Treatment of Chronic Oxaliplatin-Induced Peripheral Neuropathy: A Systematic Review

Introduction: Oxaliplatin is a platinum-derivative chemotherapeutic agent used in digestive tumours, in the adjuvant and metastatic setting. Oxaliplatin can cause a chronic peripheral sensory neuropathy which impacts the quality of life and is dose limiting. To date, no therapeutic strategies have proved effective in the treatment of oxaliplatin-induced peripheral neuropathy (OIPN). Methods: A computerized search of the literature on PubMed database was performed. Publisher original articles were included if they focused on treatment of peripheral neuropathy among patients submitted to oxaliplatin. Ele-ven out of 242 reviewed papers met our inclusion criteria and were subjected to a 19-item quality checklist. Results: The included studies differed with re-spect to study design, patient population and sample size, neuropathic symptoms assessment and efficacy measure. Most studies had an adequate quality. Ten trials tested one drug, and one pilot study tested a non-pharmacological treatment—the neurofeedback. Of these, 3 trials included only patients submitted to oxaliplatin-based chemotherapy. Duloxetine showed moderate efficacy in 3 trials. Topical treatment with capsaicin or 10% amitriptyline was promisors in 2 single-arm trials with a few samples. Conclusion: In the last decade, there wasn’t an improvement in the treatment of chronic OIPN. The duloxetine is the unique drug with moderate efficacy on the treatment of OIPN. There is insufficient evidence to support a recommendation for any other treatment.


Introduction
Oxaliplatin is a platinum-derivative chemotherapeutic agent used in digestive tumours, in the adjuvant and metastatic setting [1] [2] [3]. Neurotoxicity is the most significant adverse effect. Oxaliplatin-induced peripheral neuropathy (OIPN) occurs in two distinct forms: an acute neurotoxicity and a chronic cumulative sensory neuropathy [3] [4] [5] [6]. Acute symptoms manifest as a cold-precipitated tingling distal paraesthesia and pain, perioral paresthesias and muscle cramps, in approximately 80% of patients. These symptoms occur within hours or days after oxaliplatin course and typically resolve within a week of infusion [7]. It does not require dose reduction.
The chronic form of OIPN is a pure sensory, axonal neuropathy and is usually seen after cumulative doses of 780 -850 mg/m 2 . It can surge after discontinuation treatment [8]. OIPN may be present in 26% -46% of patients at the 12-month follow-up [8]. Typical presentation is characterized of devastating and non-cold related pain and pronounced dysesthesia/paraesthesia, according to a typically, symmetrical and distal, "stocking and glove" distribution. Sensory loss and dysfunction of fine sensory-motor coordination can also occur. Motor nerve function usually remains normal. OIPN improves, not completely, in approximately 6 -8 months after discontinuation of oxaliplatin treatment, especially in upper extremities [2] [3] [7]. OIPN leads to a reduction in dose(s) and/or discontinuation of chemotherapy (ChT), which can negatively impact cancer-related outcomes. Chronic OIPN contributes for functional difficulties with activities of daily living and a negative influence on quality of life [7] [9]. It is an important concern for long-term survivors. Beyond the established association with oxaliplatin dose, no clinical or patient-related factors were consistently associated with the incidence and severity of OIPN [10]. To date, no prevention strategies of OIPN have proved effective [2] [3] [11] [12].
Face to other neurotoxic ChT regimens (taxanes, vinca alkaloids, etc.), oxaliplatin caused a unique spectrum of clinical presentation [13] [14]. Studies suggested that OIPN occurred by a different mechanism compared to other neurotoxic agents, not yet fully understood [14] [15]. Treatment of chemotherapy-induced peripheral neuropathy (CIPN) with different drugs, with different mechanisms of action, was disappointed until now [16]. So, management of chronic OIPN is a substantial challenge for medical oncologists.
To address this issue, in this systematic review, we aimed to identify all studies, irrespective of study design, that investigated pharmacological and non-pharmacological treatments for chronic OIPN.

