This study was a multicenter, randomized, double-blind, and placebo-controlled phase II clinical trial. It was approved by the Chinese State Food and Drug Administration (approval no. 2007L03540). The clinical trial registration number is NCT01526564. The clinical study was carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments involving humans. In addition, informed consent was obtained from all participants involved in this study.

Eligible patients were aged 18–75 years without gender limitation. Eligibility criteria included: Grade ≥3 neuropathy, as determined by NCI-CTC criteria version 3.0 (9), while receiving paclitaxel, cisplatin or vinblastine treatment, and/or grade ≥2 neuropathy persisting for at least one month after the discontinuation of either drug, and neurotoxicity for <6 months; at least one abnormality on electrophysiological examination; Karnofsky physical score (KPS) of ≥60; absolute neutrophil count of ≥1.5×10⁹/l, hemoglobin count of ≥80 g/l, platelet count of ≥75×10⁹/l, total bilirubin counts of 1.5-fold less than normal value, glutamic-pyruvic transaminase (GPT/ALT) and glutamic-oxalacetic transaminease (GOT/AST) no more than 2.5-fold greater than the normal value; normal blood urea nitrogen, serum creatinine and electrocardiogram (ECG) findings. During the study, the use of steroids, analgesic or neuroprotectant drugs was not permitted. Patients were enrolled after providing written informed consent.

Exclusion criteria included: Neuropathy caused by other antineoplastic treatment except paclitaxel, cisplatin or vinblastine; pre-existing diabetes mellitus and/or neuropathy caused by vitamin deficiency, infection, trauma, poisoning, oppression, ischemia, metabolic disorders; genetic neuropathy and/or peripheral sensory nerve dysfunction due to central nervous system lesions; use of other drug therapy for neuropathy in the last 30 days (such as nerve growth factor, amifostine reduced glutathione, vitamin E or B, glucocorticoids, ethosuximide, carbamazepine, gabapentin, sodium thiosulfate, glutamic acid, lamotrigine, α-fatty acid, lithium salt, lithium salt or magnesium salt); participation in other clinical trials in the past 30 days; out of control clinical problems (such as serious mental, nerve, cardiovascular and/or respiratory system disease); pregnant or lactating women; and poor compliance.

During the trial, patients were withdrawn if serious adverse events occurred, and/or the patient became pregnant.

A total of 240 patients met the criteria and were recruited for the present study at Shanghai Changzheng Hospital (Shanghai, China) between September 4, 2010 and November 7, 2013. All patients were treated with at least one type of taxoid, either satraplatin (80 mg/m²; GPC Biotech, Munich, Germany) or vincristine (0.05 mg/kg²; Pharmachemic Hisun, Zhejiang, China). Their peripheral sensory neuropathy grading following chemotherapy was ≥ grade 3 or 2, lasted for ≥ 1 month, and the course of neuropathy was ≤6 months. The patient either no longer required chemotherapy or did not choose to undergo chemotherapy. Each patient was administered one ALC hydrochloride enteric-coated tablet (oral administration; 500 mg/tablet, 2 tablets/time) three times a day for eight consecutive weeks. The patients of the control group received a placebo (lactose, 500 mg/tablet, 2 tablets/time) in the same manner. Patients were instructed that the interval between doses should be at ≥4 h and were monitored at week 4, 8 and 12. Since drug treatment ceased after week 8, the results obtained after week 12 represented the long-term effects of ALC intervention.

Safety analysis set (SS) was an analysis in those who received at least one dose of the investigational drug and safety valuation. Full analysis set (FAS) was analysis in which patients are included in the group to which they were randomized irrespective of compliance, administrative errors (such as error in eligibility), or other protocol deviations. Per protocol set (PPS) was an analysis in which patients are included in the group corresponding to the treatment they actually received. Patient compliance and “switchovers” were considered in the analysis.

The primary endpoint of the study was to demonstrate an improvement of ≥1 grade in neurotoxicity, as determined by NCI-CTC criteria (version 3.0) at week 8 of ALC administration (9).

