Neuropathic Pain in Taxane-Induced Peripheral Neuropathy: Evidence for Exercise in Treatment

Introduction

Taxanes (docetaxel or paclitaxel) are widely used chemotherapeutic agents for the treatment of many solid tumors that prevent tumor growth through microtubule stabilizing mechanisms. A common side effect is chemotherapy-induced peripheral neuropathy (CIPN). CIPN is predominantly a small-fiber sensory neuropathy that develops in the hands/feet and worsens with increasing dose and duration of treatment. It affects the Aβ, Aδ, and C-fiber function involved in light touch and vibration sense, thermal detection, and thermal pain. This results in a variety of positive and/or negative sensory symptoms, including hypoesthesia, dysesthesias, hyperalgesia, allodynia, and neuropathic pain. This review examines current theories on the pathophysiology for taxane-induced peripheral neuropathy (TIPN), the clinical presentation and present management, the cellular mechanisms maintaining neuropathic pain, and the animal and human models of exercise and neuropathy and neuropathic symptoms. Finally, the importance of exercise in cancer prevention and recurrence and some of the barriers and facilitators that clinicians and patients face when prescribing and following through with exercise prescription are discussed.

Pathophysiology

The etiology of TIPN is still unclear; however, taxanes are microtubule-stabilizing agents that prevent division of cells.^1,2 Taxanes bind with high affinity to the interior portion of the b-tubulin subunit of the microtubule, which suppresses dynamic stability and thus stops mitosis. ³ Whereas this is effective in arresting rapidly dividing tumor cells, the same action is thought to affect cells in healthy peripheral nerves. Microtubules require dynamic stability and act as scaffolding for molecular motors to transport nutrients, neurotransmitters, and mitochondria from the cell body to the periphery via anterograde transport. Used synaptic vesicles and other proteins for degradation return to the cell soma via retrograde transport. It is thought that loss of dynamic stability in the microtubule results in loss of transport function. At the sensory endings, microtubules are also critical in directing the growth cone and filipodia, which advance neuron outgrowth. ⁴ Impairment to the function of the microtubules and breakdown of the transport process has been implicated in causing changes in the morphology and physiology of the peripheral nerve. ⁵ The results are described as a symmetrical development that moves more proximally in a stocking/glove distribution. ⁶ It is thought that longer nerves are more vulnerable to impaired transport, and this theory has been used to explain the distal axonopathy as well as the phenomenon of coasting. Coasting is the term used to describe progressive worsening of TIPN symptoms after chemotherapy. It is thought to be related to the time required for anterograde transport to move the chemotherapy drug from the dorsal root ganglion (DRG) to the distal axon.^3,4-8

Recent research by Gornstein and Swartz ⁹ suggests that paclitaxel locally affects microtubules in the distal axon, and transport deficits may not be the primary cause. The authors applied paclitaxel (10-50 nm) to the midaxon and distal axon of adult mouse DRG neurons and found the distal portion to be selectively sensitive, whereas the midaxon portion showed no significant changes compared to controls. Specifically, the authors suggest that paclitaxel disrupts growth cone dynamics in filipodial extension of pioneer microtubules in the sensory nerve. Pioneer microtubules are active and in a high metabolic state, maintaining retraction and advancement of the nerve as the epithelium regenerates; therefore, stabilization of these pioneer microtubules would directly result in impairment at the distal axon. ⁹ Other suggested primary mechanisms of TIPN include mitotoxicity (swollen and vacuolated mitochondria) present in Aδ and C fibers after chemotherapy resulting in Ca²⁺ release (opening mitochondrial permeability transition pore and neuronal excitability). Damaged cellular function leads to an increase in reactive oxidative stress, subsequent ion channel changes (TRPV1 via PKC or MAPK), release of inflammatory cytokines, and activation of caspase 3/7.^3,7,10,11 This may, in part, be related to a second low-affinity binding site of taxanes at the end of the microtubules associated with generating mediators of inflammation (IL, NO, and COX-2). ¹²

