Chris J. Somps

A current practice for predicting ocular toxicity of systemically delivered drugs

The ability to predict ocular side effects of systemically delivered drugs is an important issue for pharmaceutical companies. Although animal models involving standard clinical ophthalmic examinations and postmortem microscopic examinations of eyes are still used to identify ocular issues, these methods are being supplemented with additional in silico, in vitro, and in vivo techniques to identify potential safety issues and assess risk. The addition of these tests to a development plan for a potential new drug provides the opportunity to save time and money by detecting ocular issues earlier in the program. This review summarizes a current practice for minimizing the potential for systemically administered, new medicines to cause adverse effects in the eye.

Crizotinib Reduces the Rate of Dark Adaptation in the Rat Retina Independent of ALK Inhibition

Crizotinib (Xalkori) is a tyrosine kinase inhibitor of both anaplastic lymphoma kinase (ALK) and mesenchymal-epithelial transition factor (c-Met). Though not predicted from standard nonclinical toxicological evaluation, visual disturbance became a frequently observed adverse event in humans. To understand the possible mechanism of this vision effect, an in vivo electroretinogram (ERG) study was conducted to assess retinal functional changes following oral administration of crizotinib. Immunohistochemical (IHC) staining of ALK and c-Met in the neural retinas of human, non-human primate, dog, rat, and mouse was used to aid in the animal model selection. ALK IHC staining was identified predominantly in the ganglion cell and inner nuclear layers of most species evaluated, in the inner plexiform layer in human and rodent, and in the nerve fiber layer in human and rat only. There was no apparent staining of any layer of the neural retina for c-Met in any of the species evaluated. ERG measurements identified a significant reduction in b-wave amplitude during the initial phase of dark adaptation in the crizotinib-treated rats. ERGs were also taken following oral administration of PF-06463922 (an ALK-selective inhibitor), for an understanding of potential kinase involvement. ERG effects were not observed in PF-06463922-treated animals when comparable exposures in the vitreous humor were achieved. Collectively, our results suggest that the ERG b-wave amplitude decreases during dark adaption following crizotinib administration may be related to signaling changes within the retina in rats, likely independent of ALK inhibition.

Telithromycin blocks neuromuscular transmission and inhibits nAChR currents in vitro

Telithromycin, a ketolide antibiotic, is reported to exacerbate myasthenia gravis, potentially leading to respiratory failure and death. However, telithromycin is not associated with neuromuscular effects in animal toxicity studies. The objective of this study was to examine the effect of telithromycin on the neuromuscular junction in the isolated rat phrenic nerve-diaphragm preparation and to investigate its postsynaptic effects on the muscle-like nicotinic acetylcholine (ACh) receptors expressed on human TE671 cells. Telithromycin decreased the twitch contraction force of the rat diaphragm muscle in response to phrenic nerve stimulation in a concentration-dependent manner with an IC(50) of 22.3 microM and a maximal inhibition of approximately 70%. The trans-membrane current from the ACh receptors expressed in the TE671 neuromedulloblastoma cells was recorded in the whole-cell patch-clamp configuration. When applied to the TE671 cells, telithromycin caused a dose-dependent inhibition of the nicotinic ACh current with an IC(50) of 3.5 microM and maximal inhibition of nearly 100%. These results indicate that telithromycin inhibits postsynaptic nicotinic ACh receptors in vitro and partially blocks neuromuscular transmission in the isolated rat phrenic nerve-diaphragm preparation. Based on these findings, we propose that exacerbation of myasthenia gravis reported in some patients taking telithromycin results in part from postsynaptic neuromuscular transmission block.

Na+/H+ Exchanger-1 Inhibitors Reduce Neuronal Excitability and Alter Na+ Channel Inactivation Properties in Rat Primary Sensory Neurons

Inhibitors of the Na+/H+ exchanger isoform 1 (NHE-1) have been associated with peripheral neuropathy in rats and dogs. Recent studies suggest that NHE-1 plays an important role in mediating neuronal excitability. To investigate potential NHE-1-mediated mechanisms contributing to neuronal toxicity, we studied the effects of NHE-1 inhibitors on nerve and dorsal root ganglion (DRG) neurons isolated from the adult rat. Compound action potentials (CAPs) were recorded from electrically stimulated sections of isolated sciatic nerve/DRG/root preparations. Whole-cell patch-clamp technique was used to record fast and slow voltage-dependent Na+ currents from dissociated DRG neurons (29-41 microm). Exposures to 1 and 10 microM of a selective NHE-1 inhibitor reduced the amplitude of the CAP recorded from the dorsal root by 33% and 58%, respectively (p < 0.05). The compound had no effect on CAPs recorded from the ventral root. Perfusion of dissociated DRG neurons with NHE-1 inhibitors at 10 and 100 microM shifted voltage-dependent inactivation curves of fast Na+ current by as much as 11 mV (p < 0.001) in the hyperpolarizing direction. No shift was observed in slow Na+ currents. No statistically significant drug effects were observed on voltage-dependent activation or recovery from inactivation of either fast or slow Na+ currents. These results suggest that NHE-1 inhibitors may reduce peripheral neuronal excitability by shifting fast Na+ channels into the inactivated state under physiological conditions. Such effects may underlie peripheral neuropathies reported in rats and dogs with NHE-1 inhibitors.

