Complications of Immunosuppression for Eye Diseases

University Of Pennsylvania, Philadelphia PA

Investigators

Linked publications & trials

Abstract

DESCRIPTION (provided by applicant): Ocular inflammatory diseases are major causes of vision loss in the United States, at a relatively early age. Immunosuppression is a primary therapeutic approach for severe cases. In a, a multicenter retrospective cohort study, we linked 7,957 ocular inflammation patients seen by 2005 inclusive to the National Death Index (NDI), finding 936 deaths over 66,902 person-years of follow-up for mortality. Primary findings included: 1) a suggestion of substantially increased overall and cancer mortality with Tumor Necrosis Factor (TNF) inhibitor therapy; 2) no increased risk of overall and cancer mortality with antimetabolite, T-cell inhibitor, or systemic corticosteroid therapy; and 3) overall and cancer mortality risk very similar the general population (suggesting that inflammatory eye diseases patients may be a group in which the indications-for- treatment will not bias study of the relationship between treatment use and mortality or cancer). Millions of individuals in the United States use TNF inhibitors for a range of inflammatory diseases. We propose to extend the study for five years in order to: 1) validate preliminary findings re: overall and cancer mortality risk with TNF inhibitor therapy; 2) evaluate cancer incidence (fatal and non-fatal) for all four major classes of immunosuppressive agents; 3) generalize statistical methodology accounting for time- dependent confounding to the paired organ setting, to allow more valid analyses of treatment effects using observational ophthalmic data; and, 4) apply these methods to evaluate the relative (cost-) effectiveness of alternative immunosuppressive drugs. We will increase the cohort size by ~35% at the five original centers; the follow-up time and number of deaths observed will nearly double. TNF inhibitor-focused enrollment at ancillary centers will enhance statistical power for aim #1. We will link the enlarged database to the NDI and state cancer registries, with a new co-investigator who successfully completed a similar multiple cancer registry study. A biostatistical methodologist will generalize methods for dealing with time-dependent confounding to the paired organ setting, applying these methods along with a seasoned health economist to directly compare the costs and effectiveness of alternative immunosuppresive drugs for ocular inflammation.