Toxic neuropathy in patients with pre-existing neuropathy

Methods.

The patient demographics are given in table 1. All patients underwent a directed history, examination, and nerve conduction testing. A modified total neuropathy score (TNS) was calculated in all patients (see table 1).3

Results.

Details of the diagnosis and changes in pre- and post-chemotherapy components of the TNS evaluation are given in table 2. Five patients’ baseline evaluations (including nerve conduction studies) documented the presence of pre-existing neuropathy (1 idiopathic small fiber, 1 ataxia telangiectasia, 2 diabetic neuropathies, and 1 Charcot-Marie-Tooth). The last patient, who was a known diabetic and gave a history of heavy alcohol use, was asymptomatic. The mean TNS score was 9.5 (range 0 to 19). Chemotherapy was initiated for lung cancer (2), breast cancer (1), lymphoma (1), oropharyngeal cancer (1), and discoid lupus (1). The chemotherapeutic agents were vincristine (2), Taxol (3), cisplatin (1), and thalidomide (1). Doses of chemotherapy were as follows (Taxol [3] 125 to 175 mg/m² × four cycles; vincristine [2] 2 mg 1 patient, 5 mg 1 patient; cisplatin [1] 40 mg/m² × eight cycles; and thalidomide [1] 100 mg daily × 16 months and 50 mg daily for 8 months). All patients developed significant worsening of their neuropathy while on chemotherapy that continued to progress for up to 8 weeks (range 1 to 24 weeks) after discontinuing therapy. In three patients the neuropathic symptoms were dose limiting. The TNS for all patients post chemotherapy was 22 (range 13 to 29) (figure). For four patients (on thalidomide, Taxol, and cisplatin) the sensory symptoms (severe dysesthesia or ataxia of gait) were disabling and for two patients (on vincristine) disabling weakness were the presenting feature. Examination in all showed sensory (large and small fiber) and motor length-dependent dysfunction. Deep tendon reflexes were reduced to absent. Nerve conduction studies showed sensory and motor axonal loss in a length-dependent fashion. In one diabetic patient with discoid lupus, thalidomide had to be restarted twice because of refractory disease; each time, even with half the initial dose of thalidomide given for half the duration compared to the first time, the dysesthetic symptoms became intolerable and resulted in discontinuation of treatment.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Total neuropathy score for each of the six patients before and after chemotherapy.

Discussion.

With the advent of hematopoietic growth factors that ameliorate the neutropenia associated with chemotherapeutic agents, peripheral neuropathy has become the principal dose-limiting side effect of most chemotherapeutic agents.1 Factors that predispose patients to the development of peripheral neuropathy include the type of the drug used, dose used (both single and cumulative), history or prior exposure to known neurotoxic agents, synergistic effects of multiple neurotoxic compounds used together, and the presence of diabetes or alcohol use presumably related to the pre-existence of peripheral neuropathy.1 Given the dose-dependent nature of the toxic neuropathies, lower dosages and various alternate dosage regimens have been used to avoid the neuropathy.

The four drugs used in this study were paclitaxel, cisplatin, vincristine, and thalidomide, all of which have a known dose-dependent relationship to the development of peripheral neuropathy. Paclitaxel is an effective chemotherapeutic agent used for solid tumors. Single doses of paclitaxel greater than 250 mg/m² are followed by acute development of paresthesias and cumulative dosages of greater than 1,000 mg/m² are invariably associated with gradual development of sensory motor axonal neuropathy.4 Patients 1, 3, and 4 developed severe neuropathy after receiving paclitaxel at individual dosages of 125 mg or 175 mg and cumulative dosages of < 700 mg/m².

Cisplatin (Cis-diamine-dichloro-platinum) is an established antitumor agent that is effective in ovarian, bladder, lung, and testicular cancers. Cisplatin-induced neuropathy generally results in predominantly sensory neuropathy or neuronopathy. The severity of neuropathy is correlated to the cumulative dose, with levels above 400 mg/m² invariably associated with development of paresthesias, sensory ataxia (when severe), loss of vibration sensibility, and reduction/loss of deep tendon reflexes.4 In this study, one patient who had pre-existing risk factor of alcohol use and diabetes received cisplatin at 320 mg/m² and developed functionally disabling neuropathy.

Most patients treated with vincristine develop a dose-dependent neuropathy with sensory symptoms beginning at 5 mg and motor symptoms at higher cumulative doses around 30 to 50 mg.5 There are several reports of vincristine-induced severe paralysis at lower doses in patients with pre-existing hereditary neuropathy.6,7⇓ Our study corroborates this finding. Two patients developed severe paralysis: one patient with Charcot-Marie-Tooth disease at a dosage of 2 mg, another with ataxia telangiectasia at a dosage of 5 mg.

Thalidomide (alpha-[n-phthalimido] glutarimide), withdrawn from the market in 1961 because of its teratogenic effects, has been reintroduced for its immunomodulatory and antiangiogenic properties in various dermatologic conditions, graft-versus-host disease, rheumatoid arthritis, multiple myeloma, myelodysplastic syndromes, gliomas, Kaposi’s sarcoma, renal cell carcinoma, advanced breast cancer, and colon cancer.8 Polyneuropathy secondary to thalidomide was recognized as early as 1961 when the embryopathy was identified.9 The dose-dependent nature has not been recognized, however. We have recently reported that thalidomide-induced peripheral neuropathy also occurs in a dose-dependent fashion with cumulative dosages greater than 100 g resulting in significant neuropathy.10 The presence of pre-existing diabetic neuropathy in Patient 5 resulted in worsening of her neuropathy at a dose of 60 g.

With longer disease-free intervals due to advances in cancer therapeutics, quality of life concerns become increasingly important in the evaluation of potential chemotherapeutic regimens. There is no specific treatment for toxic neuropathy, and in most cases the neuropathic effects on quality of life are not reversible. Therefore prevention or attenuation of toxic neuropathy remains the major goal. Although close sequential monitoring and dosing within accepted safe norms may achieve this goal, in patients with pre-existing neuropathy this may not possible. In patients with pre-existing neuropathy or even at risk of neuropathy (e.g., diabetics), alternative chemotherapeutic nontoxic (to the nerve) agents may be desirable when feasible. Extremely close neurologic monitoring is recommended when neurotoxic agents such as paclitaxel, cisplatin, thalidomide, suramin, or vincristine are used in order to minimize the potential for functionally disabling neuropathy.

The mechanism of enhanced vulnerability to toxic neuropathy in the presence of pre-existing neuropathy is not known. Common sense suggests that “diseased nerves” may be more vulnerable to toxic nerve injury. An analogy might be found in the enhanced vulnerability with “double-crush” injuries. Some of the toxic drugs affect axoplasmic transport and others may affect nerve excitability by altering ion conduction of the axon or Schwann cells. In nerves with functional or structural compromise, either mechanism could potentially be detrimental at a lower dose.