).

Most CRC cases arise from adenomas through the well-established adenoma-carcinoma sequence (6). Progression from adenoma to CRC takes years to decades, a process that accelerates in the presence of DNA mismatch repair defects. Due to the slow progression from detectable precancerous lesions to invasive cancer, early identification and management of pre-malignant lesions (such as adenomas) or pre-symptomatic individuals (i.e. early stage CRC) are important measures for a successful cancer control and prevention program of CRC. It is equally important to note that the multilevel national efforts of screening implementation following evidence-based guidelines have resulted in a decline in the CRC incidence and mortality in many countries, such as the United States (7).

Fecal occult blood testing (FOBT) is the first colorectal screening test studied in a prospective randomized clinical trial. For instance, the Minnesota Colon Cancer Control Study randomized 46 551 adults to one of three arms: (i) annual FOBTs, (ii) biennial FOBT screening and (iii) usual care in which the rehydrated guaiac test was used. With 13 years of follow-up, the annual screened arm had a 33% reduction in CRC mortality compared with the usual care group (8). To date, FOBT, stool DNA testing, flexible sigmoidoscopies, colonoscopies, CT colonographies, and, occasionally, barium enemas are recommended options for CRC screening program (9). A positive FOBT, immunochemical FOBT [iFOBT or fecal immunochemical test (FIT)], fecal DNA test or sigmoidoscopy warrants a follow-up diagnostic colonoscopy.

Colonoscopy, the most effective strategy to reduce both incidence and mortality for CRC, however, is expensive, carries risks such as perforation, has a limited national capacity to implement fully and there is considerable controversy regarding an appropriate surveillance interval for adenoma patients, particularly in low-resource settings (10–13). The American Cancer Society updated its CRC screening guideline in which screening options may be chosen based on individual risk, personal preference and access (14).

Currently, CRC screening programs are undertaken in most European countries, Canada, specific regions in North and South America, other Asia countries and Oceania (15). Nevertheless, in Vietnam, due to the limited infrastructure, resources and health literacy, a population-based CRC screening program has never been conducted. The current analysis reported major findings from the first population-based CRC screening program in six districts of Hanoi, Vietnam, targeting men and women aged 40 years or older. For comparison, we also systematically reviewed CRC screening programs in the Asia-Pacific which used similar approach to ours.

Methods

Population

The current screening program was conducted among 672 742 men and women who were 40 years of age and older and were living in six districts of Hanoi, including Hai Ba Trung, Hoang Mai, Gia Lam, Thuong Tin, Thanh Oai and Chuong My between 01 January 2018 and 31 October 2019 (Supplementary Fig. 1).

Screening program

Supplementary Fig. 2 shows the operational workflow of this screening program. The program was implemented in collaboration with Hanoi Health Department (HHD), District Health Centers (DHCs) and Commune Health Stations (CHSs). Advocacy campaigns were given once to each DHCs and CHSs. Government officials from different sectors, the health center staff, commune health collaborators (CHCs) and potential participants were invited to attend the campaigns. Wherever possible, the advocacy campaign was provided further to sub-ward/hamlet, the lowest or grassroots level in Vietnam administrative system to reach out to as many people as possible. The advocacy campaign covered the following topics: (a) The benefit of early CRC screening; (b) Screening tests; (c) Screening protocol/flowchart; (d) Self-administered forms, including (1) HHD form (i.e. age, gender, insurance information and abnormality of abdominal and/or gastro-intestines and history of colorectal polyps/cancer) and (2) Questionnaire (i.e. socio-demographics, alcohol drinking status, tea and coffee drinking status, smoking status, dietary, physical activity, medication history, disease history, anthropometric information and reproductive health/history, including menstruation period, pregnancies, contraceptive/hormone replacement therapy for women only) and (e) Point of contact. Attendees of these advocacy campaigns would then go back to the communities to invite people in their communities to get screened and give them procedure information.

