Methods:

This was a prospective, single-centre, open-label, phase 3, randomized controlled trial done in a tertiary care referral centre, in India. The study was planned with an open-label design because there were practical challenges in implementing multiple-level masking, especially, as the acupuncture treatment is explicitly different than the usual care medications and difficult to introduce a placebo for acupuncture needles. This study did not involve the administration of any additional drugs beyond those being used for the patient’s standard medical care. This study was conducted in accordance with GCP as defined by the International Conference on Harmonization. The study was independently monitored by the data and safety monitoring board. Data collection and analyses were conducted in a single institution. Data quality was ensured by regular reviews and audits. Ethics approval was obtained from the Tata Memorial Centre, Mumbai, India.

Statistical methods and analysis:

The primary endpoint is to compare the change in sensory symptom QoL score from baseline to 8 weeks post intervention between the acupuncture and control group. According to Lu et.al the average raw score for sensory scale as assessed using PNQ at baseline for the usual care arm was 2.5 with a SD of 0.9. For the present study the sensory symptom score of CIPN20 was considered equivalent to PNQ sensory score. Hence assuming a baseline raw score of 2.5 for sensory scale the corresponding normalized score (according to EORTC guidelines) was assumed to be 45 at baseline for the control group of the present study. Lu et al. reported a decrease of 0.3 with a SD of 0.6 points at 8 weeks from baseline in the control group. The corresponding decline in the control group was assumed to be 38.7 at 8 weeks. A reduction of 12 points was considered clinically relevant in the acupuncture arm. Simulations were performed by assuming this difference and the respective simulated means in the acupuncture and SoC arm at 8 weeks post intervention was assumed as (27.04+ 23.43) and (38.7 + 23.4). Using ANCOVA procedure, assuming a power of 80% and significance level of 5%, drop-out rate of 10% and a moderate correlation between baseline and post intervention sensory QoL levels of 0.5, a total of 74 patients (47 patients per group) are needed. 

All raw scores were transformed to a 0 to 100 scale according to the algorithms described in the respective scoring manuals. Continuous data were represented as mean (Standard deviation) or Median [inter quartile range (IQR)] and categorical data were reported in counts (percentage) respectively. The Shapiro-Wilk’s test was used to check the normality of each variable.  An analysis of covariance (ANCOVA) model was used to assess treatment differences in the pain scores after adjusting for age and baseline scores. The model estimated the mean change in scores and was reported using 95% CI. Secondary endpoints were analyzed using Linear mixed effects model to assess the overall change in the scores between the groups adjusting for age, with time as a within-subject factor and group as a between-subject factor with respect to domains of QLQ-CIPN20, EORTC QLQ, Total pain detect scores from baseline to 6 months post-therapy. Logistic regression analysis was used to assess the impact of intervention on exploratory outcomes like incidence of Grade 2 or worse CIPN, presence of musculoskeletal symptoms, hot flashes, and mood swings at 8 weeks, 12 weeks, 18 weeks, and 6 months of treatment adjusting for the grades at baseline. Odds ratio was reported along with 95% CI. All analyses were two-sided, and the level of significance was set at 0.05. All analyses were done in the modified intention-to-treat (mITT) population. Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 25.0. Armonk, NY: IBM Corp and RStudio Version 2023.03.0.