1 INTRODUCTION

Colorectal cancer (CRC) is the second most common cancer in China, and the incidence rate of CRC has increased rapidly in recent years.¹ There is a continuing need to refine and develop effective strategies for early detection and prevention, particularly in populations at higher than average risk. Current guidelines recommend colonoscopy-based surveillance to decrease the risk of CRC among these participants with previously detected adenomatous polyps.^2-4 Surveillance combined with clinical findings at the time of colonoscopy recommends follow-up colonoscopy within 3 years after removal of above-average risk adenomas and within 5 years for low-risk adenoma participants, which may be extended up to 7–10 years in the average-risk population.^{2, 5-9} However, there is an urgent and inevitable concern is this surveillance schedule may lead to significantly increased demand for endoscopic examination, potentially exceeding the colonoscopy capacity, especially in low resource regions.

Several factors contribute to this situation. The widespread implementation of annual colorectal cancer screening programs is influenced by factors such as population growth among aging individuals and changes in lifestyles enhancing CRC screening rates.¹⁰ Moreover, individuals with increased awareness of colorectal cancer screening and prevention, along with active promotion and publicity by governments, healthcare institutions, and relevant organizations, have positioned colonoscopy examinations as the primary method for CRC screening.^{11, 12} As a result, more individuals are encouraged to seek screening services and undergo colonoscopy examinations. Regular surveillance after screening can help doctors assess the colon health of individuals and intervene in a timely manner to prevent the potential development of colorectal cancer.^{2, 5} Therefore, participants need to undergo regular colonoscopy examinations after screening to ensure timely detection and management of any abnormalities, which increased demand means that healthcare facilities and medical professionals need to provide more colonoscopy examination resources ensure that patients receive appropriate monitoring and treatment. Whereas resource limitations, including inadequate endoscopic equipment, manpower, and medical resources limit the capacity of colonoscopy and pose challenges in meeting the growing demand for colonoscopy surveillance. Such an escalation in demand for endoscopies was recently observed in Netherlands with a spike in the rates of primary care referrals for suspected CRC due to the widespread implementation of CRC screening.¹³

In addition to colonoscopy, fecal immunochemical testing (FIT) has been widely adopted in screening programs for CRC worldwide among participants with unknown risk^{14, 15} but has been largely overlooked in clinical studies as a method for surveillance among participants with elevated risk. Some recent studies reported that FIT testing could potentially serve as a viable alternative strategy for surveillance in participants with negative colonoscopy results. A study of population-based screening programs in Taiwan found that FIT testing after a negative colonoscopy may reduce the risk of developing CRC.¹⁶ Another study in the United Kingdom found that using annual FIT instead of colonoscopy every 3 years could reduce colonoscopy procedures by 71% among intermediate-risk participants (three to four small adenomas or one ≥10 mm), significantly lowering costs.¹⁷ However, data from other regions is still lacking, and furthermore, whether individuals with no neoplasia or non-advanced neoplasia based on primary colonoscopy should routinely undergo FIT screening remains controversial.

In Asia and other resource-limited regions, colonoscopy-based surveillance can potentially greatly increase clinical workload, in addition to requiring staff with advanced training, and imposing financial burdens.² Given these limitations, FIT presents an attractive alternative to CRC monitoring by colonoscopy. However, before adopting FIT for surveillance, more research is necessary to determine its performance in monitoring CRC in above-average risk populations. Currently, one CRC screening program in Tianjin differentiates it from the rest of China in that above-average risk individuals were invited to participate a follow-up CRC surveillance cohort. To our knowledge, no studies in a Chinese population have yet investigated the effectiveness of FIT surveillance among participant with no neoplasia, non-advanced colorectal neoplasia, or advanced colorectal neoplasia. Our current study, therefore, aims to investigate the performance of FIT in preventing advanced colorectal neoplasia and CRC in individuals with above-average risk.

