Symptoms: Chemotherapy-Induced Peripheral Neuropathy

Pharmacologic intervention

Neurotoxic agent

Reference

Duloxetine

Taxane or platinum

Smith et al. (2013a)

Gabapentin

Vinca or platinum or taxane

Rao et al. (2007)

Lamotrigine

Vinca or platinum or taxane

Rao et al. (2008)

Nortriptyline

Cisplatin

Hammack et al. (2002)

Vinca or platinum or taxane

Kautio et al. (2008)

Topical amitriptyline, ketamine ± baclofen (BAK)

Vinca or platinum or taxane or thalidomide

Barton et al. (2011)

Table 6.2

ASCO practice guidelines for chemotherapy induced peripheral neuropathy Hershman et al. (2014)

Prevention of CIPN

There are no established agents recommended for the prevention of CIPN in patients with cancer undergoing treatment with neurotoxic agents. This is based on the paucity of high-quality, consistent evidence and a balance of benefits versus harms

• Clinicians should not offer the following agents for the prevention of CIPN to patients with cancer undergoing treatment with neurotoxic agents:

– Acetyl-l-carnitine (ALC)

– Amifostine

– Amitriptyline

– CaMg for patients receiving oxaliplatin-based chemotherapy

– Diethyldithio-carbamate (DDTC)

– Glutathione (GSH) for patients receiving paclitaxel/carboplatin chemotherapy

– Nimodipine

– Org 2766

– All-trans-retinoic acid

– rhuLIF

– Vitamin E

– Venlafaxine is not recommended for routine use in clinical practice. Although the venlafaxine data support its potential utility, the data were not strong enough to recommend its use in clinical practice, until additional supporting data become available

– No recommendations can be made on the use of N-acetylcysteine, carbamazepine, glutamate, GSH for patients receiving cisplatin or oxaliplatin-based chemotherapy, goshajinkigan (GJG), omega-3 fatty acids, or oxycarbazepine for the prevention of CIPN at this time

Treatment of CIPN

For patients with cancer experiencing CIPN, clinicians may offer duloxetine

No recommendations can be made on the use of:

– ALC, noting that a positive phase III abstract supported its value, but this work has not yet been published in a peer-reviewed journal, and a prevention trial suggested that this agent was associated with worse outcomes

– Tricyclic antidepressants; however, based on the limited options that are available for this prominent clinical problem and the demonstrated efficacy of these drugs for other neuropathic pain conditions, it is reasonable to try a tricyclic antidepressant (e.g. nortriptyline or desipramine) in patients suffering from CIPN after a discussion with the patients about the limited scientific evidence for CIPN, potential harms, benefits, cost, and patient preferences

– Gabapentin, noting that the available data were limited regarding its efficacy for treating CIPN. However, the panel felt that this agent is reasonable to try for selected patients with CIPN pain given that only a single negative randomized trial for this agent was completed, the established efficacy of gabapentin and pregabalin for other forms of neuropathic pain, and the limited CIPN treatment options. Patients should be informed about the limited scientific evidence for CIPN, potential harms, benefits, and costs

– A topical gel treatment containing baclofen (10 mg), amitriptyline HCl (40 mg), and ketamine (20 mg), noting that a single trial indicated that this product did decrease CIPN symptoms. Given the available data, the panel felt that this agent is reasonable to try for selected patients with CIPN pain. Patients should be informed about the limited scientific evidence for the treatment of CIPN, potential harms, benefits, and costs

A major limitation for the intervention trials to date is a limited understanding of the underlying pathophysiology. If the goal of therapy is to repair an underlying problem then lumping patients with different types of sensory or motor neuropathy may be as inefficient as studying anti-HER2 therapies for all tumors regardless of HER2 status.

Assessment of Neurotoxicity and Relationship to Mechanisms

As stated above, much of what we know about the frequency and severity of CIPN is derived from the CTCAE as this has been widely employed in many of the large clinical trials that have studied these agents. There are multiple limitations to this reporting methodology (Hershman et al. 2011). This criterion is largely based on the degree of impact on functionality. This is a practical criterion when considering the need to dose reduce but is not overly helpful in distinguishing the various types/manifestations of the process. This limits the ability to fully characterize the true diversity and frequency of the toxicity. Importantly, it hinders correlative work aimed at predicting the toxicity and unveiling the mechanistic underpinnings. If, indeed, the various types of neuropathy symptoms reflect differences in unique pathophysiologies between the different neurotoxic chemotherapy agents, then correlative biomarker work requiring large sample sizes where CTCAE was used will suffer from dilution of the true associations. Additionally, recent data suggest poor concordance among raters for cancer therapy-induced toxicities (Cella et al. 2003, Cancer Therapy Evaluation Program, August 9, 2006). Further, there is discordance in agreement between clinician raters and the patients who experience these toxicities. Specific to taxane-induced peripheral neuropathy, the use of patient reported outcomes (PROs) correlate well with vibration threshold testing and are effective for both acute symptoms and over long-term follow-up (Hershman et al. 2011). With the recognition that the toxicity profile is an important piece to optimize the therapeutic index, many of the large studies have begun to incorporate PROs as a superior endpoint/phenotype. Finally, and perhaps most importantly, many trials do not capture the long-term frequency of CIPN. This is a big limitation as enduring or irreversible neuropathy may be the most clinically important phenotype to identify and study.

