The 2016 US Preventive Services Task Force (USPSTF) recommendations list 7 different screening strategies including colonoscopy, fecal immunochemical testing (FIT) for occult blood, guaiac-based FOBT (gFOBT), sigmoidoscopy alone, sigmoidoscopy plus FIT, the FIT-DNA test (multitargeted stool DNA test), and computed tomography (CT) colonography.
Based on the development of more sensitive and specific tests to detect human hemoglobin in the stool, FIT largely replaces the gFOBT strategies previously endorsed. Although the majority of clinical trials that demonstrated reduction in colorectal cancer mortality used gFOBT, it is successfully argued that using a more accurate FIT will only improve the mortality benefit from FOBT screening while reducing harms due to unnecessary follow-up colonoscopy following false-positive results. However, there is variability in the test characteristics between different FITs, and although meta-analysis estimates an overall cancer sensitivity and specificity of 79% and 94%, respectively, these range from 73% to 92% and 87% to 95%, respectively.
Flexible sigmoidoscopy was recommended in the prior recommendations at 5-year intervals, with FOBT every 3 years. Several randomized clinical trials confirm the mortality benefit of sigmoidoscopy screening, although Holme et al reported significant benefit only when FIT was added to sigmoidoscopy. The 2016 recommendations extend the sigmoidoscopy screening interval in the combined strategy to 10 years while increasing the frequency of FIT to annually. While it is difficult to exclude sigmoidoscopy from the list of recommended strategies based on the quality of evidence supporting benefit, sigmoidoscopy use in the United States has been declining—fewer than 1% of individuals reported cancer screening with sigmoidoscopy in 2012 compared with 10% who used FIT and 60% who underwent colonoscopy.
Colonoscopy screening at 10-year intervals remains the most prevalent strategy for colorectal cancer screening in the United States. There are no randomized clinical trials to demonstrate a reduction in colorectal cancer mortality with colonoscopy screening, but several are in progress that compare colonoscopy with FIT
The USPSTF recommendation statement also lists FIT-DNA, CT colonography, and a serological test. In 1 study, FIT-DNA was able to detect colorectal cancer with greater sensitivity than FOBT but with lower specificity however, there are no data to establish the morbidity, quality of life, cancer incidence, or mortality benefit of FIT-DNA in a screening population. Comparative effectiveness modeling estimates that other strategies can achieve similar benefits with fewer lifetime colonoscopies per life-year gained. Although the USPSTF report specifically does not address the economic effect of screening, the FIT-DNA test retails in the United States for $649 and is reimbursed by Medicare for $493, compared with the reimbursement for FIT ($22) or colonoscopy ($773). For these reasons FIT-DNA every 3 years is unlikely to represent a cost-effective alternative to screening with FIT or colonoscopy.
Similarly, CT colonography has not been subject to the rigor of a randomized clinical trial with clinically relevant outcomes including morbidity, quality of life, or cancer incidence and mortality. Additional concerns about CT colonography include extracolonic findings that require further evaluation accompanied by potential harms, the potential for radiation-induced cancers, and the lack of data to guide recommendations for the screening interval. A single blood test for circulating methylated septin 9 DNA has been approved by the US Food and Drug Administration for colorectal cancer screening. The sensitivity (48.2%) and specificity (91.5%) for detecting colorectal cancer are lower than for FIT, and no studies have been conducted to demonstrate mortality benefit.
The USPSTF used comparative effectiveness analysis to inform their updated colorectal cancer screening recommendations.
The absence of a USPSTF recommendation for any specific strategy leaves clinicians with a dilemma—the highest-quality data exist for gFOBT, which has been replaced by FIT, and sigmoidoscopy, which is largely unavailable. Colonoscopy is the most often used screening test, and observational studies report reductions in cancer incidence and mortality, yet validation from randomized clinical trials is lacking. The FIT-DNA should be at least as good as FIT, but there are no data to demonstrate greater reductions in cancer mortality beyond FIT. Radiographic and blood-based screening likewise have a paucity of clinical outcome data. It is in this context that the USPSTF has chosen to forgo specific test recommendations and instead highlight the advantages and disadvantages of the strategies presented in their statement. Perhaps the absence of data should not indicate the absence of benefit, and these recommendations should be viewed as a living document that is expected to change as more information become available.
Instead, the USPSTF recommendations highlight the great opportunity to reduce population mortality from colorectal cancer. National adherence to colorectal cancer screening ranges from 58% to 65% and has not increased over the past 5 years. The Affordable Care Act excluded cost sharing (deductible and copayment) for colorectal cancer screening; however, cost sharing is often imposed for colonoscopy either as a follow-up to positive FOBT findings or when a polypectomy is performed during a screening examination. These hidden costs represent a barrier to screening by any modality and should be removed to increase screening adherence.
In sum, we should focus on achieving our goal of screening “80% by 2018. While we may each have our own preference, this should be a patient-focused decision because the best test is not simply the one that gets done, but the one that gets done consistently.
