Lost in Translation: Failure of Preclinical Studies to Accurately Predict the Effect of Regional Analgesia on Cancer Recurrence

The major goal of translational research is to evaluate the efficacy and effectiveness of treatments and interventions that have emerged from exhaustive preclinical evidence. In 2007, a major clinical trial was started to investigate the impact of paravertebral analgesia on breast cancer recurrence. The trial was based on preclinical evidence demonstrating that spinal anesthesia suppressed metastatic dissemination by inhibiting surgical stress, boosting the immunological response, avoiding volatile anesthetics, and reducing opioid use. However, that trial and three more recent randomized trials with a total of 4,770 patients demonstrate that regional analgesia does not improve survival outcomes after breast, lung, and abdominal cancers. An obvious question is why there was an almost complete disconnect between the copious preclinical investigations suggesting benefit and robust clinical trials showing no benefit? The answer is complex but may result from preclinical research being mechanistically driven and based on reductionist models. Both basic scientists and clinical investigators underestimated the limitations of various preclinical models, leading to the apparently incorrect hypothesis that regional anesthesia reduces cancer recurrence. This article reviews factors that contributed to the discordance between the laboratory science, suggesting that regional analgesia might reduce cancer recurrence and clinical trials showing that it does not-and what can be learned from the disconnect. Statistical significance provides little guidance to proper interpretation of results. It is generally best to largely ignore P values and instead focus on the treatment effect and 95% CIs in the context of clinically meaningful differences.