Study: Patient Registries Inadequate to Evaluate Cost-Effectiveness of Biologics for Psoriatic Arthritis

Cost-effectiveness evidence has come to play a prominent role in decisions regarding the approval of expensive biologic treatments, such as those for psoriatic arthritis (PA). However, cost-effectiveness evidence for PA treatments has been difficult to develop because of the lack of a robust evidence base. Data from patient registries are potentially useful for fulfilling the evidence requirements of cost-effectiveness analyses (CEA) evaluating biologic therapies for PA when clinical trial data are lacking or insufficient. However, no individual patient registries were found to meet all the evidence requirements on their own, a recent review concludes. Thomas Patton, PhD, and colleagues published their findings in Clinical Rheumatology.

The report’s authors note that the short-term nature of many phase 3 trials in this area means that assumptions must be made regarding the long-term efficacy of biologics in order to investigate the cost-effectiveness of biologic therapies over the remaining lifetime of an average patient. Data from switching studies, in particular, are considered very important in treatment and marketing-approval decisions, yet there is a dearth of such clinical trial data. “Consequently, this can impose additional uncertainty surrounding the results and, ultimately, reduce confidence in a decision to accept or reject a treatment for reimbursement,” the authors note. Thus, the use of patient registries (such as the British Society for Rheumatology Biologics Registry as well DANBIO, a registry in Denmark) that capture information on PA patients receiving a variety of treatments can be useful in data-gathering efforts.

The review identified relevant literature pertaining to PA registry data and used it to establish a list of previous or ongoing patient registries worldwide. The registries were analyzed to assess whether they used measures of PA disease activity that are relevant from a clinical and economic perspective, and then were appraised according to a set of criteria that are based on methods currently used to model PA in the CEA literature. The findings were then used to inform recommendations regarding the use and collection of data in PA patient registries for the purposes of CEA, and to identify how future data collection in PA registries might be better tailored to inform CEA research.

The researchers found 21 potentially relevant PA registries in 18 different countries, and used 9 critical appraisal questions to guide their assessment. They found substantial variation regarding the extent to which the registries, as a whole, were useful for CEA studies, and identified notable disparities in terms of the accessibility of the registries to researchers. All of the registries were useful, to some degree, for the purposes of informing CEA studies in PA, but no individual registry met all requirements when considering how the disease has been modeled previously.

The majority of registries collected data on time-to-withdrawal from treatment in patients, as well as the reasons for withdrawal from treatment. Most collected at least 1 relevant outcome measure for assessing treatment effects, and most followed patients over a long time horizon, suggesting that they would be useful for estimating long-term disease progression. However, the degree of reporting on heathcare resources was poor (such reports were found in only 3 of 21 registries) and only 6 registries were potentially useful for modeling the relationship between disease severity and mortality risk. Only 2 registries were capable of estimating the effect of treatment withdrawal on disease progression.

Another problem that the study identified was accessibility of the registries, limiting the pool of registry evidence available for CEA research. In addition, the generalizability of the data across setting and jurisdictions may further constrain the usefulness of the available data depending on the specific decision problem being investigated.

The researchers recommend that future data collection in PA patient registries take place 3 months after the initiation of treatment, and that registries collect outcome measurements when patients withdraw from treatment, so that researchers can obtain empirically derived estimates of disease progression rather than relying on assumptions. To increase the robustness of registry data, the authors also recommend the collection of optimal outcomes for estimating initial response to treatment. “It is hoped that the recommendations provided will inform future data collection within a registry setting,” the researchers conclude, in order to provide the research community with a greater understanding of the current opportunities available with regard to the application of existing PA registries for CEA.