Clinical Presentation and Management

The clinical presentation of TIPN is reported as a terminal nerve ending sensory disturbance with loss of temperature and vibration sensation. ⁵ This small-fiber sensory neuropathy specifically affects the Aδ (thinly myelinated fibers involved in cold perception and thermal pain), C-Fiber (unmyelinated fibers involved in warm temperature perception and thermal pain), and the Aβ fibers (myelinated fibers involved in touch and vibration sense). Nerve conduction studies are normal during the early stages of the axonopathy because they monitor the amplitude and velocity of large myelinated fibers. Nerve conduction studies are considered neither specific nor sensitive enough to measure and monitor changes in sensory neuropathies.^5,13 Skin biopsy, quantification of intraepidermal nerve fiber density (IENFD), and quantitative sensory testing (QST) are useful in diagnosing and measuring changes in this neuropathy over time. Currently, there is no gold standard test for diagnosing small-fiber neuropathy; however, presence of symptoms mentioned above with abnormal QST and a normal nerve conduction study will confirm the diagnosis.^13,14

Large interpersonal variability exists in neuropathy location (hands or feet) and intensity of symptoms. This has been explained by variability in chemotherapy cumulative dose, schedule, combination therapy, preexisting risk factors (including diabetes, advanced age, smoking, increased alcohol consumption), increased body mass index, and genetic predisposition.^15-18 The presence of cold allodynia and cold hyperalgesia during chemotherapy has been clinically identified as a risk factor for persistent CIPN symptoms. ¹⁹ Taxane neuropathy affects 80% to 97% of patients, ⁷ and 27% of those who have neuropathy experience neuropathic pain. ²⁰

A retrospective study found that 10% of patients on docetaxel or paclitaxel had a dose reduction or treatment delay during treatment because of TIPN symptoms. ²¹ Other studies have estimated that 25% to 60% of all patients will experience lasting neuropathic symptoms months to years after the completion of chemotherapy.^22,23 The monetary burden of CIPN has been estimated to cost more than 17 000 per person per year in the United States ²⁴ and is associated with anxiety, sleep disorders, and depression. ²⁵

Reviews have summarized potential treatment approaches studying antioxidants, vitamins, and herbal medicines such as Goshajinkigan, acetyl-l-carnitine, Ca²⁺-Mg²⁺ infusions, and vitamins A, B, and E.^6,8,26,27 These potential treatments have shown promise for possible preventive management by reducing oxidative stress in cell research and animal models, and a few of these studies have led to clinical trials; however, they failed to produce either neuroprotection or alleviation of symptoms.^26,28

At present, expert opinion from the 2014 practice guidelines from the American Society for Clinical Oncology has indicated that no agents can be recommended to prevent CIPN and provides only moderate support for treatment with a serotonin-norepinephrine reuptake inhibitor, duloxetine.^29-31 Unfortunately, duloxetine is associated with side effects (nausea, insomnia, and dizziness), and a physician must closely monitor drug dosage. Furthermore, duloxetine may be more effective in treating oxaliplatin rather than taxane neuropathic symptoms. ³⁰ In a consensus statement from the Canadian Pain Society, it has been identified that physical therapy programs, exercise, and psychological treatment are needed in conjunction with duloxetine to optimize successful outcomes. ³²

In addition to duloxetine, topical agents have been commonly suggested for treatment of symptoms. Results from clinical trials regarding sensory symptom improvement are mixed. A phase III 6-week trial of topical ketamine and amitriptyline demonstrated no improvement in pain scores. ³³ Another randomized controlled trial used a topical combination of ketamine, amitriptyline, and baclofen demonstrating improved tingling sensory subscales but no improvement in numbness, thermal pain, or functional abilities. ³⁴ Topical menthol has been suggested as a possible treatment but requires further study. ³⁵

There is a theory that a single pharmacological treatment may be possible for all CIPN because of similar overlapping mechanisms of mitotoxicity. Oxaliplatin, cisplatin, vincristine, and paclitaxel all have mitotoxicity as part of the pathophysiology. ¹¹ It has been postulated that this may involve inhibiting TRPV1 or NMDAR activation with additional anti-inflammatory properties, perhaps by inhibiting p38-MAPK. ⁸

One of the major concerns of targeting the pathway to prevent or minimize TIPN is that it theoretically could also affect taxane drug efficacy. ¹⁵ Thus, it is important to find a treatment that can block TIPN or restore sensory function that is independent of the chemotherapy pathway.

