Accepted Manuscript Uptake of Colon Capsule Endoscopy vs Colonoscopy for Screening Relatives of Patients with Colorectal Cancer Zaida Adrián-de-Ganzo, MD, Onofre Alarcón-Fernández, MD, Laura Ramos, MD, Antonio Gimeno-García, MD, PhD, Inmaculada Alonso-Abreu, MD, Marta Carrillo, MD, PhD, Enrique Quintero, MD, PhD. PII: DOI: Reference:
S1542-3565(15)00908-8 10.1016/j.cgh.2015.06.032 YJCGH 54356
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 9 June 2015 Please cite this article as: Adrián-de-Ganzo Z, Alarcón-Fernández O, Ramos L, Gimeno-García A, Alonso-Abreu I, Carrillo M, Quintero E, Uptake of Colon Capsule Endoscopy vs Colonoscopy for Screening Relatives of Patients with Colorectal Cancer, Clinical Gastroenterology and Hepatology (2015), doi: 10.1016/j.cgh.2015.06.032. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. All studies published in Clinical Gastroenterology and Hepatology are embargoed until 3PM ET of the day they are published as corrected proofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs.
Uptake of Colon Capsule Endoscopy vs Colonoscopy for Screening Relatives of Patients with Colorectal Cancer ACCEPTED MANUSCRIPT
Zaida Adrián-de-Ganzo MD, Onofre Alarcón-Fernández MD, Laura Ramos MD, Antonio Gimeno-García MD, PhD, Inmaculada Alonso-Abreu MD, Marta Carrillo MD, PhD, and Enrique Quintero MD, PhD. de
Gastroenterología, Hospital
Universitario
de
Canarias.
Instituto
RI PT
Servicio
Universitario de Tecnologías Biomédicas (ITB) & Centro de Investigación Biomédica de Canarias (CIBICAN). Departamento de Medicina Interna. Universidad de La
SC
Laguna. Tenerife. España Grant support.
M AN U
This study was supported by a grant from the European Capsule Endoscopy Group (ECEG), by a grant from Caja Canarias Banca Cívica and by funds of the Department of Internal Medicine of La Laguna University. Abbreviations.
CCR: colorectal cancer
TE D
CCE: colon capsule endoscopy
CI: confidence intervals
AC C
OR: odds ratios
EP
FDR: first-degree relatives
Disclosures: None of the authors have any potential conflict of interest to disclose.
Author Contributions
Enrique Quintero was responsible for the conception and design of the trial. Zaida Adrián-de-Ganzo and Enrique Quintero took responsibility for the integrity of the data,
1
ACCEPTED MANUSCRIPT performed the analysis and interpretation of the results and were involved in the drafting of the article. Zaida Adrián-de-Ganzo, Onofre Alarcón, Antonio Z Gimeno-García, Laura Ramos, Inmaculada Alonso-Abreu and Marta Carrillo were involved in the acquisition of data. All the authors performed a critical revision of the article for
RI PT
important intellectual content and gave their final approval of the manuscript submitted.
Acknowledgements.
SC
The authors gratefully acknowledge Dr. Victor Abraira and Dr. Alfonso Muriel (Unidad de Bioestadística, Hospital Ramón y Cajal, Madrid) for assistance on the statistical
Correspondence: Enrique Quintero M.D. Department of Gastroenterology.
M AN U
analysis and critical reading of the manuscript.
TE D
Hospital Universitario de Canarias. Ctra. Ofra S/N La Cuesta,
38320 La Laguna, Tenerife, Spain.
EP
Email: [email protected]
AC C
Word count: 3,993 words
2
ACCEPTED MANUSCRIPT
ABSTRACT Background & Aims: The efficacy of screening colonoscopy in first-degree relatives (FDRs) of patients with colorectal cancer (CRC) is limited by suboptimal uptake. We compared
RI PT
screening uptake of colon capsule endoscopy (CCE) vs colonoscopy in this population.
