Society: AGA

Background and Aims: Although high body mass index (BMI) increased the risk of colorectal cancer (CRC), the impact of BMI change
on cancer risk has been rarely reported. We investigated the impact of BMI change on colorectal cancer risk.
Methods: Individuals, who underwent general health examination at 2009 and 2013 through the National Health Insurance Service System, were enrolled and followed up through 2017. Persons who had been diagnosed with any cancer until the second measurement of BMI were also excluded. BMI at baseline (2009) and follow-up (2013) were extracted and was classified according to
Asian Pacific guideline; underweight (<18.5 kg/m²), normal (18.5-22.9 kg/m²), overweight (23-24.9 kg/m²), obesity I (25-29.9 kg/m²),
and obesity II (≥30 kg/m²). To investigate cancer risk according to BMI change, proportional Cox regression analysis using hazard ratio (HR) and 95% confidence interval (CI) were conducted.
Results: Among 2.8 million persons, 12439 patients developed colorectal cancers during follow-up.
Persistent high BMI increased the risk of CRC. Adjusted HR is 1.11 (95% CI 1.06-1.18) for persistent obesity I and 1.24 (95% CI 1.08-1.41) for persistent obesity II comparing to persistent normal BMI group. Among baseline obese persons (obesity I), weight reduction decreased the risk of CRC comparing to persistent obesity I group (HR 0.79 in persons with change from obesity I to normal BMI and 0.92 in persons with change from obesity I to overweight group). Change from obesity I to obesity II increased the risk of CRC (HR 1.23; 95% CI 1.04-1.46) comparing to persistent obesity I group (Figure).
CONCLUSIONS: Persistent high BMI increased the risk of CRC. Furthermore, change of BMI dynamically reduced or increased the CRC
risk among baseline obese persons.

This work was supported by the National Research Foundation, republic of Korea (NRF-2022R1A2C2013044).

Background: Polygenic risk scores (PRS), which calculate genetic risk for colorectal cancer (CRC), may help prioritize individuals for more intensive versus less intensive CRC screening and surveillance. These tools have many potential benefits over current clinical risk stratification algorithms, but have yet to be widely validated. We investigated an existing PRS to determine associations with prevalent and incident advanced neoplasia (AN) in a population undergoing screening colonoscopy and follow up.

Methods: The CSP#380 screening colonoscopy cohort includes 10 years of clinical follow-up and a biorepository comprised of selected individuals with baseline AN (defined as CRC or adenoma ≥10mm or villous or high-grade dysplasia) and matched individuals without neoplasia. A PRS was constructed for each individual from 136 pre-specified CRC-risk single nucleotide polymorphisms. Multivariate logistic regression was used to evaluate the PRS for associations with AN prevalence at initial screening colonoscopy (n=594), or incident AN in participants with at least one follow-up colonoscopy (n=354). The PRS was analyzed categorically by quintiles (based on the distribution among controls) and as a continuous variable in stratified analyses.

Results: Participants with a PRS in the lowest quintile had more than a 70% decreased risk of AN at baseline screening colonoscopy (OR 0.29, 95% CI 0.14-0.58; p<0.001) compared to participants with a PRS in the middle quintile (Table 1). Further, only one CRC was identified in the lowest PRS quintile, compared to 19 cancers in the other PRS categories. Using the PRS positivity level of greater than 20% as high-risk, the sensitivity/specificity/NPV for AN at baseline colonoscopy is 91.8%/20.1%/85.6% and for CRC is 95.2%/17.1%/99.0%. In stratified analyses (Table 2), a one point increase in the PRS had the greatest impact on baseline AN risk in those with a family history of CRC (OR 12.6, 95% CI 1.20-15.50; p=0.04), Europeans (OR 7.80, 95% CI 2.30-27.52; p=0.001), and participants aged 65-75+ (OR 4.44, 95% CI 1.39-14.52; p=0.01). However, we did not observe a relationship between the PRS and incident AN during follow up (Table 1). Further, the PRS was not significantly associated with incident AN during follow-up in the stratified analyses by age, race, or baseline colonoscopy findings (Table 2).

