CSIRO - June 1994 BIOLOGICAL EFFECTS AND SAFETY OF EMR |
5.0 CANCER STUDIESSUMMARYThere is increasing concern about the possibility that RF exposure, particularly from cellular telephones, may play a role in the causation or promotion of cancer, particularly in blood forming areas of the CNS. Experimental evidence from in vitro and in vivo studies indicates that EMR exposure does not produce adverse genetic effects and is, therefore, unlikely to have a direct effect on tumour initiation. There is, however, some evidence of subtle changes in cell behaviour and proliferation rate that could be consistent with an effect at the level of tumour promotion. In vivo studies have shown accelerated rates of growth of certain tumours when microwave exposure is applied after the initiation of known carcinogens. Stress is an important potential confounder. Dosimetry and the effects of resonant frequency, body orientation are important factors that must be carefully controlled. Cellular studies give evidence of microwave-induced neoplastic transformation. There is no convincing evidence that athermal microwave irradiation produces clastogenic effects. The potential for direct effects on DNA through enhanced absorption of microwave frequencies has not been supported by subsequent studies. Results of studies related to cancer induction by microwave or RF radiation are equivocal. Many early studies are severely limited by inadequate dosimetry and poor histopathology. Further research in this area is needed to resolve many of the controversial issues. A number of long-term exposure studies are about to begin (three are in-progress). It will be interesting to see if there is any similarity in the design that will allow effective cross-comparisons of data. 5.1 Molecular Mechanisms in Radiation Induced CancerThe relative binding energies of EMR and the organic molecular bonds in mammalian cells provide the limitation on the likelihood of cancer production by direct action on DNA. The binding energy of typical chemical organic molecules is approximately 300 kJ per molecule, i.e. 3 eV or 10-19 J per single bond (Burkart 1993). Whereas ionizing radiation particles release large energies (approx. 5 MeV, or 10-13 J) and easily damage DNA or cellular components, the relatively low energy levels in EMR fields are generally considered to be incapable of damaging DNA directly. The photon energy is about 10-3 eV at 300 GHz and decreases linearly with decreasing frequency (NRPB 1993). However, biological response to absorbed radiation energy is complex and depends on many parameters. A common by-product of radiation interaction with water molecules is the formation of chemically-reactive free radicals. With low level ionizing radiation, the activity of free radicals is thought to constitute the major cause of cancer production while direct action on the nuclear target molecules occurs with high level exposures. Radicals can react with DNA of the cell nucleus and damage the genetic code. The hormone melatonin, has been suggested as an inhibitor of cancer by way of its ability to act as a potent free radical scavenger. Suppression of melatonin production is associated with increased breast cancers in women. Evidence is accumulating to show that such free radicals are formed in cellular responses to EMR (Liburdy). There are differences in biological sensitivity due to the type of cell and its position in the cell cycle. Cellular kinetics of tissues are important in their response to radiation. Generally, cells in S-phase or M-phase would be expected to be most sensitive to physical insult. The environmental conditions, temperature, degree of hypoxia or presence of antioxidants are important in the intracellular interactions. While it is generally accepted that incorrectly repaired or unrepaired modifications of the DNA molecule are the main cause of radiation induced cancer, other “epigenetic” mechanisms have also been recently proposed. The epigenetic effects include; initiation of membrane lipid peroxidation, or loss of intercellular signalling such as gap junction mediated transfer of messenger molecules (Lowenstein 1979). Subtle changes in the genome that do not adversely affect the proliferative capabilities but may impair regulation of cell growth can lead to the late somatic effects of cancer. Loss or alteration of crucial genetic information in gonadal cells may create a risk of congenital disease. Malignant cells are defined by certain characteristic features such as unrestrained proliferation, angiogenesis (the ability to attract blood supply), infiltration into neighbouring and distant tissues, and the evasion of attacks by the immune system. Such characteristics could be the result of genetic modifications, i.e. a somatic mutation, or result from epigenetic changes. Epigenetic factors, which do not lead to irreversible changes in the primary structure of the genetic code, could produce de-differentiation, activation and expression of normally suppressed genes involved in the production, binding or signalling of growth factors, inactivation of regulatory genes, or the loss of growth controlling cell-cell-interactions (gap junctions) between a transformed cell and its environment. It may well be that both genetic and epigenetic factors have to act in parallel to let a cell escape the division-restraining signals of its environment (UNSCEAR, 1986). The origin of cancer is not well understood. However, some quantitative models of cancer induction based on simplified hypotheses have been proposed (Whittemore 