PAIN MEDICINE Volume 2 • Number 1 • 2001

Chemotherapy-evoked Painful Peripheral Neuropathy Rosemary C. Polomano, PhD, RN,* and Gary J. Bennett, PhD† *Departments of Anesthesiology, Neuroscience, and Anatomy, Penn State Milton S. Hershey Medical Center, The Pennsylvania State College of Medicine, Hershey, Pennsylvania, and †Department of Neurology, MCP Hahnemann University, Philadelphia, Pennsylvania

ABSTRACT

Key Words. Allodynia; Chemotherapy Neurotoxicity; Hyperalgesia; Neuropathy; Paclitaxel; Painful Neuropathy; Vincristine

Paclitaxel (commercially available as Taxol) is an agent in the class of taxane antineoplastic agents. It is one of the most effective and most commonly used chemotherapeutic drugs for the treatment of solid tumors. Paclitaxel-evoked neurotoxicity typically presents as a predominantly sensory neuropathy, with the most common complaints being numbness, tingling, and burning pain. Sensory symptoms usually start symmetrically in the feet, but sometimes appear simultaneously in both hands and feet. Motor and autonomic neuropathies are also sometimes seen, especially at higher doses [1–6]. Paclitaxel is generally believed to kill tumor cells via promoting the hyperpolymerization of betatubulin, a protein that forms the mitotic spindle [7]. Hyperpolymerization is believed to interfere with spindle function and thereby arrest the process of cell division, which in turn engages the apoptotic pathway. Paclitaxel’s neurotoxic effect has also generally been assumed to be due to its binding to beta-tubulin, which is the primary component of the axonal microtubules. Disruption of microtubules is hypothesized to impair retrograde and anterograde axonal transport. However, recent evidence suggests that paclitaxel’s ability to destroy cancer cells may be due to additional mechanisms, and there is doubt about whether its neurotoxicity has anything to do with impaired axonal transport [8–9]. Vincristine, a vinca alkaloid introduced in the early 1960s, has been used to treat lymphomas, leu-

Introduction

P

eripheral neurotoxicity is a clinically significant complication of cancer chemotherapy. For several of the most effective drugs, neurotoxicity is dose limiting and sometimes forces the termination of otherwise successful therapy, or precludes the repetition of successful therapy. Sensory abnormalities from chemotherapy-evoked neurotoxicity range from mild paresthesiae or dysesthesiae to severe neuropathic pain. It is important to note that while these drugs produce a symptomatic neuropathy in many patients, only some of the affected patients develop neuropathic pain; ie, only a subset develop painful peripheral neuropathy. In some cases, sensory and motor symptoms resolve within days or weeks after the agents are discontinued, but peripheral neuropathy can be a chronic painful and disabling condition. The mechanisms that produce the nerve injury in general, and the neuropathic pain in particular, are unknown in every case. We concentrate here on the neurotoxicity seen with two commonly used antineoplastic agents, paclitaxel and vincristine.

Reprint requests to: Rosemary C. Polomano, PhD, RN, Department of Anesthesiology, The Pennsylvania State College of Medicine, Milton S. Hershey Medical Center, Hershey, PA 17033. Tel: (717) 531-4085; Fax: (717) 531-3911; E-mail: [email protected]. © Blackwell Science, Inc. 1526-2375/01/$15.00/8 8–14

8

Downloaded from http://painmedicine.oxfordjournals.org/ by guest on January 18, 2017

Vincristine and paclitaxel, two of the most effective drugs in the battle against cancer, produce a doselimiting neurotoxicity that sometimes presents as a painful peripheral neuropathy. For the first time, investigators have been able to produce these chemotherapy-evoked painful peripheral neuropathies in the laboratory rat. These new models have already begun to elucidate the causes of the neuropathic pain associated with these antineoplastic drugs, which will now make it possible to search for effective ways to prevent and treat it.

