Early Screening Is Currently Recommended for First-Degree Relatives of Individuals With Adenomas
Some guidelines recommend early colonoscopy screening for colorectal cancer (CRC) at age 40 years for first-degree relatives of individuals diagnosed with adenoma at aged < 60 years, but to our knowledge to date, population-based evidence to support this recommendation has been limited.[1, 2]
Using Population-Based Data, Tuohy and Colleagues Provide Evidence of an Increased Risk of CRC Among Relatives of Individuals With Adenomas
In this issue of Cancer, Tuohy et al have presented a population-based observational study of CRC risk among relatives of 126,936 individuals who underwent colonoscopy over a 14-year period. Compared with individuals without adenomas at colonoscopy, first-degree and second-degree relatives of individuals with adenomas were more likely to have CRC. The association was found to be stronger among first-degree relatives of individuals with villous adenomas. An increased risk of CRC among relatives was observed regardless of whether patients with adenomas were diagnosed before or after age 60 years. Among patients with adenomas who were aged < 60 years, 10% of first-degree relatives with CRC were diagnosed at an age < 50 years, before the recommended screening initiation age for individuals considered to be at average risk.
The study had important strengths. The population was drawn from 2 health systems that provide > 80% of the health care in Utah, and was uniquely linked to long-standing, detailed, population-based cancer and genealogy registries in the state. Thus, it is likely that those individuals who underwent colonoscopy were accurately characterized with respect to the presence of adenoma, and that CRC was accurately ascertained for relatives, issues that have challenged the validity and interpretability of prior studies on this topic.
A few limitations may be considered. Although population-based, the group under study was fairly homogenous, representing whites of European ancestry, which perhaps challenges generalizability. Environmental and genetic risk factors for CRC (such as smoking or the presence of Lynch syndrome) were not measured. These factors could be causative factors critical to understanding the association observed or, alternatively, could be confounders. Not all colonoscopy factors of interest were available for analysis. For example, with respect to polyps, information regarding the presence of villous features was available but not so other features such as adenoma size, presence of high-grade dysplasia, or adenoma multiplicity. It is possible that these features might have a similar or even greater risk association with CRC among relatives than was observed for individuals having an adenoma with villous features. A potentially important issue concerns timing of CRC ascertainment among relatives of individuals who underwent colonoscopy. Relatives of patients with CRC may be more likely to undergo colonoscopy due to an increased risk of CRC and guideline recommendations for screening, and are more likely to have adenomas.[4, 5] If CRC ascertainment included the time period before colonoscopy, it is possible that a disproportionate number of individuals with adenomas at the time of colonoscopy underwent colonoscopy due the outcome of interest: CRC among relatives. Such reverse causation could have biased the results toward finding a stronger association between the presence of adenoma at colonoscopy and CRC among first-degree relatives than was truly present. Indeed, others have raised a concern that including individuals undergoing colonoscopy who have had first-degree relatives with CRC before the time of the qualifying colonoscopy answers the question of whether individuals with a first-degree relative with CRC have a higher risk of adenomas rather than the question of whether individuals who have a first-degree relative with adenoma have a higher risk of developing CRC. This concern could be addressed in the current study by checking whether the results changed after limiting the ascertainment of cancer among relatives to after the date of colonoscopy for both cases and controls.
If Risk Is Elevated, Now What?
Given the consistency of their report with other studies, if we accept the strengths of the study by Tuohy et al, and assume the concerns regarding reverse causality are not highly influential, it appears that CRC is more common among first-degree relatives of individuals with adenomas at the time of colonoscopy. The question is, now what?
Early Screening Could Be Promoted, but Challenges to Effectiveness Are Substantial
As mentioned previously, guidelines have recommended early colonoscopy screening starting at age 40 years for first-degree relatives of individuals diagnosed with adenoma at age < 60 years. This approach may not be an effective public health strategy without more research for several reasons.
First, a modest 1.35-fold increased risk of CRC for first-degree relatives of patients with adenomas was observed. Using a modestly increased risk to guide screening recommendations (such as early age at screening initiation) may not be an efficient strategy for reducing the public health burden of CRC. In future work, presenting the absolute risk of CRC among the first-degree relatives of individuals with and without adenomas and conducting formal modeling studies may help to clarify this issue. Moreover, from the individual-level perspective, if the absolute risk could be effectively communicated to patients, some might not choose early screening and might prefer to wait until age 50 years.
Second, practical barriers to the implementation of this strategy may be substantial. Patients with polyps are rarely ever able to self-report the type of polyp they had, and providers do not always have easy access to colonoscopy and pathology reports.7,8. In addition, the current rate of adenoma detection among average-risk individuals is 30%. Based on the large number of individuals expected to be diagnosed with an adenoma, it has been estimated that if guidelines for early screening among relatives of patients with adenomas were followed, > 50% of the population would require early screening.
Effective Strategies for Risk Management Among Relatives of Individuals With Adenomas Must Be Pursued
Because the risk for first-degree relatives of patients with adenomas appears to be elevated, more research to inform management should be pursued. First, more studies of the prevalence of significant neoplasia among individuals aged 40 years to 49 years who are undergoing colonoscopy because they have a first-degree relative with adenoma should be done to understand the potential benefits of early screening. Prior work has been limited by small sample sizes; for example, one prior study reported the rate of adenomas, but not advanced adenomas, was higher among the first-degree relatives of patients with adenomas who were aged 40 to 49 years compared with controls of the same age, but estimates were based on 176 relatives of patients with adenomas who were undergoing colonoscopy.
Second, notwithstanding a change in current guidelines, effective strategies to communicate polyp findings to patients with adenomas and the guideline recommendation for early screening for first-degree relatives must be developed. There is some evidence that providing personalized polyp information and brief educational interventions can be effective.
Third, new solutions for identifying those patients with a truly elevated risk of CRC should be developed. Although current guidelines use family history of CRC and adenoma as well as a personal history of adenoma to determine the recommended screening regimen, it may be possible to develop more advanced methods of risk modeling that incorporate demographic factors (eg, sex), clinical factors (eg, nonsteroidal antiinflammatory drug use), and genetic risk markers (eg, APC11307K or SMAD7 variants). In addition, to the best of our knowledge, to date the main solution for populations at an increased risk of CRC has been to recommend early and/or more frequent colonoscopy. Because this may not be feasible for the growing number of relatives of patients with adenomas, we can consider investigating whether the use of blood and stool biomarkers such as fecal DNA testing or fecal immunochemical testing could be used for more precise triage of patients to early colonoscopy.
Overall, the work by Tuohy et al suggests that relatives of patients with adenomas have a modestly increased risk of developing CRC. Further studies are needed to determine how the observation of a modestly increased risk should be translated into practice to optimize the effectiveness of CRC screening and prevention, and whether guidelines should recommend special screening for relatives of patients with adenomas. However, these relatives can be counseled regarding their potential increased risk and encouraged to participate in screening, although to the best of our knowledge, the optimal age at which to initiate screening remains uncertain.