Advertisement

Merits, Challenges, and Limitations of Randomized Trials on Colorectal Cancer Screening Effectiveness

  • Hermann Brenner
    Correspondence
    Correspondence Address correspondence to: Hermann Brenner, Clinical Epidemiology and Aging Research Division Head, German Cancer Research Center (DKFZ), Im Neuenheimer Feld 581, Heidelberg, Germany.
    Affiliations
    Clinical Epidemiology and Aging Research Division, German Cancer Research Center, Heidelberg, Germany
    Search for articles by this author
  • Amanda J. Cross
    Affiliations
    Cancer Screening and Prevention Research Group, Department of Surgery and Cancer, Imperial College London, London, UK
    Search for articles by this author
Published:December 21, 2020DOI:https://doi.org/10.1053/j.gastro.2020.12.029
      See “Colorectal cancer screening with repeated fecal immunochemical test versus sigmoidoscopy: baseline results from a randomized trial,” by Randel KR, Schult AL, Botteri E, et al, on page 1085.
      Randomized controlled trials (RCTs) are commonly considered as providing the highest degree of evidence on the effectiveness of medical interventions as randomization of the intervention should prevent any bias from confounding or selective allocation of the intervention. In fact, RCTs on guaiac-based fecal occult blood tests (gFOBTs) and flexible sigmoidoscopy (FS) have provided compelling evidence of the effectiveness of these screening modalities in decreasing colorectal cancer (CRC) incidence (FS) and mortality (gFOBT, FS).
      • Shaukat A.
      • Mongin S.J.
      • Geisser M.S.
      • et al.
      Long-term mortality after screening for colorectal cancer.
      ,
      • Brenner H.
      • Stock C.
      • Hoffmeister M.
      Effect of screening sigmoidoscopy and screening colonoscopy on colorectal cancer incidence and mortality: systematic review and meta-analysis of randomised controlled trials and observational studies.
      For fecal immunochemical tests (FITs) and colonoscopy, two of the most widely recommended and applied screening options, no such evidence from RCTs is so far available for long-term outcomes, even though evidence from observational and modelling studies suggests them to be more effective and cost effective than gFOBT-based screening and similarly or even more effective and cost-effective than FS-based screening.
      • Brenner H.
      • Stock C.
      • Hoffmeister M.
      Effect of screening sigmoidoscopy and screening colonoscopy on colorectal cancer incidence and mortality: systematic review and meta-analysis of randomised controlled trials and observational studies.
      • Buskermolen M.
      • Cenin D.R.
      • Helsingen L.M.
      • et al.
      Colorectal cancer screening with faecal immunochemical testing, sigmoidoscopy or colonoscopy: a microsimulation modelling study.
      • Ran T.
      • Cheng C.Y.
      • Misselwitz B.
      • et al.
      Cost-effectiveness of colorectal cancer screening strategies-a systematic review.
      RCT-based evidence is even more limited regarding direct comparisons of the effectiveness of various screening options. In this issue of Gastroenterology, results are presented from a Norwegian RCT including almost 140,000 participants that was designed to directly compare, for the first time, the effectiveness of FIT-based screening (4 biennial rounds of FIT) with once-only FS screening with respect to reduction of CRC mortality.
      • Randel K.R.
      • Schult A.L.
      • Botteri E.
      • et al.
      Colorectal cancer screening with repeated fecal immunochemical test versus sigmoidoscopy: baseline results from a randomized trial.
      This Norwegian trial will provide most welcome and long-awaited evidence on the comparative performance of these 2 widely used screening strategies for CRC. In the design and sample size calculation, a decrease in CRC mortality (the primary end point) was assumed to be twice as great in the FS arm (30%) than in the FIT arm (15%). The first reported results on screening uptake and findings at screening show higher participation rates and higher detection rates of advanced neoplasms in the FIT compared with the FS arm after the first 3 FIT rounds in intention-to-treat analyses. These initial findings may challenge this assumption, but it will require 10-years of follow-up until the results for CRC mortality will be available. These initial findings also show that the detection rate of any adenomas was higher in the FS than in the FIT arm, as was the detection rate of advanced adenomas in per-protocol analyses. Cost-effectiveness analyses will be particularly important for this trial given that the FIT arm resulted in more than double the number of colonoscopy referrals compared with the FS arm, which is a key consideration in colonoscopy-constrained settings.
      The article on this Norwegian trial, as well as previous RCTs on CRC screening and long-term outcomes (Table 1)
      • Miller E.A.
      • Pinsky P.F.
      • Schoen R.E.
      • et al.
      Effect of flexible sigmoidoscopy screening on colorectal cancer incidence and mortality: long-term follow-up of the randomised US PLCO cancer screening trial.