2 METHODS

3 RESULTS

3.1 Baseline characteristics of study participants

The study cohort comprised a total of 12,515 individuals who had above-average CRC risk between January 2012 and December 2022. Among them, 4980 individuals (39.79%) received at least one FIT test surveillance, the remaining 7535 individuals (60.21%) did not undergo FIT test during study period (Figure 1). More than half (53.8%) were male (6734/12515) and the mean age was 60.21 years (Table 1). The main demographic characteristics were compared between the FIT surveillance group and the non-FIT surveillance groups. The analysis revealed no statistically significant differences between the groups in terms of family history of colorectal cancer, chronic diarrhea, chronic constipation, mucoid blood stool, chronic cholecystitis, or history of cancer.

TABLE 1. Baseline Characteristics of Study Participants by Surveillance Group.

Characteristic	Overall	Non-FIT surveillance	FIT surveillance	p value
Total, n	12,515	7535	4980
Age (mean (SD))	60.18 (7.21)	60.08 (7.32)	60.34 (7.03)	0.041
Sex (%)
Male	6734 (53.8)	4130 (54.8)	2604 (52.3)	0.006
Female	5781 (46.2)	3405 (45.2)	2376 (47.7)
Education (%)
High school and below	1928 (15.4)	983 (13.0)	945 (19.0)	<0.001
High school	8461 (67.6)	5118 (67.9)	3343 (67.1)
Postsecondary	2126 (17.0)	1434 (19.1)	692 (13.9)
Work (%)
Administrators and technicians	3972 (31.7)	2492 (33.1)	1480 (29.7)	<0.001
Service industry	863 (6.9)	558 (7.4)	305 (6.1)
Production and transport devices staff	4796 (38.3)	2737 (36.3)	2059 (41.4)
Unemployment and unknown	2884 (23.0)	1748 (23.2)	1136 (22.8)
Previously detected colonic polyp (%)
Yes	2679 (21.4)	1721 (22.8)	958 (19.2)	<0.001
No	9836 (78.6)	5814 (77.2)	4022 (80.8)
Family history of colorectal cancer (%)
Yes	1820 (14.5)	1095 (14.5)	725 (14.6)	0.988
No	10,695 (85.5)	6440 (85.5)	4255 (85.4)
Chronic diarrhea (%)
Yes	3292 (26.3)	1975 (26.2)	1317 (26.4)	0.786
No	9223 (73.7)	5560 (73.8)	3663 (73.6)
Chronic constipation (%)
Yes	3043 (24.3)	1817 (24.1)	1226 (24.6)	0.534
No	9472 (75.7)	5718 (75.9)	3754 (75.4)
Mucoid blood stool (%)
Yes	2721 (21.7)	1656 (22.0)	1065 (21.4)	0.445
No	9794 (78.3)	5879 (78.0)	3915 (78.6)
Chronic appendicitis or appendectomy (%)
Yes	1264 (10.1)	746 (9.9)	518 (10.4)	0.379
No	11,251 (89.9)	6789 (90.1)	4462 (89.6)
Chronic cholecystitis or cholecystectomy (%)
Yes	1257 (10.0)	755 (10.0)	502 (10.1)	0.936
No	11,258 (90.0)	6780 (90.0)	4478 (89.9)
History of cancer (%)
Yes	683 (5.5)	416 (5.5)	267 (5.4)	0.731
No	11,832 (94.5)	7119 (94.5)	4713 (94.6)

• Abbreviation: SD, standard deviation.