Chemotherapy-evoked pain

ally captured through the use of neurotoxicity grading scales. Most clinical trials involving neurotoxic agents identify and classify degrees of neurotoxicity by using these ratings, which provide only a global measure for screening and ranking neurological abnormalities. Unfortunately, they lack sensitivity and specificity for detecting early signs and symptoms of sensory abnormalities (particularly pain), and for tracking their progression. Postma et al. [17] caution that many of these toxicity rating scales are subjected to highly variable interpretations, which can result in unreliable assessments of the severity of peripheral neurotoxicity. Moreover, the effects of chemotherapy-induced neurotoxicity on other clinical outcomes, such as quality of life and patient decisions to continue treatment, have not been adequately studied. Only a few reports emphasize the disabling effects of chemotherapy-induced peripheral neuropathy [12,18,19]. Electrodiagnostic testing (electromyogram (EMG) and sensory nerve conduction velocity) is commonly used to characterize chemotherapy-evoked neuropathy. While standard electrodiagnostic tests are often useful in confirming the presence of nerve damage, they are very poor indicators of minor nerve injury and do not provide any information about the status of the small myelinated (A␦) and unmyelinated (C-fiber) sensory afferents that mediate pain sensation. Sensory deficits to thermal and mechanical stimuli can be easily measured with quantitative sensory testing (QST), a psychophysical technique for measuring deficits in the detection thresholds for warming (a C-fiber–mediated sensation), cooling (an A␦-mediated sensation), and vibration (a sensation mediated by large, myelinated A␤ afferents) [20–22]. Despite the appropriateness of these tests, QST has been employed in a small number of studies of chemotherapy-induced peripheral neuropathy [4,13]. Recent progress has been made in understanding chemotherapy-induced peripheral neuropathy through the use of animal models. Experiments using rats, mice, and other animals have enabled investigators to examine changes in sensory nerves caused by paclitaxel and vincristine and to test novel treatments [23–26]. In animal studies, analyses of abnormal pain responses such as allodynia (pain evoked by a normally innocuous stimulus) and hyperalgesia (pain of exaggerated severity in response to normally painful stimulation) to thermal (eg, cold and heat) and mechanical stimuli are sensitive measures of neuropathic pain. Animal models also permit thorough evaluations of anatomical and structural changes in primary afferents

Downloaded from http://painmedicine.oxfordjournals.org/ by guest on January 18, 2017

kemia, and some solid tumors [10–12]. Vincristine therapy generally evokes marked sensorimotor dysfunction. Autonomic neuropathy, cranial nerve palsies, and sometimes even central nervous system toxicity are also seen. Early and common symptoms include loss of deep tendon reflexes and paresthesiae in the fingers and toes. Neurological deficits in the periphery are manifested by pain, loss of touch and vibration sense, and loss of proprioception in the hands and feet. Motor symptoms can progress to muscle weakness, especially in the extensors of the hands and dorsiflexors of the feet. Vincristine, similar to other vinca alkaloids, inhibits the normal polymerization of beta-tubulin. Thus, as with paclitaxel, its antitumor effect is believed to be due to mitotic spindle dysfunction, and its neurotoxic effect to dysfunction of axoplasmic transport. However, here also there is a lack of convincing evidence to support the exact mechanisms for neurotoxicity. For paclitaxel and vincristine, the incidence of chemotherapy-induced peripheral neuropathy is influenced by a variety of factors including age, single dose intensity, cumulative dose, length of therapy, prior or concomitant administration of other neurotoxic chemotherapeutics, and pre-existing conditions such as diabetes and alcohol abuse [10–12]. Peripheral neuropathy can be expected with paclitaxel monotherapy at doses of 250 mg/m2 body surface area or greater in as many as 22% to 100% of patients [3,13]. A large majority of patients receiving vincristine report sensory disturbances. In one multicenter study evaluating the effects of vincristine to treat advanced cancers, 223 of 409 patients (57%) reported paresthesiae [14]. With a relatively high incidence of neurotoxicity, most vincristine protocols limited single doses to 2 mg, regardless of body surface area [15]. However, dosing limitations may have compromised the efficacy of treatment regimens. Full doses of 1.4 mg/m2 are now recommended in the treatment of Hodgkin’s disease [16]. The painful peripheral neuropathy associated with chemotherapy has not been adequately characterized or quantified clinically. There is a paucity of data to indicate just how many patients experience a painful neuropathy during the course of their therapy, and the subsequent short-and long-term patterns of painful experiences. Comprehensive assessments of the pain, including serial measures of its intensity, quality, and character, and assessments of changes in thermal and mechanical pain thresholds, are rarely incorporated into routine clinical care. Instead, estimates of abnormal pain associated with chemotherapy-induced neuropathy are gener-