Following the advocacy campaign, the iFOBT kits and print materials were distributed to participants and then collected by CHS staffs. Participants could also get the kits and print materials directly from the local health station and return the completed samples and forms/questionnaire by themselves. Print materials included screening program leaflet, sample collection infographic instruction, self-administered HHD form and questionnaire. The HHD form data were entered by CHS staff using the State Administration Management System for e-Government (or eSAMs software), whereas iFOBT samples, completed forms and questionnaires were then transferred to Vinmec Times City (Vinmec TC) Hospital, Hanoi, where the iFOBT analysis was performed. Subsequently, iFOBT results were sent to CHSs via eSAMs. Participants were informed their test results by CHS staff via phone. All participants with positive iFOBT were invited to have a colonoscopy at Vinmec TC Hospital.

iFOBT test and testing procedure

The automated iFOBT OC-Sensor Pledia instrument (Eiken Chemical LTD, Tokyo, Japan) was used to evaluate iFOBT samples. The instrument was calibrated using control standards provided by the manufacturer. A concentration of 100 ng Hb/ml buffer was set for the level of occult blood for a positive reaction, following the manufacturer’s recommendations. Per protocol, iFOBT samples were returned to CHS within 24 h since sample collection. CHS staff delivered the samples to Vinmec TC Hospital within the next 3 days for testing. Subsequently, samples were stored in a refrigerator (2–8°C) and were analyzed within 7 days since reception.

iFOBT OC-Sensor Pledia system works on the principle of latex agglutination immunoturbidimetry. The test uses antibodies specific to the globin moiety of human hemoglobin. Latex agglutination immunoturbidimetry was used to measure the quantity of antibody-globin complexes. The globin moiety of hemoglobin is prone to degradation by proteolytic enzymes of endogenous origin in the stomach and small intestine; therefore, iFOBT is specific to colorectal lesions. Unlike guaiac FOBT, iFOBTs do not require diet or medication restrictions (16).

Histopathologic evaluation

During the colonoscopy, all biopsies and removed polyps/adenomas were evaluated. CRC was histologically diagnosed using the International Classification Code, Tenth Revision, codes C18.8, C18.9, C19 and C20. Advanced adenomas were defined as any adenoma≥1 cm, high-grade dysplasia or with tubulovillous or villous histology, whereas non-advanced adenoma was defined as adenomas<1 cm without advanced histology (17). The anatomic location of the advanced neoplasia lesions was recorded. The proximal colon included the cecum, ascending, hepatic flexure and transverse colon, whereas distal colon included splenic flexure, descending, sigmoid colon and rectum.

Systematic review of colorectal screening program in Asia-Pacific region

Search strategy

Between June 2019 and December 2020, three investigators (M.V. N., T.T.T.V. and M. T. T.) conducted a systematic search to identify published studies up to 11 December 2020. We limited our search from 1995 to 2020, the period of use of iFOBT or FIT test in the screening program. Two databases (i.e. OvidMedline and PubMed) were searched using the following search terms: (colorectal cancer) AND (colon OR rectum) AND (screening) AND (Asia OR Pacific).

Study screening and selection criteria

We used published guidelines for reported observational studies (Strengthening the Reporting of Observational Studies in Epidemiology) (18) and systematic review and meta-analyses (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) (19) to evaluate studies to be included in this systematic review. Inclusion criteria were English language reports of original research in human subjects. We further excluded articles if they were review papers or case-report papers or experimental studies. All citations were independently reviewed by three investigators (M.V. N., T.T.T.V. and M. T. T.). Any discrepancies were resolved by discussion and consensus between the two investigators and in consultation with senior authors (T.C.T.D. and H.N.L).

Systematic review process

After the review process, data on the characteristics of the study were extracted. For each eligible study, information included in the table of characteristics were (i) first author’s name, (ii) year of publication, (iii) country of the study conducted, (iv) study setting and sample size, (v) sources of population, (vi) study period, (vii) age at the screening stage, (viii) methods/techniques or tools used for colorectal screening and (ix) main results (i.e. rates of positive tests, cancer, adenomas, diagnostic accuracy, etc.).

Statistical analysis

In our analysis, the coverage rate of the screening program was calculated as the proportion of those who took iFOBT tests over total number of persons who were 40 years of age or older or those who were eligible to attend the CRC screening program. The iFOBT positivity was calculated as the number of positive iFOBT tests over the number of iFOBT tests received. The rate of cancer and adenomas (both advanced and non-advanced adenomas) was calculated among those who underwent colonoscopy.

