- Email: [email protected]
Effect of Cotinine-verified Change in Smoking Status on Risk of Metachronous Colorectal Neoplasia After Polypectomy
Accepted Manuscript Effect of Cotinine-verified Change in Smoking Status on Risk of Metachronous Colorectal Neoplasia After Polypectomy Yoon Suk Jung, Nam Hee Kim, Mi Yeon Lee, Jung Ho Park, Dong Il Park, Chong Il Sohn
PII: DOI: Reference:
S1542-3565(19)30181-8 https://doi.org/10.1016/j.cgh.2019.02.018 YJCGH 56347
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 10 February 2019 Please cite this article as: Jung YS, Kim NH, Lee MY, Park JH, Park DI, Sohn CI, Effect of Cotinineverified Change in Smoking Status on Risk of Metachronous Colorectal Neoplasia After Polypectomy, Clinical Gastroenterology and Hepatology (2019), doi: https://doi.org/10.1016/j.cgh.2019.02.018. 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.
ACCEPTED MANUSCRIPT Effect of Cotinine-verified Change in Smoking Status on Risk of Metachronous Colorectal Neoplasia After Polypectomy
1
RI PT
Yoon Suk Jung1, Nam Hee Kim2, Mi Yeon Lee3, Jung Ho Park1, Dong Il Park1, and Chong Il Sohn1
Division of Gastroenterology, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Preventive Healthcare Center, Kangbuk Samsung Hospital, Sungkyunkwan University School of
SC
2
Medicine, Seoul, Republic of Korea
Division of Biostatistics, Department of R&D Management, Kangbuk Samsung Hospital,
M AN U
3
Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Running title: Cotinine change and metachronous CRN risk
TE D
Correspondence to: Yoon Suk Jung, MD, PhD
Division of Gastroenterology, Department of Internal Medicine Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine
EP
29, Saemunan-ro, Jongno-gu, Seoul 03181, Korea Phone: +82-2-2001-8577, Fax: +82-2-2001-2049
AC C
E-mail: [email protected]
ACCEPTED MANUSCRIPT Conflicts of interest The authors disclose no conflicts
Acknowledgements
RI PT
None
Author contributions
SC
YS Jung was involved in study concept and design, acquisition of data, analysis and interpretation of data, drafting of manuscript, and overall study supervision. NH Kim was involved in acquisition,
M AN U
analysis, and interpretation of data. MY Lee was involved in analysis and interpretation of data. JH
AC C
EP
TE D
Park, DI Park, and CI Sohn were involved in acquisition of data.
ACCEPTED MANUSCRIPT Abstract: Background & Aims: Previous assessments of colorectal neoplasia (CRN) recurrence after polypectomy used self-report to determine smoking status. We evaluated the association between change in smoking status and metachronous CRN risk after polypectomy using cotinine level in urine
RI PT
to determine tobacco exposure.
Methods: We performed a retrospective study of participants in the Kangbuk Samsung Health Study, in Korea, who underwent a screening colonoscopy examination and measurement of cotinine in urine
SC
samples. Our analysis included 4762 patients who had ≥1 adenoma detected in an index colonoscopy performed between January 2010 and December 2014, and underwent a surveillance colonoscopy, 6
M AN U
or more months later, until December 2017.
Results: Patients were classified into 4 groups based on the change in cotinine-verified smoking status from index to follow-up colonoscopy (mean interval, 3.2±1.3 years), as follows: remained nonsmokers (n=2962; group 1), smokers changed to non-smokers (n=600; group 2), non-smokers changed to smokers (n=138; group 3), and remained smokers (n=1062; group 4). After adjustment for
TE D
confounding factors, group 4 had significantly higher risk of metachronous CRN than group 1 (hazard ratio [HR], 1.54; 95% CI, 1.36–1.73) and group 2 (HR, 1.63; 95% CI, 1.39–1.99). Group 4 also had a higher risk of metachronous advanced CRN than group 1 (HR, 2.84; 95% CI, 1.79–4.53) and group 2
EP
(HR, 2.10; 95% CI, 1.13–3.89). Group 3 had a higher risk of metachronous CRN than group 1 (HR, 1.50; 95% CI, 1.14–1.97) and group 2 (HR, 1.62; 95% CI, 1.20–2.20).
AC C
Conclusion: In a retrospective study of individuals with at least one adenoma, we found that cotinineverified changes in smoking status between index and follow-up colonoscopy are associated with risk of metachronous CRN. Helping patients quit smoking is important to effective prevention of colorectal cancer.
KEY WORDS: cigarette, tobacco, colon cancer, risk factors
ACCEPTED MANUSCRIPT Introduction Cigarette smoking is an important risk factor for colorectal neoplasia (CRN).1-4 Many studies have demonstrated the association between smoking and increased risk of, colorectal adenomas , as well as colorectal cancer (CRC).1-4 In addition, several studies have shown that the risk of developing
RI PT
CRN decreased after smoking cessation.5-8 However, the information on smoking status in these studies was mostly based on self-report via questionnaires, which may not accurately reflect actual smoking status. For example, current smokers may be unresponsive or may misrepresent themselves .
SC
as nonsmokers owing to social pressures and negative image associated with smoking.
Given that cigarette smoking is an important risk factor for CRN, it can be hypothesized that it
M AN U
may also be associated with increased risk of metachronous CRN after adenoma removal. In fact, some studies have reported the association between smoking and recurrence of adenomas after polypectomy.9-13 However, all previous studies on this topic have also used self-report
to determine
smoking status, and have assessed the risk of recurrence of CRN based solely on baseline smoking status at index colonoscopy, without considering changes in smoking status. Self-reported information
TE D
may be inaccurate and limited in capturing changes in smoking status between index and follow-up colonoscopy. To date, no studies have investigated the risk of developing metachronous CRN according to changes in smoking status using an objective measure of tobacco exposure.
EP
Biomarkers for tobacco exposure have been developed to more objectively assess smoking status.14,15 Among these, cotinine, the main metabolite of nicotine, has been extensively used to
AC C
accurately assess smoking status.14,15 This study aimed to evaluate the association of urinary cotinineverified change in smoking status with the risk of metachronous CRN after polypectomy in asymptomatic examinees.
Methods Study population The Kangbuk Samsung Health Study is a cohort study comprising South Korean men and women aged ≥18 years who underwent a comprehensive annual or biennial health examination at
ACCEPTED MANUSCRIPT clinics associated with the Kangbuk Samsung Hospital Total Healthcare Center in Seoul and Suwon, South Korea. The study population comprised a subset of the Kangbuk Samsung Health Study subjects who underwent a screening colonoscopic examination and urinary cotinine measurement as a
analyzed data obtained from this prospectively established cohort.
RI PT
component of a comprehensive health examination between 2010 and 2017. We retrospectively
Patients who had ≥1 adenoma detected on an index colonoscopy performed between January 2010 and December 2014, and those who underwent a follow-up surveillance colonoscopy at an
SC
interval of >6 months until December 2017 were considered for inclusion in the study. All polyps were endoscopically removed at the index colonoscopy. Patients with a history of CRC or colorectal
M AN U
surgery, those diagnosed with CRC during index colonoscopy, those with a history of inflammatory bowel disease (IBD), and those with poor bowel preparation at index or follow-up colonoscopy were excluded. Poor bowel preparation was defined as “large amounts of solid fecal matter observed in the bowel precluding a satisfactory study, unacceptable preparation, or a condition in which <90% of the mucosa could be adequately visualized”.16
TE D
In South Korea, the Industrial Safety and Health Law mandates free annual or biennial health screening examinations of all employees. Most individuals in our study were employees of various companies and local governmental organizations, or their spouses. As part of their welfare policy,
EP
companies often subsidize comprehensive health examinations, including colonoscopy. Before colonoscopy, interviews by general practitioners were conducted to ensure that all participants were
AC C
asymptomatic (i.e., no abdominal pain, diarrhea, or hematochezia). Symptomatic participants were urged to seek medical care.
This study was approved by the Institutional Review Board of Kangbuk Samsung Hospital. The requirement for informed consent was waived because only de-identified data were retrospectively accessed.
Measurements and definitions Data on smoking status and medical history were obtained through a self-administered
ACCEPTED MANUSCRIPT questionnaire. Self-reported smoking status was categorized as never, former, or current smokers. Participants who had smoked 100 or more cigarettes in their life and who smoked cigarettes at the time of the study were defined as “current smokers.” Participants who had smoked 100 or more cigarettes in their life but said that they did not currently smoke were defined as “former smokers.”
