# 2245 Serological Pre-Hospital Detection of Colorectal Neoplasia in a
Symptomatic Cohort - the Next Step Towards a Population
Screening Programme.
Nicholah G. Hurst, Darren M. Williams, Michael J. Wakelam, Tariq
Ismail, Birmingham, United Kingdom
Colorectal cancer (CRC) is well suited to a dedicated screening programme
although none exists in the Western World. The absence of a sensitive,
specific, palatable and affordable test confounds efforts to establish such a
scheme. Traditional imaging techniques of the colorectal mucosa remain
hitherto largely invasive with associated morbidity and mortality, and may
still overlook significant lesions. Serological evaluation for detection of
primary disease has not previously been applied in colorectal cancer. Referrals
for specialist coloproctological opinion are traditionally triaged by general
practitioners as urgent or non-urgent. Failing such designation, hospital
prioritisation occurs via evaluation of symptoms and signs described in
the referring communiqué. The classical tetrad of an altered bowel habit,
rectal bleeding, abdominal pain and weight loss are currently used to identify
patients with a perceived higher risk of neoplastic pathology. Critical
analysis of patients attending our Colorectal Clinic has cast doubt on the
usefulness of symptom patterns in eliciting high-risk patients. Multi-variant
analysis of presenting symptoms versus eventual diagnosis failed to
detect a consistent relationship between presenting symptoms and the presence
of colorectal neoplasia, except for age >50 years and family history of
>1 first degree relative with CRC. It is clear from our observations that
triaging patients on the basis of symptoms alone is inefficient and may
cause delay in consulting a specialist. Serum Matrix Metalloproteinase
(MMP) estimation offers early indication of adenoma and carcinoma development.
Analysis of serum MMPs, from patients presenting at a colorectal
clinic, identifies patients who would benefit from traditional colonic imaging.
A 240 patient cohort had serum MMP9 levels measured, corrected
for age, and predictions of colorectal neoplasia were made based on the
results. Comparison of our forecasts, with diagnoses achieved by conventional
investigations, showed detection of 67/70 (95.7%) significant lesions
by serum analysis. Further improvement by comparative analysis of other
MMPs may prove possible in future assays. Referral of patients with an
accompanying serum sample could provide an objective means of classifying
patients as at high or low risk of CRC, enabling prioritisation of appointments.
Its extended application in the asymptomatic population as a
screening tool is under development.
|