      • Atkin W.
      • Wooldrage K.
      • Parkin D.M.
      • et al.
      Long term effects of once-only flexible sigmoidoscopy screening after 17 years of follow-up: the UK Flexible Sigmoidoscopy Screening randomised controlled trial.
      • Segnan N.
      • Armaroli P.
      • Bonelli L.
      • et al.
      Once-only sigmoidoscopy in colorectal cancer screening: follow-up findings of the Italian Randomized Controlled Trial--SCORE.
      • Holme Ø.
      • Løberg M.
      • Kalager M.
      • et al.
      Long-term effectiveness of sigmoidoscopy screening on colorectal cancer incidence and mortality in women and men: a randomized trial.
      • Bretthauer M.
      • Kaminski M.F.
      • Løberg M.
      • et al.
      Population-based colonoscopy screening for colorectal cancer: a randomized clinical trial.
      , illustrate some major challenges in assessing the effectiveness of CRC screening in RCTs. First, despite being one of the most common cancers, CRC mortality in the general population is relatively low, which means that huge numbers of participants are needed in an RCT to obtain a sufficient number of end points. Second, owing to the relatively slow development of most CRCs from their precursor lesions and increasing survival times after CRC diagnosis, very long follow-up periods are required, typically a minimum of 10 years from completion of recruitment. With recruitment of large study populations taking many years (7 years in the Norwegian trial, from 2012 to 2019) and additional time needed for acquisition of funding and preparation, it typically takes at least 15–20 years from RCT design until the main results are available. Given the ongoing progress in various fields, including screening test technologies and quality, the main RCT findings are at high risk of pertaining to outdated methods. Third, such huge RCTs typically are able to compare 2 or at best 3 trial arms with very specific screening modalities. However, in the field of CRC screening, there are multiple factors beside the screening test itself that matter, including (but not limited to) the age range of the target population, screening intervals, referral criteria to colonoscopy according to FS findings, FIT positivity threshold, and surveillance intervals according to findings at colonoscopy.
      • Rex D.K.
      • Boland C.R.
      • Dominitz J.A.
      • et al.
      Colorectal cancer screening: recommendations for physicians and patients from the U.S. Multi-Society Task Force on Colorectal Cancer.
      • Gupta S.
      • Lieberman D.
      • Anderson J.C.
      • et al.
      Recommendations for follow-up after colonoscopy and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer.
      • Cross A.J.
      • Robbins E.C.
      • Pack K.
      • et al.
      Long-term colorectal cancer incidence after adenoma removal and the effects of surveillance on incidence: a multicentre, retrospective, cohort study.
      Each of these factors may have a major impact on CRC mortality. A single RCT, such as that reported here, will provide a comparison of just 2 very specific screening options among a large number of possible options defined by factors such as those outlined above. Fourth, unlike other medical interventions, CRC screening does not take place in an “intervention-naive” population. Even in Norway, one of the few high-income countries where CRC screening has, despite its proven effectiveness, not been offered to the general population so far, a large proportion of the target population for screening has had a colonoscopy (probably mostly for diagnosis of symptoms). According to survey data collected in 2014, approximately 30% of the Norwegian population aged 50–74 years reported having had a colonoscopy in the preceding 10 years.
      • Cardoso R.
      • Guo F.
      • Heisser T.
      • et al.
      Utilisation of colorectal cancer screening tests in European countries by type of screening offer: results from the European Health Interview Survey.
      The majority of such people would be expected to have had a negative colonoscopy and would typically not be recommended to undergo any CRC screening within 10 years after that colonoscopy, or they would have had premalignant lesions removed, decreasing their risk of future CRC. In the Norwegian trial, participants were invited for screening regardless of history of colonoscopy, which was unknown. Even though, as the authors note, it is expected that people with a preceding colonoscopy would be equally distributed between the trial arms owing to randomization, screening adherence (and later results on mortality reduction) needs to be interpreted with caution in light of the high background colonoscopy prevalence.
      Table 1Selected Key Randomized Controlled Trials on the Effects of Colorectal Cancer Screening on Colorectal Cancer Mortality
      First author, year (reference)CountryIncluded (Prescreened)Screening OfferedRecruitment PeriodPublication of Main Results
      Shaukat et al
      • Shaukat A.
      • Mongin S.J.
      • Geisser M.S.
      • et al.
      Long-term mortality after screening for colorectal cancer.
      (2013)
      US46,551Annual/ biennial gFOBT vs none1975–19781993, 1999, 2013
      Miller et al
      • Miller E.A.
      • Pinsky P.F.
      • Schoen R.E.
      • et al.