9

10 and electrophysiological conduction defects, which are difficult or impossible to perform on humans. Experimental Models of Paclitaxel Neurotoxicity

lative doses of 8 and 16 mg/kg). When tested 2 weeks after the last dose, these animals had abnormal sensory nerve conduction velocities (SNCV) in the tail nerves. In a subsequent study that also used adult Wistar rats, Cavaletti et al. [35] gave a series of 5 intravenous (IV) injections (jugular vein) of 5 mg/kg on days 1, 2, 3, 9, and 10 (cumulative dose of 25 mg/kg). On the day following the last injection, they found a significant decrease in pain sensation in the tail flick test, abnormal SNCV in the tail nerves, and impaired performance on the rota-rod test. The SNCV demonstrates changes in large myelinated axons. Similarly, impaired performance in the rota-rod test demonstrates large fiber damage (to the axons of motor neurons and/or proprioceptive sensory axons). Campana et al. [26] used adult Sprague-Dawley rats and 2 IP injection schedules: 1.2 mg/kg 5 times per week for 3 weeks (cumulative dose of 18 mg/kg), or injections every third day (25, 12.5, and 12.5 mg/kg) for 10 days (cumulative dose of 50 mg/kg). In the tail flick test, rats receiving 5 injections per week were significantly hyposensitive on the last day of drug exposure. The animals that were dosed every 3 days were significantly hyposensitive on days 9 and 11, but showed no change in motor or sensory conduction velocities recorded from the sciatic nerve. Neither dosing schedule produced any light-microscopic evidence of abnormal anatomy in the sciatic and sural nerves. Cliffer et al. [36] used adult SpragueDawley rats and IV (tail vein) injections and examined a range of dosing regimens that varied single dose intensity, cumulative dose, and dosing schedule: 5 daily injections of 5 or 7 mg/kg (cumulative doses of 25 and 35 mg/kg); 4 to 5 weekly injections of 15 or 18 mg/kg (cumulative doses of 72 and 75 mg/kg); and 2 twice-weekly injections of 15 or 18 mg/kg (cumulative doses of 30 or 36 mg/kg). The daily dosing regimens were lethal to half the animals. General health was preserved and a dose-dependent neuropathy was induced by both the weekly and twice-weekly regimens. Animals receiving the twice-weekly injections had abnormal sensory and motor nerve conduction in both tail and hind limb nerves, and impaired rota-rod performance, but no change in tests of muscle strength or pain sensitivity. The electrophysiological abnormalities were evident 1 to 2 weeks after treatment and persisted for at least 12 weeks. The deficit in rota-rod performance was evident within one week of treatment, but resolved completely by 10 weeks after treatment. These animals had massive degeneration of myelinated axons in the dorsal roots. Boyle et al. [24] used adult rats of the dark agouti strain and 4

Downloaded from http://painmedicine.oxfordjournals.org/ by guest on January 18, 2017

Paclitaxel has been injected directly into the sciatic or the peroneal nerves of rats to evaluate morphological changes in the nerve. Investigations of subepineural injections of paclitaxel have evaluated structural abnormalities in axons and glia, and the cytoskeletal defects in axons using light and electron microscopy after local injections of the drug in doses of 0.1 mL to 0.2 mL of 1 ␮m paclitaxel in 0.1% to 0.2% dimethylsulfoxide [27–29]. In some cases the injections have been made following crush injury to the nerve intended to enhance local uptake [30,31]. The relevance of these models to the human case is highly questionable. It has not been proven that such concentrations of drug approximate the degree of exposure expected with serum levels following systemic administration. The animal studies show swollen axons, accumulations of microtubules within axons and Schwann cells, and abnormalities in the myelin sheath and internodes. However, there are no human data showing similar histological findings, except for 2 atypical cases of severe neurotoxicity following extraordinarily high doses of paclitaxel [3,32]. When injected directly into nerves, paclitaxel escapes metabolism by the liver, thus eliminating the possibility that the histopathology is due to one or more active metabolites of the drug. The results from epineural injections following neuronal crush injuries tend to be highly variable, which has been attributed to differences in paclitaxel concentrations and injury-related permeability of the axonal membrane. There are no studies of the behavioral effects of paclitaxel injected into the nerve. Until very recently (see below), studies that have evaluated sensory changes in general, and changes in pain sensitivity in particular, following systemic exposure to paclitaxel have found sensory deficits; none have demonstrated the presence of neuropathic pain. Apfel et al. [33] used adult mice of the CD1 strain and gave 6 consecutive daily intraperitoneal (IP) doses of 21.6 mg/kg (cumulative dose of 159.6 mg/kg). On the third day following the last injection, they found a profound insensitivity in the tail flick test and reduced nerve conduction velocities in the tail nerves (predominately sensory nerves). Cavaletti et al. [34] dosed adult Wistar rats with 4 weekly IP injections of 2 mg/kg or 4 mg/kg (cumu-

