ABSTRACT
Background and Aims
Patients with synchronous conventional adenomas and serrated polyps are at increased risk for metachronous neoplasia. We used New Hampshire Colonoscopy Registry data to examine the prevalence, pathway, and severity of synchronous lesions detected on colonoscopy after positive multi‐target‐stool DNA (mt‐sDNA) or Fecal Immunochemical Tests (FIT), compared to colonoscopies without prior stool test.
Methods
We estimated the risk of synchronous conventional adenomas and serrated polyps relative to no lesions, adenomas only, and serrated polyps only, across the three study cohorts using multinomial logistic regression.
Results
We included 1089 mt‐sDNA+ patients, 436 FIT+ patients, and 43 379 patients with screening colonoscopy only without previous stool test. Patients with mt‐sDNA+ tests had a higher risk of synchronous lesions compared to FIT+ patients (OR = 2.34; 95% CI: 1.70–3.28) and colonoscopy‐only patients (OR = 2.92; 95% CI: 2.50–3.40). Mt‐sDNA+ patients also had significantly higher risks of serrated polyps versus no polyps (RRR = 2.92; 95% CI: 2.39–3.58) and synchronous polyps versus adenomas alone (RRR = 2.00; 95% CI: 1.67–2.4) compared to colonoscopy‐only patients; no such differences were found in the FIT+/colonoscopy‐only comparison. Most mt‐sDNA+ and FIT+ patients with synchronous lesions had at least one advanced lesion, while the majority of colonoscopy‐only patients with synchronous lesions had no advanced lesion.
Conclusions
The yield of synchronous conventional adenomas and serrated polyps is higher in patients with positive mt‐sDNA tests compared to FIT+ or colonoscopy‐only patients, likely due in part to the higher yield of serrated polyps in mt‐sDNA+ patients, which persists even among patients with adenomas.
Abbreviations
- AA
advanced adenoma
- BMI
body mass index
- CI
confidence interval
- CRC
colorectal cancer
- FIT
fecal immunochemical test
- HgB
hemoglobin
- IBD
inflammatory bowel disease
- IRB
institutional review board
- mt‐sDNA
multi‐target stool DNA
- NHCR
New Hampshire Colonoscopy Registry
- NSAID
nonsteroidal anti‐inflammatory drug
- OR
odds ratio
- SD
standard deviation
- USMSTF
United States Multi‐Society Task Force on Colorectal Cancer
1. Introduction
Stool‐based tests such as the fecal immunochemical test (FIT) and the multi‐target stool DNA (mt‐sDNA) test are associated with an increased yield of important polyp findings at follow‐up colonoscopy [1, 2, 3, 4]. In particular, published data have demonstrated that patients with positive stool tests have higher rates of adenomas and serrated polyps than patients having colonoscopy without previous stool tests [1, 4, 5]. These data support the use of stool tests to identify patients who will benefit most from endoscopic evaluation to identify and resect these polyps.
Both FIT and mt‐sDNA tests detect human hemoglobin (HgB) in stool, which can be shed from large adenomas and cancers [6, 7, 8]. In contrast to adenomas, serrated polyps are commonly flatter and much less likely to bleed. Therefore, FIT, which only measures fecal HgB, has an inherently low sensitivity for serrated polyp detection [9, 10]. In addition to HgB, the mt‐sDNA test also detects DNA mutation (KRAS) and methylation (BMP3, NDRG4) markers associated with colorectal neoplasia, including serrated polyps [11]. Previous studies have demonstrated that patients with colonoscopy after positive mt‐sDNA test have an increased yield of serrated polyps as compared to patients with colonoscopy after a positive FIT [2, 11]. Because the serrated pathway may account for a large proportion of CRC [12, 13] and serrated polyps are present in a high percentage of colonoscopy patients [14, 15, 16], using stool tests to help identify patients with serrated polyps could improve CRC prevention. One subset of patients who appear to be at particularly increased risk for CRC is individuals with synchronous adenomas and serrated polyps detected at the same colonoscopy. Previous research from the New Hampshire Colonoscopy Registry (NHCR) showed a significant increase in the risk for metachronous advanced neoplasia in patients with synchronous adenomas and serrated polyps [17].
To date, few studies have examined the yield of synchronous adenomas and serrated polyps in patients with positive stool tests. Our aim was to provide real‐world evidence on the yield of synchronous lesions in patients with positive stool tests using data from the NHCR, a statewide population‐based registry. We hypothesized that both mt‐sDNA‐positive and FIT‐positive patients would be more likely to have synchronous lesions identified at follow up colonoscopy, as compared to patients evaluated by colonoscopy without a prior positive stool test result. We further hypothesized that, given the additional markers present in the mt‐sDNA test, patients with a positive mt‐sDNA test result would have a higher yield of synchronous lesions than FIT+ patients. We also sought to evaluate differences between study cohorts (mt‐sDNA+, FIT+, and colonoscopy only) in the distribution of patients with synchronous polyps versus polyps in just one pathway (adenomatous or serrated), as well as the level of severity of findings within each pathway among patients with synchronous polyps.
2. Methods
2.1. Population
In this study we used data from the NHCR, which collects data on demographics, lifestyle, and health behaviors, medical history, and family history from colonoscopy patients at centers across New Hampshire. The NHCR also obtains detailed colonoscopy data, including information on exam indication, completion status, bowel preparation quality, and the size, histology, and location of all polyp findings, recorded by endoscopists or endoscopy nurses during or immediately following the exam. Pathology results are obtained directly from pathology labs, and trained NHCR abstractors match the outcomes on those reports to the polyp‐level findings recorded by the endoscopist or nurse during the colonoscopy.
