当前位置: 首页 > 期刊 > 《新英格兰医药杂志》 > 2005年第13期 > 正文
编号:11325913
Fecal DNA for Colorectal-Cancer Screening
http://www.100md.com 《新英格兰医药杂志》
     To the Editor: Although comparing the performance of a test of fecal DNA with that of a fecal occult-blood test is appealing, it would have been better if Imperiale et al. (Dec. 23 issue)1 had chosen to use fecal occult-blood tests with higher sensitivity, such as tests that measure hemoglobin with an enhancer, which allows the detection of lower levels of peroxidase, fluorimetric tests, and immunochemical tests. An immunochemical test has a sensitivity that ranges from 69 percent to 97 percent.2,3 This test has been chosen as the reference test in Japan, where its use has been shown to be cost-effective.4 Before extending the use of fecal DNA testing worldwide, other crucial points, including the variable performance of the test and its price, should be discussed. Assuming that the cost for the fecal DNA test is about $300 and that it is about $600 for a colonoscopy performed in the United States ($400 in Europe and much lower in Asia), the Markov model has shown fecal DNA testing performed every 4 years to be less effective and less cost-effective than colonoscopy performed every 10 years.5

    Jean-Louis Frossard, M.D.

    Raymond de Peyer, M.D.

    Geneva University Hospital

    1211 Geneva 14, Switzerland

    jean-louis.frossard@hcuge.ch

    References

    Imperiale TF, Ransohoff DF, Itzkowitz SH, et al. Fecal DNA versus fecal occult blood for colorectal-cancer screening in an average-risk population. N Engl J Med 2004;351:2704-2714.

    St John DJ, Young GP, Alexeyeff MA, et al. Evaluation of new occult blood tests for detection of colorectal neoplasia. Gastroenterology 1993;104:1661-1668.

    Allison JE, Tekawa IS, Ransom LJ, Adrian AL. A comparison of fecal occult-blood tests for colorectal-cancer screening. N Engl J Med 1996;334:155-160.

    Shimbo T, Glick HA, Eisenberg JM. Cost-effectiveness analysis of strategies for colorectal cancer screening in Japan. Int J Technol Assess Health Care 1994;10:359-375.

    Song KS, Ladabaum U. Potential cost-effectiveness of molecular stool marker testing compared to conventional colorectal cancer (CRC) screening methods. Gastroenterology 2003;124:Suppl 1:A-76. abstract.

    To the Editor: Imperiale et al. acknowledge that participation is a problem in colorectal-cancer screening. Their study excluded persons who did not collect an adequate stool sample for fecal DNA analysis, fecal occult-blood analysis, or both. The specific fecal occult-blood test that was studied required drug and dietary restrictions and samples of three stools. The fecal DNA analysis required persons to collect and refrigerate an entire stool.1 A significantly greater number did not collect an appropriate sample for DNA analysis (641 subjects) than did not collect the samples for the test for fecal occult blood (426 subjects) (chi-square = 47.99, P<0.001), suggesting that the fecal DNA test was less acceptable to the study population. The option to use an analysis of fecal DNA is an exciting development, but there is a need for equally sophisticated behavioral research that may lead to better methods to gain people's compliance with stool testing, without which the DNA test may indicate only "did not attempt."

    Daniel L. Worthley, M.B., B.S.

    Stephen R. Cole, B.Sc.

    Graeme P. Young, M.D.

    Flinders University of South Australia

    Adelaide, SA 5042, Australia

    danielworthley@bigpond.com

    Dr. Young reports having received research funds and consulting fees from Enterix.

    References

    Woolf SH. A smarter strategy? Reflections on fecal DNA screening for colorectal cancer. N Engl J Med 2004;351:2755-2758. [Full Text]

    The authors reply: In response to Drs. Frossard and de Peyer: we chose Hemoccult II as the comparator fecal occult-blood test because it has been proved to reduce mortality from colorectal cancer1,2; it is probably the most widely used fecal occult-blood test in the United States; and it has been recommended in published guidelines.2 We agree that other fecal occult-blood tests may be promising and deserve consideration.

    The estimates in the studies cited of 69 to 97 percent for the sensitivity of other fecal occult-blood tests are probably inflated. One study enrolled subjects with symptomatic colorectal cancer,3 who may be more likely to have a positive fecal occult-blood test and who are not representative of a screened population. In the other study, colonoscopy was performed only when the fecal occult-blood test was positive, rather than in all subjects4 — a strategy that inflates sensitivity because subjects with nonbleeding cancers are not identified.

    We agree with Drs. Frossard and de Peyer that both the consistency of test performance and the cost require consideration. All fecal DNA testing for our study was performed in one location. We suggested in our article that, with more widespread use of the test, it would be necessary to monitor variability in test performance. Although the cost-effectiveness analysis by Song et al.5 showed that colonoscopy every 10 years was more cost-effective than fecal DNA testing every 5 years, the fecal DNA test was cost-effective as compared with no screening.

    In response to Worthley and colleagues, dietary and medication restrictions were suggested but not required before testing with Hemoccult II. We required six samples for Hemoccult II because the six-specimen test has been shown to be associated with a reduction in mortality due to colorectal cancer.1,2 For fecal DNA analysis, the subjects did not have to refrigerate a stool specimen; after sealing the specimen-collection container, subjects placed it in a shipping box along with two freezer packs. We agree that further research is needed to understand and address patients' preferences and behavior regarding screening tests for colorectal cancer. Our study was a first step toward understanding how fecal DNA testing may be applied to an average-risk population.

    Thomas F. Imperiale, M.D.

    Indiana University

    Indianapolis, IN 46202

    David F. Ransohoff, M.D.

    University of North Carolina at Chapel Hill

    Chapel Hill, NC 27514

    Steven H. Itzkowitz, M.D.

    Mount Sinai School of Medicine

    New York, NY 10029