DESCRIPTION OF WORK PACKAGES IN THE PROJECT

WP1 - Management and communication
WP Leader: Antony Young of KCL

WP 2 - Personal dosimeter
WP Leader: Jakob Heydenreich of BBH

With EU funding (ENV 4), Participant 2 (BBH) developed a personal, electronic dosimeter “SunSaver” as shown below. The dosimeter comprises a sensor and a data logger. It is housed together with a digital watch and serves as a wristwatch. A Silicon Carbide Photodiode (JECF1-IDE; Laser Components; Germany) was chosen as sensor, which is only sensitive in the range 200-400 nm. The sensor has a built-in diffuser and has cosine response. The spectral response is similar to the CIE erythema action spectrum (McKinlay and Diffey, 1987). The data logger controls the sensor which was set to measure every 8th second and to store an average of the last 75 measurements every 10 minutes along with the time. The measurement range of the dosimeter is 0.1 SED/hour to 23 SED/hour. The UVR dosimeter is battery driven. It can run for 145 days without maintenance, and the data can be transferred to a personal computer. This UV dosimeter in an updated version will be used in the field studies.

Figure 2: Personal electronic UVR dosimeter “SunSaver” worn on the wrist. 1: The housing, 2: The UVR data logger with the sensor and battery, 3: The watch, a separate unit (Heydenreich J, Wulf HC. Miniature personal electronic UVR dosimeter with erythema response and time-stamped readings in a wristwatch. Photochemistry and Photobiology 2005;81(5):1138-44).

McKinlay AF, Diffey BL. A reference spectrum for ultraviolet induced erythema in human skin. CIE J 1987;6:17-22.

Time-stamped UVR measurements and time-stamped diary data including behaviour and ground surface information will be used together with albedo, temperature and weather data obtained from WP 4. This work package will also establish a quality assurance system to validate the data and a database to process the data so that it may be used in modelling. These data are of importance for all the population groups examined in WP 3 and for the evaluation of the effect of UVR on biomarkers in WP 5 and health outcome in WP 7. The daily and long-term dose distribution will be of fundamental importance for health risk assessment (WP 7) and the modelling based on individual exposure data (WP 6).

WP 3 - Population UVR exposure
WP Leader: Elizabeth Thieden of BHH

Objective: To carry out population UVR exposure studies in 4 European countries in people in different work and leisure situations, and in children

Earlier personal UVR dosimeter studies have shown huge individual ranges of UVR exposure. Some people get many fold higher or lower UVR doses than their peers (Diffey et al, 1996; Thieden et al, 2004 - 2006). To study the individual’s relationship between sun exposure behaviour and UVR doses it is therefore important to conduct UV dosimeter studies measuring time-stamped UVR doses (WP 2) continuously over a longer period as a summer season to be able to investigate what the great individual differences imply in relation to the UVR-induced biomarkers in WP 5 and to the health risk in general (WP 7). Since outdoor working is a high UVR risk activity we have chosen full-time farmers as the example of outdoor workers. For comparison we will include the farmers’ spouses, who often work outside the farm and 1-2 children who attend school in 4 countries. This gives us the opportunity to follow and compare a broad range of children, adolescents and adults and to identify what cultural, geographical as well individual behavioral differences implies in relation to UVR exposure.

Self reported Skin Type according to (Fitzpatrick, 1988) will be collected from all volunteers. Objective and reliable measures of individual sun sensitivity, expressed as the pigment protection factor (PPF), will also be assessed by a skin reflectance meter designed by Participant 2 (UV-Optimize Model 666, Chromo-Light, Denmark) (Lock-Andersen et al, 1998). It measures the pigment protection factor (PPF), that equals the number of SED needed to elicit just perceptible erythema on unexposed buttocks but can also be used to indicate the UVR sensitivity on other body locations. The PPF values will be measured on the volunteers before and after the studies. We will use the PPF measured on the buttock to express the constitutive skin type. While the PPF on the shoulder will be used as an indicator of the pigmentation acquired during sun risk behaviour as the shoulder is a body part considered only UVR exposed during sun risk behaviour.

In addition we want to investigate the acute effect of UVR on biomarkers, i.e. DNA damage (WP 5), in short term studies of high UVR risk leisure activities at the beach and during skiing. In addition these studies will allow us to monitor the influence of different albedo (WP 4). The data derived from the dosimeters and corresponding diaries in the field studies together with the Biomarker data will form the basis that enable the development of a humanized radiative transfer based model to assess the future impact on UVR exposure (WP 6).