All data collection and study procedures were approved by the Dartmouth College and Dartmouth‐Health IRBs in accordance with the Belmont Report and US Common Rule.
2.2. Study Cohorts
Patients with positive mt‐sDNA tests within the NHCR catchment area (New Hampshire as well as the neighboring states of Massachusetts, Maine, and Vermont) were provided to the NHCR by Exact Sciences Laboratories LLC (Madison, Wisconsin) under an IRB‐approved protocol. NHCR participants with positive mt‐sDNA results and a subsequent follow up colonoscopy between 2/1/2015 and 9/1/2024 were identified by NHCR staff. Our sample included all average‐risk equivalent (no personal history of polyps or colorectal cancer and no family history of colorectal cancer) patients identified as having a positive stool test (mt‐sDNA or FIT) and a complete colonoscopy with adequate bowel preparation recorded in the NHCR, as well as a reference group of average‐risk patients with a complete screening colonoscopy without a prior positive stool test (hereafter referred to as “colonoscopy only”). We excluded exams performed on patients under 45 years old, who are ineligible for screening colonoscopy under current guidelines, as well as those performed on patients with inflammatory bowel disease (IBD) or a genetic syndrome (e.g., Lynch Syndrome), since these conditions require distinct surveillance.
2.3. Outcomes
Our primary outcome in this study was the occurrence of synchronous adenomatous and serrated polyps, or in other words the detection of both adenomatous and serrated polyps during a single colonoscopy, both as a binary variable (yes/no synchronous polyps) and as a partially ordered categorical variable to reflect the underlying structure of the outcome as the co‐occurrence of two separate polyp types. The categories for this variable were (1) no polyps, (2) adenomatous polyps only, (3) serrated polyps only, and (4) synchronous adenomatous and serrated polyps. In our database we have classified serrated polyps as recommended by the 2010 WHO recommendations into hyperplastic polyps, sessile serrated polyps, and traditional serrated adenomas [18]. We defined adenomas to include any villous, tubulovillous, or tubular adenoma, while serrated polyps included any traditional serrated adenoma (TSA), sessile serrated polyp (SSP), or hyperplastic polyp (HP). We included HPs since they may be associated with a higher risk for metachronous large serrated polyps [19] and some HPs may actually be SSPs [20]. We also performed a sensitivity analysis restricting serrated polyps to SSPs and TSAs.
In a secondary analysis restricted to patients with synchronous polyps, we evaluated the severity of the polyps in both the adenomatous and serrated pathways using another partially‐ordered categorical variable with the following categories: (1) neither pathway advanced, (2) advanced adenoma and non‐advanced serrated polyp, (3) advanced serrated polyp and non‐advanced adenoma, and (4) advanced adenoma and advanced serrated polyp. We defined an advanced adenoma as any adenoma at least 1 cm in diameter, with villous or tubulovillous histology, or with high‐grade dysplasia; we defined advanced serrated polyps as any serrated polyp at least 1 cm in diameter, any TSA, or any SSP.
2.4. Statistical Approach
We compared patient and exam characteristics across our three study cohorts (mt‐sDNA+, FIT+, and colonoscopy only) using Fisher's exact test for binary and categorical variables and Kruskal‐Wallis tests for continuous variables (Table 1). We then calculated the proportion of exams with synchronous polyps as well as the proportion of underlying individual polyp outcomes (adenomas and serrated polyps) across the three groups (Table 2A).
TABLE 1.
Patient characteristics by study cohort.
| Characteristic | mt‐sDNA+ (N = 1089) | FIT+ (N = 436) | Colonoscopy only (N = 43 379) | p |
|---|---|---|---|---|
| N (% or SD) | N (% or SD) | N (% or SD) | ||
| Age (years, continuous) | 64.8 (8.6) | 65.2 (8.9) | 58.6 (7.8) | 0.000 |
| Patient sex | ||||
| Male | 438 (40.2) | 216 (49.5) | 20 379 (47) | < 0.001 |
| Female | 651 (59.8) | 220 (50.5) | 23 018 (53.0) | |
| Patient race | ||||
| Caucasian | 822 (97.3) | 345 (96.6) | 34 985 (96.1) | 0.228 |
| Not Caucasian | 23 (2.7) | 12 (3.4) | 1402 (3.9) | |
| BMI (continuous) | 29.1 (8.4) | 29.3 (7.9) | 28.8 (7.3) | 0.370 |
| Smoking status | ||||
| Never | 426 (47.9) | 180 (49.5) | 22 838 (60.8) | < 0.001 |
| Former | 361 (40.6) | 147 (40.4) | 12 069 (32.1) | |
| Current | 103 (11.6) | 37 (10.2) | 2651 (7.1) | |
| Self‐reported health status | ||||
| Good to excellent | 794 (90.4) | 322 (88.2) | 35 658 (95.0) | < 0.001 |
| Fair | 77 (8.8) | 36 (9.9) | 1745 (4.6) | |
| Poor | 7 (0.8) | 7 (1.9) | 125 (0.3) | |
| Aspirin or NSAIDs use | 225 (28.4) | 131 (39.3) | 11 055 (31.6) | 0.002 |
| Blood thinner use | 76 (8.9) | 32 (8.7) | 826 (2.2) | < 0.001 |
| History of prior colonoscopy | 461 (42.3) | 233 (53.4) | 20 003 (46.1) | < 0.001 |
| Endoscopist ADR (median [IQR]) | 42 (14) | 39 (16) | 38 (15) | < 0.001 |
Note: Continuous variables compared using the Kruskal–Wallis test; categorical variables compared using Fisher's exact test. Percentages of missing data (mt‐sDNA, FIT, colonoscopy only): Race (22, 18, 16), BMI (27, 24, 21), Smoking status (18, 17, 13), Health status (19, 16, 14), Aspirin or NSAIDS (27, 24, 20), Blood thinners (21, 16, 13).
