Follow-up of patients with ECL cell-derived tumours

Abstract

Objectives: To review the presentation, treatment and outcome of patients with type 1 gastric carcinoid tumours.

Material and methods: We retrospectively reviewed medical records and re-evaluated histopathological specimens of 26 patients with type 1 gastric carcinoids treated at a single tertiary referral centre from 1993 to 2013, with median time of follow-up 52.5 months (IQR 90.8).

Results: Seven patients (27%) had single tumours and 19 patients (73%) multiple tumours at the time of diagnosis. The median number of tumours and median diameter of largest tumour were 2.5 (IQR 3.2) and 6.0 mm (IQR 9.5) respectively. Median serum gastrin was 321.0 pmol/l (IQR 604.0) and median serum chromogranin A 7.7 nmol/l (IQR 5.3). Three patients had metastatic disease at the time of diagnosis and two developed metastases during follow-up. Patients with metastatic disease had larger primary tumours than the others (20.0 mm (IQR 28.5) vs. 5.0 mm (IQR 5.5), p = 0.04). There was a positive correlation between age and tumour size (r = 0.44, p = 0.03) and between serum chromogranin A and serum gastrin at diagnosis (r = 0.76, p = 0.001). Patients were either treated with surgery (n = 8 (31%)), a long-acting somatostatin analogue and/or gastrin receptor antagonist (n = 10 (39%)) for a period of time, or were observed without treatment (n = 8 (31%) with close endoscopic follow up.

Conclusions: Although gastric carcinoids have an overall good prognosis, a significant proportion develops metastatic disease. As partial and total gastrectomy is associated with major side effects, treatment with long-acting a somatostatin analogue or gastrin antagonist should be considered.

Keywords:

• ECL cell
• gastrin
• gastric carcinoids
• somatostatin analogues

Acknowledgements

We thank Bjørn Munkvold and Britt Schulze for technical assistance.

Disclosure statement

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.