Repeated exposure to nanosecond high power pulsed microwaves increases cancer incidence in rat.

High-power microwaves are used to inhibit electronics of threatening military or civilian vehicles. This work aims to assess health hazards of high-power microwaves and helps to define hazard threshold levels of modulated radiofrequency exposures such as those emitted by the first generations of mobile phones. Rats were exposed to the highest possible field levels, under single acute or repetitive exposures for eight weeks. Intense microwave electric fields at 1 MV m-1 of nanoseconds duration were applied from two sources at different carrier frequencies of 10 and 3.7 GHz. The repetition rate was 100 pps, and the duration of train pulses lasted from 10 s to twice 8 min. The effects on the central nervous system were evaluated, by labelling brain inflammation marker GFAP and by performing different behavioural tests: rotarod, T-maze, beam-walking, open-field, and avoidance test. Long-time survival was measured in animals repeatedly exposed, and anatomopathological analysis was performed on animals sacrificed at two years of life or earlier in case of precocious death. Control groups were sham exposed. Few effects were observed on behaviour. With acute exposure, an avoidance reflex was shown at very high thermal level (22 W kg-1); GFAP was increased some days after exposure. Most importantly, with repeated exposures, survival time was 4-months shorter in the exposed group, with eleven animals exhibiting a large sub-cutaneous tumour, compared to two in the sham group. A residual X-ray exposure was also present in the beam (0.8 Gy), which is probably not a bias for the observed result. High power microwaves below thermal level in average, can increase cancer prevalence and decrease survival time in rats, without clear effects on behaviour. The parameters of this effect need to be further explored, and a more precise dosimetry to be performed.

Introduction

High power microwaves (HPM) are used to inhibit the electronic systems of threatening vehicles. Concern has arisen as to whether HPM could lead to health hazards for operators of emitting systems and for staff members exposed in targeted vehicles. The health effects of HPM have been studied since the discovery of radar in the middle of the past century. Many experiments have been performed with microsecond pulses at levels of several hundred kilovolts per meter. Some studies have been published, but many others have been presented only as reports or at scientific meetings.

Studies performed with a whole-body specific absorption rate (SAR) above the thermal threshold of 4 W kg-1 have shown biological effects. Below 4 W kg-1, for studies showing effects, the pulse duration of single pulses was between 40 ns and 10 μs, and temporal peak-SAR was from 5 to 20 MW kg-1. Half of the studies on HPM addressed behavioural endpoints, reviewed by D’Andrea [1]. Others bear on the cardio-vascular [2,3], visual [4], and auditory systems [5]. Only sparse work concerned blood-brain-barrier permeability [6,7], DNA damage [8,9], carcinogenesis [10,11], or cellular or sub-cellular mechanisms [12]. Another paper by Dorsey et al. [13] on carcinogenesis used ultrawideband (UWB) exposure. Concerning cancer, Zhang et al. [14] and Devyatkov et al. [10] reported protective effects at levels above thermal threshold, with smaller tumours and a 30% increase of survival rate in the exposed group. Focusing on UWB exposures, several years after Seaman’ article [15], a recent review by Schunck et al. reported only one new paper in 2009, and no other effects on cancer were reported [16]. However, durations of exposure in those studies were often short.

Using a realistic source of HPM, this study investigated whether the highest possible exposure levels could produce behavioural or functional effects in rats acutely exposed, or chronic pathological effects with a repetitive exposure for two months. We assessed effects of 3.7 and 10 GHz nanosecond pulsed HPM around 1 MV m-1 on the health and lifespan of male Sprague-Dawley rats.

Methods

Literature survey

We looked at the scientific and medical literature (NCBI-PubMed, Current Contents and Science Direct, more recently Web of Science) and at specialized databases of papers, scientific meetings and reports (EMF Database, IEEE ICES EMF Literature Database and WHO- EMF-Portal). The following keywords were used: HPM, high power microwave, high peak, electromagnetic pulse, microwave radiation, high exposure microwave, HPPP, EHPP, high intensity microwave.

Exposure systems

The sources of high-power microwaves (HPM) were two superradiance generators, one in X-band at 10 GHz (SRX) with pulses of 1 ns, the other in S-band at 3.7 GHz (SRS) with pulses of 2.5 ns. The strongest possible microwave electric fields were applied, of about 1 MV m-1, at a repetition rate of 100 pps (Table 1). The “SINUS type” electron accelerator of this system is made of a Tesla generator and a continuous formation line. A great advantage of this system is its small size. The superradiance source is derived from the back-wave oscillator, with the following characteristic features: ultrashort microwave pulses, and very high peak power. Pulses were gated sinewaves, with a bandwidth around 10% and rise/fall-times of about 0.2ns.

Table 1

Exposure parameters of the two HPM sources.