Means and standard deviation (SD) were calculated for continuous variables, whereas counts and proportions were computed for categorical variables. The t-test and χ² test were used to compare the distributions of continuous and categorical variables, respectively, between comparative groups. All statistical analyses were conducted using SAS, version 9.4 (SAS Institute Inc., Cary, NC).

Results

CRC screening program in Hanoi

Screening coverage rate and iFOBT positive rate

According to the 2017 population data of Hanoi from the General Statistics Office of Vietnam (20), there were 672 742 people aged 40 years and older in the six targeted districts. A total of 103 542 kits were distributed and 84 369 (81.5%) completed kits were returned. Among 84 369 returned kits, 80 330 kits were from individuals aged 40 years and older, making the overall screening coverage rate 11.9% (Table 1). The coverage rate was higher among women (14.5%) than among men (9.2%) (P < 0.0001). It also varied by age group, being lowest in those older than 70 years (6.7%) and higher in those aged 50–59 years old and 60–69 years old (14.7 and 14.9%, respectively) (P < 0.0001) (Table 1).

Table 1.

iFOBT coverage rate and iFOBT positivity in the current colorectal cancer screening program in Vietnam

	Population ≥ 40	iFOBT received (n, %)	iFOBT positivity (n, %)
Age
Age (Mean ± SD)		55.90 ± 9.0	57.23 ± 10.6
Total	672 742	80 330 (11.9)	4887 (6.1)
40–49	233 066	24 242 (10.4)	1346 (5.6)
50–59	191 949	28 261 (14.7)	1569 (5.6)
60–69	136 675	20 419 (14.9)	1331 (6.5)
≥70	111  052	7408 (6.7)	641 (8.7)
P value		<0.0001	<0.0001
Gender
Male	321 592	29 433 (9.2)	2170 (7.4)
Female	351 150	50 897 (14.5)	2717 (5.3)
P value		<0.0001	<0.0001

iFOBT: immunological fecal occult blood test; SD: standard deviation

Of 80 330 individuals who returned iFOBT and within screening targeted age range, 50 897 (63.4%) were women. The overall mean age ± SD were 55.9 ± 9.0; 30.2, 35.2, 25.4 and 9.2% being aged 40–49, 50–59, 60–69 and 70 years old or older, respectively (P < 0.0001).

Among 80 330 returned iFOBT, 4887 (or 6.1%) had a positive test. The respective iFOBT positive test among individuals aged 40–49, 50–59, 60–69 and 70 years old or older were 5.6, 5.6, 6.5 and 8.7% (P < 0.0001). The iFOBT positive rate was higher in men (7.4%) than in women (5.3%) (P < 0.0001) (Table 1).

All individuals with a positive iFOBT test, who did not actively contact the program staff themselves to schedule a follow-up colonoscopy as instructed, were reminded by phone. Among all contacted individuals, 581 were unreachable due to wrong phone number or not answering the phone (at least 3 phone calls were given to each of those persons) and 2028 individuals did not take follow-up colonoscopy in our screening program (Fig. 1).

Figure 1.

Work flow in the colorectal cancer screening program, Hanoi, Vietnam.

Colonoscopy and pathology findings

Among 2278 persons (or 46.6% of those with iFOBT positive results) who underwent colonoscopy, 485 persons (21.3%) were histologically diagnosed with advanced neoplasia, defined as advanced adenoma or adenocarcinoma. Men had higher rate of advanced neoplasia (33.8%) than women (11.5%) (P < 0.0001). In stratified analysis by age-specific among advanced neoplasia only, male in all age-specific groups had higher proportion of advanced neoplasia than women (Table 2). Individuals aged 40–49 years old had a lower proportion of advanced neoplasia than non-advanced neoplasia (i.e. 9.0 vs. 27.9%; P < 0.0001). On the other hand, the rate of advanced neoplasia was comparable to the rate of non-advanced neoplasia in individuals aged 70 years old or older (i.e. 35.3 vs. 40.8%, P > 0.05) (data not shown).

Table 2.

