Background/Purpose: During randomized clinical trials (RCTs) with pegloticase, the incidence of infusion-related reactions (IRs) was 26% including 5 cases of anaphylaxis (determined post-hoc using published criteria).^1,2 Post-hoc analyses revealed a relationship between loss of urate-lowering response and IRs. This relationship could not have been detected during the trials as investigators were blinded to pre-infusion uric acid (UA) values. These analyses led to guidance on measuring UA levels as a biomarker of therapeutic response and IR risk, and included the recommendation that patients discontinue pegloticase if serum uric acid (UA) was >6 mg/dL, particularly at 2 consecutive q2wk infusions.³ In order to determine whether this guidance was effective in reducing the incidence of IRs, the company reviewed IRs from post-approval safety data.

Methods: We estimated the incidence of IRs by comparing usage information from the RCTs and the post-approval period (Sept 14, 2010 to June 1, 2012). An estimate of the total number of infusions given in the post-approval period was derived from the number of vials sold during that period compared to the number administered to a defined number of patients in the 6-month RCTs.

Results:  852 individual vials for infusion were administered to the 85 patients receiving the approved dose of pegloticase during the RCTs (8 mg every 2 weeks) and there were 22 reports of IRs (see Table). During the post-approval period, 5727 vials were sold and the company received 58 reports of IRs and anaphylaxis (43 reports of IRs, 15 reports of anaphylaxis). The above information was used to estimate the relative risk reduction for IRs during the post-approval period compared with the RCTs period. When determined this way, there was a 61% reduction (95% CI: 36.3 to 75.9) in the risk of IR during the post-approval period vs. the RCTs.

While providing evidence for relatively low rates of IRs, these estimates have substantial limitations. The actual number of patients receiving infusions is difficult to estimate from vials sold.  In addition, the number of IRs depends on voluntary reporting and some adverse events are reported to the FDA and not to the company. Finally, these estimates are valid only for the time period of collection as practice patterns may change with accrued clinical experience.

Conclusion: The risk of IRs estimated for pegloticase during a defined post-approval period was reduced by 61% relative to rates of IRs seen in the RCTs. Post-approval surveillance will continue to assess whether IR risk is adequately mitigated by adherence to the recommended stopping rules based on elevated serum UA. Additional real world data with pegloticase, collected over a longer time period, should provide more robust estimates of IRs that can be compared to the clinical trials experience.

References:

1. Sundy et al. JAMA. 2011;306:711-720.

2. Sampson et al. Annals Emerge Med. 2006;47:373-380.

3. KRYSTEXXA prescribing information.