Methods: We performed a prospective study of 329 asymptomatic FDRs of patients with CRC who were randomly assigned to groups examined by CCE (PillCam, 2nd generation;
SC
n=165) or colonoscopy (n=164) at a tertiary hospital in Spain, from July 2012 through December 2013. Crossover was permitted for patients who did not wish to undergo the
M AN U
assigned procedure. Subjects assigned to CCE who had a significant lesion (polyp ≥10 mm, >2 polyps of any size, or CRC) were invited to undergo colonoscopy. Results: One-hundred twenty subjects in the CCE group and 113 in the colonoscopy group were eligible for inclusion. In the intention-to-screen analysis, uptake was similar between
TE D
groups (55.8% CCE vs 52.2% colonoscopy; odds ratio [OR], 0.86; 95% confidence interval [CI], 0.51−1.44; P=.57); 57.4% of subjects crossed-over from the CCE group and 30.2% crossed-over from the colonoscopy group (OR, 3.11; 95% CI, 1.51−6.41; P=.002).
EP
Unwillingness to repeat bowel preparation in the case of a positive result was the main reason that subjects assigned to the CCE group crossed over; fear of colonoscopy was the reason that
AC C
most patients in this group crossed over. A significant lesion was detected in 14 subjects (11.7%) in the CCE group and 13 subjects (11.5%) in the colonoscopy group (OR, 1.02; 95% CI, 0.45−2.26; P=.96).
Conclusion: In a prospective study, similar numbers of FDRs of patients with CRC assigned to undergo CCE or colonoscopy agreed to participate, but most preferred to undergo colonoscopy. CCE was as effective as colonoscopy in detecting significant lesions; it could be a valid rescue strategy for subjects who reject screening colonoscopy.
3
ACCEPTED MANUSCRIPT
ClinicalTrials.gov number NCT01557101.
RI PT
KEY WORDS: participation; colon cancer; colon capsule-2; detection; tumor
INTRODUCTION.
Family history is a major risk-factor for developing CRC.1 Risk increases two to five-fold in FDR of patients with CRC if the index-case is younger than 60 years or when 2, 3
Based on this premise, clinical practice guidelines
SC
there are two or more FDR affected.
M AN U
empirically recommend colonoscopy every five years, starting at the age of 40 or ten years less than the index-case at first diagnosis, as the first-line screening strategy in this setting.4-6
Colonoscopy is the gold standard procedure for detecting CRC and advanced adenomas. The advantage of colonoscopy over other screening tests is that polypectomy can
TE D
be performed, thus breaking the adenoma-carcinoma sequence and reducing subsequent development of CRC. However, it is far from being an ideal screening technique: it fails to identify as many as 12% of advanced adenomas and 5% of CRC, 7,
8
it is an invasive or
EP
procedure that carries an increased risk for complications, especially bleeding
perforation following caecal polypectomy,9 and screening uptake is suboptimal with
AC C
participation rates below 40% in the familial risk population. 10-12 For these reasons, alternative-screening strategies are currently under investigation. Colon capsule endoscopy (CCE) is a minimally invasive tool with similar accuracy
for detecting advanced neoplasia as optical colonoscopy.13-16 Currently, CCE is used in clinical practice as an alternative diagnostic tool in patients with non- occlusive, incomplete optical colonoscopy.17 Recently, it has been suggested that CCE may be cost-effective as compared to colonoscopy if it is able to increase CRC screening uptake by 20%. 18 In a pilot
4
ACCEPTED MANUSCRIPT
study in which all individuals were invited to undergo CCE followed by colonoscopy, the authors reported a four-fold higher screening uptake for CCE than for colonoscopy, with both techniques showing the same diagnostic yields for detecting adenomas. However,
RI PT
there are no prospective, randomized, controlled studies comparing CRC screening uptake between colonoscopy and CCE.
To test the hypothesis that CCE may improve CRC screening uptake in the familial-
SC
risk population, we designed a pragmatic, prospective, randomized open trial to compare
M AN U
screening uptake of colonoscopy versus CCE in screening naïve FDR of patients with CRC.
METHODS. Subjects
The study was conducted in a tertiary hospital serving approximately 375.000
TE D
inhabitants between July 2012 and December 2013. Consecutive patients with nonsyndromic CRC diagnosed during the previous two years were interviewed at the high-risk CRC clinic of our hospital. The trial registered as ClinicalTrials.gov number NCT01557101
EP
complied with the CONSORT extension for pragmatic trials.19 Inclusion criteria were: asymptomatic FDR ≥40 years of age or ten years less
AC C
than the youngest relative with CRC. Exclusion criteria were: previous CRC screening; history of colorectal neoplasia or inflammatory bowel disease; hereditary CRC; gastrointestinal surgery; pregnancy; allergy or other severe comorbidity (mean life expectancy of 2 polyps of any size, or CRC); and Group 2: screening with optical colonoscopy. Sealed envelopes were prepared for eligible FDR in the family and
AC C
given to the index-case for their delivery. The envelopes contained a leaflet with detailed information about the trial, the strategy allocated, the informed consent document to participate in the study and a contact telephone number to arrange an appointment at the High Risk CRC Clinic. The possibility of changing group was not mentioned in the information leaflet, to minimize as much as possible crossover in both arms. In that event, a structured questionnaire was used to determine the reason for crossover between the assigned strategies (Table 1).