1978), based on the concept of multi-stage carcinogenesis, initially developed as a model for skin cancer in mice (Farber 1980). The currently accepted model of carcinogenesis involves a multistage process (NRPB 1993; Cridland 1993) of at least three stages: initiation, involving genetic mutation of one or more cells; promotion, involving the multiplication and accumulation of damaged cells; progression during which further genetic abnormalities accumulate resulting in increased malignancy. In addition, increased proliferation may be associated with carcinogenesis by fixing and amplifying naturally occurring genetic damage, and thus may serve a role as a co-promoter. It is believed that initiation can occur in response to a single brief exposure to an agent, and that it becomes a permanent change occurring within a single mitotic cycle. It probably involves the production of a stable genetic mutation. Initiation, is generally considered to occur in the genome. Further steps towards an overt malignancy are generally assumed to be epigenetic in origin, promotion and progression. An alternative hypothesis can be based on two initiation events, the activation of an oncogene and inactivation of a tumour suppressor gene. Promotion is typically a more protracted process requiring repeated exposure to the promoting agent. It is usually reversible if the promoting agent is removed and is therefore unlikely to result from genetic mutation. Promoting agents induce cellular proliferation which allows initiated cells to multiply, expressing an altered phenotype. The best characterised pathway by which promoters can affect cell proliferation involves interference with normal cellular control system through cell surface receptors. Cell growth factors bind to specific receptors which transfer the signal across the plasma membrane and activate signal transduction, biochemical pathways for cell growth and DNA synthesis. Many cellular phenomena may also be relevant in the development of a transformed cell. At the level of DNA, any change affecting the primary structure, and hence its function, may also affect growth control. Secondary influences, such as endogenous or exogenous growth factors, e.g. female steroid hormone in breast tissue, or levels of melatonin, may increase the probability of initiated cells passing through the stages of promotion and progression to cancer. The administration of exogenous drugs can also affect promotion. Oncogene activation involves the induction or enhanced expression of gene products in growth and differentiation.
5.2 Cellular studiesCancers result from the multiplication of cells which exhibit abnormal, malignant behaviour and tend to invade and destroy adjacent tissue, or by metastasising they invade other body tissues. Usually, these cells grow at an accelerated rate and often appear to be less differentiated than normal cells. There have been suggestions that exposure to EM fields may result in an increased risk of cancer. The equivocal nature of such epidemiological reports at ELF has sustained the debate, and recent claims of brain tumours caused by use of cellular telephones have fuelled speculation. It is important that biological aspects of EMF exposure are fully investigated to determine whether any mechanism exists by which such fields could affect carcinogenesis. Biological tests for identifying potential carcinogens involve either whole animal or cell culture systems which, ideally, should be used in a complementary manner. Each test type has advantages and limitations, with neither being ideal systems. Cellular studies have the advantage that they provide a rapid screening method for potential effects, have greater sensitivity to weak carcinogens, and are more amenable to detailed molecular analysis, thereby allowing study of underlying mechanisms of interaction. Cell transformation assays represent a method for assaying changes consistent with tumorigenesis without knowing the genetic nature of the damage giving rise to the change. After plating at low density, transformed cells, which have an altered phenotype, may be identified using morphological criteria. The problem with existing cell transformation assays is that they are generally based on established cell lines which are known to be atypical. The C3H/10T1/2 transformation assay has been used to investigate the effects of electric and magnetic fields. Signal transduction pathways are complicated systems, but some important elements include inositol phosphate metabolism, intracellular calcium ion concentrations, and activation of specific protein kinase enzymes. As these represent important aspects of normal cell growth, detection methods have been developed in assays for study of the cellular effects of EM field exposures as a potential promoting agent, apart from the endpoint of increased proliferation. Tests for elevated enzyme activity such as ornithine decarboxylase, 51 - nucleotidase, and ATP-ase are commonly used. DNA synthesis is an essential prerequisite for cell division and is, therefore, one of the most commonly used endpoints of enhanced proliferation. In the process leading to tumour progression, an initiated cell is acted on by a tumour promoting agent and produces a clone of cells with altered genotype, or genetic complement. During progression these cells change from pre-malignant to increasingly malignant phenotype by undergoing loss of growth control and acquisition of invasive behaviour. This may require chromosomal aberrations that inactivate tumour suppressor genes (Cerutti 1988).
