Influence of Colorectal Cancer Risk Factors on Predictive Value of a Positive Multitarget Stool DNA Test

GOALS

We analyzed if the predictive value of multitarget stool-based DNA (mt-sDNA) varied when patients had pre-existing known colorectal cancer (CRC) risk factors.

BACKGROUND

mt-sDNA test is approved for CRC screening in average-risk patients. Whether patients with a personal history of adenomatous colon polyps or a family history of CRC (CRC risk factors) benefit from mt-sDNA testing is unknown.

STUDY

We reviewed charts for all positive mt-sDNA referrals between 2017 and 2021. Diagnostic colonoscopy adherence rates were calculated. In those who had colonoscopy, we compared detection rates for any colorectal neoplasia (CRN), multiple (3 or more) adenomas, sessile serrated polyps (SSP), advanced CRN, and CRC between patients with and without known CRC risk factors.

RESULTS

Of 1297 referrals for positive mt-sDNA, 1176 (91%) completed a diagnostic colonoscopy. The absence of neoplasia was noted in 27% of colonoscopy exams. When neoplasia was identified, findings were as follows: any CRN (73%), multiple adenomas (34%), SSP (23%), advanced CRN (33%), and CRC (2.5%). One or more CRC risk factors were present in 229 (19%) of cases. In the CRC risk factor subgroup, patients having a prior history of adenomatous polyps or a family history of CRC were no more likely to have any CRN, multiple adenomas, SSP, advanced CRN, or CRC compared to average-risk patients when mt-sDNA was positive.

CONCLUSIONS

In this real-world analysis of positive mt-sDNA referrals, adherence to subsequent diagnostic colonoscopy recommendations was high. The presence of pre-existing CRC risk factors did not affect the positive predictive value of mt-sDNA.

Utility of esophageal biopsy in the diagnosis of nonerosive reflux disease

This study evaluated the accuracy of esophageal biopsy for the diagnosis of nonerosive reflux disease (NERD) in adults. Thirty-five patients with reflux symptoms and a normal endoscopy were prospectively evaluated using esophageal biopsies, 24-h ambulatory pH monitoring and symptomatic response 4 weeks after an increase in antireflux therapy. Biopsies were scored for the total number of typical histologic reflux features seen. Patients were considered to have NERD if both pH-metry was positive and step-up therapy was successful. Biopsies were then compared to this predefined gold standard. Biopsy was most sensitive (62%) but poorly specific (27%) when one or more histologic reflux features were seen. A threshold of three or more histologic features improved the specificity (91%) but reduced sensitivity (31%). Response to step-up therapy was associated with 100% sensitivity and 100% negative predictive value when compared to biopsy and pH-metry as an alternate combined gold standard. In conclusion, biopsy is insensitive in diagnosing NERD but reasonably specific if three or more typical histologic reflux features are present.