Circulating nerve growth factor level changes during oxaliplatin treatment-induced neurotoxicity in the rat

BACKGROUND

Oxaliplatin neurotoxicity represents a clinically-relevant problem and its etio-pathogenesis is still unknown. We explored the possible role of some neuronal growth factors ("neurotrophins") during the course of oxaliplatin sensory neuronopathy.

MATERIALS AND METHODS

In our rat model two different doses of oxaliplatin were used (2 and 3 mg/kg i.v. twice weekly for 9 times). The neurotoxicity of the treatment was assessed with neurophysiological and pathological methods and serum neurotrophin levels were measured by ELISA.

RESULTS

Both oxaliplatin-treated groups showed the neurophysiological and neuropathological changes which mimic the chronic effects of oxaliplatin administration in humans, e.g. reversible sensory impairment due to dorsal root ganglia neuron damage. These changes were associated with a significant and dose-dependent reduction only in the circulating level of nerve growth factor (NGF), which returned to normal values after neurophysiological and pathological recovery.

CONCLUSION

This specific association between neurological impairment and NGF modulation indicates that NGF impairment has a role in the neurotoxicity of oxaliplatin.

Cisplatin-induced peripheral neurotoxicity in rats reduces the circulating levels of nerve growth factor

The pathogenesis of the neurotoxicity of most antineoplastic drugs is unknown. Recent reports suggest that changes in the circulating levels of nerve growth factor (NGF) might be related to the dorsal root ganglia sensory neuron damage induced by cisplatin (CDDP), the first member of a family of widely used and very effective platinum-derived anticancer agents. Using a well-characterized model of CDDP neurotoxicity, we demonstrated that the NGF circulating level decreased during chronic CDDP administration in close accordance with the clinical course and returned to normal levels after recovery from the neurotoxic damage. Moreover, these changes were restricted to NGF and did not involve other trophic factors of the same neurotrophin family. Our findings are in agreement with previous in vitro and in vivo results and further suggest that NGF plays a specific role in the course of CDDP-induced primary sensory neuron damage.

Women with pregnancy-related polymyositis and high serum CK levels in the newborn

Two previously healthy women developed an inflammatory myopathy before the term of their first pregnancy. Skeletal muscle biopsy led to a diagnosis of T cell-mediated polymyositis. Both babies were healthy, but their serum creatine kinase levels remained elevated for a few months after birth. Their mothers did well after corticosteroid treatment.

Effects of different schedules of oxaliplatin treatment on the peripheral nervous system of the rat

The aim of this study was to determine the influence of oxaliplatin scheduling on the onset of peripheral neurotoxicity and ototoxicity in a rat model. Animals were treated with four different schedules of oxaliplatin using two cumulative doses (36 and 48 mg/kg intraperitoneally (i.p.)). The neuropathological examination evidenced dorsal root ganglia (DRG) nucleolar, nuclear and somatic size reduction with nucleolar segregation in the treated rats. Sensory nerve conduction velocity (SNCV) was reduced after oxaliplatin treatment, while the auditory pathway was unaffected. After treatment, platinum was detected in the kidney, DRG and sciatic nerve. After a 5-week follow-up period, recovery of the pathological changes in the DRG and sciatic nerves occurred, although platinum was still detectable in these tissues. The following conclusions may be drawn: the main targets of oxaliplatin neurotoxicity were the DRG; the shorter the interval between the injections, the higher the severity of peripheral neuropathy and this was also related to the cumulative oxaliplatin dose; the peripheral neurotoxicity tended to be reversible; ototoxicity was absent even with high cumulative doses of oxaliplatin.

Immunomodulating effects of extracorporeal photochemotherapy in rat experimental allergic encephalomyelitis

Experimental allergic encephalomyelitis (EAE) is a well-established model of human multiple sclerosis that is commonly used to evaluate the possible effectiveness of new treatments in this disease. Extracorporeal photochemotherapy (ECP) is an immunomodulating procedure currently used in several non-neurological diseases that, like multiple sclerosis, are likely to be due to T-cell-mediated autoimmunity. In this study we examined the effect of ECP using the EAE paradigm in the Lewis rat. In our model, ECP induced a significant modulation in peripheral blood T-cell distribution, changes which are typical of EAE. Remarkably, this effect was closely correlated with the clinical and pathological results, which showed reduced severity of the disease in the ECP-treated EAE animals vs. the EAE alone rats. We conclude that ECP induces modifications in the immunological events that occur during the course of EAE in rats, thus giving support to the hypothesis that it could be used in the treatment of multiple sclerosis.

Effect on the peripheral nervous system of systemically administered dimethylsulfoxide in the rat: a neurophysiological and pathological study

The issue of dimethylsulfoxide (DMSO) neurotoxicity is an important one, given its wide use in experimental toxicology as a solvent for hydrophobic substances. We examined the effect of the intraperitoneal administration of different DMSO solutions (1.8-7. 2%) on the peripheral nervous system of Wistar rats treated for 10 consecutive days and followed-up for an additional 45 days. DMSO administration induced a dose-dependent reduction in nerve conduction velocity, with complete recovery occurring in the follow-up. No structural changes were found in the sciatic nerve at 1.8% and 3.6% DMSO concentrations, suggesting that the mechanism of action of DMSO involves a functional impairment (i.e. conduction block) similar to that already described for this substance in isolated systems. However, when DMSO was administered at the 7.2% concentration, evident structural changes were observed in the sciatic nerve, with myelin disruption and uncompacted myelin lamelle. The neurophysiological and pathological changes observed in our study are severe enough to merit careful consideration in the course of experimental studies involving DMSO as a solvent for drugs which are under evaluation for their potential neurotoxicity.