Mechanisms Maintaining Neuropathic Pain

Depending on the chemotherapy agent used, severity of symptoms can vary. Platinum-based drugs, specifically oxaliplatin, can cause acute severe sensory disturbances immediately following chemotherapy administration and are thought to be the most neurotoxic of all the drugs. By comparison, taxane drug neurotoxicity is as prevalent as that of platinum-based drugs, affecting 80% to 97% of patients, but the lasting symptoms are thought to be less severe with a reported maximum duration of 4.75 years. Of the taxane drugs, paclitaxel neuropathy symptoms are reportedly more severe and persistent when compared with those of docetaxel. ⁷

Two different sensory phenotypes are described in TIPN being either positive or negative.^5,6 A negative sensory profile is related to the loss of sensation that is associated with the loss of intraepidermal nerve fiber density, loss of myelination, and dying back of the sensory axon and symptoms of “numbness.” A positive sensory profile is associated with hypersensitivity potentially caused by an increased number of ion channels, altered or impaired Ca²⁺ signaling, neuroinflammation, and activation of adjacent silent nociceptors. Neuropathic pain is part of the positive sensory symptom profile of TIPN and includes sensations of stabbing, burning, shooting, tingling, allodynia, hyperalgesia, and pins and needles. These symptoms are thought to be maintained by peripheral and central sensitization.^2,7,36

Because taxanes do not cross the blood-brain barrier, spinal cord and central involvement are thought to originate from information provided from peripheral afferent input. Maintaining neuropathic pain potentially may arise initially from peripheral nociceptive input to the spinal cord. Furthermore, chemotherapy induces mitochondrial damage, which directly contributes to abnormal spontaneous discharge of Aδ and C-fibers with mechano and thermal hypersensitivity via TRPA1 (mechano and cold sensitivity) and TRPV1 (heat hyperalgesia) receptors at the distal terminal sensory ending. ¹¹ This distal axon mitotoxicity could be the cause of peripheral nociceptive input that would explain the sequence of events leading to spontaneous nerve firing (shooting pain), sensitized receptors (cold and hot hyperalgesia), and release of proinflammatory cytokines (allodynia and neuroinflammation).^4,11,37 Constant peripheral input or impaired processing can maintain spinal cord dorsal horn excitability (glia cell activation) resulting in cortical changes, central sensitization, and impaired descending noxious inhibitory control.^38-41 Activated glia (microglia and astrocytes) at the dorsal horn seem to be central in maintaining neuropathic pain, increasing proinflammatory cytokines, and brain-derived neurotrophic factor (BDNF). ⁴²

Animal Models of Exercise and Neuropathic Pain

Animal models have been developed to attempt to understand how exercise improves peripheral nerve regeneration and neuropathic pain. Chronic constriction injury (CCI) is a common model used to induce neuropathic pain by tying a suture around the sciatic nerve with resultant mechanical allodynia and thermal hyperalgesia. Research has shown that aerobic exercise and neural movement can reduce allodynia (measured by paw withdrawal to von Frey hairs) and thermal hyperalgesia (measured by infrared heat or cold plate withdrawal thresholds).^43-46

Armada-da-Silva et al ⁴³ have shown that physical exercise increases both the number of axons and rate of axonal elongation. Additionally, exercise prior to injury may help precondition the nervous system to be able to rehabilitate more effectively after injury. ⁴³ In a CCI rat model of neuropathic pain, voluntary wheel running before injury suppressed allodynia postinjury. ⁴⁷ There were 10 neural mobilization treatments (a commonly used nerve treatment technique in physical therapy) that decreased glial activity (GFAP expression, OX-42, and BDNF immunoreactivity) in the periaqueductal gray (PAG) and thalamic nuclei, which are related to maintaining neuropathic pain. ⁴⁴ Swim therapy (1 hour daily for 6 weeks) has been shown to significantly reduce allodynia and hyperalgesia in another rat model of neuropathic pain. ⁴⁵

Regular aerobic exercise has been shown to attenuate neuropathic pain mediated by endogenous opioid systems. ⁴⁶ Using a spinal nerve ligation model of neuropathic pain, mechanical and thermal hyperalgesia was reversed within 3 weeks with a 5-week forced treadmill training session postligation. The positive effects of exercise on neuropathic pain was intensity (not frequency) dependent, and the hypersensitivity returned to preexercise levels within 8 days of discontinuing the exercise. Results suggest that maintaining regular exercise of moderate intensity is required to preserve the protective effects. Mechanisms mediating this exercise-induced modulation of neuropathic pain are thought to be a result of increased endogenous opioid levels in the PAG and rostral ventralmedial medulla, and this was confirmed using an intracerebroventricular injection of naloxone. ⁴⁶