Conclusions: Our results provide evidence which support the ability of a PRS to identify patients at risk for prevalent AN, particularly in those with a family history of CRC. Ongoing work will determine whether this PRS can identify a subset of individuals at sufficiently low risk across diverse populations who could safely delay or undergo less frequent or non-invasive screening. However, more research is needed to augment genetic tools to predict the likelihood of developing incident AN and CRC during long term follow up.

Background: Colonic polyposis of unknown etiology (CPUE) is defined as having ≥10 cumulative colonic adenomas without a detectable germline pathogenic variant. Colorectal and extra-colonic surveillance for those with 100 adenomas is recommended to be similar to familial adenomatous polyposis. The utility of extra-colonic screening in patients with 10- <100 adenomas is not well established.
Method: All patients with CPUE and negative germline testing for APC and MUTYH and/or multigene panel testing, who were seen at our center between 2003 and 2022 were included. Patients were categorized based on colonic polyp count into 3 groups: 10- 19, 20 - 99, and ≥100 adenomas. Baseline characteristics as well as extra-colonic manifestations including upper endoscopy and thyroid imaging if performed were collected. Categorical data were analyzed using Chi Square and Fisher Exact tests, while continuous data via ANOVA testing.
Results: 154 patients were identified of which 20 (13.0%) had 10-19 cumulative adenomas, 83 (53.9%) had 20-99 adenomas and 51(33.1%) had >100 adenomas. Compared to patients with 10-19 and 20-99 adenomas, those with >100 adenomas were younger (median age 51 vs 52 vs 42 years, respectively, p<0.001). Sex, race, body mass index, social and family history were similar among the 3 groups (Table 1). Of the patients who underwent upper endoscopy, duodenal adenomas were found in 5 (41.7%), 18 (25.7%) and 21 (45.7%) of patients with 10-19, 20-99 and >100 adenoma respectively, p=0.065. Ampullary adenomas, in particular, were significantly more common in the >100 adenoma group 15.2%, p= 0.017. Although fundic gland polyps were more often seen in patients with > 100 adenomas compared to the 10-19 and 20-99 adenoma groups (56.5% vs. 32.4% vs 41.7% respectively, p=0.035), gastric adenomas were detected in all 3 groups (p =0.166). Of those who underwent thyroid ultrasound, thyroid nodules >1 cm were not detected in patients with 10-19 adenomas but found in 8 (23.5%) and 5 (14.7%) of patients with 20-99 and >100 adenomas respectively (p=0.037) (Table 2). Thyroid cancer was diagnosed in one patient with 20-99 adenomas.
Conclusion: In our cohort, duodenal and gastric adenomas occurred similarly in CPUE patients with adenoma count 10 to >100 at a relatively high proportion. We therefore advocate for a baseline upper endoscopic exam in all patients with CPUE. While clinically significant thyroid nodules were not detected in patients with 10-19 adenomas, they occurred in about the fifth of the patients with >20 adenomas, indicating that a thyroid ultrasound is prudent in these patients.

Table 1: Baseline demographics in patients with CPUE. (* = p-value ≤ 0.05).

Table 1: Baseline demographics in patients with CPUE. (* = p-value ≤ 0.05).

Table 2: Extra-colonic manifestations in patients CPUE. (* = p-value ≤ 0.05).

Table 2: Extra-colonic manifestations in patients CPUE. (* = p-value ≤ 0.05).

Background: There is limited data on the safety of endoscopic surveillance in patients with MUTYH-Associated Polyposis (MAP) with some studies suggesting accelerated adenoma-carcinoma progression. We hypothesized that endoscopic management of polyposis and close surveillance by experts is safe and can help in avoiding or deferring prophylactic colectomy without an increased risk of death from cancer.