Diffey BL, Gibson CJ, Haylock R, McKinlay AF. Outdoor ultraviolet exposure of children and adolescents. British Journal of Dermatology 1996;134(6):1030-4.

Fitzpatrick TB. The Validity and Practicality of Sun-Reactive Skin Type-I Through Type-Vi. Archives of Dermatology 1988;124(6):869-71.

Lock-Andersen J, Drezewicki KT, Wulf HC. The measurement of constitutive and facultative skin pigmentation and estimation of sun exposure in Caucasians with basal cell carcinoma and cutaneous malignant melanoma. British Journal of Dermatology 1998;139:610-617.

Thieden E, Philipsen PA, Sandby-Møller J, Heydenreich J, Wulf HC. Proportion of lifetime UV dose received by children, teenagers and adults based on time-stamped personal dosimetry. Journal of Investigative Dermatology 2004;123(6):1147-50.

Thieden E, Philipsen PA, Heydenreich J, Wulf HC. UV radiation exposure related to age, sex, occupation, and sun behavior based on time-stamped personal dosimeter readings. Archives of Dermatology 2004;140(2):197-203.
Thieden E, Collins SM, Philipsen PA, Murphy GM, Wulf HC. Ultraviolet exposure patterns of Irish and Danish gardeners during work and leisure. British Journal of Dermatology 2005;153(4):795-801.

Thieden E, Philipsen PA, Wulf HC. Ultraviolet radiation exposure pattern in winter compared with summer based on time-stamped personal dosimeter readings. British Journal of Dermatology 2006;154(1):133-8.

WP 4 - Satellite and ground station data
WP Leader: Paul Eriksen of DMI

Objective: To provide ambient UVR measurements, environmental and climatological data that will be incorporated with the personal UVR exposure data of WP 3 and used in the modelling of WP 6

A key objective of the project is to model personal exposure. It is therefore necessary to be able to model UV irradiance with as high a degree of confidence as possible and thus necessary to describe the atmosphere (at the location of exposure) in as much detail as possible. There are two important personal dosimeter exposure scenarios in the project: in one scenario (field study dosimeter measurements) the exposure is obtained by a small group of persons at a rather specific geographic location over a short period of time (a week), and in the other scenario the exposure is obtained by a large group of persons in a long period of time (several months). For the first scenario it is possible to accurately describe the atmosphere at the specific locations by taking detailed measurements in order to support the exposure modelling, i.e., by providing detailed input to the model, whereas in the latter scenario we will have to rely on satellite measurements of as many parameters as possible together with as many locally measured parameters that may be available, typically from weather services and available data bases, e.g., Geomon, NDACC, WMO. To best support the modelling of exposure in the dosimeter field study measurements we will include measurements taken with instruments brought to the specific locations (ground station dosimeters, spectroradiometers and different radiometers). These Ground Station data are crucial for the exposure modelling because we need detailed measurements of irradiance, particularly on inclined surfaces, in order to develop the personal exposure model.

WP 5 - Biomarkers
WP Leader: Antony Young of KCL

Objective: To measure biomarkers of the adverse and beneficial effects UVR exposure in the population studies of WP 3 and correlate these with UVR exposure. In addition studies will be done to determine the relationship between erythema and immunosuppression and vitamin D synthesis

Solar UVR causes adverse and beneficial effects, It is therefore important to understand the balance between theses outcomes for any risk benefit analysis. It recognised that the formation of cyclobutane pyrimidine dimers (CPD) plays a major role in skin cancer because they are mutagenic, and the “signature mutations” of these lesions have been found in p53 on non-melanoma cancers. Furthermore, CPD appear to play an important role in immunosuppression by UVR. Most studies on the effects of UVR on skin DNA in vivo have been under controlled laboratory conditions and have been done on skin biopsies, mostly using semi-quantitative immunological techniques (i.e. monoclonal antibodies that recognize the lesions). The assessment of DNA damage in the skin requires the taking of biopsies that gives rise to ethical problems and cannot be done with children under normal circumstances. Participant 4 has developed a technique that enables CPD detection in urine that has obvious advantages. This a quantitative sensitive radiolabelling technique assessed by HPLC that was initially developed for skin and the groups of Beneficiaries 1 and 4 have worked together in the past using this technique. At, present the relationship between DNA photodamage in the skin and in the urine is not known and this relationship will be determined in WP 5. Furthermore, we do not know the relationship between the immunological and the HPLC techniques in the skin and this will also be addressed with the help of Participant 3 in one of the field studies in WP 3. Overall, these studies will result in the better definition of the relationship between real life UVR exposure and DNA damage, using a non-invasive technique.