Distribution of advanced neoplasia by sex and age-specific groups among colonoscopy cases

	Colonoscopy cases, both sexes (n, % among iFOBT positive)	Advanced neoplasia, both sexes (n, % in age-specific groupa)	Advanced neoplasia – males (n, % in age-specific groupa)	Advanced neoplasia –females (n, % in age-specific groupa)
Age (Mean ± SD)		61.7 ± 8.6*	61.6 ± 8.6	62.0 ± 8.5
Age-group
40–49	520 (38.6)	47 (9.0)	29 (17.1)	18 (5.1)
50–59	747 (47.6)	147 (19.7)	114 (33.0)	33 (8.2)
60–69	739 (55.5)	195 (26.4)	130 (37.5)	65 (16.6)
≥70	272 (42.4)	96 (35.3)	64 (47.8)	32 (23.2)
All	2278 (46.6)	485 (21.3)	337 (33.8)	148 (11.5)

^aProportion over number of counts in age-specific group

^* P < 0.0001 (comparison to those in non-advanced neoplasia).

iFOBT: immunological fecal occult blood test; SD: standard deviation

The anatomic location of advanced neoplasia was reported in Table 3. The majority (88%) of 485 persons with advanced neoplasia had their advanced lesions found in one side only; of which distal colonic advanced neoplasia were predominant (P < 0.0001). Male had a significant higher proportion of both-sided advanced neoplasia than female (13.6 vs. 6.8%) but a significantly lower proportion of advanced neoplasia in proximal colon only than female (17.8 vs. 24.3%) (P < 0.0001). Male also had a significantly higher proportion of ≥ two polyps/tumors than female (84.3 vs. 68.2%) but significantly lower proportion of one polyp/tumor than female (15.7 vs. 31.8%) (P < 0.0001) (Table 3).

Table 3.

Distribution of advanced neoplasia by location and number of polyps and/or tumors identified in the current colorectal cancer screening program in Vietnam

	Total (n, %)	Male (n, %)	Female (n, %)	P value
Location
One sided
Proximal onlya	96 (19.8)	60 (17.8)	36 (24.3)	<0.0001
Distal onlyb	331 (68.2)	229 (68.0)	102 (68.9)
Subtotal	427 (88.0)	289 (85.8)	138 (93.2)
Both sided	56 (11.5)	46 (13.6)	10 (6.8)
Unknown	2 (0.4)	2 (0.6)	0 (0.0)
Number of polyps/Tumors
1	100 (20.6)	53 (15.7)	47 (31.8)	<0.0001
2–5	288 (59.4)	204 (60.5)	84 (56.8)
>5	97 (20.0)	80 (23.8)	17 (11.4)

^aProximal colon includes cecum, ascending, hepatic flexure and transverse colon

^bDistal colon includes splenic flexure, descending, sigmoid colon and rectum

Table 4 presents further details on the histological classification of detected colorectal lesions. Among 2278 colonoscopies, 79 (3.5%) were histologically confirmed as CRC (71 cases) or suspected cancer (8 cases). Also, 406 (17.8%) were histologically diagnosed with advanced adenomas and 526 (23.1%) with non-advanced adenomas. Male had a significantly higher proportion of advanced adenomas and cancer/suspected cancer than those in female (i.e. 29.2 vs. 9.0% and 4.6 vs. 2.6%; respectively, P < 0.0001).

Table 4.

Pathology findings of the current colorectal cancer screening program in Vietnam

	Total (N = 2278)	Male (n = 996) (n, %)	Female (n = 1282) (n, %)	P value
No adenomas	294 (12.9)	140 (14.1)	154 (12.0)	<0.0001
Non-advanced adenomas	526 (23.1)	255 (25.6)	271 (21.1)
Advanced adenomas	406 (17.8)	291 (29.2)	115 (9.0)
Cancer and suspected cancer	79 (3.5)	46 (4.6)	33 (2.6)
Othersa	53 (2.3)	14 (1.4)	39 (3.0)
Colonoscopy only
No pathologyb	920	250	670

^aOthers include but not limited to chronic mucosal inflammation; leiomyoma; lymphangiectasia; eosinophilic colitis; typical carcinoid

^bNo pathology includes those (1) who did not have polyp/tumor detected under colonoscopy (n = 905); (2) who had polyp/tumor that was not biopsied for some reasons (n = 15)

Systematic review of CRC screening in Asia-Pacific region

We found 17 articles/studies describing the CRC screening in Asia-Pacific region with FIT or iFOBT screening coupled with colonoscopy (Fig. 2), of which one study in Hong Kong (21), four studies in Japan (22–25), five studies in Korea (26–30), one study in Iran (31), three study in Singapore (32–34), two studies in Taiwan (35,36) and one study in Thailand (37) (Supplementary Table 1).