Methods: MAP patients were identified through our institutional review board (IRB) approved inherited colorectal cancer registry. We included all patients with confirmed biallelic MUTYH pathogenic variants (PVs) who have more than one lower endoscopy documented at our institution over at least 2 years, including patients with and without prior segmental colectomy. The primary outcome assessed was the development of cancer while under surveillance. Secondary outcome of interest was the need for surgery while under surveillance.

Results: 68 patients enrolled in our registry were identified with biallelic MUTYH PVs. We excluded 33 patients who were not undergoing surveillance at our institution. A total of 35 patients underwent endoscopic colorectal cancer surveillance at our institution following their diagnosis of MAP with a total of 363 lower endoscopies performed. The median interval between endoscopies was 30.7 months (IQR 19.3-37.1). The median endoscopic follow up was 10 years (IQR=7-14) with a median of 9 (IQR=7-14) lower endoscopies per patient. 9/35 (44%) patients had a diagnosis of colorectal cancer. Of these patients, colorectal cancer was diagnosed prior to the start of surveillance in 7/9 (78%). 2 (6%) patients were diagnosed with colorectal cancer while under surveillance, neither of whom had previous colorectal surgery. Both cancers were stage 1 at resection, and the patients remained cancer free at last follow up, 8 and 13 years after resection. 18/35 (51%) patients never required colorectal resection. Of the 17 patients who required surgery, 9 had undergone segmental resection prior to being diagnosed with MAP. Of these 4/9 (44%) required subsequent operation but no patients developed metachronous cancer. In total 12/38 (32%) patients underwent additional surgery while under surveillance. One surgery was a colonic interposition, the remainder were for endoscopically unmanageable polyps.

Conclusion: Cancer in MAP patients is usually diagnosed on presentation. The vast majority of patients undergoing careful surveillance will not develop cancer although approximately one third will require additional surgery. For patients who present without cancer, or after previous segmental resection, careful endoscopic surveillance with endoscopic polyp clearance is a reasonable alternative to colectomy.

Study Flowchart

Background: Current guidelines for screening all newly diagnosed individuals with colorectal cancer (CRC) for Lynch syndrome (LS) recommend tumor testing for microsatellite instability (MSI) or immunohistochemistry (IHC) for the protein expression of LS-associated mismatch repair genes. Use of alternative strategies including up-front germline testing and/or tumor sequencing to identify LS have been debated. The cost-effectiveness of various strategies to optimally identify LS in CRC patients, and the impact of cascade testing in at-risk family members of newly identified carriers, have not been explored. We assessed cost-effectiveness of multiple screening strategies to identify LS and the number of cases identified per strategy, including the cost of each newly identified carrier.

Methods: We developed a decision analytic model of the natural history of LS related CRC to evaluate eight screening strategies; the model cohort started at 25 years and cycled annually until age 75 (end of screening) or death using aggregated data regarding lifetime risk and life expectancy for women and men. The strategies included upfront testing by: (1) Germline (GL); (2) IHC, then GL (IHCGL); (3) MSI, then GL (MSIGL); (4) Tumor sequencing + GL (TSGL); (5) MSI; (6) MSI w/ Double Somatic (MSIDS); (7) IHC, then GL w/ Double Somatic (IHCGL);
(8) MSI, then GL w/ Double Somatic (MSIGLDS).

Cascade testing of at-risk relatives was added to each strategy; average number of first and second-degree relatives was eight with 52% adherence, as per prior publications. The primary endpoint was to determine the optimal strategy to identify LS in CRC patients and relatives, yielding the highest QALYs with an incremental cost-effectiveness ratio (ICER) below a willingness-to-pay threshold (WTP) of $100,000. Unadjusted life years gained was also evaluated.

Results: GL was the cost-effective strategy, at a cost of $61,357 per QALY when compared to MSI which yielded fewer QALYs. Additionally, GL had the lowest proband screening cost per LS identified at $3,778 per LS carrier. MSI was also included on the efficiency frontier with an ICER of $6,644 compared to MSIGL where all other strategies were dominated and excluded. GL and TSGL identified the most LS carriers among CRC patients and relatives, with 6,875 and 6,832 respectively.