Lipid-free versus lipid-bound P2 protein-induced experimental allergic neuritis: clinicopathological, neurophysiological, and immunological study

The P2 protein of the peripheral nervous system myelin is a neuritogenic protein capable of inducing experimental allergic neuritis (EAN) in the Lewis rat. It has been suggested that the addition of some lipids to the protein isolated in the lipid-free form might enhance its immunogenicity. In this study, we compared lipid-free P2 (the EAN factor) and the corresponding lipid-bound form of the protein regarding their ability to induce EAN. Lipid-bound P2, copurified with all the myelin lipids, shows a conformation different from that of LF-P2. The timing of disease and the clinical scores of lipid-bound P2-induced EAN animals (n = 23) did not differ statistically from those injected with lipid-free P2 (n = 23), with only a tendency to higher clinical severity in the former group. Tail nerve conduction velocities did not differ in the two groups and in both were significantly lower in comparison to Freund adjuvant controls (n = 8). Inflammation and demyelination predominated in the spinal roots and were less evident in the sciatic nerve for both groups of animals. The ELISA determination of antibodies to lipid-free and lipid-bound P2 revealed the development of antibodies recognizing the lipid-free form of the protein in both groups of animals. Our results stand in contrast to results of previous studies performed after addition of exogenous lipids to the P2 purified in the lipid-free form and indicate that lipid-bound P2 is not significantly more immunogenic than lipid-depleted P2.

Creutzfeldt-Jakob disease with a novel four extra-repeat insertional mutation in the PrP gene

OBJECTIVE

To investigate the role of a short insertional mutation in the prion protein (PrP) gene (PRNP) in prion disease pathogenesis.

BACKGROUND

The genetic forms of Creutzfeldt-Jakob disease (CJD) are associated with point or insertional mutations in PRNP. Whereas patients with five, six, seven, eight, and nine extra octapeptide repeats show an autosomal dominant pattern of inheritance and features of CJD, Gerstmann-Sträussler-Scheinker disease, or atypical dementia, patients with one, two, or four extra repeats have typical CJD and lack a family history of neurologic disorder.

METHODS

A genetic, neuropathologic, and biochemical study was carried out in a 65-year-old patient with clinical features of sporadic CJD.

RESULTS

A novel four extra-repeat insertional mutation of PRNP was found in the patient and in his 59-year-old healthy sister. The patient showed spongiosis, nerve cell loss, and gliosis associated with diffuse PrP immunoreactivity in the cerebral cortex, subcortical gray structures, and cerebellum. A peculiar aspect was the presence of focal PrP deposits in the basal ganglia and hypothalamus, superimposed to diffuse PrP immunoreactivity. The biochemical analysis revealed that both mutant and wild-type PrP participated in the pathologic process, and that the protease-resistant core of the altered PrP isoforms was distinct from that observed in sporadic, acquired, and other genetic forms of CJD.

CONCLUSION

These findings support the view that the four extra-repeat insertion in PRNP is a pathogenic mutation with low penetrance rather than a benign polymorphism, and suggest that this mutation results in the formation of a distinct PrP conformer.

Distribution of paclitaxel within the nervous system of the rat after repeated intravenous administration

The distribution of paclitaxel (Taxol) within the central and peripheral nervous system after repeated administration of this antineoplastic agent is still largely unknown. In this study we determined for the first time paclitaxel tissue concentration in the brain, spinal cord, dorsal root ganglia (DRG) and sciatic nerve using an experimental paradigm in the rat which reproduces the features of paclitaxel peripheral neurotoxicity in humans. Pathological confirmation of the onset of paclitaxel-induced peripheral neurotoxicity was performed. In order to achieve reliable results even with low concentrations of paclitaxel, a newly reported analytical method (high-performance liquid chromatography with tandem mass spectrometry) was used. We demonstrated that paclitaxel has easy access to the DRG, where it accumulates, while the lowest concentrations of the drug were measured in the brain. The intermediate concentrations of paclitaxel observed in the sciatic nerve and spinal cord may be due to paclitaxel transport along the centrifugal and centripetal branches of the DRG neuron axons.