RI PT
Participants who had never smoked a cigarette or who had smoked fewer than 100 cigarettes in their life were defined as “never smokers”.17
Urinary cotinine level was measured before colonoscopy on the day of colonoscopy. It was
SC
determined using DRI Cotinine Assay (Microgenics Corp., Fremont, CA, USA) with a modular P800 analyzer (Roche Diagnostics, Tokyo, Japan). Cotinine-verified smokers were defined as participants
M AN U
having a urinary cotinine level ≥50 ng/mL, while cotinine-verified non-smokers were defined as participants having a urinary cotinine level <50 ng/mL.18
Family history of CRC was defined as the presence of CRC in at least one first-degree relative(s) of any age. Self-reported regular use of nonsteroidal anti-inflammatory drugs (NSAIDs) during the previous month was also assessed. Obesity was defined as body mass index ≥25 kg/m2, which is the
TE D
proposed cutoff for the diagnosis of obesity in Asians.19
Colonoscopic and histological examinations
EP
All colonoscopic examinations were performed using the EVIS LUCERA CV-260 colonoscope (Olympus Medical Systems, Tokyo, Japan) by board-certified endoscopists. Bowel cleansing was
AC C
performed using 4 L of polyethylene glycol solution and a split-dose preparation was employed. Suspicious neoplastic lesions were removed via biopsy, snare polypectomy, or endoscopic mucosal resection and histologically assessed by experienced pathologists. Overall CRN was defined as cancer or adenoma, and advanced CRN (ACRN) was defined as cancer or advanced adenoma. Advanced adenoma was defined as the presence of one of the following features: a lesion measuring ≥10 mm in diameter, showing a tubulovillous or villous structure, and high-grade dysplasia. Low-risk adenoma (LRA) was defined as 1–2 adenomas measuring <10 mm in size, and high-risk adenoma (HRA) was defined as advanced adenomas or ≥3 adenomas.20,21 Sessile
ACCEPTED MANUSCRIPT serrated polyps and traditional serrated adenomas were included in the definition of CRN. However, hyperplastic polyps were not considered neoplastic lesions.
Statistical analysis
RI PT
Data were stratified into four groups based on the change in cotinine-verified smoking status between index colonoscopy and follow-up colonoscopy. Groups 1, 2, 3, and 4 were classified as “from cotinine-verified non-smokers at the time of index colonoscopy to cotinine-verified non-
SC
smokers at the time of follow-up colonoscopy,” “from cotinine-verified smokers to cotinine-verified non-smokers,” “from cotinine-verified non-smokers to cotinine-verified smokers,” and “from
M AN U
cotinine-verified smokers to cotinine-verified smokers,” respectively.
Baseline characteristics between the four groups were compared using chi-square test and 1-way analysis of variance for categorical variables and continuous variables, respectively. The cumulative incidence of metachronous CRN was calculated using the Kaplan–Meier method, and the differences between groups were determined using the log-rank test. To compare the risk for development of
TE D
metachronous CRN between groups, we also performed multivariable analysis using the Cox proportional hazards regression model after adjusting for potential confounders, including age, sex, family history of CRC, NSAID use, obesity, hypertension, diabetes mellitus, and baseline adenoma
EP
characteristics (LRA and HRA). All reported P values were two tailed, and P < 0.05 was considered statistically significant. SPSS software version 21 (IBM Corp., Armonk, NY, USA) was used to
AC C
perform all statistical analyses.
Results
Baseline patient characteristics A total of 12,457 patients with ≥1 adenoma at the time of index colonoscopy who underwent follow-up surveillance colonoscopy were eligible for inclusion in this study. Among these, 2313 patients were excluded owing to the following reasons: history of CRC or colorectal surgery (n = 97), diagnosis of CRC at the time of index colonoscopy (n = 9), history of IBD (n = 35), and poor bowel
ACCEPTED MANUSCRIPT preparation at index or follow-up colonoscopy (n = 2172). Of the remaining 10,144 patients, 5382 patients were excluded because they did not undergo urinary cotinine measurements at the index or follow-up colonoscopy. A total of 4762 patients who underwent urinary cotinine measurements at both the index and follow-up colonoscopy were included. The mean age was 40.1 ± 6.3 years, and
RI PT
83.8% were male. The mean interval between index and follow-up colonoscopy was 3.2 ± 1.3 (range, 0.5–6.4) years.
At index colonoscopy, among cotinine-verified non-smokers, the proportion of self-reported
SC
never, former, and current smokers was 48.9%, 35.7%, and 6.6%, respectively, while among cotinineverified smokers, the proportion of self-reported never, former, and current smokers was 3.4%, 5.9%,
M AN U
and 84.5%, respectively (Table 1).
Patients were divided into four groups based on the change in cotinine-verified smoking status: cotinine-verified non-smokers at the time of index colonoscopy → cotinine-verified non-smokers at the time of follow-up colonoscopy (group 1), cotinine-verified smokers → non-smokers (group 2), cotinine-verified non-smokers → smokers (group 3), and cotinine-verified smokers → smokers
TE D
(group 4). Groups 1, 2, 3, and 4 consisted of 2962 (62.2%), 600 (12.6%), 138 (2.9%), and 1062 (22.3%) patients, respectively. Table 2 shows the baseline characteristics of patients in the four groups. Mean patient age was highest in group 4 and lowest in group 2. The proportion of men,
EP
obesity, and diabetes was the lowest in group 1. The proportion of patients with HRA at index
AC C
colonoscopy was high in groups 2 and 4.
Risk of metachronous neoplasia based on change in cotinine-verified smoking status The comparison of the risk of metachronous CRN on follow-up colonoscopy between the four groups is shown in Figure 1. Among the LRA group identified at index colonoscopy, the cumulative risk of metachronous CRN was higher in group 4 than in groups 1 and 2 (P < 0.001 and P < 0.001, respectively), and higher in group 3 than in groups 1 and 2 (P = 0.007 and P = 0.010, respectively), whereas the risk did not differ between groups 1 and 2 (P = 0.686) and between groups 3 and 4 (P = 0.613) (Fig. 1A). Additionally, among the HRA group identified at index colonoscopy, the cumulative
ACCEPTED MANUSCRIPT risk of metachronous CRN was higher in group 4 than in groups 1 and 2 (P < 0.001 and P < 0.001, respectively), and it was higher in group 3 than in groups 1 and 2 (P = 0.008 and P = 0.004, respectively); the risk did not differ between groups 1 and 2 (P = 0.590) and between groups 3 and 4 (P = 0.432) (Fig. 1B).
RI PT
For metachronous ACRN, the cumulative risk in the LRA group was higher in group 4 than in groups 1 and 2 (P < 0.001 and P = 0.033, respectively), whereas the risk was not different between the other groups (Fig. 2A). Among the HRA group, the cumulative risk for metachronous ACRN did not
SC
differ between the groups (Fig. 2B).
Similar results for the risk of metachronous CRN were obtained after adjusting for confounding
M AN U
factors (Table 3). Compared to group 1, the risk of metachronous CRN in groups 3 (adjusted hazard ratio [HR], 1.50; 95% confidence interval [CI], 1.14–1.97) and 4 (adjusted HR, 1.54; 95% CI, 1.36– 1.73) was higher, but the risk in group 2 was not significantly different. Using group 2 as a reference group, the risk of metachronous CRN in groups 3 (adjusted HR, 1.62; 95% CI, 1.20–2.20) and 4 (adjusted HR, 1.63; 95% CI, 1.39–1.99) was higher. Additionally, the risk of metachronous ACRN in
Discussion
EP
2.10; 95% CI, 1.13–3.89).
TE D
group 4 was higher than that in groups 1 (adjusted HR, 2.84; 95% CI, 1.79–4.53) and 2 (adjusted HR,
In this longitudinal study, we found that cotinine-verified smokers had a higher risk of
AC C
metachronous CRN than cotinine-verified non-smokers did, at both index and follow-up colonoscopies. They also had a higher risk of metachronous CRN than patients who switched from “cotinine-verified smokers at index colonoscopy” to “cotinine-verified non-smokers at follow-up colonoscopy.” More importantly, patients who switched from “cotinine-verified non-smokers at index colonoscopy” to “cotinine-verified smokers at follow-up colonoscopy” had a higher risk of metachronous CRN than the risks of patients who switched from “cotinine-verified smokers at index colonoscopy” to “cotinine-verified non-smokers at follow-up colonoscopy” as well as those of cotinine-verified non-smokers at both index and follow-up colonoscopies. In addition, the risk of
ACCEPTED MANUSCRIPT metachronous ACRN in cotinine-verified smokers at both index and follow-up colonoscopies was higher than the risk in patients who switched from “cotinine-verified smokers” to “cotinine-verified non-smokers” as well as cotinine-verified non-smokers at both index and follow-up colonoscopies. These results indicate that the risk of metachronous CRN after polypectomy may be affected by
RI PT
changes in smoking status between index colonoscopy and follow-up colonoscopy.
