P0263 - Burden-to-Benefit Ratios Differ by Adenoma Size: Results From the CRC-MAPS(TM) Model

Girish Putcha, MD, PhD¹, Lauren N. Carroll, MS, MPH¹, Signe Fransen, MSc², Ben Wilson, PhD¹, Andrew Piscitello, MAT³, Tarun Chandra, PhD^3
1Freenome, South San Francisco, CA; ²Freenome Inc, South San Francisco, CA; ³EmpiriQA, LLC, Long Grove, IL

Introduction: While the benefits of colorectal cancer (CRC) screening are widely accepted, the benefit of detecting small and medium adenomas can be controversial. Adenoma size and histology significantly affect surveillance intervals recommended by clinical guidelines due to their impact on CRC risk. However, removing all adenomas also carries risks. We previously showed that, of all test performance metrics, changes in adenoma sensitivity had the largest impact on CRC incidence reduction (IR) and mortality reduction (MR), with increasing benefits for decreasing adenoma size. Here, we examine the tradeoffs of test burden and benefits for a hypothetical blood-based test.

Methods: A semi-Markov microsimulation of the adenoma-carcinoma pathway was developed and calibrated to autopsy, SEER, and CT colonography data. The model demonstrated good internal validity, and the model's cumulative lifetime outcomes were consistent with existing validated models. This study simulated perfect adherence to a hypothetical annual, blood-based CRC screening test among previously unscreened individuals free of diagnosed CRC. Outcomes were aggregated from age 40 to death, and individuals were screened from age 45 to 75. The base case assumed size-specific adenoma sensitivities (1-5mm: 15%; 6-9mm: 20%; ≥10mm: 30%), 90% CRC sensitivity, and 90% specificity. Primary outcome measures were the number of colonoscopies (COL), and CRC IR and MR, compared to no screening. The number of additional COL per percentage change in CRC IR and MR (burden-to-benefit ratio) was calculated for changes (+/-5%) to each performance metric.

Results: The base case resulted in 2,824 COL per 1,000 individuals as well as an 83.4% CRC IR and an 87.9% CRC MR, compared to no screening. The burden-to-benefit ratios for CRC IR and MR were greatest for changes in 1-5mm adenoma sensitivity and much lower for 6-9mm and ≥10mm adenomas (Table). The burden-to-benefit ratios of CRC sensitivity were negligible. Changes to specificity had a large impact on the number of COL but only a moderate impact on CRC IR and MR, resulting in large negative burden-to-benefit ratios.

Discussion: This microsimulation study of a hypothetical blood-based CRC screening test provides insights into the different burden-to-benefit ratios for small, medium, and large adenomas. This work underscores meaningful tradeoffs with the detection and removal of different sizes of adenomas, with the greatest burden for small adenomas.

		CRC Incidence Reduction		CRC Mortality Reduction
Performance Metric (range)	Change in #COL	Change in CRC IR	Burden-to-Benefit	Change in CRC MR	Burden-to-Benefit
All adenoma size groups	287.3	5.7%	50.0	3.8%	76.0
1-5mm Sensitivity (10% to 20%)	238.8	2.6%	93.3	1.8%	133.1
6-9mm Sensitivity (15% to 25%)	34.5	2.0%	17.0	1.3%	26.9
≥10mm Sensitivity (25% to 35%)	12.3	1.0%	12.6	0.6%	21.5
CRC Sensitivity (85% to 95%)	< < 0.1	< -0.1%	-0.4	0.1%	< 0.1
Specificity (85% to 95%)	-623.1	0.8%	-764.8	0.8%	-779.9

Table: Table. Burden-to-benefit Ratios Across Performance Metrics


Disclosures:
Girish Putcha: Freenome – Employee.
Lauren Carroll: Freenome – Employee.
Signe Fransen: Freenome – Employee.
Ben Wilson: Freenome – Employee.
Andrew Piscitello: BlueStar Genomics – Consultant. Exact Sciences – Consultant. Freenome – Consultant. Viome, Inc. – Consultant.
Tarun Chandra: BlueStar Genomics – Consultant. Exact Sciences – Consultant. Freenome – Consultant. Viome, Inc. – Consultant.


Girish Putcha, MD, PhD¹, Lauren N. Carroll, MS, MPH¹, Signe Fransen, MSc², Ben Wilson, PhD¹, Andrew Piscitello, MAT³, Tarun Chandra, PhD³. P0263 - Burden-to-Benefit Ratios Differ by Adenoma Size: Results From the CRC-MAPS(TM) Model, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.