Figure 2.

Flowchart for screening and eligibility evaluation process of the systematic review, colorectal cancer screening program in Asia-Pacific region.

FIT test and colonoscopy

Studies with concentration of 100 ng Hb/ml with one-time testing procedure, the positive rate ranged from 5.0 (35) to 9.1% (31), whereas in study with concentration of 100 ng/ml using two-time testing procedures, the positive rate ranges from 6.4 (33) to 10.6% (32). Likewise, studies using 50 ng Hb/ml concentrations as the cut-off for positive reaction had the lowest positive rate of 4.5% (37) and highest positive rate of 10.4% (35). For studies using other cut-off of concentrations, the iFOBT (or FIT) positive rate was from 4.1 (28) to 8.0% (29) (Fig. 3).

Figure 3.

FIT test positive results – a systematic review of Asia-Pacific region. ^aOne-specimen; ^bTwo-specimen.

Once colonoscopy was performed subsequently, the rate of cancer among FIT positive participants was from 1.7 (36) to 16.4% (37) and that of advanced adenomas was from 7.1 (35) to 23.1% (21,32,35) (Fig. 4).

Figure 4.

Percentage of advanced adenomas and cancer among individuals with FIT test positive who underwent colonoscopy – a systematic review of Asia-Pacific region. Note: Study without data on advanced adenomas or cancer has data on one outcome only (i.e. either advanced adenomas or cancer) but not both outcomes. ^aOne-specimen; ^bTwo-specimen; ^c100 ng Hb/m; ^d75 ng Hb/m; ^e50 ng Hb/m.

Discussion

The current study presents the main findings of the first population-based CRC screening program among 103 542 individuals aged 40 years old or older who are residing in Hanoi, Vietnam. We found the attendance rate was 81.5% and the iFOBT test positive rate was 6.1%. Among those who underwent colonoscopy, 21.3% of total individuals were histologically diagnosed with advanced neoplasia (i.e. 3.5% CRC and 17.8% advanced adenomas) and 23.1% with non-advanced adenomas. Males were found to have significant higher proportions of iFOBT positive test, advanced-adenomas and/or CRC diagnosis and had two or more polyps/tumor than females. These major findings were consistent with studies found in the systematic review in Asia-Pacific region using similar modality (i.e. iFOBT/FIT test in combination with colonoscopy).

We showed that even though overall screening coverage rate, or the proportion between those who took iFOBT (n = 80 330) over those who were eligible to attend the CRC screening program (N = 672 742) is low (11.9%), the attendance rate for iFOBT test, defined as total number of individuals who returned the iFOBT test over total number of individuals who were distributed and consented for iFOBT, in our screening program is substantially high (81.5%). This rate is consistent with a prior study in Asia (38). In a mass screening program among 350 581 individuals living in Guangzhou, China, Lin et al. (38) found that the coverage rate for a FIT (or iFOBT) test was 14%. The rate of colonoscopy in our screening program was 46.6% (i.e. 2278 colonoscopies over 4887 iFOBT positive cases), which is consistent with the recent meta-analysis of CRC screening programs in China (39) showing the overall rate of colonoscopy at 44%; however, this rate was lower than that in Japan or European countries (i.e. >60–70% in a year). Note that no statistically significant difference in the rate of colonoscopy for people with positive iFOBT test was found between male and female (P = 0.59; Data not shown). However, there was a significant relationship between age group and rate of colonoscopy (P < 0.0001, Table 2). The rate was higher in the age groups of 50–59 years old (i.e. 47.6%) and 60–69 years old (i.e. 55.5%) and was lowest in the age group 40–49 years old (i.e. 38.6%, Table 2). Those who did not undergo colonoscopy in our screening program had different reasons such as they had colonoscopy at other hospitals, or had contraindication to colonoscopy, or refusal to colonoscopy, etc.

