Endocrine Abstracts

Background/Aims: Prognosis in midgut carcinoid tumours (MGCs) is difficult to predict. Studies of traditional clinical parameters give conflicting results. This is a large retrospective survival analysis of both clinical and biochemical data collected at diagnosis before treatment.
Methods: Sequential cases of MGCs were identified using a database in Belfast from 1978. Clinical notes and pre-treatment plasma Neurokinin A (NKA) concentrations, were reviewed for each patient and survival was calculated using all cause mortality or survival to January 2002 as endpoints. Age, gender, symptoms, primary site, size and number, histological depth of penetration and metastases at diagnosis were recorded. Surgical success and the presence of a second neoplasm were also documented.
Results:
One hundred and seventeen patients were identified. Univariate analysis revealed shorter survival to be associated with advanced age (p=0.0017); flushing at presentation (p=0.005); more than 4 liver metastases (p=0.0006); primary tumour not resected (p=0.0078) and plasma NKA greater than 50 nanograms per litre (upper limit normal 20 nanograms per litre) (p<0.0001).
Multivariate analysis of clinical data alone (113 patients) showed only more than 4 liver metastases (p=0.047) and age greater than 70 (p<0.0001) to be independent markers of poor prognosis. For multivariate analysis of both biochemical and clinical data (44 patients), only plasma NKA exceeding 50 nanograms per litre was significantly associated with poor prognosis (p=0.005).
Conclusions
Advanced age and more than four liver metastases at diagnosis are the best clinical markers of poor prognosis. Plasma neurokinin A concentration greater than 50 nanograms per litre is a more powerful predictor of poor prognosis for midgut carcinoid tumours than traditional clinical markers.
This work was supported by an educational grant from the Royal Victoria Hospitals Research Fellowships Committee and an educational grant from Novartis Pharmaceuticals.