Sage Journals: Discover world-class research

Abstract

Objective

Roll-out of population-based colorectal cancer (CRC) screening with faecal immunochemical test (FIT) is limited by availability of further investigations, particularly colonoscopy and examination of excised lesions. Our objective was to assess whether variation in number of faecal samples and threshold adjustment can optimise resource utilisation and CRC detection rate.

Methods

Three different screening strategies were compared for the same FIT threshold using a quantitative FIT system: one FIT, positive when >20 µg Hb/g faeces; two FIT, positive when either was >20 µg Hb/g faeces; and two FIT, positive when the mean was >20 µg Hb/g faeces. We calculated changes in the size of population the provider could invite to screening for an equal number of screening positive results, and CRC and adenoma detected.

Results

In our setting, Region of Murcia, south of Spain (not fully rolled out screening programme), changing the usual strategy of two FIT, positive when either to positive when the mean was >20 µg Hb/g faeces, would increase population invited by 37.81% with the same number of positive results (which would generate a CRC detection rate of 19.2%). In a fully rolled out programme, changing the strategy from one to two FIT (positive when the mean is >20 µg Hb/g faeces), would increase CRC detection rate by 4.64% with an increase of only 13.34% in positive FIT.

Conclusions

In a population-based CRC screening programme, smart use of number of FITs and positivity threshold can increase population invited and CRC detection without increasing the number of colonoscopies and pathological examinations needed.

Keywords

Colorectal cancer colorectal neoplasia faecal immunochemical test faecal haemoglobin screening

Introduction

Colorectal cancer (CRC) has high incidence and age-adjusted mortality world-wide: in 2018, there were over 1,849,000 new cases and more than 880,000 deaths.¹ CRC fulfils the criteria proposed by the World Health Organization (WHO) for the implementation of population-based screening programmes.² Several randomised controlled trials and meta-analyses have demonstrated the efficacy of such programmes.^3,4 In consequence, in 2003, the European Council recommended that CRC should be subject to population-based screening campaigns in EU Member States and a test for the presence of occult blood in faeces offered to the population aged between 50 and 74 years, adjusting the age range according to the local epidemiology of CRC and each country's priorities.⁵ In response to this recommendation, several States developed their own strategies and guidelines,⁶ including Spain,^7,8 France,⁹ United Kingdom,¹⁰ The Netherlands¹¹ and Italy.¹²

One important aspect of implementation of such programmes is to offer the participants a two-fold guarantee of continuity¹³ to ensure that (a) participants will be invited over the entire recommended age range and (b) if the screening test result is positive, the system must facilitate access to diagnostic confirmation techniques, adequate treatment and subsequent surveillance. For CRC, once the initial faecal test results are available, colonoscopy is usually offered to participants whose results suggest a greater likelihood of CRC.⁶ Screening colonoscopy is a complex investigation which requires specific instrumentation, highly trained staff and considerable patient involvement time, since it might involve the excision and retrieval of polyps and other lesions, and biopsies of suspicious findings. Therefore, it is understandable that health services’ capacities to cope with this type of follow-up may be limited. Developing strategies to increase the detection of CRC when colonoscopy capacity is constrained is necessary. One reasonable approach to solving this problem is optimising the faecal test strategy.

Several approaches to optimisation have been taken, most importantly the use of faecal immunochemical tests (FITs) versus the traditional guaiac-based faecal occult blood tests,^14–16 but also a two-tier reflex approach,¹⁷ varying screening frequency¹⁸ or the threshold used with FIT,^19,20 personalising the threshold by including factors such as sex and age in a risk-score,^21–24 or going further, for example through neural network risk modelling.²⁵ The European Commission's scientific advisors advocate the approach of threshold individualisation as a way to improve CRC screening programmes.²⁶ Implementing a programme means deciding about the type of FIT to be used, the faecal haemoglobin concentration (f-Hb) threshold for declaring a result as positive, the number of FITs to be used and the format of the FIT. Few studies have compared one to two or three FITs²⁷ and only one has compared three alternatives.²⁸

The objective of this study was to compare three alternatives for the same positive FIT threshold (>20 µg Hb/g faeces): one FIT; two FIT, positive when either result is >20 µg Hb/g faeces; and two FIT when the mean result is >20 µg Hb/g faeces. The aim was to examine the variations in CRC and adenoma detection in two situations: (a) programmes not rolled out to the entire target population, maintaining the positivity and therefore the number of colonoscopies required, and also investigating the effect on the invited population; and (b) programmes with population coverage (population invited in relation to population registered in the census) close to 100%.

Methods

The colorectal cancer screening programme of the region of Murcia (Spain)

In the Region of Murcia, following recommendations,²⁹ the pilot project for the CRC screening programme with FIT and colonoscopy began in 2006.³⁰ Women and men aged between 50 and 69 years in two Health Areas were invited, except for the population in the old part of the city of Murcia: 16,975 were invited in 2006. The roll-out of the programme to other Health Areas began in 2008 and was further extended in 2009. In 2018, coverage had been extended to 130,929, 39.5% of the target population of the Region. The details of the biennial invitation screening methodology are fully described in Supplement 1. Exclusion criteria are detailed in Supplement 2.

Faecal immunochemical tests (FIT)

This study is based on data available in the management and assessment application of the CRC screening programme of the Region of Murcia (PCAColon) since the introduction of FIT (HM-JACKarc, Minaris Medical Co., Ltd, Tokyo, Japan) from 01 July 2014 to 31 December 2018. The data concern participants who submitted testable FIT devices. The two FIT samples are taken from successive bowel motions. Registry order is random, it doesn't take defecation date into account, although it is checked that they belong to successive motions.

Full details of what each variable means, further variable transformation, and generation of estimates of the variations compared for each strategy and estimates in fully rolled out programmes are shown in Supplement 3.