Material and Methods
Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement [17] was used as a guide and template for every step of the study.
An electronic search of the PubMed database was performed to obtain key literature in the field of OIPN. Final access dates February 2, 2020 . Only one selection filter was used, which was a "clinical trial".
Studies that met the following criteria were included: 1) if CIPN was assessed among cancer patients treated with oxaliplatin; 2) if the study focus is the treatment of CIPN; 3) if the publication was an original study (e.g. no review, poster abstracts, editorials, letters to the editor, etc.); 4) if they were published in peer-reviewed journals and, 5) if they were written in English. Studies were included irrespective of study design, cancer diagnosis or year of publication. Additional studies were included after analysed of the references of the selected full texts, based on the same inclusion criteria. In summary, therapeutic phase II or III clinical trials investigating the efficacy of pharmacological or non-pharmacological treatment of CIPN, among patients treated with oxaliplatin, published in English, were considered for this review. Table 1 illustrates the detailed criteria for quality assessment of the eligible studies was conducted, which they based on criteria adapted from a previously published systematic review on OIPN [10] and the CONSORT statement checklist [18].
Detailed data from study design, type of study (single, multicentric, or population-based cohort), characteristics of the patient population (cancer diagnosis, the regimen of chemotherapy and other), treatment (pharmacological or non-pharmacological management), and treatment outcomes, were extracted from the eligible studies and resumed in Table 2. Figure 1 summarizes the study selection process in a stepwise fashion with reasons for exclusion at each step. Overall, a total of 242 papers were retrieved. A preliminary review of the title and abstracts led to exclusion of 233 papers for 3 reasons: studies do not focus on OIPN; studies focus on prevention of OIPN and studies focus on other themes like pathophysiology, OIPN assessment methods, etc. For the remaining 9 articles were reviewed to determine eligibility. Two relevant non-English studies were excluded. One article was eliminated due to address not-human participants. After reviewing the reference text of the remaining 6 articles [19]- [24], 5 additional relevant articles were included [25] [26] [27] [28] [29]. Table 2 illustrates a summary of the key characteristics of the reviewed trials as well as their studied population, intervention and important outcomes.
Across the 11 studies reviewed here, there was a large variation in the methods used to assess peripheral neuropathy, in terms of the criteria employed and effi- Chinese patients with chronic neuropathy and colon cancer in stage III and IV, after treatment with oxaliplatin-based ChT [19]. Concomitant antidepressants or anticonvulsants weren't permitted. After 12 weeks of treatment with duloxetine (60 mg/day, orally), there was a significant reduction in neurotoxicity: 63% of patients have a VAS score decreased by more than 30% at the end of the 12 weeks, and 47% of patients have an NCI-CTCAE v3.0 grade improvement (a lower grade at the end of treatment compared to the grade at the beginning, which mean impairment of daily function). The major limitations of this study were the little size sample and the absence of a controlled arm.
The benefit of duloxetine (60 mg/day) for treating chronic CIPN was confirmed by a multicentric, randomised, placebo-controlled crossover trial, in 2013 [20]. This study enrolled 231 patients with any cancer diagnosis and any stage, who received taxane or platinum-based ChT (129 patients were undergoing to oxaliplatin). After the initial 5 weeks of treatment, cross-over occurred following a 2-weeks wash-out period for a total study duration of 14 weeks. Chronic CIPN was defined by the presence of the symptoms for ≥3 months beyond ChT completion [20]. Five weeks of duloxetine treatment resulted in a statistically and clinically significant improvement in pain, as like as, it improved function and quality of life. In the analysis by subgroup, it was also found that the effectiveness of the treatment was greater in the group that had previously undergone oxaliplatin-based ChT, rather than taxane-induced painful neuropathy. In this review, this is the unique phase III trial that elucidates an effective intervention in the control of neuropathic pain secondary to OIPN.
The effectiveness of duloxetine was also supported by a small, randomised Japanese trial, employing 34 patients, comparing duloxetine to vitamin [21]. The patients were divided into 2 groups with equivalent clinical and demographic characteristics in both arms. After the initial 4 weeks, cross-over occurred following 2 to 4 weeks of wash-out period. After 4 weeks of treatment, in the 2 arms, duloxetine was shown to be more effective than vitamin B12. Unlike previous studies, the dose of duloxetine used was 40 mg/day (a dose regulated in Japan Unfortunately, the magnitude of benefit of duloxetine is modest and much less than is desirable. Relatively to safety, the discontinuation rate was significant and ranged from 11% to 23.1%. The adverse effects more frequent were fatigue (7% -18%), nausea (5% -10%), insomnia (5%) and somnolence (5%). Most of the adverse effects were manageable. It should be noted that the discontinuation occurred only in the first 3 weeks of treatment with duloxetine at 60 mg/day [19]. It means that if the patients can tolerate the duloxetine dosages at the beginning of treatment, they can be expected to maintain duloxetine without discontinuation.