6
ACCEPTED MANUSCRIPT
Subjects assigned to CCE received information about the procedure (bowel preparation, duration and possible adverse events) and were told that they would be scheduled to undergo colonoscopy in the event of positive result. By contrast, subjects
RI PT
assigned to colonoscopy received detailed information about the technique (bowel cleansing, sedation, advantages of polypectomy in the event of polyp detection, and possible complications). Specific informed consent for the agreed strategy (CCE or conventional
SC
colonoscopy) was sought before the procedure. Both screening procedures were free of charge. The Clinical Research Ethics Committee of our center approved the study protocol
M AN U
and all authors had access to the study data and reviewed and approved the final manuscript
Colon capsule endoscopy.
CCE was performed with second-generation PillCam™ CCE system (PillCam® colon
TE D
capsule, Given Imaging Inc, Yoqneam, Israel) previously described. 13 Colon cleanliness for CCE was scheduled in one-day, low-volume preparation, as previously described, to facilitate adherence to screening. 22, 23 Colon cleanliness was graded for the following colonic
EP
segments: right colon, transverse colon, left colon and rectum, and classified using a validated 4-point scale as previously described:13 (1) excellent (no material or liquid material covering
AC C
10% of the mucosal surface), fair (solid material impossible to suction, covering 10% of the mucosal surface). In the current study a colon preparation rated as “excellent” or “good” was considered as “adequate” whereas colon cleansing rated as “fair” and “poor” was classified as “inadequate”. CCE readings were performed by one independent blinded observer (AG or OA), as previously described.23 Polyp size was estimated by using the polyp size estimation
7
ACCEPTED MANUSCRIPT
tool included in the RAPID software (Given Imaging Ltd, Yoqneam, Israel, version 7.0). CCE findings were collected and a global lesion evaluation for each patient was recorded as: negative examination (no polyps or CCR), polyps, CRC or miscellaneous (diverticula,
RI PT
angiomas, inflammation or hemorrhoids). CCE examination was considered to be valid when the capsule was excreted (or hemorrhoidal plexus was observed), had adequate colonic cleansing or a significant lesion was detected. Subjects with an incomplete CCE
SC
exploration due to poor bowel cleansing were scheduled for optical colonoscopy. Colonoscopy
M AN U
Colonoscopies were carried out by experienced endoscopists as previously described.24 The endoscopic report contained the parameters for quality assessment (caecal intubation, quality of cleanliness for each colorectal segment, number of polyps identified, removed and recovered for histological examination and immediate complications).
TE D
Colonoscopy was considered complete if the cecum was reached and bowel cleansing was good or excellent.24 Colon cleansing was defined as inadequate when less than 90% of colorectal mucosa could be properly evaluated due to the presence of faeces. Patients with
colonoscopy.
EP
incomplete colonoscopy because of poor bowel preparation were re-scheduled to
AC C
Adenomas ≥10 mm in diameter, with tubulovillous architecture, high-grade
dysplasia or intramucosal carcinoma were classified as advanced adenomas. Invasive cancer was considered when malignant cells were observed beyond the muscularis mucosa. Tumor staging was performed according to the classification system of the American Joint Committee on Cancer 25 and patients were classified according to the most advanced lesion. Data collection Demographic characteristics of participants, kinship, degree of bowel cleanliness for
8
ACCEPTED MANUSCRIPT
each procedure, colonoscopy and CCE completion, colonoscopy or CCE findings, reason for failure and reason for crossover between the assigned groups was recorded. The level of patient satisfaction and potential procedure-related adverse effects were assessed using a
RI PT
questionnaire administered by telephone interview two weeks later (Supplementary Table 1). Sample Size
We hypothesized that uptake of CCE screening should be at least 20% higher
SC
than that of colonoscopy, on the basis that this is the minimal estimated difference that would make CCE cost-effective.18 Assuming a participation rate of 38% for colonoscopy12 and of
M AN U
58% for CCE screening, with an alpha risk of 0.05 and a beta risk of 0.20 in a two -sided test, 108 subjects in each arm of the study were needed to find a statistically significant difference.
