Commentary: Preventable causes of gastric cancer may also operate in adult life

David Coggon

MRC Environmental Epidemiology Unit, Southampton General Hospital, Southampton SO16 6YD, UK.

‘Jack Spratt would eat no fat, his wife would eat no leanAnd so between the two of them they licked the platter clean.’

The children's rhyme about Jack Spratt and his wife will be familiar to British readers, but Nanni and colleagues believe that married couples such as the Spratts are an exception, at least in northern Italy.1 They assume that during adult life the exposures of husbands and wives to dietary and other risk factors for stomach cancer will generally be similar. Thus, they interpret their observation that spouses of stomach cancer cases experience mortality from the disease similar to that of the general population as evidence that the main events in gastric carcinogenesis occur before adult life. The implication is that the search for effective means of prevention should focus on the environment in childhood, or perhaps even earlier.

Certainly there is strong evidence from migrant studies that major risk factors for stomach cancer, or at least for the histological subset of intestinal-type tumours, operate at young ages.2,3 One such factor is likely to be infection by Helicobacter pylori. Several prospective cohort studies have indicated that H. pylori infection carries an increased risk of later gastric cancer.4–6 Furthermore, prevalent H. pylori infection in adults has been linked with domestic crowding in childhood;7 and in England and Wales, mortality from stomach cancer has been correlated geographically with crowding in homes some 40 years earlier.8

However, it does not necessarily follow from Nanni's findings that important preventable causes for stomach cancer do not act also during adult life. It may just be that, in the Italian population studied, these risk factors did not vary greatly from person to person, or at least no more than between husbands and their wives. Indeed, case-control studies have found quite consistently that higher consumption of fresh fruit and vegetables as an adult is associated with a lower risk of gastric malignancy.9

Two other conceptual points from Nanni's paper are worth highlighting. One is the reference to a long induction period being a feature of stomach cancer. Strictly, an induction period is not a characteristic of a disease, but of the relationship between a disease and one of its causes. Thus, the same disease may have one cause that acts with a long induction period, and another that alters risk within a short time from exposure.

The other contentious point is the assumption that people who have died from stomach cancer must have been heavily exposed to risk factors, since their cumulative risk of death from the disease was 100%. It is correct that they must have been exposed to a combination of causes sufficient to induce the disorder, but we should not assume that these causes are necessarily identifiable or measurable risk factors. They could, for example, be unobservable ‘chance’ events at a molecular level. An analogy can be drawn with the tossing of a coin. The fact that a coin has come down heads does not imply that it must have been heavily exposed to any risk factors for this outcome. Rather, the observation may reflect a combination of a complex set of circumstances (the dimensions, shape and weight of the coin; the exact forces acting on it and its height above ground when tossed; the viscosity of the air and wind speed; the contour and rigidity of the surface onto which it falls etc), none of which on its own could be shown to have a consistent impact on whether coins come down heads or tails.

As a corollary, the fact that the causes of a disease are only ‘partially understood’ and that we do not know why some people get it and others do not (even when their exposure to established risk factors appears identical), does not necessarily imply that there must be other major causes waiting to be identified. Of course, it may still be worth looking for such causes in case they exist.

References

1 Nanni O, Zoffoli G, Scarpi E et al. Gastric cancer mortality in the spouses of patients who died from gastric cancer. Int J Epidemiol 2002; 31:468–72.[Abstract/Free Full Text]

2 Coggon D, Acheson ED. The geography of stomach cancer. Br Med Bull 1984;40:335–41.[ISI][Medline]

3 Coggon D, Osmond C, Barker DJP. Stomach cancer and migration within England and Wales. Br J Cancer 1990;61:573–74.[ISI][Medline]

4 Forman D, Newell DG, Fullerton F et al. Association between infection with Helicobacter pylori and risk of gastric cancer: evidence from a prospective investigation. BMJ 1991;302:1302–05.[ISI][Medline]

5 Parsonnet J, Friedman GD, Vandersteen DP et al. Helicobacter pylori infection and risk of gastric carcinoma. N Engl J Med 1991;325:1127–31.[Abstract]

6 Nomura A, Stemmermann GN, Chyou PH, Kato I, Perez-Perez GI, Blaser MJ. Helicobacter pylori infection and gastric carcinoma among Japanese Americans in Hawaii. N Engl J Med 1991;325:1170–71.[ISI][Medline]

7 Mendall MA, Goggin PM, Molineaux N et al. Childhood living conditions and Helicobacter pylori seropositivity in adult life. Lancet 1992;339:896–97.[ISI][Medline]

8 Barker DJP, Coggon D, Osmond C, Wickham C. Poor housing in childhood and high rates of stomach cancer in England and Wales. Br J Cancer 1990;61:575–78.[ISI][Medline]

9 Palli D. Epidemiology of gastric cancer: an evaluation of available evidence. J Gastroenterol 2000;35(Suppl.12):84–89.[Medline]





This Article
Extract
FREE Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Request Permissions
Google Scholar
Articles by Coggon, D.
PubMed
PubMed Citation
Articles by Coggon, D.