Microwave / Radiofrequency ( MW / RF ) Radiation Exposure and Cancer Risk : Meta-Analysis of Accumulated Empirical Evidence

C l i n M e d International Library Citation: Atzmon I, Linn S, Richter ED, Portnov B (2016) Microwave/Radiofrequency (MW/ RF) Radiation Exposure and Cancer Risk: Meta-Analysis of Accumulated Empirical Evidence. Int J Cancer Clin Res 3:040 Received: October 21, 2015: Accepted: January 28, 2016: Published: January 30, 2016 Copyright: © 2016 Atzmon I, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Atzmon et al. Int J Cancer Clin Res 2016, 3:040


Introduction
The World Health Organization's (WHO's) International Agency for Cancer Research (IARC) classified both Extremely Low Frequency (ELF) Magnetic Fields and Radio Frequency Radiation (RFR) in Group 2B Human Carcinogen [1][2][3].
Scientific debate over the potential risk attributed to MW/RF radiation has increased dramatically in recent years due to increased concerns as wireless technologies saturate the home and workplace.Deshmuckh et al. [4] found DNA strand breaks in rat brains after being exposed to low intensity microwave radiation at the lower, middle and upper frequencies used in mobile telecommunication.The researchers concluded that although microwave energy is not sufficient to break the chemical bonds in DNA directly, genotoxic effects may be mediated by indirect mechanisms, such as generation of oxygen free radicals or a disturbance in DNA-repair processes (ibid).To the best of our knowledge, the present study is the first of its kind, which aims to quantify the evidence of association between MW/ RF radiation and cancer, in both occupational and environmental settings, using meta-analysis tools.

Association ratios
We considered all types of association ratios reported in the literature, including Odds Ratio (OR) reported in 19 studies; Relative Risk or Rate Ratio (RR) reported in 11 studies, Standardized Incidence Ratio (SIR) used in 4 studies, Proportional Incidence Ratio (PIR) used in 1 study, Proportional Mortality Ratio (PMR) reported in 5 studies, Proportional Registration Ratio (PRR) reported in 1 study, Standardized Mortality/Morbidity Ratio (SMR; O/E = observed/expected) reported in 10 studies, Mortality Rate Ratio (MRR) reported in 1 study, and crude death rate per 1000 reported in 1 study.The ratio of the 'observed to expected' is called SMR (standardized mortality ratio), SIR (standardized incidence ratio) or when death is the outcome, the standardized mortality ratio, SMR [19].Proportional Mortality Ratio (PMR) gives results similar to SMR (personal communication with Dr. Sam Milham; Decoufle 1980) [20], while for a disease like cancer, the RR is approximately equal to the OR [21].The calculation that is used in WinPepi© is valid in a meta-analysis where different units of measurements are used in different studies (WinPepi© manual).

Data sources
We analyzed the peer reviewed papers published between 1982 and 2012 dealing with the association between MW/RF radiation and cancer in both environmental and occupational settings.The studies were found using the Medline and Google scholar search engines or based on previous knowledge.Altogether we reviewed 57 papers that included different end points (morbidity, mortality and 1 survival study) in different ages (children and adults).
Out of the 57 studies collected, 4 studies were excluded because they did not contain epidemiologic analysis [22,23] reported a very high result that masked results of other studies in disproportion [there was one underlying cause of death due to leukemia compared with 0.2 expected (standard mortality ratio [SMR] = 437, 95% confidence interval [CI] = 11-2433), and two multiple listed causes of death due to leukemia compared with 0.3 expected (SMR = 775, 95% CI = 94-2801) [24]] or used a case series method with no epidemiological ratios and/or probability values [25].Out of the remaining 53 studies, described in Table 1, we discarded for use in the meta-analysis 6 studies without information on confidence intervals or indications on how to extract them [26][27][28][29][30][31] and one study in which the CI was reported to be higher than the actual measure of association [RR = 4.15 95% CI 40.1, 217.2 [17]].From the total collected, we used 46 studies (81%) for the meta-analysis.

Use of WinPepi© program
The analysis was performed by the meta-analysis module of the WinPepi program COMPARE2 [32] using a fixed-effect model.In the fixed-effect model, it is assumed that the individual studies provide estimates of the same results [33].
The WinPepi computers programs for epidemiologists were designed as learning or teaching aids for use in practice and research in the health field.WinPepi is the Windows version of the DOS-based PEPI (an acronym for Programs for EPIdemiologists) package, which grew from a set of programs for programmable pocket calculations published in 1983 [34].