TE D
Statistical Analysis.
The primary objective of this study was to compare screening uptake of CCE versus optical colonoscopy in FDR of patients with CRC. Screening uptake was defined as
EP
number of invited subjects screened by CCE or colonoscopy as a proportion of eligible subjects. The principal analysis consisted of an intention-to-screen comparison of the uptake
AC C
proportions between both study arms (subjects were analyzed according to their randomly assigned strategy, irrespective of whether they actually received it or not). The detection rate of neoplastic lesions (number of subjects with true positive results divided by the number of subjects who actually underwent testing) was based on as-screened analyses (subjects that underwent screening by colonoscopy or CCE, independently to their randomly assigned strategy). Statistical comparisons for secondary endpoints (efficacy of each procedure to
9
ACCEPTED MANUSCRIPT
explore the entire colon, and acceptance of each strategy) were performed according to asscreened analysis. Comparisons between groups were carried out by logistic-regression analysis and reported as odds ratios (OR) with 95% confidence intervals (CI). Continuous
RI PT
variables were expressed as means and standard deviation. Categorical variables were expressed as frequencies and percentages. Chi-square test was used to compare categorical variables. Differences with a P value 60 g/d, n (%)
36 (56.3)
0.29
68 (51.5)
8 (11.8)
8 (12.5)
0.90
16 (12.1)
0.38
27.9 ± 5.1
19 (29.7)
0.68
37 (28.0)
9 (14.1)
0.73
20 (15.2)
76 (67.3)
0.97
18 (26.5)
Antiplatelet therapy, aspirin
11 (16.2)
TE D
CRC2 Proband, n (%) 81 (67.5) 56.8 ± 8.7
27.2 ± 4.8
57.2 ± 9.19
96 (80.0)
91 (80.5)
24 (20.0)
22 (19.5)
0.57
0.87
112 (48.1)
157
67.4
57 ± 8.95
187
(80.3)
46
(19.7)
AC C
Two or more
EP
One
121 (51.9)
0.22
32 (47.1)
Smoking, n (%)
Relatives with CRC, n (%)
54.9 ± 11.6
59 (52.2)
28.6 ± 5.5
Mean age ± SD
(n=233)
53 (44.2)
BMI ± SD
Age < 60 years
0.91
SC
Male
54 (47.8)
M AN U
Female
p
RI PT
Category
Total
1
CCE = Colon capsule endoscopy
2
CRC = colorectal cancer
23
ACCEPTED MANUSCRIPT
Table 3.- Diagnostic yield of colon capsule endoscopy (CCE) and colonoscopy according to the
CCE Group
Rate
Subjects
Rate
Most advanced lesion
no.
%
no.
%
Intention-to-screen analysis
113
Non-advanced adenomaω
12
10.6
9
≥ 3 non-advanced adenomas
5
4.4
4
Advanced adenoma
8
7.1
9
Significant lesion§
13
11.5
As-screened analysis
81
Non-advanced adenomaω
16
≥ 3 non-advanced adenomas
4
Advanced adenoma
12
Significant lesion §
16
120
Odds Ratio+
M AN U
Subjects
SC
Colonoscopy Group
RI PT
intention-to-screen and as-screened analyses.*
95% CIξ
P value
0.68
0.27 to 1.68
0.41
3.3
0.74
0.19 to 2.84
0.67
7.5
1.06
0.39 to 2.86
0.92
11.7
1.02
0.45 to 2.26
0.96
5
11.1
0.50
0.17 to 1.49
0.22
4.9
5
11.1
2.41
0.61 to 9.46
0.21
14.8
5
11.1
0.72
0.23 to 2.19
0.56
19.8
11∞
24.4
1.31
0.54 to 3.14
0.54
TE D
7.5
14∞
AC C
19.8
EP
45
*In the intention-to-screen analysis the detection rate was calculated as the number of subjects with true positive results divided by the number of subjects who were eligible to undergo testing. In the as-screened analysis the
24
ACCEPTED MANUSCRIPT
RI PT
detection rate was calculated as the number of positive results divided by the number of subjects who actually underwent screening. Subjects were classified according to the most advanced lesion. +
ξ
CI denotes confidence interval.
