Optimal Screening Method for Colorectal Cancer Still Debated

June 2011, Vol 2, No 3 -


San Francisco, CA—Colorectal cancer is preventable but remains the secondleading cause of cancer death, because of persistently low screening rates. At the 2011 Gastrointestinal Cancers Symposium, colorectal cancer screening experts debated the relative merits of 2 modalities competing for patients and reimbursement dollars, both claiming to be cost-effective.

Computed Tomographic Colonography for Screening
Two- and 3-dimensional computed tomographic colonography (CTC), also known as virtual colonoscopy, holds advantages over conventional optical colonoscopy (OC) for population screening, according to Perry J. Pickhardt, MD, of the University of Wisconsin School of Medicine and Public Health, Madison. However, politics and “turf issues” unrelated to medical evidence have delayed widespread acceptance and third-party coverage, he claims.

The most current versions of CTC can detect 90% of polyps ≥10 mm, which includes most early cancers. According to Dr Pickhardt, CTC offers the following benefits compared with traditional colonoscopy:

  • Less invasive, with little or no risk
  • No need for sedation, pain medication, or recovery time
  • Allows for extracolonic evaluation (finding abdominal aortic aneur – ysms, other cancers)
  • Increases overall screening adherence
  • More cost-effective, because it re – serves invasive colonoscopy for treat ment (polypectomy).

Nearly all studies focusing on the cost-effectiveness of CTC for screening have shown it is more cost-effective than not screening. “CTC compares favorably with colonoscopy when appropriate input assumptions (ie, real values) are applied,” Dr Pickhardt said. Unlike OC, CTC does not pick up the diminutive (≤5 mm) lesions that pose little to no threat and are overtreated.

“Ignoring isolated diminutive lesions at CTC screening has been identified as a drawback by some, but I believe this approach results in a significant cost benefit, without any real penalty in terms of clinical efficacy,” he said. “If we sent every diminutive lesion we detected to colonoscopy, the cost would be exorbitant, but with a focus on large lesions, there’s a good chance that CTC screening can even be cost-saving.”

In a model Dr Pickhardt developed, an endoscopist would need to find 2353 diminutive lesions to identify 1 cancer among them, compared with 297 small lesions (6-9 mm) and just 11 large lesions (≥10 mm). The respective incremental cost ratios were $464, $59, and – $151 per person screened, respectively.

Dr Pickhardt and colleagues compared CTC with OC for the detection of advanced neoplasia in a parallel community screening program of more than 3000 patients per arm (Kim DH, et al. N Engl J Med. 2007;357:1403- 1412). Of the CTC-screened patients, 8% then had OC for further evaluation. Although more adenomatous polyps were removed in the OC arm, slightly more advanced cancers were found with CTC, and 3 times more cancers were detected. Furthermore, no serious complications occurred with CTC compared with 7 with OC.

Based on meta-analyses, Dr Pickhardt calculates a sensitivity of 96.1% with CTC and 94.7% with OC for cancer detection; CTC also detects more proximal (right-sided) lesions, which can often be missed. Although concerns have been raised that CTC misses flat (sessile) polyps, “in our experience these are encountered at both conventional and virtual colon – oscopy screening at a much lower frequency than polypoid lesions and tend to be less aggressive,” he said.

Unresolved Issues
But David Lieberman, MD, of Oregon Health and Science University, Portland, expressed concerns about CTC, including the “threshold for referral to colonoscopy.”

Current guidelines recommend that all patients with polyps ≥6 mm (up to 25% of screened populations) be offered OC. Under this scenario, most decision models show that CTC would be more costly than other screening programs, he said.

He also expressed concerns that small polyps may be identified but never referred for further evaluation by OC, and that flat polyps, which are not easily detected by CTC, will frequently be missed. The need for a second bowel preparation in patients referred to OC is also problematic, and although same-day procedures would circumvent this, few institutions offer this, Dr Lieberman said.

Frequent extracolonic findings on CTC also trigger additional workups, usually for inconsequential reasons. The appropriate interval for screening with CTC also has not been determined.

“CTC provides an excellent structural exam alternative to colonoscopy, but programmatic costs are higher, and are related to the cost of the CTC, rate of referral for colonoscopy, and the evaluation of extracolonic findings,” Dr Lieberman said. “Because of uncertainty about the benefit and risk of extraintestinal evaluations, the US Preventive Services Task Force has not endorsed CTC for screening.”

Colonoscopy Not Perfect Either
“Clouds over colonoscopy” still keep it from being the near-perfect screening modality, Dr Lieberman added. These include the occurrence of interval cancers (ie, those found soon after screening) and low utilization rate, potential for harm, and screening quality in some centers.

Interval cancers—those that appear 3 to 5 years after colonoscopy and polypectomy—are observed in 0.3% to 0.9% of patients. “If colonoscopy were a perfect screening test, we would not expect to see this rate,” he noted.

These may be new, fast-growing lesions or a result of incomplete removal of neoplastic polyps. Of even greater concern, these lesions might have been missed during colonoscopy, which occurs for >10% of polyps >1 cm, Dr Lieberman said. “Missed lesions are the elephant in the endos – copy suite, and it raises questions about the efficacy of colonoscopy.”

Dr Lieberman stressed the importance of a complete examination— navigating the entire colon—that results in greater cancer detection. “We need to do a better job of documenting quality and improving colonoscopy, which is a good test, although not perfect.”

Current Colorectal Cancer Screening* Prevalence