  |
|
|
|||||||||||||||||||||||||||||||||||||||||
|
|||||||||||||||||||||||||||||||||||||||||||
| Journal of the New Zealand Medical Association,
22-September-2006, Vol 119 No 1242
|
|||||||||||||||||||||||||||||||||||||||||||
|
Confirming a diagnosis of hereditary colorectal cancer: the
impact of a Familial Bowel Cancer Registry in New Zealand
Paul James, Susan Parry, Julie Arnold, Ingrid Winship
Familial forms of colorectal cancer account for 5–10%
of bowel cancers in countries of high incidence of sporadic colorectal cancer.
Although only a minority of the total cases, this represents around 200 cancers
per year in New Zealand which through timely investigation and intervention
could potentially be prevented.1
Consequently, specialised services have been developed to
investigate and manage affected families. Familial Bowel Cancer Registries
facilitate the collection of accurate family data, identify at-risk individuals,
and coordinate the provision of evidence-based colonoscopic surveillance for
this cohort.
Registry-based programmes have been effective in achieving
gratifying improvements in both the incidence of colorectal cancer and the
overall survival in families affected by either of the two most common genetic
conditions causing bowel cancer—namely, hereditary nonpolyposis colorectal
cancer (HNPCC) and familial adenomatous polyposis (FAP).2–5
A Familial Bowel Cancer Registry was established as a
clinical service in Auckland in 1996. To evaluate the impact of registry
assessment on the management of families with an inherited predisposition to
bowel cancer, we used the registry records to compare the available information,
and resulting management advice, at completion of assessment with that available
at the time of initial referral.
MethodsThe study was performed through an audit of the records
of the Familial Bowel Cancer Registry of patients referred for evaluation, by
the Registry team.
The multidisiplinary Registry team is comprised of a
clinical geneticist, gastroenterologist, registry coordinator, and a researcher,
as well as associated members from the colorectal and oncology teams.
Following referral to the service, a family pedigree
involving at least three generations was constructed initially using information
provided by the proband and augmented by other available family members who
consent to be involved. Details of all cancers in the pedigree were confirmed
where possible through medical records, death certificates, or the National
Health Index database—all of which were accessed with written consent.
In each instance where a cancer was confirmed,
histology was sought for review. The Registry team then reviewed all information
and an assessment of the families’ colorectal cancer (CRC) risk status
made. Specific recommendations regarding colonoscopic and appropriate
extracolonic surveillance procedures were then advised.
Twenty-five consecutive families who completed their
assessment from November 2000 until June 2002 were included in the review.
Information compiled during the assessment was compared to the family history as
known at the time of referral and to any information present in the medical
records of the proband.
ResultsIn the 25 families reviewed there were 90 cancers known to
the referring doctor at the time of referral. Following assessment by the
Registry, a history of 130 cancers was established in these families—an
average of just over 5 cancers per family (Table 1). In a further 9 cases there
was a history of colonic polypectomy for adenomas associated with at least
moderate dysplasia. In two-thirds (85/130) of the cases of malignancy it was
possible to obtain direct confirmation of the diagnosis. In 56 of the 66
confirmed cases of colorectal cancer histology was obtained for review.
Table 1. Number of cancers per family and median age at
diagnosis
Amst.=Amsterdam criteria;
+ve=positive; HNPCC=hereditary nonpolyposis colorectal cancer.
The large majority of the confirmed cancers (87%) belonged
to the spectrum of malignancy that makes up the diagnostic criteria for HNPCC.6
Ten of the 25 families had histories of extracolonic tumours from this spectrum.
Furthermore, 3 cases of endometrial adenocarcinoma, 2 small bowel tumours, and 1
case each of gastric cancer and transitional cell carcinoma were confirmed
(Table 2).
Table 2. HNPCC-associated extracolonic tumours per
family
The age and site distribution of the observed colorectal
cancers differed from that described in the general population (Table 3).7 The
median age at diagnosis tended to be lower (median 60 years). In addition, an
increase in right-sided colorectal cancers was found when compared to available
data on the site distribution of colorectal cancers in the New Zealand
population (Table 3).7
Table 3. Site of colorectal cancer (CRC) in families
compared with site distribution CRC in general population
RC=right colon (caecum and
transverse colon to the splenic flexure); LC=left colon.
Eight of the 25 families at completion of assessment met the
modified Amsterdam Criteria for the diagnosis of HNPCC compared to four families
at the time of referral. These 8 families were referred directly for genetic
testing to detect mutations in the mismatch repair genes identified to cause
HNPCC.
In 6 of the 25 families, the Amsterdam criteria for a
diagnosis of HNPCC were not met—but in line with the Bethesda Criteria,
tumour immunohistochemistry (or microsatellite status), to support the
involvement of the mismatch repair genes in these families, was requested.8 In
one other family, review of the histology led to a diagnosis of FAP.
164 asymptomatic individuals were identified on the basis of
their family history or the age (<55 years) at which a first-degree relative
developed CRC to have a least a moderate increase in their lifetime risk for
developing CRC. Of these, 48 individuals were from families meeting the
Amsterdam criteria.
Individuals were referred to the Registry mostly by their
oncologist or surgeon, and accounted for two-thirds of referrals. Other
individuals were referred severally by gastroenterologists, general
practitioners, or self-referral.
DiscussionCausative factors for familial bowel cancer include several
inherited disorders that are distinct at both the clinical and molecular level
from the common forms of sporadic bowel cancer.