ω
One or two adenoma 2 polyps of any size or cancer.
M AN U
§
SC
Odds ratios were adjusted according to age, gender, age (
S1542-3565(15)00908-8 10.1016/j.cgh.2015.06.032 YJCGH 54356
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 9 June 2015 Please cite this article as: Adrián-de-Ganzo Z, Alarcón-Fernández O, Ramos L, Gimeno-García A, Alonso-Abreu I, Carrillo M, Quintero E, Uptake of Colon Capsule Endoscopy vs Colonoscopy for Screening Relatives of Patients with Colorectal Cancer, Clinical Gastroenterology and Hepatology (2015), doi: 10.1016/j.cgh.2015.06.032. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. All studies published in Clinical Gastroenterology and Hepatology are embargoed until 3PM ET of the day they are published as corrected proofs on-line. Studies cannot be publicized as accepted manuscripts or uncorrected proofs.
Uptake of Colon Capsule Endoscopy vs Colonoscopy for Screening Relatives of Patients with Colorectal Cancer ACCEPTED MANUSCRIPT
Zaida Adrián-de-Ganzo MD, Onofre Alarcón-Fernández MD, Laura Ramos MD, Antonio Gimeno-García MD, PhD, Inmaculada Alonso-Abreu MD, Marta Carrillo MD, PhD, and Enrique Quintero MD, PhD. de
Gastroenterología, Hospital
Universitario
de
Canarias.
Instituto
RI PT
Servicio
Universitario de Tecnologías Biomédicas (ITB) & Centro de Investigación Biomédica de Canarias (CIBICAN). Departamento de Medicina Interna. Universidad de La
SC
Laguna. Tenerife. España Grant support.
M AN U
This study was supported by a grant from the European Capsule Endoscopy Group (ECEG), by a grant from Caja Canarias Banca Cívica and by funds of the Department of Internal Medicine of La Laguna University. Abbreviations.
CCR: colorectal cancer
TE D
CCE: colon capsule endoscopy
CI: confidence intervals
AC C
OR: odds ratios
EP
FDR: first-degree relatives
Disclosures: None of the authors have any potential conflict of interest to disclose.
Author Contributions
Enrique Quintero was responsible for the conception and design of the trial. Zaida Adrián-de-Ganzo and Enrique Quintero took responsibility for the integrity of the data,
1
ACCEPTED MANUSCRIPT performed the analysis and interpretation of the results and were involved in the drafting of the article. Zaida Adrián-de-Ganzo, Onofre Alarcón, Antonio Z Gimeno-García, Laura Ramos, Inmaculada Alonso-Abreu and Marta Carrillo were involved in the acquisition of data. All the authors performed a critical revision of the article for
RI PT
important intellectual content and gave their final approval of the manuscript submitted.
Acknowledgements.
SC
The authors gratefully acknowledge Dr. Victor Abraira and Dr. Alfonso Muriel (Unidad de Bioestadística, Hospital Ramón y Cajal, Madrid) for assistance on the statistical
Correspondence: Enrique Quintero M.D. Department of Gastroenterology.
M AN U
analysis and critical reading of the manuscript.
TE D
Hospital Universitario de Canarias. Ctra. Ofra S/N La Cuesta,
38320 La Laguna, Tenerife, Spain.
EP
Email: [email protected]
AC C
Word count: 3,993 words
2
ACCEPTED MANUSCRIPT
ABSTRACT Background & Aims: The efficacy of screening colonoscopy in first-degree relatives (FDRs) of patients with colorectal cancer (CRC) is limited by suboptimal uptake. We compared
RI PT
screening uptake of colon capsule endoscopy (CCE) vs colonoscopy in this population.
Methods: We performed a prospective study of 329 asymptomatic FDRs of patients with CRC who were randomly assigned to groups examined by CCE (PillCam, 2nd generation;
SC
n=165) or colonoscopy (n=164) at a tertiary hospital in Spain, from July 2012 through December 2013. Crossover was permitted for patients who did not wish to undergo the
M AN U
assigned procedure. Subjects assigned to CCE who had a significant lesion (polyp ≥10 mm, >2 polyps of any size, or CRC) were invited to undergo colonoscopy. Results: One-hundred twenty subjects in the CCE group and 113 in the colonoscopy group were eligible for inclusion. In the intention-to-screen analysis, uptake was similar between
TE D
groups (55.8% CCE vs 52.2% colonoscopy; odds ratio [OR], 0.86; 95% confidence interval [CI], 0.51−1.44; P=.57); 57.4% of subjects crossed-over from the CCE group and 30.2% crossed-over from the colonoscopy group (OR, 3.11; 95% CI, 1.51−6.41; P=.002).