Polomano et al.

Chemotherapy-evoked pain

Experimental models of vincristine neurotoxicity

While vincristine peripheral neurotoxicity has been studied extensively in animals, there are few studies

that have assayed sensory changes [24,38–42]. As with studies of paclitaxel, the vincristine studies that have examined sensory function have used drug exposures that produce severe neurological impairment and significant toxicities leading to deterioration of the animal’s general health and even to death. The relevance of these studies to the clinical problem is obscure. It is thus noteworthy that a model of vincristine-evoked painful peripheral neuropathy in the rat has been described. Aley et al. [43] injected 20, 100, or 200 ␮g/kg vincristine intravenously 5 days per week for 2 weeks. They found that vincristine treatment produced both acute and chronic effects on pain sensitivity. Doses of 100 or 200 ␮g/kg produced an acute mechano-hyperalgesia (assessed with the RandalSellito method). The effect was first observed after the second dose and continued for each of the subsequent 8 doses. The effect was detectable within less than 3 minutes after the injection and reached peak severity in approximately 30 minutes. An acute mechano-allodynia (von Frey hair method) was also detected (only the 100-␮g/kg dose was examined). No data were obtained concerning the possibility of an acute effect on heat-evoked pain. The clinical relevance of these acute effects is unclear. While pain at the injection site (presumably due to vascular irritation) is commonplace, we have been unable to find any clinical report of vincristine-evoked acute hyperalgesia or allodynia at locations distant from the injection. On the other hand, the chronic effects of vincristine on pain hypersensitivity in this model have obvious clinical relevance. Vincristine doses of 20, 100, or 200 ␮g/kg produced a clear-cut chronic mechano-hyperalgesia. The effect was first noted following the fifth injection and it persisted for 2 to 3 weeks; thresholds were normal thereafter. The 100-␮g/kg dose was examined for a chronic effect on mechano-allodynia. Mechano-allodynia was also found with an onset after the fifth injection and a duration of 2 to 3 weeks with a normalized threshold thereafter. A small but statistically significant chronic heat-hyperalgesia was also reported. Chronic vincristine treatment with the doses used by Aley et al. had relatively few nonsensory effects. Two weeks after beginning dosing, rota-rod performance was unaffected by the 100-␮g/kg dose, but severely disrupted by 200 ␮g/kg. Rats given 100 ␮g/kg did not gain weight during the dosing period and those given 200 ␮g/kg lost weight, but both groups put on weight when dosing ceased. Tanner et al. [44] did not find any anatomical evidence of pathology at the light-microscopic level

Downloaded from http://painmedicine.oxfordjournals.org/ by guest on January 18, 2017