Results
The results are presented in Table 2, Figure 1, Figure 2, Figure 3 and Figure 4. Overall, MW/RF radiation was recorded in studies to be associated with a statistically significant increased risk of morbidity of 2-3 fold compared to the controls.This study found that especially low intensity MW/RF radiation (0.25 µW/cm 2 ; SAR = 3 µW/kg) led to oxidative DNA damage.Based on the evidence, the authors of the study concluded that the oxidative changes may develop in pathology leading to oncogenic transformation of cells.Yakymenko et al. [6] reviewed epidemiological evidence from radars and mobile communication systems studies.The researchers found that under certain conditions, exposure to long term low intensity MW/RF radiation led to initiation and promotion of cancer.Yakymenko concluded that recent data strongly indicate the need for concern and transparency of the current safety limits for nonionizing radiation using recently obtained knowledge.Summarizing laboratory evidence on oxidative damage to cells, Yakymenko et al. [7] reviewed 80 peer reviewed publications, of which 76 (92.5%) reported the detection of significant oxidative stress from MW/RF radiation.The authors noted that significantly increased levels of reactive oxygen species (ROS) in living cells caused by low intensity MW/RF radiation exposure, may promote mutagenic effects through oxidative damage in the DNA.They also noted, that overproduction of ROS in living cells under low intensity exposure could cause a broad spectrum of health disorders and diseases, including cancer in humans.Within one year, the number of studies they reviewed increased to 100, of them 93 confirmed that MW/RF radiation induces oxidative effects in biological systems.Molecular effects induced by low-intensity in living cells included significant activation of key pathways generating ROS, activation of peroxidation, oxidative damage of DNA and changes in the activity of antioxidant enzymes.The authors concluded that oxidative stress should be recognized as one of the primary mechanisms of the biological activity of this kind of radiation [8].
Levitt and Lai [9] found that broadcast exposures had been found unsafe even at regulated thresholds, noting significant increases for all cancers in both men and women living near broadcast towers and leukemia clusters in children and adults.They found also, that 56 of 56 studies reported biological effects at very low intensities of MW/RF radiation, including DNA damage in human glial and leukemia cells, effect on the DNA repair mechanism and indication of an increase in glioma cells division.
Regarding epidemiological evidence, reviews of occupational and environmental studies of cancer-related effects from exposure to MW/RF radiation, varied in conclusions.Elwood [10], Ahlbom et al. [11] and Breckenkamp et al. [12] found inconsistent epidemiological evidence.
Carpenter [13,14] stated in his reviews, that excessive exposure to MW/RF radiation increased risk of cancer, with the strongest evidence coming from studies on cell phone users, whereas results were not consistent across all studies that reported elevations in both leukemia and brain tumors among individuals with occupational exposure to MW/RF radiation.The author pointed to more recent reports that found elevated rates of leukemia among children who lived near AM radio transmitter sites.
Kundi and Hutter [15] reviewed effects of base stations, and found that only two studies had been published on cancer [16,17].Both of them found increased risk, but had no individual data and therefore were considered to provide limited evidence, and no firm conclusions could be drawn.Khurana et al. [18] included the same two studies in the review, concluding that altogether increased prevalence of adverse neurobehavioral symptoms or cancer in populations living at distances of < 500 meters from base stations prevalence of adverse neurobehavioral symptoms were found in 80% of the available studies.The authors recognized methodological weaknesses, especially since exposure to MW/RF radiation was not always measured.However, they noted that none of the studies they reviewed, revealing adverse health effects from base stations, reported exposures to MW/RF radiation above accepted international guidelines.The researchers concluded, that if such findings continue to be reproduced, current breast cancer morbidity, brain and CNS cancer morbidity, all cancer sites morbidity.For testicular cancer, borderline significant morbidity risk was found.No statistically significant association was found for brain and central nervous system (CNS) tumors morbidity in children, lymphoma and multiple myeloma mortality, and mortality from brain and CNS tumors.For all cancer sites mortality reduced risk was found, on the basis of 2 results that showed reduced risk and 5 results that showed increased risk.