Prior to ALC administration at 4, 8 and 12 weeks, electrophysiological examinations were carried out at each time point, which included examination of the nerves (handedness, median nerve, ulnar nerve, common peroneal nerve, tibial nerve, superficial peroneal nerve and sural nerve). Three neural electricity experts compared the electrophysiology examination results at the 8th week (including nerve conductive velocity, latency period and amplitude) to the results obtained prior to ALC administration. The unified curative effect standard to centralized assess was followed and the evaluation results were recorded and signed by the experts. The results were categorized into improved, effective improved and no change, based on the efficient rate which was calculated by the following formula: Efficient rate = (Number of patients that improved + Number of effective improved patients) / (All patients - Number of patients that could not be evaluated) × 100%.

The case histories of the patients, previous chemotherapy information, and disease and treatment histories were recorded during screening. Prior to 4, 8, and 12 weeks of testing, all patients underwent baseline assessment, which included comprehensive physical examinations, laboratory examinations, and electrophysiological and 12-lead ECG examination. The physical examinations included the measurement of vital signs (temperature, pulse, respiration and blood pressure), the grading of neurotoxicity (9), cancer-associated fatigue classification (10) and KPS assessment (11). Laboratory examination included a routine blood panel, liver and kidney function (12) and fasting glucose measurement (13). Electrophysiology examination was performed as described and the results were categorized into improved, effective improved and no change. During the study, all disease factors, drug combinations and adverse events were recorded in the case report form (CRF).

NCV was examined to evaluate the changes prior to ALC administration and after 8 weeks of ALC administration. Treatment was defined as effective if the NCV was improved after ALC administration, whereas, it was defined as ineffective if the NCV was reduced or unchanged. Effective analysis was conducted according to the results using the following formula: Efficient rate = number of patients effectively treated/(number of patients that could not be evaluated) ×100%.

Statistical analysis was performed using SAS software version 9.2 (SAS Institute, Cary, North Carolina, USA). Independent sample t-tests or Wilcoxon tests were used to compare continuous variables between groups, and paired t-test or Wilcoxon tests were used for comparison within groups. For the categorical variable analysis, a χ² test or exact test were used for comparisons between groups. The Cochran-Mantel-Haenszel-χ² test was used to analyze ordinal categorical data. Data were presented as the mean±standard deviation. P<0.05 was considered to indicate a significantly significant result.

A total of 240 patients were originally eligible for this study, although the final full study group included 239 CRF subjects. The safety analysis set (SS) contained 236 patients, 118 in the experimental group and 118 in the control group. The full analysis set (FAS) contained 225 patients, 109 in the experimental group and 116 in the control group. The per-protocol set (PPS) contained 203 patients, 95 in the experimental group and 108 in the control group.

As shown in Table I, in the FAS, at the 8th week of the study, ALC treatment reduced neurotoxicity in 50.5% of the experimental patients, compared with a 24.1% reduction in the control group. The difference between the ALC and placebo groups was statistically significant [95% confidence interval (CI), 14.1–38.5%; P<0.001]. The 8-week efficacy in the PPS was 51.6% (ALC) and 23.1% (control), which was significantly different (95% CI, 15.6–41.2%; P<0.001). The statistical significance of these findings was maintained following correction for the effect of different hospital centers (P<0.001; data not shown). The reduction of neurotoxicity was time-dependent for the patients in both treatment groups. A comparison of the effect of ALC between the 4th, 8th, and 12th week revealed that the improvement in neurotoxicity was significantly different between the experimental and control group (P<0.05; Table I). As shown in Table II, there was a observable improvement on the 4th week, with a decrease in the number of patients displaying grade ≥3 neuropathy, and an increase in the number of patients displaying grade 2, although the differences between the two groups did not reach statistical significance (P>0.05). By the 8th week of treatment, the difference between the experimental and control group was statistical significant (P<0.05).

To further examine whether the efficacy of ALC was affected by chemotherapy treatment and cancer type, the cancer type in the treatment and placebo groups were classified prior to comparison of ALC efficacy. In the FAS, ALC was able to significantly improve the peripheral neuropathy grading in all cancer types, results which were not observed in the placebo group (P<0.05; Fig. 1). ALC's effect was not affected by treatment with platinum-containing chemotherapy (Fig. 1). In addition, chemotherapy had no difference on the effect of ALC, compared with the (Fig. 2).