twice-weekly injections of 9 mg/kg (cumulative dose of 36 mg/kg). At the time of the last dose, they found gait abnormalities, significant deficits in rotarod performance, and significant hyposensitivity to pain in the tail flick test. It is important to note that there are many differences in the methods used in the studies described above. In particular, they used different single dose intensities, different cumulative doses, and different dosing schedules; clinical experience suggests that each of these variables influences the appearance of neuropathy. Although the results of these studies are not entirely consistent, there is a clear trend towards observations of extensive damage to large myelinated axons and either no effect on pain perception or a decreased sensitivity to pain. These findings may be relevant to human cases with severe paclitaxel-evoked neuropathy, but they do not appear to be relevant to the typical clinical presentation, nor to the patient with paclitaxelevoked neuropathic pain. It is thus significant that Polomano et al. (unpublished data, 2001; [37]) have recently described a paclitaxel-evoked painful peripheral neuropathy that is not associated with any evidence of axonal injury or dysfunction. It is probable that the important methodological difference in this new work is the use of low single dose intensities and low cumulative doses. Adult male Sprague-Dawley rats were treated with vehicle or paclitaxel (0.5, 1.0, or 2.0 mg/kg) via 4 IP injections given on alternate days. The total cumulative doses were 2.0, 4.0, or 8.0 mg/kg. All 3 doses produced clear-cut signs of neuropathic pain: heat-hyperalgesia, mechano-hyperalgesia, mechano-allodynia, and cold-allodynia. The abnormal pain sensations began within several days of the initiation of dosing and lasted for at least several weeks after dosing was completed. None of the doses produced any retardation of weight gain or any other sign of significant generalized debilitation. The highest dose (the only one tested) had no effect on rota-rod performance. Light-microscopic analyses of the sciatic nerve, lumbar dorsal and ventral roots, lumbar dorsal root ganglia, and lumbar segments of the spinal cord revealed no sign of structural damage. It is probable that this model is relevant to the early symptoms of paclitaxel-evoked neuropathy, especially in those cases that develop neuropathic pain.

11

12

change seen in rats with experimental painful diabetic neuropathy [46]. Authier et al. [47] have repeated the studies of Aley et al. [43], but with a slightly different dosing schedule: 10 consecutive daily doses, rather than 5 consecutive daily doses with 2 days off and then another 5 consecutive doses. With the 100-␮g/kg dose used by Aley et al., they found a significant deterioration of general health and significant mortality during the dosing period. A dose of 75 ␮g/kg produced a significant reduction in the rate of weight gain and prevented deaths during the dosing period, but a quarter of the animals died after dosing ceased. A dose of 50 ␮g/kg did not produce any mortality at any time, and although the rate of weight gain was significantly reduced (⫺9% versus controls), the animals otherwise appeared to be in good health. Authier et al. found that doses of 50 and 75 ␮g/kg produced significant mechano-hyperalgesia beginning around the time of the last injection on day 10 and continuing for at least 12 days after dosing ceased. Significant mechano-allodynia with a similar time course was seen only with 75 ␮g/kg. Interestingly, both doses produced a significantly increased threshold to heat-evoked pain (ie, hypoalgesia). The reason for the discrepancies between the results of Authier et al. and Aley et al. are not known. However, clinical experience suggests that even the seemingly minor difference in dosing schedules might be significant. Conclusions

Because chemotherapy-evoked painful peripheral neuropathy is a common and serious clinical problem, there is an urgent need for well-designed clinical studies to fully characterize the severity of pain and sensory deficits. In order to elucidate the clinical course of chemotherapy-induced painful neuropathies, routine clinical care must include serial self-report measurements of pain and frequent neurological examinations to detect sensory and motor abnormalities. When appropriate, quantitative sensory testing (QST) can be used to assess alterations in sensory thresholds to innocuous and noxious stimuli such as allodynia and hyperalgesia. Patterns for the onset and severity of pain should be closely monitored and tracked throughout the course of therapy, and even after therapy is discontinued. Emphasis on the clinical diagnosis of large fiber dysfunction simply misses the hallmarks of pain and other sensory disturbances, making it imperative to focus on assessments that evaluate the status of the

Downloaded from http://painmedicine.oxfordjournals.org/ by guest on January 18, 2017