WP 5 also addresses the relationship between UVR exposure and vitamin D status, which will be done by measuring blood calcidiol, which is a routine procedure. DNA damage and vitamin D data from volunteers with a known UVR exposure profile will enable a better understanding of the risk/benefit evaluation from UVR exposure. Sampling will be done before, during and after exposure. Apart from UVR dose, spectral quality is important in photobiological outcomes. Erythema is a widely used clinical endpoint but its spectral relationship with immunosuppression and actual vitamin D status is poorly understood. We will design experiments based on solar simulated radiation, in which we alter the spectrum with cut-off filters that selectively attenuate UVB (this in effect simulates the sun at different zenith angles). We will given equivalent erythemal exposures (in terms of SED) and determine the effects of such exposures on vitamin D status and the suppression of cell-mediated immunity using a standard contact hypersensitivity (CHS) assay. These studies will show the spectral relationship between erythema, vitamin D status and immunity and will be valuable for the interpretation of overall risk when the erythema is used as a spectral weighting function. The CHS assay will also be used in one the studies in WP 3 carried out by Participant 3. Beneficiaries 1 and 3 have considerable experience with this assay and will use a similar protocol. In essence, this assesses the ability of UVR to inhibit the normal sensitisation to an antigen (hapten). In the absence of UVR prior to sensitisation, the immune system mounts a response when the skin encounters this antigen 2-3 weeks after sensitisation. However, exposure of the skin to UVR prior to sensitisation alters the processing of the antigen and the immune response (skin swelling) to a second encounter with the antigen is diminished. Suppression of the sensitisation phase of the CHS response is regarded as a model for the immunosuppression that is important in skin cancer in mice.

Urinary thymidine dimer as a marker of total body burden of UV-inflicted DNA damage in humans. Kotova N, Hemminki K, Segerbäck D. Cancer Epidemiol Biomarkers Prev. 2005 Dec;14(12):2868-72.

WP 6 - Analysis and modelling of data from WP 2-4
WP Leader: Paul Eriksen of DMI

Objective: To develop a personalized UVR exposure model in order to model personal UVR exposure of small or large population groups taking into account the population behaviour and culture and to develop a variant of the above model that is coupled to, or imbedded in, a regional climate model (HIRHAM) to better represent radiation in a climate model and for prediction of future UVR exposure related to a changing climate
For state-of-the-art radiative transfer programs the accuracy of modelled UV irradiance is highly dependent on how accurately the atmosphere can be described: the input to the model must be the atmospheric parameters that match the real conditions as best as possible. If the atmosphere is well described the modelled UV irradiance is generally within 5-10% of accurately measured irradiance (http://www.atmos-chem-phys.net/7/2817/2007/acp-7-2817-2007.pdf) and several radiative models are available that are considered candidates to this project and which the applicants are familiar with (Koepke et al, 1998). Since an important objective of the project is to model personal exposure it is therefore necessary to be able to model UV irradiance with as much confidence as possible and it is therefore necessary to describe the atmosphere as well as possible at the location of exposure. Given the complexity of personal UV exposure, comprising both complex exposure geometry and the more subtle exposure aspects of behaviour and culture, the development of the UV exposure model must take into account both. To the best of our knowledge this has not been done before. We will try to model the personal UV exposure by breaking the input to the model into one part that describes the atmosphere and the exposure geometry and another part that describes behaviour and culture by parameterisation. The UV exposure model, or a variant of it, will finally be coupled to, or imbedded in, a state-of-the-art regional climate model (HIRHAM). This has, to the best of our knowledge, not been attempted before. This last development may perhaps not reach a final state but it will enable us to compare the results obtained from the dosimeter studies with results obtained with a “run” of the climate model. It will finally enable us to assess the influence of future climate changes on personal UV exposure and distinguish between the importance of climate and behaviour in personal exposure. This may be utilised to develop EU policies related to health impacts.

Koepke P., Bais A., Balis D., Buchwitz M., De Backer H., de Cabo X., Eckert P., Eriksen P., Gillotay D., Heikkila A., Koskela T., Lapeta B., Litynska Z., Lorente J., Mayer B., Renaud A., Ruggaber A., Schauberger G., Seckmeyer G., Seifert P., Schmalwieser A., Schwander H., Vanicek K., and Weber M.: Comparison of models used for UV index calculations, Photochemistry and Photobiology, 67, 657–662, 1998.