Several studies have shown that smoking is associated with an increased risk of recurrence of adenomas after polypectomy.9-13 However, the previous studies used self-reported questionnaires to
SC
assess smoking status and evaluated the association between smoking and recurrence of adenoma based on smoking status at baseline, without dynamic assessment of changes in status over time.
M AN U
Some smokers at the index colonoscopy may have stopped smoking during the follow-up period and may have become non-smokers at the follow-up colonoscopy; similarly, some non-smokers at the index colonoscopy may start smoking during the follow-up period and may become smokers at the follow-up colonoscopy. In fact, our study showed that some patients switched from “cotinine-verified smokers at index colonoscopy” to “cotinine-verified non-smokers at follow-up colonoscopy” (12.6%),
TE D
or from “cotinine-verified non-smokers” to “cotinine-verified smokers” (2.9%). In this regard, to better understand the association between smoking and the risk of metachronous CRN after polypectomy, their associations should be assessed based on the change in smoking status between
EP
index colonoscopy and follow-up colonoscopy, and not just the smoking status at index colonoscopy. Current postpolypectomy surveillance guidelines recommend that patients undergo follow-up
AC C
surveillance colonoscopy 3 and 5 years after HRA and LRA removal, respectively.20,21 Therefore, in our study, it can be assumed that the interval between index colonoscopy and follow-up colonoscopy was 3–5 years. It may be difficult to capture changes in smoking status through self-reported questionnaires during this short period. Moreover, assessment of smoking status based on selfreported questionnaires may be inaccurate, and it may induce recall biases and reporting errors. In fact, our previous study showed that some self-reported never smokers were cotinine-verified smokers and had an increased risk of CRN.22 These results suggest that cotinine-captured smoking status exhibits associations with CRN more accurately and objectively than self-reporting does.22 Considering these
ACCEPTED MANUSCRIPT points, in this study, we attempted to identify the changes in smoking status using cotinine, an objective biomarker for tobacco exposure. An interesting result of our study is that compared to “cotinine-verified non-smokers → smokers,” “cotinine-verified smokers → non-smokers” group had a lower risk of metachronous CRN. The same
RI PT
results were obtained when subgroup analysis was conducted for each of the high-risk groups and the low-risk group at the index colonoscopy (Fig. 1). As all adenomas detected at the time of the index colonoscopy were removed, it may be assumed that the two groups are similar at baseline. The risk of
SC
metachronous CRN appeared to increase in “cotinine-verified non-smokers → smokers” group, whereas it appears to decrease in “cotinine-verified smokers → non-smokers” group. Our results
M AN U
suggest that the impact of smoking on CRN may be reversible.
More importantly, our study showed that changes in smoking status affect the risk of developing metachronous ACRN as well as metachronous CRN. Compared to “cotinine-verified non-smokers → non-smokers” group, “cotinine-verified smokers → smokers” group had a higher risk of metachronous ACRN (adjusted HR, 2.84; 95% CI, 1.79–4.53). These results suggest that the
TE D
postpolypectomy surveillance interval should consider the smoking status of the patients as well as the characteristics of the adenoma identified at the time of index colonoscopy. The interval may have to be reduced for patients who continue to smoke. In addition, we found that compared to “cotinine-
EP
verified smokers → smokers” group, “cotinine-verified smokers → non-smokers” group had a lower risk of metachronous ACRN. These results strongly suggest that smoking cessation may help prevent
AC C
CRC. The “cotinine-verified smokers → non-smokers” group might also have healthier lifestyles after smoking cessation (for example, increased physical activity, improved eating habits, and weight loss). Although we did not consider these lifestyle changes during the follow-up period, lifestyle improvements as well as smoking cessation might have resulted in low risk for metachronous CRN in this group. However, among the HRA group, the risk for metachronous ACRN did not differ between the 4 groups based on the change in cotinine-verified smoking status. These results might be because of the small number of patients who developed metachronous ACRN in this cohort. The number of patients
ACCEPTED MANUSCRIPT with HRA was relatively small (n=674), and only 30 (4.5%) of them developed metachronous ACRN. To clarify the relationship between the change in smoking status and the risk of metachronous ACRN in HRA group, studies which include large numbers of patients with HRA are necessary. Meanwhile, we found that among cotinine-verified smokers, the percentage of self-reported
RI PT
current smokers was 84.5%, and among cotinine-verified non-smokers, the percentage of selfreported current smokers was 6.6%. There appears to be good correlation between self-reported smoking information and status based on cotinine verification.
SC
To the best of our knowledge, this is the first study to identify the relationship between cotinineverified change in smoking status and the risk of metachronous CRN. Nevertheless, the current study
M AN U
has several limitations. First, most subjects in our study were employees of various companies. As a result, there was likely some degree of selection bias. Second, the interval between the index and follow-up colonoscopy was relatively short (mean, 3.2 years). A proportion of adenomas found at the time of surveillance colonoscopy might have been missed during the index colonoscopy. However, the follow-up interval was less than one year in only 97 patients (2.0%). Additionally, considering that in
TE D
our center, the overall adenoma detection rate in subjects aged ≥50 years at index colonoscopy (between 2010 and 2014) was 32% (39% for men and 22% for women), exceeding the recommended thresholds for colonoscopy quality,23 the possibility of missed lesions at the index colonoscopy is
EP
attenuated . Further, the follow-up interval was sufficient in the HRA group, even though it might be short in the LRA group. Lastly, urinary cotinine levels were measured only once at the index
AC C
colonoscopy and at the follow-up colonoscopy. This one-time cotinine level may be unable to closely reflect the increase or decrease in the quantity of cigarettes smoked during the interval between the index and follow-up colonoscopy, owing to its short half-life (about 20 hours).24,25 Despite these limitations, our data provides a better understanding of the risk of metachronous CRN according to change in smoking status. The “cotinine-verified smokers at index colonoscopy → cotinine-verified smokers at follow-up colonoscopy” had a higher risk of metachronous ACRN as well as metachronous CRN than the risks of “cotinine-verified non-smokers → non-smokers” and “cotinine-verified smokers → non-smokers.” Further, “cotinine-verified non-smokers at index
ACCEPTED MANUSCRIPT colonoscopy → smokers at follow-up colonoscopy” had a higher risk of metachronous CRN than the risks of “cotinine-verified non-smokers → non-smokers” and “cotinine-verified smokers → nonsmokers.” Our results indicate that changes in smoking status between index and follow-up
AC C
EP
TE D
M AN U
SC
and that smoking cessation may help prevent CRC.
RI PT
colonoscopy have an impact on the risk of developing metachronous CRN following polypectomy
ACCEPTED MANUSCRIPT Figure legends Fig. 1. Cumulative incidence of metachronous colorectal neoplasia based on changes in cotinineverified smoking status in the low-risk (A) and high-risk adenoma group (B) Fig. 2. Cumulative incidence of metachronous advanced colorectal neoplasia based on changes in
AC C
EP
TE D
M AN U
SC
RI PT
cotinine-verified smoking status in the low-risk (A) and high-risk adenoma group (B)
ACCEPTED MANUSCRIPT References 1.
Botteri E, Iodice S, Raimondi S, et al. Cigarette smoking and adenomatous polyps: a metaanalysis. Gastroenterology 2008;134:388-395.
2.
Botteri E, Iodice S, Bagnardi V, et al. Smoking and colorectal cancer: a meta-analysis. Jama
3.
RI PT
2008;300:2765-2778.
Peng L, Weigl K, Boakye D, et al. Risk Scores for Predicting Advanced Colorectal Neoplasia in the Average-risk Population: A Systematic Review and Meta-analysis. Am J Gastroenterol 2018. Tsoi KK, Pau CY, Wu WK, et al. Cigarette smoking and the risk of colorectal cancer: a meta-
SC
4.
analysis of prospective cohort studies. Clin Gastroenterol Hepatol 2009;7:682-688.e681-685. Hannan LM, Jacobs EJ, Thun MJ. The association between cigarette smoking and risk of
M AN U
5.
colorectal cancer in a large prospective cohort from the United States. Cancer Epidemiol Biomarkers Prev 2009;18:3362-3367. 6.
Shrubsole MJ, Wu H, Ness RM, et al. Alcohol drinking, cigarette smoking, and risk of colorectal adenomatous and hyperplastic polyps. Am J Epidemiol 2008;167:1050-1058. Martinez F, Fernandez-Martos C, Quintana MJ, et al. APC and KRAS mutations in distal
TE D
7.
colorectal polyps are related to smoking habits in men: results of a cross-sectional study. Clin Transl Oncol 2011;13:664-671.
Nishihara R, Morikawa T, Kuchiba A, et al. A prospective study of duration of smoking cessation
EP
8.
and colorectal cancer risk by epigenetics-related tumor classification. Am J Epidemiol
9.