	SRX	SRS acute	SRS avoidance	SRS repeated
Beam diameter at output	13 cm	22 cm
Frequency	10 GHz	3.7 GHz
Total emitting power	350 MW	500 MW
Pulse duration	1 ns	2.5 ns
Train duration	10 s	Continuous
Emission duration	Every 5 min for 1 h	2 x 8 min	14 min	2 x 8 min
Peak surface power at output	20 GW m-2	2 GW m-2
Distance from the horn		0.60 m	0.13 m	2.5 m
Peak E-field	3 MV m-1	1.7 MV m-1	2.9 MV m-1	0.56 MV m-1
Peak SAR	95 MW kg-1	31 MW kg-1	90 MW kg-1	3.33 MW kg-1
Average SAR over total exposure	0.34 W kg-1	4.7 W kg-1	22 W kg-1	0.83 W kg-1

Animals

Six-weeks-old Sprague Dawley male rats were purchased from Charles River, L’Arbresle, France. Identification of animals was made with a marker pen on the tail for acute studies, or by tattoo on the ear pavilion during the long-term follow-up. Rats were accustomed to the animal facility for 5 days before handling. They were kept under specific pathogen free (SPF) controlled environmental conditions (ambient temperature 22±1 ◦C, 12-h light/12-h dark cycle) and received food and tap water ad libitum, except during exposure. Rats were housed in groups of 2 animals per enriched cage for social comfort. Enrichment consisted in adding corn chips for animals to nest and play. The day before exposure, rats were handled and accustomed to the experimenter. Further, they were either exposed (N = 150) or sham-exposed (N = 144). The protocol was reviewed and approved by INERIS ethical committee. Animals were monitored clinically and for mortality once a day. Closer surveillance was performed in case of clinical observations, such as behaviour changes, dull hair or upon appearance of a tumour. In the study with repetitive exposure, the criteria to determine when animals should be ethically euthanized during the follow-up were: weight loss (more than 20% as compared to the week before), ulceration of a tumour, tumour size larger than 8 cm, impaired movement, loss of spontaneous activity or loss of reactions to stimulus.

Animal dedicated to immunohistochemical analyses, to avoid pain and distress, were injected with an overdose of pentobarbital (i.p.; 50 mg kg-1) until brain-dead stage.

Exposure protocol

Two types of acute exposures were carried out. The SRX exposure lasted 10 s every 5 min for one hour, and the SRS exposure lasted 2 x 8 min with 10 min interval (26 min total). Rats were transferred in plastic boxes with filtering covers from the animal facility through an airlock to the shielded exposure room 50 m apart. For acute exposures, rats were exposed one by one in their cage directly at the horn output (246 animals in total, 12 per group, except one group of 6 for preliminary viability tests). Besides, one protocol of repeated exposures was used with SRS source for 48 rats. The 26 minutes-exposure was repeated each day, 5 days/week for 8 weeks. When performing mean term repetitive exposures every day with a realistic source that cannot easily be duplicated, there is a need for optimization of the design to expose many animals at the same time. The circular beam produced by the TM01 mode of the waves was adapted to this goal, with a beam width large enough to simultaneously expose 12 animals at 2.5 m from the SRS output. Animals were exposed 2 by cage in six cages placed each day at different positions on the circle. Then every day, 2 series of 12 animals were exposed, alternating with 2 series of sham exposure in-between to allow time for the equipment to cool down between two successive exposure sessions. In total, two groups of 24 rats received a repeated exposure, either real, either sham.

After the end of repeated exposures, the animals were observed and followed up to 2 years of age. Lifespan was recorded, and anatomo-pathological examination was performed at the animal death.

For each test, a group of 12 exposed animals was compared to a similar-sized group of sham-exposed animals, put in the same place and under the same ambient conditions than the exposed animals, but without emission from the source.

Investigations on the central nervous system

After one acute or the last repetitive exposure, different behavioural tests were performed blindly: beam-walking, rotarod, T-maze, open-field. An avoidance test was also performed during an acute SRS exposure, applied continuously for 14 minutes. In the avoidance test, animals can choose to spend time in two parts of a box. One part is protected against the beam (shielded), the other is not. Animals behaviour is recorded with a video camera. Upon analysis of the movies, the time spent in the non-protected side is recorded.

After the behavioural tests were performed, animals were sacrificed 2 days after the SRX and the SRS exposures, and 7 days after the SRX exposure, and an immunohisto-chemical labelling of the brain inflammation marker GFAP was achieved on 40 μm thick slices for 5 parts or regions of the brain: frontal cortex, gyrus dentate, putamen, pallidum and cerebellar cortex.

The global design of this study and the sample size in each test are summarized in Table 2.

Table 2

Global design of the study.

	SRX	SRS
Exposure	Acute	acute	repeated
Emission duration	10 s every 5 min for 1 h	2 x 8 min with 10 min interval Avoidance: 14 min continuous	2 x 8 min with 10 min interval 5 days/week for 8 weeks
Age at experiment	6 weeks	6 weeks	6 weeks
N animals/exposition	1	1	12
Behavioural tests	Beam walking (n = 13/11a), rotarod (n = 12/12), T-maze (13/11), open field (n = 12/12), avoidance (n = 11/10)	Beam walking, rotarod, T-maze, avoidance (n = 12/12)	T-maze (n = 8/8, wk14b), Beam walking (n = 12/12, wk15), rotarod (n = 24/24, wk16)
GFAP staining (J = exposition day n = nb animals)	J+2n = 12/12 J+7 n = 11/11	J+2 n = 12/12
Anatomo-pathology (HES)	104 weeks or at death	104 weeks or at death	104 weeks or at death
Lifespan recording (up to 2 years)			n = 24/24

an = n1/n2 = [number of exposed animals] / [number of sham animals].

bwk = age of animals at the date of test.

Immunohistology

Brain preparation

An intra-cardiac perfusion was performed with a 0.9% NaCl solution, followed by a 4% paraformaldehyde solution in 0.1 M phosphate buffer (pH 7.6) at a flow rate of 30 ml min-1. The brain was dissected and immerged in a 30% sucrose/4% paraformaldehyde solution for 48 h under agitation at 4 °C to cryo-protect the cerebral tissue Using a freezing microtome, 40 μm-thick sagittal brain slices were obtained and processed as free-floating sections for GFAP immunodetection [17].

