|
|
Colorectal cancer (CRC) is the second most common cause of cancer
death in Australia.1 While motor vehicle
accidents cause one death about every five hours, and breast cancer
causes one death every four hours, CRC causes one every two hours.
Australia has made a major government-sponsored effort to reduce
mortality from motor vehicle accidents, and screening to prevent
breast cancer mortality is an accepted government-sponsored
initiative.
Why then is there still confusion and argument about CRC prevention?
There is convincing evidence that finding and removing adenomas in
individuals at increased risk for bowel cancer prevents development
of subsequent cancer in most.2 This would seem logical
given the acceptance of the polyp-cancer sequence.3 In this issue of
the Journal, Croese clearly demonstrates the
potential for improving mortality.4 Using a community-based
open-access colonoscopy service in Townsville, he showed that
patients over 50 years of age who had undergone colonoscopy (with
polypectomy when necessary) were less likely to be subsequently
diagnosed with CRC than the remaining community in the same age group.
Most of his repeat-colonoscopy patients had higher than average risk
for developing CRC, which he defined as having one or more
first-degree relatives with CRC or polyps, or ulcerative colitis,
including quiescent pancolitis or active limited colitis. His
message is simple: CRC can be prevented if those at increased risk are
alerted to the need to enter a colonoscopic surveillance program.
The strengths of the study are that it reports the outcome of "real
world" colonoscopy practice from a relatively confined geographic
area and provides details of cancers occurring during surveillance.
The population was isolated, and the author was able to
comprehensively cross-check data, making the information
particularly valuable. The study's weaknesses -- a heterogeneous,
unmatched population and retrospective comparisons -- were
comprehensively addressed by the author.
Although colonoscopic surveillance of those at higher risk of CRC
is justified, surveillance intervals and starting age remain
controversial. Timing of repeat colonoscopy will be partly
influenced by the possibility of metachronous lesions, although
these have been documented to occur in fewer than 1% of
patients.5 The age at which to begin
colonoscopic surveillance is also debated and, as about 8% of cancers
develop in people aged under 50 years, it would not seem reasonable to
withhold educational information from this group despite any
perceived increase in cost. However, despite the simplicity of the
message, many authorities in Australia still disagree on the need to
deliver it.
We need a coordinated, sponsored public education campaign to inform
our community that an important step to reduce mortality from CRC is
for those at increased risk (eg, first-degree relatives of people
with CRC or polyps) to see a medical practitioner for advice and
referral to an appropriate colonoscopic surveillance program. At
the moment, the message is confused, as so well illustrated by
Ward.6
Can we also shed some light on screening to prevent bowel cancer in the
average-risk individual in our community? Setting aside the issue of
mass screening for now and focusing on case-finding (ie, giving the
appropriate advice to individuals who seek it or who may be receptive
to it), there are four options for prevention or, at least, early
diagnosis.
Screening based on faecal occult blood testing (FOBT) reduces
mortality from CRC. Studies showed a 16% reduction in mortality with
biennial screening in the United Kingdom7 and Denmark,8 while a 33%
reduction was seen with annual screening in the United
States.9 Despite this well-designed
research, the Australian Health Technology Advisory Committee has
recently recommended further pilot studies on the efficacy of FOBT
screening in the over-50 years age group. What other evidence they
require remains a mystery.
Flexible sigmoidoscopy is proposed for screening by many cancer
authorities worldwide, usually in conjunction with FOBT. The
combined approach recognises the limitations of flexible
sigmoidoscopy, which may miss 50% of polyps and CRCs. Several
retrospective studies have found that, in patients with proximal
colon cancers, only 17%-30% of adenomas are in reach of the flexible
sigmoidoscope.10 A prospective
colonoscopy study showed that only 35% of 105 patients with proximal
colon cancer had adenomas distal to the splenic flexure.11 These studies
confirm that rectosigmoid adenomas ("sentinel" polyps) are an
insensitive marker for proximal colon cancer, and that most proximal
colonic neoplasms are not associated with distal polyps or cancer.
These conclusions are supported by recent Australian
data.12 Screening by flexible
sigmoidoscopy alone would fail to detect 70%-80% of proximal
cancers. Addition of annual FOBT would increase the diagnostic
yield, but at increased cost.