One interesting point is that although prior studies showed male tend to have higher participation for FOBT or FIT test, particularly in Europe or North America, we found that in our screening program, females had a significantly higher coverage rate than males (14.5 vs. 9.2%), which is consistent with the findings from a mass screening program in Guangzhou, China (38) (19.2 vs. 11.5%). One possible explanation is that females maybe more health conscious and comply better with medical advice than males in Asian countries (38). We also found that males appeared to have a significantly higher proportion of advanced adenomas and/or CRC than females (i.e. 29.2 male vs. 9.0% female and 4.6 male vs. 2.6% female, respectively) which are consistent with prior studies that were included in our systematic review. The aforementioned Guangzhou study (38) showed that the rates of advanced adenomas were 19.9% in males and 7.9% in females and of CRC were 5.1% in male and 2.8% in female participants, respectively.

Mass FOBT screening usually resulted in a low coverage rate for colonoscopy. For instance, results from the Nationwide Screening Colonoscopy Program for Colorectal Cancer in Germany (40) showed that the cumulative participation rate was 17.2 and 15.5% among male and female aged 55–74 years old, respectively. Financial issue is always a major challenge for the attendance or participation into colonoscopy in which patients have to pay for the procedure. Although there is no comparable evidence in the Asia-Pacific region, findings from a study by Guessous et al. (41) in the United States showed that lack of health insurance and dual coverage with Medicare and Medicaid were the most frequently reported barriers, whereas Medicare’s coverage of screening colonoscopy was consistently reported as a facilitator. In our screening program, the cost of colonoscopy was covered and thus might explain the higher rate of colonoscopy (i.e. 2278/4887 = 46.6%).

We found that the iFOBT positive rate was 6.1% which is consistent with the results reported by Lin et al. (38) (i.e. 6.2% in the Guangzhou study), by Chew et al. (33) (i.e. 6.4% in the screening program of 1048 Singaporean aged 40 years old or older) or by Kim et al. (30) (i.e. 6.2% in the screening program of 52 376 Koreans aged 50 years old or older). This iFOBT positivity rate (6.1%) found in our study is, however, lower than that in the study by Tan et al. (34) in the screening program among 20 989 Singaporeans aged ≥50 years old (8.1%) or in the study by Fu et al. (32) (10.6% in the screening program among 540 Singaporeans aged 41–78 years old). In our systematic review, the iFOBT (or FIT) positive rate ranged from the lowest of 4.1% (28) to the highest of 10.6% (32). The difference between our findings and findings from other studies, particularly in the systematic review might be attributable to the difference in screening settings in which ours is population-based screening program versus hospital-based screening program in prior studies (22,26,31,32,35,36). This difference might also be attributable to the difference in age cut-off for eligible participants in the screening program, in which we included those who were aged 40 years or older versus prior studies which included those who were aged 50 years or older (27,32,35,37), or study including participants of all ages (22), or study with participants aged 45 years old or older (31). In addition, the cut-off of concentration for positive reaction in our study was 100 ng/ml, which is different from those used in prior studies, either 50 ng/ml (21,36,37) or other concentrations (22,26,27,42), which produced results differently from ours.

The histologically confirmed rate of CRC among those who underwent colonoscopy in our screening program was 3.5%, which is consistent with prior studies by Tatsumi et al. (24) who reported the screening program among 29 900 Japanese and Wong et al. (21) in a screening program among 5343 individuals aged 50–70 years old who resided in Hong Kong. Our findings is higher than that found in study in Taiwan (1.7%) (36) but lower in different studies in Japan [i.e. 4.2% in a study by Morikawa at el. (22)], in Singapore [i.e. 5.8% in a study by Fu et al. (32)] and in South Korea [i.e. 4.8% in a study by Kim et al. (26)].

We also found that the percentage of advanced adenomas diagnosed in our study (17.8%) is in the mid-range of prior studies included in our systematic review, with the lowest rate at 7.1% (35) and the highest rate at 23.1% (21,32). Again, the difference from our study to these studies might be due to the difference in study settings (i.e. population-based vs. hospital-based), age range to be used for eligible participants in the screening program (i.e. ≥40 years of age in our study versus ≥50 years of age or ≥ 45 years of age or all age group in these studies) as well as the difference in concentration to be considered as positive reaction between our study (100 ng Hb/ml) and these studies (50 ng Hb/ml or other concentrations) as aforementioned.