WP 7 - Health risk assessment of data from WP 3-6
WP Leader: Mark Nieuwenhuijsen of CREAL

Objective: To systematically review the literature on UVR exposure and adverse health effects to obtain exposure response data. To analyse data on immune response within the EC Respiratory Health Survey (ECHRS) and validate questionnaire data and, to integrate our personal monitoring and modelling (WP 3-6) data in existing epidemiological studies to estimate the degree of measurement error and any effects on the observed exposure response relationships of UVR exposure and health effects.

One of the limitations in epidemiological work has been the exposure assessment; generally surrogates for UV exposure have been used such as latitude or ambient radiation levels, without taking into account behavioural factors such as sun avoidance, sunscreen and sunglasses use, clothing etc, which may lead to differences between ambient UV levels and personal exposure and dose. Furthermore, self reported UV exposure has been used without further validation. All this may lead to exposure misclassification, or measurement error, and may mask the true variability of exposures, thus attenuating the effect of exposures in effects models. This may lead to bias as well as loss of power in epidemiological studies. Major sources of misclassification can stem from errors in estimating ambient UV radiation levels at a particular location of exposure, and inadequate assessment of the location at which individuals are exposed (ie related to activity patterns), or personal behaviour. Furthermore, in typical studies of health and environment we wish to make inference on the relationship between individual-level quantities using aggregate, or ecological, exposure data. Such ecological inference is often subject to bias and imprecision, due to the lack of individual-level information in the data. Conversely, individual-level survey data often have insufficient power to study small-area variations in health. Such problems can be reduced by supplementing the aggregate-level exposure data with small samples of data from individuals within the areas, which directly link exposures and outcomes.

The aim and objectives of this WP is to:
• systematically review the literature on UV exposure and adverse health effects, specifically skin cancer, immune suppression, cataracts and vitamin D to obtain exposure response data
• Analyse data on immune response and UV exposure within ECHRS and validate questionnaire data on UV exposure
• integrate our personal monitoring and modelling data in existing epidemiological studies to estimate the degree of measurement error and any effects on the observed exposure response relationships of UV exposure and health effects.

We will systematically review the literature on exposure to UV and health effects, specifically skin cancer, cataracts, immune response and Vitamin D (and related diseases) and obtain exposure response relationships for UV exposure and the various health outcomes. We will review recent IARC and NRPB reports together with literature searches in databases such as MEDLINE/PUBMED. We will estimate measurement error from our work on personal monitoring, biomonitoring and modelling (WP 3-6) and integrate this with exposure response data of existing studies on health effects of UV exposure that have used a variety of surrogate measures of exposure to understand better the relationships and, where possible, improve on these. We will outline a hierarchical model framework for estimating individual-level associations using a combination of a large dataset of aggregate and a small subset of individual data. The goal is to estimate the uncertainty associated with ambient UV radiation or estimates and with the use of ambient levels as a proxy for personal exposures, and to account and correct for the measurement error in exposure response models. Methods to incorporate measurement error information in health effects models can be built upon Bayesian Hierarchical frameworks developed for time series studies of air pollution by Samet et al (2000) or Holloman et al (2004), or by Jackson et al (2006). for the improvement of ecological inference using individual-level data and estimation of measurement error. We will review such existing methods, and develop a framework appropriate in the current studies of health effects of UV exposure, perform a comprehensive simulation study, under a variety of realistic conditions, to determine when aggregate data are sufficient for accurate inference, and when we also require individual-level information, and apply the method in a variety number of studies. Since there is fairly good data for skin cancer and eye, but fairly weak data for immune response we will aim to conduct an epidemiological analysis within the European Community Respiratory Health Survey (ECRHS) to assess the relationship between UV exposure, as determined by latitude, season and UV radiation models and outcomes such as IgE, allergic disease, asthma and COPD. The EC funded ECRHS is a large follow up study conducted in 34 centres in 15 countries in Europe. The initial study took place in the 1990s and a follow up too place in 2001-2003. For the study extensive data was collected on specific immunoglobulin E, allergies and respiratory disease and confounding variables from 11,215 subjects, aged 20-44 yr. Furthermore, to validate questionnaire data we will conduct a small validation study within ongoing studies of chronic obstructive pulmonary disease (COPD) patients (N=350) and a case control study on colon cancer study (N=1000) in Barcelona. We will take personal UVR measurements in a subset of the population using similar methodology to that described in WP 3. Both studies collect information on UV exposure, a supposedly protective factor in both diseases, from questionnaires and will be able to provide measurements of agreement between questionnaire data and measured personal data to be inputted in the measurement error models. The outcome will be improved risk assessment that will include personal exposure data.