TABLE 2A.
Frequencies and proportions of polyp outcomes with 95% CIs for proportions, by study cohort.
| Polyp outcomes | mt‐sDNA+ (N = 1089) | FIT+ (N = 436) | Colonoscopy only (N = 43 379) | |||
|---|---|---|---|---|---|---|
| N | % (95% CI) | N | % (95% CI) | N | % (95% CI) | |
| Any adenoma | 615 | 56.5 (53.5, 59.4) | 217 | 49.8 (45.0, 54.6) | 14 854 | 34.2 (33.8, 34.7) |
| Advanced adenoma | 206 | 18.9 (16.6, 21.4) | 78 | 17.9 (14.4, 21.8) | 2374 | 5.5 (5.3, 5.7) |
| Any serrated polyp | 546 | 50.1 (47.1, 53.1) | 136 | 31.2 (26.9, 35.8) | 12 480 | 28.8 (28.3, 29.2) |
| Advanced serrated polyp | 259 | 23.8 (21.3, 26.4) | 55 | 12.6 (9.6, 16.1) | 4018 | 9.3 (9.0, 9.5) |
| Synchronous adenoma and serrated polyp | 319 | 29.3 (26.6, 32.1) | 77 | 17.7 (14.2, 21.6) | 5131 | 11.8 (11.5,12.1) |
For our primary analysis, we performed logistic regressions to estimate the difference in the odds of having synchronous polyps (versus no polyps or only a single polyp type) between each pair of study cohorts (colonoscopy only versus FIT+, colonoscopy only versus mt‐sDNA+, and FIT+ versus mt‐sDNA+), first without adjustment and then with adjustment for patient age (centered at 50, in terms of 5‐year increments), sex, and smoking status (current smoker versus former or non‐smoker) (Table 2B). We then used multinomial logistic regression to estimate the differences in the risks of a detailed range of possible outcomes (no polyps, adenomas only, serrated polyps only, or synchronous adenomas and serrated polyps) for each pair of study cohorts, both unadjusted and adjusted, as above. Table 3A presents the relative risk of having either an adenoma only or a serrated polyp only (versus no polyps), exploring differences in the “first step” toward synchronous polyps, in which an individual develops polyps in one pathway (either adenomatous or serrated). Table 3B presents the relative risk of having synchronous lesions versus having only adenomas or only serrated polyps, exploring differences in the “second step,” in which the second pathway is added to the first, resulting in synchronous lesions.
TABLE 2B.
Logistic regression of odds of synchronous polyps versus any other outcomes.
| Term | Synchronous polyps versus any other outcome | |
|---|---|---|
| Unadjusted OR (95% CI) | Adjusted OR (95% CI) | |
| FIT+ vs. colonoscopy only (ref) | 1.60 (1.24–2.04) | 1.25 (0.92–1.66) |
| mt‐sDNA+ vs. colonoscopy only (ref) | 3.09 (2.70–3.53) | 2.92 (2.50–3.40) |
| mt‐sDNA+ vs. FIT+ (ref) | 1.93 (1.47–2.57) | 2.34 (1.70–3.28) |
TABLE 3.
Unadjusted and adjusted multinomial logistic regression results describing the relative risk of having outcomes in the comparison versus base category, by study cohort.
| Outcome base category | Outcome comparison category | Study cohort reference | Study cohort comparison | Unadjusted relative risk ratio (95% CI) | Adjusted relative risk ratio (95% CI) |
|---|---|---|---|---|---|
| A | |||||
| No polyps | Adenoma only | Colonoscopy only | FIT+ | 1.91 (1.52–2.4) | 1.65 (1.29–2.12) |
| mt‐sDNA+ | 2.61 (2.2–3.1) | 2.63 (2.18–3.18) | |||
| FIT+ | mt‐sDNA+ | 1.37 (1.03–1.82) | 1.59 (1.17–2.17) | ||
| Serrated only | Colonoscopy only | FIT+ | 1.06 (0.79–1.44) | 1.10 (0.79–1.53) | |
| mt‐sDNA+ | 2.65 (2.21–3.18) | 2.92 (2.39–3.58) | |||
| FIT+ | mt‐sDNA+ | 2.49 (1.76–3.53) | 2.66 (1.81–3.89) | ||
| Synchronous adenoma and serrated | Colonoscopy only | FIT+ | 1.99 (1.51–2.61) | 1.51 (1.09–2.08) | |
| mt‐sDNA+ | 5.33 (4.51–6.32) | 5.27 (4.35–6.4) | |||
| FIT+ | mt‐sDNA+ | 2.68 (1.95–3.69) | 3.49 (2.41–5.06) | ||
| B | |||||
| Adenoma only | Synchronous adenoma and serrated | Colonoscopy only | FIT+ | 1.04 (0.79–1.38) | 0.91 (0.66–1.27) |
| mt‐sDNA+ | 2.04 (1.74–2.4) | 2.00 (1.67–2.4) | |||
| FIT+ | mt‐sDNA+ | 1.96 (1.42–2.7) | 3.87 (3.47–4.32) | ||
| Serrated only | Colonoscopy only | FIT+ | 1.87 (1.33–2.63) | 1.37 (0.93–2.02) | |
| mt‐sDNA+ | 2.01 (1.69–2.39) | 1.80 (1.48–2.2) | |||
| FIT+ | mt‐sDNA+ | 1.08 (0.74–1.57) | 2.19 (1.52–3.17) | ||
For our secondary analysis exploring the presence of advanced lesions in each pathway among patients with synchronous polyps, we calculated the proportion of exams with no advanced polyps, with advanced adenomas only, with advanced serrated polyps only, and with advanced lesions in both pathways across the three study cohorts (Table 4). We also report on the proportion of exams with neither adenomas nor serrated polyps, with adenomas only, with serrated polyps only, and with both adenomatous and serrated polyps across the cohorts (Table S1). We again used multinomial logistic regression to estimate the differences in severity in the adenomatous and serrated pathways between each pair of study cohorts, first without and then with adjustment as above (Table 5).