AC C
2013;178:84-100.
Jacobson JS, Neugut AI, Murray T, et al. Cigarette smoking and other behavioral risk factors for recurrence of colorectal adenomatous polyps (New York City, NY, USA). Cancer Causes Control 1994;5:215-220.
10. Reid ME, Marshall JR, Roe D, et al. Smoking exposure as a risk factor for prevalent and recurrent colorectal adenomas. Cancer Epidemiol Biomarkers Prev 2003;12:1006-1011. 11. Kim MC, Jung SW, Kim CS, et al. Metabolic syndrome is associated with increased risk of recurrent colorectal adenomas in Korean men. Int J Obes (Lond) 2012;36:1007-1011.
ACCEPTED MANUSCRIPT 12. Lin CC, Huang KW, Luo JC, et al. Hypertension is an important predictor of recurrent colorectal adenoma after screening colonoscopy with adenoma polypectomy. J Chin Med Assoc 2014;77:508-512. 13. Saiken A, Gu F. Lifestyle and lifestyle-related comorbidities independently associated with
RI PT
colorectal adenoma recurrence in elderly Chinese people. Clin Interv Aging 2016;11:801-805. 14. Connor Gorber S, Schofield-Hurwitz S, Hardt J, et al. The accuracy of self-reported smoking: a
Nicotine Tob Res 2009;11:12-24.
SC
systematic review of the relationship between self-reported and cotinine-assessed smoking status.
15. Cross AJ, Boca S, Freedman ND, et al. Metabolites of tobacco smoking and colorectal cancer
M AN U
risk. Carcinogenesis 2014;35:1516-1522.
16. Soweid AM, Kobeissy AA, Jamali FR, et al. A randomized single-blind trial of standard diet versus fiber-free diet with polyethylene glycol electrolyte solution for colonoscopy preparation. Endoscopy 2010;42:633-638.
2009;123:598-601.
TE D
17. Rodu B, Cole P. Smoking prevalence: a comparison of two American surveys. Public Health
18. Biochemical verification of tobacco use and cessation. Nicotine Tob Res 2002;4:149-159. 19. Wen CP, David Cheng TY, Tsai SP, et al. Are Asians at greater mortality risks for being
EP
overweight than Caucasians? Redefining obesity for Asians. Public Health Nutr 2009;12:497-506. 20. Lieberman DA, Rex DK, Winawer SJ, et al. Guidelines for colonoscopy surveillance after
AC C
screening and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2012;143:844-857. 21. Yang DH, Hong SN, Kim YH, et al. Korean guidelines for postpolypectomy colonoscopy surveillance. Clin Endosc 2012;45:44-61. 22. Kim NH, Jung YS, Yang HJ, et al. Association Between Cotinine-verified Smoking Status and Risk of Colorectal Neoplasia. J Clin Gastroenterol 2018. 23. Rex DK, Boland CR, Dominitz JA, et al. Colorectal Cancer Screening: Recommendations for Physicians and Patients from the U.S. Multi-Society Task Force on Colorectal Cancer. Am J
ACCEPTED MANUSCRIPT Gastroenterol 2017;112:1016-1030. 24. Benowitz NL. Cotinine as a biomarker of environmental tobacco smoke exposure. Epidemiol Rev 1996;18:188-204. 25. Jaakkola MS, Jaakkola JJ. Assessment of exposure to environmental tobacco smoke. Eur Respir
AC C
EP
TE D
M AN U
SC
RI PT
J 1997;10:2384-2397.
ACCEPTED MANUSCRIPT
Cotinine-verified non-smokers
Cotinine-verified smokers†
Self-reported never smokers
1515 (48.9)
Self-reported former smokers
1106 (35.7)
Self-reported current smokers
204 (6.6)
Non-respondents
275 (8.9)
RI PT
Table 1. Frequency of cotinine-verified and self-reported smoking status at the time of index colonoscopy
Total
56 (3.4)
98 (5.9)
1405 (84.5) 103 (6.2)
SC
Smoking status
3100
1662
AC C
EP
TE D
M AN U
Data are presented as number (%). † Cotinine-verified current smoker was defined as an individual having a urinary cotinine level ≥50 ng/mL.
ACCEPTED MANUSCRIPT
Group 4 Cotinine-verified smokers → smokers
138
1062
39.8 ± 5.3
40.7 ±5.9
<0.001
134 (97.1)
1039 (97.8)
<0.001
7 (5.1)
56 (5.3)
0.982
6 (4.3)
19 (1.8)
0.271
P value
600
40.2 ± 6.6
39.0 ± 5.9
2234 (75.4)
582 (97.0)
149 (5.0)
29 (4.8)
71 (2.4)
14 (2.3)
Obesity (BMI ≥25 kg/m )
1075 (36.3)
281 (46.8)
69 (50.0)
543 (51.1)
<0.001
Hypertension
432 (14.3)
92 (15.3)
26 (18.8)
181 (17.0)
0.103
Diabetes mellitus
131 (4.4)
40 (6.7)
9 (6.5)
77 (7.3)
0.002
2574 (86.9)
507 (84.5)
119 (86.2)
888 (83.6)
0.047
93 (15.5)
19 (13.8)
174 (16.4)
3.5 ± 1.3
3.1 ± 1.3
3.0 ± 1.3
Age (years) Men Family history of CRC Use of NSAIDs 2
TE D
Baseline adenoma characteristics Low-risk adenoma
SC
2962
M AN U
No. of patients
Group 3 Cotinine-verified non-smokers → smokers
RI PT
Table 2. Baseline characteristics according to change of cotinine-verified smoking status Group 1 Group 2 Cotinine-verified non-smokers Cotinine-verified smokers Variable → non-smokers → non-smokers
EP
High-risk adenoma 388 (13.1) Interval between index and 3.2 ± 1.3 follow-up colonoscopy (years) Data are presented as number (%) or mean ± standard deviation
AC C
CRC, colorectal cancer; NSAIDs, nonsteroidal anti-inflammatory drugs; BMI, body mass index
<0.001
ACCEPTED MANUSCRIPT
Table 3. Multivariable analysis for the relationship of change in cotinine-verified smoking status with the risk of metachronous colorectal neoplasia Overall CRN
1 (Reference) 0.92 (0.78–1.09)
Group 3 (Cotinine-verified non-smokers → smokers)
1.50 (1.14–1.97)
Group 4 (Cotinine-verified smokers → smokers)
1.54 (1.36–1.73)
Group 1 (Cotinine-verified non-smokers → non-smokers)
1.09 (0.92–1.28)
M AN U
Group 2 (Cotinine-verified smokers → non-smokers)
Adjusted HR (95% CI)
RI PT
Group 1 (Cotinine-verified non-smokers → non-smokers)
P value
P value
1 (Reference)
0.338
1.36 (0.72–2.55)
0.344
0.004
1.17 (0.28–4.87)
0.834
<0.001
2.84 (1.79–4.53)
<0.001
0.338
0.74 (0.39–1.39)
0.344
SC
Adjusted HR (95% CI)
ACRN
Group 2 (Cotinine-verified smokers → non-smokers)
1 (Reference)
1 (Reference)
Group 3 (Cotinine-verified non-smokers → smokers)
1.62 (1.20–2.20)
0.002
0.86 (0.19–3.80)
0.841
Group 4 (Cotinine-verified smokers → smokers)
1.63 (1.39–1.99)
<0.001
2.10 (1.13–3.89)
0.019
TE D
Values were adjusted for age, sex, family history of CRC, NSAID use, obesity, hypertension, diabetes mellitus, and baseline adenoma characteristics (lowand high-risk adenoma)
AC C
EP
CRN, colorectal neoplasia; ACRN, advanced colorectal neoplasia; HR, hazard ratio; CI, confidence interval
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
What You Need to Know
RI PT
Background: Previous assessments of colorectal neoplasia (CRN) recurrence after polypectomy used self-reported, initial smoking status to measure of determine smoking status. We performed a retrospective study of patients undergoing screening colonoscopies and surveillance colonoscopies after polypectomy to evaluate the association between change in smoking status and metachronous CRN. We used cotinine levels in urine as a marker of tobacco exposure.
M AN U
SC
Findings: Cotinine-verified smokers at index colonoscopy who were cotinine-verified smokers at the follow-up colonoscopy had a higher risk of metachronous CRN than cotinine-verified nonsmokers who became non-smokers or cotinine-verified smokers who became non-smokers. Cotinine-verified smokers also had a higher risk of metachronous advanced CRN than cotinineverified non-smokers who became non-smokers or cotinine-verified smokers who became nonsmokers.
AC C
EP
TE D
Implications for patient care: Changes in smoking status between index and follow-up colonoscopy after polypectomy are associated with risk of metachronous CRN. Helping patients quit smoking or following-more frequent screening/surveillance of smokers might prevent colorectal cancer.