GlyT1 inhibitor reduces oscillatory potentials of the electroretinogram in rats

Abstract Context: Selective inhibitors of glycine transporter type 1 (GlyT1) increase synaptic glycine concentrations and are being developed to treat cognitive and negative symptoms of schizophrenia. However, increases in systemic glycine levels have been associated with visual disturbances and electroretinogram (ERG) alternations. Objective: To determine whether the selective GlyT1 inhibitor PF-03463275 causes changes in ERG responses in albino rats. Materials and methods: Male Sprague-Dawley rats were administered PF-03463275 subcutaneously at 1, 3 and 10 mg/kg 1 h prior to ERG acquisition. Scotopic and photopic luminance responses, photopic adaptometry and flicker responses were measured. Plasma and vitreous samples were obtained at necropsy for determination of PF-03463275 concentrations. Results: A dose-dependent reduction (up to ∼70%) in the amplitude of the scotopic ERG oscillatory potentials (OPs) was observed following PF-03463275 administration. The amplitude of the OPs was also negatively correlated to the concentration of PF-03463275 in the vitreous humor (r = −0.64, p < 0.0001). With the exception of a small increase in scotopic ERG a-wave amplitude and latency no effects were observed on other ERG parameters tested. Conclusions: We conclude that inhibition of the GlyT1 transporter in the retina causes ERG changes which may underlie recent reports of visual disturbance with GlyT1 inhibitors in clinical trials.

Neurophysiological assessment of sympathetic cardiovascular activity after loss of postganglionic neurons in the anesthetized rat.

The goal of this study was to determine the degree of sympathetic postganglionic neuronal loss required to impair cardiovascular-related sympathetic activity. To produce neuronal loss separate groups of rats were treated daily with guanethidine for either 5days or 11days, followed by a recovery period. Sympathetic activity was measured by renal sympathetic nerve activity (RSNA). Stereology of thoracic (T13) ganglia was performed to determine neuronal loss. Despite loss of more than two thirds of neurons in T13 ganglia in both treated groups no effect on resting blood pressure (BP) or heart rate (HR) was detected. Basal RSNA in rats treated for 5days (0.61±0.10μV∗s) and 11days (0.37±0.08μV∗s) was significantly less than vehicle-treated rats (0.99±0.13μV∗s, p<0.05). Increases in RSNA by baroreceptor unloading were significantly lower in 5-day (1.09±0.19μV∗s) and 11-day treated rats (0.59±0.11μV∗s) compared with vehicle-treated rats (1.82±0.19μV∗s, p<0.05). Increases in RSNA to chemoreceptor stimulation were significantly lower in 5-day treated rats (1.54±0.25μV∗s) compared with vehicle-treated rats (2.69±0.23μV∗s, p<0.05). Increases in RSNA in 11-day treated rats were significantly lower (0.75±0.15μV∗s, p<0.05) compared with both vehicle-treated and 5-day treated rats. A positive correlation of neurons to sympathetic responsiveness but not basal activity was detected. These data suggest that diminished capacity for reflex sympathetic responsiveness rather than basal activity alone must be assessed for complete detection of neurophysiological cardiovascular impairment.

Physiological approaches to assess diminished sympathetic activity in the conscious rat

The purpose of this study was to evaluate functional measures of diminished sympathetic activity after postganglionic neuronal loss in the conscious rat. To produce variable degrees of sympathetic postganglionic neuronal loss, adult rats were treated daily with toxic doses of guanethidine (100mg/kg) for either 5days or 11days, followed by a recovery period of at least 18days. Heart rate, blood pressure, cardiac baroreflex responsiveness, urinalysis (for catecholamine metabolite, 3-methoxy-4-hydroxyphenylethylenglycol; MHPG), and pupillometry were performed during the recovery period. At the end of the recovery period stereology of superior cervical ganglia (SCG) was performed to determine the degree of neuronal loss. Total number of SCG neurons was correlated to physiological outcomes using regression analysis. Whereas guanethidine treatment for 11days caused significant reduction in the number of neurons (15,646±1460 vs. 31,958±1588), guanethidine treatment for 5days caused variable levels of neuronal depletion (26,009±3518). Regression analysis showed that only changes in urinary MHPG levels and systolic blood pressure significantly correlated with reduction of SCG neurons (r2=0.45 and 0.19, both p<0.05). Although cardiac baroreflex-induced reflex tachycardia (345.7±19.6 vs. 449.7±20.3) and pupil/iris ratio (0.50±0.03% vs. 0.61±0.02%) were significantly attenuated in the 11-day guanethidine treated rats there was no significant relationship between these measurements and the number of remaining SCG neurons after treatment (p>0.05). These data suggest that basal systolic blood pressure and urinary MHPG levels predict drug-induced depletion of sympathetic activity in vivo.