Strategy comparison

As stated above, the aim was to compare three alternative strategies for the same positive FIT threshold (>20 µg Hb/g faeces): one FIT, two FIT (mean result), and two FIT (higher result). Among these, it is important to note that the one FIT strategy was examined by selecting one of the two FIT results available, not by requesting a single sample from the participants. The selected FIT was the first introduced by our staff into the database (which was done in no particular order). Internal analysis showed no significant differences in the strategy comparison by choosing the other sample instead.

Statistical analysis

For each strategy, the positivity (absolute value) and the positivity as a percentage of positive FIT results in the total number of participants was obtained; the values of the diagnostic categories (CRC; high risk adenoma, HR; intermediate risk adenoma, IR; and low risk adenoma, LR), their detection rates as the percentage of the total number of participants with testable FIT and the positive predictive values (PPV) for each of the diagnostic categories, as the percentage of each diagnosed category out of the total number of positive results. For programmes that are not fully rolled out, each of the strategies was compared to other two, calculating the percentage increase in the invited population and the results expected from the new strategy with regard to the reference strategy: this is the proportional (algebraic) increase of the invited population in the strategy we want to compare with that of the reference strategy expressed as a percentage. With A being the population of the reference strategy and B the new population estimated from A, the aim is to calculate this increase provided that: 1) the participation rate is the same in both strategies; 2) the rate of positives is the one of the strategy being compared; and 3) the number of positives is the same as in A (hence, with the same number of colonoscopies). Therefore, the result obtained is a percentage (positive or negative) that shows the variation of the invited population (B) with respect to the invited population in the reference population (A). For programmes that have been extended to the entire target population, the percentage increases in positive results and the detection of each diagnostic category were calculated regarding those which corresponded to the reference strategy in the comparison. In other words, we estimated the corresponding change in the size of the population which the provider could invite to screening while maintaining the same number of positive results (i.e. the same number of colonoscopies). The 95% confidence intervals were calculated.

Results

A total of 114,049 people participated during the study period (participation rate varied between 44.0% and 53.9% in the years studied); 42.9% (Table 1) participated for the first time (initial screening), whereas the remainder had already participated (successive screening).

Table 1.

Results of the colorectal cancer screening programme from 01 July 2014 to 31 December 2018 in the region of murcia, Spain: participants, detected cancers and participants classified as high (HR), intermediate (IR) or low (LR) risk from the characteristics of the adenomas detected during the study period, distributed by type of screening, age and sex.

			Testable FIT	FIT > 20 µg Hb/g faeces		Cancer		HR^b		IR^c		LR^d		Did not complete colonoscopy
Screening type^a	Age (years)	Sex	n	n	%	n	%	n	%	n	%	n	%	n	%
Initial	<60	Women	9522	620	6.5%	13	0.14%	42	0.44%	99	1.04%	84	0.88%	38	0.40%
		Men	8982	883	9.8%	22	0.24%	123	1.37%	213	2.37%	159	1.77%	72	0.80%
	Total < 60		18,504	1503	8.1%	35	0.19%	165	0.89%	312	1.69%	243	1.31%	110	0.59%
	≥60	Women	16,187	1636	10.1%	54	0.33%	165	1.02%	271	1.67%	365	2.25%	114	0.70%
		Men	14,188	2054	14.5%	113	0.80%	446	3.14%	431	3.04%	405	2.85%	144	1.01%
	Total ≥ 60		30,375	3690	12.1%	167	0.55%	611	2.01%	702	2.31%	770	2.53%	258	0.85%
Initial – Total			48,879	5193	10.6%	202	0.41%	776	1.59%	1014	2.07%	1013	2.07%	368	0.75%
Successive	<60	Women	16,222	794	4.9%	3	0.02%	34	0.21%	89	0.55%	159	0.98%	33	0.20%
		Men	13,598	892	6.6%	14	0.10%	63	0.46%	181	1.33%	224	1.65%	52	0.38%
	Total < 60		29,820	1686	5.7%	17	0.06%	97	0.33%	270	0.91%	383	1.28%	85	0.29%
	≥60	Women	19,746	1290	6.5%	14	0.07%	70	0.35%	186	0.94%	308	1.56%	71	0.36%
		Men	15,604	1303	8.4%	33	0.21%	166	1.06%	284	1.82%	312	2.00%	77	0.49%
	Total ≥ 60		35,350	2593	7.3%	47	0.13%	236	0.67%	470	1.33%	620	1.75%	148	0.42%
Successive – Total			65,170	4279	6.6%	64	0.10%	333	0.51%	740	1.14%	1003	1.54%	233	0.36%
Total			114,049	9472	8.3%	266	0.23%	1.109	0.97%	1.754	1.54%	2.016	1.77%	601	0.53%

Two classes: Initial, the participant undergoes screening for the first time; and Successive, if the participant has already participated.

High risk (HR). Presence of an adenoma ≥ 20 mm or at least five adenomas of any size.

Intermediate risk (IR). Presence of an adenoma ≥10 mm and less than 20 mm, or three or four adenomas less than 10 mm, or at least one adenoma less than 20 mm with high-grade dysplasia or a villous component.

Low risk (LR). Presence of one or two adenomas less than 10 mm and without high-grade dysplasia or a villous component.

FIT: faecal immunochemical test.

By sex, while almost 54.1% of the participants and 45.8% of those with positive results were women, we found a lower proportion of women with CRC (31.5%). Differences by sex were similar with high-risk adenomas (HRA), intermediate-risk adenomas (IRA) and low-risk adenomas (LRA) too (percentage of women: 28.1%, 36.7% and 45.6%, respectively).

By age, 42.4% of the participants were under 60 years. Of those with positive FIT, only 33.7% belonged to this age group. We found a lower percentage of CRC and related lesions among them, as well (cancer: 19.5%; HR: 23.6%; IR: 33.2%; LR: 31.1%), compared to the ≥ 60 years group.