Discussion
To the best of our knowledge, our study is the first metaanalysis focusing on the association between exposure to MW/RF (younger have higher risk) are reported in table 3. Occupational military studies found increased risk of cancer mortality or morbidity with decreasing age [35][36][37][38].There was higher risk of all cancer sites for the age groups 20-39 compared to older age groups [37] and latency period shortened in exposed vs. non exposed [38].Environmental studies found indications for promotional effect (acceleration) and higher risk in younger age, with regard to cancer morbidity, mortality and survival.
In a study from Germany, the average age of developing cancer was 8.5 years earlier in the exposed vs. the non exposed population.For breast cancer, the average age in the exposed area was approximately 13 years younger than the average age of developing cancer in the exposed, and approximately 20 years less than in the non exposed population, as well as 12 years younger than the national average age for developing breast cancer in Germany [16].An Israeli study on exposure to cell towers, found an extremely short latency period [17].Risk increased and survival decreased for exposed vs. non exposed in children [39].In a study from Korea, higher risk was found under 30 years old [40].

ISSN: 2378-3419
Atzmon et al.Int J Cancer Clin Res 2016, 3:040 radiation and cancer risk.Our results are consistent with the review by Yekymenko et al. [6] and (Levitt and Lai 2010) [9], who revealed evidence about increased risk of cancer associated with exposure to MW/RF radiation.Reviews by Elwood [10], Breckenkamp et al. [12] and Ahlbom [11], who found that evidence for cancer associated with MW/RF radiation was inconsistent just a decade ago, were published before more up to date results on increased risk of all site cancers were published [16,17,31,41], as well as specific cancer studies [35,42].Other researchers have recognized the need for a more precautionary approach vs. the current standards, after taking into account recent data [6,8,13,14].According to Carpenter [13,14], leukemia is the cancer most likely to indicate elevated risk from whole body exposure to electromagnetic fields (EMFs) of any frequency, since the same cancer is elevated following exposure to power-line frequencies.
Meta analyses and reviews may be complementary one to one each other, but serve an important purpose in contrasting or combining results from other studies, in an effort to identify patterns and sources of conflict between their results, or patterns of intriguing relationships that may come to light in the context of multiple studies [19].While a review helps to explain results in terms of biology and bias, quantitative analysis has more precision with small but important associations or subtle patterns in the material [19].We detected the main patterns and biases in the body of evidence during the preparation of this work.In environmental studies: ecological bias, i.e., without unique information on exposure levels, the use of an average exposure measure; the use of distance vs. radiation exposure.
In occupational studies: the healthy worker effect bias, involvement of other risk factors that can act in synergy or mask the risk without a separate analysis.In both occupational and environmental studies, confounder analysis was lacking at the individual level.Despite the limitations of individual studies, an increased risk was found for several types of cancer.Though brain cancer morbidity was elevated (1.11) for childhood brain tumors, it was not statistically significant.The lack of a sufficient latency period may be the reason.In a recent review on ionizing radiation-induced malignant gilomas, brain tumors occurred within 15 years in 82% of the patients and in 18% of the patients they developed > 15 years after radiation therapy [43].Only 26 cases of radiation-associated meningiomas occurring in the pediatric population have been reported in the English literature, they are very rare [44].In one study meningioma following radiation therapy occurred after 6-13 years and in other studies after a latency period of 3-63 years [44].The association for brain tumor morbidity was received after conservative analysis, i.e., it did not include risk results [45] from low power transmitters.After we removed the breast cancer mortality category because of too few studies for analysis, a correlation was found [40,46,47].No correlation was found for lymphoma mortality on the basis of the low number of studies.Borenstein et al. [33] and Abramson [32], both experts of meta-analyses, related to the common criticism on meta analyses about comparing diverse studies, using different methodologies.They disagree that it is necessarily a disadvantage.According to Abramson [32], diverse studies may help to explain differences in results and provide useful additional information.Such comparisons may be the main purpose of a meta-analysis and can be the power that drives the study in the sense of understanding the phenomenon.Heterogeneity can be seen as an opportunity rather than a problem.Pooling of results may reveal an effect that individual trials do not clearly show and it may also indicate that results seen in isolated trials may be a false/positive caused by random error [32].

