CRC-AIM

The Colorectal Cancer-Adenoma Incidence and Mortality (CRC-AIM) model was developed to address limitations of earlier colorectal cancer screening modeling analyses. Existing CRC models did not adequately adjust for things like real-world adherence to stool-based screening strategies and follow-up, adenoma sensitivity, or lifetime patient screening burden.1 Explore the content below to learn more about the development and impact of the CRC-AIM model.

Ideally, randomized clinical trials (RCTs) would evaluate the comparative effectiveness of different CRC screening strategies

Unfortunately, conducting RCTs for all the various CRC screening strategies is challenging, so microsimulation models are used instead2,3

CRC-AIM compares the relative screening burden between alternative screening modalities1

The CRC-AIM model is a microsimulation model based on the Colorectal Cancer-Simulated Population model for Incidence and Natural history (CRC-SPIN) and includes natural history and CRC screening components.1 CRC-AIM has been cross-validated against Cancer Intervention and Surveillance Modeling Network (CISNET) models.1 Predicted outcomes were simulated for 4 million individuals born in 1975 who were undiagnosed with CRC at age 40 and screened between ages 45-75 years or 50-75 years.1

At real-world adherence rates, multitarget stool DNA (mt-sDNA) resulted in an 8.4% to 19.1% increase in the number of life-years gained (LYG) compared with fecal immunochemical test (FIT) for screening ages 50-75 years, with similar results for ages 45-75 years1

At equivalent numbers of completed tests over the screening window, mt-sDNA resulted in greater LYG and reductions in CRC-related mortality and incidence compared with FIT4

With a 12, 18, and 24-month screening delay, reduction in CRC-related incidence and mortality and increase in LYG was greater with triennial mt-sDNA than annual FIT5

Estimating the Impact of Imperfect Adherence to Stool-Based Colorectal Cancer Screening Strategies on Comparative Effectiveness Using the CRC-AIM Model1,a

  • At real-world adherence rates of ~70%b for mt-sDNA6 and 40-48% for FIT,1,6,7 mt-sDNA resulted in an 8.4% to 19.1% increase in the number of LYG compared with FIT for screening ages 50-75 years. Similar results were observed for screening ages 45-75 years
  • At imperfect adherence rates in individuals screened between ages 50-75 or 45-75, the reduction in CRC-related incidence and mortality was higher for mt-sDNA than FIT (at 70% vs 40-48% real-world adherence rates, respectively)1,7
  • At equivalent adherence rates, mt-sDNA resulted in a higher number of total required colonoscopies and a lower number of tests per 1000 individuals compared with FIT1

Medicare: Real-World Screening + Follow-Up

Figure shows a cluster of five bar graphs that are comparing key measures of health outcomes across three colorectal cancer screening tests: the mt-sDNA test, the fecal occult blood test, and the fecal immunochemical test. These five key measures are CRC cases per one thousand, CRC deaths per one thousand, life-years gained per one thousand, mortality reduction percentage, and CRC incidence reduction percentage. A general trend emerges from these graphs showing the mt-sDNA test health outcomes results in much lower CRC cases and deaths per one thousand. The mt-sDNA test also shows substantial increases in the areas of life-years gained per one thousand, mortality reduction, and CRC incidence reduction when compared to the fecal occult blood test and the fecal immunochemical test.
Figure shows a cluster of four bar graphs comparing the costs associated with four types of screening choices for colorectal cancer: the mt-sDNA test, the fecal occult blood test, the fecal immunochemical test, and no test. The costs being compared are screening costs, colonoscopy complications costs, and CRC management costs. In descending order, the total combined costs for each are $6,525 for the mt-sDNA test, $6,013 for no test, $5,801 for the fecal occult blood test, and $5,797 for the fecal immunochemical test. The overall trend is while the mt-sDNA test generally has higher screening costs, it has a lower overall CRC management cost compared to other screening choices.

More FIT Tests Were Needed to Match the Clinical Benefit of Equivalent Numbers of mt-sDNA Tests4

Predicted LYG & CRC Incidence/Mortality Reduction by Screening Strategy

Screening strategy (50-75) Randomly assigned number of tests CRC incidence reduction CRC mortality reduction
mt-sDNA Up to 1 16% 20%
FIT Up to 1 8% 11%
mt-sDNA Up to 5 49% 57%
FIT Up to 5 31% 39%
mt-sDNA Up to 9 63% 71%
FIT Up to 9 44% 54%

Table adapted from Saoud, 20204

Detecting SSPs in CRC-AIM – An important consideration

Sessile serrated polyps (SSPs) are a type of advanced precancerous lesion10

CRC-AIM recognizes the significance of SSPs and their impact on CRC incidence rates10

SSPs account for approximately 20-30% of all CRC11

  • SSPs are difficult to detect by stool-based FIT tests due to lack of bleeding compared to adenomas10
  • The mt-sDNA test works by detecting biomarkers from DNA, such as methylated genes, shed into the stool from colorectal cancers and precancerous lesions10
  • When detecting SSPs that are 10mm or greater, mt-sDNA tests shows superior sensitivity compared to FIT (42.4% vs 5.1%)12