A mouse model of TIPN and exercise has been developed. ⁴⁸ Treadmill exercise began 1 week prior to paclitaxel and continued 7 d/wk for 4 weeks. Daily exercise prevented thermal hyperalgesia, and histological evaluations confirmed preservation of unmyelinated axons. The authors conclude that the impact of exercise demonstrates “a robust neuroprotective effect” and that an upregulation of BDNF in motor axons may assist in protecting the sensory axons from the neurotoxic effects of paclitaxel. ⁴⁸

Human Models of Exercise and Peripheral Neuropathy

Exercise has anti-inflammatory effects on the nervous system that may be critical in understanding how exercise can attenuate symptoms of neuropathic pain. Decreasing proinflammatory cytokines, normalizing glial cell activation and BDNF, and upregulating other neurotrophic factors along with other proposed mechanisms are being revealed in experimental studies. ⁴⁹ Exercise research on peripheral neuropathy has directed current clinical recommendations aiming to change sedentary behaviors. This includes high-risk diabetic populations where previously, because of safety concerns, exercise was prescribed with caution.^50,51

A recent systematic review on exercise intervention and peripheral neuropathy identified 18 clinical trials (majority studying diabetic neuropathy with 1 CIPN) and found that “exercise is safe, feasible, and beneficial (p. 1293).” ⁵² Significant reductions in pain along with increased branching of the intraepidermal nerve fibers on skin biopsy were observed after a 10-week aerobic and strengthening program for diabetic neuropathy. ⁵³ Another study using skin biopsy found that exercise and diet counseling improved pain, sensory amplitudes of the sural nerve, and IENFD. ⁵⁴ A 16-week aerobic exercise program for painful diabetic neuropathy symptoms found improved scores in pain interference with walking, work, sleep, and relationships measured by the Brief Pain Inventory, without a change to overall pain intensity. ⁵⁵ Balance, reaction time, and improvements to gait pattern and speed were observed after a 12-week exercise program in patients with diabetic neuropathy. ⁵⁶

Epidemiological research supports a possible neuroprotective role of exercise in CIPN. Mustafa et al ⁵⁷ examined risk factors for peripheral neuropathy in breast cancer survivors and found an association between women who reported exercising (at least 30 minutes on most days) with a 12% lower risk of peripheral neuropathy. ⁵⁷

A 36-week exercise program improved CIPN symptoms specific to patients treated for lymphoma on balance, mobility, and quality of life. ⁵⁸ Similarly, improvements in patient-reported quality-of-life measures and CIPN symptoms were reported after a 12-week exercise intervention for CIPN symptoms in cancer survivors. In addition to subjective improvements, objective scores on walk and balance tests improved. ⁵⁹ A recent 6-week home-based progressive resistance and walking program has demonstrated improvements in patient-reported CIPN symptoms, including numbness/tingling and hotness/coldness, compared with controls who did not exercise. ⁶⁰ Patients who exercised while receiving treatment for stage IV colorectal cancer demonstrated stable CIPN symptoms compared with controls who experienced worsening CIPN symptoms. ⁶¹ Specific to CIPN neuropathic pain, a pilot study used the McGill pain questionnaire and Leeds Assessment of Neuropathic Symptoms and Signs (S-LANSS) to evaluate a 10-week home exercise program in breast cancer patients experiencing painful CIPN. Enrolment was extremely low (n = 14), and adherence was worse, with a 50% drop-out rate (total of n = 6). Although the participants’ pain and S-LANSS scores were reduced, adherence to exercise must improve for this treatment to be beneficial. ⁶²

Physical therapy programs for cancer rehabilitation educate on the importance of exercise specific to neuropathic pain and TIPN. These programs often include postural correction, generalized stretching, gait retraining, cardiac conditioning, and strengthening. Incorporated in this training are exercises aimed at improvements in balance, fine dexterity, reaction time, coordination, and overall gross motor function following sensory deficits and motor weakness resulting from the peripheral neuropathies.