In the programme's usual strategy, (two samples, positive when at least one was ≥20 µg Hb/g faeces), the positivity was 8.31% (Table 2); 10.6% if initial participation and 6.6% if successive participation. This pattern was repeated with the CRC detection rate, with a total of 0.23%, 0.41% initial and 0.10% successive. Similar occurred with the PPV, which was higher for all groups in the first versus successive participations.

Table 2.

Results of the colorectal cancer screening programme from 01 July 2014 to 31 December 2018 in the Region of Murcia: neoplasia detection rates and positive predictive values (PPV) with 95% confidence intervals (CI) depending on the screening strategy.

Two FIT: positive when either is ≥20 µg Hb/g faeces
Results	Initial^a	Successive^a	Total	Detection rates with 95% CI						PPV with 95% CI
	n	n	n	Initial	CI	Successive	CI	Total	CI	Initial	CI	Successive	CI	Total	CI
Testable FIT	48,879	65,170	114,049
Positive FIT result	5193	4279	9472	10.62%	± 0.27	6.57%	± 0.19	8.31%	± 0.16
Cancer	202	64	266	0.41%	± 0.06	0.10%	± 0.02	0.23%	± 0.03	3.89%	± 0.53	1.50%	± 0.36	2.81%	± 0.33
High risk^b	775	333	1108	1.59%	± 0.11	0.51%	± 0.05	0.97%	± 0.06	14.92%	± 0.97	7.78%	± 0.80	11.70%	± 0.65
Intermediate risk^c	1014	739	1753	2.07%	± 0.13	1.13%	± 0.08	1.54%	± 0.07	19.53%	± 1.08	17.27%	± 1.13	18.51%	± 0.78
Low risk^d	1013	1003	2016	2.07%	± 0.13	1.54%	± 0.09	1.77%	± 0.08	19.51%	± 1.08	23.44%	± 1.27	21.28%	± 0.82
Two FIT: positive when the mean is ≥20 µg Hb/g faeces
Results	Initial^a n	Successive^a n	Total n	Detection rates with 95% CI						PPV with 95% CI
Results	Initial^a n	Successive^a n	Total n	Initial	CI	Successive	CI	Total	CI	Initial	CI	Successive	CI	Total	CI
Positive FIT result	3856	2905	6761	7.89%	± 0.24	4.46%	± 0.16	5.93%	± 0.14
Cancer	193	55	248	0.39%	± 0.06	0.08%	± 0.02	0.22%	± 0.03	5.01%	± 0.59	1.89%	± 0.41	3.67%	± 0.38
High risk^b	672	256	928	1.37%	± 0.10	0.39%	± 0.05	0.81%	± 0.05	17.43%	± 1.03	8.81%	± 0.85	13.73%	± 0.69
Intermediate risk^c	802	516	1318	1.64%	± 0.11	0.79%	± 0.07	1.16%	± 0.06	20.80%	± 1.10	17.76%	± 1.15	19.49%	± 0.80
Low risk^d	700	663	1363	1.43%	± 0.11	1.02%	± 0.08	1.20%	± 0.06	18.15%	± 1.05	22.82%	± 1.26	20.16%	± 0.81
One FIT ≥20 µg Hb/g faeces
	Initial^a	Successive^a	Total	Detection rates with 95% CI						PPV with 95% CI
Results	n	n	n	Initial	CI	Successive	CI	Total	CI	Initial	CI	Successive	CI	Total	CI
Positive FIT result	3398	2561	5959	6.95%	± 0.23	3.93%	± 0.15	5.22%	± 0.18
Cancer	183	54	237	0.37%	± 0.05	0.08%	± 0.02	0.21%	± 0.04	5.39%	± 0.61	2.11%	± 0.43	3.98%	± 0.39
High risk^b	611	215	826	1.25%	± 0.10	0.33%	± 0.04	0.72%	± 0.07	17.98%	± 1.04	8.40%	± 1.83	13.86%	± 0.70
Intermediate risk^c	711	471	1182	1.45%	± 0.11	0.72%	± 0.07	1.04%	± 0.08	20.92%	± 1.11	18.39%	± 1.16	19.84%	± 0.80
Low risk^d	650	579	1229	1.33%	± 0.10	0.89%	± 0.07	1.08%	± 0.08	19.13%	± 1.07	22.61%	± 1.25	20.62%	± 0.81

Screening type. Two classes: Initial, the participant undergoes for the first time; and Successive if the participant has already participated.

High risk (HR). Presence of an adenoma ≥ 20 mm or at least five adenomas of any size.

Low risk (LR). Presence of one or two adenomas less than 10 mm and without high-grade dysplasia or a villous component.

FIT: faecal immunochemical test.

In the next strategy (two samples, positive when the mean was ≥20 µg Hb/g faeces), positivity fell to 6.0% (Table 3), being 8.1% in initial and 4.5% in successive participants. However, the CRC detection rate only fell from 0.23% to 0.22%; from 0.41% to 0.39% in initial and from 0.10% to 0.08% in successive participants. In contrast, positivity for HRA fell from 0.97% to 0.81%; from 1.59% to 1.37% in initial and from 0.51% to 0.39% in successive participants. The differences in the IRA and LRA were greater.

Table 3.

Distribution of each studied strategy. By screening type, age, and sex, of the main results: Positivity.

