Persistence to Biologic Therapy Among Patients With Ankylosing Spondylitis: An Observational Study Using the OPAL Dataset

Abstract

Objective To describe the treatment response and persistence to biologic DMARD (bDMARD) therapy in patients with ankylosing spondylitis (AS) in a real-world Australian cohort. Methods This was a retrospective, non-interventional cohort study that extracted data for patients with AS from the Optimising Patient outcomes in Australian rheumatology (OPAL) dataset for the period Aug-2006 to Sep-2019. Patients were classified as either bDMARD initiators if they commenced a bDMARD during the sampling window, or bDMARD naïve if they did not. Results were summarised descriptively. Treatment persistence was calculated using Kaplan-Meier methods. Differences in treatment persistence were explored using log-rank tests. Results 5048 patients with AS were identified. 2597 patients initiated bDMARDs and 2451 remained bDMARD naïve throughout the study window. Treatment with first, second and third line bDMARDs significantly reduced disease activity. Median persistence on first line bDMARDs was 96 months (95% CI 85 to 109), declining to 19 months (95% CI 16 to 22) in second line, and 14 months (95% CI 11 to 18) in third line therapy. Median persistence was longest for the golimumab treated group in all lines of therapy and shortest for the etanercept group. Differences in persistence rates according to the time-period that bDMARDs were prescribed (pre-and post-2012) were also seen for etanercept and adalimumab. Conclusion In this cohort all bDMARDs effectively reduced disease activity. Patients remained on their first bDMARD longer than subsequent agents. Median persistence was longest for the golimumab treated group in all lines of therapy and shortest for the etanercept group.

Dr. Griffiths et al reply

“…We acknowledged that the KM curves are less reliable at later time-points, in particular for the GOL-treated group (as is reflected in the numbers at risk in Figure 2). 2 As recommended by Pocock et al, 4 we have included the numbers at risk under the KM plots to convey the increasing unreliability of estimates as time from initiation of treatment increases.…”

“…Log-rank tests based on the entire time period (as recom-mended in Pocock at al 4 ) were used to investigate differ-ences between patients receiving different lines of treatment (P < 0.001) and for groups of patients receiving different treat-ments as first-line therapy (P < 0.001). 2 Given the baseline differences in the patients receiving different treatments, we included the results from a Cox proportional hazards analysis (prespecified in the analysis plan) in the article as this allowed adjustment to be made for important baseline confounders. This model again showed significant differences between groups of patients receiving different treatments.…”

“…In summary, we agree that the differing follow-up times causes issues with comparisons across the groups of patients treated with different biologic disease-modifying antirheumatic drugs, and we have drawn attention to the differences in data maturity in the discussion section. 2 The focus of the study was on what was observed in these groups of patients, and we recognize the limitations of formal comparisons for nonrandomized studies such as this.…”

Some Key Issues Relating to the Reporting and Interpretation of Time-to-Event Data

“…In conclusion, there are some important statistical issues relating to time-to-event analysis that have been overlooked by Griffiths et al 1 Addressing these issues will improve the quality of the paper and give readers more faith in the paper's conclusions. Some further questions relating to the analyses presented in Table 3 data have been communicated directly to the corresponding author.…”

“…Griffiths et al recently reported that in a cohort of Australian patients with ankylosing spondylitis included in the Optimising Patient outcomes in Australian RheumatoLogy (OPAL) dataset, the median persistence (persistence defined as the time to discontinuation of treatment) was longest for patients treated with golimumab (GOL) in all lines of therapy, and shortest for those treated with etanercept (ETN). 1 In drawing this conclusion, the authors have overlooked some statistical aspects relating to the reporting of time-to-event data that make it difficult to evaluate the robustness of their conclusions.…”