To our knowledge, along with the investigation among large population in Vietnam, the current study provided the most comprehensive and up-to-date systematic reviews of CRC screening program, using iFOBT/FIT test in combination with colonoscopy, in the Asia-Pacific region. We found that two-steps modality using colonoscopy with other screening modalities, either chemical FOBT or FIT test has been one of the most prevalent modalities of colorectal screening strategy in the region.

Screening for CRC to early identify and manage pre-symptomatic individuals remains an important priority for a successful cancer control and prevention program. Colonoscopy is an invasive and expensive procedure but the only recommended screening modality for individuals with an increased risk of CRC, meaning those who have personal or family history of cancer (43). Recent evidence showed that adherence to colonoscopy in the CRC screening program is usually sub-optimal in an average and/or increased risk population (44). The efficacy of screening program using colonoscopy is undermined and therefore it is required to have an alternative modality that may limit the use of colonoscopy only to those with positive test (31,45). A detailed review of the Food and Drug Administration’s Clinical Laboratory Improvement Amendments (CLIA) database of FOBT tests, the 2014 iFOBT proficiency testing program results and the literature related to iFOBT accuracy published in 2017 showed that automatic iFOBT (or FIT) has the best characteristics for advanced adenomatous polyps and/or CRC (46). In a recent meta-analysis of 12 studies, Katsoula et al. (45) showed that FIT test has high overall diagnostic accuracy for CRC with the average sensitivity of 93% (95% CI: 53–99%) and average specificity of 91% (95% CI: 89–92%); however, moderate accuracy for advanced neoplasia in patients at above-average personal or family risk [i.e. average sensitivity: 48% (95% CI: 39–57%) and average specificity: 93% (95% CI: 91–94%)]. Taken together, this suggest that a two-step modality (or sequential modality) with FIT test and colonoscopy be an efficient screening approach for limited-resource settings such as Vietnam.

The current study has several limitations. First, the coverage rate (for iFOBT) in our screening program was low (11.9%). However, it is important to note that this rate does not necessarily reflect the actual screening uptake of the target population since not all eligible individuals were delivered information about the screening program by CHS staff. In the future program, the outreach coverage by CHS staff should be increased as it is one important channel to deliver CRC screening information. Second, selection bias might be introduced as individuals with chronic intestinal discomfort such as coprostasis, diarrhea, or blood fecal or those who had a first-degree relative with CRC tend to participate into the screening program. This might result in a higher positive rate of iFOBT test and requires a cautious interpretation. Third, our systematic data could not address an important question whether the two-step modality is more cost-effective and efficient in detecting CRC and adenomas than other screening methods. Future studies are thus warranted to answer this question.

In summary, in the first population-based CRC screening program among 103 542 Vietnamese aged 40 years or older, we used two-step modality (i.e. iFOBT test and colonoscopy) and showed 6.1% positive rate of iFOBT test, 3.5% CRC, 17.8% advanced adenomas and 23.1% with non-advanced adenomas. These rates are generally in agreement with those seen in other Asian countries with similar or different CRC screening modalities. Our study affirmed the usage of the two-step modality of iFOBT and colonoscopy as one of the promising modalities for CRC screening in limited resource settings.

Funding

The screening program was funded by Vinmec Health Care System–Vingroup (Grant No. IOC.17.01, PI: Luu). Luu HN is also partially supported by the University of Pittsburgh Medical Center (UPMC) Start-up grant and the National Institutes of Health (NIH) (Grant No. 3P30CA047904-32S2, PI: Ferris). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Conflicts of interest

None to declare.

Data Availability

Data are available from the corresponding author and from the Research Management Department, Vinmec Healthcare System, 458 Minh Khai Street Vinh Tuy Ward, Hai Ba Trung District, Hanoi, Vietnam. Phone: +84–243–975-0028; Fax: +84–243–974-3557. Email: v.rmo@vinmec.com.