Immunohistochemistry

Brain slices were washed in Phosphate Buffer Saline (PBS; pH 7.6) 3×20 min and incubated for 30 min at room temperature in 1% H2O2 to block endogenous peroxidases. Background noise was saturated by incubation for one hour in a solution of goat serum (5%), bovine serum albumin (2%) and Triton (0.3% Triton). Slices were then incubated overnight at 4 ◦C with the primary polyclonal antibody anti-GFAP (rabbit antibody, 1/8000 in PBS). On the second day, the slices were washed with PBS-Triton (3×20 min) and incubated for 1 h with the secondary antibody (biotinylated goat anti-rabbit, ABC kit 1/500 in PBS). Sections were washed in PBS (3x20 min). Immunostaining was revealed with an ABC/VIP kit (Vectastain, Vector). Slices were mounted on slides, dehydrated then mounted with DPX before image analysis. Surfaces of labelled areas have been quantified using an optical microscope (Zeiss) coupled with a Colour Camera 3 Charge Coupled Device (CCD) (Sony) and the Visilog 6.2 (NOESIS society, Les Ulis, France) analyser system.

Anatomopathology

Animals involved in repeated exposures were followed up to 2 years of age. Individual lifespan was recorded for animals needing an ethical euthanasia. Either at this time or at 104 weeks of age, animals were sacrificed with a lethal overdose of pentobarbital (5.0 ml kg-1 IP), organs were collected and cut into smaller pieces if larger than 5 mm in order to optimize tissue fixation in 4% buffered formalin for 48 to 72h. All tissue samples were included in paraffin blocks.

Five μm slices were obtained with a microtome and two slides per organ were prepared. An anatomo-pathological examination was performed on one slide per organ, that was coloured with haematoxylin-eosin stain (HES). The second section was stored in case of need for any other specific labelling.

Dosimetry

Electric field has been measured at the actual exposure distance of 2.5 m from the source output with a germanium detector and calculated for closer distances. As numerical computation of specific absorption rate (SAR) by FDTD was not available, the whole-body specific absorption rates (SAR) (defined as electromagnetic power absorbed per unit of tissue mass) were calculated for each condition from the rat’s position and size as described by Gandhi [18] and Durney et al. [19]. Time-averaged SARs were 0.83 W kg-1 for the repeated exposure, and between 0.34 and 22 W kg-1 for acute exposures. Temporal peak SARs during the pulses ranged between 3.3 and 95 MW kg-1 (Table 1). Some residual X-rays came out from the device, for 20 mGy/day (total 0.8 Gy). Numerical and experimental dosimetry and thermometry are needed to strengthen the results of this study.

Statistics

Percentages of time spent in the exposed or in the blinded box (avoidance study) were compared by two-way ANOVA with two factors: exposure (exposed or sham) and period (habituation or exposure). Percentages of labelled areas for GFAP were compared by two-way ANOVA with two factors: exposure (exposed or sham) and localisation (brain area) with R software (Version 2.5.0). Survival rates of repetitively exposed animals were compared with Prism 5 v5.02 by the log-rank test (Mantel-Cox), with calculation of two-tail p value, of the median survival and of the hazard ratio between both groups by the Mantel-Haenszel method.

Results

Behavioural tests

First, behavioural tests to evaluate cognitive and sensori-motor functions were performed. No effects were observed after acute or repetitive exposures on behavioural results in beam-walking, T-maze and open-field tests. After repeated exposure to a superradiance S source (SRS) of HPM, rotarod performance was assessed: rats had to stay for three minutes on an axis rotating at 16 turns per minute. They underwent one training session and one test session. In the test session, rotarod performance was significantly enhanced in exposed animals: 41/72 exposed animals succeeded, compared to 23/72 sham animals—p < 0.001 (S1 Table).

During an acute SRS exposure, with a choice for rats to stay in an exposed or a shielded compartment (avoidance test), exposed animals spent 3.7% ± 3.5% of time on the exposed side, compared to 21.9% ± 9.1% for the sham group (mean ± SD; p < 0.001) (S1 Fig and S2 Table).

Brain inflammation

Then, brain inflammation was assessed by measuring glial fibrillary acidic protein (GFAP) levels, indicative of damaged or dysfunctional cerebral tissue. With a superradiance X source (SRX) of HPM, expression of GFAP was not increased two days (D2), but was increased seven days (D7), after an acute exposure, in all brain areas, except the cerebellum cortex (+50.0% ± 7%—mean ± SD; p < 0.02). With SRS, GFAP expression was increased two days after acute exposure in all brain areas (D2 - +112% ± 19%—mean ± SD; p < 0.001) (Fig 1 and S3 and S4 Tables).

Fig 1

GFAP expression after repeated exposure to SRS source.

GFAP immunohistochemical labelling in different brain areas two days after exposure with Superradiance S source (% labelled area—Mean ± SD). White = sham (n = 12); black = exposed (n = 12). One slice per area per animal. *** p < 0.001.

Lifespan

Most strikingly, six exposed animals deceased early between 33 and 47 weeks, leading to a 4-months decrease of lifespan in the repetitively exposed group (n = 24) compared to the sham group (n = 24) (Fig 2). The median lifespan was 590 days for the exposed group, compared to 722 days for the sham group—p < 0.0001. One sham rat was used as sentinel for sanitary control, eleven sham animals survived at the end of the experiment, whereas only two animals survived in the exposed group. The hazard ratio was 4.1 [CI = 2.0–8.6].

