Background Adverse events of continuous subcutaneous infusion of lidocaine in people with neuropathic cancer pain have rarely been collected in a systematic way despite its use in clinical practice. The LiCPAIN trial aimed to determine the feasibility of conducting a double-blind randomized controlled trial of continuous subcutaneous lidocaine for neuropathic pain in people with advanced cancer. This analysis aimed to describe and compare treatment emergent adverse events (TEAEs) in the lidocaine and placebo arms.

Methods Adverse events (AE) were collected twice a day using a structured checklist based on the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 4, daily ECG and vital signs were reviewed. Severity, causality and the action taken with respect to study treatment were determined by the investigator. The proportion of people experiencing each TEAE were compared between lidocaine and placebo groups using Chi-squared tests.

Results Of the 17 participants who completed the trial, 11 (65%) were allocated to lidocaine infusion and 6 (35%) to placebo. There were 63 events experienced during the trial, 43 in the lidocaine group and 20 in the placebo group. Overall, 10 (91%) people in the lidocaine group and four (67%) in the placebo group experienced one or more TEAE (p=0.21), all of which included ‘possibly, probably or definitely related’ AE events.

The most common TEAE for the lidocaine group were tremor (n=5), fatigue (n=3), somnolence (n=3), dizziness (n=3), ataxia (n=3) and neoplasms (n=3). Only one of these events (neoplasm) was reported in the control group. AE that were classified as neoplasms represented disease progression in an advanced cancer population. Grade 3 or worse TEAE occurred in four participants in the lidocaine group: during the intervention (fatigue), during follow-up (depressed level of consciousness, neoplasms, terminal agitation) and post-study (neoplasms). They occurred in two participants in the placebo group: during the intervention (abdominal pain, hypertension) and follow-up (neoplasms, hypertension, biliary sepsis), p=0.90. There were four serious AE of neoplasm leading to death: in the lidocaine arm one occurred during follow up and two were an incidental finding post-study; in the placebo arm one death occurred during follow up (p=0.62). One of these deaths was determined to have a ‘possible’ causal relation to the lidocaine infusion with the others. No AE were detected on ECG.

Discussion Structured assessment found no significant difference between lidocaine and placebo groups regarding the number of participants experiencing one or more TEAE, related events, grade 3 or worse or serious AE. This study was not powered to compare AE between groups, therefore all p-values are considered nominal and caution is needed when interpreting this data to inform clinical practice. AE were common but mostly graded low severity in both intervention and placebo groups. Clinicians should monitor carefully for neurological symptoms as part of a structured assessment. A number of AE including those graded serious occurred in the follow up or post-study phase, which may be expected in this population. Future studies may further characterise the safety profile of lidocaine infusions in people with advanced cancer.