Strategy: Two FIT: positive when either ≥20 µg Hb/g faeces
Variables	Initial^a							Successive^a							Total
	<60 years			≥60 years			Initial – Total	<60 years			≥60 years			Successive – Total
	Women	Men	Total	Women	Men	Total	Initial – Total	Women	Men	Total	Women	Men	Total	Successive – Total
Positive FIT RESULT	6.5%	9.8%	8.1%	10.1%	14.5%	12.1%	10.6%	4.9%	6.6%	5.7%	6.5%	8.4%	7.3%	6.6%	8.3%
Cancer	0.14%	0.24%	0.19%	0.33%	0.80%	0.55%	0.41%	0.02%	0.10%	0.06%	0.07%	0.21%	0.13%	0.10%	0.23%
HR^b	0.44%	1.37%	0.89%	1.02%	3.14%	2.01%	1.59%	0.21%	0.46%	0.33%	0.35%	1.06%	0.67%	0.51%	0.97%
IR^c	1.04%	2.37%	1.69%	1.67%	3.04%	2.31%	2.07%	0.55%	1.33%	0.91%	0.94%	1.82%	1.33%	1.14%	1.54%
LR^d	0.88%	1.77%	1.31%	2.25%	2.85%	2.53%	2.07%	0.98%	1.65%	1.28%	1.56%	2.00%	1.75%	1.54%	1.77%
Strategy: Two FIT: positive when the mean ≥20 µg Hb/g faeces
	Initial							Successive							Total
	<60 years			≥60 years				<60 years			≥60 years			Successive – Total
Variables	Women	Men	Total	Women	Men	Total	Initial – Total	Women	Men	Total	Women	Men	Total	Successive – Total
Positive FIT RESULT	4.6%	7.5%	6.0%	7.5%	11.4%	9.3%	8.1%	3.3%	4.6%	3.9%	4.4%	5.8%	5.0%	4.5%	6.0%
Cancer	0.13%	0.23%	0.18%	0.32%	0.76%	0.53%	0.39%	0.01%	0.10%	0.05%	0.06%	0.17%	0.11%	0.08%	0.22%
HR^b	0.36%	1.28%	0.81%	0.86%	2.71%	1.73%	1.38%	0.17%	0.33%	0.24%	0.29%	0.81%	0.52%	0.39%	0.81%
IR^c	0.76%	1.88%	1.30%	1.33%	2.43%	1.85%	1.64%	0.44%	0.95%	0.67%	0.62%	1.24%	0.89%	0.79%	1.16%
LR^d	0.58%	1.22%	0.89%	1.51%	2.04%	1.76%	1.43%	0.65%	1.13%	0.87%	0.99%	1.33%	1.14%	1.02%	1.20%
Strategy: One FIT ≥20 µg Hb/g faeces
	Initial							Successive							Total
	<60 years			≥60 years				<60 years			≥60 years			Successive – Total
Variables	Women	Men	Total	Women	Men	Total	Initial – Total	Women	Men	Total	Women	Men	Total
Positive FIT RESULT	4.1%	6.6%	5.3%	6.8%	10.1%	8.4%	7.2%	2.9%	4.1%	3.5%	3.9%	5.1%	4.5%	4.0%	5.4%
Cancer	0.12%	0.20%	0.16%	0.28%	0.74%	0.50%	0.37%	0.02%	0.10%	0.05%	0.05%	0.18%	0.11%	0.08%	0.21%
HR^b	0.37%	1.19%	0.77%	0.81%	2.43%	1.57%	1.26%	0.15%	0.36%	0.25%	0.28%	0.69%	0.46%	0.36%	0.75%
IR^c	0.71%	1.63%	1.16%	1.25%	2.17%	1.68%	1.48%	0.36%	0.84%	0.58%	0.58%	1.15%	0.83%	0.72%	1.05%
LR^d	0.46%	1.02%	0.73%	1.33%	1.69%	1.50%	1.21%	0.60%	0.96%	0.76%	0.95%	1.15%	1.04%	0.91%	1.04%

Detection rates for neoplasia and participants with adenoma classified as high (HR), intermediate (IR), and low (LR) risk according to the characteristics of the adenoma detected during the study period (01 July 2014 to 31 December 2018) in the Region of Murcia, Spain.

Screening type. Two classes: Initial, the participant undergoes for the first time; and Successive if the participant has already participated.

High risk adenomas (HRA). Presence of an adenoma ≥ 20 mm or at least five adenomas of any size.

Intermediate risk adenomas (IR). Presence of an adenoma ≥10 mm and less than 20 mm, or three or four adenomas less than 10 mm, or at least one adenoma less than 20 mm with high-grade dysplasia or a villous component.

Low risk adenomas (LR). Presence of one or two adenomas less than 10 mm and without high-grade dysplasia or a villous component.

FIT: faecal immunochemical test.

In the one-test strategy (in which one of the two FIT results was examined), with the same threshold (Table 3), the positivity fell to 5.4%; 7.2% in initial and 4.0% in successive participants. Here, the CRC detection rate fell from 0.23% to 0.21%; from 0.41% to 0.37% in initial and from 0.10% to 0.08% in successive participants. The detection rates of HRA, IRA and LRA showed a noteworthy reduction.

The estimates of the possible changes in strategy maintaining the f-Hb threshold at ≥20 µg Hb/g faeces (Table 4) showed:

Population A (two FIT, positive when either is ≥20 µg Hb/g faeces) versus Population B (positive when the mean is ≥20 µg Hb/g faeces): with the same positivity, the invited population may increase by 37.81%, also achieving, if the correction for false negative FIT results is not included in the comparison strategy with regard to the reference strategy, an increase of 28.49% in CRC detected, 15.43% more HR, 3.62% more IR and reducing the detection of LR by 6.82%.

Population A (two FIT, positive when one ≥20 µg Hb/g faeces) versus Population B (one FIT ≥20 µg Hb/g faeces): with the same positivity, the invited population might increase by 56.20%, achieving an increase of 39.17% in CRC detected, and 16.45% more in HR, 5.32% more in IR and a 4.78% reduction in LR.

Population A (two FIT, positive when the mean is ≥20 µg Hb/g faeces) versus Population B (one FIT ≥20 µg Hb/g faeces): with the same positivity, the invited population might increase by 13.34%, achieving an increase of 8.31% in CRC detected, 0.88% in HR, 1.65% in IR and 2.20% in LR.

Table 4.

Distribution of the proportional detection rate increases (Δ%) by changing the screening strategy on invited population; invasive cancer detection, and high, intermediate, and low risk adenoma detected.