Anti-NGF monoclonal antibody muMab 911 does not deplete neurons in the superior cervical ganglia of young or old adult rats

Nerve growth factor (NGF) blocking therapies are an emerging and effective approach to pain management. However, concerns about the potential for adverse effects on the structure and function of the peripheral nervous system have slowed their development. Early studies using NGF antisera in adult rats reported effects on the size and number of neurons in the sympathetic chain ganglia. In the work described here, both young adult (6-8 week) and fully mature (7-8 month) rats were treated with muMab 911, a selective, murine, anti-NGF monoclonal antibody, to determine if systemic exposures to pharmacologically active levels of antibody for 1 month cause loss of neurons in the sympathetic superior cervical ganglia (SCG). State-of-the-art, unbiased stereology performed by two independent laboratories was used to determine the effects of muMab 911 on SCG neuronal number and size, as well as ganglion size. Following muMab 911 treatment, non-statistically significant trends toward smaller ganglia, and smaller and fewer neurons, were seen when routine, nonspecific stains were used in stereologic assessments. However, when noradrenergic neurons were identified using tyrosine hydroxylase (TH) immunoreactivity, trends toward fewer neurons observed with routine stains were not apparent. The only statistically significant effects detected were lower SCG weights in muMab 911-treated rats, and a smaller volume of TH immunoreactivity in neurons from younger rats treated with muMab 911. These results indicate that therapeutically relevant exposures to the anti-NGF monoclonal antibody muMab 911 for 1 month have no effect on neuron numbers within the SCG from young or old adult rats.

A novel endpoint for the assessment of chemotherapy‐induced peripheral neuropathy in rodents: biomechanical properties of peripheral nerve

Chemotherapy‐induced peripheral neuropathy (CiPN) is a frequent adverse effect in patients and a leading safety consideration in oncology drug development. Although behavioral assessment and microscopic examination of the nerves and dorsal root ganglia can be incorporated into toxicity studies to assess CiPN risk, more sensitive and less labor‐intensive endpoints are often lacking. In this study, rats and mice administered vincristine (75 μg kg−1 day−1, i.p., for 10 days in rats and 100 μg kg−1 day−1, i.p., for 11 days in mice, respectively) were employed as the CiPN models. Behavioral changes were assessed during the dosing phase. At necropsy, the sural or sciatic nerve was harvested from the rats and mice, respectively, and assessed for mechanical and histopathological endpoints. It was found that the maximal load and the load/extension ratio were significantly decreased in the nerves collected from the animals dosed with vincristine compared with the vehicle‐treated animals (P < 0.05). Additionally, the gait analysis revealed that the paw print areas were significantly increased in mice (P < 0.01), but not in rats following vincristine administration. Light microscopic histopathology of the nerves and dorsal root ganglia were unaffected by vincristine administration. We concluded that ex vivo mechanical properties of the nerves is a sensitive endpoint, providing a new method to predict CiPN in rodent. Gait analysis may also be a useful tool in these pre‐clinical animal models.

Abstract 4385: Effect on retinal function as a mechanism for vision disorders with crizotinib (PF-02341066)

Crizotinib is a small-molecule receptor tyrosine kinase inhibitor of anaplastic lymphoma kinase (ALK) and c-Met. Crizotinib has been shown to be effective in patients with ALK-positive non-small cell lung cancer. One common adverse event that has been observed is a grade 1 vision disorder (∼45% of patients) characterized as seeing shadows or streaks during changes in lighting conditions. Onset occurs during the first weeks of treatment, with most cases improving with continued treatment and not requiring treatment discontinuation. Such vision disorders were not predicted from preclinical toxicological evaluation of crizotinib. To further understand the possible mechanism of this vision disorder, a rat electroretinography (ERG) study was conducted to determine if retinal function is altered following repeat-dose administration of crizotinib. Male Long-Evans rats (8/group) were administered crizotinib by oral gavage at a dose of 100 mg/kg/day for up to 29 days, during which an evaluation of in-life parameters (clinical signs, body weight, and food consumption), ophthalmic examinations, and ERG were performed. As part of the ERG assessment, five light-stimulation protocols were applied to test scotopic (dark-adapted) and photopic (light-adapted) luminance responses, photopic and scotopic adaptometry, and photopic flicker responses. Terminal plasma and vitreous fluid were examined for crizotinib concentrations in main study animals, and a toxicokinetic profile was obtained from a satellite group of 6 animals. Results of the ERG measurements identified a statistically significant (∼30%) reduction in mean b-wave amplitude relative to control during the initial 16 and 32 minutes of dark adaptation on Days 15 (p Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4385. doi:10.1158/1538-7445.AM2011-4385