in rats with a painful peripheral neuropathy due to vincristine (100 ␮g/kg) treatment. Quantitative ultrastructural studies [44,45] have confirmed that there is no loss of myelinated or unmyelinated axons. However, electron microscopic analyses of vincristine-treated (100 ␮g/kg) animals have found signs of intra-axonal pathology in both myelinated and unmyelinated axons, as well as pathology in about 30% of the large cells in the dorsal root ganglion. The intra-axonal pathology was similar in myelinated and unmyelinated fibers. There was a decrease in the density of microtubules that was almost completely accounted for by axonal swelling. In addition, there was an increase in the number of tangentially sectioned microtubules in nerve crosssections, indicative of disruption of the normal linear organization of the microtubules that runs parallel to the long axis of the axon. In the dorsal root ganglia, vincristine treatment caused neuronal swelling and an accumulation of neurofilaments, but this was apparent in only about 30% of the large cell bodies (those that give rise to large myelinated axons that are touch-sensitive afferents), and not in the small cell bodies associated with pain fibers. It is important to note how subtle these changes are. They were revealed only with quantitative, highmagnification electron microscopy. The vincristine model has also made an important contribution to our understanding of the neurophysiological changes that may underlie chemotherapy-evoked neuropathic pain. Recordings from single C-fiber nociceptors in vincristine-treated rats (100 ␮g/kg according to the protocol of Aley et al. [43]) revealed that about half responded to a sustained (1 minute) suprathreshold mechanical stimulus with a discharge frequency that was clearly greater than normal [44]. A smaller percentage of C-nociceptors also responded in an exaggerated fashion to suprathreshold heat stimulation. However, there was no obvious change in the nociceptor thresholds to mechanical or heat stimulation, and these fibers did not develop spontaneous discharge. It is important to note that the vincristine-evoked change is thus quite different from the primary afferent sensitization that occurs with inflammation. Spontaneous discharge, a decreased threshold, and a leftward shift of the stimulus response curve characterize primary afferent sensitization. Only the latter change is comparable to the vincristineevoked hyper-responsiveness to suprathreshold stimulation. Interestingly, the change in C-nociceptor responsiveness seen in vincristine-treated rats appears to be very similar, if not identical, to the

Polomano et al.

13

Chemotherapy-evoked pain

1) Graded dosing schedules should be selected to include a minimum total dose that exceeds the threshold for nerve toxicity and a maximum total dose that is below the threshold for severe toxicities such as hypoesthesia, paralysis, and deterioration of general health. 2) Similar routes of administration (IV or IP) and similar dosing schedules should be used when comparing results from evaluations of behavioral, electrophysiological, and anatomical outcomes. 3) Behavioral assays for detecting neuropathic pain (both tests for mechano- and thermal hypersensitivity) should be performed. The tail-flick test by itself is woefully inadequate. The status of the motor system should also be examined, at least with respect to rota-rod performance, but ideally with electrophysiological examination of sensorimotor reflexes as well. 4) Results from electrophysiological testing, morphological evaluations, neurochemical evaluations, and other assessments should be correlated to the results of behavioral tests of sensory status. The development of clinically relevant animal models of chemotherapy-evoked painful peripheral neuropathy is essential if we are to understand the pathophysiological mechanisms that are involved and if we are to develop treatments to prevent or minimize the nerve damage. Now more than ever, attention must be paid to problems that limit the usefulness of effective chemotherapeutic agents. Improvements in chemotherapy continue to prolong life but at a cost of seriously compromising the quality of life [48]. Chemotherapy-evoked peripheral neurotoxicity is a problem that will only

worsen as advancements in the minimizing lifethreatening side effects such as granulocyte-colony stimulating factor to combat neutropenia now permit more aggressive therapy.

