Toxic medications in Charcot–Marie–Tooth patients: A systematic review

Several widely used medications, with a relevant efficacy profile, are toxic to the peripheral nervous system and an even larger number of agents are suspected to be neurotoxic. There are concerns about the use of these drugs in patients with Charcot–Marie–Tooth disease (CMT), a hereditary motor and sensory neuropathy. This review provides evidence‐based updated recommendations on this clinically relevant topic.


| INTRODUCTION
Several widely used medications, with a relevant efficacy profile, are toxic to the peripheral nervous system (PNS) 1 and an even larger number of agents are suspected to be neurotoxic. Concerns about the use of these drugs refer to subjects with an intact PNS, but even more to those patients with pre-existing peripheral neuropathy. Among this category, subjects with inherited neuropathies have been suggested to be at potentially higher risk. 2 Charcot-Marie-Tooth disease (CMT) is a hereditary motor and sensory neuropathy. CMT affects around 1 in 2500 people, 3 million worldwide, and is the most commonly inherited neurological disorder.
The overall incidence of CMT and closely related inherited peripheral neuropathies, such as hereditary neuropathy with liability to pressure palsies (HNPP), distal hereditary motor neuropathies (dHMN or HMN), and hereditary sensory (and autonomic) neuropathies (HSN or HSAN) has been estimated to be 17.69/100 000, although prevalence estimates vary across studies up to 82.3/100 000. 3,4 The Charcot-Marie-Tooth Association (CMTA), a leading global CMT patient advocacy organization, has a "Medical Alert" list of potentially neurotoxic medications published on its webpage (https:// www.cmtausa.org/living-with-cmt/managing-cmt/medications/). This list is also reported in a brochure that is distributed along with a "Dear Medical Professional" letter advising physicians treating CMT patients that they should consider the potential risk of prescribing drugs known to have neurotoxic properties.
This information influences patients and their treating physicians, particularly if they are not neurologists. Among the listed medications only vinca alkaloids and taxanes are considered drugs with a definite high risk if used in CMT patients. Several other drugs are reported to carry different levels of risk, ranging from "negligible or doubtful" up to "moderate to significant," contributing to a certain level of alarm regarding these drugs.
The CMTA's published list was based on a 2006 paper by Louis H. Weimer and David Podwall. 2 After an extensive literature search for reported cases of drug effects in CMT patients, Weimer and Podwall stated that in unexpectedly severe cases "the vast majority concerned excessive vincristine toxicity in patients with undiagnosed demyelinating forms of CMT," referring specifically to CMT type 1A, and most likely HNPP.
However, the authors also state that "use of other agents in the significant risk category and use of vincristine in other CMT subtypes should be considered with caution" although "this recommendation is based on very limited direct evidence in patients with CMT." Given the clear importance of a reliable information supporting or negating a higher peripheral neurotoxicity we performed a systematic review of the scientific literature regarding the severity of druginduced side effects in inherited neuropathies to provide evidencebased updated recommendations.

| SEARCH METHODS
To update and revise the CMTA neurotoxic medication list, we performed a systematic review of the scientific literature searching for papers reporting only clinical data and written in English, according to the PRISMA 2020 guidelines (see Figure 1). 5 The PubMed website was interrogated in the period July-   12 See Table 1 See Table 1 See Table 1 See Table 1 See   Table 1.
For taxanes, 52 papers were retrieved, with 6 of them fulfilling the selection criteria reported above. The results for the other drugs on the CMTA neurotoxic medication list are reported in Table 2. In Table 3 we report the number of references found for all drugs other than vincristine and taxanes, which were used to build Table 2.

| Vincristine
In the reported cases, vincristine use was associated with sudden worsening of peripheral neuropathy, often with prominent motor impairment and involvement of the cranial nerves. Most reports occurred in patients with CMT1A, but there are also reports in subjects with the inherited neuropathies CMT2, HNPP, CMT4, and CMT1X (Table 1).
Autonomic failure (a very typical side effect of vincristine) was reported, in association with voice changes and limb weakness, in only one subject with inconclusive genetic testing for CMT, who was also treated with intrathecal methotrexate and cytarabine (Ara-C). 31 Since  Table 1.

| Taxanes
Increased severity of paclitaxel-induced peripheral neurotoxicity was associated with specific polymorphisms in ARHGEF10 50,51 and T A B L E 3 Summary of the entry found to build  Table 1) cohort. 53 One case report suggesting an atypically rapid onset of paclitaxel-related worsening of peripheral neuropathy was reported in a 60-year-old woman with a "past medical history significant for CMT", without any genetic testing available: in this case no alternative diagnosis was searched. 54 A summary of the data linking taxane administration to a possible increased severity of peripheral neurotoxicity in CMT patients is reported in Table 4.

| Other neurotoxic agents
Cisplatin administration was associated with unusual rapid and severe onset of sensorimotor neurotoxicity in CMT1A patients. 35  and no association was found between 49 canonical CMT-associated genes and risk of severe oxaliplatin-induced peripheral neurotoxicity. 45 A summary of the results of our search excluding vincristine and taxanes, but with results potentially indicating a role in more severe neurotoxicity in CMT patients is reported in Table 2.
We also explored if Immune Checkpoint Inhibitors and antibodydrug conjugates 57 could be considered at increased risk in CMT patients, but no reports suggesting worsening of peripheral neuropathy in these patients were found.

| DISCUSSION
Apart from a set of drugs with a clear peripheral neurotoxicity profile (e.g., several anticancer chemotherapy drugs), the entire issue of drugrelated peripheral nerve damage is complex and not completely settled in subjects with or without pre-existing peripheral neuropathy.
Most of the drugs with a suggested, but uncertain, peripheral neurotoxicity profile including several antimicrobial, antiepileptic, antiarrhythmic, and antineoplastic agents, have been described in small cohorts or, more frequently, in single case reports. When welldesigned, large, carefully investigated series were used to confirm the preliminary observations the results were negative in most cases. This suggests no significant evidence for the claim or that the neurotoxic effect was not clinically relevant in the vast majority of patients.
The definite neurotoxic drugs have a well-established clinical phenotype and course, although the severity of PNS damage can be markedly different among treated patients. The strongest evidence for the wide range of severity can be seen in anticancer drug-treated patients, 57,58 representing the largest cohort of subjects exposed to the risk of severe The assumption that patients with CMT may be more susceptible to increased severity of additional toxic PNS damage is reasonable.
However, for chemotherapy-induced peripheral neurotoxicity (CIPN; one of the most likely candidates for unexpectedly severe PNS Our search was mainly based on the CMTA neurotoxic medication list and largely supported by the Weimer and Podwall review, 2 so it included a large number of putative peripheral neurotoxic agents. However, it is interesting to consider the findings of a recent review on the broader topic of toxic neuropathies conducted by Peters and Staff. 1 In their paper, the authors limit the list of definite peripheral neurotoxic drugs to three main classes