Is 45 the new 50? Controversies in lowering the screening age for colorectal cancer

KEYWORDS:

• Colon cancer
• screening
• young
• outcomes
• cancer

Colorectal cancer (CRC) remains both the third most common and deadly cancer in the United States in 2018 [1]. This is despite overall decreases in incidence and mortality, presumably due to increased screening efforts and early detection [1,2]. Recently, the American Cancer Society (ACS) published guidelines recommending a decrease in the starting age for CRC screening from 50 to 45 years of age in the average-risk adult population [3]. These recommendations have since established a rich forum for discussion of their merits, especially given the break from the previously unanimous recommendation of beginning screening at 50 years of age as outlined by both the United States Preventive Services Task Force (USPSTF) and the Multi-Specialty Task Force (MSTF) [4–6]. These qualified recommendations were generated both by predictive modeling analyses and age-cohort epidemiological evidence of a relative increase of 51% in CRC incidence in persons younger than 50 between 1974 and 2013 [7,8]. This latter statistic is almost entirely driven by a significant increase in the left colon and rectal malignancies in persons younger than 50, termed early-onset CRC (eoCRC) [7]. Even more, the guideline underscored the phenomenon of a birth-cohort effect as well as the fact that contemporary CRC incidence burden in patients aged 45 nearly mirrors that of patients aged 50 in the early 1990s. The modeling studies predicted that initiating screening at age 45 instead of age 50 would result in approximately 25 more life-years gained per 1000 individuals screened. Possible mechanisms cited for reduced risk include early detection of CRC and detection and removal of adenomatous polyps in the 45–49 year olds that would reduce the incidence of CRC in those 50 and older. It is clear that investigations must be designed and implemented to examine both the dynamics driving the increased incidence of colorectal cancer in younger persons and prospective studies focusing on how lowering the screening age will influence this trend, if at all.

Pros of an approach of initiating CRC screening at age 45 instead of 50 include detection of advanced adenomas that may prove lethal, the potential for boosting screening rates in individuals ≥50 year olds as well as more timely screening for African Americans, where the recommended starting age is 45 already. Also, preventing young CRC is a desirable goal, as the societal impact of CRC death at a young age is particularly devastating. Further, it has been projected that expanding the eligible screening population to 45 from 50 would prevent 29,400 cases of colorectal cancer and subsequently 11,100 deaths from this disease over the next 5 years [9]. Undoubtedly, these are appealing projections from a population standpoint. Further, the ACS guideline update may serve as an impetus to design inquiries that would probe overscreening behavior and promote guideline adherence such that resource allocation could be mote equitable where access to CRC screening is limited. The above benefits would support the widespread adoption of the new ACS guideline update.

Cons of such an approach are the utilization patterns and the resources required to screen an additional 21 million individuals between the ages of 45–49, and the potential gains from that effort. Data from the SECAP survey in conjunction with recent CRC screening capacity analyses from the CDC would inform us that we do indeed have sufficient resources to screen all eligible patients age 50–75 [10]. Even more, perhaps we are overscreening and this is the mechanism driving resource displacement (i.e. adherence to guidelines for who and when to screen, especially given current screening rates in the eligible population). The issue that remains would be one of the accesses, in that simply being able to perform the requisite amount of testing does not equate to even socioeconomic and geographic distribution of these services [11].

Furthermore, there is no direct evidence of a reduction in CRC mortality or incidence among individuals <50 years of age aside from what can be extrapolated from prior FOBT trials conducted in Nottingham and Minnesota, although these trials have a small fraction of participants age 45–49. Whereas there is established evidence of the benefit of screening in those 50–75, and adherence to screening in this group needs to be improved. While a laudable goal, the 80% CRC screening rate by 2018 projected by the National Colorectal Cancer Roundtable may not come to fruition. Many centers and specific areas did, in fact, reach 80% screening and it remains to be seen how the final National Health Interview Survey (NHIS) and Behavioral Risk Factor Surveillance System (BRFSS) will reflect final 2018 data. That being said, contemporary rates of screening uptake hover around 60% (57.9% ages 50–64, 62.4% ages 50–75) [12]. Given this, expanding the population to be screened may prove to worsen the issue. Efforts to narrow the unscreened gap in individuals aged 50–75 should arguably be more imperative. In fact, a recent Markov analysis aimed at the population-level impact of implementing the new ACS recommendations found that while it may be cost-effective to begin younger, elevating the screening rate in persons 50–75 to the target of 80% would prevent threetimes as many deaths attributed to CRC for approximately 66% less cost [9]. Perhaps a subgroup analysis of patients aged 50–75 could identify the most cost-effective cohort of patients to screen such that resources and supplies could be spread further to account for the cases in the newly appreciated younger cohort of at-risk patients.