Strategy change: from two FIT, positive when either FIT is ≥20 µg Hb/g faeces, to two FIT, positive when the mean FIT is ≥20 µg Hb/g faeces.
Results	Δ% Non-Corrected	CI	Δ% Corrected	CI
Invited population	37.81%	± 0.28	37.8%	± 0.28
Cancer	28.49%	± 0.04	19.2%	± 0.04
High risk^a	15.43%	± 0.08	−7.0%	± 0.07
Intermediate risk^b	3.62%	± 0.09	−30.6%	± 0.09
Low risk^c	−6.82%	± 0.10	−51.5%	± 0.09
Strategy change: from two FIT, positive when either FIT is ≥20 µg Hb/g faeces, to one FIT, positive when ≥20 µg Hb/g faeces.
Results	Δ% Non-Corrected	CI	Δ% Corrected	CI
Invited population	56.20%	± 0.29	56.2%	± 0.29
Cancer	39.17%	± 0.04	22.1%	± 0.04
High risk^a	16.45%	± 0.07	−23.3%	± 0.06
Intermediate risk^b	5.32%	± 0.08	−45.6%	± 0.07
Low risk^c	−4.78%	± 0.08	−65.8%	± 0.07
Strategy change: from two FIT, positive when the mean FIT is ≥20 µg Hb/g faeces, to one FIT, positive when ≥20 µg Hb/g faeces.
Results	Δ% Non-Corrected	CI	Δ% Corrected	CI
Invited population	13.34%	± 0.20	13.34%	± 0.20
Cancer	8.31%	± 0.04	3.29%	± 0.04
High risk^a	0.88%	± 0.07	−11.57%	± 0.07
Intermediate risk^b	1.65%	± 0.08	−10.05%	± 0.08
Low risk^c	2.20%	± 0.08	−8.94%	± 0.08

Data shown with corrected and non-corrected false negative FIT results, with 95% CI. Data from Region of Murcia's colorectal cancer screening programme (01 July 2014 to 31 December 2018).

High risk. Presence of an adenoma ≥ 20 mm or at least five adenomas of any size.

Intermediate risk. Presence of an adenoma ≥10 mm and less than 20 mm, or three or four adenomas less than 10 mm, or at least one adenoma less than 20 mm with high-grade dysplasia or a villous component.

Low risk. Presence of one or two adenomas less than 10 mm and without high-grade dysplasia or a villous component.

FIT: faecal immunochemical test.

In populations for which the programme is not fully rolled out, by changing from the one to two FIT strategy, positive when the mean is ≥20 µg Hb/g faeces (Table 5), there would be a proportional increase in positivity of 13.34%. For CRC, there would be an increase in detected cases of 4.64% and a reduction in the PPV for CRC of 7.68%. The increase in the detection rate of HR would be 12.35% and of IR 11.51%.

Table 5.

Screening strategy change effects on FIT positivity, cancer detection, and adenoma detection (3 risk categories), shown as proportional detection rate variation (Δ%) and positive predictive value variation (Δ%).

Strategy change: from two FIT, positive when the mean FIT is ≥20 µg Hb/g faeces, to two FIT, positive when either is ≥20 µg Hb/g faeces.
Results	Δ% Rates	CI	Δ% PPV	CI
Positivity	37.81%	± 0.28
Cancer	7.26%	± 0.13	−22.17%	± 0.51
High risk^a	19.40%	± 0.20	−13.36%	± 1.00
Intermediate risk^b	33.00%	± 0.24	−3.49%	± 1.21
Low risk^c	47.91%	± 0.26	7.32%	± 1.27
Strategy change: from one FIT, positive when ≥20 µg Hb/g faeces, to two FIT, positive when either FIT is ≥20 µg Hb/g faeces.
Results	Δ% Rates	CI	Δ% PPV	CI
Positivity	56.20%	± 0.25
Cancer	12.24%	± 0.15	−28.15%	± 0.53
High risk^a	34.14%	± 0.23	−14.12%	± 1.04
Intermediate risk^b	48.31%	± 0.24	−5.05%	± 1.25
Low risk^c	64.04%	± 0.23	5.02%	± 1.32
Strategy change: from one FIT, positive when ≥20 µg Hb/g faeces, to two FIT, positive when the mean FIT is ≥20 µg Hb/g faeces.
Results	Δ% Rates	CI	Δ% PPV	CI
Positivity	13.34%	± 0.25
Cancer	4.64%	± 0.12	−7.68%	± 0.64
High risk^a	12.35%	± 0.19	−0.88%	± 1.18
Intermediate risk^b	11.51%	± 0.19	−1.62%	± 1.38
Low risk^c	10.90%	± 0.19	−2.15%	± 1.38

Data assumes the full extension of the programme with the same number of participants in the three compared strategies (95% CI). Data from Region of Murcia's colorectal cancer screening programme (01 July 2014 to 31 December 2018).

High risk. Presence of an adenoma ≥ 20 mm or at least five adenomas of any size.

Low risk. Presence of one or two adenomas less than 10 mm and without high-grade dysplasia or a villous component.

FIT: faecal immunochemical test.

Discussion

The design and delivery of the screening programme studied here differ somewhat from others in Spain,³¹ and other countries.³² The positivity was greater, as were detection rates (particularly those in participants with HR and IR though less significantly for CRC), particularly when compared to longer running programmes, perhaps mainly due to the different screening strategies. Although the f-Hb threshold for positivity is often ≥20 µg Hb/g faeces, our programme uses two FITs collected from consecutive bowel motions, being termed positive when either is ≥20 µg Hb/g faeces. This yields a high positivity, which necessitates a high number of colonoscopies: this has hampered the rapid roll-out of the programme to the entire target population in the Region. As in other programmes, the positivity and neoplasia detection rates varied depending on the screening round, age, and sex. In general, these were greater in those undergoing initial as opposed to successive screenings, men as opposed to women, and participants ≥60 years versus those <60 years. In the first round, all participants are undergoing an initial (prevalence) screening, whereas in the successive (incident) rounds there are increasingly more participants (Table 1); moreover, participants in initial screening are younger, meaning that there are important overall reductions in positivity and detection rates over time.