Conclusion: Up-front universal GL testing is cost-effective compared to current recommendations of universal MSI and/or IHC testing among CRC patients and provides the lowest cost per LS carrier identified, including at-risk relatives. GL testing and TSGL identify a near comparable number of CRC patients with LS. Both approaches also identify the most relatives with LS, thereby providing an opportunity to prevent CRC and other LS related cancers. The potential benefit of precision treatment that may be related to TSGL in patients with CRC was not assessed.

COST EFFECTIVE ANALYSIS OF SCREENING STRATEGIES FOR THE IDENTIFICATION OF LYNCH SYNDROME IN CRC PATIENTS AND RELATIVES

COST EFFECTIVE ANALYSIS OF SCREENING STRATEGIES FOR THE IDENTIFICATION OF LYNCH SYNDROME IN CRC PATIENTS AND RELATIVES

Background: Uptake of genetic testing for cancer susceptibility in family members of cancer patients is suboptimal with cascade testing rates of approximately 30%. Up to 10% of patients with pancreatic ductal adenocarcinoma (PDAC) have a pathogenic germline variant (PGV) in a cancer susceptibility gene. Given the poor prognosis of PDAC, offering timely genetic testing is recommended for all patients and their first-degree relatives. However, uptake remains low and at-risk relatives encounter many barriers to undergoing genetic testing. The GENetic Education, Risk Assessment, and TEsting (GENERATE) study evaluated two strategies of remote genetic education and testing in relatives of pancreatic ductal adenocarcinoma (PDAC) patients.

Methods: Recruitment occurred via collaborating study sites, social media, and PDAC advocacy organizations. Enrollment was via the study website (https://generatestudy.org/). Eligible participants had; a first-degree relative with PDAC or a known PGV in one of thirteen PDAC predisposition genes (APC, ATM, BRCA1, BRCA2, CDKN2A, EPCAM, MLH1, MSH2, MSH6, PALB2, PMS2, STK11, or TP53) and a first- or second-degree relative with PDAC. Participants were cluster-randomized by family into one of two arms. All study procedures were done remotely. Arm 1 included an interactive telemedicine session with a genetic counselor, followed by genetic testing with an at-home saliva kit that was returned via mail to a commercial laboratory. Arm 2 involved remote online genetic education and testing without the interactive session. The primary outcome was genetic testing uptake, which was compared by permutation tests and mixed-effects logistic regression models.

Results: Between 5/8/2019 and 6/1/2021, 424 families were randomized, including 601 participants (n=296 Arm 1; n=305 Arm 2). The uptake of genetic testing was 87% (257/296) in Arm 1 and 93% (284/305) in Arm 2 (p=0.014) (Table 1). Participants in Arm 1 were significantly less likely to obtain genetic testing compared to Arm 2 (Adjusted ratio [Arm1/Arm2] 0.90, 95% confidence interval 0.78-0.98) (Table 1). There were no statistically significant differences in the uptake of genetic testing based on age, sex, or socioeconomic status, but there was a significant difference across ethnicity, with uptake of genetic testing being lower among Hispanic compared to non-Hispanic participants (Adjusted ratio 0.73 (95% confidence interval 0.36-0.99) (Table 2). BRCA2, ATM, CDKN2A and PALB2 were the most common PDAC susceptibility genes in which PGVs were identified.

Conclusions: Remote methods of genetic education and testing are successful alternatives to overcome barriers to traditional in-person genetic counseling and testing for PDAC. These remote models should be considered a new standard-of-care option for delivery of genetics care for PDAC and leveraged for use across other hereditary cancers.

Table 1. Characteristics and Uptake of Genetic Testing Among Randomized Participants<b> </b>

Table 1. Characteristics and Uptake of Genetic Testing Among Randomized Participants

Table 2. Adjusted Ratio of Uptake Rate of Genetic Testing Among Randomized Study Participants

Table 2. Adjusted Ratio of Uptake Rate of Genetic Testing Among Randomized Study Participants

Michael Goggins

The Johns Hopkins University

Appearances