Can human's best friend cause cancer?.

Do dogs harbour risk factors for human breast cancer?



Introduction

Environmental and genetic factors have been identified in many types of cancers. Genetic risk factors in breast cancer are mutations of BRCA1 and BRCA2 genes, the Li-Fraumeni syndrome associated with p53 tumor suppressor gene mutations and the ataxia-teleangiectasia gene. Only 5–10% of all breast cancers are hereditary in nature. Several environmental risk factors have been proposed for human sporadic breast cancer including mouse mammary tumor virus (MMTV)-like sequences, Epstein-Barr virus and human papilloma virus (HPV). Breast cancer presents many perplexing features. For example, the hormones obviously influence disease and disease course because male breast carcinomas account only for 1% of breast cancer cases; there is also some influence on disease susceptibility of early/late menarche, early/late menopause and childbearing before 30/after 30 years of age. Estrogen-receptor negative tumors have, independently of tamoxifen treatment, a less favorable prognosis than estrogen-receptor positive tumors. Furthermore, the outcome in male breast cancer is worse than in females, despite treatment with tamoxifen, although most male breast carcinomas are estrogen-receptor positive. Similarly confusing, bilateral oophorectomy has been shown to equal the benefit of tamoxifen therapy, but the incidence of breast cancer increases significantly in postmenopausal women with increasing age, despite silent ovaries. Although diagnostic procedures and systemic and adjuvant therapeutic modalities have provided progress, much more could be achieved through the identification of environmental risk factors.

One tantalizing risk factor found in animals is the mouse mammary tumor virus (MMTV), which is an oncogenic B retrovirus inducing breast cancer in mice that can be isolated either as endogenous or as exogenous virus. More than 30 years ago, Howard and Schlom isolated mouse MMTV variants by serial passage that could productively replicate in canine, murine, feline, rat and human cells. Several groups have focussed on different aspects of MMTV and possible links to human breast cancer. There are many reports from different groups identifying MMTV-like sequences that are highly expressed in human breast cancer, although two groups could not find such sequences. Amazingly, the presence of MMTV-like DNA in breast carcinomas of dogs or in their genome has not been investigated, despite a high frequency of breast carcinomas in dogs.

Recently, a population-based incidence of malignant breast tumors was reported to be 53.3% in 14,401 Norwegian dogs. Another microbial factor could be bartonella species. Bartonella henselae, for example, has been Identified as cause of cat scratch disease but is also found in dogs. Bartonella henselae uses the same strategies as helicobacter pylori to infect cells and to prevent cellular apoptosis.

The mosaic of published results and of own observations lead us to propose that dogs may harbour carcinogenic risk factors:

Hypothesis

Dogs harbour carcinogenic risk factors for human breast cancer
(1) Dogs (and maybe other pets) harbour and transmit MMTV or MMTV-like viruses that can induce human breast cancer.
(2) Dogs (and maybe other pets) transmit non-MMTV viral or bacterial cofactors promoting the development of breast carcinoma.
(3) Several microbial cofactors of dogs and MMTV-like viruses induce human breast cancer in a synergistic manner.

Patients, materials and methods

Patients

Sixty nine female patients with breast carcinoma from southern Germany, age 30 and older, were interviewed about keeping household pets. We wanted to know whether they had pets (dogs, cats or other animals) in the moment of consultation, whether and how long they ever had pets in their lifetime and whether they had/have any contact with pets.

The patients gave informed consent to use the obtained information for publication, in accordance with the declaration of Helsinki, including updates from Tokio 1995, Hong Kong 1989, Sommerset West 1996 and Edinburgh – Scotland 2000.

Statistics

Data from all consulting patients with breast carcinoma irrespective of the stage of disease (N = 69) were compared to interview data of the public-opinion poll company Allensbach from 1320 Bavarian women of the same age group using 2 × 2 tables of the computer program SPSS (version 10.1; SPSS München).