Sensitivity and Specificity Model Inputs for mt-sDNA, FIT, and Colonoscopy in the Detection of Conventional Adenomas or SSPs

Sensitivity
Adenomas, mm
Screening modality Adenoma type Specificity Cancer <6c 6-9c ≥10d
mt-sDNA Conventional 89.8% 92.3% 17.2% 17.2% 42.4%
SSP 17.2% 17.2% 42.4%
Colonoscopya Conventional 86.0% 95.0%b 75.0% 85.0% 95.0%
SSP 65.0% 75.0% 85.0%
FIT Conventionale 96.4% 73.8% 8.0% 8.0% 26.6%
SSP 5.1% 5.1% 5.1%
aIt was assumed that the same sensitivity and specificity for screening colonoscopies applied to follow-up colonoscopies.
bBy assumption.
cSensitivity for mt-sDNA and FIT in persons with non-advanced adenomas.
dSensitivity for mt-sDNA and FIT in persons with advanced adenomas (i.e., adenomas>= 10mm or adenomas with advanced histology).
eSensitivity of FIT for conventional adenomas was derived from the study by Imperiale et al, 2014,12 after adjusting for the proportion of SSPs. 		Table adapted from Kisiel, 202210

Related Topics

Footnotes

  1. Assumptions: All CRC screening test performance assumptions (sensitivity, specificity, and complications) were intentionally unchanged from the CISNET modeling analyses used to inform recent guideline recommendations. CRC microsimulation models used to inform CRC screening guidelines assume perfect (100%) adherence with all CRC screening, follow-up, and surveillance procedures over each individual’s lifetime. Adherence was set by assuming a fixed annual likelihood to comply with each stool-based screening strategy ranging from 0% to 100%, in 10% increments. It was assumed that patients were offered a stool-based screening test yearly unless they were not due for screening. Real-world first-round differential adherence rates of 40% for annual fecal immunochemical test (FIT) and 70% for triennial multitarget stool DNA (mt-sDNA) were considered likely imperfect rates. Limitations: None reported.
  2. 71% mt-sDNA adherence rate in Medicare Beneficiaries cohort.

List of definitions

CISNET: Cancer Intervention and Surveillance Modeling Network; CRC: colorectal cancer; CRC-AIM: Colorectal Cancer-Adenoma Incidence and Mortality; CRC-SPIN: Colorectal Cancer-Simulated Population model for Incidence and Natural history; FIT: fecal immunochemical test; FOBT: fecal occult blood test; LYG: life-years gained; mt-sDNA: multitarget stool DNA; RCT: randomized clinical trial; SSP: sessile serrated polyps.


References

  1. Piscitello A, Saoud L, Fendrick AM, et al. Estimating the impact of differential adherence on the comparative effectiveness of stool-based colorectal cancer screening using the CRC-AIM microsimulation model. PLoS One. 2020;15(12):e0244431.
  2. Wolf AMD, Fontham ETH, Church TR, et al. Colorectal cancer screening for average-risk adults: 2018 guideline update from the American Cancer Society. CA Cancer J Clin. 2018;68(4):250-281.
  3. Knudsen AB, Zauber AG, Rutter CM, et al. Estimation of benefits, burden, and harms of colorectal cancer screening strategies: modeling study for the US Preventive Services Task Force. JAMA. 2016;315(23):2595-2609.
  4. Saoud L, Fendrick AM, Piscitello A, et al. More fecal immunochemical tests are needed to match the clinical benefit of equivalent numbers of multitarget stool DNA tests. Gastroenterol. 2020;158(6):S642-S643.
  5. Limburg P, Saoud L, Borah B, et al. Higher impact on clinical outcomes from delays in colorectal cancer screening with the fecal immunochemical test vs multitarget stool DNA: CRC-AIM microsimulation model results. Gastroenterol. 2020;158(6):S-1176.
  6. Weiser E, Parks PD, Swartz RK, et al. Cross-sectional adherence with the multi-target stool DNA test for colorectal cancer screening: real-world data from a large cohort of older adults. J Med Screen. 2021;28(1):18-24.
  7. Jensen CD, Corley DA, Quinn VP, et al. Fecal immunochemical test program performance over 4 rounds of annual screening: a retrospective cohort study. Ann Intern Med. 2016;164(7):456-463.
  8. Fisher DA, Karlitz JJ, Jeyakumar S, et al. Real-world cost-effectiveness of stool-based colorectal cancer screening in a Medicare population. J Med Econ. 2021;24(1):654-664.
  9. Data on File. Exact Sciences Corporation. Madison, WI.
  10. Kisiel JB, Itzkowitz SH, Ozbay AB, et al. Impact of the sessile serrated polyp pathway on predicted colorectal cancer outcomes. Gastro Hep Advances. 2022;1(1):55-62.
  11. Obuch JC, Pigott CM, Ahnen DJ. Sessile serrated polyps: detection, eradication, and prevention of the evil twin. Curr Treat Options Gastroenterol. 2015;13(1):156-170.
  12. Imperiale TF, Ransohoff DF, Itzkowitz SH, et al. Multitarget stool DNA testing for colorectal-cancer screening. N Engl J Med. 2014;370(14):1287-1297.