P1318 - Multitarget Stool DNA Testing Has a High Positive Predictive Value for Colorectal Neoplasia on the Second Round of Testing

Monday, October 25, 2021

10:30 AM – 4:15 PM PT

Location: Shoreline Exhibit Hall

Has Audio

Award: Presidential Poster Award

Presenting Author(s)

Jordan K. Voss, MD, MS

Mayo Clinic
Rochester, MN, United States

Award: Presidential Poster Award

Jordan K. Voss, MD, MS¹, Derek Ebner, MD¹, Douglas Mahoney, MS¹, Kelli N. Burger, BS¹, John B. Kisiel, MD²
¹Mayo Clinic, Rochester, MN; ²Mayo Clinic, Rocheter, MN

Introduction: Multitarget stool DNA (mt-sDNA) screening for colorectal cancer is recommended every 1-3 years by the United States Preventative Services Task Force (USPSTF) and covered every 3 years by Medicare. This recommendation is based on modeling and 1^st round cross-sectional test performance. Real-world longitudinal performance data are lacking. We aimed to measure the test positive rate and the positive predictive value (PPV) of a 2nd mt-sDNA test to inform programmatic screening effectiveness estimates.

Methods: This was a multi-site, retrospective cohort study of average risk patients who had ≥2 rounds of mt-sDNA screening from 2014-2019. Those at increased risk or with symptoms were excluded. Patients with a negative 1st mt-sDNA and a completed 2^nd mt-sDNA test underwent individual chart review. Findings at post-mt-sDNA diagnostic colonoscopy (polyp number, location, and histology) were recorded. The test-positivity rate and PPV for endpoints of colorectal cancer (CRC), advanced colorectal neoplasia (ACRN), and any colorectal neoplasia (CRN) by site and sessile serrated polyp (SSP) sub-types were recorded for the 2^nd mt-sDNA. Nominal comparison was made to a historical cohort of patients having 1^st mt-sDNA testing using Fisher’s exact test.

Results: 2113 patients met study inclusion criteria. Median age was 64 years (IQR 56-69). 96% of patients were white. 37% were male. Of the 2113 patients, 319 (15%) had a positive 2nd mt-sDNA. Median time to 2^nd test was 3.2 years (IQR 3.0,3.4). Among those patients with a positive test, 300 (94%) underwent diagnostic colonoscopy, the reports from which were available for 298. The PPV of 2^nd mt-sDNA was 1 (0.34%) for CRC, 71 (24%) for ACRN, 201 (67%) for any CRN (Table). Of those with any CRN, ≥1 lesion was right-sided in 169/201 (84%) and ≥1 SSP was found in 78/201 (39%). After 2^nd mt-sDNA, only 4.6% (97/2113) patients screened completed a colonoscopy with non-neoplastic findings.

Discussion: Mt-sDNA test positivity rate was similar to that reported previously for 1^st round screening and has a high PPV for both any CRN (67%) and advanced CRN (24%) on the 2^nd round of testing. PPV for CRC may have been lowered by reduction in prevalence by negative 1^st round screen. These findings are consistent with the previously reported PPV of mt-sDNA for the detection of advanced neoplasia on 1^st round testing (Table). These observations provide essential confirmation of mt-sDNA test performance which is of importance to estimating programmatic test effectiveness.

	1st round	2nd round	P value
	Eckmann et.al.^b	Current study
N, average risk mt-sDNA positive patients	1,558	300^c
Positive predictive values, % (95% CI)
Any CRN	67% (65-69%)	67% (62-73%)	0.95
CRC	1% (0.5-2%)	0.34% (0.008-2%)	0.49
Advanced CRN^a	28% (26-31%)	24% (19-29%)	0.12
Non-advanced CRN	39% (36-41%)	44% (38-49%)	0.12
Right-sided CRN, %	53% (51-56%)	57% (51-62%)	0.23
Sessile serrated polyps (any), %	28% (26-31%)	26% (21-32%)	0.62

Table: Comparison of 1st and 2nd round mt-sDNA testing outcomes.
a) Advanced CRN includes cancers and polyps >1cm or with villous features on histology.
b) Eckmann, J. D. et al. Am J Gastroenterol 115, 608-615 (2020).
c) Colonoscopy reports were missing for 2/300 patients.

Disclosures:

Jordan Voss indicated no relevant financial relationships.

Derek Ebner indicated no relevant financial relationships.

Douglas Mahoney: Exact Sciences – Patent Holder, Mr. Mahoney is listed as an inventor on intellectual property jointly owned by Mayo Clinic and Exact Sciences and may receive royalties under Mayo Clinic policy..

Kelli Burger: Exact Sciences – Other Financial or Material Support, Contractual research services paid to my employer, Mayo Clinic.

John Kisiel: Exact Sciences – Consultant, Grant/Research Support, Patent Holder, Other Financial or Material Support, royalties, paid to Mayo Clinic.

Jordan K. Voss, MD, MS¹, Derek Ebner, MD¹, Douglas Mahoney, MS¹, Kelli N. Burger, BS¹, John B. Kisiel, MD². P1318 - Multitarget Stool DNA Testing Has a High Positive Predictive Value for Colorectal Neoplasia on the Second Round of Testing, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.