Results

Incidence of animal keeping

We asked 69 patients with breast carcinoma seeking consultation on immunotherapy, whether they kept household pets. We compared this information with data from 1320 Bavarian women of the same age group (30 years and older), interviewed by the public-opinion poll company Allensbach. More than twice the number of patients kept dogs as compared to control individuals. Precisely, 37.7% of the patients (26 out of 69) kept a dog at the time of consultation and throughout the previous 10 years, compared to 14.8% (195 out of 1320) in the control population at the time of the interview (χ2 = 25.72). Another 15.9% of patients with breast carcinoma kept a dog for at least one year during the 10 years preceding their diagnosis of breast carcinoma, but they did not have a dog at the time of consultation. These patients were not entered in the statistical evaluation because of the lack of proper controls; nevertheless these patients had many chances to come into contact with a putative contagious carcinogenic factor. Also not included in the statistical analysis were another 16 patients (23.1%), who had intensive contact with dogs, either as teenagers because their fathers were foresters or farmers, or because their husbands were hunters, etc. Two additional patients not having regular contacts with dogs had received dog bites. Taken together, 79.7% of all patients had intensive contact with dogs before the diagnosis of breast carcinoma encompassing a time period up to 30 years. Whether other pets might also confer some risk requires a detailed epidemiological study. Patients kept cats as often as the control population (20.6% of patients, 21% of controls. However, only 4.4% of the patients (N = 3) did not have pets at any time, as compared to 57.3% (N = 756) of the control population at the time of their interview. This would result in a high risk (29-fold) for pet owners but would require knowledge on long-term ownership in the control group.

Discussion

We found a significantly increased frequency of dog owners among female patients with breast cancer as compared to an age matched group of the female population (χ2 = 25.72). In contrast, female patients kept cats not more often than female controls. Howard and Schlom isolated mouse MMTV variants by serial passage that could productively replicate also in canine and human cells. Can we equate the incidence of breast cancer in dogs with the presence of MMTV? Since approximately half of the Norwegian dogs had breast cancer, approximately half of the dogs would carry MMTV or a MMTV-like virus, assuming that MMTV or MMTV-like virus was the main carcinogenic factor.

Differences in tumor incidence were noted in different breeds of dogs, suggesting that additional factors such as susceptibility to infection, can influence outcome. In human, the prevalence of MMTV-like sequences, which were found in breast cancer tissues but not in normal breast tissues of the same individuals, also varied between different ethnic groups. This would fit best with the model of an infectious exogenous agent and immunogenetic factors.

Why are MMTV-like sequences not present in all breast carcinoma patients? Immunogenetic analyses in cervical carcinomas revealed that susceptibility to or resistance against certain HPV types depended on certain HLA constellations.

Immunogenetic factors may also influence susceptibility to MMTV. In Western countries, a high incidence of human breast cancer has been found to correlate with a dominant occurrence of a particular mouse strain, Mus domesticus. Based on this correlation, a transmission of MMTV from this mouse strain to man was proposed. Information is not available on a particular susceptibility of this mouse strain for MMTV or a higher frequency of breast cancer in this mouse strain. What would be the route of transmission to man and particularly to those with a higher living standard? It is plausible to assume that an animal with frequent human contacts like the dog could be the link between mouse and man and transmit MMTV or a MMTV-like virus to man after they themselves had acquired the infection. Dogs themselves would not need to develop breast cancer themselves in order to spread the virus. Dogs follow the footsteps of animals with the nose on the ground and could thereby inhale infectious excretions, including those from mice. It was demonstrated that nasal-associated lymphoid tissue of adult mice can act as entry site for the mouse mammary tumor virus. If we accept dogs as prime transmitters of MMTV or MMTV-like viruses to man, another observation can be explained: the increased susceptibility to breast cancer of oriental women who move to Western countries.

Asian or Oriental women seldom keep dogs as pets, rather serve dogs as one source of meat. Migration to Western countries may change life style, including the keeping of pet dogs. This might contribute to the increased incidence of breast cancer in these women after their move to Western countries.