TABLE 4.
Contingency table for advanced adenomas and serrated polyps within patients with synchronous polyps by study cohort.
| No advanced serrated | Advanced serrated | |||
|---|---|---|---|---|
| # | % (95% CI) | # | %(95% CI) | |
| Mt‐sDNA+ | ||||
| No advanced adenoma | 114 | 35.7% (30.5–41.3) | 109 | 34.2% (29–39.7) |
| Advanced adenoma | 61 | 19.1% (15–23.9) | 35 | 11% (7.8–14.9) |
| FIT+ | ||||
| No advanced adenoma | 28 | 36.4% (25.7–48.1) | 24 | 31.2% (21.1–42.7) |
| Advanced adenoma | 16 | 20.8% (12.4–31.5) | 9 | 11.7% (5.5–21) |
| Colonoscopy only | ||||
| No advanced adenoma | 2976 | 58% (56.6–59.4) | 1289 | 25.1% (23.9–26.3) |
| Advanced adenoma | 534 | 10.4% (9.6–11.3) | 332 | 6.5% (5.8–7.2) |
TABLE 5.
Unadjusted and adjusted multinomial logistic regression results describing the relative risk of having outcomes in the comparison versus base category, by study cohort.
| Outcome base category | Outcome comparison category | Study cohort reference | Study cohort comparison | Unadjusted relative risk ratio (95% CI) | Adjusted relative risk ratio (95% CI) |
|---|---|---|---|---|---|
| Neither advanced | Both advanced | Colonoscopy only | FIT+ | 2.88 (1.35–6.16) | 2.97 (1.24–7.13) |
| mt‐sDNA+ | 2.75 (1.85–4.09) | 3.02 (1.94–4.69) | |||
| FIT+ | mt‐sDNA+ | 0.96 (0.41–2.22) | 1.01 (0.39–2.63) | ||
| Advanced adenoma only | Colonoscopy only | FIT+ | 3.18 (1.71–5.93) | 2.80 (1.30–6.03) | |
| mt‐sDNA+ | 2.98 (2.16–4.12) | 3.23 (2.22–4.70) | |||
| FIT+ | mt‐sDNA+ | 0.94 (0.47–1.86) | 1.15 (0.50–2.66) | ||
| Advanced serrated only | Colonoscopy only | FIT+ | 1.98 (1.14–3.43) | 1.72 (0.89–3.32) | |
| mt‐sDNA+ | 2.21 (1.68–2.89) | 2.08 (1.52–2.84) | |||
| FIT+ | mt‐sDNA+ | 1.12 (0.61–2.04) | 1.21 (0.59–2.47) | ||
| Advanced adenoma only | Both advanced | Colonoscopy only | FIT+ | 0.90 (0.40–2.07) | 1.06 (0.40–2.86) |
| mt‐sDNA+ | 0.92 (0.60–1.43) | 0.93 (0.57–1.53) | |||
| FIT+ | mt‐sDNA+ | 1.02 (0.41–2.55) | 0.88 (0.30–2.56) | ||
| Advanced serrated only | Colonoscopy only | FIT+ | 1.46 (0.67–3.16) | 1.73 (0.70–4.29) | |
| mt‐sDNA+ | 1.25 (0.84–1.86) | 1.45 (0.93–2.28) | |||
| FIT+ | mt‐sDNA+ | 0.86 (0.36–2.01) | 0.84 (0.31–2.24) |
3. Results
Our analysis included 1089 patients with colonoscopy following positive mt‐sDNA tests, 436 with colonoscopy after positive FIT, and 43 379 patients with colonoscopy only without previous positive stool tests. Table 1 presents the demographics and risk factors of all three groups.
In Table 2A, we present the frequencies and proportions of outcomes with 95% confidence intervals by study cohort. The mt‐sDNA+ cohort had the highest proportions for all polyp outcomes (any adenoma, advanced adenoma, any serrated polyp, advanced serrated polyp, and synchronous adenoma and serrated polyp). After logistic regression (Table 2B), we observed that the odds for synchronous polyps compared to any other outcome were significantly higher in mt‐sDNA+ patients than in FIT+ and colonoscopy‐only patients. The odds for synchronous polyps were also significantly higher in FIT+ patients than in colonoscopy‐only patients.