Amitriptyline
Amitriptyline was also tested in the same setting of patients. Arandomized, double-blind, placebo-controlled trial enrolled to 44 patients with CIPN (numbness, tingling, pain) with a severity of ≥3/10 [29]. Patients, who reported neuropathic symptoms and were submitted at least 2 months of neurotoxic ChT, were included. In this trial, amitriptyline was started at 10 mg/day, oral, and if tolerated, This results probably due to low dose of the amitriptyline and small sample size.
Amitriptyline was well tolerated on the target dose. Only 14 patients were submitted to platinum agents; but the number of patients submitted to oxaliplatin was not mentioned, so the amitriptyline isn't an effective treatment of OIPN.

Gabapentin
Gabapentin is an antiepileptic, GABA analogue, which is effective in treating symptoms from several neuropathic syndromes. However, a phase 3 randomised, double-blind, placebo-controlled, crossover trial failed to prove the efficacy of gabapentin in the treatment of CIPN [25]. In this trial included 115 patients with symptomatic CIPN for ≥1 month and pain scores of NRS ≥4 or Eastern Cooperative Oncology Group (ENS) sensory neuropathy ≥1, who were randomly assigned gabapentin (target dose = 900 mg tid) or placebo. After the initial six weeks, cross-over occurred following a 2-week washout period. Changes in symptom severity, measured by NRS and ENS, were similar in both groups.
Adverse events were mild and similar in both groups. So, the study was not able to confirm the benefit of the use of gabapentin in ameliorating peripheral neu- to oxaliplatin, and there weren't reported outcomes in this subgroup.

Pregabalin
Pregabalin is a GABA analogue, structurally related to gabapentin, but it has rapid dose-independent absorption and a more tolerable side effect profile. Pregabalin achieved moderate efficacy in a non-randomised trial, which included 23 patients with gastrointestinal cancer undergoing oxaliplatin-based ChT [26]. In this trial, pregabalin was started at 50 mg, oral, three times a day (tid), and if tolerated, the dose was escalated to a maximum of 150 mg tid. In the majority of patients (48%), neuropathy improved by 1 to 2 grades (NCI-CTCAE v3.0), with the onset of benefit observed in 2 -6 weeks. The best benefit was observed at a dose of 150 mg tid, but patients experience significant symptomatic relief even at lower doses of 50 mg tid. The intensity pain reduction with pregabalin wasn't reported. So, the efficacy measure used does not allow direct comparison with other studies. Besides, although the ChT regimens were described, the cumulative dose of oxaliplatin wasn't reported. The chronicity of the symptoms is not explicit in this study.

High-Dose 8% Capsaicin Patch
Based on its effectiveness on the treatment of neuropathic pain in HIV-associated neuropathy and post-herpetic neuralgia, high-dose topical capsaicin was tested on the treatment of CIPN-associated pain. A single-centre trial, the application of high-dose capsaicin patch resulted in significant pain relief, without considerable adverse events [23]. In a sample of 18 colon cancer patients undergoing oxaliplatin-based chemotherapy, with painful peripheral polyneuropathy, there was an average pain score in approximately 84% to 97% patients after 12 weeks of application of capsaicin patch. There was a reduction in pain intensity from 7.45 ± 1.14 to 0.20 ± 0.41 (after 12 weeks), using NRS. It is not clearly described how many patients experienced pain relief. Patients with lower sensitivity to neurotoxic agents (patients whom the symptoms appeared after the treatment with higher cumulative doses of oxaliplatin: ≥648.07 mg/m 2 ) have a better response to treatment and pain reduction. About 50% of patients were being treated with gabapentinoids and/or antidepressants. We do not know in what context these drugs were prescribed and what effect this has on neuropathic pain previously. Although the symptoms were described, the chronicity of these wasn't explicit in this study, neither the oxaliplatin-based ChT regimens. A small sample is the major limitation of this study.