Cisplatin-induced DNA-platination in experimental dorsal root ganglia neuronopathy

The mechanism(s) and site(s) of the neurotoxic effect of cisplatin (CDDP) are still not entirely elucidated. A more detailed knowledge of these aspects of CDDP treatment might be useful to obtain a better understanding of the pathogenesis of its peripheral neurotoxicity, which is the dose-limiting side effect of CDDP. In the present study, the occurrence of CDDP-induced DNA-platination in dorsal root ganglia (DRG) of rats was evaluated in relation to DRG neuron pathological changes and CDDP-induced neuronopathy. Eight adult Wistar rats were treated with 2 mg/kg i.p. CDDP twice weekly for 9 times to induce sensory peripheral neuropathy. DNA-platination in specimens of DRG and kidney was measured immunohistochemically, with a polyclonal antibody (GPt) detecting CDDP-induced Pt-DNA adducts. Results were compared with those of untreated rats. Chronic CDDP-induced neurotoxicity, in a well described experimental model of chronic CDDP neurotoxicity in the Wistar rat, was confirmed by sensory DRG neuronopathy with secondary neuropathy, and demonstrated by reduced pain detection, decreased nerve conduction velocity in the tail nerve as well as morphological and morphometric changes in DRG neurons. Nuclear immunostaining for Pt-DNA adducts was observed in tubular cells of the kidney in 75% of the evaluated CDDP-treated rats, while in DRG cells CDDP-induced Pt-DNA adducts formation was found in 43% of the evaluated CDDP-treated rats. CDDP-induced DNA-platination was demonstrated in rat DRG neurons using a schedule of chronic CDDP administration which induced the onset of a sensory neuronopathy with secondary peripheral neuropathy. This finding further supports the hypothesis that CDDP is neurotoxic because it directly damages the DRG neurons.

Clinical presentation and outcome of Guillain-Barré and related syndromes in relation to anti-ganglioside antibodies

We correlated the clinical features of 78 patients with Guillain-Barré syndrome (GBS) or related variants, with the presence of serum antibodies to the gangliosides GM1, GM2, GD1a, GD1b and GQ1b in order to determine whether these antibodies may influence the clinical presentation or outcome of GBS. Sixty-three patients had typical GBS (81%), nine a pure motor form (11%), three a paraparetic form (4%), and three had Miller Fisher syndrome (MFS). IgG or IgM (or both) anti-ganglioside antibodies were found by ELISA in 37% of patients, including 36% with typical, 33% with pure motor and 100% with MFS. Beside the constant occurrence of anti-GQ1b antibodies in patients with MFS (P<0.00001), the other clinical forms were not associated with a specific anti-ganglioside reactivity. Anti-GM1 and anti-GD1a antibodies tended to be associated with a worse disability at 6 month than other or no reactivity and, similarly to anti-GM2 antibodies, with a more frequent respiratory impairment. Anti-GM2 and anti-GD1b antibodies were always associated with typical GBS and, in all but one patient, with a complete recovery; still they were found in only 13 and 3%, respectively, of the patients with this presentation. Anti-GQ1b antibodies, though always associated with ophthalmoplegia and ataxia in both MFS and GBS, were found in only 36 and 26%, respectively, of patients with these symptoms. Even if different anti-ganglioside antibodies tend to be associated with some clinical features possibly suggesting that they may influence the clinical presentation or outcome, with the exception of anti-GQ1b antibodies for ophthalmoplegia and ataxia, they do not permit to predict the clinical presentation or outcome in individual patients.

Effect of recombinant human nerve growth factor on cisplatin neurotoxicity in rats

In this study we evaluated the effect of recombinant human nerve growth factor (rhNGF) on cisplatin (CDDP)-induced sensory neuronopathy in an experimental paradigm in the rat. Young adult female Wistar rats were treated with CDDP (2 mg/kg ip twice weekly for nine times) alone or in combination with rhNGF (1 mg/kg sc on alternate days). The effect of CDDP +/- NGF treatment was evaluated with behavioral (tail-flick test) and neurophysiological (nerve conduction velocity in the tail) methods immediately after treatment and after a follow-up period of 6 weeks. Pathological and morphometrical examinations of the dorsal root ganglia (DRG) and sciatic and saphenous nerves were also performed. rhNGF treatment induced a significant reduction in the CDDP-induced decrease in nerve conduction velocity (P < 0.05), and this was associated with a significant protection against the decrease in somatic (P < 0.05), nuclear (P < 0.05), and nucleolar size (P < 0.01) caused by CDDP treatment. However, for each of the parameters examined the neuroprotection obtained with rhNGF treatment was not complete. At the follow-up examination no differences between the three groups were observed in tail-flick test and nerve conduction velocity. We conclude that rhNGF, administered according to the schedule used in this experiment, exerts a biologically significant neuroprotective effect against CDDP peripheral neurotoxicity.

Resveratrol, map kinases and neuronal cells: might wine be a neuroprotectant?

Alcohol is noxious to the brain and peripheral nervous system. However, wine contains substances that may have positive biological and pharmacological effects. Resveratrol is the most studied and probably the most active of these substances. This naturally occurring compound, which is present in wine and grapes, reduces oxidative stress in neuronal-like cell cultures. We have shown that resveratrol induces phosphorylation of the mitogen-activated protein (MAP) kinase family members, extracellular regulated kinase 1 (ERK1) and ERK2, in the human neuroblastoma SH-SY5Y cells in vitro at much lower concentrations than those found in the plasma of rats after oral wine administration. MAP kinases are involved in numerous different aspects of signal transduction in the cells. In particular, phosphorylation of ERK2 has been related to the synaptic changes at the basis of memory and learning processes. These findings, together with our own, on resveratrol-induced activation of MAP kinases in human neuronal-like cells, and previously published epidemiological studies which have demonstrated an inverse relationship between moderate wine intake and dementia, suggest that wine (not alcohol) may have a positive effect on nervous cells.

