P0330 - Comparing Yield of Colonoscopy After Abnormal Multi-Target Stool DNA Test Compared to Fecal Immunochemical Test for Average-Risk Patients Undergoing Screening

Sunday, October 22, 2023

3:30 PM – 7:00 PM PT

Location: Exhibit Hall

Has Audio

Presenting Author(s)

Mudassar Sandozi, DO

MercyHealth
Rockford, IL

Mudassar Sandozi, DO¹, Justin Clark, ², Benjamin Bauer, BS³, Andrew Geiser, BS³, Kyle Rudser, PhD⁴, Irfan K. Sandozi, MD, MS⁵, Aasma Shaukat, MD, MPH⁶
¹MercyHealth, Rockford, IL; ²University of Minnesota School of Public Health, Minneapolis, MN; ³MNGI Digestive Health, Minneapolis, MN; ⁴University of Minnesota, Minneapolis, MN; ⁵Minnesota Gastroenterology, New Brighton, MN; ⁶New York University Grossman School of Medicine, New York, NY

Introduction: Over the last 3 decades, Colorectal Cancer (CRC) related incidence and mortality has declined 1.7% and 3.2% per year, respectively. The decline has been attributed to early detection through CRC screening and the removal of precancerous polyps via colonoscopy. Utilization of non-invasive, stool-based CRC screening modalities, with follow up colonoscopy when indicated, have been utilized to increase participation in screening. Results from implementation of these tests remains an area of active research. This retrospective review of patient encounters compared findings of colonoscopies conducted as routine screening vs those conducted after a positive fecal immunochemical test (FIT) or multitarget stool DNA test (mts-DNA).

Methods: Patients 40 years and older presenting to MNGI Digestive Health between January 1, 2017 - December 31, 2021 for a colonoscopy were studied. Three groups were compared: those presenting for routine screening, after an abnormal FIT, or after an abnormal mts-DNA. Pathology reports describing adenoma, advanced adenoma, sessile serrated adenoma or cancer were deemed positive for neoplastic lesions. A logistic regression model adjusted for age, sex, ethnicity, and race determined the relative odds of finding neoplastic lesions by comparing screening colonoscopy patients to the positive FIT and mts-DNA patients.

Results: Out of 127,986 patient encounters 47.81% of patients presenting for routine screening colonoscopies were found to have a neoplastic lesion as compared to 65.39% and 68.94% of patients presenting after a positive FIT or mts-DNA, respectively. Adjusted for age, sex, ethnicity, and race, the odds of finding neoplastic lesions was 1.75 times higher [1.39-2.21; 95% CI] in colonoscopies indicated by abnormal FIT and 2.13 times higher [1.93-2.34; 95% CI] in colonoscopies indicated by abnormal mts-DNA than routine screening colonoscopies.

Discussion: The odds of finding neoplastic lesions were highest in the mts-DNA group, and higher in the FIT group than the screening colonoscopy group. The variation in odds were likely a function of both differences in the underlying prevalence of neoplasms between the groups and the sensitivities of the screening modalities. Future work should investigate the relative contributions of these two factors on the diagnostic yield of stool-based screening.

Figure: Bar graph displaying proportion of encounters in each screening group with documented neoplastic lesion (adenoma, advanced adenoma, sessile serrated adenoma or cancer) vs no neoplastic lesion.