Experimental evidenceInitial studies claiming preferential absorption of microwaves by DNA molecules, and therefore giving evidence of direct interaction, have not been confirmed in subsequent studies in different laboratories. It was reported that microwave absorption in purified DNA was greater than for water at 8-12 GHz (Swicord & Davis 1982). Later studies using plasmid DNA (ranging in size from 5 to 30 kb) in aqueous solution found no evidence of enhanced absorption of microwaves over a range of frequencies from 0.1 to 12 GHz (Gabriel et al 1989; Foster et al 1984). Studies on the potential mutagenic effect of RF radiation have used fungi and yeast and detected no effect at frequencies of 2.45, 8.7175, 9.4, 17, and 70-75 GHz with SAR of 10-30 W/kg. Determining the rate of DNA synthesis provides a measure of cell proliferation. Before cells can undergo normal mitotic division they must replicate their DNA during the well-defined S-phase stage in the cell cycle. Incorporation of 3H-thymidine is a commonly used assay. Cultures of normal human lymphocytes are widely used. Transcriptional regulation is a key factor in the control of cellular growth. A number of oncogene products, including those encoded by c-myc, -fos and -jun have been shown to function as transcription factors. The cell studies that are continually referred to as evidence of potential carcinogenic effects of microwave radiation are those showing enhanced proliferation (Cleary et al 1990 a, b) and cell transformations (Balcer-Kubiczek & Harrison 1985, 1989). Cleary et al reported increased proliferation in human lymphocytes and increased rate of gross transcription in LN71 human glioma cells. However, a recent study, given as an offered presentation at the BEMS conference (Stagg et al 1994) showed the absence of a convincing effect at exposure conditions similar to that emitted by cellular telephones. It is typical of this field of research that similar studies never actually attempt to replicate the protocol used in studies reporting an effect. The work of Cleary et al used frequencies of 27 and 2450 MHz, cw, applied for 2 h and quoted estimated SAR for so-called isothermal exposures at relatively high levels of exposure. They used human glioma and T-lymphocyte cells. The study by Stagg et al used a similar assay for incorporated 3H- tdr and quoted values for power densities up to 3.7 mW/cm2 RMS values. The RF field was 837 MHz, TDMA pulse modulated and was applied for 24 h to two cell types. Different results were obtained for each cell type. Primary glial cells (rats) showed no effect on DNA synthesis. However, C6 glioma cells showed a significant increase in thymidine incorporation, although curiously there was no significant difference in cell doubling times. Cell transformation assays test for genetic alteration that gives rise to an altered phenotype that is readily identified by the morphological appearance of the transformed cells. The main problem with this test system is that the cell lines used are generally atypical. The strongest evidence of microwave radiation producing malignant transformation abnormalities comes from the work of Balcer-Kubiczek and Harrison (1985, 1989, 1991). They exposed the mouse embryonal fibroblast line C3H/10T1/2, which are commonly used in this assay, and which contain already genetically altered chromosomes. Exposure for 24 h to 2.45 GHz, pulse modulated (120 pps) and SAR 4.4 W/kg, together with X-irradiation, promoted transformation of cells. The effect was enhanced significantly when the chemical promoter TPA was added. Irradiation with microwaves and X-rays produced additive effects, suggesting that they produce effects at different targets, possibly supporting the concept of epigenetic interaction of microwaves. A recent study (funded by Motorola) has failed to show any effect on proliferation rate of glial cells following 24 h exposure to up to 3.7 mW/cm2 at 837 MHz for 24 h (Stagg et al 1994). Co-promotion of transformed fibroblasts was not enhanced (Cain et al 1994). However, there was some doubt about whether or not his exposure elicited specific (c-os, c-jun) oncogenes (Phillips et al 1994). The results of a study by Prausnitz and Susskind (1962) at 9.27 GHz pulsed radiation (2 µs pulses at 500 pps) average power density 1 kW/m2, contribute very little to the debate. Swiss mice were exposed for 4.5 min per day for 5 days per week for 59 weeks. Rectal temperatures rose by an average of 3.3°C in the exposed group which initially comprised 200 mice, while the control group comprised only 100 mice. The study was abbreviated by an outbreak of pneumonitis and terramycin was administered during the last 3 months of irradiation. A total of 132 mice were sacrificed 7, 17 or 19 months after the initial exposure. 168 mice died, (68 without adequate post-mortem analysis). Monocytic or lymphatic "leucosis" (defined as a non-circulating neoplasm of white blood cells) and lymphatic or myeloid leukaemia (defined as a circulating "leucosis") occurred in 35% of the 60 exposed mice compared with 10% of the 40 control mice. Interpretation of the results is difficult and the study has been extensively criticised (NRPB 1991) on the grounds that "leucosis" and "leukaemia" were inadequately defined and identified, that the infection may have confounded the results, that a large proportion of mice (23%) died without a cause of death being identified and that statistical analysis was absent. An analysis (chi-square contingency table) carried out on the two groups with elevated levels of leucosis found the results to be of marginal significance (0.1 > p > 0.05). Continued....
[CONTENTS] [GLOSSARY] [REFERENCES] Current Research Abstracts Index
Please e-mail comments, information and updates to DON MAISCH: |