Chronic axonal sensory and autonomic polyneuropathy without motor involvement: a new 'chronic inflammatory neuropathy?'

We report the case of a woman with axonal sensory and autonomic neuropathy lasting several months who improved in association with steroid administration. During the course of her disease and in the follow-up, the patient underwent repeated cerebrospinal fluid (CSF) examinations, neurophysiological somatic, autonomic nervous system studies and sural nerve biopsy. Clinical and laboratory assessments demonstrated the occurrence of a monophasic, chronic sensory and autonomic neuropathy. A sural nerve biopsy suggested an axonopathy. After a progressive worsening of symptoms lasting about 6 months, steroid treatment was started and within 6 months a complete recovery, with normalization of the CSF findings, was observed. Although the 'chronic inflammatory neuropathies' are still debated entities, the features of this chronic, exclusively sensory and autonomic neuropathy are new, and the occurrence of a high protein level in the CSF, together with the favorable outcome associated with steroid treatment, suggests that our case might be another variant in this debated area.

Neuro- and ototoxicity of high-dose carboplatin treatment in poor prognosis ovarian cancer patients

BACKGROUND

Hematopoietic toxicity of high-dose carboplatin (HD-CBDCA) chemotherapy can be managed effectively with autologous blood cell support, but no conclusive data are available on its neuro- and ototoxicity.

PATIENTS AND METHODS

We determined the neuro- and ototoxicity of HD-CBDCA in 10 patients affected by advanced ovarian cancer. HD-CBDCA was delivered as 24-hour continuous infusion or as 5-day schedules. Each patient underwent an extended clinical and instrumental neurological and otological evaluation before, during and after treatment.

RESULTS

After HD-CBDCA only 1 patient had a clinically-evident peripheral neuropathy, while 3 additional patients had only distal paresthesias. Neurophysiological examination evidenced mild, although diffuse, sensory nerve impairment. Motor nerve impairment was also occasionally observed. All the sensory and motor pathological changes had a favorable course during the follow-up period. Ototoxicity was more severe than neurotoxicity and, in one case it was dose-limiting and audiologic impairment tended to remain constant also in the follow-up period.

CONCLUSIONS

HD-CBDCA treatment can be tolerated by most of the patients, but careful monitoring of neuro- and, especially, ototoxicity should be planned.

Retinoic acid differentiated SH-SY5Y human neuroblastoma cells: an in vitro model to assess drug neurotoxicity

PURPOSE

To evaluate whether retinoic acid (RA) differentiated human neuroblastoma (HN) SH-SY5Y cells are a suitable and reliable model to test the neurotoxicity of chemotherapic drugs without the confusing effects of the neurotrophic factors commonly used to induce neuronal differentiation.

METHODS

Cultures of the SH-SY5Y cell line were exposed to RA to induce neuronal differentiation which was assessed by measuring the neurite length. The effect of increasing concentrations of cisplatin (CDDP) on neurite outgrowth was determined. Cyclophosphamide (CTX) was used as negative control.

RESULTS

CDDP induced a significant reduction in the mean neurite length in a dose dependent manner. The neurotoxic effect of CDDP was reversible. Cyclophosphamide did not induce changes in RA differentiated HN cells.

CONCLUSIONS

RA differentiated HN cells are sensitive to the neurotoxic effect of CDDP and the course of the changes is similar to that observed in clinical practice and in in vivo experimental models. Therefore, this model is proposed as a screening method to test the neurotoxicity of chemotherapy drugs and the possible effect of neuroprotectant molecules and drugs.

Polyneuropathy due to cobalamin deficiency in the rat

In the present study, we investigated the peripheral nervous system (PNS) (both in terms of its ultrastructure and in terms of its function) of rats made cobalamin (Cbl)-deficient either through total gastrectomy or through prolonged feeding on a Cbl-deficient diet. In both these types of Cbl-deficient neuropathies we found: (a) ultrastructurally, intramyelin and endoneural edema, with no or minimal axonal damage in the PNS, in dorsal root ganglia, and the ventral and dorsal rootlets of the spinal cord; (b) electrophysiologically, a significant reduction in the nerve conduction velocity, consistent with that reported in (a); (c) morphometrically, a significant reduction in the density of myelinated fibers both in the sciatic nerve and in the peroneal nerve. All these pathological changes were reversed by chronic postoperative administration of Cbl into totally gastrectomized (TGX)-rats, hinting at the specificity of the damage itself in relation to the permanent Cbl-deficient status of the TGX-rats. No signs of segmental demyelination or remyelination were found. We also observed a turning of type I fibers into type II fibers in the soleus muscle of all our Cbl-deficient rats, however the Cbl deficiency had been induced. This muscular change was still present in TGX- and Cbl-treated rats, and it cannot be related to a malnutrition status, since it has been observed also in rats fed a Cbl-deficient diet. All these results demonstrate that Cbl deficiency strongly affects rat PNS within different parameters.

Carboplatin toxic effects on the peripheral nervous system of the rat

BACKGROUND

The most striking of carboplatin's advantages (CBDCA) over cisplatin (CDDP) is its markedly reduced rate of neurotoxic effects. However, the use of CBDCA higher-intensity schedules and the association with other neurotoxic drugs in polychemotherapy may cause some concern about its safety with respect to peripheral nervous system damage.