References

1 Manfredi JJ, Horowitz SB. Taxol: an antimitotic agent with a new mechanism of action. Pharmacol Ther 1984;25:83–125. 2 Rowinsky EK, Donehower RC. Paclitaxel (taxol). N Engl J Med 1995;332:1004–14. 3 Wiernik PH, Schwartz EL, Einzig A, et al. Phase 1 trial of taxol given as a 24-hour infusion every 21 days: Responses observed in metastatic melanoma. J Clin Oncol 1987;5:1232–9. 4 Chaudhry V, Rowinsky EK, Sartorius SE, et al. Peripheral neuropathy from taxol and cisplatin combination chemotherapy: clinical and electrophysiological studies. Ann Neurol 1994;35:304–11. 5 Einzig AI, Wiernik PH, Sasloff J, et al. Phase II study and long-term follow-up of patients treated with taxol for advanced ovarian carcinoma. J Clin Oncol 1992;10:1748–53. 6 Kaplan JG, Einzig AI, Schaumburg HH. Taxol causes permanent large fiber peripheral nerve dysfunction: a lesson for preventative strategies. J Neurooncol 1993;16:105–7. 7 Parness J, Horowitz SB. Taxol binds to polymerized tubulin in vitro. J Cell 1991;9:479–87. 8 Fan W. Possible mechanisms of paclitaxel-induced apoptosis. Biochem Pharmacol 1999;57:1215–21. 9 Komiya Y, Tashiro T. Effects of taxol on slow and fast axonal transport. Cell Motil Cytoskeleton 1988; 11:151–6. 10 Pratt WB, Ruddon RW. The Anticancer Drugs. New York: Oxford University Press; 1974:221–33. 11 Legha SS. Vincristine neurotoxicity: pathophysiology and management. Med Toxicol 1986;1:421–7. 12 Hussain M, Wozniak AJ, Edelstein MB. Neurotoxicity of antineoplastic agents. Crit Rev Oncol Hematol 1993;14:61–75. 13 Forsyth PA, Balmaceda C, Peterson K, et al. Prospective study of paclitaxel-induced peripheral neuropathy with quantitative sensory testing. J Neurooncol 1997;35:47–53. 14 Holland JF, Scharlau C, Gailani S, et al. Vincristine treatment of advanced cancer: a cooperative study of 392 cases. Cancer Res 1973;33:1258–64. 15 Chabner BA, Myers CE. Clinical pharmacology of cancer chemotherapy. In: DeVita VT, Hellman S, Rosenberg SA, eds. Principles and Practice of Oncology. Vol. 1. Philadelphia: JB Lippincott Company; 1985:287–328. 16 Haim N, Epelbaum R, Ben-Shahar M, et al. Full dose vincristine (without 2-mg dose limit) in the treatment of lymphomas. Cancer 1994;73:2515–19. 17 Postma TJ, Heimans JJ, Muller MJ, et al. Pitfalls in

Downloaded from http://painmedicine.oxfordjournals.org/ by guest on January 18, 2017

small sensory fibers. Subtle sensory abnormalities are likely to appear soon after the initiation of chemotherapy; by the time the ankle jerk reflex is gone and the patient’s foot is numb, late-stage pathology is almost certain. Progress made with the use of animal models is hampered unless clear parallels can be drawn to the clinical problem. In the case of paclitaxel- and vincristine-evoked painful peripheral neuropathy, we now have a better understanding of the sensory abnormalities in rats than we do of those in people. The ability to translate research findings from laboratory animals to clinical practice is dependent on the clinical relevancy and meaningfulness of animal results in explaining human experiences. Ideally, animal experiments with chemotherapy-induced neurotoxicity should address the following issues:

14

18 19

20 21 22

24 25 26 27 28

29

30 31 32 33 34

grading severity of chemotherapy-induced peripheral neuropathy. Ann Oncol 1998;9:739–44. Jerian SM, Sarosy GA, Link CJ, et al. Incapacitating autonomic neuropathy precipitated by taxol. Gynecol Oncol 1993;51:277–80. van den Bent MJ, van Raaij-van den Aarssen VJ, Verweiji J, et al. Progression of paclitaxel-induced neuropathy following discontinuation of treatment. Muscle Nerve 1997;20:750–2. Gruener G, Dyck PJ. Quantitative sensory testing: methodology, applications, and future directions. J Clin Neurophysiol 1994;11:568–83. Yarnitsky D. Quantitative sensory testing. Muscle Nerve 1997;20:198–204. Zaslansky R, Yarnitsky D. Clinical applications of quantitative sensory testing (QST). J Neurol Sci 1998;153:215–38. Hol EM, Mandys V, Sodaar P, et al. Protection by an ACTH4-9 analogue against the toxic effects of ciplatin and taxol on sensory neurons and glial cells in vitro. J Neurosci Res 1994;39:178–85. Boyle FM, Wheeler HR, Shenfield GM. Amelioration of experimental cisplatin and paclitaxel neuropathy with glutamate. J Neurooncol 1999;41:107–16. Jackson DV, Wells HB, Atkins NJ, et al. Amelioration of vincristine neurotoxicity by glutamic acid. Am J Med 1988;84:1016–22. Campana WM, Eskeland N, Calcutt NA, et al. Prosaptide prevents paclitaxel neurotoxicity. Neurotoxicology 1998;19:237–44. Röyttä M, Raine CS. Taxol-induced neuropathy: further ultrastructural studies of nerve fibre changes in situ. J Neurocytol 1985;14:157–75. Vuorinen V, Röyttä M., Raine CS. The long-term cellular response to taxol in peripheral nerve: Schwann cell and endoneurial cell changes. J Neurocytol 1989;18:785–94. Nennesmo I, Reinholt FP. Effects of intraneural injection of taxol on retrograde axonal transport and morphology of corresponding nerve cell bodies. Virchows Arch B Cell Pathol Incl Mol Pathol 1988;55: 241–6. Vuorinen V, Röyttä M, Raine CS. The acute effect of taxol upon regenerating axons after nerve crush. Acta Neuropathol (Berl) 1988;76:26–34. Vuorinen V, Röyttä M. Taxol-induced neuropathy after nerve crush: long-term effects on regenerating axons. Acta Neuropathol (Berl) 1990;79:663–71. Sahenk Z, Barohn RJ, New P, et al. Taxol neuropathy. Electrodiagnostic and sural nerve biopsy findings. Arch Neurol 1994;51:726–9. Apfel SC, Lipton RB, Arezzo JC, et al. Nerve growth factor prevents toxic neuropathy in mice. Ann Neurol 1991;29:87–90. Cavaletti G, Tredici G, Braga M, et al. Experimental peripheral neuropathy induced in adult rats by re-