PII: DOI: Reference:
S1542-3565(19)30181-8 https://doi.org/10.1016/j.cgh.2019.02.018 YJCGH 56347
To appear in: Clinical Gastroenterology and Hepatology Accepted Date: 10 February 2019 Please cite this article as: Jung YS, Kim NH, Lee MY, Park JH, Park DI, Sohn CI, Effect of Cotinineverified Change in Smoking Status on Risk of Metachronous Colorectal Neoplasia After Polypectomy, Clinical Gastroenterology and Hepatology (2019), doi: https://doi.org/10.1016/j.cgh.2019.02.018. 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.
ACCEPTED MANUSCRIPT Effect of Cotinine-verified Change in Smoking Status on Risk of Metachronous Colorectal Neoplasia After Polypectomy
1
RI PT
Yoon Suk Jung1, Nam Hee Kim2, Mi Yeon Lee3, Jung Ho Park1, Dong Il Park1, and Chong Il Sohn1
Division of Gastroenterology, Department of Internal Medicine, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Preventive Healthcare Center, Kangbuk Samsung Hospital, Sungkyunkwan University School of
SC
2
Medicine, Seoul, Republic of Korea
Division of Biostatistics, Department of R&D Management, Kangbuk Samsung Hospital,
M AN U
3
Sungkyunkwan University School of Medicine, Seoul, Republic of Korea
Running title: Cotinine change and metachronous CRN risk
TE D
Correspondence to: Yoon Suk Jung, MD, PhD
Division of Gastroenterology, Department of Internal Medicine Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine
EP
29, Saemunan-ro, Jongno-gu, Seoul 03181, Korea Phone: +82-2-2001-8577, Fax: +82-2-2001-2049
AC C
E-mail: [email protected]
ACCEPTED MANUSCRIPT Conflicts of interest The authors disclose no conflicts
Acknowledgements
RI PT
None
Author contributions
SC
YS Jung was involved in study concept and design, acquisition of data, analysis and interpretation of data, drafting of manuscript, and overall study supervision. NH Kim was involved in acquisition,
M AN U
analysis, and interpretation of data. MY Lee was involved in analysis and interpretation of data. JH
AC C
EP
TE D
Park, DI Park, and CI Sohn were involved in acquisition of data.
ACCEPTED MANUSCRIPT Abstract: Background & Aims: Previous assessments of colorectal neoplasia (CRN) recurrence after polypectomy used self-report to determine smoking status. We evaluated the association between change in smoking status and metachronous CRN risk after polypectomy using cotinine level in urine
RI PT
to determine tobacco exposure.
Methods: We performed a retrospective study of participants in the Kangbuk Samsung Health Study, in Korea, who underwent a screening colonoscopy examination and measurement of cotinine in urine
SC
samples. Our analysis included 4762 patients who had ≥1 adenoma detected in an index colonoscopy performed between January 2010 and December 2014, and underwent a surveillance colonoscopy, 6
M AN U
or more months later, until December 2017.
Results: Patients were classified into 4 groups based on the change in cotinine-verified smoking status from index to follow-up colonoscopy (mean interval, 3.2±1.3 years), as follows: remained nonsmokers (n=2962; group 1), smokers changed to non-smokers (n=600; group 2), non-smokers changed to smokers (n=138; group 3), and remained smokers (n=1062; group 4). After adjustment for
TE D
confounding factors, group 4 had significantly higher risk of metachronous CRN than group 1 (hazard ratio [HR], 1.54; 95% CI, 1.36–1.73) and group 2 (HR, 1.63; 95% CI, 1.39–1.99). Group 4 also had a higher risk of metachronous advanced CRN than group 1 (HR, 2.84; 95% CI, 1.79–4.53) and group 2
EP
(HR, 2.10; 95% CI, 1.13–3.89). Group 3 had a higher risk of metachronous CRN than group 1 (HR, 1.50; 95% CI, 1.14–1.97) and group 2 (HR, 1.62; 95% CI, 1.20–2.20).
AC C
Conclusion: In a retrospective study of individuals with at least one adenoma, we found that cotinineverified changes in smoking status between index and follow-up colonoscopy are associated with risk of metachronous CRN. Helping patients quit smoking is important to effective prevention of colorectal cancer.
KEY WORDS: cigarette, tobacco, colon cancer, risk factors
ACCEPTED MANUSCRIPT Introduction Cigarette smoking is an important risk factor for colorectal neoplasia (CRN).1-4 Many studies have demonstrated the association between smoking and increased risk of, colorectal adenomas , as well as colorectal cancer (CRC).1-4 In addition, several studies have shown that the risk of developing
RI PT
CRN decreased after smoking cessation.5-8 However, the information on smoking status in these studies was mostly based on self-report via questionnaires, which may not accurately reflect actual smoking status. For example, current smokers may be unresponsive or may misrepresent themselves .
SC
as nonsmokers owing to social pressures and negative image associated with smoking.
Given that cigarette smoking is an important risk factor for CRN, it can be hypothesized that it
M AN U
may also be associated with increased risk of metachronous CRN after adenoma removal. In fact, some studies have reported the association between smoking and recurrence of adenomas after polypectomy.9-13 However, all previous studies on this topic have also used self-report
to determine
smoking status, and have assessed the risk of recurrence of CRN based solely on baseline smoking status at index colonoscopy, without considering changes in smoking status. Self-reported information
TE D
may be inaccurate and limited in capturing changes in smoking status between index and follow-up colonoscopy. To date, no studies have investigated the risk of developing metachronous CRN according to changes in smoking status using an objective measure of tobacco exposure.
EP
Biomarkers for tobacco exposure have been developed to more objectively assess smoking status.14,15 Among these, cotinine, the main metabolite of nicotine, has been extensively used to
AC C
accurately assess smoking status.14,15 This study aimed to evaluate the association of urinary cotinineverified change in smoking status with the risk of metachronous CRN after polypectomy in asymptomatic examinees.
Methods Study population The Kangbuk Samsung Health Study is a cohort study comprising South Korean men and women aged ≥18 years who underwent a comprehensive annual or biennial health examination at
ACCEPTED MANUSCRIPT clinics associated with the Kangbuk Samsung Hospital Total Healthcare Center in Seoul and Suwon, South Korea. The study population comprised a subset of the Kangbuk Samsung Health Study subjects who underwent a screening colonoscopic examination and urinary cotinine measurement as a
analyzed data obtained from this prospectively established cohort.
RI PT
component of a comprehensive health examination between 2010 and 2017. We retrospectively
Patients who had ≥1 adenoma detected on an index colonoscopy performed between January 2010 and December 2014, and those who underwent a follow-up surveillance colonoscopy at an
SC
interval of >6 months until December 2017 were considered for inclusion in the study. All polyps were endoscopically removed at the index colonoscopy. Patients with a history of CRC or colorectal
M AN U
surgery, those diagnosed with CRC during index colonoscopy, those with a history of inflammatory bowel disease (IBD), and those with poor bowel preparation at index or follow-up colonoscopy were excluded. Poor bowel preparation was defined as “large amounts of solid fecal matter observed in the bowel precluding a satisfactory study, unacceptable preparation, or a condition in which <90% of the mucosa could be adequately visualized”.16
TE D
In South Korea, the Industrial Safety and Health Law mandates free annual or biennial health screening examinations of all employees. Most individuals in our study were employees of various companies and local governmental organizations, or their spouses. As part of their welfare policy,
EP
companies often subsidize comprehensive health examinations, including colonoscopy. Before colonoscopy, interviews by general practitioners were conducted to ensure that all participants were
AC C
asymptomatic (i.e., no abdominal pain, diarrhea, or hematochezia). Symptomatic participants were urged to seek medical care.
This study was approved by the Institutional Review Board of Kangbuk Samsung Hospital. The requirement for informed consent was waived because only de-identified data were retrospectively accessed.
Measurements and definitions Data on smoking status and medical history were obtained through a self-administered
ACCEPTED MANUSCRIPT questionnaire. Self-reported smoking status was categorized as never, former, or current smokers. Participants who had smoked 100 or more cigarettes in their life and who smoked cigarettes at the time of the study were defined as “current smokers.” Participants who had smoked 100 or more cigarettes in their life but said that they did not currently smoke were defined as “former smokers.”