Fig 2

Survival proportion of rats after repeated daily exposure to SRS source.

Empty circles = sham; black circles = exposed. Survival curves were significantly different (p < 0.001).

Lethal tumours and anatomopathology

For tumours identified as the cause of death, eleven of the exposed animals showed one or two large sub-cutaneous tumours of different types (five were malignant—seven appeared before 20 months) (Table 3 and one illustration Fig 3), compared to only two such tumours in the sham group (both malignant, first one at 22 months).

Table 3

Lifespan to death or sacrifice and anatomopathological diagnostic of lesions.

Rat #	Age at death (weeks)	Exposed group	Sham group
4	33	Internal mass, adenopathiesa …
10	38	Fibrosarcomab
21	39	Internal mass, adenopathies, spleen
8	43	Fibrosarcoma
20	45	Fibroma, ulcerated
5	47	Posterior limbs paralysed
36	60		/c
3	66	Internal mass, spleen, pancreas
6	66	Subcutaneous adenocarcinoma
27	73		Large cystic kidneys
23	79	Fibroma, ulcerated
17	82	/
37	83		/
25	84		Posterior limbs paralysed
46	84		/
2	84	Pituitary tumour
11	84	Spongy/granulous kidneys
19	85	Mesenteric massc, ileum (lysed organs)
45	86		Large preputial glands
34	88		Sentinel animal
24	88	Polycystic kidneys
7	88	Osteosarcoma
16	88	Fibro-epithelial polypus, cystic and spongy kidneys
14	91	Jejunal mass
9	92	Mass: adrenal/kidney/spleen (internal bleeding)
48	92		Large preputial glands
13	94	Zymbal's gland adenoma, ulcerated ear area, pituitary tumour, cystic and spongy kidneys
1	94	Fibro-adenoma, cystic and spongy kidneys, tracheo-bronchial ganglions large and inflammatory
15	95	/
29	97		Large cystic and spongy kidneys, white lung masses
35	98		Cystic and spongy kidneys, duodenum dark and spongy content
26	98		Subcutaneous schwannoma, large spleen, large left preputial gland
40	99		/
12	99	FibrosarcomaDark abdominal cavity, testes soft small and dark. Soft brain, small spleen, large left adrenal gland, external part of lungs grey/brown
18	103	/
22	103	Fibro-adenoma
39	104		Fibrosarcoma
28, 30–33 41–44, 47	103–104		Sacrificed, no tumour

Left column: exposed group (# = 1–24); right column: sham group (# = 25–48).

aitalic: internal masses found at death or at 24 months;

bbold: large external masses leading to ethical sacrifice;

c/: no macroscopic abnormality. Only two exposed rats survived at the end of the experiment at 103 weeks: #18, 22. Eleven rats of the sham group survived to the end of the experiment: #28, 30–33, 38, 39, 41–44 and 47 were sacrificed at 103 or 104 weeks, 10 of them without any tumour.

Fig 3

(a) Picture of one exposed rat with two fibromas (left)–(b) macroscopic view of the femoral tumour (right).

The two fibromas were in the axillary and the femoral area, the macroscopic view was taken at autopsy (death at 18.5 months).

One of the exposed animals with an external tumour also had a pituitary tumour, and at death, six other exposed animals had abdominal masses and one had a pituitary tumour. Tumour types and lifespan are detailed in Table 3 and histological lesions are further listed in S5 Table.

Discussion

Although they are hugely far above environmental levels, the question has been raised whether intense and very short pulses (nanosecond range) could have health effects. Old studies considered typical radar modulation of 1/1000th (1 μs every millisecond, i.e. repetition rate of 1 kHz). Below the thermal level of 4 W kg-1, no specific effect of modulation had been proven, so this modulation factor of 1/1000th has been considered as safe in the public health standards [20]. Recent studies on high intensity nanosecond pulsed microwaves have been performed on cells and their electrophysiological properties, or on membrane permeability, but none on animals with repeated exposures [16].

Although some stress could be induced by the microwave auditory effect or by possible corona effect following field enhancement near protrusions on animals' bodies (e.g. toes, nose), no effect was observed on learning abilities or anxiety level. However, a positive effect was found in the rotarod test, which mainly assesses sensori-motor activity. This could be due to a slight heating at the SAR of 4.7 W kg-1 of the SRS source. Such a heating effect has been hypothesized by Preece who observed an increased reactivity (shorter reaction time) in human volunteers exposed to mobile phones with a local SAR of 1.7 W kg-1 [21]. Avoidance of the SRS beam was significant in exposed animals subjected to a thermal SAR of 22 W kg-1, which is high above the thermal threshold of 4 W kg-1 identified by ICNIRP [20]. This expected result confirms the relevance of this threshold.

The large GFAP level increase in different brain structures after SRS exposure reflects the higher average SAR of 4.7 W kg-1 allowed by a continuous emission of SRS pulses instead of spaced 10 s pulse trains with average SAR of 0.34 W kg-1 in the SRX. These results are consistent with those previously found with a much lighter modulation of microwaves 1/8th of the time, such as the one produced by GSM mobile phones [22]. In comparison, the modulation of HPM is 1 ns at 100 pps, i.e. a ratio of 10−7. Alterations in the glial cell marker GFAP could represent a marker of a long-term risk in rats, but this has yet not been shown. Mainly known as a marker of traumatic injury, GFAP has been considered by previous studies as non-specific, therefore compromising its prognostic power [23].