3.2 Association between FIT surveillance and outcomes

The total follow-up time was 48,278 person-years for the 12,515 participants (Table 2). The median follow-up times of 2.9 years for CRC and 3.0 years for advanced colorectal neoplasia. During the follow-up period, CRC was diagnosed in 80 subjects in the study population, with a cumulative incidence of 184.85 per 100,000 person-years. Among these patients, 51 CRC cases occurred in the non-FIT surveillance group (incidence rate, 233.88 per 100,000 person-years), while 29 CRC cases were detected in the FIT surveillance group (incidence rate, 184.35 per 100,000 person-years), resulting in an incidence rate ratio (IRR) of 0.58 (95% CI, 0.37–0.91). During the study period, 428 advanced colorectal neoplasia cases were reported in the non-FIT surveillance group, while 269 cases occurred in the FIT surveillance group. A significantly lower incidence of advanced colorectal neoplasia was observed in the FIT surveillance group (IRR: 0.64; 95% CI, 0.55–0.74). The incidence rates of CRC and advanced colorectal neoplasia were higher in males and individuals older than 60 years.

TABLE 2. Incidence of CRC and advanced colorectal neoplasia by surveillance group.

	Total (n = 12,515)			Non-FIT surveillance (n = 7535)			FIT surveillance (n = 4950)			IRR (95% CI)a
	Cases, no. (%)	Person years	Incidence density	Cases, no. (%)	Person years	Incidence densityb	Cases, no. (%)	Person years	Incidence densityb
Colorectal cancer
Total	80 (0·6%)	43,278	184.85	51 (0·7%)	21,806	233.88	29 (0·6%)	21,472	135.06	0.58 (0.37–0.91)*
Age
40–59	8 (0·1%)	17,904	44.68	3 (5.9%)	9605	31.23	5 (17.2%)	8299	60.25	1.93 (0.46–8.07)
60–74	72 (0·5%)	25,375	283.75	48 (94.2%)	12,202	393.38	24 (82.8%)	13,173	182.19	0.46 (0.28–0.76)*
Sex
Male	50 (0.4%)	22,759	219.69	30 (58.2%)	11,571	259.27	20 (69.0%)	11,188	178.76	0.69 (0.39–1.21)
Female	30 (0.2%)	20,519	146.20	21 (51.2%)	10,235	205.17	9 (31.0%)	10,284	87.52	0.43 (0.20–0.93)*
Advanced colorectal neoplasia
Total	697 (5.6%)	43,278	1610.52	428 (5·7%)	21,806	1962.72	269 (5.4%)	21,472	1252.80	0.64 (0.55–0.74)*
Age
40–59	205 (1.7%)	17,904	1145.02	114 (26.2%)	9605	1186.88	91 (33.8%)	8299	1096.57	0.92 (0.70–1.22)
60–74	492 (3.9%)	25,375	1938.93	314 (73.4%)	12,202	2573.35	178 (66.2%)	13,173	1351.21	0.53 (0.44–0.63)*
Sex
Male	429 (3.4%)	22,759	1884.95	258 (60.3%)	11,571	2229.68	171 (63.6%)	11,188	1528.42	0.69 (0.57–0.83)*
Female	268 (2.2%)	20,519	1306.09	170 (39.7%)	10,235	1660.92	98 (36.4%)	10,284	952.94	0.57 (0.45–0.74)*

• Abbreviations: CI, confidence interval; CRC, colorectal cancer; FIT, fecal immunochemical test; IRR, incidence rate ratio.
• ^a FIT surveillance versus non-FIT surveillance.
• ^b Incidence density was the number of cases per 100,000 person-years; all incidence densities are crude rates.
• *Indicate statistically significant (p < 0.05).

During the entire follow-up, the cumulative incidence of CRC and advanced colorectal neoplasia was higher in the non-FIT surveillance group than in the FIT group (Figure 2). After 10 years of follow-up, the cumulative incidence of CRC was 0.66% (95% confidence interval [CI], 0.71–0.95) in the non-FIT surveillance group and 0.56% (95% CI, 0.96–0.98) in the FIT surveillance group (Figure 2A). CRC developed in 50 participants without FIT surveillance and in 28 participants with FIT surveillance. The 10-year cumulative incidence of advanced colorectal neoplasia was 5.63% (95% CI, 0.42–0.60) without FIT surveillance and 5.07% (95% CI, 0.66–0.76) with FIT surveillance (Figure 2B). Colorectal neoplasia developed in 424 participants without FIT surveillance and in 266 participants with FIT surveillance. Among the different groups defined by baseline colonoscopy findings, the advanced adenoma group had the highest number of cumulative events, followed by the non-advanced adenoma group, and finally the negative colonoscopy group (Figure S2).