In Table 3, we present the results of unadjusted and adjusted multinomial logistic regression models comparing the risk of four possible outcomes—no polyps, adenoma only, serrated polyp only, or synchronous adenomas and serrated polyps—across all three study cohorts. These results are estimates of the relative difference in risk between each of the three cohorts for each pair of possible outcomes (e.g., no polyps versus adenomas only). We present estimates of the relative difference in risk of having no polyps versus each individual polyp type (adenoma OR serrated) alone and then versus synchronous polyps in Table 3A. Table 3B explores the risk of individuals with polyps in one pathway developing polyps in the second pathway, comparing the risk of having an adenoma only or serrated polyps only to having synchronous lesions.
FIT+ and mt‐sDNA+ patients had a higher risk of adenomas alone (versus no polyps) than colonoscopy‐only patients; mt‐sDNA+ patients were also more likely than FIT+ patients to have an adenoma versus no polyps (Table 3A). Mt‐sDNA+ patients were also significantly more likely to have serrated polyps alone (versus no polyps) than both colonoscopy‐only and FIT+ patients, but there was no difference between FIT+ and colonoscopy‐only patients.
Table 3B compares the relative risk of synchronous polyps versus each individual polyp type across the three study cohorts. We found that mt‐sDNA+ patients were significantly more likely to have synchronous polyps than adenomas alone compared to both the FIT+ and colonoscopy‐only groups, whereas there was no difference between FIT+ and colonoscopy‐only in the relative risk of those two outcomes. However, both FIT+ and mt‐sDNA+ patients had a significantly higher risk of synchronous polyps relative to serrated polyps alone compared to colonoscopy‐only patients, while there was no difference between the two positive stool test groups. We also found that mt‐sDNA+ patients had a higher risk of synchronous polyps versus no polyps compared to the other two groups, and that both positive stool test groups had a higher risk of synchronous polyps versus no polyps than the colonoscopy‐only group (Table 3A). In particular, patients with mt‐sDNA+ tests were at 5.27 times higher risk (95% CI: 4.35–6.40) of having a synchronous adenoma and serrated polyp compared to no polyps than the colonoscopy‐only group and 3.49 times higher risk (95% CI: 2.41–5.06) than FIT+ patients. FIT+ patients were at 1.51 times higher risk (95% CI: 1.09–2.08) of having synchronous polyps compared to no polyps than colonoscopy‐only patients. Our sensitivity analysis restricting serrated polyps to SSPs and TSAs had similar results as shown in Tables S2A, S2B, and S3.
Table 4 presents results from our analysis of advanced lesions in each pathway among patients with synchronous polyps. Most mt‐sDNA+ and FIT+ patients with synchronous polyps had at least one advanced polyp, whereas the majority of colonoscopy‐only patients with synchronous polyps had no advanced polyps.
Table 5 presents relative risk ratios and associated 95% confidence intervals from two multinomial logistic regression models, one unadjusted and one adjusted as above, with advanced polyp histology (no advanced polyp, advanced adenoma only, advanced serrated polyp only, or advanced lesions in both pathways) as the outcome. When compared to colonoscopy only patients, both mt‐sDNA+ and FIT+ patients with synchronous polyps were significantly more likely to have advanced polyps in both pathways (mt‐sDNA OR 3.02; 95% CI: 1.94–4.69; FIT OR 2.97; 95% CI: 1.24–7.13) or advanced adenomas (mt‐sDNA OR 3.23; 95% CI: 2.22–4.70; FIT OR 2.80; 95% CI: 1.30–6.03) versus neither pathway advanced. Mt‐sDNA+ patients were also more likely than colonoscopy only patients to have advanced serrated polyps versus neither pathway advanced (OR 2.08; 95% CI: 1.52–2.84); this was not the case for FIT+ patients. However, neither mt‐sDNA+ nor FIT+ patients were at increased risk of having synchronous advanced polyps in both pathways relative to advanced polyps in a single pathway compared to colonoscopy‐only patients.
4. Discussion
In this analysis we examined the risk for synchronous serrated polyps and conventional adenomas, a diagnosis which may place an individual at higher risk for metachronous neoplasia [17, 21]. Patients who have both synchronous serrated polyps and conventional adenomas have been shown in a few studies to have an increased risk for metachronous advanced adenomas [17, 21]. We found that positive stool tests are associated with higher polyp yield at follow‐up colonoscopy as previously reported [1, 4, 5, 22]. Moreover, the yield for synchronous polyps was significantly higher after a positive mt‐sDNA test compared to either FIT+ or colonoscopy only patients, while FIT+ patients were also found to have a higher yield of synchronous polyps compared to colonoscopy only patients. Employing mt‐sDNA or, to a lesser extent FIT, as the initial CRC screening strategy could therefore be utilized to target patients at increased risk of synchronous lesions and improve the efficiency of CRC prevention.
In our study we observed that almost 1/3 of the patients with positive mt‐sDNA tests (29.3%) had a synchronous serrated polyp and conventional adenoma. The rate was higher as compared to FIT+ (17.7%) and colonoscopy‐only (11.8%) patients, a relationship which persisted after adjustment in our logistic regression model (Table 2B), with mt‐sDNA+ patients having a higher risk of synchronous lesions when compared to FIT+ patients (OR = 2.27; 95% CI: 1.74–2.96) or colonoscopy‐only patients (OR = 3.30; 95% CI: 2.82–3.88).