Lifespan, lethal tumours and anatomopathology

More importantly, an increased and early rate of sarcomas and fibrosarcomas, as well as higher associated mortality, were observed in animals exposed to repetitive sessions at an average SAR of 0.8 W kg-1, five times below the thermal threshold (Table 1). The spontaneous rate of fibrosarcomas found in old Sprague Dawley rats at termination of two-year studies is usually around 1 to 3%, and tumours are rarely reported as cause of life shortening [24]. Several studies have tested the impact of HPM, but chronic exposure was only performed with continuous waves, radar-type microwave pulses of the order of microseconds [11,25,26], or mobile phone-type exposures [27]. Effects on cancer and lifespan were reported with SARs close to or above the thermal level [25,28,29]. Only Chou et al. reported an increase in primary malignancies, without life shortening, with pulsed waves at low SAR levels (0.15–0.4 W kg-1) and exposures lasting 21.5 h/day for 25 months [26]. Recently, a NIEHS study of the National Toxicological Program reported an increased incidence of heart schwannoma and glioma in whole-body exposed rats to phone-type microwaves at much higher SARs than those used in humans [27]. Although experiments with newer extremely short pulses (a few ns long) have been performed, HPM had only been used in acute experiments, and most studies reporting an effect looked at physiological reactions, without addressing genotoxicity or carcinogenicity endpoints. This work therefore corresponds to the first report with in vivo exposure to extremely short duration peak pulses, with a high repetition rate, and with a design of repeated exposure for eight weeks.

Tinkey showed that very high doses of X-rays (> 46 Gy) were needed to induce sarcomas in Sprague-Dawley rats [30]. To our knowledge, no study showed increased tumour rates in rats exposed to lower doses of X-rays; then the low 0.8 Gy residual X-ray level of this study would probably not explain the observed early tumour increase. This hypothesis would need to be tested with positive controls exposed at 0.8 Gy expanded over the same 2-months period. Therefore, this study shows that the observed tumours and decreased lifespans were due to repeated exposures at a SAR below the known health threshold of 4 W kg-1, given that the peak SAR was of the order of 3.3 MW kg-1 (E-field above 0.5 MV m-1).

Conversely, some studies would support a protective effect of HPM against cancer. Devyatkov found a decrease in tumour cell proliferation in vitro and an increase in survival time of rats implanted with a liver carcinosarcoma and exposed to 10 ns pulses at 9 GHz, which paradoxically was beneficial [10]. The peak power was 100 MW, but the electric field or SAR was not specified.

More recently, although using UWB pulses which have a different frequency spectrum, after 16–1000 ns pulses with a frequency of 0.6–1.0 GHz, a duration of 4–25 nanosecond, an amplitude of 0.1–36 kV cm-1, and a pulse repetition rate of 13 pulses per second (pps), Zharkova also found an inhibition of mitochondrial activity which has been interpreted rather as an anti-tumoral activity [31]. However, other studies with UWB bring some mechanistic explanation that could support a cancerogenic effect. Dorsey found an increase in mitogenic activity of mouse hepatocytes [13]. Natarajan published genotoxic effects [32] and Shckorbatov showed some changes in chromatin [33], which studies bring arguments rather in favour of a carcinogenic effect. Whether mechanisms are common between HPM and UWB exposures will have to be further explored.

Observed tumours were mostly subcutaneous, but were also ubiquitous, which is not indicative of a specific mechanism or sensitivity of a given tissue or organ. This means that inflammatory processes or genotoxic effects should be investigated in the different target tissues where tumours appeared: connective tissue, muscle, fat, vessels, pituitary gland, lymph nodes, etc. To check if this is a general phenomenon or a strain/specie specific effect, this experiment should be repeated with other rat strains and different animal species (e.g. mice, and/or rabbits) which are usual models for human toxicology.

The actual corner stones of guidelines for RF exposures consider behavioural effects as the most sensitive biological endpoint that had yet been observed as a deleterious effect on health. Up to date, this decreased behavioural performance is today attributed to the temperature elevation produced in rodents or primates, consecutive to the above-mentioned dielectric absorption, only linked to the average absorbed power (rms SAR). This study shows that a repeated and prolonged exposure to extremely high intensity microwave pulses, around one million volts per meter (1 MV m-1), comparable to those that have in part been used in the Gulf War, produce a clear increased incidence of cancer in exposed animals. Furthermore, it tells that even an aggressive damage such as cancer can occur without so much decreased cognitive performance, even at a level below the known thermal threshold of whole-body SAR (4 W kg-1). Then the peak SAR should be re-considered in the definition of guidelines. However, and luckily, humans are not exposed for such durations at close distance from HPM emitters.

The original hypothesis was: is there an effect of high-power microwaves? In which conditions? If yes, does it obey to a classical thermal mechanism or a mechanism other than thermal? This study showed: i) few behavioural effects from either acute or repeated exposure; ii) an inflammatory effect of acute exposure to HPM; and iii) a surprising increase of lethal cutaneous or subcutaneous tumour incidence of sarcoma or fibrosarcoma type, in the repetitively exposed group (46% versus 8% in the sham-control group). This increased cancer incidence was associated with decreased lifespan in rats exposed to HPM with an average SAR level below the thermal threshold of 4 W kg-1. Furthermore, this effect was not associated with clear effects on behaviour, as could have been expected from previous knowledge. The underlying mechanisms are likely to be different from thermal effects and need to be further explored. Also, the thresholds or dose-responses in SAR level, duration and number of exposure sessions need to be defined before establishing limit values for human exposure.

Avoidance test during SRS exposure: % time spent in the shielding box in habituation and exposure periods.

(TIF)

Click here for additional data file.

