Ann NY Acad Sci 1988;534: 1-30. 3 Choudary JB. Supervisory p armacologists addendum .... M I CARTER. Luton and Dunstable Hospital,. Luton,. Bedfordshire ...
Omeprazole certainly induces a proliferative response in the gastric mucosa of rats and all other species of animals in which it has been tested and reported. One third of patients treated for more than four years with omeprazole have also developed proliferation of the endocrine enterochromaffin-like cells of the stomach. It is now suggested that the proliferative changes found in patients are attributable to the development of atrophic gastritis in those affected (although most of these patients did not show the changes of atrophic gastritis). The mechanisms by which omeprazole produces gastric mucosal proliferation in animals are not clear. The claim that the gastric mucosal proliferation is caused by hypergastrinaemia has been rejected by the scientists of the Food and Drug Administration3 since the lowest doses of omeprazole used in the rats (doses within the human therapeutic range) produced tumours of the stomach without producing hypergastrinaemia. Current arguments about the magnitude of the human hypergastrinaemic response to omeprazole are not very valuable because the sensitivity of the human gastric mucosa to the proliferation-inducing effects of gastrin (if any) have not been defined. It is not yet clear how pronounced, or otherwise, hypergastrinaemia has to be to trigger clinical concern. Finally, the use of an animal carcinogen to treat benign chronic disease raises important issues of principle that have not yet been satisfactorily addressed. K G WORMSLEY
Ninewells Hospital and Medical School, Dundee DDl 9SY 1 Correspondence.
activity had increased by 1500 dpm in 50 [tg DNA, which is not a marginal rise. Experience gained in testing known carcinogens and non-carcinogens over several years gives us confidence that this method provides a reasonable estimate of unscheduled DNA synthesis, notwithstanding the incomplete inhibition of semiconservative DNA synthesis by hydroxyurea. It is also important to emphasise that our experiments are designed specifically to separate cell proliferative effects from any genotoxic effect.4 Omeprazole did not cause an increase in semiconservative DNA synthesis four hours after its administration-that is, there was no measurable mitotic effect at this time. In conclusion, our studies take account of the incomplete inhibition by hydroxyurea and show that omeprazole induces unscheduled DNA synthesis in the rat pyloric mucosa at a time when there is no measurable increase in normal semiconservative DNA synthesis.
1 Goodlad R, Wright NA. Omeprazole BM7 1991;303:851. (5 October.) 2 Furihata C, Matsushima T. Reply to Professor Evans. Mutat Res 1991;264:89-9 1. 3 Furihata C, Yamawaki Y, Jin S-S, Moriya H, Kodama K, Matsushima T, et al. Induction of unscheduled DNA synthesis in rat stomach mucosa by glandular stomach carcinogens. J Natl Cancer Inst 1984;72:1327-34. 4 Furihata C, Hirose K, Matsushima T. Genotoxicity and cell proliferative activity of omeprazole in rat stomach mucosa. Mutation Research 1991;262:73-6.
BMJ 1991;303:850-1. (5
1991;264:89-91. 5 Rosenkranz HS, Klopman G. Omeprazole: an exploration of its reported genotoxicity. Alutagenesis 1991;6:381-4. 6 Sharma BK, Santana IA, Wood EC, Walt RP, Pereira M, Noone P, et al. Intragastric bacterial activity and nitrosation before, during, and after treatment with omeprazole. BMJ 1984;289: 717-9. 7 Creutzfeldt W, Lamberts R. Is hypergastrinaemia dangerous to man? Scand7 Gastroenterol 1991;26 (suppl 180):179-91.
SIR,-The letter from R Goodlad and N A Wright in the recent debate on omeprazole contains several factual inaccuracies with regard to our research.' I would like to draw readers' attention to a more extensive discussion of our data contained in recent correspondence on this matter2 and to readdress the main points in the letter. Goodlad and Wright raise a pertinent question regarding the efficiency of hydroxyurea as an inhibitor of semiconservative DNA synthesis, which we have considered ourselves.3 They imply that the marginal rise in tritiated thymidine incorporation is due to an induction of normal cell proliferation by omeprazole which may be concealed within the inherent variation of the assay. It is possible to test for this by calculating the uptake of radiolabel that could be expected if 5% of semiconservative DNA synthesis is insensitive to hydroxyurea. The calculated value can then be compared with the actual counts observed in the presence of hydroxyurea. Whereas in control animals observed and expected values agree closely, in animals treated with omeprazole at a dose of 100 mg/kg the observed values exceed the expected values in every case, amounting to a mean incorporation of radiolabel that is 2 - 3 times greater than expected (p