A SMART design study using patient-based algorithms for multicomponent therapy in chronic low back pain and beyond

University Of Michigan At Ann Arbor, Ann Arbor MI

Investigators

Abstract

PROJECT SUMMARY / ABSTRACT A wide array of evidence-based approaches is commonly used for the treatment of chronic low back pain (cLBP). Yet, most studies show that only about one third of patients with cLBP, or really any chronic pain condition, benefits appreciably from any particular treatment. Given the largely underwhelming effects of current therapies, chronic pain remains a serious public health problem and there must be a cultural transformation in how pain is understood, assessed and treated. One possible explanation for the small effect sizes seen with most of the current treatments for cLBP is that patients are not being adequately matched to appropriate interventions. Individual factors such as demographics; social determinants of health (SDOH); cognitive, affective and social factors; and especially, pain mechanisms (e.g., inflammatory, neuropathic, nociplastic) all impact the experience of pain, level of disability, and response to treatment. Thus, the critical challenge our team has been addressing for over a decade is: What treatment is best for a particular person? The need for a personalized medicine approach is at the heart of the research underway at the University of Michigan (UM) Back Pain Consortium (BACPAC) Mechanistic Research Center (MRC). To address this critical clinical challenge, our team performed an “interventional response phenotyping study” consisting of a sequential, multiple assignment, randomized trial (SMART) to evaluate treatments for cLBP from four key domains: pharmacological, physical therapy and exercise, cognitive-behavioral therapies (CBT), and mHealth pain self-management. Our preliminary data show that while about 25-30% of participants in our SMART mechanistic study respond to an assigned treatment, there are powerful predictors of treatment response that are not always intuitive. For INTERACT, we will expand on and enhance our ongoing longitudinal data collection and analysis for our unique mechanistic cohort where richly phenotyped participants undergo two treatments. First, we will enhance the data from the current UM BACPAC SMART participants by recontacting them and then collecting a more robust measurement of SDOH and early life experiences including assessment of childhood pain and trauma, as well as collecting additional follow-up data to assess long-term treatment trajectories (Aim 1). Next, we will leverage the clinical, biomechanical, MRI, performance testing, and survey data from the UM BACPAC SMART study to develop individualized phenotyping treatment rules (Aim 2) that will better match patients to treatments. We will then recruit and assess another 500 participants with cLBP and conduct a phenotype-informed SMART design study to evaluate the added benefit of matching participants to treatments using the individualized phenotyping treatment rules (Aim 3). Lastly, as the top preforming site in the BACPAC collaborative trial known as BEST, the UM INTERACT team will be enthusiastic participants in an INTERACT collaborative trial and offer ideas, infrastructure, and leadership to the network.