As expected, a change of strategy meant changes in the positivity, neoplasia detection rates and PPV (Table 2). Thus, on changing from the two FIT strategy in which the result is regarded as positive when at least one is ≥20 µg Hb/g faeces to positive when the mean is ≥20 µg Hb/g faeces, or to a one FIT ≥20 µg Hb/g faeces strategy, there is a reduction in positivity (8.31%; 5.93% and 5.22%, respectively) which is statistically significant; the CI do not overlap. In contrast, the CRC detection rates show very small differences (0.23%; 0.22% and 0.21%), with overlapping CI, although the 0.02% difference in the rate is equivalent to 29 fewer CRC detected during the study. However, the reduction in the aggregated HR plus IR detection rates is significant: 2.51%, 1.97% and 1.76%, whose CI do not overlap, which could lead to a loss in programme capacity to reduce incidence.³² Finally, there is a gain in PPV (for all CRC, HR, IR plus LR combined detection): 54.3%; 57.1% and 58.3% respectively, and therefore a somewhat greater clinical efficiency of colonoscopy. Interestingly, the results of the one FIT strategy are comparable to those of other programmes that use the same strategy^31,32: it is therefore plausible that our results and others³³ using two FITs are transferable to other programmes.

Toyoshima et al. have recently undertaken a comparison of the results obtained using two FITs, with a result being declared positive when either f-Hb is above the threshold, with using one FIT and no FIT. As expected, using two FITs gives a much higher yield of more advanced colorectal neoplasia than the other two strategies. In contrast to our work, the capacity of the health system to cope with the increase in colonoscopies due to the higher detection rate was not considered.³⁴ Turvill et al. have also recently contributed showing no improvement in diagnostic yield using one or two FITs, but their study assessed FIT in patients presenting with symptoms, not in an asymptomatic screening population.³⁵

A germane question is to what extent would a change in strategy that reduces the positivity and allows an increase in the participating population without changing the total number of participants with positive FIT results (and therefore the same colonoscopy requirement) affect neoplasia detection rates. This study has shown (Table 4) that either of the two alternatives substantially increases the proportion of those that can be invited and the detection of CRC, HR and LR when there is no correction for false negative FIT results, although the increased detection of HR and IR is lost when the correction is applied. In our opinion, these results justify a change of strategy, both on account of the greater capacity to detect neoplasia and increasing equality, since coverage can be broadened with the same colonoscopy requirement. Thus, which of the two alternatives should be chosen? If false positive test result correction is not used, the one FIT alternative is clearly better. On the other hand, if correction is used, both alternatives have their pros and cons. In the alternative where the mean of two FITs is used, the increased population coverage and CRC detection rate are somewhat lower, although presenting a greater detection rate for participants with HR and IR, likely meaning that the primary prevention capacity is greater.

Finally, when the programme has reached, or approached, 100% coverage rate, is a change of strategy reasonable? According to this study, perhaps. Indeed, in one FIT programmes, switching to two FIT, where the result is positive when the mean is above the set threshold (Table 5), increases the CRC detection rate (4.64%) and that of other significant neoplasia (12.35% HR and 11.51% IR), whose excision has been associated with a lower incidence of CRC.^36,37 Although the number of positive FIT results increases (13.34%), this does not have a substantial effect on colonoscopy (PPV: −7.68%, −0.88% and −1.62%, respectively). In fact, this increase in colonoscopy requirement can be absorbed by the programme's diagnostic confirmation strategies, since the increased workload is lower than the reduction in participants with FIT positive results due to a greater population undergoing successive screening. Indeed, as shown (Table 2), the positivity in participants undergoing initial screening when one FIT is used (6.95%) is greater than the total positivity when two FITs are used and the mean is used for classification (5.93%). Moreover, as the rounds in a programme progress, the population ≥60 years is almost exclusively undergoing successive screening (with lower positivity) whereas the younger population is undergoing initial screening and the positivity is also lower than at older ages.

One very interesting aspect is to consider the distribution of the results of the three strategies across the different categories of type of screening, age, and sex. That said, a weakness of this study is our lack of data in some of the subgroups (divided by sex, age and colonoscopy findings), which makes our statistical analysis less powerful. Another drawback is that the study was performed using data from an organised programme and not designed as a specific research study. Therefore, although there do exist protocols for performance of colonoscopy, there may be greater inter-observer differences than in research studies. However, this could also be a strength of the study since the data are from an organised programme, reflecting actual practice.

It is important to consider that the characteristics of the one FIT strategy, in which one FIT result is selected, could impact the participation rate in the screening programme, since greater participation might be expected with one FIT required than with two, and therefore this could result in an artificially reduced CRC and adenoma detection rate. Nevertheless, careful revision of similar studies showed that the differences in participation in screening programmes using either one or two FITs were either insignificant or low, and thus the number of FITs would not be likely to have a major impact on our findings.^38–42

In conclusion, in a situation in which colonoscopy resources are limited, thus hampering the rapid roll-out of a screening programme to the entire population, beginning the programme with one FIT seems a good strategy, with the threshold for positivity being >20 µg Hb/g faeces. In programmes that have been initiated with two FITs but are not fully rolled out, it is reasonable to use the mean of two FITs as the criterion for positivity. In fully rolled out programmes, switching from one to two FITs, considering the FIT result as positive when the mean is above the threshold, improves detection rates of CRC, HR, and IR without increasing the colonoscopy demand for further investigation, treatment and surveillance.