RI PT
Participants who had never smoked a cigarette or who had smoked fewer than 100 cigarettes in their life were defined as “never smokers”.17
Urinary cotinine level was measured before colonoscopy on the day of colonoscopy. It was
SC
determined using DRI Cotinine Assay (Microgenics Corp., Fremont, CA, USA) with a modular P800 analyzer (Roche Diagnostics, Tokyo, Japan). Cotinine-verified smokers were defined as participants
M AN U
having a urinary cotinine level ≥50 ng/mL, while cotinine-verified non-smokers were defined as participants having a urinary cotinine level <50 ng/mL.18
Family history of CRC was defined as the presence of CRC in at least one first-degree relative(s) of any age. Self-reported regular use of nonsteroidal anti-inflammatory drugs (NSAIDs) during the previous month was also assessed. Obesity was defined as body mass index ≥25 kg/m2, which is the
TE D
proposed cutoff for the diagnosis of obesity in Asians.19
Colonoscopic and histological examinations
EP
All colonoscopic examinations were performed using the EVIS LUCERA CV-260 colonoscope (Olympus Medical Systems, Tokyo, Japan) by board-certified endoscopists. Bowel cleansing was
AC C
performed using 4 L of polyethylene glycol solution and a split-dose preparation was employed. Suspicious neoplastic lesions were removed via biopsy, snare polypectomy, or endoscopic mucosal resection and histologically assessed by experienced pathologists. Overall CRN was defined as cancer or adenoma, and advanced CRN (ACRN) was defined as cancer or advanced adenoma. Advanced adenoma was defined as the presence of one of the following features: a lesion measuring ≥10 mm in diameter, showing a tubulovillous or villous structure, and high-grade dysplasia. Low-risk adenoma (LRA) was defined as 1–2 adenomas measuring <10 mm in size, and high-risk adenoma (HRA) was defined as advanced adenomas or ≥3 adenomas.20,21 Sessile
ACCEPTED MANUSCRIPT serrated polyps and traditional serrated adenomas were included in the definition of CRN. However, hyperplastic polyps were not considered neoplastic lesions.
Statistical analysis
RI PT
Data were stratified into four groups based on the change in cotinine-verified smoking status between index colonoscopy and follow-up colonoscopy. Groups 1, 2, 3, and 4 were classified as “from cotinine-verified non-smokers at the time of index colonoscopy to cotinine-verified non-
SC
smokers at the time of follow-up colonoscopy,” “from cotinine-verified smokers to cotinine-verified non-smokers,” “from cotinine-verified non-smokers to cotinine-verified smokers,” and “from
M AN U
cotinine-verified smokers to cotinine-verified smokers,” respectively.
Baseline characteristics between the four groups were compared using chi-square test and 1-way analysis of variance for categorical variables and continuous variables, respectively. The cumulative incidence of metachronous CRN was calculated using the Kaplan–Meier method, and the differences between groups were determined using the log-rank test. To compare the risk for development of
TE D
metachronous CRN between groups, we also performed multivariable analysis using the Cox proportional hazards regression model after adjusting for potential confounders, including age, sex, family history of CRC, NSAID use, obesity, hypertension, diabetes mellitus, and baseline adenoma
EP
characteristics (LRA and HRA). All reported P values were two tailed, and P < 0.05 was considered statistically significant. SPSS software version 21 (IBM Corp., Armonk, NY, USA) was used to
AC C
perform all statistical analyses.
Results
Baseline patient characteristics A total of 12,457 patients with ≥1 adenoma at the time of index colonoscopy who underwent follow-up surveillance colonoscopy were eligible for inclusion in this study. Among these, 2313 patients were excluded owing to the following reasons: history of CRC or colorectal surgery (n = 97), diagnosis of CRC at the time of index colonoscopy (n = 9), history of IBD (n = 35), and poor bowel
ACCEPTED MANUSCRIPT preparation at index or follow-up colonoscopy (n = 2172). Of the remaining 10,144 patients, 5382 patients were excluded because they did not undergo urinary cotinine measurements at the index or follow-up colonoscopy. A total of 4762 patients who underwent urinary cotinine measurements at both the index and follow-up colonoscopy were included. The mean age was 40.1 ± 6.3 years, and
RI PT
83.8% were male. The mean interval between index and follow-up colonoscopy was 3.2 ± 1.3 (range, 0.5–6.4) years.
At index colonoscopy, among cotinine-verified non-smokers, the proportion of self-reported
SC
never, former, and current smokers was 48.9%, 35.7%, and 6.6%, respectively, while among cotinineverified smokers, the proportion of self-reported never, former, and current smokers was 3.4%, 5.9%,
M AN U
and 84.5%, respectively (Table 1).
Patients were divided into four groups based on the change in cotinine-verified smoking status: cotinine-verified non-smokers at the time of index colonoscopy → cotinine-verified non-smokers at the time of follow-up colonoscopy (group 1), cotinine-verified smokers → non-smokers (group 2), cotinine-verified non-smokers → smokers (group 3), and cotinine-verified smokers → smokers
TE D
(group 4). Groups 1, 2, 3, and 4 consisted of 2962 (62.2%), 600 (12.6%), 138 (2.9%), and 1062 (22.3%) patients, respectively. Table 2 shows the baseline characteristics of patients in the four groups. Mean patient age was highest in group 4 and lowest in group 2. The proportion of men,
EP
obesity, and diabetes was the lowest in group 1. The proportion of patients with HRA at index
AC C
colonoscopy was high in groups 2 and 4.
Risk of metachronous neoplasia based on change in cotinine-verified smoking status The comparison of the risk of metachronous CRN on follow-up colonoscopy between the four groups is shown in Figure 1. Among the LRA group identified at index colonoscopy, the cumulative risk of metachronous CRN was higher in group 4 than in groups 1 and 2 (P < 0.001 and P < 0.001, respectively), and higher in group 3 than in groups 1 and 2 (P = 0.007 and P = 0.010, respectively), whereas the risk did not differ between groups 1 and 2 (P = 0.686) and between groups 3 and 4 (P = 0.613) (Fig. 1A). Additionally, among the HRA group identified at index colonoscopy, the cumulative
ACCEPTED MANUSCRIPT risk of metachronous CRN was higher in group 4 than in groups 1 and 2 (P < 0.001 and P < 0.001, respectively), and it was higher in group 3 than in groups 1 and 2 (P = 0.008 and P = 0.004, respectively); the risk did not differ between groups 1 and 2 (P = 0.590) and between groups 3 and 4 (P = 0.432) (Fig. 1B).
RI PT
For metachronous ACRN, the cumulative risk in the LRA group was higher in group 4 than in groups 1 and 2 (P < 0.001 and P = 0.033, respectively), whereas the risk was not different between the other groups (Fig. 2A). Among the HRA group, the cumulative risk for metachronous ACRN did not
SC
differ between the groups (Fig. 2B).
Similar results for the risk of metachronous CRN were obtained after adjusting for confounding
M AN U
factors (Table 3). Compared to group 1, the risk of metachronous CRN in groups 3 (adjusted hazard ratio [HR], 1.50; 95% confidence interval [CI], 1.14–1.97) and 4 (adjusted HR, 1.54; 95% CI, 1.36– 1.73) was higher, but the risk in group 2 was not significantly different. Using group 2 as a reference group, the risk of metachronous CRN in groups 3 (adjusted HR, 1.62; 95% CI, 1.20–2.20) and 4 (adjusted HR, 1.63; 95% CI, 1.39–1.99) was higher. Additionally, the risk of metachronous ACRN in
Discussion
EP
2.10; 95% CI, 1.13–3.89).
TE D
group 4 was higher than that in groups 1 (adjusted HR, 2.84; 95% CI, 1.79–4.53) and 2 (adjusted HR,
In this longitudinal study, we found that cotinine-verified smokers had a higher risk of
AC C
metachronous CRN than cotinine-verified non-smokers did, at both index and follow-up colonoscopies. They also had a higher risk of metachronous CRN than patients who switched from “cotinine-verified smokers at index colonoscopy” to “cotinine-verified non-smokers at follow-up colonoscopy.” More importantly, patients who switched from “cotinine-verified non-smokers at index colonoscopy” to “cotinine-verified smokers at follow-up colonoscopy” had a higher risk of metachronous CRN than the risks of patients who switched from “cotinine-verified smokers at index colonoscopy” to “cotinine-verified non-smokers at follow-up colonoscopy” as well as those of cotinine-verified non-smokers at both index and follow-up colonoscopies. In addition, the risk of
ACCEPTED MANUSCRIPT metachronous ACRN in cotinine-verified smokers at both index and follow-up colonoscopies was higher than the risk in patients who switched from “cotinine-verified smokers” to “cotinine-verified non-smokers” as well as cotinine-verified non-smokers at both index and follow-up colonoscopies. These results indicate that the risk of metachronous CRN after polypectomy may be affected by
RI PT
changes in smoking status between index colonoscopy and follow-up colonoscopy.
Several studies have shown that smoking is associated with an increased risk of recurrence of adenomas after polypectomy.9-13 However, the previous studies used self-reported questionnaires to
SC
assess smoking status and evaluated the association between smoking and recurrence of adenoma based on smoking status at baseline, without dynamic assessment of changes in status over time.
