Periodic Reporting for period 1 - ACCEPT (Artificial Intelligence in Colonoscopy for Cancer Prevention)
Période du rapport: 2021-06-01 au 2023-05-31
Removal of precancerous polyps (so-called adenomas) during colonoscopy reduces colorectal cancer incidence and mortality. We recently developed artificial intelligence (AI) systems which optimise colonoscopy quality by aiming at increasing the detection of adenomas. However, it was unknown if this benefit of AI translated into improved cancer prevention. The overarching goal of the present project was to uncover the value of AI in colorectal cancer prevention. To achieve this goal, we conducted microsimulation modelling with use of most reliable real-world data and established an infrastructure and conducted a pilot for a large-scale randomized controlled trial with colorectal cancer incidence and mortality as an endpoint.
As a first project, we conducted a Markov model microsimulation of using colonoscopy with and without AI for colorectal cancer screening for individuals at average risk. We ran the microsimulation in a hypothetical cohort of 100 000 individuals in the USA aged 50–100 years. The primary analysis investigated screening colonoscopy with versus without AI every 10 years starting at age 50 years and finishing at age 80 years, with follow-up until age 100 years, assuming 60% screening population uptake. In secondary analyses, we modelled once-in-life screening colonoscopy at age 65 years in adults aged 50–79 years at average risk for colorectal cancer. The main outcome measures included the incremental effect of AI-assisted colonoscopy versus standard (no-AI) colonoscopy on colorectal cancer incidence and mortality, and cost-effectiveness of screening. In the primary analyses, compared with no screening, the relative reduction of colorectal cancer incidence with screening colonoscopy without AI tools was 44·2% and with screening colonoscopy with AI tools was 48·9% (4·8% incremental gain). Compared with no screening, the relative reduction in colorectal cancer mortality with screening colonoscopy with no AI was 48·7% and with screening colonoscopy with AI was 52·3% (3·6% incremental gain). AI detection tools decreased the discounted costs per screened individual from $3400 to $3343 (a saving of $57 per individual). Results were similar in the secondary analyses modelling once-in-life colonoscopy. Our findings suggested that implementation of AI detection tools in screening colonoscopy was a cost-saving strategy to further prevent colorectal cancer incidence and mortality. (Areia-Mori, et al. Lancet Digital Health 2022).
In line with this first project, we established a study infrastructure of a large-scale randomized trial with colorectal incidence and death as an endpoint. To generalize study results as much as possible we targeted only people subject to cancer screening programmes with a pragmatic study design (Mori, et al. Lancet Gastroenterol Hepatol 2022). The scale of the study seemed gigantic and ambitious given its endpoint; therefore, we were determined to do a relatively small-scale pilot to confirm the feasibility of the study before we went for a large-scale trial. This first phase of the trial was done in Poland including more than 1,000 participants from December 2021 after confirming approval of relevant ethics committees and establishing a pragmatic randomization design earlier in 2021. The inclusion and randomization of the patients went well with an interim analysis done in September 2022 in a blinded fashion. The interim analysis (Unpublished) supported continuation and expansion of the study to a long-term inclusion and follow-up phase.
In line with this first project, we established a study infrastructure of a large-scale randomized trial with colorectal incidence and death as an endpoint. To generalize study results as much as possible we targeted only people subject to cancer screening programmes with a pragmatic study design (Mori, et al. Lancet Gastroenterol Hepatol 2022). The scale of the study seemed gigantic and ambitious given its endpoint; therefore, we were determined to do a relatively small-scale pilot to confirm the feasibility of the study before we went for a large-scale trial. This first phase of the trial was done in Poland including more than 1,000 participants from December 2021 after confirming approval of relevant ethics committees and establishing a pragmatic randomization design earlier in 2021. The inclusion and randomization of the patients went well with an interim analysis done in September 2022 in a blinded fashion. The interim analysis (Unpublished) supported continuation and expansion of the study to a long-term inclusion and follow-up phase.
To our knowledge, no one has attempted to establish the benefit of AI for colorectal cancer prevention. This ambitious project is the first to rigorously test AI in screening for colorectal cancer, thus is of high value for patients and society. Furthermore, the project has two distinct impacts on health care. First, around 5% absolute reduction of colorectal cancer could be statistically demonstrated; prevention is a main focus of cancer control. Second, the study results (i.e. use of AI during colonoscopy) can be directly integrated into the next generation screening service because the present study is done with the screening population as an analytic subject. These results are also transferable to other countries in Europe and other regions by utilizing effective dissemination and communication channels. The results of the project has been thoroughly disseminated throughout the project period.