      Effect of flexible sigmoidoscopy screening on colorectal cancer incidence and mortality: long-term follow-up of the randomised US PLCO cancer screening trial.
      (2019)
      US154,900 (NR)Repeat FS vs none1993–20012012, 2019
      Atkin et al
      • Atkin W.
      • Wooldrage K.
      • Parkin D.M.
      • et al.
      Long term effects of once-only flexible sigmoidoscopy screening after 17 years of follow-up: the UK Flexible Sigmoidoscopy Screening randomised controlled trial.
      (2017)
      UK170,432 (368,142)Single FS vs none1994–19992010, 2017
      Segnan et al
      • Segnan N.
      • Armaroli P.
      • Bonelli L.
      • et al.
      Once-only sigmoidoscopy in colorectal cancer screening: follow-up findings of the Italian Randomized Controlled Trial--SCORE.
      (2011)
      Italy34,272 (236,568)Single FS vs none1995–19992011
      Holme et al
      • Holme Ø.
      • Løberg M.
      • Kalager M.
      • et al.
      Long-term effectiveness of sigmoidoscopy screening on colorectal cancer incidence and mortality in women and men: a randomized trial.
      (2018)
      Norway98,678Single FS vs none1999–20002009, 2014, 2018
      Bretthauer et al
      • Bretthauer M.
      • Kaminski M.F.
      • Løberg M.
      • et al.
      Population-based colonoscopy screening for colorectal cancer: a randomized clinical trial.
      (2016)
      Multiple European94,959Colonoscopy vs none2009–2014(not before mid 2020s)
      Randel et al
      • Randel K.R.
      • Schult A.L.
      • Botteri E.
      • et al.
      Colorectal cancer screening with repeated fecal immunochemical test versus sigmoidoscopy: baseline results from a randomized trial.
      (2020)
      Norway139,291Single FS vs 4 biennial FITs2012–2019(not before end 2020s)
      FIT, fecal immunochemical test; FS, flexible sigmoidoscopy; gFOBT, guaiac-based fecal occult blood test.
      Notwithstanding the undisputed merits of RCTs, such as the one presented in this issue, researchers, clinicians, and public health authorities need to be aware of the limitations of this apparent gold standard design in the evaluation of CRC screening programs. Comprehensive consideration of evidence from CRC screening studies beyond RCTs, taking into account the pros and cons of each design, is required for informed judgement of the evidence and for decisions on screening implementation and refinement. Given the many factors potentially influencing screening effectiveness, comprehensive comparative evaluation of all options by long-term RCTs would neither be feasible nor desirable. Comprehensive modelling of different screening options, with models informed by results of both RCTs and observational studies to the best possible extent, can be a promising approach to this end and should be pursued with priority.
      • Buskermolen M.
      • Cenin D.R.
      • Helsingen L.M.
      • et al.
      Colorectal cancer screening with faecal immunochemical testing, sigmoidoscopy or colonoscopy: a microsimulation modelling study.
      ,
      • van Hees F.
      • Saini S.D.
      • Lansdorp-Vogelaar I.
      • et al.
      Personalizing colonoscopy screening for elderly individuals based on screening history, cancer risk, and comorbidity status could increase cost effectiveness.
      • Ladabaum U.
      • Mannalithara A.
      Comparative effectiveness and cost effectiveness of a multitarget stool DNA test to screen for colorectal neoplasia.
      • Naber S.K.
      • Kuntz K.M.
      • Henrikson N.B.
      • et al.
      Cost effectiveness of age-specific screening intervals for people with family histories of colorectal cancer.
      • Chen C.
      • Stock C.
      • Hoffmeister M.
      • et al.
      Optimal age for screening colonoscopy: a modeling study.
      • Ladabaum U.
      • Mannalithara A.
      • Meester R.G.S.
      • et al.
      Cost-effectiveness and national effects of initiating colorectal cancer screening for average-risk persons at age 45 years instead of 50 years.
      • Heisser T.
      • Weigl K.
      • Hoffmeister M.
      • et al.
      Age-specific sequence of colorectal cancer screening options in Germany: a model-based critical evaluation.

      References

        • Shaukat A.
        • Mongin S.J.
        • Geisser M.S.
        • et al.
        Long-term mortality after screening for colorectal cancer.
        N Engl J Med. 2013; 36: 1106-1114
        • Brenner H.
        • Stock C.
        • Hoffmeister M.
        Effect of screening sigmoidoscopy and screening colonoscopy on colorectal cancer incidence and mortality: systematic review and meta-analysis of randomised controlled trials and observational studies.
        BMJ. 2014; 348: g2467
        • Buskermolen M.
        • Cenin D.R.
        • Helsingen L.M.
        • et al.