Exercise and Cancer

International guidelines, World Health Organization recommendations, and American College of Sports Medicine guidelines reinforce the importance of exercise for general health. It is believed that there is a near linear relationship between the benefits of exercise and health in a dose-dependent manner. A review of evidence-informed guidelines for Canadians identifies that if Canadians followed current guidelines (30-40 minutes of moderate-intensity exercise on most days), deaths could be reduced by 33% for coronary heart disease, 25% for stroke, 20% for colon cancer, 20% for diabetes, 14% for breast cancer, 20% for hypertension, and 25% for osteoporosis.^63,64 Despite this knowledge, in the general population, 48% of men and 54% of women are physically inactive. ⁶⁴

Cancer-specific population estimates reveal higher levels of obesity with lower levels of activity. Patient preference for type of exercise and social, cognitive, demographic, environmental, and medical variables are all important considerations when prescribing exercise during chemotherapy.^65,66 Specifics of barriers to exercise that patients report include treatment side effects, time constraints, and fatigue. Interestingly, the reported barriers and facilitators for exercise are often the very symptoms that can be improved with exercise.^67,68

For many cancer survivors, restoring physical function after diagnosis and treatment is difficult. A prospective breast cancer study (n = 267) measuring physical activity confirmed, as expected, that physical activity declined significantly postoperatively. Activity levels measured included occupation, sport, and household activity as well as an overall activity measure. Follow-up at 1 year found that these women had not returned to their preoperative physical activity levels. ⁶⁹

Physical activity during and after medical treatment is important to maintain and improve cardiovascular fitness, muscle strength, balance, flexibility, and range of motion; reduce fatigue and depression; and generally improve quality of life.^52,58,70,71 Moderate-intensity exercise improves musculoskeletal fitness and positive mental health (including reduced rates of both depression and anxiety) that are extremely important to maximize quality of life. ⁶⁴ Possible mechanisms for improved health with exercise include lower serum concentrations of sex hormones, improved insulin sensitivity, reduction in systemic inflammation, and improved immune function. ⁷² A Cochrane review specific to exercise and breast cancer found moderate support that exercise can lessen fatigue and improve physical fitness, quality of life, and cognitive function. ⁷³ Physical activity literature specific to cancer, identifies that a reduction of adipose tissue, reduction in circulating hormones, lower production of insulin, and reduction of reactive oxygen species and other inflammatory markers with natural killer cell activation may be possible mechanisms that might reduce cancer risk or recurrence.^72,74,75

The clinical oncology society of Australia recently published a position statement on exercise in cancer care. Their recommendations advocate that all members of the oncology team support exercise for patients. When possible, the current guidelines (150 minutes of moderate activity or 75 minutes of high-intensity aerobic activity and at least 2-3 resistance training sessions weekly) should be encouraged in order to reduce cancer-related side effects and disease burden. Although clinical trials and epidemiological data have correlated physical activity to improved disease-free and overall survival, larger phase III studies are required to confirm this association.^76,77

Conclusion

The pathophysiology of TIPN is complex and poorly understood. Several theories outline possible mechanisms of action, but few effective treatments exist. The same can be said for neuropathic pain originating from other injuries. Poorly understood mechanisms inciting and maintaining neuropathic pain are complicated, and successful management often requires a combination of physical therapy, pharmacological drugs, and psychological intervention to minimize symptoms. The general health benefits of exercise are well established. The past 2 decades of research on cancer and physical activity has positively correlated the recovery of treatment side effects, including neuropathy.^57,64,73,76 The evidence for the role of exercise in primary cancer prevention and secondary recurrence has been illustrated in epidemiological research and a few clinical trials; however, a causal mechanism has not been identified. The results from animal models of exercise and neuropathic pain are promising. A growing body of evidence that supports exercise for peripheral neuropathy is emerging. It is possible that exercise can assist in attenuating pain in cancer patients; minimize neuropathy symptoms and other known taxane side effects, including fatigue; and improve mental alertness, depression, general strength, endurance, and flexibility. Overlapping mechanisms in diabetic neuropathy and CIPN may help guide clinical recommendations for exercise in both fields. The goal of future research needs to identify the specifics of exercise intensity, frequency, duration, and type to provide the most benefit for prevention and treatment of TIPN and neuropathic pain.