EP
Unwillingness to repeat bowel preparation in the case of a positive result was the main reason that subjects assigned to the CCE group crossed over; fear of colonoscopy was the reason that
AC C
most patients in this group crossed over. A significant lesion was detected in 14 subjects (11.7%) in the CCE group and 13 subjects (11.5%) in the colonoscopy group (OR, 1.02; 95% CI, 0.45−2.26; P=.96).
Conclusion: In a prospective study, similar numbers of FDRs of patients with CRC assigned to undergo CCE or colonoscopy agreed to participate, but most preferred to undergo colonoscopy. CCE was as effective as colonoscopy in detecting significant lesions; it could be a valid rescue strategy for subjects who reject screening colonoscopy.
3
ACCEPTED MANUSCRIPT
ClinicalTrials.gov number NCT01557101.
RI PT
KEY WORDS: participation; colon cancer; colon capsule-2; detection; tumor
INTRODUCTION.
Family history is a major risk-factor for developing CRC.1 Risk increases two to five-fold in FDR of patients with CRC if the index-case is younger than 60 years or when 2, 3
Based on this premise, clinical practice guidelines
SC
there are two or more FDR affected.
M AN U
empirically recommend colonoscopy every five years, starting at the age of 40 or ten years less than the index-case at first diagnosis, as the first-line screening strategy in this setting.4-6
Colonoscopy is the gold standard procedure for detecting CRC and advanced adenomas. The advantage of colonoscopy over other screening tests is that polypectomy can
TE D
be performed, thus breaking the adenoma-carcinoma sequence and reducing subsequent development of CRC. However, it is far from being an ideal screening technique: it fails to identify as many as 12% of advanced adenomas and 5% of CRC, 7,
8
it is an invasive or
EP
procedure that carries an increased risk for complications, especially bleeding
perforation following caecal polypectomy,9 and screening uptake is suboptimal with
AC C
participation rates below 40% in the familial risk population. 10-12 For these reasons, alternative-screening strategies are currently under investigation. Colon capsule endoscopy (CCE) is a minimally invasive tool with similar accuracy
for detecting advanced neoplasia as optical colonoscopy.13-16 Currently, CCE is used in clinical practice as an alternative diagnostic tool in patients with non- occlusive, incomplete optical colonoscopy.17 Recently, it has been suggested that CCE may be cost-effective as compared to colonoscopy if it is able to increase CRC screening uptake by 20%. 18 In a pilot
4
ACCEPTED MANUSCRIPT
study in which all individuals were invited to undergo CCE followed by colonoscopy, the authors reported a four-fold higher screening uptake for CCE than for colonoscopy, with both techniques showing the same diagnostic yields for detecting adenomas. However,
RI PT
there are no prospective, randomized, controlled studies comparing CRC screening uptake between colonoscopy and CCE.
To test the hypothesis that CCE may improve CRC screening uptake in the familial-
SC
risk population, we designed a pragmatic, prospective, randomized open trial to compare
M AN U
screening uptake of colonoscopy versus CCE in screening naïve FDR of patients with CRC.
METHODS. Subjects
The study was conducted in a tertiary hospital serving approximately 375.000
TE D
inhabitants between July 2012 and December 2013. Consecutive patients with nonsyndromic CRC diagnosed during the previous two years were interviewed at the high-risk CRC clinic of our hospital. The trial registered as ClinicalTrials.gov number NCT01557101
EP
complied with the CONSORT extension for pragmatic trials.19 Inclusion criteria were: asymptomatic FDR ≥40 years of age or ten years less
AC C
than the youngest relative with CRC. Exclusion criteria were: previous CRC screening; history of colorectal neoplasia or inflammatory bowel disease; hereditary CRC; gastrointestinal surgery; pregnancy; allergy or other severe comorbidity (mean life expectancy of 2 polyps of any size, or CRC); and Group 2: screening with optical colonoscopy. Sealed envelopes were prepared for eligible FDR in the family and
AC C
given to the index-case for their delivery. The envelopes contained a leaflet with detailed information about the trial, the strategy allocated, the informed consent document to participate in the study and a contact telephone number to arrange an appointment at the High Risk CRC Clinic. The possibility of changing group was not mentioned in the information leaflet, to minimize as much as possible crossover in both arms. In that event, a structured questionnaire was used to determine the reason for crossover between the assigned strategies (Table 1).
