"Experimental Design: The study was conducted blind, and only the project engineers and dosimetry monitoring staff were informed of the exposure levels, which were decoded after the hisopathological studies had been completed. Fifty male and 50 female Pim1 mice, randomly selected by means of a computerized stratified sequenced randomization program, were exposed at each exposure level (0.5, 1.4 or 4 W/kg), were sham exposed, or were used as cage controls. The exposure was performed 1 h/day, 7 days/week for 18 consecutive months. In the previous experiment by Repacholi et al. (4), the 1-h daily exposure was split into two 30-min sessions (morning and afternoon). Exposure The exposure requirements included the same signal conditions as used by Repacholi et al. (4) (carrier frequency: 900 MHz, pulse width: 0.577 ms; pulse repetition rate: 217 Hz) but with substantially improved exposure conditions, i.e., constant age-independent and uniform exposure. Since the SAR levels in the experiment of Repacholi et al. (4) varied greatly, it was decided to select the highest RF-field exposure level at the thermal regulation threshold, i.e. 4 W/kg (6). This level results in tissueaveraged exposures that are higher than the peak spatial SAR limits for environmental human exposure (1) of 2 W/kg averaged over 10 g of tissue. To obtain information about a possible dose–response relationship, a medium- and a low-dose SAR were set at 1.4 and 0.5 W/kg, respectively. Further requirements were blinded exposure, minimal space, and compatibility with the experimental requirements of U.S. National Toxicology Program (NTP). The dosimetry was assessed according to the methodology of Kuster et al. (8) and provided information about wholebody, organ and peak spatial averaged SAR as well as about uncertainty, instant and lifetime variations. The exposure system developed for mice consisted of a signal generation unit, control and monitoring unit and four ‘‘Ferris wheel’’ exposure units (Fig. 1). Each exposure unit, enabling exposure of up to 65 animals, was excited at a different SAR (high, medium, low and sham). The Ferris wheel concept developed by Balzano et al. (9) was adopted and optimized for this study. Each wheel consisted of two parallel, circular, stainlesssteel metal plates 117 mm apart, with a conical antenna in their center and stainless-steel poles constituting a short-cut of the cylindrical cavity at a radius of 755 mm. The positions of the animals, restrained in tubes similar to those approved and used for inhalation studies, were optimized for maximum uniformity (Fig. 2). To maintain a symmetrical load, missing animals were replaced by conical plastic tubes filled with 36 ml of liquid, simulating the dielectric parameters of muscle tissue at 900 MHz (conductivity: 0.98 S/m, relative permittivity: 55.6). The ventilation system guaranteed an air flow of 1 liter per minute at the snout of the animal. The wheels were excited by a 200-W amplifier (LSE, Sweden)."

"Exposure: Each of the four active bursts (slots 0, 2, 4, 6) per frame of the output signal was sequentially multiplexed (4-channel PIN-diode switch, LSE, Sweden) to the different wheels; i.e., each wheel was excited by one slot. The frame signal was produced by an arbitrary waveform generator (Agilent 33120A) combined with a signal generator (SML01, Rhode & Schwarz, Germany). The amplifiers were water-controlled to enable placement close to the wheels and therefore the use of short cables. The time multiplexing and the low losses in the cables reduced the peak power requirement and thus also the costs of the amplifiers by a factor of almost eight. Throughout the experiment, the exposure levels were controlled and monitored with two electric-field sensors inside each wheel. Exposure levels were adjusted automatically (sampling rate 10 s) whenever there was any drift/change, ensuring stable exposure conditions inside the setup. The incident field was adjusted as a function of weight to obtain an age independent dose. Environmental parameters such as relative humidity, air temperature and oxygen level were also recorded for each wheel at all times, even when no exposure was taking place. In addition to the ambient parameters, all hardware communication was saved to be able to reconstruct each experiment. To maximize data security, all recorded data were saved encoded at different physical locations, Istituto di Ricerche Biomediche LCG-RBM (Torino) and at IT’IS Foundation/ETH (Zurich), and were transmitted automatically each night."

See Table 2 for Spatial Peak and Organ-Averaged SAR Relative to the Whole-Body Average Values of 4 W/kg (high-dose group), 1.4 W/kg (medium-dose group) and 0.5 W/kg (low-dose group).

E mu pim-1 mice, female

HAWC