Rotarod test data after SRS exposure: Time spent on the rods at training and at test.

(PDF)

Click here for additional data file.

Avoidance test during SRS exposure: Time and % time spent in the shielding box in habituation and exposure periods.

(PDF)

Click here for additional data file.

GFAP immunohistochemical labelling after SRX exposure—Raw values.

Magnification x10.

(PDF)

Click here for additional data file.

GFAP immunohistochemical labelling after SRS exposure—Raw values.

Magnification x10.

(PDF)

Click here for additional data file.

Synthesis of histological lesions in sham (Sh) and exposed (Ex) groups after repeated SRS exposure.

(PDF)

Click here for additional data file.

4 Feb 2020

PONE-D-19-32924

Repeated exposure to nanosecond high power pulsed microwaves increases cancer incidence in rat

PLOS ONE

Dear Dr De Seze,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Specifically:

some precisions needs to be given in the material section

some modifications are needed in the text

some references could be added to reinforce the message

We would appreciate receiving your revised manuscript by Mar 20 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Christophe Egles, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

1. At this time, we request that you  please report additional details in your Methods section regarding animal care, as per our editorial guidelines:

(1) Please state whether the provided ethics committee contains animal welfare experts or whether an animal ethics or IACUC committee reviewed and approved the study. Please provide the full name of the committee that reviewed and approved the study

(2) Please state the number of mice used in the study

(3) Please provide details of animal welfare (e.g., shelter, food, water, environmental enrichment)

(4) Please describe any steps taken to minimize animal suffering and distress, such as by administering anaesthesia  or analgesia

Thank you for your attention to these requests.

2. To comply with PLOS ONE submission guidelines, in your Methods section, please provide additional information regarding your statistical analyses. For more information on PLOS ONE's expectations for statistical reporting, please see https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Worthy but small study by an experienced group that explores new grounds in RF bioeffects. Several comments

1. Please describe the exposure facility better, Were any conductive objects (e.g. metal tubes from water bottles) in the field? Describe the waveforms of the pulses (I assume they were gated sinewave with a spectral range that is small compared to the carrier frequency - is that correct?) Were the animals "chipped"? (a potential source of enhanced local exposure within the body)

The exposures are extraordinarily large (MV/m) and not far from the breakdown threshold for air. This makes me wonder if corona was involved. There will be field enhancement near protrusions on animals' bodies (e.g. toes, nose), possible sites for corona which would be a source of stress to the animals. Would the authors comment on this as a possible mechanism for the observed effects.

2. The animals clearly found the stimulus to be obnoxious (avoided exposure when given the choice). I note that the fluence (of the order of tens of J/m^2) is far higher than the threshold for the microwave auditory effect (tens of millijoules/m^2). So the animals were undoubtedly perceiving strong audible sensations from the exposure. Is there any evidence for this as a source of stress?

less major:

Writing is imprecise.

1. It is not correct to compare the present study (pulse modulated microwaves) with those using ultrawideband pulses (e.g. Ref. 16; p. 17). The latter have a different frequency spectrum including in some cases a DC component, which will produce quite different effects. Please avoid this confusion.

2. Authors repeatedly refer to "SAR". Please make it clear every time whether they refer to whole body SAR or partial body SAR in an identified region of the body, or spatially peak SAR at any place in the body. Preece (ref 20) was referring to partial body SAR from a phone in localized regions of the head averaged over time, no whole body SAR as referred to in this study.

3. p. 19 "comparable to those that have in part been used in the Gulf War" - were humans exposed to RF pulses comparable in strength to those in this study? For how long and under what circumstances?

Reviewer #2: The authors present a very interesting and important paper, which has to be published after some amendments. These includs some modification as well as more information within the Materials and Methods.

• Legends to the figures are missing

• Were the experiments performed in a blinded manner?

• Regarding the GFAP-staining, it is not clear how this protocol was applied. Here clear information need to be included (which kind of antibody was used, how it was applied, concentration, incubation time etc.).

• In the discussion (line 233) the authors cite a single one author, namely Tinkey (29) as the one proving that >46 Gy X-ray is needed to induce sarcomas in rats. They write that this the prove that 0.8 Gy cannot be the reason for the increased tumor rate. A single study cannot be used as a “prove”, so please change this sentence accordingly. It could have been helping using positive controls at 0.8 Gy level.

• In Table 2: please change in second row “10mn” to “10 min”; and in the 4th row “Nb” to “N”

• There are also some more typographic errors and some linguistic problems that need to be improved

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Kenneth Foster

Reviewer #2: Yes: Myrtill Simko

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

27 Feb 2020

Response to reviewers

Editor:

1. At this time, we request that you please report additional details in your Methods section regarding animal care, as per our editorial guidelines:

(1) Please state whether the provided ethics committee contains animal welfare experts or whether an animal ethics or IACUC committee reviewed and approved the study. Please provide the full name of the committee that reviewed and approved the study

At the time of the study (2005-2007), the ethics committees were not formalized in France. The ethics file has been submitted to the INERIS ethics committee, based on experimented animal technicians and on good laboratory practices labelled at INERIS. The study was named “MFP study”, but not numbered.

(2) Please state the number of rats used in the study

Altogether, 294 rats were used in this published part of the study. Numbers have been given page 7 lines 76, and page 8 lines 91-92 for exposed (144), sham (144) and test (6) animals.