Colonoscopy Type	Routine Screening	Post(+) mts-DNA (Cologuard)	Post(+) FIT	Total
Number of Encounters	125,075	2,466	445	127,986
Age
Mean Age (SD)	57.65 (8.79)	64.35 (8.32)	62.53 (8.87)
Sex
Female	66,401 (53%)	1,471 (60%)	251 (56%)
Male	58,649 (47%)	994 (40%)	194 (44%)
Other	2 (< 0.1%)	0 (0%)	0 (0%)
Unspecified	23 (< 0.1%)	1 (< 0.1%)	0 (0%)
Race
American Indian Or Alaska Native	411 (0.3%)	5 (0.2%)	2 (0.5%)
Asian	4,369 (3.6%)	51 (2.2%)	20 (5.0%)
Black Or African American	4,698 (3.9%)	40 (1.7%)	18 (4.5%)
Declined to specify	1,139 (0.9%)	20 (0.9%)	8 (2.0%)
Native Hawaiian Or Other Pacific Islander	111 (< 0.1%)	2 (< 0.1%)	1 (0.2%)
Patient Wishes Not To Disclose	1,515 (1.3%)	26 (1.1%)	7 (1.7%)
Unknown	1,529 (1.3%)	18 (0.8%)	8 (2.0%)
White	107,285 (89%)	2,176 (93%)	339 (84%)
NA	4,018	128	42
Ethnicity
Declined to specify	1,852 (1.7%)	55 (2.8%)	5 (1.6%)
Hispanic or Latino	1,892 (1.8%)	18 (0.9%)	8 (2.6%)
Not Hispanic or Latino	87,336 (82%)	1,570 (79%)	250 (80%)
Other	264 (0.2%)	6 (0.3%)	1 (0.3%)
Patient Wishes Not To Disclose	3,369 (3.1%)	73 (3.7%)	14 (4.5%)
Unknown	7,974 (7.4%)	214 (11%)	26 (8.3%)
Unknown / Not Reported	4,469 (4.2%)	48 (2.4%)	8 (2.6%)
NA	17,919	482	133
Pathology Result
Adenoma	38,343 (31%)	981 (40%)	207 (47%)
Sessile Serrated	19,360 (15%)	715 (29%)	75 (17%)
Advanced Adenoma	17,001 (14%)	679 (28%)	133 (30%)
Cancer	59 (< 0.1%)	4 (0.2%)	1 (0.2%)
Other	41,591 (33%)	1,092 (44%)	185 (42%)
Had Neoplastic Lesion	59,804 (48%)	1,700 (69%)	291 (65%)
No Path Report	41,983 (34%)	373 (15%)	75 (17%)
Presence of Neoplastic Lesion
TRUE	59,804 (47.81%)	1,700 (68.94%)	291 (65.39%)	61,795 (48.28%)
FALSE	65,271 (52.19%)	766 (31.06%)	154 (34.61%)	66,191 (51.72%)
Relative Odds of Neoplastic Lesion
Unadjusted Odds Ratio	—	2.42	2.06
95% Confidence Interval	—	[2.22 - 2.64]	[1.70 - 2.51]	p-value < 0.001
Odds Ratio adjusted for age, sex, ethnicity, and race	—	2.13	1.75
95% Confidence Interval	—	[1.93 - 2.34]	[1.39 - 2.21]	p-value < 0.001

Table: Table summarizing findings of colonoscopy divided by screening modality (i.e. as routine screening colonoscopy, or surveillance post positive multitarget stool DNA test, or post fecal immunochemical test). Age reported as mean with standard deviation parenthesized. Sex, race, ethnicity, and pathology results categories summarize number and percentages of encounters reporting specified findings. Neoplastic lesion categories describe findings of adenoma, advanced adenoma, sessile serrated adenoma or cancer. Odds ratios compared findings of neoplastic lesion in FIT or mts-DNA compared with routine screening colonoscopy. Adjusted odds ratios were determined by logistic regression adjusted for age, sex, ethnicity, and race.
mts-DNA - multitarget stool DNA test (commercialized as Cologuard), FIT - fecal immunochemical test (not standardized to one commercial test), SD - standard deviation within 95% confidence interval

Disclosures:

Mudassar Sandozi indicated no relevant financial relationships.

Justin Clark indicated no relevant financial relationships.

Benjamin Bauer indicated no relevant financial relationships.

Andrew Geiser indicated no relevant financial relationships.

Kyle Rudser indicated no relevant financial relationships.

Irfan Sandozi indicated no relevant financial relationships.

Aasma Shaukat: Freenome Inc – Consultant. Iterative Health – Grant/Research Support. Medtronic – Consultant.

Mudassar Sandozi, DO¹, Justin Clark, ², Benjamin Bauer, BS³, Andrew Geiser, BS³, Kyle Rudser, PhD⁴, Irfan K. Sandozi, MD, MS⁵, Aasma Shaukat, MD, MPH⁶. P0330 - Comparing Yield of Colonoscopy After Abnormal Multi-Target Stool DNA Test Compared to Fecal Immunochemical Test for Average-Risk Patients Undergoing Screening, ACG 2023 Annual Scientific Meeting Abstracts. Vancouver, BC, Canada: American College of Gastroenterology.