Colonoscopy is the third screening option, but has been criticised
because of its cost and the failure to demonstrate that it improves
mortality. Croese found that, in people aged over 50 years, the rate of
cancer diagnosis in the unscreened population was double that in
individuals who had previously had colonoscopy (for whatever
reason). Almost half the cancers in the unscreened population were
Dukes stage C or D, compared with only 16% in the previous-colonoscopy
group.4 Australian data confirm
that the cost-effectiveness of colonoscopy at both five- and 10-year
intervals is almost identical to that of annual FOBT.13 Flexible
sigmoidoscopy, alone or combined with FOBT, was found to be
significantly less cost effective.
Barium enema remains the fourth cost-effective diagnostic option,
although it suffers from the fact that at least 20% of individuals will
have a lesion identified which requires subsequent colonoscopy.
A cohesive, unified and comprehensive public education campaign
about CRC and the potential for its prevention is needed. This should
emphasise the common nature of CRC and should target higher-risk
groups, who can be offered colonic surveillance. This would be a start
in reducing the current high mortality rate. When screening
strategies for early diagnosis or prevention of CRC are chosen,
compliance, costs and efficacy are all key issues. The fact that we
have four effective options now allows the practitioner to offer
individuals a menu from which they can select a test, depending on
their preference and perceived compliance.
Terry D Bolin
Associate Professor, Gastrointestinal Unit
Prince of Wales Hospital, Sydney, NSW
Melvyn G Korman
Associate Professor, Gastroenterology Unit
Monash Medical Centre, Melbourne, VIC
- Anti-Cancer Council of Victoria. Canstat 1997; 26: 2.
-
Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer
by colonoscopic polypectomy. The National Polyp Study Workgroup.
N Engl J Med 1993; 329: 1977-1981.
-
Cotton S, Sharp L, Little J. The adenoma-carcinoma sequence and
prospects for the prevention of colorectal neoplasia. Crit Rev
Oncol 1996; 7: 293-342.
-
Croese J. Colorectal cancer after open-access colonoscopy: a
community and case survey. Med J Aust 1999; 170: 251-254.
-
Leggett BA, Cornwell M, Thomas LR, et al. Characteristics of
metachronous colorectal carcinoma occurring despite colonoscopic
surveillance. Dis Colon Rectum 1997; 40: 603-608.
-
Ward M. Preventing colon cancer: the problem with guidelines or
The perils of prevention. Med J Aust 1997; 166:
201-204.
-
Hardcastle JD, Chamberlain JO, Robinson MH, et al. Randomised
controlled trial of faecal-occult-blood screening for colorectal
cancer. Lancet 1996; 348: 1472-1477.
-
Kronborg O, Fenger C, Olsen J, et al. Randomised study of screening
for colorectal cancer with faecal-occult-blood test. Lancet
1996; 348: 1467-1471.
-
Mandel JS, Bond JH, Church TR, et al. Reducing mortality from
colorectal cancer by screening for fecal occult blood. Minnesota
Colon Cancer Control Study. N Engl J Med 1993; 328: 1365-1371.
-
Lemmel GT, Haseman JH, Rex DK, Rahmani E. Neoplasia distal to the
splenic flexure in patients with proximal colon cancer.
Gastrointest Endosc 1996; 44: 109-111.
-
Rex D, Chak A, Sack L, et al. Prospective determination of distal
colon findings in patients with proximal colon cancer.
Gastrointest Endosc 1998; 47: AB103.
-
Nicholson FB, Stern AI, Korman MG, Hansky J. Colorectal cancer
screening -- are proximal polyps missed by using flexible
sigmoidoscopy? Digestion 1998 Suppl 3: 730.
-
Bolin TD, Korman MG, Stanton R, et al. Positive cost effectiveness
of early diagnosis of colorectal cancer. Colorectal Dis
1999; 1: 2.
-
Reprints: Associate Professor T D Bolin, GI Unit, Prince of Wales
Hospital, High Street, Randwick, NSW 2031.
©MJA 1999
Make a
comment
Reprints: Associate Professor T D Bolin, GI Unit, Prince of Wales Hospital, High street, Randwick, NSW 2031.
|