Cox regression analyses that FIT surveillance was significantly associated with lower risk of CRC or advanced colorectal neoplasia compared with non-FIT surveillance (Figure 3). In participants who received FIT surveillance after baseline colonoscopy, risk of CRC was significantly lower than in the non-FIT group; risk of advanced colorectal neoplasia was similarly reduced in the FIT surveillance group. After adjusting for contributing baseline factors (age, sex, employment, education level, family history of CRC, and digestive tract symptoms), participants in the FIT surveillance group had a 54% lower risk of developing CRC (HR, 0.46; 95% CI, 0.29–0.74) and a 45% lower risk of developing advanced colorectal neoplasia (HR, 0.55; 95% CI, 0.47–0.64) compared to those in the non-FIT surveillance group (Figure 3).

Unadjusted models based on findings in the baseline colonoscopy indicated that FIT surveillance was associated with a lower risk of CRC in subjects with advanced adenoma in the initial colonoscopy (HR, 0.24; 95% CI, 0.07–0.84) (Figure 4A). Furthermore, this significant correlation was robust to adjustments for age, sex, or multivariable adjustment. In addition, the risk of CRC was also significantly reduced in individuals with non-advanced adenoma at baseline colonoscopy. There was a reduction in the risk of colorectal cancer in individuals with no neoplasia findings in baseline colonoscopy, but it did not reach statistical significance. FIT surveillance not only reduced the occurrence of CRC but also reduced the incidence of advanced colorectal neoplasia in participants with advanced adenoma, no neoplasia, or non-advanced adenoma at baseline (Figure 4B).

4 DISCUSSION

In this regional retrospective cohort of participants with an above-average risk of developing colorectal neoplasia, we observed a significant reduction in the incidence of colorectal cancer and advanced colorectal neoplasia among participants who underwent surveillance with the FIT test. These results highlight the importance of tailored surveillance approaches in reducing the burden of colorectal cancer, particularly in above-average risk groups.

According to the current guidelines, colonoscopy is an important component of CRC surveillance,^{14, 18, 19} and post-surgery surveillance colonoscopy has been previously associated with significantly reduced incidence of CRC.²⁰ However, colonoscopy is accompanied by risks, and colonoscopy-related complications, over-diagnosis, and over-treatment of CRC can result in a substantial depletion of resources for a public health system.^{21, 22} By contrast, FIT is a relatively simple, low-cost tool capable of detecting fast-growing or overlooked lesions that were missed during surveillance colonoscopy.^23-25 Several other studies have reported evidence supporting the value of FIT test in the intervening years between surveillance colonoscopies, especially since negative FIT is also associated with lower risk for advanced neoplasia.^{26, 27} Despite data indicating its informative value, FIT has yet to be adopted as a routine measure in CRC surveillance because its clinical benefits in decreasing CRC incidence and potentially mortality, have remained unclear.^{2, 28, 29} Thus, the benefits of FIT surveillance are worth exploring given the potential for to improve CRC guidelines.

Our study showed that subgroups of the population who had a greater risk for advanced neoplasia in those with a diagnosis of advanced adenoma at the prime colonoscopy, those without a completed FIT in the interval, those aged over 60 years of age, and male. Previous studies have shown that individuals with above-average risk received subsequent FIT after a negative colonoscopy can remarkably reduce the risk of incident CRC.^{16, 17} However, evidence based on previous colonoscopy findings related to FIT between colonoscopy interval and other important risk factors of CRC, namely age and sex remain limited. There was the highest increased risk of CRC and advanced neoplasia in those individuals who were diagnosed with advanced adenoma at the prime colonoscopy and did not receive subsequent FIT surveillance. In our study, FIT surveillance after reduced the risk of incident CRC and advanced neoplasia stratifying participants according to prime colonoscopy findings. Overall, we found that those with advanced adenoma were more likely to highly benefit from FIT surveillance.