(3) Please provide details of animal welfare (e.g., shelter, food, water, environmental enrichment)

Rats were accustomed to the animal facility during 5 days before handling. They were kept under specific pathogen free (SPF) controlled environmental conditions (ambient temperature 22±1 ◦C, 12-h light/12-h dark cycle) and received food and tap water ad libitum, except during exposure. They were housed in groups of 2 animals per enriched cage for social comfort. Enrichment consisted in adding corn chips for animals to nest and play. The day before exposure, rats were handled and accustomed to the experimenter. This has been added page 7 lines 70-75.

(4) Please describe any steps taken to minimize animal suffering and distress, such as by administering anaesthesia or analgesia

No experimental procedure was painful and needed analgesia. Upon sacrifice before biochemical analyses, to avoid pain and distress, animals were anaesthetized with an overdose of pentobarbital (i.p.; 50 mg kg-1) until brain-dead stage. This has been added on page 7, lines 83-84.

2. To comply with PLOS ONE submission guidelines, in your Methods section, please provide additional information regarding your statistical analyses. For more information on PLOS ONE's expectations for statistical reporting, please see https://journals.plos.org/plosone/s/submission-guidelines.#loc-statistical-reporting

The used softwares have been defined (page 13 line 168), and results have been further defined (mean) and completed with SD where useful (behavioural tests page 14, lines 181-183 and 188-190).

Reviewer #1: Worthy but small study by an experienced group that explores new grounds in RF bioeffects. Several comments

1. Please describe the exposure facility better,

The rats were transferred from the animal facility through an airlock to the shielded exposure room 50m from there in plastic boxes with filtering covers. A thin metallic grid supported the filtering cover. The cages were set on a plastic support for acute exposure just in front of and perpendicular to the emitting horn of the HPM generator. Then exposure came to the animal from the side. Details have been added page 8 lines 88-90.

Were any conductive objects (e.g. metal tubes from water bottles) in the field?

The food grid and the water bottles were removed from the cages for the exposures.

Describe the waveforms of the pulses (I assume they were gated sinewave with a spectral range that is small compared to the carrier frequency - is that correct?)

At 10 GHz, the SRX source emits about 10 oscillations of 1 ns. Spectrum width is about 10%, i.e. 1 GHz (inverse of the duration, obtained by Fourier Transform). What about the rise-time and fall-time? Rise-time and fall-time are about 0.2 ns.

At 3.7 GHz, the SRS source emits about 9 oscillations of 2.5 ns. Spectrum width is about 400 MHz (inverse of the duration, obtained by Fourier Transform). What about the rise-time and fall-time? Same as for SRX? Rise-time and fall-time are about 0.2 ns.

Details have been added page 5 lines 63-64.

Were the animals "chipped"? (a potential source of enhanced local exposure within the body)

The animals were not chipped but labelled with a marker pen for the duration of the exposure period, and ears were tattooed for the lifetime follow-up. Then there was no metallic component close to the rat’s body. This has been added page 7 lines 69-70.

The exposures are extraordinarily large (MV/m) and not far from the breakdown threshold for air. This makes me wonder if corona was involved. There will be field enhancement near protrusions on animals' bodies (e.g. toes, nose), possible sites for corona which would be a source of stress to the animals. Would the authors comment on this as a possible mechanism for the observed effects.

We thank the reviewer for this interesting comment. Indeed, at this field level, a field enhancement near protrusions on animals' bodies (e.g. toes, nose) is possible, and corona effect could be a source of stress to the animals. However, the cognitive tests measuring anxiety (T-Maze, open-field) did not show any change of anxiety level in exposed rats compared to sham-controls.

This concern has been addressed together with the next one: one sentence and precision have been added page 19 lines 231-234

2. The animals clearly found the stimulus to be obnoxious (avoided exposure when given the choice). I note that the fluence (of the order of tens of J/m^2) is far higher than the threshold for the microwave auditory effect (tens of millijoules/m^2). So, the animals were undoubtedly perceiving strong audible sensations from the exposure. Is there any evidence for this as a source of stress?

The avoidance test was the only one with the maximum thermal exposure (22 W/kg). We agree with the reviewer that the exposure level in our experiments (20 J/m² with the SRX source, 2 J/m² with the SRS source) is higher than the threshold for the auditory effect (100-400 mJ/m²). However, the auditory effect is described to be perceivable between 200 MHz and 6.5 GHz (ICNIRP, 1998), then it should not be detected with the SRX source, and it is potentially strongly perceived with the SRS source, 5-fold above the threshold level. Although this is a possible source of stress, there again cognitive tests did clearly not show increased anxiety in exposed rats compared to sham-controls.

To address these two last concerns, one sentence and precision have been added page 17 lines 231-234

less major:

Writing is imprecise.

1. It is not correct to compare the present study (pulse modulated microwaves) with those using ultrawideband pulses (e.g. Ref. 16; p. 17). The latter have a different frequency spectrum including in some cases a DC component, which will produce quite different effects. Please avoid this confusion.

When UWB have been used, clarifications have been made in the manuscript:

On page 3, the paper by Dorsey has been individualized (lines 37-38), and it has been specified that reviews by Seaman and Shunck focused on UWB (line 40).

On page 22, it has been specified when studies have been performed with UWB exposures: Dorsey, Zharkova, Natarajan and Shckorbatov (lines 282-290).

2. Authors repeatedly refer to "SAR". Please make it clear every time whether they refer to whole body SAR or partial body SAR in an identified region of the body, or spatially peak SAR at any place in the body. Preece (ref 20) was referring to partial body SAR from a phone in localized regions of the head averaged over time, no whole-body SAR as referred to in this study.