Even more, it is well established that CRC screening uptake is markedly depressed in racial/ethnic minorities, recent immigrants to the United States, those with low socioeconomic status, and those with suboptimal insurance coverage [12]. The new guidelines comment on the fact that incidence patterns driving the recommendation to screen African-American patients at 45 (already in place) are now nearly mirrored in Caucasian patients. The country as a whole already struggles to allocate surveillance resources and access to those known to be chronically underserved. It is possible that the 45–49-year-old population in the newly proposed CRC screening catchment would not be part of any of the aforementioned groups above and thus may aggravate this ongoing disparity as overscreening of well-insured individuals continues [4]. Furthermore, the models used to support ACS recommendations predicted an incremental improvement of 3 CRC cases averted and 1 CRC death prevented per 1000 people screened compared to a strategy of screening colonoscopy every 10 years from age 50 to 75 years [5]. In addition to these small differences, the model assumed perfect adherence to screening and follow-up and was unable to consider the displacement of screening resources from higher- to lower-risk groups.

The question whether eoCRC and sporadic CRC in patients older than 50 share similar pathobiology also remains to be answered. Are they mechanistically distinct or do they both adhere to the classically accepted adenoma-carcinoma pathway? The answers to these questions have important implications as it relates to why and how we screen for CRC as we currently are. Traditional risk factors (obesity, sedentism, and dietary factors) for the development of sporadic CRC in individuals older than 50 years old have not been adequately studied to date in eoCRC, making it difficult to target a specific group in this younger population based on behaviors [3,13]. Recently, various risk scores have proven to be superior to the longstanding application of family history as a means to target CRC screening in younger patients. Specifically, a model incorporating genetic risk and environmental factors in tandem with a family history has proven more accurate than family history alone in establishing the age at which CRC screening should begin [14]. Harnessing scoring systems such as this may allow for targeted screening inclusion of patients 45–50 years old (and possibly even younger) while sparing those unlikely to harbor disease.

In summary, the present phenomenon of increasingly early CRC is an important challenge moving forward for clinicians and investigators alike. There exists a paucity of data, especially prospective, indicating that CRC screening in persons younger than 50 has a meaningful impact on incidence and survival. It will continue to exist as a legitimate public health concern until the driving factors are better appreciated and population-based prospective studies are complete identifying meaningful avenues for detection and prevention. While the updated ACS guidelines have indeed shone a light on this important issue and garnered significant attention in the literature thus far, we would hope that it would not alter practice patterns before necessary investigations can be completed. Further, it would behoove us to improve our current surveillance efforts in the 50–75-year-old population with well-understood benefits to be gained prior to expanding the population to be screened. Also, expansion of the screening population may widen the already concerning socioeconomic and racial/ethnic disparity in CRC screening. These recommendations should, however, begin the conversation for targeted and perhaps even individualized screening of individuals younger than 50 in a more systematic fashion. Finally, we would hope that the publication of this guideline update would implore funding agencies to support prospective CRC screening investigations including patients younger than 50 years, which to date are lacking and have been a significantly missed opportunity.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Two of the three peer reviewers for this manuscript have disclosed that they have participated in work contributing to the formulation of the ACS guidelines discussed in this manuscript. Peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose.

Additional information

Funding

This paper was supported by the Department of Veterans’ Affairs, USA (United States of America): CX 008-16F.