M AN U
Some smokers at the index colonoscopy may have stopped smoking during the follow-up period and may have become non-smokers at the follow-up colonoscopy; similarly, some non-smokers at the index colonoscopy may start smoking during the follow-up period and may become smokers at the follow-up colonoscopy. In fact, our study showed that some patients switched from “cotinine-verified smokers at index colonoscopy” to “cotinine-verified non-smokers at follow-up colonoscopy” (12.6%),
TE D
or from “cotinine-verified non-smokers” to “cotinine-verified smokers” (2.9%). In this regard, to better understand the association between smoking and the risk of metachronous CRN after polypectomy, their associations should be assessed based on the change in smoking status between
EP
index colonoscopy and follow-up colonoscopy, and not just the smoking status at index colonoscopy. Current postpolypectomy surveillance guidelines recommend that patients undergo follow-up
AC C
surveillance colonoscopy 3 and 5 years after HRA and LRA removal, respectively.20,21 Therefore, in our study, it can be assumed that the interval between index colonoscopy and follow-up colonoscopy was 3–5 years. It may be difficult to capture changes in smoking status through self-reported questionnaires during this short period. Moreover, assessment of smoking status based on selfreported questionnaires may be inaccurate, and it may induce recall biases and reporting errors. In fact, our previous study showed that some self-reported never smokers were cotinine-verified smokers and had an increased risk of CRN.22 These results suggest that cotinine-captured smoking status exhibits associations with CRN more accurately and objectively than self-reporting does.22 Considering these
ACCEPTED MANUSCRIPT points, in this study, we attempted to identify the changes in smoking status using cotinine, an objective biomarker for tobacco exposure. An interesting result of our study is that compared to “cotinine-verified non-smokers → smokers,” “cotinine-verified smokers → non-smokers” group had a lower risk of metachronous CRN. The same
RI PT
results were obtained when subgroup analysis was conducted for each of the high-risk groups and the low-risk group at the index colonoscopy (Fig. 1). As all adenomas detected at the time of the index colonoscopy were removed, it may be assumed that the two groups are similar at baseline. The risk of
SC
metachronous CRN appeared to increase in “cotinine-verified non-smokers → smokers” group, whereas it appears to decrease in “cotinine-verified smokers → non-smokers” group. Our results
M AN U
suggest that the impact of smoking on CRN may be reversible.
More importantly, our study showed that changes in smoking status affect the risk of developing metachronous ACRN as well as metachronous CRN. Compared to “cotinine-verified non-smokers → non-smokers” group, “cotinine-verified smokers → smokers” group had a higher risk of metachronous ACRN (adjusted HR, 2.84; 95% CI, 1.79–4.53). These results suggest that the
TE D
postpolypectomy surveillance interval should consider the smoking status of the patients as well as the characteristics of the adenoma identified at the time of index colonoscopy. The interval may have to be reduced for patients who continue to smoke. In addition, we found that compared to “cotinine-
EP
verified smokers → smokers” group, “cotinine-verified smokers → non-smokers” group had a lower risk of metachronous ACRN. These results strongly suggest that smoking cessation may help prevent
AC C
CRC. The “cotinine-verified smokers → non-smokers” group might also have healthier lifestyles after smoking cessation (for example, increased physical activity, improved eating habits, and weight loss). Although we did not consider these lifestyle changes during the follow-up period, lifestyle improvements as well as smoking cessation might have resulted in low risk for metachronous CRN in this group. However, among the HRA group, the risk for metachronous ACRN did not differ between the 4 groups based on the change in cotinine-verified smoking status. These results might be because of the small number of patients who developed metachronous ACRN in this cohort. The number of patients
ACCEPTED MANUSCRIPT with HRA was relatively small (n=674), and only 30 (4.5%) of them developed metachronous ACRN. To clarify the relationship between the change in smoking status and the risk of metachronous ACRN in HRA group, studies which include large numbers of patients with HRA are necessary. Meanwhile, we found that among cotinine-verified smokers, the percentage of self-reported
RI PT
current smokers was 84.5%, and among cotinine-verified non-smokers, the percentage of selfreported current smokers was 6.6%. There appears to be good correlation between self-reported smoking information and status based on cotinine verification.
SC
To the best of our knowledge, this is the first study to identify the relationship between cotinineverified change in smoking status and the risk of metachronous CRN. Nevertheless, the current study
M AN U
has several limitations. First, most subjects in our study were employees of various companies. As a result, there was likely some degree of selection bias. Second, the interval between the index and follow-up colonoscopy was relatively short (mean, 3.2 years). A proportion of adenomas found at the time of surveillance colonoscopy might have been missed during the index colonoscopy. However, the follow-up interval was less than one year in only 97 patients (2.0%). Additionally, considering that in
TE D
our center, the overall adenoma detection rate in subjects aged ≥50 years at index colonoscopy (between 2010 and 2014) was 32% (39% for men and 22% for women), exceeding the recommended thresholds for colonoscopy quality,23 the possibility of missed lesions at the index colonoscopy is
EP
attenuated . Further, the follow-up interval was sufficient in the HRA group, even though it might be short in the LRA group. Lastly, urinary cotinine levels were measured only once at the index
AC C
colonoscopy and at the follow-up colonoscopy. This one-time cotinine level may be unable to closely reflect the increase or decrease in the quantity of cigarettes smoked during the interval between the index and follow-up colonoscopy, owing to its short half-life (about 20 hours).24,25 Despite these limitations, our data provides a better understanding of the risk of metachronous CRN according to change in smoking status. The “cotinine-verified smokers at index colonoscopy → cotinine-verified smokers at follow-up colonoscopy” had a higher risk of metachronous ACRN as well as metachronous CRN than the risks of “cotinine-verified non-smokers → non-smokers” and “cotinine-verified smokers → non-smokers.” Further, “cotinine-verified non-smokers at index
ACCEPTED MANUSCRIPT colonoscopy → smokers at follow-up colonoscopy” had a higher risk of metachronous CRN than the risks of “cotinine-verified non-smokers → non-smokers” and “cotinine-verified smokers → nonsmokers.” Our results indicate that changes in smoking status between index and follow-up
AC C
EP
TE D
M AN U
SC
and that smoking cessation may help prevent CRC.
RI PT
colonoscopy have an impact on the risk of developing metachronous CRN following polypectomy
ACCEPTED MANUSCRIPT Figure legends Fig. 1. Cumulative incidence of metachronous colorectal neoplasia based on changes in cotinineverified smoking status in the low-risk (A) and high-risk adenoma group (B) Fig. 2. Cumulative incidence of metachronous advanced colorectal neoplasia based on changes in
AC C
EP
TE D
M AN U
SC
RI PT
cotinine-verified smoking status in the low-risk (A) and high-risk adenoma group (B)
ACCEPTED MANUSCRIPT References 1.
Botteri E, Iodice S, Raimondi S, et al. Cigarette smoking and adenomatous polyps: a metaanalysis. Gastroenterology 2008;134:388-395.
2.
Botteri E, Iodice S, Bagnardi V, et al. Smoking and colorectal cancer: a meta-analysis. Jama
3.
RI PT
2008;300:2765-2778.
Peng L, Weigl K, Boakye D, et al. Risk Scores for Predicting Advanced Colorectal Neoplasia in the Average-risk Population: A Systematic Review and Meta-analysis. Am J Gastroenterol 2018. Tsoi KK, Pau CY, Wu WK, et al. Cigarette smoking and the risk of colorectal cancer: a meta-
SC
4.
analysis of prospective cohort studies. Clin Gastroenterol Hepatol 2009;7:682-688.e681-685. Hannan LM, Jacobs EJ, Thun MJ. The association between cigarette smoking and risk of
M AN U
5.
colorectal cancer in a large prospective cohort from the United States. Cancer Epidemiol Biomarkers Prev 2009;18:3362-3367. 6.
Shrubsole MJ, Wu H, Ness RM, et al. Alcohol drinking, cigarette smoking, and risk of colorectal adenomatous and hyperplastic polyps. Am J Epidemiol 2008;167:1050-1058. Martinez F, Fernandez-Martos C, Quintana MJ, et al. APC and KRAS mutations in distal
TE D
7.
colorectal polyps are related to smoking habits in men: results of a cross-sectional study. Clin Transl Oncol 2011;13:664-671.
Nishihara R, Morikawa T, Kuchiba A, et al. A prospective study of duration of smoking cessation
EP
8.
and colorectal cancer risk by epigenetics-related tumor classification. Am J Epidemiol
9.
AC C
2013;178:84-100.