MATERIALS AND METHODS

Two different schedules of CBDCA administration (10 mg/kg and 15 mg/kg i.p. twice a week for nine times) were evaluated in Wistar rats. Neurotoxicity was assessed for behavioral (tail-flick test), neurophysiological (nerve conduction velocity in the tail nerve), morphological, morphometrical and analytical effects.

RESULTS

CBDCA administration induced dose-dependent peripheral neurotoxicity. Pain perception and nerve conduction velocity in the tail were significantly impaired, particularly after the high-dose treatment. The dorsal root ganglia sensory neurons and, to a lesser extent, satellite cells showed the same changes as those induced by CDDP, mainly affecting the nucleus and nucleolus of ganglionic sensory neurons. Moreover, significant amounts of platinum were detected in the dorsal root ganglia and kidney after CBDCA treatment.

CONCLUSIONS

CBDCA is neurotoxic in our model, and the type of pathological changes it induces are so closely similar to those caused by CDDP that it is probable that neurotoxicity is induced in the two drugs by the same mechanism. This model can be used alone or in combination with other drugs to explore the effect of CBDCA on the peripheral nervous system.

Subacute combined degeneration in totally gastrectomized rats: an ultrastructural study

Severe permanent cobalamin (Cbl) deficiency was induced in rats either by total gastrectomy (TG) or through prolonged dietary Cbl deprivation. This paper deals with an ultrastructural investigation of different parts of the central nervous system (CNS) of rats made Cbl-deficient through one of these of two procedures. In both totally gastrectomized (TGX) rats and in rats chronically fed a Cbl-deficient diet, we observed intramyelin edema, with splitting of the lamellae, and interstitial edema affecting the white matter, mainly in the spinal cord (SC). These lesions were also present in the subcortical white matter, although to a lesser degree. In both TGX-rats and in rats chronically fed a Cbl-deficient diet the pyramidal tract and the optic nerve were completely spared. Vascular lesions were never observed. Intramyelin edema and interstitial edema of the white matter account for the patchy myelopathic spongy vacuolation, which is the histological hallmark of human subacute combined degeneration and has been previously seen in SC white matter of TGX-rats. Macro- and micro-glial cells in the white matter were activated, at least as seen ultrastructurally. Interestingly enough, there were activated glial cells even in the gray matter, in which neurons showed absolutely no alterations. Chronic subcutaneous Cbl administration of TGX-rats partially repaired the CNS damage. However, the amelioration produced by this treatment was greater when Cbl was given shortly after TG than when given three and four months after TG, i.e. when the lesions have already been formed.

Experimental cisplatin neuronopathy in rats and the effect of retinoic acid administration

Peripheral nervous system alterations were induced in adult rats by administration of cisplatin (CDDP) 2 mg/kg twice weekly for 4.5 weeks. Dorsal root ganglion neurons showed pathological changes. Morphometric determinations demonstrated a reduction in size of the somatic, nuclear and nucleolar area. The nucleoli were the most involved subcellular structures. Nerve conduction velocity and the tail-flick test for pain were both significantly altered in CDDP treated rats, whereas the rota-rod test for coordination revealed no changes in either treated or control rats. Analytical determinations demonstrated platinum accumulation in the DRG of CDDP treated rats. Spontaneous recovery, demonstrated by morphometric, electrophysiological, functional and analytical determinations, occurred after treatment discontinuation within about 7 weeks. A pilot study of the possible neuroprotective action of retinoic acid (RA) was also performed with this model of cisplatin neuronopathy. The rationale for using RA was based on its assumed antioxidant and neurotrophic effect. However, RA failed to prevent morphometric, electrophysiological, functional and analytical alterations induced by CDDP on DRG neurons. RA induced only a mild generalized protective effect.

Antibodies of the anti-Yo and anti-Ri type in the absence of paraneoplastic neurological syndromes: a long-term survey of ovarian cancer patients

In recent years several authors have described a close correlation between circulating antineuronal antibodies of different types and the occurrence of paraneoplastic neurological syndromes. Because this has not been widely accepted, we screened 300 serum samples from 181 ovarian cancer patients for the presence of circulating antineuronal antibodies by immunofluorescence. The findings were confirmed by immunoblotting. In 11 patients circulating antineuronal antibodies were detected. In 4 patients they were classified as anti-Yo and in 7 as anti-Ri, titres ranging from 1:400 to 1: 204,800. All the patients underwent thorough neurological and neurophysiological investigations, with special regard to paraneoplastic syndrome. None of them had symptoms pointing to a paraneoplastic neurological syndrome, although patients were followed up to 2 years after the first examination. Thus the frequency of circulating antineuronal antibodies in ovarian cancer patients is higher than the frequency of paraneoplastic syndromes, and antibody positivity does not necessarily lead to the appearance of a neurological paraneoplastic syndrome.

Neurotoxicity and ototoxicity of cisplatin plus paclitaxel in comparison to cisplatin plus cyclophosphamide in patients with epithelial ovarian cancer

PURPOSE

To compare the neurotoxicity and ototoxicity of combination cisplatin plus paclitaxel versus cisplatin plus cyclophosphamide using extensive clinical and instrumental evaluation.