        Colorectal cancer screening with faecal immunochemical testing, sigmoidoscopy or colonoscopy: a microsimulation modelling study.
        BMJ. 2019; 367: l5383
        • Ran T.
        • Cheng C.Y.
        • Misselwitz B.
        • et al.
        Cost-effectiveness of colorectal cancer screening strategies-a systematic review.
        Clin Gastroenterol Hepatol. 2019; 17: 1969-1981.e15
        • Randel K.R.
        • Schult A.L.
        • Botteri E.
        • et al.
        Colorectal cancer screening with repeated fecal immunochemical test versus sigmoidoscopy: baseline results from a randomized trial.
        Gastroenterology. 2021; 160: 1085-1096
        • Miller E.A.
        • Pinsky P.F.
        • Schoen R.E.
        • et al.
        Effect of flexible sigmoidoscopy screening on colorectal cancer incidence and mortality: long-term follow-up of the randomised US PLCO cancer screening trial.
        Lancet Gastroenterol Hepatol. 2019; 4: 101-110
        • Atkin W.
        • Wooldrage K.
        • Parkin D.M.
        • et al.
        Long term effects of once-only flexible sigmoidoscopy screening after 17 years of follow-up: the UK Flexible Sigmoidoscopy Screening randomised controlled trial.
        Lancet. 2017; 389: 1299-1311
        • Segnan N.
        • Armaroli P.
        • Bonelli L.
        • et al.
        Once-only sigmoidoscopy in colorectal cancer screening: follow-up findings of the Italian Randomized Controlled Trial--SCORE.
        J Natl Cancer Inst. 2011; 103: 1310-1322
        • Holme Ø.
        • Løberg M.
        • Kalager M.
        • et al.
        Long-term effectiveness of sigmoidoscopy screening on colorectal cancer incidence and mortality in women and men: a randomized trial.
        Ann Intern Med. 2018; 168: 775-782
        • Bretthauer M.
        • Kaminski M.F.
        • Løberg M.
        • et al.
        Population-based colonoscopy screening for colorectal cancer: a randomized clinical trial.
        JAMA Intern Med. 2016; 176: 894-902
        • Rex D.K.
        • Boland C.R.
        • Dominitz J.A.
        • et al.
        Colorectal cancer screening: recommendations for physicians and patients from the U.S. Multi-Society Task Force on Colorectal Cancer.
        Gastroenterology. 2017; 153: 307-323
        • Gupta S.
        • Lieberman D.
        • Anderson J.C.
        • et al.
        Recommendations for follow-up after colonoscopy and polypectomy: a consensus update by the US Multi-Society Task Force on Colorectal Cancer.
        Gastroenterology. 2020; 158: 1131-1153.e5
        • Cross A.J.
        • Robbins E.C.
        • Pack K.
        • et al.
        Long-term colorectal cancer incidence after adenoma removal and the effects of surveillance on incidence: a multicentre, retrospective, cohort study.
        Gut. 2020; 69: 1645-1658
        • Cardoso R.
        • Guo F.
        • Heisser T.
        • et al.
        Utilisation of colorectal cancer screening tests in European countries by type of screening offer: results from the European Health Interview Survey.
        Cancers. 2020; 12: 1409
        • van Hees F.
        • Saini S.D.
        • Lansdorp-Vogelaar I.
        • et al.
        Personalizing colonoscopy screening for elderly individuals based on screening history, cancer risk, and comorbidity status could increase cost effectiveness.
        Gastroenterology. 2015; 149: 1425-1437
        • Ladabaum U.
        • Mannalithara A.
        Comparative effectiveness and cost effectiveness of a multitarget stool DNA test to screen for colorectal neoplasia.
        Gastroenterology. 2016; 151: 427-439.e6
        • Naber S.K.
        • Kuntz K.M.
        • Henrikson N.B.
        • et al.
        Cost effectiveness of age-specific screening intervals for people with family histories of colorectal cancer.
        Gastroenterology. 2018; 154: 105-116.e20
        • Chen C.
        • Stock C.
        • Hoffmeister M.
        • et al.
        Optimal age for screening colonoscopy: a modeling study.
        Gastrointest Endosc. 2019; 89: 1017-1025.e12
        • Ladabaum U.
        • Mannalithara A.
        • Meester R.G.S.
        • et al.
        Cost-effectiveness and national effects of initiating colorectal cancer screening for average-risk persons at age 45 years instead of 50 years.
        Gastroenterology. 2019; 157: 137-148
        • Heisser T.
        • Weigl K.
        • Hoffmeister M.
        • et al.
        Age-specific sequence of colorectal cancer screening options in Germany: a model-based critical evaluation.
        PLoS Med. 2020; 17e1003194