Overall, 3–5% of bowel cancers are due to HNPCC and a
further 0.5–1% to FAP.5 The remaining familial cases are due to either
rare genetic conditions such as Peutz-Jeghers syndrome or familial hyperplastic
polyposis (HP), or represent familial clustering for which the underlying
inherited basis is unknown.
HNPCC and FAP are autosomal dominant single gene disorders
where first-degree relatives of an affected individual have a 50% risk of
inheriting the predisposing mutation. Research and confirmation of the family
cancer history is an essential component of the investigation of affected
families. Histological review of the location and nature of colonic tumours and
confirmation of the profile of extracolonic tumours allows a more accurate
assessment of the true likelihood of an autosomal dominant cancer syndrome in a
family and facilitates the diagnosis of the more unusual polyp
syndromes—e.g. attenuated FAP or HP.
Verification of both the tumour site and age at diagnosis is
a requirement for a diagnosis of HNPCC according to the modified Amsterdam
criteria.6
This study has shown that referral to the Familial Bowel
Cancer Registry has a large impact on the amount of information available for
the assessment and management of individuals belonging to families with a
history of CRC. In the 25 families studied, the Registry assessment process
identified 40 (44% of total) additional cancers, to the 90 reported at the time
of referral. This is consistent with other studies which have shown that as many
as 25% of individuals are unaware of a diagnosis of colorectal cancer in a
first-degree relative.9
At the completion of assessment, one-third of our referred
families met the modified Amsterdam criteria for a diagnosis of HNPCC. This is a
high proportion in comparison with data from other familial cancer registries,
thus suggesting that (in our local setting) referrals to the Registry are only
made when there is a high index of suspicion for the existence of a familial
syndrome.
We also found that the Registry has an important public
health function in that a large number of relatives were identified on the basis
of their family history or the age (<55 years) at which a first-degree
relative developed CRC to have a least a moderate increase in their lifetime
risk for developing CRC.
Of the 164 at-risk first-degree relatives identified from
the 25 families and offered colonoscopic surveillance, less than a third (48 of
164 individuals) were from families meeting the modified Amsterdam criteria for
HNPCC. Registry assessment restricted the recommendation for intensive
colonoscopic surveillance to this smaller group in line with the evidence that
in HNPCC families such surveillance results in a significant reduction in both
the incidence of colorectal cancer and overall mortality.3
For the majority of at-risk first-degree relatives
identified at assessment (116 of 164), 5-yearly colonoscopic surveillance
beginning at the age of 50 years (or at an age 10 years before the earliest age
at which CRC was diagnosed in the family) was advised. In some cases this
represented a reduction in their current surveillance recommendations. The
Registry also provided education and reassurance to many individuals in the
wider families of people affected by bowel cancer, although this benefit is
difficult to quantify.
HNPCC has been documented to be caused by mutations in one
of four mismatch repair genes, with the majority of mutations being identified
in two of the four genes (hMLH1 and hMSH2).10 These genes normally repair errors
that occur in DNA as a result of cell replication.
In New Zealand, the limited resources for predisposition
genetic testing need to be targeted to those most likely to benefit. Registry
assessment facilitated this with the eight families (meeting the modified
Amsterdam criteria for HNPCC) being referred directly for genetic testing to
detect germline mismatch repair gene mutations.
Molecular techniques that detect microsatellite instability
or immunohistochemical tests revealing loss of expression for hMLH1, hMSH2 or
hMSH6 proteins can aid in the detection of tumours resulting from defective
mismatch repair. This is particularly helpful in families where the Amsterdam
criteria for HNPCC are not met, but clinical features (as outlined in the
Bethesda Criteria)8 suggest this diagnosis.
Of the 25 families assessed, 6 are in this category but
genetic testing will only be requested if tumour immunohistochemistry or
microsatellite status supports the involvement of the mismatch repair genes.
Referral to the Familial Bowel Cancer Registry facilitated
the diagnosis of a dominantly inherited bowel cancer syndrome with appropriate
targeting of specialised genetic testing and intensive colonoscopic
surveillance.
Ongoing coordination of colonoscopic surveillance by the
Registry, for those individuals identified to have disease-causing mutations or
to be at-risk, is anticipated to reduce the number of deaths from CRC in these
families.
The New Zealand Guidelines on the surveillance and
management of groups at increased risk of CRC advise offering individuals or
families with hereditary CRC syndromes referral to a Familial Bowel Cancer
Registry.11 To facilitate these referrals, a National Registry based in both
Auckland and Christchurch is now being established.
Author information:
Paul A James, Nuffield Medical Fellow, Department of Human Anatomy and Genetics,
Oxford University, Oxford, UK; Susan Parry, Gastroenterologist, Familial Bowel
Cancer Registry, Northern Regional Genetic Services, Auckland City Hospital,
Auckland; Julie Arnold, Coordinator, Familial Bowel Cancer Registry, Northern
Regional Genetic Services, Auckland City Hospital, Auckland; Ingrid Winship,
Clinical Geneticist, Victorian Genetic Health Service, Royal Melbourne
Children’s Hospital, Melbourne, Australia
Correspondence: Dr
Susan Parry, Northern Regional Genetic Service, Building 18, Auckland Hospital,
Auckland. Fax: 09 307 4978; email: sparry@adhb.govt.nz
References
|
|
|||||||||||||||||||||||||||||||||||||||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|