PATIENTS AND METHODS

Forty-six of 51 consecutive patients affected by-epithelial ovarian cancer seen in our institution between October 1994 and August 1995 entered the study. After randomization, they were assigned to receive cisplatin 75 mg/m2 every 3 weeks associated with cyclophosphamide 750 mg/m2 (CC group, n = 22) or paclitaxel 175 mg/m2 over a 3-hour infusion (CP group, n = 24). Treatment was repeated six times in 43 patients and nine times in 25. Before treatment and after three, six, and nine courses of chemotherapy, patients underwent clinical and instrumental neurologic and otologic examinations.

RESULTS

Mild sensory impairment was evident even after only three courses of both treatments and signs and symptoms were more severe at the end of treatment. On clinical grounds only, it was possible to demonstrate after six and nine courses a difference between CC and CP treatment, due to the involvement in some CP patients of pain and thermal sensory modalities. However, the overall severity of the neuropathy was similar. Audiometric parameters demonstrated a more negative outcome after treatment in CC compared with CP patients. However, the different severity of the involvement was closely correlated to this initial difference in audiologic performance.

CONCLUSION

Up to nine courses of chemotherapy, the CC and CP schedules are similar in terms of severity of neurotoxicity and ototoxicity when patients are evaluated during and immediately after treatment. With the doses used in our study, these toxicities are not dose-limiting. Our results suggest that most of the toxic effects observed during the treatment were due to cisplatin.

Effect on the peripheral nervous system of the short-term intravenous administration of paclitaxel in the rat

The effectiveness of paclitaxel (Taxol) in the treatment of different tumors is well-known but, on the other hand, there is little information regarding its neurotoxicity and the mechanism(s) underlying this potentially severe side effect. In this study, using behavioral, neurophysiological, morphological and morphometric methods, we evaluated the effect of intravenous administration of paclitaxel on the rat nervous system. After 2 pilot studies, 40 female Wistar rats were treated with intravenous paclitaxel via a catheter placed in the jugular vein, while 20 animals were used as controls. Paclitaxel dissolved in ethanol/Tween 80/saline (5/5/90%) was administered 5 times over a period of 10 days. At the end of the experiment half the surviving animals in each group were evaluated and sacrificed (day 11), while the rest of the rats were evaluated and sacrificed on day 25. On day 11 the treated animals had significant impairment in pain perception (tail-flick test), coordination (rota-rod test) and nerve conduction velocity in the tail nerve. At the light microscope minimal axonal damage and Schwann cell activation were observed in the sciatic nerve. At the electron microscope microtubular accumulation was present within the axon in dorsal and ventral spinal roots and in the sciatic nerve. On day 25 the behavioral tests were normal in treated rats, while the nerve conduction velocity was still moderately reduced in comparison with the controls. At the electron microscope a morphological examination evidenced that microtubular accumulation was less severe, but still evident, especially in the sciatic nerve. Morphometric determinations performed on days 11 and 25 did not evidence differences between paclitaxel-treated rats and controls. The results of this study, the first in which an extended examination of the nervous system of animals treated intravenously with paclitaxel has been carried out, suggest that short-term administration of the drug induces mainly reversible changes in the peripheral nerves and spinal roots. Microtubules seem to be the main target of paclitaxel neurotoxicity, in much the same way as has been described for its antineoplastic activity. Finally, no pathological changes were seen in the neuronal bodies of the spinal cord and dorsal root ganglia. This model may be used for further studies with combination treatments with other antineoplastic or neuroprotective agents.

Neuroprotectant drugs in cisplatin neurotoxicity

Cisplatin (CDDP) is one among the most effective and widely used anticancer drugs. Its use is, however, often limited by its peripheral neurotoxicity, which may be severely disabling and sometimes not reversible. To prevent or reduce CDDP peripheral neurotoxicity several "neuroprotective" drugs have been proposed. This goal is of extreme importance in the treatment of cancer patients, especially in view of the better results obtained by anticancer chemotherapy in terms of longer disease-free survival which has made even more crucial than in the past the point of the quality of life of the long-surviving patients. The data of pre-clinical and clinical studies with neuroprotectant agents are often conflicting, in some cases because of inadequate methods of evaluation and/or study design used to examine their effectiveness. The aims of this review will be 1) to discuss and describe the most appropriate methods of evaluation of CDDP and neuro-protectant drugs in experimental in vitro and in vivo pre-clinical studies 2) to evaluate critically the results of the clinical trials reported so far with the combined treatment and 3) to explore the possible future strategies to achieve neuroprotection during high-dose CDDP treatment in humans.

Neuropeptides and morphological changes in cisplatin-induced dorsal root ganglion neuronopathy

Dorsal root ganglia (DRG) neuronopathy was induced in rats by chronic treatment (2 mg/kg twice a week for nine injections) with the antineoplastic drug cisplatin. Morphological alterations and changes in peptide [calcitonin gene-related peptide (CGRP), substance P, galanin (Gal), and somatostatin] concentration were studied in the DRG, the spinal cord, and the sciatic nerve. Peptide concentration was increased in DRG neurons, with CGRP and Gal showing the highest increase. Conversely, in the sciatic nerve there was a general decrease in peptide content. In DRG a reduction in the nuclear, cytoplasmic, and nucleolar areas of primary sensory neurons was evident and was accompanied by clear-cut aspects of nucleolar structural damage. In peripheral nerves only extensive morphometric determinations could evidence a reduction in nerve conduction velocities and impairment in pain detection and coordination. Some of the nerve fibers presented axonal and adaxonal accumulations, suggesting the presence of an axonopathy. These results confirm that DRG cells are the primary target of cisplatin-induced neurotoxicity. Milder alterations can be detected in peripheral nerves. The increase in peptide concentration in DRG is probably due to cisplatin-related damage to the axonal transport system rather than to an increased synthesis.