FITs present important value in colonoscopy surveillance, and might have identified the missed lesion to prevent colorectal cancer When an advanced adenoma was previously missed by colonoscopy,^{30, 31} especially in some critical settings when resources are constrained by healthcare funding or events, such as the pandemic, implementing FIT to extend colonoscopy intervals would help relieve and lessen the backlog of colonoscopy and distribute health resources for ensuring rational.³² Therefore, based on the results of our present study, we propose that above-average risk individuals after colonoscopy should routinely receive a subsequent FIT to detect missed cancers and advanced adenomas.

This study has several strengths and limitations. First, it was based on a large sample size, community-based recruitment; validated approaches to capture pathology data and colonoscopy examinations; adjustment of major factors that might have affected the risk for incident CRC to make precise estimations. Second, these enrolled subjects underwent FIT between the repeat colonoscopies. The conclusion presented in this study may be applicable to similar situations for colonoscopy surveillance programs in different areas, including individuals with a family history of colorectal cancer, personal history of colorectal adenomas, or positive fecal occult blood test as well. Third, the sufficient follow-up time of the cohort is taken into consideration, which is needed for the occurrence and progression of outcome events and the sojourn time of preclinical colorectal cancer. In addition, our study included diverse evaluation of assumptions through sensitivity analyses. Our study also has some limitations. First, although we found that FIT surveillance after colonoscopy is associated with the risk of CRC, the result cannot prove causality arises from limitations of any observational study. Second, one limitation of our study is self-selection bias. Besides, some important confounders such as adenoma detection rate and lifestyle factors such as smoking or physical activity have not been measured and adjusted in this retrospective cohort study. Therefore, the residual confounding by measured factors may still be present. Third, the effect of more than one subsequent FIT on incident CRC has yet to be explored. Going forward, we will persist the surveillance cohort and keep providing up-to-date and vital information, including tumor stages and treatments and survival rates based on longer follow-up periods, controlling for more objective measurements and adjusting covariates. Further, we will explore to clarify the appropriate timing and frequency of FIT follow-up and identify the population that may benefit from this approach.

5 CONCLUSION

This cohort study supports FIT as surveillance tool for above-average risk individuals, as they have a higher risk of developing colorectal neoplasia than the general population. Additionally, FIT surveillance is associated with a reduction in CRC and advanced colorectal neoplasia incidence. Our study supporting the potential of FIT surveillance as an effective strategy for reducing the incidence of colorectal cancer and advanced colorectal neoplasia in above-average risk population.

AUTHOR CONTRIBUTIONS

Hai Qin: Data curation (equal); formal analysis (equal); writing – original draft (equal); writing – review and editing (equal). Mingqing Zhang: Data curation (supporting); formal analysis (equal); writing – original draft (equal); writing – review and editing (equal). Guanglu Zhang: Formal analysis (equal); visualization (equal); writing – original draft (equal); writing – review and editing (equal). Lizhong Zhao: Data curation (supporting); investigation (equal); validation (equal). Huan Zhang: Data curation (supporting); investigation (supporting); validation (equal). Weituo Zhang: Validation (equal); writing – review and editing (supporting). Yijia Wang: Formal analysis (supporting); writing – review and editing (supporting). Xipeng Zhang: Conceptualization (equal); methodology (equal); writing – review and editing (supporting). Li Xie: Formal analysis (equal); methodology (equal); writing – original draft (equal); writing – review and editing (equal). Biyun Qian: Conceptualization (lead); funding acquisition (lead); methodology (lead); project administration (lead); resources (lead); supervision (lead); writing – review and editing (lead).

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no conflicts of interest.