Topical Amitriptyline
The effects of topical amitriptyline in the treatment of CIPN have been evaluated in 3 trials: 2 randomized, double-blind, placebo-controlled trials and 1 clinical trial with a single arm. The first 2 trials assessed topical amitriptyline adminis-  [28]. The third trial assessed topical amitriptyline in high concentrations (10%) [24]. These 3 preparations can only be manufactured by a compounding pharmacy. Barton  Other study tested other topical treatment -2% ketamine plus 4% amitriptyline cream (KA cream), on 462 cancer survivors with CIPN [28]. The patients applied 4 g of cream (using a measuring device), twice daily, during 6 weeks, to each area with pain, numbness, and/or tingling at a single time. They completed a seven-day daily pain, numbness and tingling diary one week before study entry, and at 3 and 6 weeks after enrolment. The average score at 6 weeks was the primary outcome. This study was negative because the KA cream application wasn't associated with a decrease in CIPN symptoms (p = 0.363). Secondary analyses assessing pain alone, using NRS score, also showed no benefit for KA cream (p = 0.400). Despite gastrointestinal tumours being the second most common neoplasm (27%), the number of patients submitted to oxaliplatin was not mentioned.
Finally, a pilot study tested the high concentration amitriptyline cream in the treatment of patients with newly diagnosed CIPN for less than 1 month (group 1) or CIPN for more than 1 month who had not to respond to previous pharmacologic treatment (antidepressants, anticonvulsants, opioids) and discontinue this previous treatment (group 2) [24]. Eligible patients were asked to apply a thin layer of 1 g of 10% amitriptyline cream, and gently rub it in, twice a day to the affected areas. They should leave the cream on for 30 minutes before hand and experienced no recrudescence in CIPN symptoms, that suggesting a long-lasting post-therapy effect. Second, only 1 patient discounted the topical treatment due to skin irritability and this patient had a previous skin disorder; so this treatment appears safety. And finally, reduced initial ChT doses in 11 patients, as well as, ChT discontinued in 5 patients were resumed after 1 month of topical amitriptyline application which provided efficient pain control (reduction of VAS pain score to a range of 0 to 3). Large and randomised studies are needed to confirm these 3 topics.

Neurofeedback
A pilot study examined whether electroencephalogram (EEG) neurofeedback (NFB) could alleviate CIPN symptoms [22]. This non-pharmacological treatment was a neuromodulatory intervention which can teach participants to interpret pain signals differently by affecting changes in brain regions that are active during pain conditions. In this randomized study, patients who had CIPN at least 3 months after completion of ChT were included. They were randomised in two groups: NFB group (30) or a wait-list control (WLC) group (32). The NFB group underwent 20 sessions of NFB, twice a week, over a maximum of 10 weeks. In each session, the participants watched and responded to their own EEGs while playing a game for 45 minutes per session; when they maintained their EEG waveform amplitude over a chosen threshold and inhibited less desirable waveforms, they were given rewarding feedback with a picture and a beep; the game paused when the participant did not match the thresholds preprogramed into the software, and no auditory or visual feedback was given. The primary endpoint was the change in worst pain related to CIPN, assessed by Brief Pain Inventory short form (BPI-SF) worst-pain item, from the baseline to the end of the treatment period (10 weeks

Limitations
Principal review-level limitations include the exclusion of possible relevant studies not published in English and the exclusion of retrospective works and clinical case series. Besides that, the authors admit possible selection bias, since only one literature database was used, and a search filter was used. Thus, it is assumed that there may be relevant tests that were not automatically included.
Other limitations are dependent of each trial and concern about study design, sample size and CIPN assessment. Most studies were unicentric and single-arm, with a little sample and medium quality. Direct comparative outcomes were prohibited by the variety of methods on participation selection, neuropathy assessment criteria and on efficacy assessment measure. Additionally, most of the studies used a clinician-based assessment. This type of evaluation can compromise the recognition of the true symptomatic burden of neurotoxicity and functional consequences. Consequently, it can also compromise the assessment of treatment effectiveness.

Conclusion
Unfortunately, in the last decade, there wasn't an improvement in the treatment of oxaliplatin-induced neuropathy. Until now, the duloxetine is the unique drug with moderate effects on the treatment based on a multicentric, randomised, double-blind, placebo-controlled crossover trial with significant sample size.
There is insufficient evidence to support a recommendation for any other treatment. Larger placebo-controlled trials to validate the effectiveness and safety of the other treatment strategies are warranted in patients treated with oxaliplatin.