Cryoglobulinaemic neuropathy: lack of progression in patients with good haematological control

Nine consecutive patients with essential mixed cryoglobulinaemia type II (EMCII) were investigated by a follow-up study of 3-6 years (mean 5.3), to verify the long-term outcome of the peripheral nerve involvement in the disease. At the beginning of the study, a peripheral neuropathy, mild or moderate, mainly sensory, was present approximately in 50% of cases. All patients maintained good general conditions, low-staging of EMCII, were well controlled by therapy or did not need any medical treatment. During the neurological follow-up all patients had regular controls, usually at 1-year intervals, and only slight changes were seen with respect to the first evaluation. Therefore, we conclude that peripheral neuropathy during EMCII is less progressive that it can be thought at least in patients with benign course of the disease.

An ultrastructural study of neuronal changes in dorsal root ganglia (DRG) of rats after chronic cisplatin administrations

In humans, the main dose-limiting side-effect of cisplatin (CDDP) treatment is a peripheral sensory neuropathy secondary to dorsal root ganglion (DRG) neuron involvement. To investigate further for neuronal alterations responsible for CDDP neurotoxicity we undertook the present experimental ultrastructural study, based on observations of 3 different groups of rats (6 animals in each group). Group A rats were treated with 1 mg/kg weekly for 9 weeks; Group B with 2 mg/kg weekly for 9 weeks; and group C rats served as untreated controls. At the end of the experiment, rats were perfused with 3% glutaraldehyde and lumbar DRGs were prepared for electron microscopic observations. In CDDP-treated rats somatic, nuclear and, above all, nucleolar size was reduced. Ultrastructurally, the nucleolus was the most affected structure. Nucleolar alterations were quantified morphometrically. Less marked changes were seen in the nucleus and in the RER and Golgi apparatus of the cytoplasm. The number of lysosomes and lipofuscins was greatly increased in CDDP-treated rats. The ultrastructural alterations observed in CDDP rats suggest that CDDP may be neurotoxic due to a reduction in protein synthesis. This assumption would explain why cells such as neurons, which are non replicating, but which have a high rate of protein synthesis, may be the target of the neurotoxic action of CDDP. The lack of an efficient blood/nerve barrier in the DRG explains the involvement of this particular type of neuron.

Ultrastructural aspects of DRG satellite cell involvement in experimental cisplatin neuronopathy

Different substances may induce neurological impairment, clinically expressed as peripheral neuropathies, due to damage of the neuronal bodies (neuronopathy) of sensory or motor neurons. Neuronopathies have generally been studied referring to neurons, although other cellular components may also be damaged. Cisplatin (CDDP) is known to be neurotoxic to the neurons of the dorsal root ganglia (DRG). The scarcity of information as to the possible involvement and role played by dorsal root ganglion (DRG) satellite cells in neuronopathies prompted this study using the chronic DRG neuronopathy induced by the repeated administration of CDDP in rats as a model. Eighteen female Wistar rats were treated according to 3 different schedules of CDDP administration (6 rats for each group). Six further animals were used as controls. At the end of the experiment the L4-L5-L6 dorsal root ganglia were examined at the light and electron microscope. Ag-NOR reaction was also examined in 4 further CDDP-treated rats and 4 controls. Pathological changes in satellite cells of animals treated with CDDP were remarkable in the nucleus where heterochromatin clumps were reduced or even completely absent. Morphometric analysis of the area occupied by heterochromatin indicated that this nuclear component decreased in an exposure-time dependent manner. Frequently, nucleolar-like structures became apparent in the nucleus of the rats treated with the higher doses of CDDP. Ag-NOR positive regions in the nuclei of treated rats were increased with respect to the controls. Cytoplasmic changes in DRG satellite cells of CDDP treated rats were limited, being characterized by an increased electron-density of the matrix. In treated rats deep invaginations between satellite cells and the neuronal surface were evident, leading to the formation of vacuoli. The interstitial connective space often showed edematous areas. Our observations demonstrate that in chronic cisplatin neuronopathy, DRG satellite cells are also involved in the pathological changes induced by drug exposure, and that these changes may be interpreted as being mainly reactive.

Experimental peripheral neuropathy induced in adult rats by repeated intraperitoneal administration of taxol

Taxol, a natural extract with antineoplastic properties, is known to be neurotoxic in humans. Its neurotoxicity after systemic administration, however, has never been studied in detail at the morphological level in humans and in animals. In this study we administered taxol intraperitoneally to female Wistar rats and we performed an extended neurophysiological and morphological examination of the peripheral nerves, dorsal root ganglia, spinal rootlets, and spinal cord. The results obtained in this experimental model indicate that taxol induces pathological changes mainly in the peripheral nerves, but they are present also in the ventral and dorsal spinal rootlets and spinal dorsal column fibers. The dorsal root ganglia and spinal cord neurons were, on the contrary, unaffected. The most impressive change induced by systemic taxol administration was intraaxonal neurotubule accumulation. Schwann cells showed signs of "activation" but clear demyelination was not observed. We conclude that with the use of this model it is possible to induce a peripheral neuropathy in the Wistar rat which resembles that reported in humans and which can, therefore, be used to better understand the basic mechanism(s) of taxol toxicity and to evaluate protective strategies in an attempt to reduce it.

