"Central Nervous System Tumors The incidence of brain, nerve and spinal cord tumors is recorded in Table 2. There was no statistically significant difference between males and females in any neoplastic feature studied (except where noted), so the sexes are com?bined for appropriate analyses.

Characteristics of Brain Tumors Two hundred thirty-five (235) brain tumors were ob?served in 183 rats. As expected, the incidence was strongly associated with the dose of ENU. The zero-dose ENU groups had 22 solitary brain tumors in 300 rats (7.3%). The 400 rats in the 2.5 mg/kg ENU groups had 55 brain tumors (13%), and the high ENU dose groups had 158 brain tumors in 112 rats from a total of 180 rats (88%). The incidence of brain tumors was higher than expected in the 300 rats given no ENU; the natural occurrence of tumors in the brains of rats is about 1–2% (44) compared to 7.3% in this study. The difference is probably because many more sec?tions of brain (18–26) were studied histologically in this experiment compared to the three sections per brain rou?tinely used in most safety assessment studies. Brain tumors were located in the periventricular white matter and less commonly in the cortex or nuclei of the cerebrum. Most tumors present in the medulla or cerebellum appeared to have originated from a primary location in the cerebrum. The occurrence of brain tumors by histological type is given in Table 3. The tumors were composed of neoplastic oligodendroglia, astrocytes or more commonly both. Mixed gliomas were defined as containing less than 80% of either neoplastic astrocytes or oligodendroglia (of having less than 20% of the minor cellular component), as described pre?viously (44). Immunohistochemical reactions were very helpful in classifying anaplastic brain tumors (44). Astrocytomas were the most common brain tumors in the 300 rats that were not exposed to ENU; 20 developed as?trocytomas and 2 had oligodendrogliomas. The astrocyto?mas in these rats differed from those that occurred in rats exposed to ENU in that what few oligodendroglia (0–8%) that were contained in these tumors appeared to be normal cells entrapped by infiltrating neoplastic astroglial cells. Male rats in the zero-dose ENU group had more astrocy?tomas (15) than females (5). The tumor incidence in rats exposed to ENU was essentially equally distributed by sex, so the males and females were combined for other analyses. Oligodendrogliomas varied greatly in volume. Very small tumors were readily distinguished from reactive gli?osis, since oligodendroglia do not generally accumulate at sites of brain injury. Oligodendrogliomas also varied wide?ly with regard to malignancy. The mean malignancy score was similar to that of astrocytomas and was found in pre?vious studies (44) to be significantly lower than for mixed gliomas. Mixed gliomas were more malignant, larger, and more likely to be fatal, and rats that bore them had the shortest survival. The incidence of mixed gliomas was clearly related to the dose of ENU (Table 3), but no sig?nificant differences in incidence were found between the PRF (group 3) and sham (group 4) groups (log-rank test, P 5 0.21). Differences in brain tumor malignancy between study groups showed no clear pattern (Table 4) and were not sta?tistically significant (P 5 0.17). The malignancy score was directly related (P 5 ,0.001) to survival: the higher the score (the more malignant), the shorter the survival. Malig?nancy of brain tumors was also positively related to tumor volume (r 5 0.75, P 5 ,0.001): the higher the malignancy grade, the larger the tumor. Multiple brain tumors (2–4/rat) were common (20%) in rats that received the highest ENU dose, were rare in rats receiving 2.5 mg/kg ENU, and were not observed in rats that were not given ENU (Table 4). In most rats with multiple brain tumors, the volume, malig?nancy and histological type of tumor varied considerably between tumors in the same rat. Group 3 (PRF) contained 15 rats with 37 tumors, compared to group 4 (sham), in which 12 rats had 25 brain tumors, and group 15 (cage controls), in which 9 rats had 20 brain tumors. Oligoden?drogliomas occurred with about equal incidence to mixed gliomas in rats with multiple tumors, whereas astrocytomas were uncommon. Brain tumors ranged in volume from 0.1 mm3 to 610 mm3 (Table 4). Small clusters of glial cells were determined to be neoplasms if they appeared to be growing masses compressing surrounding tissue and there was no apparent cause for gliosis (44). Small tumors were usually incidental findings in rats killed because of other lesions. Diminutive brain tumors were found in rats that died at all ages, in?cluding 2-year-olds; however, most occurred at sites of pre?dilection for ENU-induced brain tumors in rats less than 1 year old. There