Serum levels of rapamycin predict the response of Graves' orbitopathy to sirolimus.

Objective: Compared with methylprednisolone, a greater response of Graves' orbitopathy (GO) at 24 weeks was reported in patients treated with sirolimus (rapamycin) for 12 weeks. We investigated whether serum levels of sirolimus at week-12 predict the outcome of GO at week-24.

Methods: Retrospective investigation in 30 patients [males: 4; females: 26; age 61.3 (9.62) yr] with moderate-to-severe, active GO, treated with sirolimus (2 mg on day-one, followed by 0.5 mg QD for 12 weeks).

Primary outcome: serum sirolimus at week-12 in week-24 GO responders vs nonresponders, assessed by a composite evaluation.

Secondary outcome: establishment of a cut-off value of week-12 sirolimus concentrations predictive of response to treatment.

Results: At week-24, overall GO responders and nonresponders were 18 (60%) and 12 (40%), respectively. Week-12 serum sirolimus was higher in week-24 responders [4.6 (2.2) ng/mL vs 2.8 (1.6) ng/mL in nonresponders; mean difference 1.7; 95% CI from 0.1 to 3.1; P = 0.034)]. A cut-off value of week-12 serum sirolimus was established at 3.5 ng/mL; levels ≥ 3.5 ng/mL were more frequent in week-24 GO responders than in nonresponders [13/18 (72.2%) vs 3/12 (25%); OR: 7.8; 95% CI from 1.4 to 41.2; P = 0.015], with a positive predictive value of 81.2% (95% CI from 60.9 to 92.3), a sensitivity of 72.2% (95% CI from 46.5 to 90.3), and a specificity of 75% (95% CI from 42.8 to 94.5).

Conclusions: Serum levels of sirolimus at week-12 are predictive of GO response to treatment at week-24, suggesting a dose-dependent relationship with GO outcome.