6
ACCEPTED MANUSCRIPT
Subjects assigned to CCE received information about the procedure (bowel preparation, duration and possible adverse events) and were told that they would be scheduled to undergo colonoscopy in the event of positive result. By contrast, subjects
RI PT
assigned to colonoscopy received detailed information about the technique (bowel cleansing, sedation, advantages of polypectomy in the event of polyp detection, and possible complications). Specific informed consent for the agreed strategy (CCE or conventional
SC
colonoscopy) was sought before the procedure. Both screening procedures were free of charge. The Clinical Research Ethics Committee of our center approved the study protocol
M AN U
and all authors had access to the study data and reviewed and approved the final manuscript
Colon capsule endoscopy.
CCE was performed with second-generation PillCam™ CCE system (PillCam® colon
TE D
capsule, Given Imaging Inc, Yoqneam, Israel) previously described. 13 Colon cleanliness for CCE was scheduled in one-day, low-volume preparation, as previously described, to facilitate adherence to screening. 22, 23 Colon cleanliness was graded for the following colonic
EP
segments: right colon, transverse colon, left colon and rectum, and classified using a validated 4-point scale as previously described:13 (1) excellent (no material or liquid material covering
AC C
10% of the mucosal surface), fair (solid material impossible to suction, covering 10% of the mucosal surface). In the current study a colon preparation rated as “excellent” or “good” was considered as “adequate” whereas colon cleansing rated as “fair” and “poor” was classified as “inadequate”. CCE readings were performed by one independent blinded observer (AG or OA), as previously described.23 Polyp size was estimated by using the polyp size estimation
7
ACCEPTED MANUSCRIPT
tool included in the RAPID software (Given Imaging Ltd, Yoqneam, Israel, version 7.0). CCE findings were collected and a global lesion evaluation for each patient was recorded as: negative examination (no polyps or CCR), polyps, CRC or miscellaneous (diverticula,
RI PT
angiomas, inflammation or hemorrhoids). CCE examination was considered to be valid when the capsule was excreted (or hemorrhoidal plexus was observed), had adequate colonic cleansing or a significant lesion was detected. Subjects with an incomplete CCE
SC
exploration due to poor bowel cleansing were scheduled for optical colonoscopy. Colonoscopy
M AN U
Colonoscopies were carried out by experienced endoscopists as previously described.24 The endoscopic report contained the parameters for quality assessment (caecal intubation, quality of cleanliness for each colorectal segment, number of polyps identified, removed and recovered for histological examination and immediate complications).
TE D
Colonoscopy was considered complete if the cecum was reached and bowel cleansing was good or excellent.24 Colon cleansing was defined as inadequate when less than 90% of colorectal mucosa could be properly evaluated due to the presence of faeces. Patients with
colonoscopy.
EP
incomplete colonoscopy because of poor bowel preparation were re-scheduled to
AC C
Adenomas ≥10 mm in diameter, with tubulovillous architecture, high-grade
dysplasia or intramucosal carcinoma were classified as advanced adenomas. Invasive cancer was considered when malignant cells were observed beyond the muscularis mucosa. Tumor staging was performed according to the classification system of the American Joint Committee on Cancer 25 and patients were classified according to the most advanced lesion. Data collection Demographic characteristics of participants, kinship, degree of bowel cleanliness for
8
ACCEPTED MANUSCRIPT
each procedure, colonoscopy and CCE completion, colonoscopy or CCE findings, reason for failure and reason for crossover between the assigned groups was recorded. The level of patient satisfaction and potential procedure-related adverse effects were assessed using a
RI PT
questionnaire administered by telephone interview two weeks later (Supplementary Table 1). Sample Size
We hypothesized that uptake of CCE screening should be at least 20% higher
SC
than that of colonoscopy, on the basis that this is the minimal estimated difference that would make CCE cost-effective.18 Assuming a participation rate of 38% for colonoscopy12 and of
M AN U
58% for CCE screening, with an alpha risk of 0.05 and a beta risk of 0.20 in a two -sided test, 108 subjects in each arm of the study were needed to find a statistically significant difference.