Supplemental Material

sj-docx-1-msc-10.1177_09691413221094919 - Supplemental material for One or two faecal immunochemical tests in an organised population-based colorectal cancer screening programme in Murcia (Spain)

Supplemental material, sj-docx-1-msc-10.1177_09691413221094919 for One or two faecal immunochemical tests in an organised population-based colorectal cancer screening programme in Murcia (Spain) by Carlos Tourne-Garcia, Francisco Perez-Riquelme, Olga Monteagudo-Piqueras, Callum G Fraser and Pedro Yepes-Garcia in Journal of Medical Screening

Supplemental Material

sj-docx-2-msc-10.1177_09691413221094919 - Supplemental material for One or two faecal immunochemical tests in an organised population-based colorectal cancer screening programme in Murcia (Spain)

Supplemental material, sj-docx-2-msc-10.1177_09691413221094919 for One or two faecal immunochemical tests in an organised population-based colorectal cancer screening programme in Murcia (Spain) by Carlos Tourne-Garcia, Francisco Perez-Riquelme, Olga Monteagudo-Piqueras, Callum G Fraser and Pedro Yepes-Garcia in Journal of Medical Screening

Supplemental Material

sj-docx-3-msc-10.1177_09691413221094919 - Supplemental material for One or two faecal immunochemical tests in an organised population-based colorectal cancer screening programme in Murcia (Spain)

Supplemental material, sj-docx-3-msc-10.1177_09691413221094919 for One or two faecal immunochemical tests in an organised population-based colorectal cancer screening programme in Murcia (Spain) by Carlos Tourne-Garcia, Francisco Perez-Riquelme, Olga Monteagudo-Piqueras, Callum G Fraser and Pedro Yepes-Garcia in Journal of Medical Screening

Footnotes

Authors’ Contributions

CTG was the main contributor on the Methods section and one of the coordinators of the screening programme. FPR was the main contributor on the Results and Discussion section, and the main responsible of the screening programme. OMP contributed significantly to the analysis, interpretation of the data and quality control of the study database, and was the main coordinator of the screening programme. CGF contributed significantly to the analysis and interpretation of the data and to the writing of the paper. PYG was the main contributor on the Introduction section and helped with the coordination of all the previous authors. All authors approved of the final draft submitted.

Patient Consent for Publication

Not required. It was not appropriate or possible to involve patients or the public in the design, or conduct, or reporting, or dissemination plans of our research.

Data Availability Statement

Data are available upon reasonable request to the corresponding author.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Callum G Fraser

Pedro Yepes-Garcia

Supplemental material

Supplemental material for this article is available online.

References

Kim

. Colorectal cancer globocan 2018. IARC. WHO. 2020; 66: 1–9.

Wilson

JMG

Jungner

. Principles and practice of screening for disease. WHO Public Paper 34. Geneva: Geneva World Health Biblio Organ, 1968, p. 168.

Zhang

Cheng

, et al. Effectiveness of screening modalities in colorectal cancer: a network meta-analysis. Clin Colorectal Cancer [Internet]. 2017; 16: 252–263.

Von Karsa

Patnick

Segnan

, et al. European Guidelines for quality assurance in colorectal cancer screening and diagnosis: overview and introduction to the full supplement publication. Endoscopy 2013; 45: 51–59.

The Council of the European Union. Council recommendation of 2 December 2003 on cancer screening. Off J Eur Union 2003; 327: 34–38.

Cardoso

Guo

Heisser

, et al. Colorectal cancer incidence, mortality, and stage distribution in European countries in the colorectal cancer screening era: an international population-based study. Lancet Oncol 2021; 22: 1002–1013.

Grupo de trabajo de la Ponencia de Cribado de la Comisión de Salud Pública. Documento marco sobre cribado poblacional. Minist Sanid y Política Soc España [Internet]. 2010; 35: 3–10. [Accessed on 29 Mar 2022]. Available from: https://www.sanidad.gob.es/profesionales/saludPublica/prevPromocion/Cribado/docs/Cribado_poblacional.pdf

Ministerio de Sanidad. Informe del grupo de expertos sobre concreción de cartera común de servicios para cribado de cáncer. Minist Sanid y Política Soc España [Internet]. 2013; 20: 14–16. [Accessed on 29 Mar 2022]. Available from: https://www.sanidad.gob.es/profesionales/saludPublica/prevPromocion/Cribado/docs/ResumenEjecutivoCribadoCancer.pdf

Haute Autorité du Santé France. Cancer colorectal: modalités de dépistage et de prévention chez les sujets à risque élevé et très élevé. [Internet]. 2017. [Accessed on 29 Mar 2022]. Available from: https://www.has-sante.fr/upload/docs/application/pdf/2017-06/dir71/fiche_memo_ccr.pdf

10.

Bowel Cancer UK. Bowel cancer screening. [Internet]. [Accessed on 29 Mar 2022]. Available from: https://www.bowelcanceruk.org.uk/about-bowel-cancer/screening/

11.

Toes-Zoutendijk

van Leerdam

Dekker

, et al. Real-time monitoring of results during first year of Dutch colorectal cancer screening program and optimization by altering fecal immunochemical test cut-off levels. Gastroenterology 2017; 152: 767–775.e2.

12.

Dorizzi

. Screening oncologici. Riv Ital Della Med di Lab 2007; 3: 222–223.

13.

Perez-Riquelme

Peris-Tuser

Cruzado-Quevedo

, et al. Recommendations of the national panel of experts for the planning and start-up of demographic based programmes on colorectal cancer prevention. Murcia. Perez-Riquelme F, editor. Educación para la Salud informes. Consejería de Sanidad. 2007. 1–48p.

14.

Steele

McDonald

Digby

, et al. Clinical outcomes using a faecal immunochemical test for haemoglobin as a first-line test in a national programme constrained by colonoscopy capacity. United Eur Gastroenterol J 2013; 1: 198–205.

15.

Mousavinezhad

Majdzadeh

Sari

, et al. The effectiveness of FOBT vs. FIT: a meta-analysis on colorectal cancer screening test. Med J Islam Repub Iran 2016; 30: 366.

16.

Tinmouth

Lansdorp-Vogelaar

Allison

. Faecal immunochemical tests versus guaiac faecal occult blood tests: what clinicians and colorectal cancer screening programme organisers need to know. Gut 2015; 64: 1327–1337.

