Working Hand in Glove to Prevent Chemotherapy-Induced Peripheral Neuropathy

Interventions that prevent and treat the adverse effects of cancer therapy may improve both quality of life and survival by increasing the likelihood that patients complete their treatment. Over the last decade, substantial improvements have been made in preventing chemotherapy-induced emesis, neutropenic fever, and, more recently, chemotherapy-induced alopecia (1–3). However, despite numerous efforts and multiple randomized trials, no intervention has successfully prevented the development of chemotherapy-induced peripheral neuropathy (CIPN), one of the most common and debilitating treatment toxicities (4). In the current issue of the Journal, Hanai etal. report results of the first self-controlled trial of cryotherapy for the prevention of taxane-induced peripheral neuropathy (5). Women with breast cancer undergoing at least 12 treatments of planed weekly paclitaxel (80 mg/m2) wore frozen gloves and socks on their dominant side before, during, and after each infusion. Objective and subjective outcome measures were compared between the treatment and control sides, so that the patient served as their own control with regard to risk and exposure. CIPN was predefined as any decline in tactile sensation assessed by the Semmes-Weinstein monofilament test, an outcome measure routinely used in studies of diabetic neuropathy (6). The primary outcome of this trial was met. On the control, nondominant side, 80.6% and 63.9% of patients had a decrease in tactile sensation in their hands and feet, respectively. However, on the intervention, dominant side, only 27.8% and 25.0% were found to have a decrease in tactile sensation in their hands and feet, respectively. Subjects were also found to have less CIPN on the intervention as opposed to the control side, as measured by the patient neuropathy questionnaire. Differences in perception of warmth and reaction speed were also observed, but no differences in vibration threshold or electrophysiologic testing were detected between the intervention and control sides. Designing studies to evaluate interventions to prevent the development of toxicities can be methodologically challenging. The first issue involves blinding. Studies involving physical modalities such as exercise, topical therapies, and behavioral interventions are often unblinded or difficult to effectively blind. This can result in reporting bias by both subjects and investigators on both objective and subjective measures. Systemic interventions are often placebo controlled and, as a result, often have a large placebo effect. A large placebo effect can make it difficult to find a meaningful difference between groups. In addition, due to heterogeneity in baseline risk for developing the toxicity and differences in drug exposure, clinical trials often require large sample sizes. While the current trial suffered from an inability to blind, the outcome (CIPN) was common, the effect size was large, and a difference between the intervention and control was observed in both subjective and objective outcome measures. In addition, while the study was small, the within-subject comparison removed bias from unmeasured risk factors and potential differences in drug (ie, taxane) exposure or metabolism. Studies of CIPN have been limited by the absence of a clearly defined phenotype for CIPN. There are several methods available to assess CIPN including objective assessments, such as clinical or neurophysiological examinations, and subjective assessments, such as the National Cancer Institute–Common Terminology Criteria for Adverse Events (NCI-CTCAE) grading scale and a variety of patient-reported outcome measures. The primary outcome for this study was any decrease in tactile sensation, assessed by the Semmes-Weinstein monofilament test. This is the most widely used test to diagnose the loss of protective sensation, which is one component of CIPN (6). CIPN typically induces a sensory neuropathy, with symptoms that are thought to result from an increase in the function of sensory neurons such as paresthesias, tingling, and pain, as well as symptoms that likely reflect loss of function such as numbness and dulled sensation and a loss of position and/or vibratory sense (7,8). It is not clear that all of these symptoms could be assessed by the measures used in this trial. While the mechanism explaining the efficacy of topical cooling is unknown, it is thought that hypothermia decreases blood flow to the skin and hair follicles, and therefore reduces local exposure and subsequent toxicity (9). A similarly designed trial of hand cryotherapy during docetaxel treatment reported a statistically significant reduction of onycholysis (grade 0, 89% vs 49%) and skin toxicity (grade 0, 73% vs 41%) on the intervention side vs control side, respectively (10). In addition, the results from two prospective clinical trials showing benefits of scalp-cooling systems to prevent chemotherapy-induced alopecia were recently reported (2,3). In both trials, hair retention among patients receiving taxanes was 60% to 65% in patients receiving scalp cooling. It is unclear why a similar topical approach would prevent CIPN because CIPN is not thought to be a local toxicity. The pathophysiology of CIPN appears to vary depending upon the chemotherapeutic agent. Hypothesized mechanisms of taxane-induced peripheral neuropathy include the disruption of the axonal microtubule structure and a deficit in axonal energy supply through the toxic effect of chemotherapy on mitochondria in primary afferent neurons (11). It is also unclear if cryotherapy interventions would provide benefits to patients undergoing platinum therapy, another common agent causing CIPN. Platinum agents are thought to cause CIPN by exerting damage in the dorsal root ganglion through neuronal apoptosis, either by DNA cross-linking or oxidative stress, and mitochondrial dysfunction (11). Larger randomized trials of cryotherapy, with outcome measures that can differentiate sensory, motor, and autonomic neuropathy, are needed to answer these questions. To date, agents that have been most successful in the treatment of CIPN, such as duloxetine (12), have been directed toward the painful component of CIPN. However, even for painful neuropathy, the uptake of these agents is limited due to cost and side effects. If the results are confirmed, cryotherapy has the advantage of a limited side effect profile, it is low cost, and it appears to prevent components of CIPN other than neuropathic pain. Ultimately, a better understanding of the biologic mechanisms causing CIPN will improve our ability to effectively prevent and treat all components of this toxicity. The author has no conflicts of interest to declare.