We also used multinomial logistic regression to model the synchronous polyp outcome as the co‐occurrence of two separate polyp outcomes, adenomatous and serrated. Our results from this analysis expand on past evidence contrasting how adenomatous and serrated polyp yield among mt‐sDNA+ and FIT+ differ relative to colonoscopy‐only patients.
When we compared outcomes that differ by the presence of an adenoma (no polyps vs. adenoma only, serrated only versus synchronous polyps), relative risks within the FIT+ cohort were significantly higher than in the colonoscopy‐only cohorts, whereas when we compare outcomes that differ by the presence of a serrated polyp (e.g., no polyp versus serrated polyp, adenoma only versus synchronous) there is no significant difference in relative risk between the two cohorts (FIT+ and colonoscopy‐only, Table 3). In contrast, relative risks were significantly higher in both the serrated and adenomatous pathways in mt‐sDNA+ patients relative to colonoscopy‐only patients. These differences in performance between FIT+ and mt‐sDNA+ patients reflect the inclusion of molecular markers (in combination with hemoglobin) in the latter assay.
Finally, we also compared the level of severity of the adenomatous and serrated pathways among patients with synchronous polyps across the three cohorts. Most stool‐test‐positive patients with synchronous polyps had at least one advanced polyp (Table 4), whereas most colonoscopy‐only patients with synchronous polyps had no advanced polyps. Overall, we observed that mt‐sDNA+ patients with synchronous adenomas and serrated polyps had a significantly higher relative risk of advanced polyps of all kinds vs. no polyps than did colonoscopy‐only patients, while for FIT+, as above, this relationship is comparison‐dependent and stronger in the adenomatous pathway.
An important strength of our analysis is that our models were controlled for age, sex, and smoking. Smoking is an important risk factor for patients with serrated polyps and advanced adenomas [23, 24, 25, 26, 27, 28, 29], and our published data suggest that smoking may be a stronger risk factor for patients with synchronous adenomas and serrated polyps than for those who have serrated polyps or adenomas alone [23]. Another strength is that our population‐based data incorporate multiple health centers and systems and reflects real‐world practice.
One limitation is that our data are from a single state which lacks significant racial diversity. However, there is substantial ethnic, socioeconomic, and rural/urban variation present within NHCR data [30]. Nonetheless, future research examining the prevalence of patients with synchronous lesions in other populations is needed. Although the mt‐sDNA–positive patients showed a higher detection of synchronous serrated polyps and adenomas than FIT‐positive patients, as shown in Table 4 the proportion of advanced lesions among patients with synchronous polyps is similar between the mt‐sDNA and FIT groups. Thus, despite the increased serrated polyp detection with mt‐sDNA, the test may be similar to FIT in detecting advanced neoplasia. Both stool tests had a higher proportion of advanced lesions than colonoscopy.
In our analysis we have excluded patients with poor bowel preparation but we did not assess withdrawal time, and therefore could not account for the possibility that endoscopists may have spent more time inspecting the colon in patients with positive stool tests as has been shown in a previous paper [31]. However, as shown in Table 1, median ADRs were high across all cohorts, exceeding the currently recommended benchmark of 35% [32], with only slightly increased ADR in the mt‐sDNA+ and FIT+ cohorts over colonoscopy only (42%, 39%, and 38%, respectively).
In summary, we observed an increased prevalence of synchronous adenomas and serrated polyps in patients who underwent colonoscopy following a positive stool test result relative to screening colonoscopy‐only patients, with the highest rate of both individual polyp outcomes and synchronous polyps found in mt‐sDNA positive patients. This is likely due to the higher yield of serrated polyps in mt‐sDNA+ patients, which persists even among patients with adenomas. Compared to both the FIT+ and colonoscopy‐only groups, mt‐sDNA+ patients were significantly more likely to have synchronous polyps than adenomas alone. Both FIT+ and mt‐sDNA+ patients were significantly more likely to have synchronous polyps relative to serrated polyps alone compared to colonoscopy‐only patients. Finally, most mt‐sDNA+ and FIT+ patients with synchronous polyps had at least one advanced polyp, while the majority of colonoscopy‐only patients with synchronous polyps had no advanced polyps. Using stool tests, in particular mt‐sDNA, as the initial CRC screening test could help target patients at increased risk of synchronous lesions, especially those with advanced polyps, improving the efficacy of screening for CRC prevention.
Funding
Exact Sciences provided funding support for this analysis to the New Hampshire Colonoscopy Registry (staffed by L. Butterly, W. Hisey, and C. Robinson). The funding agreement ensured that the NHCR authors had independence in designing the study, conducting analyses, and writing and publishing the results. The contents of this work do not represent the views of the Department of Veterans Affairs or the United States Government.
Ethics Statement
All data collection and study procedures were approved by the Committee for the Protection of Human Subjects at Dartmouth College (CPHS#00015834) or the Dartmouth Health IRB (Study 02002782) in accordance with the Belmont Report and the US Common Rule.
Conflicts of Interest
P.J.L. serves as Chief Medical Officer for Screening at Exact Sciences and holds stock in the company. B.L.K. is an employee at Exact Sciences.
Supporting information
Table S1: Contingency table of adenomatous and serrated polyps, by study cohort.