was no significant association between tu?mor volume (log transformed) and exposure group (P 5 0.74). There was no evidence that the brain tumors were closer to the antenna in some study groups than in others (P 5 0.24). Effects of RF-Field Exposure on Brain Tumors The primary study hypothesis contrasted rats that were given a high dose of ENU and exposed to a PRF field (group 3) with those given a high dose of ENU and sham exposure (group 4). PRF-exposed rats had a trend toward a higher incidence of fatal brain tumors than the sham and cage control groups, but these differences were not statis?tically significant (relative hazard 5 1.63, 95% CI: 0.85– 3.13, P 5 0.14; and relative hazard 5 1.51, 95% CI: 0.81–2.82, P 5 0.19, respectively). Cumulative survival curves for the three groups are shown in Fig. 4. In addition, no statistically significant differences were found between groups 3 and 4 in early death (P 5 0.22), incidence of fatal spinal or cranial nerve tumors (P 5 0.50), or incidence of mixed gliomas (P 5 0.21). For the groups given 0 and 2.5 mg/kg ENU, where brain tumors were uncommon, there was no trend toward more fatal brain tumors in the RF?exposed groups (Table 4). Comparisons of the overall rate of tumor incidence (for tumors of any type) were conducted between the RF-exposed and control groups for each ENU dose. None of the six comparisons of were statistically sig?nificant. Spinal Cord Tumors Spinal cord tumors were observed in 10 rats: 4/180 rats (2.2%) given 10 mg/kg ENU, 5/418 (1.2%) in rats given 2.5 mg/kg ENU, and 1/130 (0.3%) in rats given 0 mg/kg ENU. Nine of the 10 spinal cord tumors were gliomas; one was a granular cell tumor. Seven tumors were observed in RF-exposed rats, 2 were seen in sham-exposed rats, and 1 was present in cage controls. Most spinal cord tumors oc?curred in the thoracic or lumbar enlargements, where the SAR of the RF field was only 0.4 and 0.2 W/kg, respec?tively. Peripheral Nervous System Tumors Cranial and spinal nerve tumors were all histologically malignant schwannomas. They were composed mostly of Antoni type B cell pattern and had many mitotic figures and necrotic or cystic areas. Larger tumors, especially those in the sacral area or subcutis, readily invaded skeletal mus?cle; none metastasized. The combined incidence of cranial and spinal schwannomas was 0/300 in the zero-dose ENU rats, 39/418 (9.3%) in the group given 2.5 mg/kg ENU, and 57/180 (32%) in the group given 10 mg/kg ENU (Table 2). Of the 60 spinal nerve schwannomas observed, 33 arose from the roots of the spinal nerves in the cranial, lumbar or less commonly thoracic or sacral regions of the spinal cord. Many of these tumors compressed the spinal cord, causing paresis or paralysis of limbs. Other tumor locations included the subcutis (12), thoracic cavity (8) and abdom?inal cavity (7). Many of the subcutaneous and thoracic tu?mors arose from nerves associated with the brachial plexus. Nerve tumors in the body cavities often grew very large (see below) before signs such as dyspnea and enlarged ab?domen were observed. In the rats exposed to ENU, the incidence of spinal nerve tumors in RF-exposed rats was 25/240 (10%) compared to 19/238 (8.0%) in sham controls and 16/120 (13%) in cage controls. Multiple tumors occurred in 7 rats, 4 RF-exposed and 3 non-RF-exposed. The average distance from the an?tenna to the tumor was 150 mm for RF-exposed rats com?pared to 152 for nonexposed rats. The volume of spinal nerve tumors ranged from 4.0 to 170,000 mm3 ; the mean volume for all RF-exposed rats was 35,000 6 60,000 mm3 compared to 14,000 6 26,000 mm3 for sham-exposed rats, but the difference was not statistically significant (Kruskal?Wallis test, P 5 0.20). Survival analysis revealed no sig?nificant difference between RF- and sham-exposed (P 5 0.50) groups given a high ENU dose or between RF-ex?posed and cage control groups (P 5 0.21) on a composite fatal spinal or cranial nerve tumor outcome. Cranial nerve tumors arose from the trigeminal nerves in 31 rats and from the optic nerve and facial nerve in one rat each. Trigeminal nerve tumors tended to compress the brain or invade the pituitary and meninges of the brain. Sometimes the tumor cells invaded the brain through the perivascular spaces. Among ENU-exposed rats, the inci?dence of cranial nerve tumors was 13/240 (5.4%) in RF?exposed rats, 13/238 (5.5%) in the sham-exposed rats, and 7/120 (5.8%) in cage controls. There were 7 rats with more than one tumor, 2 in the RF-exposed, 2 in the sham-ex?posed, and one in the cage control groups. There was no obvious difference in the incidence between rats in the PRF and CWRF groups. The mean tumor volumes were 490 6 630 mm3 in the RF rats and 340 6 440 mm3 in the sham-exposed rats."

Sprague-Dawley rats - CWRF, 860 MHz

HAWC