17.

Fraser

Digby

McDonald

, et al. Experience with a two-tier reflex gFOBT/FIT strategy in a national bowel screening programme. J Med Screen 2012; 19: 8–13.

18.

Gibson

Mooney

, et al. Impact of a higher fecal immunochemistry test cut-off on pathology detected in subsequent rounds of a colorectal screening program. Gastrointest Endosc 2019; 89: 518–522.

19.

Selby

Jensen

Lee

, et al. Influence of varying quantitative fecal immunochemical test positivity thresholds on colorectal cancer detection: a community-based cohort study. Ann Intern Med 2018; 169: 439–447.

20.

Cha

Lee

JIl

Joo

, et al. Use of a low cut-off value for the fecal immunochemical test enables better detection of proximal neoplasia. Dig Dis Sci 2013; 58: 3256–3262.

21.

Arana-Arri

Idigoras

Uranga

, et al. Population-based colorectal cancer screening programmes using a faecal immunochemical test: should faecal haemoglobin cut-offs differ by age and sex? BMC Cancer 2017; 17: 1–13.

22.

McDonald

Strachan

Digby

, et al. Faecal haemoglobin concentrations by gender and age: implications for population-based screening for colorectal cancer. Clin Chem Lab Med 2012; 50: 935–940.

23.

Kapidzic

van der Meulen

Hol

, et al. Gender differences in fecal immunochemical test performance for early detection of colorectal neoplasia. Clin Gastroenterol Hepatol 2015; 13: 1464–1471.

24.

Chen

SLS

Hsu

Yen

AMF

, et al. Demand for colonoscopy in colorectal cancer screening using a quantitative fecal immunochemical test and age/sex-specific thresholds for test positivity. Cancer Epidemiol Biomarkers Prev 2018; 27: 704–709.

25.

Cooper

Parsons

Stinton

, et al. Risk-adjusted colorectal cancer screening using the FIT and routine screening data: development of a risk prediction model. Br J Cancer 2018; 118: 285–293.

26.

Scientific Advice for Policy by European Academies. Cancer screening – SAPEA. [Internet]. Germany. [Accessed on 3 Mar 2022]. Available from: https://sapea.info/topic/cancer-screening/

27.

Lee

Liles

Bent

, et al. Accuracy of fecal immunochemical tests for colorectal cancer. Ann Intern Med 2014; 160: 171–181.

28.

Goede

Van Roon

AHC

Reijerink

JCIY

, et al. Cost-effectiveness of one versus two sample faecal immunochemical testing for colorectal cancer screening. Gut 2013; 62: 727–734.

29.

Perez-Riquelme

Navarro-Sanchez

Chirlaque

, et al. Informe sobre prevención del cáncer de colon y recto en la Región de Murcia. Cons Sanid Región Murcia. 2004; 1–50.

30.

Pérez-Riquelme

Cruzado-Quevedo

Gutierrez-Garcia

. La Prevención del Cáncer de Colon y Recto en la Región de Murcia. In: Perez-Riquelme

Cruzado-Quevedo

Gutierrez-Garcia

(eds) Educación para la Salud informes. Vol. 50. Murcia: Consejería de Sanidad, 2008.

31.

Red de programas de Cribado de Cáncer. Evaluación programas de cribado de cáncer colorrectal. Informe de Evaluación 2017. 2019; 2017: 1–53. Available from: www.cribadocancer.es

32.

Navarro

Nicolas

Ferrandez

, et al. Colorectal cancer population screening programs worldwide in 2016: an update. World J Gastroenterol 2017; 23: 3632–3642.

33.

Allison

Ransohoff

Pignone

, et al. The effect of fecal occult-blood screening on the incidence of colorectal cancer. N Eng J Med 2001; 345: 1850.

34.

Toyoshima

Yamaji

Nishizawa

, et al. Priority stratification for colonoscopy based on two-sample faecal immunochemical test screening: results from a cross-sectional study at an endoscopy clinic in Japan. BMJ Open 2021; 11: e046055.

35.

Turvill

Mellen

Jeffery

, et al. Diagnostic accuracy of one or two faecal haemoglobin and calprotectin measurements in patients with suspected colorectal cancer. Scand J Gastroenterol 2018; 53: 1526–1534.

36.

Mandel

Church

Bond

, et al. The effect of fecal occult-blood screening on the incidence of colorectal cancer. N Engl J Med 2000; 343: 1603–1607.

37.

Shaukat

Shyne

Mandel

, et al. Colonoscopy with polypectomy reduces long-term incidence of colorectal cancer in both men and women: extended results from the Minnesota Colon Cancer Control Study. Gastroenterology 2021; 160: 1397–1419.

38.

van Roon

Wilschut

Hol

, et al. Diagnostic yield improves with collection of 2 samples in fecal immunochemical test screening without affecting attendance. Clin Gastroenterol Hepatol. 2011; 9: 333–339.

39.

Mosen

Liles

Feldstein

, et al. Participant uptake of the fecal immunochemical test decreases with the two-sample regimen compared with one-sample FIT. Eur J Cáncer Prev. 2014; 23: 516–512.

40.

van Roon

AHC

Goede

van Ballegooijen

, et al. Random comparison of repeated faecal immunochemical testing at different intervals for population-based colorectal cancer screening. Gut. 2013; 62: 409–415.

41.

Schreuders

Grobbee

Nieuwenburg

SAV

, et al. Multiple rounds of one sample versus two sample faecal immunochemical test-based colorectal cancer screening: a population-based study. Lancet Gastroenterol Hepatol. 2019; 4: 622–631.

42.

Red de programas de Cribado de Cáncer. Indicadores cáncer colorrectal – red de programas de Cribado de Cáncer [internet]. Spain. [Accessed on 3 Mar 2022]. Available from: https://cribadocancer.es/indicadores-cancer-colorrectal/

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB