Protective Effects of Vitamin E on Chemotherapy-Induced Peripheral Neuropathy: A Meta-Analysis of Randomized Controlled Trials

Chemotherapy-induced peripheral neuropathy (CIPN) is one of the most common adverse effects of chemotherapeutic agents, which confines the effects of chemotherapy largely. The incidence of this disabling pain reaches up to 68% in the short term when receiving chemotherapy [1]. The severe pain of CIPN causes the loss of functional abilities and lower patients’ quality of life. The additional financial burden and the waste of medical resources caused by CIPN also cannot be ignored [2].

However, the neuropathogenesis of CIPN is still unclear. Multifactorial pathological processes may be involved, including oxidative stress, apoptosis, immune response, and neuroinflammation [3]. Accumulating studies have indicated that the production of reactive oxygen species is a significant cause of peripheral neuropathy induced by chemotherapy [4], and many researchers have proposed that an antioxidant substance could provide effective protection against peripheral neuropathy [5]. Currently, there is no effective therapy demonstrated effective in the prevention of CIPN [6].

Vitamin E is regarded as a treatment for neuropathy and is useful for the alleviation of chemotherapy-related toxicities [7]. It has been shown that the plasma levels of vitamin E in patients who developed severe CIPN following cisplatin were significantly reduced [8], and vitamin E supplementation ameliorated neurotoxicity in animals receiving cisplatin by eliminating oxygen-free radicals [7]. Pace et al. [9, 10] suggested that vitamin E lower the incidence and severity of cisplatin- and paclitaxel-induced peripheral neurotoxicity. Similarly, Argyriou et al. [11] conducted a randomized controlled trial (RCT) and reported that vitamin E had a protective effect on chemotherapy-induced peripheral nerve damage. A recent study also found that vitamin E helped paclitaxel-induced neuropathy and improved sensory nervous system function [12]. However, the negative findings about the effect of vitamin E on CIPN were reported in some clinical trials [13-15]. Thus, whether vitamin E can improve or prevent CIPN is still controversial. Our study was aimed to summarize the currently available evidence, and conduct a meta-analysis to further study the effect of vitamin E on the prevention of CIPN, providing the convincing medical advice for clinicians to prevent CIPN.

Two independent reviewers completed a systematic search in PubMed, EMBASE, Cochrane databases, CNKI, Wanfang, and CHKD from the inception dates to December 31, 2019. All disagreements were resolved by discussion between the reviewers. The search was restricted to human subjects with the following keywords: chemotherapy, neuropathy or peripheral neuropathy or peripheral nerve toxicity or neurotoxicity or CIPN, and vitamin E or tocopherol. There were no language limitations.

The titles and abstracts identified from the above databases were browsed independently by two reviewers. A consensus was reached by consultation. When no agreement was reached, a third reviewer was involved in the decision. The following studies were excluded (1) not RCTs (2) patients included in the trials without chemotherapy (3) trials without vitamin E supplement and placebo for comparison (4) study population with diabetes.

Two reviewers collected and organized the following data from all included trials independently. Items included were first author name, publication year, country, number of total participants, cancer type, vitamin E prevention regimen, chemotherapy regimen, number of participants in each group, and outcomes. The characteristics of the included articles were summarized.

The methodological quality of RCTs was assessed by two researchers (M.H.K. and L.R.Z.) independently to avoid the risk of overestimation of the intervention effects. The specific assessment methods were based on Cochrane risk-of-bias criteria: (1) generation of the allocation sequence, (2) allocation concealment, (3) blinding of participants and personnel, (4) blinding of outcome assessment, (5) incomplete outcome data, and (6) selective reporting. Each was graded as low-risk, high-risk, or unclear risk.

The main outcomes were the incidence of peripheral neuropathy (IPN) induced by chemotherapy, sural amplitude (SA), median amplitude (MA), and other evaluation parameters, including neurotoxicity score, the incidence of reflexes and distal paraesthesias (RDP). Review Manager (RevMan) version 5.3 was used for data processing. For dichotomous variables, risk ratios (RRs) with 95% confidence intervals (CIs) were calculated. For continuous variables, mean differences with 95% CIs were calculated. If there were no significant heterogeneities in the data of the different trials, the fixed-effects model was applied. Otherwise, we used the random-effects model [16]. Subgroup analyses by the dose of vitamin E (either 600 mg/day or 400 mg/day) and the type of chemotherapy regimen (either cisplatin or oxaliplatin) were performed to evaluate the efficacy of vitamin E in the prevention of CIPN. In addition, subgroup analysis of the number of chemotherapy cycles (either 3 or 6 cycles) was applied to assess the parameters of SA. The I² was calculated, which can be interpreted as the percentage of the variation between studies that is attributed to heterogeneity rather than chance. I² = 0% means no heterogeneity, I² > 50% indicates a high heterogeneity; and I² = 100% means that all variation derives from study heterogeneity. Sensitivity analyses were performed to exclude specific studies and to evaluate if any study could significantly influence the overall results. Publication bias was evaluated via Funnel plots made by RevMan5.3. The statistical significance was determined as p < 0.05.

A total of 514 references were identified in the initial search from the PubMed, EMBASE, Cochrane, CNKI, Wanfang, and CHKD databases. No references were gained from non-English databases, according to the excluded criteria. After getting rid of the duplicate references, 458 references were remained to be browsed. Due to the exclusion of studies with improper design, there were 40 references left to be assessed for full-text review. Finally, 8 references were considered for inclusion after excluding the other 32 references for other reasons. The entire search progress is described in Figure 1.

Among the final 8 references, 4 studies were conducted in Europe, 2 in Greece [11, 17] and 2 in Italy [9, 10] and for non-European studies, 2 studies were conducted in Iran [12, 15], 1 in the USA [13] and 1 in Brazil [14]. The major characteristics of the included studies are summarized in Table 1. In all, 488 patients were included in the meta-analysis, and the number of participants in each arm ranged from 13 to 96. The experimental intervention was vitamin E supplementation as an adjuvant to cisplatin, paclitaxel, and other chemotherapies. In 4 trials, 210 patients received vitamin E treatment at a dose of 400 mg/day [10, 12, 14, 15], and in 3 trials, 251 patients received 600 mg/day vitamin E [11, 13, 17]. Furthermore, only in 1 trial, 27 patients received 300 mg/day vitamin E in the experimental group [9]. The durations of the treatment with vitamin E varied from initiation of chemotherapy to 3 months after the cessation of chemotherapy.

Overall, 5 trials (62.5%) had adequate sequence generation, 4 trials (50%) had adequate concealed allocation, and the outcomes of 4 trials (50%) were judged to be adequately blinded. However, most of these studies (87.5%) did not have enough information about the risk of incomplete outcome data, and none of them mentioned selective reporting bias. The risk of bias is shown in online suppl. Table 1; for all online suppl. material, see www.karger.com/doi/10.1159/000515620.

The IPN was the primary outcome which directly reflected the occurrence of CIPN. Seven of the 8 studies were available to investigate the IPN, and the total number of participants was 206 and 212 for the vitamin E group and placebo group, respectively. The pooled results found a significantly lower IPN with vitamin E than placebo, while the high heterogeneity should not be ignored (RR 0.82; 95% CI 0.68–0.98; p = 0.03; I² = 88%) (Fig. 2a). Considering that different doses of vitamin E and various chemotherapy regimens were used in different studies, we conducted the parallel subgroup analysis with the fixed-effects model. The subgroup analysis results revealed that no significant difference in the IPN in patients treated with vitamin E 600 mg/day compared with placebo, with high heterogeneity (RR 0.81; 95% CI 0.57–1.15; p = 0.24; I² = 79%). Because of the apparent heterogeneity in the group receiving vitamin E 600 mg/day, we did a sensitivity analysis by excluding the study of Kottschade et al. [13] due to its low methodological quality. The sensitivity analysis suggested a significantly positive effect of vitamin E 600 mg/day on preventing CIPN as compared to placebo (RR 0.31; 95% CI 0.14–0.65; p = 0.002; I² = 0%) (Fig. 2b). However, the vitamin E 400 mg/day group displayed no difference in the reduction of the IPN compared with the placebo group (RR 0.93; 95% CI 0.79–1.10; p = 0.39; I² = 82%) (Fig. 2b). Besides, due to the study conducted by Pace et al. [9], the vitamin E 300 mg/day on the prevention of peripheral neuropathy might have significant clinical value.

Similarly, subgroup analysis was conducted by grouping the type of chemotherapy in the fixed-effects model (Fig. 2c). The combination of vitamin E and cisplatin group [9-11] displayed a significantly lower IPN with no heterogeneity than the combination of placebo and cisplatin group (RR 0.28; 95% CI 0.14–0.54; p = 0.0001; I² = 0%). However, for the combination of vitamin E and oxaliplatin group, the result showed that there was no difference in the IPN compared with the placebo group (RR 1.08; 95% CI 0.95–1.22; p = 0.24; I² = 34%).

SA is the action potential amplitude of the sural muscle, which is a reliable index for evaluating the condition of peripheral neuropathy. Five studies measured changes in SA before and after supplementation with vitamin E. We used the fixed-effects model to merge the mean difference values, and the pooled mean difference was 0.58 (95% CI −0.24 to 1.40; p = 0.17) with high heterogeneity (I² = 72%) (Fig. 3a), which indicated that vitamin E did not improve SA compared to placebo. Thus, a subgroup analysis was conducted based on the number of chemotherapy cycles. Further analysis indicated that vitamin E could reduce the extent of SA in patients after 3 rounds of chemotherapy compared with the placebo group, but high heterogeneity still existed (mean difference = 1.06; 95% CI 0.14–1.98; p = 0.02; I² = 82%) (Fig. 3b). Subsequently, we performed a sensitivity analysis after ruling out a study conducted by Shamsaei et al. [12], which included Asian instead of European in the study cohort. The sensitivity analysis demonstrated a significantly protective effect on SA with vitamin E after three rounds of chemotherapy compared with placebo (mean difference = −2.66; 95% CI −5.09 to −0.24; p = 0.03; I² = 0%). Nonetheless, significant improvement in SA was not detect with vitamin E supplementation after 6 rounds of chemotherapy (mean difference = −1.28; 95% CI −3.11 to 0.54; p = 0.17; I² = 40%) (Fig. 3b).

Similar to SA, MA is another indicator that reflects the severity of peripheral neuropathy. Two previous studies [9, 10] reported MA were available to measure the change in MA before and after the application of vitamin E. Compared to the participants treated with placebo, vitamin E was more effective in remission for MA with a pooled mean difference of −3.00 (95% CI −5.32 to −0.68; p = 0.01) with low heterogeneity (I² = 0%) (Fig. 3c).

The secondary outcomes in the meta-analysis were the neurotoxicity score, sural sensory conduction velocity (SSCV), and the incidence of RDP. A composite neurotoxicity score was calculated for each patient in three included studies [9, 10, 15], according to the severity of neuropathic symptoms and electrophysiologic changes [18]. Analysis showed that vitamin E significantly decreased the neurotoxicity score and displayed moderate heterogeneity (mean difference = −1.47; 95% CI −2.45 to −0.49; p = 0.003; I² = 67%) (Fig. 4a). Then, we carried out a sensitivity analysis by excluding the study conducted by Salehi et al. [15], as it had a different chemotherapy regimen. The results found a significantly improved effect on the neurotoxicity score with vitamin E compared to placebo (mean difference = −2.65; 95% CI −4.01 to −1.29; p = 0.0001; I² = 0%) (Fig. 4a). The SSCV is the action potential conduction velocity of the sural muscle in electromyography. The pooled results presented that there was no significant difference in the vitamin E group compared with the placebo group (mean difference = −0.39; 95% CI −1.82 to 1.03; p = 0.59; I² = 52%) (Fig. 4b). RDP reflected the physiological status of limb muscles. The pooled result revealed that vitamin E played a significant role in the lower incidence of RDP than placebo (mean difference = 0.50; 95% CI 0.29–0.87; p = 0.01; I² = 8%) (Fig. 4c).

The Egger’s plot based on the IPN is shown in Figure 5. The result showed that no obvious reporting bias was observed (p = 0.408).

Results of this meta-analysis showed that vitamin E supplementation can confer modest improvement in the prevention of CIPN. Patients who received vitamin E supplementation of 600 mg/day had a significantly lower incidence of CIPN than the placebo group. Vitamin E played a key role in reducing the IPN in the cisplatin chemotherapy group.

Peripheral neuropathy is one of the most debilitating toxicities in cancer patients receiving chemotherapy [19]. The incidence of CIPN varies with different antineoplastic agents, namely 28–100% of patients who receive cisplatin [20] and 57–83% of patients who receive paclitaxel [21]. In addition to chemotherapy regimens, the incidence of CIPN was often related to the therapeutic dose [22]. Severe acute CIPN required chemotherapy dose reduction or cessation, leading to the progression of cancer and eventually a poor prognosis in patients [23].

Many pharmacological agents, such as glutathione [24], calcium, and magnesium infusions [25], and venlafaxine [26] have been studied for the treatment of CIPN. The effect of vitamin E on the improvement of CIPN is still controversial. The results of the animal study proved the positive effect of vitamin E, which urged more researchers to seek its validity in clinical trials [5, 27]. A previous review made by Bove et al. [28] concluded that the severity of peripheral neuropathy varies inversely with the level of vitamin E. The vitamin E deficiency caused by chemotherapy made the nervous system more vulnerable to damage. A review, published in 2011, included 2 RCTs that contained vitamin E supplementation and proved the preventative effect of vitamin E against CIPN [22]. However, only 2 RCT studies seem not reaching a convincing conclusion. In another published meta-analysis, which included 5 available RCTs, the positive neuroprotective effects of vitamin E were confirmed. In addition, the study concluded that vitamin E supplementation could attenuate the development of CIPN [29]. Whereas, Huang’s meta-analysis included 6 RCTs and failed to demonstrate a statistically significant effect of vitamin E on the incidence of CIPN [30].

In our study, we collected all the newest published RCTs and tried to make a higher evidence-based quality analysis. Our primary outcomes included the IPN, which directly reflected the occurrence of CIPN in cancer patients after chemotherapy [31]. According to our analysis, vitamin E supplementation provided a significant reduction in the IPN in the cisplatin group, which was consistent with the meta-analysis by Eum et al. [29]. They found a significant preventive effect of vitamin E on CIPN with a probability of benefit greater than 95%. In the meta-analysis worked by Huang et al. [30] in 2016, 353 patients of 6 articles were involved in the study. They failed to report a statistically significant effect of vitamin E supplementation on the incidence of CIPN. The possible reason which accounted for the result was explored. We realized that Huang et al. [30] have searched the database from their inception to December 31, 2013. However, 2 more newly developed RCTs (Salehi et al. [15] and Shamsaei et al. [12]) were involved in the present study. Salehi et al. [15] reported that there was no difference in the mean peripheral neuropathy score changes between vitamin E and control groups. Nevertheless, Shamsaei et al. [12] reported that the delta amplitude of sural nerve was significantly lower among patients taking vitamin E supplements. Due to the various inclusions of studies, the discrepancy between 2 meta-analyses appears. Furthermore, the results that preventive effect of vitamin E supplementation on cisplatin was associated with neurotoxicity in Huang’s report were consistent with our findings, which showed that the beneficial effect of vitamin E on the prevention of cisplatin-associated neurotoxicity.

The exact mechanisms of neuropathy associated with various chemotherapy drugs remains to be fully elucidated. Researchers tended to assume that cisplatin-induced neuropathy was dose-related, and the symptom manifested after an accumulative cisplatin dose [22]. Bove et al. [28] reported that the plasma concentrations of vitamin E fluctuated over the process of cisplatin treatment and returned to a favourable level, which suggested that vitamin E exhaustion after cisplatin treatment had a potential association with the occurrence of peripheral neuropathy. However, our results demonstrated that vitamin E did not play a role in reducing the IPN in the paclitaxel group. Although paclitaxel has been confirmed to cause dose-dependent distal axonal sensorimotor polyneuropathy, both of the RCTs [12, 17] in our meta-analysis revealed that no remarkable difference was found between the vitamin E and placebo groups after treatment with paclitaxel. More high-quality RCTs are needed to investigate the influence of vitamin E on the peripheral neuropathy induced by paclitaxel.

Furthermore, in our study, only 600 mg/day vitamin E for 3 months presented significantly protective effects on the prevention of CIPN. It is worth noting that the dose and regimen of vitamin E were distinguished from existing studies. A study showed that an overdose of vitamin E could be harmful and increase the risk of overall mortality [32, 33]. However, other studies have argued that vitamin E supplementation did not have an effect on all-cause mortality, regardless of the dosage [34]. No adverse effects on vitamin E supplementation were reported by any of the reviewed studies. The use of vitamin E 600 mg/day for 3 months did not cause any adverse effects in our study. Generally, the regimen of vitamin E 600 mg/day for 3 months might be preferred for the prevention of CIPN.

The evaluation methodology of CIPN not only involves symptoms of pain and paralysis but also changes in electromyography [35]. In the meta-analysis, other outcomes measured by electromyography were also analysed. We found that vitamin E provided improvement of SA for patients who finished 3 cycles of chemotherapy, while there was no significant effect on SA for patients who finished the 6 entire cycles of chemotherapy. The original target of medical treatment is to cure cancer or delay progression of the disease, but current attention is focused more on the long-term influence for patients. Our study showed that vitamin E supplementation was not enough to improve the SA of patients who finished the 6 entire cycles of chemotherapy. To thoroughly evaluate the effect of vitamin E on the improvement of CIPN, we also conducted a meta-analysis to assess other indicators reflecting the severity of peripheral neuropathy, including MA, neurotoxicity score, SSCV, reflexes, and distal paraesthesias. Vitamin E played a key role in the improvement of MA, neurotoxicity score, RDP, and all evidence proved the efficacy of vitamin E in the prevention of CIPN was reliable. Furthermore, the nutrition status could also influence the side effect symptom in chemotherapy patients. Malnutrition is a prognostic factor for poor clinical outcome and severe side effect during chemoradiotherapy [36]. Many indicators could be used to assess the nutrition status in cancer patients. The Patient-Generated Subjective Global Assessment Short Form (PG-SGA SF), NRS2002 scores, and PG-SGA score could be used for descriptive nutrition screening [37] and are useful prognostic factors for patients after chemoradiotherapy [38, 39]. However, none of the 8 studies included in our analysis involved the nutrition status as an outcome indicator, which might be a meaningful covariate in the analysis. The nutrition status should be paid more attention in the future RCT studies to provide more meaningful message for the physicians.

There are some limitations in our analysis should not be ignored. Firstly, only 8 RCTs were included in the analysis. The methodological quality of the RCTs involved was not satisfactory, and only 2 RCTs were of high methodological quality. Secondly, varied cancer populations across the studies influenced the consistency of the research. In addition, clinical outcomes, such as the disease response, adverse effects, and conditions of prognosis, were not reported clearly, impending further analyses due to the insufficient data.

Vitamin E might contribute to prevent CIPN from patients underwent neurotoxic chemotherapy. The vitamin E supplement as the standard of care for prevention of CIPN is moderately recommended in our study. More high-quality trials with standardized reporting of clinical outcomes about peripheral neuropathy are needed to explore the exact role of vitamin E in the prevention of CIPN.

The authors have no ethical conflicts to disclose.

No authors report any conflict of interest.

This work was supported by the National Natural Science Foundation of China (81871986).

Huikai Miao and Rongzhen Li conceived and designed the study. Dongni Chen, Jia Hu, and Youfang Chen performed the analysis, prepared the figures and tables, and wrote the main manuscript. Chunmei Xu and Zhesheng Wen were involved in critically revising the manuscript. All the authors read and approved the final manuscript.