SAR has been defined in this study as whole-body SAR (page 3 line 31) and peak SAR as temporal peak-SAR (page 3 line 33, page 13 line 160). Where different, this has been pointed out (Preece’s study: with a local SAR of …; page 18 line 237)

3. p. 19 "comparable to those that have in part been used in the Gulf War" - were humans exposed to RF pulses comparable in strength to those in this study? For how long and under what circumstances?

In the context referring to Gulf War the pulses were not directed to humans at such a close distance. The emissions were shorter, as it does not take long to inhibit electronics.

The addressed question is: do HPM exposures induce cancer? At which intensity and which duration?

After the results of this study, a dose-response relationship must be established, and once threshold for a risk will have been defined, safety margins will have to be chosen to protect exposed humans (operators or in targeted vehicles).

To avoid a too quick extrapolation to human, it has been added on pages 22-23:

Lines 302-303: “This study shows that a repeated and prolonged exposure to extremely high intensity microwave pulses”

Lines 308-309: “However and luckily, humans are not exposed for such durations at close distance from HPM emitters.”

Lines 320-321: duration and number of exposure sessions need to be defined “before to allow establishing limit values for human exposure.”

Reviewer #2: The authors present a very interesting and important paper, which has to be published after some amendments. These include some modification as well as more information within the Materials and Methods.

• Legends to the figures are missing

Following PLOS One instructions to authors, we placed figure captions “in the manuscript text in read order, immediately following the paragraph where the figure is first cited. Do not include captions as part of the figure files or submit them in a separate document.”

• Were the experiments performed in a blinded manner?

For behavioural tests and biochemical analyses, the experiments were blinded (added line 112 page 10).

For the two-year follow-up, the exposed and the sham group had been identified.

• Regarding the GFAP-staining, it is not clear how this protocol was applied. Here clear information needs to be included (which kind of antibody was used, how it was applied, concentration, incubation time etc.).

Thanks for asking these precisions. We did not want to be too long in methods description, but here is our detailed protocol:

• Brain preparation

Animals were anaesthetized with pentobarbital (i.p.; 50 mg kg-1) until brain-dead stage 2 or 7 days after exposure. An intra-cardiac perfusion was then performed with a 0.9% NaCl solution, followed by a 4% paraformaldehyde solution in 0.1 M phosphate buffer (pH 7.6) at a flow rate of 30 ml min-1. The brain was dissected and immerged in a 30% sucrose/4% paraformaldehyde solution for 48 h under agitation at 4 ◦C to cryo-protect the cerebral tissue. Using a cryostat microtome, 40 µm-thick sagittal brain slices were obtained and processed as free-floating sections for GFAP immunodetection [17].

• Immunohistochemistry

Brain slices were washed in phosphate buffer saline (PBS; pH 7.6) 3×20 min and incubated for 30 min at room temperature in 1% H2O2 to block endogenous peroxidases. Background noise was saturated by incubation for one hour in a solution of goat serum (5%), bovine serum albumin (2%) and triton (0.3% Triton). Slices were then incubated overnight at 4 ◦C with the primary polyclonal antibody anti-GFAP (rabbit antibody, 1/8000 in PBS. On the second day, the slices were washed with PBS-triton (3×20 min) and incubated for 1 h with the secondary antibody (biotinylated goat anti-rabbit, ABC kit 1/500 in PBS; pH 7.6). Sections were washed in PBS (3x20 min; pH 7.6). Revelation of immunostaining was revealed with a ABC/VIP kit (Vectastain, Vector). Slices were mounted on slides, dehydrated then mounted with DPX before image analysis. Surfaces of labelled areas have been quantified using an optical microscope (Zeiss) coupled with a Colour Camera 3 Charge Coupled Device (CCD) (Sony) and the Visilog 6.2 (NOESIS society, Les Ulis, France) analyser system.

The protocol has been described page 11-12 lines 123-142.

• In the discussion (line 233) the authors cite a single one author, namely Tinkey (29) as the one proving that >46 Gy X-ray is needed to induce sarcomas in rats. They write that this proves that 0.8 Gy cannot be the reason for the increased tumour rate. A single study cannot be used as a “prove”, so please change this sentence accordingly. It could have been helping using positive controls at 0.8 Gy level.

It is right that a single study cannot be a proof and that it could have been helping using positive controls at 0.8 Gy level. However, we carefully looked at the literature and did not find that lower levels of X-rays could cause an increase of early tumour rates during the lifetime of the animals. Usually, studies with X-rays show increase of tumours only at the anatomopathological examination of the animals post-mortem. However, to satisfy this comment, text has been changed accordingly (page 21, lines 271-274): “To our knowledge, no study showed increased tumour rates in rats exposed to lower doses of X-rays; then the low 0.8 Gy residual X-ray level of this study would probably not explain the observed early tumour increase. This hypothesis would need to be tested with positive controls exposed at 0.8 Gy expanded over the same 2-months period.”

• In Table 2: please change in second row “10mn” to “10 min”; and in the 4th row “Nb” to “N”

Done

• There are also some more typographic errors and some linguistic problems that need to be improved

The whole paper has carefully been reread and corrections have been made where felt to be needed.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

18 Mar 2020

Repeated exposure to nanosecond high power pulsed microwaves increases cancer incidence in rat

PONE-D-19-32924R1

Dear Dr. De Seze,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Christophe Egles, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

26 Mar 2020

PONE-D-19-32924R1

Repeated exposure to nanosecond high power pulsed microwaves increases cancer incidence in rat

Dear Dr. De Seze:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Christophe Egles

Academic Editor

PLOS ONE