35

36

37 38 39 40

41

42

43

44

45

46

47 48

peated intraperitoneal administration of taxol. Exp Neurol 1995;133:64–72. Cavaletti G, Cavalletti E, Montaguti P, et al. Effect on the peripheral nervous system of the short-term intravenous administration of paclitaxel in the rat. Neurotoxicology 1997;18:137–45. Cliffer KD, Siuciak JA, Carson SR, et al. Physiological characterization of taxol-induced large-fiber sensory neuropathy in the rat. Ann Neurol 1998;43: 46–55. Polomano RC, Mannes AJ, Bennett GJ. Paclitaxelinduced painful peripheral neuropathy in rats. Soc Neurosci Abstr 1998;24:381. Shelanski ML, Wisniewski H. Neurofibrillary degeneration induced by vincristine therapy. Arch Neurol 1969;20:199–206. Schlaepfer WW. Vincristine-induced axonal alterations in rat peripheral nerve. J Neuropathol Exp Neurol 1971;30:488–505. Todd GC, Griffing WJ, Gibson WR, et al. Animal models for the comparative assessment of neurotoxicity following repeated administration of vinca alkaloids. Cancer Treat Rep 1979;63:35–41. Rebert CS, Pryor GT, Frick MS. Effects of vincristine, maytansine, and cis-platinum on behavioral and electrophysiological indices of neurotoxicity in the rat. J Appl Toxicol 1984;4:330–8. Fiori MG, Schiavinato A, Lini E, et al. Peripheral neuropathy induced by intravenous administration of vincristine sulfate in the rabbit. An ultrastructural study. Toxicol Pathol 1995;23:248–55. Aley KO, Reichling DB, Levine JD. Vincristine hyperalgesia in the rat: a model of painful vincristine neuropathy in humans. Neuroscience 1996;73: 259–65. Tanner KD, Levine JD, Topp KS. Microtubule disorientation and axonal swelling in unmyelinated sensory axons during vincristine-induced painful neuropathy in rat. J Comp Neurol 1998;395:481–92. Topp KS, Tanner KD, Levine JD. Damage to the cytoskeleton of large diameter sensory neurons and myelinated axons in vincristine-induced painful peripheral neuropathy in the rat. J Comp Neurol 2000; 424:563–76. Ahlgren SC, White DM, Levine JD. Increased responsiveness of sensory neurons in the saphenous nerve of streptoxotocin-diabetic rat. J Neurophysiol 1992; 68:2077–85. Authier N, Coudore F, Eschalier A, et al. Pain related behaviour during vincristine-induced neuropathy in rats. NeuroReport 1999;10:965–8. Sigurdardottir V, Bolund C, Sullivan M. Quality of life evaluation by the EORTC questionnaire technique in patients with generalized malignant melanoma on chemotherapy. Acta Oncol 1996;35: 149–58.

Downloaded from http://painmedicine.oxfordjournals.org/ by guest on January 18, 2017

23

Polomano et al.