Table S2A: Frequencies and proportions of polyp outcomes with 95% CIs for proportions, by study cohort, modified definition of serrated polyps (SSPs and TSAs only).
Table S2B: Logistic regression of odds of synchronous polyps versus any other outcomes, modified definition of serrated polyps (SSPs and TSAs only).
Table S3: Unadjusted and adjusted multinomial logistic regression results describing the relative risk of having outcomes in the comparison versus base category, by study cohort, modified definition of serrated polyps (SSPs and TSAs only).
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Anderson J. C., Robinson C. M., Hisey W. M., et al., “Colorectal Neoplasia Detection in Individuals With Positive Multitarget Stool DNA Tests: Data From the New Hampshire Colonoscopy Registry,” Journal of Clinical Gastroenterology 56, no. 5 (2022): 419–425. [DOI] [PubMed] [Google Scholar]
- 2. Bosch L. J. W., Melotte V., Mongera S., et al., “Multitarget Stool DNA Test Performance in an Average‐Risk Colorectal Cancer Screening Population,” American Journal of Gastroenterology 114 (2019): 1909–1918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Anderson J. C., Hisey W. M., Robinson C. M., Limburg P. J., Kneedler B. L., and Butterly L. F., “Serrated Polyp Yield at Colonoscopy in Patients With Positive FIT, Positive Mt‐sDNA, and Colonoscopy Only: Data From the New Hampshire Colonoscopy Registry,” Cancer Epidemiology, Biomarkers & Prevention 32 (2023): 226–232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Butterly L. F., Hisey W. M., Robinson C. M., Limburg P. J., Kneedler B. L., and Anderson J. C., “What Do ‘false‐Positive’ Stool Tests Really Mean? Data From the New Hampshire Colonoscopy Registry,” Preventive Medical Reports 35 (2023): 102309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Anderson J. C., Robinson C. M., Hisey W. M., et al., “Colorectal Neoplasia Detection in Individuals With Positive Multitarget Stool DNA Tests: Data From the New Hampshire Colonoscopy Registry,” Journal of Clinical Gastroenterology 56 (2022): 419–425. [DOI] [PubMed] [Google Scholar]
- 6. Digby J., Fraser C. G., Carey F. A., et al., “Faecal Haemoglobin Concentration Is Related to Severity of Colorectal Neoplasia,” Journal of Clinical Pathology 66 (2013): 415–419. [DOI] [PubMed] [Google Scholar]
- 7. Heigh R. I., Yab T. C., Taylor W. R., et al., “Detection of Colorectal Serrated Polyps by Stool DNA Testing: Comparison With Fecal Immunochemical Testing for Occult Blood (FIT),” PLoS One 9 (2014): e85659. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. van Doorn S. C., Stegeman I., Stroobants A. K., et al., “Fecal Immunochemical Testing Results and Characteristics of Colonic Lesions,” Endoscopy 47 (2015): 1011–1017. [DOI] [PubMed] [Google Scholar]
- 9. Chang L. C., Shun C. T., Hsu W. F., et al., “Fecal Immunochemical Test Detects Sessile Serrated Adenomas and Polyps With a Low Level of Sensitivity,” Clinical Gastroenterology and Hepatology 15 (2016): 872–879.e1. [DOI] [PubMed] [Google Scholar]
- 10. Mowat C., Digby J., Strachan J. A., Steele R. J. C., and Fraser C. G., “Low Sensitivity of Fecal Immunochemical Tests (FIT) for Detection of Sessile Serrated Adenomas/Polyps Confirmed Over Clinical Setting, Geography, and FIT System,” Digestive Diseases and Sciences 64 (2019): 3024–3026. [DOI] [PubMed] [Google Scholar]
- 11. Imperiale T. F., Ransohoff D. F., Itzkowitz S. H., et al., “Multitarget Stool DNA Testing for Colorectal‐Cancer Screening,” New England Journal of Medicine 370 (2014): 1287–1297. [DOI] [PubMed] [Google Scholar]
- 12. Holme O., Bretthauer M., Eide T. J., et al., “Long‐Term Risk of Colorectal Cancer in Individuals With Serrated Polyps,” Gut 64 (2015): 929–936. [DOI] [PubMed] [Google Scholar]
- 13. Rex D. K., Ahnen D. J., Baron J. A., et al., “Serrated Lesions of the Colorectum: Review and Recommendations From an Expert Panel,” American Journal of Gastroenterology 107 (2012): 1315–1329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Anderson J. C., “Pathogenesis and Management of Serrated Polyps: Current Status and Future Directions,” Gut Liver 8 (2014): 582–589. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Anderson J. C., Butterly L. F., Goodrich M., Robinson C. M., and Weiss J. E., “Differences in Detection Rates of Adenomas and Serrated Polyps in Screening Versus Surveillance Colonoscopies, Based on the New Hampshire Colonoscopy Registry,” Clinical Gastroenterology and Hepatology 11 (2013): 1308–1312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Anderson J. C., Butterly L. F., Weiss J. E., and Robinson C. M., “Providing Data for Serrated Polyp Detection Rate Benchmarks: An Analysis of the New Hampshire Colonoscopy Registry,” Gastrointestinal Endoscopy 85 (2017): 1188–1194. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Anderson J. C., Butterly L. F., Robinson C. M., Weiss J. E., Amos C., and Srivastava A., “Risk of Metachronous High‐Risk Adenomas and Large Serrated Polyps in Individuals With Serrated Polyps on Index Colonoscopy: Data From the New Hampshire Colonoscopy Registry,” Gastroenterology 154 (2018): 117–127.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Snover D. C., Ahnen D. J., Burt R. W., et al., “Serrated Polyps of the Colon and Rectum and Serrated (‘Hyperplastic’) Polyposis,” in WHO Classification of Tumours of the Digestive System, 4th ed., ed. Bosman F. T., Carneiro F., Hruban R. H., and Theise N. D. (IARC, 2010), 160–165. [Google Scholar]