Supporting evidence for promotional effects and higher risk for young ages
Findings of Tillmann et al. [48] were recently confirmed in a study by Lerchl et al. [49].In that study, mice exposed in the womb to a cancer agent and then exposed to a cell phone signal, had significantly higher rates of liver and lung tumors, as well as lymphoma compared to the cancer agent without the cell phone signal.This study employed radiation levels that do not cause thermal reactions and are well below current safety standards.The current standards are designed for prevention of an acute/immediate effect of heat damage (thermal effect) from short term exposure whereas the entire population is exposed for years/life-time (long-term/chronic exposure) which standards state are not applicable.In a study that evaluated the effect of 900 MHz generated by mobile base stations on hematological parameters and cellular composition of bone marrow in mature and immature rats, exposure to a mobile base station was found to have a deleterious effect on hematological parameters and bone marrow composition; this effect was more severe in immature animals [50].Brautbar [51] reported a rapid development of brain tumors in two cell phone testers with occupational exposure to cell phones and testing equipment, in whom brain tumors appeared within less than 5 years of the first exposure, and on the same side of the head that the phones were used.Richter et al. [52] reported a short latency period of brain tumors with occupational exposure to cell phones, suggesting that earlier-reported individual cases characterized by short latencies in young persons with high military occupational exposure, serve as predictors of increased group risk for exposures to RF/MW radiation.Earlier findings [53] suggested that young persons exposed to high levels of MW/RF radiation for long periods were at increased risk of cancer.In the Hardell group's studies on the possible association between brain tumors and mobile/ cordless telephone use, the highest risks were associated with > 5-year latency period in the 20-29-year age group (OR = 4.30, 95% CI = 1.22-15) for cordless phones [54], and the risk of astrocytoma grade I-IV was highest for cases with first use < 20 years of age, for mobile phone OR = 5.2 (95% CI = 2.2-12) and cordless phone OR = 4.4 (95% CI=1.9-10) [55].For the aviation technicians group, that had the highest exposure, death rates were significantly higher than those for the remaining men for all deaths, disease-related deaths, deaths from malignance and deaths from malignancy of the lymphatic and hematopoietic systems.The same group had a younger mean age (23.4) than the average for aviation electrician's mates AE (24.7), a category that showed no increase in deaths from any malignancy or from other diseases Park [40] Higher risk of mortality from leukemia and multiple myeloma in people under 30 years old compared to other age groups Hocking [39] Increased incidence and decreased survival among children who resided near TV transmitters, supported the possibility that RFR acts as a facilitator of cancer Eger [16] Average age of developing cancer was 64.1 years in the inner area (exposed) and 72.6 years in the outer area, a difference of 8.5 years.The average age of patients that developed breast cancer in the inner area was 50.8 years.In the outer area the average age was 69.9 years, approximately 20 years less.In Germany the average age for developing breast cancer is about 63 years.
Wolf [17] The authors suggested a strong cancer promoting effect at very low radiation, because the study found an extremely short latency period of less than 2 years.According to the authors, this short latency period indicated, that if there is a causal association between RF/MW radiation emitted from base stations and cancer as they believe, then there is a very strong promotional effect at low levels of exposure Considering that the mankind is exposed to this radiation increasingly from multiple sources, with multiple ambient frequencies, on a daily basis, especially considering that this new combination of frequencies has not been studied (4G, together with smart meters, Wi-Fi, together with previous cellular generations that are currently in use), it is imperative to build revised measures in the current standards, taking into account the reality of chronic exposure and providing robust protection to the public.

Conclusions
Accumulated empirical evidence to date and summarized in this analysis using meta-analysis tools, found an increased risk of morbidity and/or mortality from lymphoma, leukemia, melanoma and brain/CNS cancers, following exposure to MW/RF radiation.Evaluating this evidence, current standards are clearly not sufficient to protect the public and workers from exposure to MW/RF radiation and should take into account long term effect of increased risk of cancer.Promotional effects and special sensitivity in young people supports the restriction of exposure in children to wireless technologies including Wi-Fi in schools.Exposure assessment limitations in empirical studies also postulate an under-estimation of the level of risk.It is crucial that future studies take into account individual specific data, both on confounders and exposures.

Figure 1 :
Figure 1: Cancer of the Brain and Central Nervous System Morbidity.Risk of brain and central nervous system cancer was found to increase in relation to MW/RF radiation exposure

Figure 2 :Figure 3 :Figure 4 :
Figure 2: Lymphoma Morbidity Risk of lymphoma increased in relation to MW/RF radiation exposure

Table 1 :
General description of studies

Study Study design, types of cancer outcome and type of exposure Confounders No. of Subjects/population Source of data
ISSN: 2378-3419 Atzmon et al.Int J Cancer Clin Res 2016, 3:040 ISSN: 2378-3419 Atzmon et al.Int J Cancer Clin Res 2016, 3:040 Wright [28] Letter Proportional Incidence Ratio (PIR) leukemia, occupational Specific for age, gender and race 1 no. of cases (TV and radio repair) Cancer Surveillance Program information on Los Angeles County

Table 2
reports statistically significant increased risk for lymphoma morbidity, childhood leukemia morbidity, all leukemia adult morbidity, all leukemia adult mortality, melanoma morbidity,

Table 2 :
Results of meta analysis

Table 3 :
Promotional effect and higher risk for young vs. older ages