TE D
Statistical Analysis.
The primary objective of this study was to compare screening uptake of CCE versus optical colonoscopy in FDR of patients with CRC. Screening uptake was defined as
EP
number of invited subjects screened by CCE or colonoscopy as a proportion of eligible subjects. The principal analysis consisted of an intention-to-screen comparison of the uptake
AC C
proportions between both study arms (subjects were analyzed according to their randomly assigned strategy, irrespective of whether they actually received it or not). The detection rate of neoplastic lesions (number of subjects with true positive results divided by the number of subjects who actually underwent testing) was based on as-screened analyses (subjects that underwent screening by colonoscopy or CCE, independently to their randomly assigned strategy). Statistical comparisons for secondary endpoints (efficacy of each procedure to
9
ACCEPTED MANUSCRIPT
explore the entire colon, and acceptance of each strategy) were performed according to asscreened analysis. Comparisons between groups were carried out by logistic-regression analysis and reported as odds ratios (OR) with 95% confidence intervals (CI). Continuous
RI PT
variables were expressed as means and standard deviation. Categorical variables were expressed as frequencies and percentages. Chi-square test was used to compare categorical variables. Differences with a P value 60 g/d, n (%)
36 (56.3)
0.29
68 (51.5)
8 (11.8)
8 (12.5)
0.90
16 (12.1)
0.38
27.9 ± 5.1
19 (29.7)
0.68
37 (28.0)
9 (14.1)
0.73
20 (15.2)
76 (67.3)
0.97
18 (26.5)
Antiplatelet therapy, aspirin
11 (16.2)
TE D
CRC2 Proband, n (%) 81 (67.5) 56.8 ± 8.7
27.2 ± 4.8
57.2 ± 9.19
96 (80.0)
91 (80.5)
24 (20.0)
22 (19.5)
0.57
0.87
112 (48.1)
157
67.4
57 ± 8.95
187
(80.3)
46
(19.7)
AC C
Two or more
EP
One
121 (51.9)
0.22
32 (47.1)
Smoking, n (%)
Relatives with CRC, n (%)
54.9 ± 11.6
59 (52.2)
28.6 ± 5.5
Mean age ± SD
(n=233)
53 (44.2)
BMI ± SD
Age < 60 years
0.91
SC
Male
54 (47.8)
M AN U
Female
p
RI PT
Category
Total
1
CCE = Colon capsule endoscopy
2
CRC = colorectal cancer
23
ACCEPTED MANUSCRIPT
Table 3.- Diagnostic yield of colon capsule endoscopy (CCE) and colonoscopy according to the
CCE Group
Rate
Subjects
Rate
Most advanced lesion
no.
%
no.
%
Intention-to-screen analysis
113
Non-advanced adenomaω
12
10.6
9
≥ 3 non-advanced adenomas
5
4.4
4
Advanced adenoma
8
7.1
9
Significant lesion§
13
11.5
As-screened analysis
81
Non-advanced adenomaω
16
≥ 3 non-advanced adenomas
4
Advanced adenoma
12
Significant lesion §
16
120
Odds Ratio+
M AN U
Subjects
SC
Colonoscopy Group
RI PT
intention-to-screen and as-screened analyses.*
95% CIξ
P value
0.68
0.27 to 1.68
0.41
3.3
0.74
0.19 to 2.84
0.67
7.5
1.06
0.39 to 2.86
0.92
11.7
1.02
0.45 to 2.26
0.96
5
11.1
0.50
0.17 to 1.49
0.22
4.9
5
11.1
2.41
0.61 to 9.46
0.21
14.8
5
11.1
0.72
0.23 to 2.19
0.56
19.8
11∞
24.4
1.31
0.54 to 3.14
0.54
TE D
7.5
14∞
AC C
19.8
EP
45
*In the intention-to-screen analysis the detection rate was calculated as the number of subjects with true positive results divided by the number of subjects who were eligible to undergo testing. In the as-screened analysis the
24
ACCEPTED MANUSCRIPT
RI PT
detection rate was calculated as the number of positive results divided by the number of subjects who actually underwent screening. Subjects were classified according to the most advanced lesion. +
ξ
CI denotes confidence interval.
ω
One or two adenoma 2 polyps of any size or cancer.
M AN U
§
SC
Odds ratios were adjusted according to age, gender, age (