Peripheral neurotoxicity of taxol in patients previously treated with cisplatin

BACKGROUND

Taxol is a new anticancer drug that acts as a tubulin polymeration enhancer. Its major toxicities are myelosuppression, hypersensitivity, and mucositis, but it also induces peripheral nerve damage. The use of taxol has recently been proposed for platinum-resistant cancers, but in these cases there is a possibility of cumulative toxicity in the peripheral nervous system.

METHODS

Twenty-two patients affected by a relapse of cisplatin-treated ovarian cancer were examined clinically and neurophysiologically to determine the evolution of taxol-induced peripheral somatic and autonomic neurotoxicity and the possible cumulative effect of a combination of taxol and cisplatin. Each patient was examined before, during, and after taxol treatment (using a dose of 135 or 175 mg/m2 in 3 hours every 3 weeks).

RESULTS

No patients were excluded from the study because of unacceptable toxicities of any kind. The serial examinations demonstrated that taxol induced onset of (or worsening of preexisting) neuropathic symptoms and signs in almost all the patients. The features were those of a distal, symmetrical, sensory polyneuropathy due to an axonopathy. Motor nerves and the autonomic nervous system were unaffected. Taxol neurotoxicity appeared early in the course of the treatment (i.e., after three courses) and was not severely disabling. In most cases after the early onset of peripheral neuropathy, stabilization of this side effect occurred.

CONCLUSIONS

Considering the low doses of taxol used in this study, the sensory nerve damage was unexpectedly severe. It appears that a cumulative, but not dose-limiting, neurotoxic effect occurs using taxol in patients previously treated with cisplatin.

Weekly cisplatin +/- glutathione in relapsed ovarian carcinoma

On the basis of experimental data showing the efficacy of glutathione (GSH) as a protective agent on cisplatin-induced neurotoxicity and the clinical evidence of the low incidence of neurotoxicity in high-dose cisplatin + GSH treated patients we evaluated the neuroprotective effect of GSH in a randomized phase II study. Thirty-three patients with relapsed ovarian cancer after a disease-free interval of at least 1 year and a cumulative dose of prior cisplatin ranging 450-650 mg m-2 were randomized to receive cisplatin 50 mg m-2 weekly +/- 2.5 g GSH for 9 consecutive weeks. Clinical and instrumental neurologic and otologic evaluations were made at the baseline and at the end of the study. Overall response rate in 31 evaluable patients was: 9/15 in group A and 12/16 in group B, including 4/15 vs 7/16 complete responses. The administered dose intensity of cisplatin was higher in the GSH treated patients (100% dose intensity was received by 56% vs 27%). A trend in terms of neuroprotection was detected in the GSH treated group, and no major difference was observed in the other toxicities between the two groups. It is concluded that possible benefit can be expected from the concomitant administration of GSH and cisplatin in patients at high risk of developing neurotoxicity, without decreasing the anti-tumor activity.

Frequency and clinical correlates of anti-neural IgM antibodies in neuropathy associated with IgM monoclonal gammopathy

We studied the frequency and clinical correlates of different IgM specificities in 75 patients with neuropathy associated with IgM monoclonal gammopathy. Patients were tested for IgM reactivity with the myelin-associated glycoprotein, P0, neurofilaments, and tubulin by immunoblot; with GM1, asialo-GM1, GM2, GD1a, GD1b, sulfatide, and chondroitin sulfate C by enzyme-linked immunosorbent assay; and with brain and nerve glycolipids by overlay high-performance thin-layer chromatography. Forty-two patients (56%) had high titers of IgM antibodies to MAG; 4 (5%), to sulfatide (1 also to myelin-associated glycoprotein); 4 (5%), to the 200-kd neurofilament (2 also to myelin-associated protein); and 1 each, to GD1b and chondroitin sulfate C. No reactivity was found in 26 patients (35%). More patients with anti-myelin-associated glycoprotein IgM (62%) than with unknown IgM reactivity (31%) had a predominantly sensory neuropathy (p < 0.025). Nerve conduction findings were consistent with a demyelinating neuropathy in 77% of patients reactive to myelin-associated glycoprotein and 24% with unknown reactivity (p < 0.0001) and the mean conduction velocity of peroneal nerve was lower in the former group (22.9 m/sec) than in the latter group (39.6 m/sec) (p < 0.000001). Patients with anti-sulfatide IgM had a sensorimotor neuropathy with morphological evidence of demyelination while anti-neurofilament IgM was not associated with homogeneous findings. Patients with anti-GD1b or anti-chondroitin sulfate C IgM had a predominantly motor impairment. The frequent occurrence of anti-neural IgM antibodies in neuropathy associated with IgM gammopathy, and their frequent, though not constant association with similar neuropathy features, support their possible pathogenetic role in the neuropathy.