Reviewed by Dr. Branden Tarlow
Corley, Douglas A., et al. Jama 317.16 (2017): 1631-1641

Every PCIM and PRIME resident knows that in a busy clinic, a screening colonoscopy can be a hard sell. Alternative colon cancer screening methods like annual fecal immunohitochemistry testing (e.g. FIT) are becoming increasingly popular due to ease of use, low cost, and effectiveness. But when you’re on a MICU rotation and a positive FIT result lands in your “In Basket,” how quickly must you schedule your patient for colonoscopy? This large retrospective cohort study in the Kaiser Permanente system in California examined over 81,000 positive FIT tests in 1.2 million patients (8.5% positive rate). The primary outcome was diagnosis of colorectal adenocarcinoma at any stage diagnosed within 6 months after colonoscopy. The authors observed patients with positive FIT and separated them into cohorts based on the time to follow up colonoscopy: within 8 to 30 days, 2 months, 3 months, 4 to 6, 7 to 9, 10 to 12, or more than 12 months. Most patients received early follow-up with colonoscopy (63% within 2 months, 75% within 3 months, 83% within 12 months). The overall cancer detection rate within 12 months was 4.9% in patients with a positive FIT test. Compared with patients who received follow up within 1 month, patients who had colonoscopy within 9 months did not have an increased risk of any colorectal cancer or advanced stage cancer. Those who had a colonoscopy more than 10 months later had a higher risk of diagnosis with colon cancer (OR, 1.48 [95% CI, 1.05-2.08]). These differences persisted after controlling for factors such as age, sex, ethnicity, smoking status, and anemia in the year before the FIT.

This study provides reassurance to doctors, patients and health systems like Parkland and the VA that there is no urgent need to rush to colonoscopy after a positive FIT test. A recent study at a VA hospital showed the median follow-up time between positive FIT and colonoscopy was 101 days, which is acceptable according Corley et. al; similar median follow-up times are reported in county health systems. Several societies and systems like the VA recommend colonoscopy within 30-60 days. This may be unnecessary. While the authors report that a 10-month delay was associated with an increase in the risk of cancer diagnosis, some caution is needed even in this interpretation. Two limitations of the study include the facts that the study was observational, and the reason for colonoscopy was not recorded. A possible explanation is that patients who underwent colonoscopy in the late follow-up period disproportionately did so due to symptoms of cancer; in fact, 14% of patients with a positive FIT never had a colonoscopy. A randomized trial would be needed to tease out this association (but this may be unlikely to occur). On a more basic level, these results are compatible with our mechanistic understanding of colon cancer as a slow step-wise process resulting from accumulating genetic mutations and/or epigenetic alterations over decades.

Mortality rates from colon cancer have decreased significantly over the past several decade (see figure) driven by multiple factors including improved screening and early detection. But improved treatments such as better surgical practices, highly effective neoadjuvant chemotherapy regimens, and better diets also have contributed significantly.