AIMS: Haptoglobin (HP) phenotype has been reported to modulate fenofibrate benefit on coronary artery disease in type 2 diabetes. It is unknown whether HP phenotype and levels modulate fenofibrate benefit on sight-threatening diabetic retinopathy (STDR). METHODS: In plasma from 8,047 Australasian adults with type 2 diabetes in the FIELD trial, HP phenotype was determined, and HP levels were measured at baseline and after six-week fenofibrate run-in. RESULTS: There were 307 new first on-trial STDR events over five years. Baseline HP levels and phenotype were not related to STDR risk. Fenofibrate benefit on STDR vs. placebo (-32 % overall), was greatest in participants with the lowest baseline HP level tertile (hazard ratio [95 % CI] 0.41 [0.26-0.65], vs. 0.82 [0.56-1.21] and 0.84 [0.56-1.27] for tertiles 2 and 3 respectively, P for heterogeneity = 0.019). During run-in, fenofibrate reduced HP levels by 20.7 %. However, fenofibrate benefit on STDR did not differ significantly by HP phenotype or change in HP levels during run-in after adjustment for confounding factors. CONCLUSIONS: Regarding STDR, fenofibrate benefit is greatest in type 2 diabetes patients with the lowest baseline HP levels, which may reflect patients more susceptible to oxidative retinal injury. All HP phenotypes benefit from fenofibrate.