Cancer in belugas from the St. Lawrence Estuary. (Correspondence).

In their paper published in the March 2002 issue of EHP, Martineau et al. (1) reported the results of 100 necropsies of St. Lawrence Estuary (SLE) belugas found dead, drifting, or stranded along the shoreline during 1983-1999. They gathered a remarkable set of pathology information on these dead animals and are to be commended for their work.

In their report, Martineau et al. (1) used epidemiologic techniques to compare cancer rates among the belugas with those of other animal species and humans. Because their work involved a cohort of dead animals found on shoreline, comparison with cohorts of live or hunted animals or with human statistics is indeed challenging. These comparisons raise the questions of whether the information collected is representative of the source population and whether the referent groups used for the comparisons are valid.

In Figure 2 of their paper, Martineau et al. (1) compared the age structure of SLE belugas found dead with that of northwest Alaska (NA) belugas obtained through subsistence hunting. They concluded,

   Thus, the age structure of SLE beluga dying of natural causes appears
   clearly different from that of NA beluga, principally because most SLE
   belugas die at an earlier age.

We do not concur with this interpretation. The age distribution of the NA belugas that died of hunting and the age distribution of belugas found stranded on the SLE shoreline are two very different things. To conclude that SLE belugas died at an earlier age, one would need to know the age structure of the two populations--the NA belugas and the SLE belugas--and the death rate by age categories in each of them. The authors did not have this information and therefore could not state that the "SLE belugas die at an earlier age."

Martineau et al. (1) estimated that the annual cancer mortality rate of the SLE belugas was 570 per 100,000, a crude rate. It would have been more informative to present age-specific cancer rates. The crude rate may simply be reflective of the age distribution of the beluga population. If a large proportion of the population is in the age groups with elevated cancer rates, the overall cancer rate will be high. Comparing these crude rates to those of other animal species or humans without taking the age distribution of the population into consideration is bound to lead to erroneous interpretations.

In their Figure 4, Martineau et al. (1) compared the annual cancer rate in the SLE belugas with that of humans and domestic animals. They concluded that

   ... The AAR [adjusted annual rate] in SLE beluga is much higher than that
   observed in cattle, horse, and sheep examined in veterinary hospitals,
   higher than the rate observed in dogs and cats examined in veterinary
   hospitals, and higher than the rate in humans.

Here again, we do not concur with this interpretation. On one hand, the comparisons are made without considering the age distribution of these groups, and on the other hand, the source of information, the profile of life and death of these groups, and the way the information was collected for each comparison groups (sheep, horses, cats, dogs, cattle, and humans) are totally foreign to the life profile of the SLE belugas and the way the information about them was obtained. These were simply not comparable data; thus the authors' conclusion is not sustainable.

Martineau et al. (1) referred to our studies on bladder and lung cancers among aluminum workers and the association that we reported of these cancers with exposure to polycyclic aromatic hydrocarbons (PAHs) in the workplace (2,3) as an argument in support of their hypothesis that cancer in belugas was caused by PAHs in the sediments of the river. We are not sure of the relevance of our study findings to their hypothesis because the exposure that led to increased lung and bladder cancers in workers involved the inhalation of high concentrations of airborne PAHs over several years.

Martineau et al. (1) referred to a higher incidence of stomach, digestive tract, and breast cancer in members of the Saguenay population not working in the aluminum industry, and inferred that this may be related to drinking water that they claim is contaminated with PAHs. We have reservations on these views. First, the referred statistic may be inflated because it does not exclude workers. Second, the hypothesis of a relationship between high incidence of cancers of these sites and drinking water seems highly improbable in light of the confirmation of the Ministry of Environment of Quebec that the norm of PAHs (benzo[a]pyrene) (10 ng/L) is present in untreated water as well as in water from the filtration plants (4). Third, to date, no association has been observed between PAH and stomach cancer among aluminum workers, even if they drank the local water and worked in the aluminium industry environment.

In spite of our reservation concerning the epidemiologic aspects of this paper by Martineau et al. (1), it remains an important contribution on the pathological aspects of cancer in wildlife.

REFERENCES AND NOTES

(1.) Martineau D, Lemberger K, Dallaire A, Labelle P, Lipscomb TP, Michel P, Mikaelian I. Cancer in wildlife, a case study: beluga from the St Lawrence Estuary, Quebec, Canada. Environ Health Perspect 110:285-292 (2002).

(2.) Armstrong B, Tremblay C, Baris D, Theriault G. Lung cancer mortality and polynuclear aromatic hydrocarbons: a case-cohort study of alumninum production workers in Arvida, Quebec, Canada. Am J Epidemiol 139:250-262 (1994).

(3.) Tremblay C, Armstrong B, Theriault G, Brodeur J. Estimation of risk of developing bladder cancer among workers exposed to coal tar pitch volatiles in the primary aluminum industry. Am J Ind Med 27:335-348 (1995).

(4.) Bleau H. Teneur en BPC, HAP, Dioxines et Furannes de quatre rivieres du Saguenay a la suite des inondations de juillet 1996 [in French]. Vecteur Environ 33(6):67-71 (2000).

The authors are currently involved in the conduct of a mortality study of the aluminum workers from the Saguenay region as a fellow-up to their previous studies. The study is financed to G. Gibbs by Alcan Inc.; C. Tremblay and G. Theriault are consultants on the project.

Gilles Theriault
Department of Occupational Health
McGill University
Montreal, Quebec, Canada
E-mail: [e-mail]

Graham Gibbs
Safety Health Environment International
Consultants Corp.
Devon, Alberta, Canada

Claude Tremblay
National Institute of Public Health
Quebec, Quebec, Canada