Although we assume that MMTV-like sequences will be found in canine breast carcinoma, other microbial factors from the dog like chlamydiaceae or bartonella could suppress peptide presentation via the MHC. Such suppression could interfere in MMTV-peptide presentation and deeply influence the multistep process of carcinogenesis in human breast cancer. Both chlamydiaceae and bartonella can survive as intracellular parasites after infection of antigen presenting cells (APC). We found a significantly increased frequency of both chlamydiaceae and bartonella infections in breast cancer patients (B.Fellerhoff, R.Wank, personal communication). Both chlamydiaceae and bartonella have been identified in dogs.

Different combinations of carcinogenic factors may promote the development of breast carcinomas. Whether one or several microbial factors transmitted by dogs contribute to the carcinogenesis of breast cancer requires further research. We subsequently propose methods to test the discussed carcinogenic risk factors harboured by dogs.

Testing the hypothesis

Epidemiological and molecular investigations and a MMTV vaccine could prove or disprove our hypothesis.
(1) Informative on all three suppositions of the hypothesis could be a detailed epidemiological analysis of pets in larger panels of breast carcinoma patients and in a control population.
(2) The molecular analysis of dog breast carcinomas with primers identifying MMTV and MMTV like sequences, and a comparison of the homology shared between MMTV, the putative dog MMTV-like sequences and human MMTV-like sequences would be informative on the first supposition of the hypothesis.
(3) Development of a vaccine against breast cancer in dogs would provide proof or disproof of the hypothesised transmission of a MMTV-like, human breast cancer-inducing pathogen by dogs. Prevention of breast cancer in dogs by a vaccine against MMTV or MMTV-like viruses should be followed by a lower incidence of human breast cancer.
(4) The occurrence and types of chlamydiaceae and bartonella should be identified in human and in dogs.
(5) Simultaneous treatment with antibiotics of infected pets and pet owners should be accompanied by PCR monitoring of infections each year. This should reduce the incidence of breast cancer.

Implications

Transmission of MMTV or a MMTV-like virus or bacterial risk factors by pet dogs offers a possible explanation for the increased incidence of breast cancer in Western countries and its correlation with a higher living standard. It does not exclude pet animals or mice as additional sources of MMTV or MMTV-like viruses or bacterial risk factors. Development of a tumor vaccine against dog breast cancer would be of commercial interest and would also serve as a model for a human breast cancer vaccine. Simultaneous antibiotic treatment of pet owners and pets with chlamydiaceae and bartonella infections may also be important in order to assist immune responses against viral carcinogenic factors.

Summary

We ask consulting patients regularly whether they keep pets in order to identify zoonotic factors. It became apparent that patients with breast carcinoma (N = 69) owned significantly more often dogs but not cats compared to age matched female controls.

We compared the frequencies of dog and pet ownership with data from public available statistics on women (N = 1320) of the same age group in Bavaria. The most striking result was that more than twice the number of patients kept dogs permanently in the last 10 years and at the time of interrogation as compared to control individuals at the time of interrogation. Further internet search on the morbidity of breast carcinoma showed in dogs a protracted course of disease and metastases into lung, liver and bones, resembling the course of disease in human breast cancer. In contrast with this, breast cancer presented in cats a dramatically short course and the main but unusual location of metastasis presents in the hind legs. A recent publication in Norway reported on a high frequency (53.3%) of breast carcinomas in 14,401 investigated dogs. Which transmissible factor or factors come into question? Variants of the mouse mammary tumor virus (MMTV) can productively replicate in human cells and in different animals, including dogs. Many investigators, but not all, could identify MMTV-like sequences in sporadic human breast cancer.

MMTV or MMTV-like sequences have not been investigated in canine breast carcinomas until now. It is also conceivable that other microbes from the dog, for example bacteria, could participate in the first steps of carcinogenesis in human. It was recently shown that bartonella species promote vascularization and prevent apoptosis of infected cells with the same methods as helicobacter pylori. Our considerations require further research. Epidemiologic cohort studies and identification of potential carcinogenic microbial factors will prove or disprove our hypothesis that risk factors from dogs could contribute to the carcinogenesis of human breast cancer.