In the last years infection from Herpes virus 8 (HHV8) has been often found in Kaposi’s Sarcoma patients so that a lot of epidemiologic studies have been performed to study in depth this association that can be useful to clarify also KS etiology. HHV8 has been studied also in KS patients and their family members: the very low level of transmission of this virus was assessed. A subsequent preliminary study, anti-HHV8 antibodies were found in 95% of the patients, in 39% of their family members and in 11% of the controls. This last value between controls resulted anyway high with respect to the rest of Italy where it is equal to 4%.
So a further study was done in North Sardinia on broad population samples: it was shown that HHV8 prevalence rose where KS incidence was higher and, analogously to KS, it was greater for subjects older than fifty. HHV8 seemed not to be associated with sex, while KS is more frequent in males. Anyway the main result was that seroprevalence for the general population in North Sardinia was higher than previously reported, with a ratio of 3:1 with respect to Central Italy. It has to be clarified how much these facts can be genetically explained by selection of characters predisposing to both classic KS and HHV8 infection and how much can be due to environmental factors.
The main task of Verona unit is to prepare the study design, to standardise the methods, to collect and analyse survey data in evaluating the different contribution of genetic and environmental factors in HHV8 infection.
In epidemiologic research, especially when studying risk factors for cancer, migrant studies are shown to be very important and extremely useful to understand the influence of environment and behaviour on development of cancer pathology. One of the first studies of this type was performed on stomach and colon cancer of Japanese migrants living in California: incidence of stomach cancer, very high in Japan, decreased in the first and, more, in the second generation of Japanese migrants; the opposite trend was observed for colon cancer rates: from low values they reached Californian incidences. This phenomenon can be obviously explained by the loss of bad dietary behaviour on one side and the loss of protective behaviour on the other. However, in spite of these findings, the hypothesis of an ethnic susceptibility for stomach cancer could not be completely excluded, but further deepening was required.
So it can be very useful to study Sardinian migrants living in Italian mainland and Europe: if the prevalence of HHV8 in migrant communities is similar to Sardinina’s prevalence, genetic contribution will be considered strong; while if HHV8 infection is less frequent, this will imply a main role of environmental and behavioural factors.
The Sardinian population is ideal for this type of extended study considering its long history of emigration. Actually in the 50’s, due to crisis in the agricultural and shepherding sector, this emigration became a veritable tide mainly towards other Italian region and, to a less extent (25%), abroad. Those who emigrated to other European countries and to the industrialised North Italy saw a complete change in life style both in terms of occupation and environment; on the other hand, those emigrating in Central Italy continued to practice animal breeding and to conduct a life very similar to that they left behind in Sardinia.
The aim of the present research is to estimate HHV8 prevalence in Sardinian migrants to possibly correlate it with environmental risk factors and to evaluate the share due to genetic components. To reach this aim, the study will be developed as follows:
1) Choice of representative geographic areas regarding the different situations of Sardinian emigrations. Diverse communities will be selected: one community of an European country, one living in Northern Italy where migrants live in an industrial contest completely different from the rural one they came from; another community will be chosen in Central Italy where a way-of-life similar to the usual in Sardinia is maintained.
2) Determination of the number of subjects to be included in the sample and stratification by age and gender. Depending on the number of subjects in the communities, sampling fraction shall be decided; male/female ratio shall be 1:1 as no evidence of an association between HHV8 and gender has been shown till now. As HHV8 infection is more frequent in people older than 50, this shall be kept into account in the sample choice. Once chosen the most representative communities and once identified the eligible subjects (Sardinian migrant of first and second generation), random samples stratified according to what written above will be drawn.
3) Tools for collecting personal and environmental data (questionnaires) will be developed. Pilot studies to evaluate such tools will be carried out. Emigrants could be identified by their nationality when they keep it, and by their place of birth for the first generation. Surname can also be included among the criteria for identification.
4) Setting up of a data-base for data collection from the different communities. The free statistical package EPI INFO (OMS) shall be given to each centre together with the appropriate mask for data entering; the collected data will be sent to this centre for control and analyses. EPI INFO allows for a first automatic control during data entering and furnishes also a basal program of statistical analysis (ANALYSIS) useful to perform a second deeper check on the quality of collected data, so that corrections to the centers can be required, if needed. For the true statistical analysis, more advanced software as SPSS, STATA, S-PLUS will be used.
An Internet site will be set up to allow quick information exchange between participating centers and for the easy monitoring of the successive phases of the surveys.
5) Development of specific models for migrant studies. Migrant studies are based on the idea that emigrants are, by their life style and culture, exposed to risk factors which differs from those prevailing in the country of origin. Thus evidence for risk levels specific to immigrants can indirectly suggest or confirm etiological hypotheses. In general, the risk to which immigrants are subject is recognised by comparison with the host country or with the risk in the country of origin: this second chance will be used in this survey, even if a comparison with the host country could be done when there are available data. Moreover, the distinction between first-generation emigrants and second-generation emigrants can provide useful information on the effects of behaviour changes resulting from the cultural integration, that acts more profoundly in the second generation.
To obtain risk estimates, ratios between HHV8 prevalence, standardised by age and gender, of the different community respect to the Sardinian ones will be calculated.
The opportunity of applying to the data log-linear modelling based on Poisson distribution will be considered.