- 19. Anderson J. C., Butterly L. F., and Robinson C. M., “Are Proximal 5‐9 Mm Hyperplastic Polyps Significant Lesions? Data Examining Metachronous Risk of Large (≥1cm) Serrated Polyps and Advanced Conventional Adenomas for Patients With These Lesions in the New Hampshire Colonoscopy Registry,” Gastrointestinal Endoscopy 91 (2020): AB110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Anderson J. C., Lisovsky M., Greene M. A., Hagen C., and Srivastava A., “Factors Associated With Classification of Hyperplastic Polyps as Sessile Serrated Adenomas/Polyps on Morphologic Review,” Journal of Clinical Gastroenterology 52 (2018): 524–529. [DOI] [PubMed] [Google Scholar]
- 21. Laven‐Law G., Symonds E. L., Simpson K., et al., “Risk of Advanced Colorectal Neoplasia at Follow‐Up Colonoscopy After Synchronous Adenoma and Clinically Significant Serrated Polyp,” Clinical Gastroenterology and Hepatology 24 (2025): 544–553. [DOI] [PubMed] [Google Scholar]
- 22. Anderson J. C., Robinson C. M., Hisey W., Limburg P. J., and Butterly L. F., “Colonoscopy Findings in FIT+ and Mt‐sDNA+ Patients Versus in Colonoscopy‐Only Patients: New Hampshire Colonoscopy Registry Data,” Cancer Prevention Research (Philadelphia, Pa.) 15 (2022): 455–464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Anderson J. C., Calderwood A. H., Christensen B. C., Robinson C. M., Amos C. I., and Butterly L., “Smoking and Other Risk Factors in Individuals With Synchronous Conventional High‐Risk Adenomas and Clinically Significant Serrated Polyps,” American Journal of Gastroenterology 113 (2018): 1828–1835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Anderson J. C., Moezardalan K., Messina C. R., Latreille M., and Shaw R. D., “Smoking and the Association of Advanced Colorectal Neoplasia in an Asymptomatic Average Risk Population: Analysis of Exposure and Anatomical Location in Men and Women,” Digestive Diseases and Sciences 56 (2011): 3616–3623. [DOI] [PubMed] [Google Scholar]
- 25. Anderson J. C., Stein B., Kahi C. J., Rajapakse R., Walker G., and Alpern Z., “Association of Smoking and Flat Adenomas: Results From an Asymptomatic Population Screened With a High‐Definition Colonoscope,” Gastrointestinal Endoscopy 71 (2010): 1234–1240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Anderson J. C., Attam R., Alpern Z., et al., “Prevalence of Colorectal Neoplasia in Smokers,” American Journal of Gastroenterology 98 (2003): 2777–2783. [DOI] [PubMed] [Google Scholar]
- 27. Anderson J. C., Latreille M., Messina C., et al., “Smokers as a High‐Risk Group: Data From a Screening Population,” Journal of Clinical Gastroenterology 43 (2009): 747–752. [DOI] [PubMed] [Google Scholar]
- 28. Anderson J. C., Pleau D. C., Rajan T. V., et al., “Increased Frequency of Serrated Aberrant Crypt Foci Among Smokers,” American Journal of Gastroenterology 105 (2010): 1648–1654. [DOI] [PubMed] [Google Scholar]
- 29. Anderson J. C., Weiss J. E., Robinson C. M., and Butterly L. F., “Adenoma Detection Rates for Screening Colonoscopies in Smokers and Obese Adults: Data From the New Hampshire Colonoscopy Registry,” Journal of Clinical Gastroenterology 51 (2017): e95–e100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Rice K., Sharma K., Li C., Butterly L., Gersten J., and DeGroff A., “Cost‐Effectiveness of a Patient Navigation Intervention to Increase Colonoscopy Screening Among Low‐Income Adults in New Hampshire,” Cancer 125 (2019): 601–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Johnson D. H., Kisiel J. B., Burger K. N., et al., “Multitarget Stool DNA Test: Clinical Performance and Impact on Yield and Quality of Colonoscopy for Colorectal Cancer Screening,” Gastrointestinal Endoscopy 85 (2017): 657–665.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Rex D. K., Anderson J. C., Butterly L. F., et al., “Quality Indicators for Colonoscopy,” American Journal of Gastroenterology 119 (2024): 1754–1780. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Contingency table of adenomatous and serrated polyps, by study cohort.
Table S2A: Frequencies and proportions of polyp outcomes with 95% CIs for proportions, by study cohort, modified definition of serrated polyps (SSPs and TSAs only).
Table S2B: Logistic regression of odds of synchronous polyps versus any other outcomes, modified definition of serrated polyps (SSPs and TSAs only).
Table S3: Unadjusted and adjusted multinomial logistic regression results describing the relative risk of having outcomes in the comparison versus base category, by study cohort, modified definition of serrated polyps (SSPs and TSAs only).
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.