Dose-response relationships epidemiological studies

According to EPA (IRIS 1999), the available human epidemiological studies lack quantitative exposure data for lead and for possible confounding exposures (e.g., arsenic, smoking). Cancer excesses in the lung and stomach of lead-exposed workers that are reported are relatively small, dose-response relationships are not demonstrated neither is there consistency in the site of cancers reported. EPA (IRIS 1999) concluded that the human data are inadequate to refute or demonstrate the potential carcinogenicity of lead exposure. 

Pharmacokinetics has played a crucial and somewhat unusual role in the assessment of health risks from methylmercury. Some of the epidemiology studies of this fish contaminant involved the measurement of mercury levels in the hair of pregnant women, and subsequent measurements of health outcomes in their offspring (Chapter 4). Various sets of pharmacokinetic data allowed estimation of the level of methylmercury intake through fish consumption (its only source) that gave rise to the measured levels in hair. In this way it was possible to identify the dose-response relationship in terms of intake, not hair level. Once the dose-response relationship was established in this way, the EPA was able to follow its usual procedure for establishing an RfD (which is 0.1 ag/(kg b.w. day)). 

We still lack an adequate dose-response relationship for humans exposed to ozone, particularly at concentrations less than about 0.2 ppm. The data base for the development of such a relationship for both short-and long-term exposures is inadequate. Although some data from controlled studies are available for concentrations above 0.3 ppm, methods for extrapolating to lower concentrations are needed. Moreover, it is not clear how to weight the results of pulmonary function tests on humans, animal studies, and epidemiologic studies in a general dose-response relationship. 

There is a class of curvilinear dose-response relationships in toxicological and epidemiological studies that may be described as U-shaped or J-shaped curves. Other terms such as biphasic, and more recently hormesis, have been used to refer to paradoxical effects of low-level toxicants. In brief, these dose-response curves reflect an apparent improvement or reversal in the effect of an otherwise toxic agent. These... 

For any hazardous substance, estimates of the relationship of dose to response in humans are based on either animal or human data. For example, only about 20 of the approximately 300 chemical carcinogens regulated by EPA have dose-response relationships based on human data from epidemiologic studies the remainder are based on animal bioassays. In contrast, the dose-response relationships for radiation are based primarily on the results of human epidemiologic studies. 

UCL takes into account measurement uncertainty in the study used to estimate the dose-response relationship, such as the statistical uncertainty in the number of tumors at each administered dose, but it does not take into account other uncertainties, such as the relevance of animal data to humans. It is important to emphasize that UCL gives an indication of how well the model fits the data at the high doses where data are available, but it does not indicate how well the model reflects the true response at low doses. The reason for this is that the bounding procedure used is highly conservative. Use of UCL has become a routine practice in dose-response assessments for chemicals that cause stochastic effects even though a best estimate (MLE) also is available (Crump, 1996 Crump et al., 1976). Occasionally, EPA will use MLE of the dose-response relationship obtained from the model if human epidemiologic data, rather than animal data, are used to estimate risks at low doses. MLEs have been used nearly universally in estimating stochastic responses due to radiation exposure. 

The prime objective in epidemiologic studies is to associate particular exposures with potential health effects and thus to define cause-effect relationships. Since this process is an indirect assessment, it is highly dependent on the accuracy and specificity of observations recorded both for exposure and outcome. It is a more powerful study if dose-response relationships can be shown, that is, if increasing levels of exposure are associated with increasing frequency of the health effects in individuals. 

The 2 boxes in the upper right corner of Figure 5.11 reflect a situation in which the mixture of concern is well characterized, for example, because a lot of information is available about its composition, its origin, and its dose-response relationship. A well-characterized mixture can be thought of as a commonly occurring mixture with a stable chemical composition, which is more or less known, for example, coke oven emissions. It is often infeasible to determine the exact chemical composition of the mixture at hand because the mixture contains hundreds or thousands of different components. This is also unnecessary because dose-response data on the mixture of concern are available from previous studies, for example, epidemiological data on coke oven emissions. A mixture is also considered well characterized if it can be... 

The clinical and epidemiological evidence is summarized below. The deficiencies in the studies include the lack of appropriate sampling techniques, exposure determinations, mortality standards, and other aspects of experimental design or methodology. Additionally, intermittent exposures to benzene made it difficult to assume that the average concentrations of benzene measured in a workplace actually indicated the true exposure experienced by each worker (Goldstein 1985). A cause-effect relationship between benzene and leukemia is sufficiently clear however, there are few data from which dose-response relationships can be established. 

For both humans and laboratory animals, one cannot currently distinguish between a radiation-induced cancer and a spontaneously occurring cancer (i.e., from an unknown cause). Therefore, statistical methods are used to determine whether radiation exposure is associated with an increase in cancer in a given study population. There have been several epidemiological studies in which definite dose-response relationships have been established for radiation-induced cancers. The best studied populations include atomic bomb survivors, Tinea capitis irradiation patients, ankylosing spondylitis irradiation patients, radium dial painters, radium therapy radium-224 patients, Thorotrast patients, uranium miners, Chernobyl fallout victims, and Mayak plutonium facility workers. 

These concerns are based on some epidemiological studies whose results were barely significant statistically and of questionable biological importance. (Their results lacked dose-response relationships, and were inconsistent and ambiguous. Further whether inanimate talc particles can translocate from the perineum to the ovaries, a precondition if they were to cause ovarian cancer, remains unresolved.)... 

For those forms of toxicity for which the threshold hypothesis may not hold - the important case being carcinogenicity - the problem of high-to-low dose extrapolation is more bewitching. Recall that in typical animal cancer studies, and in epidemiology investigations as well, excess lifetime cancer risks lower than about 10% (1/10) are undetectable. What dose-response relationship holds below the region of direct observation And why should we care ... 

---