The greater colorectal cancer incidence and mortality in African Americans, compared with other races, may not be due to increased colon polyp prevalence, according to a study from the Medical University of South Carolina in Charleston. The authors said that it was not clear where along the adenoma-carcinoma sequence the African-American population's excess cancer risk lies.

JOSEPH ROMAGNUOLO, MD

Eli Penn, MD, Joseph Romagnuolo, MD, and colleagues from the university's Digestive Disease Center searched their institution's detailed endoscopy database and found 3,732 patients with minimal or no symptoms who underwent their first screening colonoscopy or colonoscopy due to a positive fecal occult blood test (FOBT) from 1996 to 2006. Of these patients, 761 (20.4%) had polyps.

Multivariate analysis showed that age older than 65 years and male sex, but not race, independently predicted the presence of polyps (odds ratio 1.35, 1.67, and 1.04, respectively). In addition, male sex, but not race, predicted right-sided polyps, and the presence of three or more polyps; the latter was also predicted by a positive FOBT result (Arch Intern Med 170:1127-1132, 2010).

Of the 761 patients with polyps, 60% had adenomas, and 57 of these patients had a follow-up colonoscopy at the university a median of 3.6 years after their initial screening. Polyps recurred in 35 patients (71.4% of African Americans [AAs] and 60% of whites). Neither male sex nor race significantly predicted polyp recurrence. A random sample of 100 pathology reports (50 AAs and 50 whites) of polypectomy specimens confirmed no significant racial difference in the proportion of those with polyps having at least one adenomatous polyp (68% of whites vs 60% of AAs, P= .60).

Based on these findings, the authors suggest that colon cancer screening guidelines should be stratified by sex but not race. Due to the small number of patients who underwent follow-up colonoscopy, more data are needed to make recommendations about stratified surveillance intervals by sex or race, but the limited data from this study do not support stratification at this time, the authors noted.

"A significantly higher polyp prevalence and incidence do not seem to serve as the explanation [for AAs' more advanced cancers at presentation and presentation at a younger age]," the authors concluded. They speculated that the rate of malignant transformation of adenomas may be different in AAs or that other advanced colon cancer features are more common in AAs.

VANTAGE POINT

Personalized CRC screening still in the future

HEMANT K. ROY, MD

LAURA K. BIANCHI, MD

These study results are important but there are a number of problems that may limit their relevance, said Drs. Roy and Bianchi in an accompanying editorial. They noted that the use of polyps as an endpoint is "imprecise," that only 100 of the 761 polyp cases underwent histopathologic review, and that adenoma size was not considered.

"Race and sex probably play a role as predictors but are not powerful enough to be relied on solely," wrote Dr. Roy and Dr. Bianchi, who are from the department of medicine, NorthShore University Health System and the University of Chicago Pritzker School of Medicine.

They asked if there are better ways to stratify patients for colon cancer risk. Validated stratification factors include family history, smoking status, diet, exercise level, and NSAID use. Other approaches, including FOBTs and DNA analysis, have low sensitivity for advanced adenomas, and genetic factors, such as single-nucleotide polymorphisms, may not be powerful enough for widespread screening. "Clearly, much more work needs to be done to reach the goal of personalized [colorectal cancer] screening," they wrote.

Drs. Roy and Bianchi stressed that, even if biological differences cannot be identified, the colorectal cancer disparities for African Americans must be addressed, potentially via medical outreach and risk factor modification. They also noted the complexity of defining risk in women, who appear to have fewer adenomas than men but an equal risk of colorectal cancer and a greater risk of proximal colon cancer, which is less easily detected by colonoscopy. "Thus, it seems imprudent to advocate for less aggressive screening in women," they wrote (Arch Intern Med 170:1132-1134, 2010).