The design and rationale of the Biomarkers for Evaluating Spine Treatments trial: a sequential multiple assignment randomized trial

Abstract

Objective: Chronic low back pain (cLBP) is a common condition that impacts quality of life and function. There are many evidence-based treatments to address cLBP; however, treatment effects are modest, perhaps in part due to individual variation in treatment response. The Biomarkers for Evaluating Spine Treatments (BEST) trial was designed as the collaborative centerpiece of the Back Pain Consortium (BACPAC) research program. This consortium was sponsored by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) as part of the Helping to End Addiction Long-term (HEAL) Initiative.

Design: The BEST trial was a sequential multiple assignment randomized trial (SMART) designed with the primary goal of identifying in whom different treatments show optimal response. The primary focus of the study was to use patient features, including biomarkers and phenotypic measures, to identify subsets of persons with cLBP who respond best to specific common treatments.

Methods: Four interventions were chosen for the trial: Enhanced Self-Care, Acceptance and Commitment Therapy, Duloxetine, and Evidence-Based Exercise and Manual Therapy. Following a run-in period and baseline assessment, participants were randomized to 1 of the 4 treatments for the first 12-week intervention period. Participants were reassessed and based on their self-reported response to initial treatment, continued that initial treatment, were augmented with an additional randomly assigned treatment, or were switched to a new treatment.

Conclusion: This trial was designed to deliver rich phenotypic data that will both potentially aid in the discovery of phenotypic characteristics that predict treatment response and provide a greater mechanistic understanding of cLBP.

Clinical trial registration number: The Biomarkers for Evaluating Spine Treatments (BEST) trial is registered on ClinicalTrials.gov (Registration number: NCT05396014; https://clinicaltrials.gov/study/NCT05396014).

Keywords: SMART; chronic low back pain; chronic pain; clinical trial; low back pain; phenotyping.

Figure 1.

Study design. The study design schematic describes the clinical protocol of the Biomarkers for Evaluating Spine Treatments (BEST) trial. Participants enrolled in the study and participated in informed consent. Participants progressed through a 2-week run-in period followed by a baseline visit (Visit 0). At this visit, phenotyping measures were collected, and participants were randomized to 1 of 4 treatments for a 12-week period. Participants were reassessed and phenotyping measures were collected at the end of the 12-week treatment period. Participants were then assigned a second treatment based on the response to the first intervention. Participants were continued on their period 1 intervention, augmented with an additional randomly assigned intervention, or randomized to a different intervention. This decision tree is described in Figure 2. Participants were then reassessed, and phenotyping measures collected at the end of the second 12-week intervention period. Participants were then followed through 36 weeks. (PGIC: Patient global impression of change, PEG: Pain, enjoyment of life and general activity scale, Tx: Treatment, ESC: Enhanced self-care, ACT: Acceptance and commitment therapy, EBEM: Evidence-Based Exercise and Manual Therapy, R: Randomization).

Figure 2.

Second-stage randomization scheme. Participants were assigned to second-stage intervention activities based on the response to the Patient Global Impression of Change (PGIC) after Treatment 1. If the participants noted an improvement on the PGIC (rating improvement as a 1 or 2), they were either assigned to maintain the first stage intervention or be randomized to treatment augmentation based on the 12-week Pain, Enjoyment of Life and General Activity (PEG) scale. Individuals who minimally improved or demonstrated no change in the first stage intervention (3-4 on the PGIC) were randomized to either an additional randomly assigned treatment to augment the first stage intervention or switched to a new randomly assigned intervention. Finally, participants who worsened (5-7 on the PGIC) were randomized to 1 of the 3 other interventions for the second-stage treatment period. This was intended to emulate clinical practice and reasoning. Participants initially randomized to Enhanced Self-Care (ESC) with 12-week PGIC scores of 3-7 were randomized to augmentation, not switching. This is because the intervention was not able to be stopped given the self-care activities could not be unlearned.

Figure 3.

Map of the BEST trial enrolling sites. Map displaying the clinical centers who enrolled participants in the Biomarkers for Evaluating Spine Treatments (BEST) trial. The sites were distributed across the United States and were selected based on their ability to recruit participants with chronic low back pain into the trial and demonstrated capacity to deliver interventions and perform phenotyping activities.