I read with interest the article by Grant et al.(1) regarding the effects of octreotide on severe nonproliferative and early proliferative diabetic retinopathy. However, I have a few observations and questions for Grant et al. First, I noticed that the baseline Early Treatment of Diabetic Retinopathy Study (ETDRS) scores of the two groups were not compared. Did these two groups have baseline ETDRS scores that were not significantly different? The results of the study cannot be interpreted without the knowledge of the baseline ETDRS scores of the two groups. In addition, it appears that the photos were not judged by a reading center, as is traditional for most diabetic retinopathy studies. Also, there is no comment regarding the potentially confounding ocular factors, such as cataract extraction or yttrium:aluminum:garnet (YAG)laser capsulotomy, that might cause the diabetic retinopathy to progress independently of glycemic control(2). Finally, I noticed that all of the patients in the octreotide treatment group had a decrease in HbA1c levels, averaging 1.2%. In the control group, however, the average HbA1c level of the group remained essentially unchanged throughout the study, and in some subjects, HbA1c actually increased during the study. At the end of the study, the mean HbA1cin the treatment group was 7.1%, whereas in the control group it was 8.3%. This discrepancy is similar to the difference in HbA1c levels between the conventional and the more intense treatment groups in the U.K. Prospective Diabetes Study, in which the end point of needing retinal laser was decreased by 29% in patients with more intensive control and lower HbA1c levels (3). Because Grant et al.'s study was open label, it was possible that patients receiving study drugs were more motivated to be vigilant about their diabetes control than those who were treated conventionally. Therefore, we may actually be seeing an epi-phenomenon in which the difference in the need for panretinal photocoagulation between the treatment and control groups was actually due to the improved glycemic control in the treatment group. It is not clear from this study whether improved glycemic control is attributable to octreotide or to patient-initiated measures.
The use of octreotide to inhibit IGF-1 and its effects on diabetic retinopathy is an interesting and promising avenue of therapy. I look forward to the results of ongoing randomized clinical trials that use a reading center and are conducted in a prospective, masked fashion. The authors of this study appropriately noted that their results only suggest that the use of octreotide may be helpful in delaying the progression of retinopathy. However, this report does not support the use of octreotide for diabetic retinopathy outside the context of a randomized, prospective, controlled, masked clinical trial.
K.M.G. has been a paid investigator for Novartis.