Newer Research Strategies

There are a number of ongoing clinical trials to further identify agents that might prevent or treat CIPN. In addition, other modalities are currently undergoing testing such as acupuncture, topical menthol and cutaneous electro-stimulation devices. With anticipated new insights into the mechanism of CIPN, additional advances can be made with rational selection of drugs. The identification of common mechanisms of neuropathy caused across drug classes will be important for the development of drugs that can be implemented broadly. Additionally, understanding the pathophysiology for the spectrum of CIPN symptoms will be critical to optimal drug development.

Identifying Host Factors Related to Risk for CIPN

Another evolving area of research is the use of germline genetic variability (i.e. SNPs, copy number variations, etc.) to predict drug-induced toxicity. This has become a provocative area of research as variants can impact the metabolism, transport, and excretion of drugs, but can also impact the target tissue (i.e., neurons). A candidate study from an institutional series demonstrated an association between a variant in a paclitaxel metabolizing enzyme, CYP2C8*3 and neuropathy (Hertz et al. 2013). Another candidate approach from the adjuvant breast cancer trial SWOG-0221, demonstrated an association between a SNP in FANCD2 and taxane induced neuropathy (Sucheston et al. 2011). Recently, several large genome wide association studies (GWAS) have been conducted from large clinical trials involving taxanes. The CALGB-40101 investigators identified a SNP in FDG4 that correlated with increased likelihood of paclitaxel-induced neuropathy (Baldwin et al. 2012). FDG4 is associated with the hereditary neuropathy condition of Charcot-Marie-Tooth disease. Subsequent pathway and modeling work with this data set have suggested that a hereditable predisposition to this toxicity may lie in genes involved in axon outgrowth (Chhibber et al. 2014). Another trial specifically focused on more rare variants using massively parallel sequencing of 20,794 genes associated with heredity neuropathy from patients who had received paclitaxel-based chemotherapy (Beutler et al. 2013). The investigators reported an association between EPHA5, ARHGEF10, and PRX and paclitaxel-induced neuropathy. It is hopeful that the identification of genomic predictors might not only play a useful role in predicting who will be at high or low risk of this toxicity but may also shed insight into the biological underpinnings to help with future drug development. Further, the integration of these genetic factors with other predictors such as race, weight, or co-morbid states might allow for a truly personalized and successful predictor for likelihood of this toxicity.

Future Directions

There are multiple areas of research progress to improve our approach to those who might be at risk for CIPN and here we outline three immediate areas of need: education, drug development, and predictive biomarkers.

For any toxicity, patient and physician education are paramount to successful management. Physicians who fail to prioritize toxicities cannot adequately counsel the true risk to benefit ratio of the therapies they plan to deliver. This is particularly true in the case of adjuvant chemotherapy, where the vast majority of women can expect to be long-term survivors. Additionally, this makes it difficult for patients to prepare themselves mentally for potential challenges and hurdles that await them. Recent data demonstrated that the counseled frequency and severity of neuropathy could impact the specific regimen a patient might choose (Smith et al. 2013b). Further, patients who had previously experienced neuropathy were actually more likely to choose (less worried about) a regimen that would cause mild neuropathy and markedly less likely to choose (more worried about) a regimen that would cause severe neuropathy (Smith et al. 2013b). These data demonstrate that those patients who experienced the toxicity are more nuanced in their decision-making based on their personal understanding of the toxicity. This implies that physicians must strive to educate patients in a more detailed fashion so that they might make the best possible decision for their therapy.

Improved drug development is also drastically needed. The bar is high, however, for drugs that prevent or treat a specific side effect. The first hurdle is to identify drugs that are highly effective. As outlined above this has not been an easy task, to date, in part because the underlying pathophysiology is poorly understood. Second, the preventive drug must not have significant toxicity. Many patients are willing to accept substantial drug-induced toxicity for the payoff of increased cure rate, but few like the idea of trading one drug side effect for another. Finally, the preventive drug, ideally, will be affordable. In our current healthcare economic environment, extremely expensive supportive care drugs will be less likely to be paid for by insurance without substantial evidence of benefit and this may leave many patients unable to afford the option.

Only gold members can continue reading. Log In or Register to continue