Jacobson JS, Neugut AI, Murray T, et al. Cigarette smoking and other behavioral risk factors for recurrence of colorectal adenomatous polyps (New York City, NY, USA). Cancer Causes Control 1994;5:215-220.
10. Reid ME, Marshall JR, Roe D, et al. Smoking exposure as a risk factor for prevalent and recurrent colorectal adenomas. Cancer Epidemiol Biomarkers Prev 2003;12:1006-1011. 11. Kim MC, Jung SW, Kim CS, et al. Metabolic syndrome is associated with increased risk of recurrent colorectal adenomas in Korean men. Int J Obes (Lond) 2012;36:1007-1011.
ACCEPTED MANUSCRIPT 12. Lin CC, Huang KW, Luo JC, et al. Hypertension is an important predictor of recurrent colorectal adenoma after screening colonoscopy with adenoma polypectomy. J Chin Med Assoc 2014;77:508-512. 13. Saiken A, Gu F. Lifestyle and lifestyle-related comorbidities independently associated with
RI PT
colorectal adenoma recurrence in elderly Chinese people. Clin Interv Aging 2016;11:801-805. 14. Connor Gorber S, Schofield-Hurwitz S, Hardt J, et al. The accuracy of self-reported smoking: a
Nicotine Tob Res 2009;11:12-24.
SC
systematic review of the relationship between self-reported and cotinine-assessed smoking status.
15. Cross AJ, Boca S, Freedman ND, et al. Metabolites of tobacco smoking and colorectal cancer
M AN U
risk. Carcinogenesis 2014;35:1516-1522.
16. Soweid AM, Kobeissy AA, Jamali FR, et al. A randomized single-blind trial of standard diet versus fiber-free diet with polyethylene glycol electrolyte solution for colonoscopy preparation. Endoscopy 2010;42:633-638.
2009;123:598-601.
TE D
17. Rodu B, Cole P. Smoking prevalence: a comparison of two American surveys. Public Health
18. Biochemical verification of tobacco use and cessation. Nicotine Tob Res 2002;4:149-159. 19. Wen CP, David Cheng TY, Tsai SP, et al. Are Asians at greater mortality risks for being
EP
overweight than Caucasians? Redefining obesity for Asians. Public Health Nutr 2009;12:497-506. 20. Lieberman DA, Rex DK, Winawer SJ, et al. Guidelines for colonoscopy surveillance after
AC C
screening and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer. Gastroenterology 2012;143:844-857. 21. Yang DH, Hong SN, Kim YH, et al. Korean guidelines for postpolypectomy colonoscopy surveillance. Clin Endosc 2012;45:44-61. 22. Kim NH, Jung YS, Yang HJ, et al. Association Between Cotinine-verified Smoking Status and Risk of Colorectal Neoplasia. J Clin Gastroenterol 2018. 23. Rex DK, Boland CR, Dominitz JA, et al. Colorectal Cancer Screening: Recommendations for Physicians and Patients from the U.S. Multi-Society Task Force on Colorectal Cancer. Am J
ACCEPTED MANUSCRIPT Gastroenterol 2017;112:1016-1030. 24. Benowitz NL. Cotinine as a biomarker of environmental tobacco smoke exposure. Epidemiol Rev 1996;18:188-204. 25. Jaakkola MS, Jaakkola JJ. Assessment of exposure to environmental tobacco smoke. Eur Respir
AC C
EP
TE D
M AN U
SC
RI PT
J 1997;10:2384-2397.
ACCEPTED MANUSCRIPT
Cotinine-verified non-smokers
Cotinine-verified smokers†
Self-reported never smokers
1515 (48.9)
Self-reported former smokers
1106 (35.7)
Self-reported current smokers
204 (6.6)
Non-respondents
275 (8.9)
RI PT
Table 1. Frequency of cotinine-verified and self-reported smoking status at the time of index colonoscopy
Total
56 (3.4)
98 (5.9)
1405 (84.5) 103 (6.2)
SC
Smoking status
3100
1662
AC C
EP
TE D
M AN U
Data are presented as number (%). † Cotinine-verified current smoker was defined as an individual having a urinary cotinine level ≥50 ng/mL.
ACCEPTED MANUSCRIPT
Group 4 Cotinine-verified smokers → smokers
138
1062
39.8 ± 5.3
40.7 ±5.9
<0.001
134 (97.1)
1039 (97.8)
<0.001
7 (5.1)
56 (5.3)
0.982
6 (4.3)
19 (1.8)
0.271
P value
600
40.2 ± 6.6
39.0 ± 5.9
2234 (75.4)
582 (97.0)
149 (5.0)
29 (4.8)
71 (2.4)
14 (2.3)
Obesity (BMI ≥25 kg/m )
1075 (36.3)
281 (46.8)
69 (50.0)
543 (51.1)
<0.001
Hypertension
432 (14.3)
92 (15.3)
26 (18.8)
181 (17.0)
0.103
Diabetes mellitus
131 (4.4)
40 (6.7)
9 (6.5)
77 (7.3)
0.002
2574 (86.9)
507 (84.5)
119 (86.2)
888 (83.6)
0.047
93 (15.5)
19 (13.8)
174 (16.4)
3.5 ± 1.3
3.1 ± 1.3
3.0 ± 1.3
Age (years) Men Family history of CRC Use of NSAIDs 2
TE D
Baseline adenoma characteristics Low-risk adenoma
SC
2962
M AN U
No. of patients
Group 3 Cotinine-verified non-smokers → smokers
RI PT
Table 2. Baseline characteristics according to change of cotinine-verified smoking status Group 1 Group 2 Cotinine-verified non-smokers Cotinine-verified smokers Variable → non-smokers → non-smokers
EP
High-risk adenoma 388 (13.1) Interval between index and 3.2 ± 1.3 follow-up colonoscopy (years) Data are presented as number (%) or mean ± standard deviation
AC C
CRC, colorectal cancer; NSAIDs, nonsteroidal anti-inflammatory drugs; BMI, body mass index
<0.001
ACCEPTED MANUSCRIPT
Table 3. Multivariable analysis for the relationship of change in cotinine-verified smoking status with the risk of metachronous colorectal neoplasia Overall CRN
1 (Reference) 0.92 (0.78–1.09)
Group 3 (Cotinine-verified non-smokers → smokers)
1.50 (1.14–1.97)
Group 4 (Cotinine-verified smokers → smokers)
1.54 (1.36–1.73)
Group 1 (Cotinine-verified non-smokers → non-smokers)
1.09 (0.92–1.28)
M AN U
Group 2 (Cotinine-verified smokers → non-smokers)
Adjusted HR (95% CI)
RI PT
Group 1 (Cotinine-verified non-smokers → non-smokers)
P value
P value
1 (Reference)
0.338
1.36 (0.72–2.55)
0.344
0.004
1.17 (0.28–4.87)
0.834
<0.001
2.84 (1.79–4.53)
<0.001
0.338
0.74 (0.39–1.39)
0.344
SC
Adjusted HR (95% CI)
ACRN
Group 2 (Cotinine-verified smokers → non-smokers)
1 (Reference)
1 (Reference)
Group 3 (Cotinine-verified non-smokers → smokers)
1.62 (1.20–2.20)
0.002
0.86 (0.19–3.80)
0.841
Group 4 (Cotinine-verified smokers → smokers)
1.63 (1.39–1.99)
<0.001
2.10 (1.13–3.89)
0.019
TE D
Values were adjusted for age, sex, family history of CRC, NSAID use, obesity, hypertension, diabetes mellitus, and baseline adenoma characteristics (lowand high-risk adenoma)
AC C
EP
CRN, colorectal neoplasia; ACRN, advanced colorectal neoplasia; HR, hazard ratio; CI, confidence interval
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
What You Need to Know
RI PT
Background: Previous assessments of colorectal neoplasia (CRN) recurrence after polypectomy used self-reported, initial smoking status to measure of determine smoking status. We performed a retrospective study of patients undergoing screening colonoscopies and surveillance colonoscopies after polypectomy to evaluate the association between change in smoking status and metachronous CRN. We used cotinine levels in urine as a marker of tobacco exposure.
M AN U
SC
Findings: Cotinine-verified smokers at index colonoscopy who were cotinine-verified smokers at the follow-up colonoscopy had a higher risk of metachronous CRN than cotinine-verified nonsmokers who became non-smokers or cotinine-verified smokers who became non-smokers. Cotinine-verified smokers also had a higher risk of metachronous advanced CRN than cotinineverified non-smokers who became non-smokers or cotinine-verified smokers who became nonsmokers.
AC C
EP
TE D
Implications for patient care: Changes in smoking status between index and follow-up colonoscopy after polypectomy are associated with risk of metachronous CRN. Helping patients quit smoking or following-more frequent screening/surveillance of smokers might prevent colorectal cancer.