Robots for Retina?

On Saturday, Prethy Rao, MD, MPH, received the 2017 The Raymond R. Margherio Award.  She presented her project entitled “Real World Vision in Age-related Macular Degeneration Patients Treated with Bevacizumab, Ranibizumab or Afibercept only Over One Year in a Cohort of the IRIS Registry”.  We caught up with Dr. Rao to discuss. EN: Congratulations on receiving the Margherio Award, Dr. Rao.  Can you tell us the motivation for doing this project? PR: We knew the IRIS Registry had the capability to answer thousands of questions. So, that was a given. David Parke said it best when he stated that the IRIS has the ability to generate 100's of PhD level theses. However, we wanted to make the question meaningful and not just another study.  We wanted to report something that would help guide clinical practice--after all, that's the main goal of IRIS. It's a database from, and for, the practitioners that are on the front lines. EN: What was the study design? PR: It was a retrospective review of patients who were treated with a single anti VEGF subtype only for one year. So, either bevacizumab only, ranibizumab only, or aflibercept only without switching for 1 year.  In total, 13,859 eyes from an equal number of patients were included. EN: How were you able to access the data from the iris registry? PR: I was fortunate enough to take a trip down to SF to work with the data analysts that helped construct the unique dataset that we wanted. Then, we were able to analyze the data remotely in a safe way. EN: What were the key results? PR: All 3 drugs improved vision from baseline to 1 year--as we expected. Surprisingly, there was no statistical difference in the final or magnitude of change in vision after multivariate adjustment. EN: Was there a difference in the frequency of injections for the different agents? PR: Overall, no. When we broke it down by vision losses and gains (ie. 3 line improvement), we found that the ranibizumab group received more number of injections compared to the other drug subtypes. EN: Did you notice any trends in the choice of drug based on presenting visual acuity or other factors? PR: It appeared as though the patients in the bevacizumab group tended to have worse baseline vision compared to the other groups. Even after controlling for this, there was no difference in the drug groups in terms of final and mean change in vision. Something that was pretty interesting to me (and that we didn't discuss in the presentation) is that the aflibercept group had a greater percentage of patients with Medicare/Medicaid and private insurance compared to the other two drugs. As we would expect, it seems that cost of drug and/or coverage may have an impact in clinical decision making. However, that is more of an extrapolation and personal interpretation. Further studies would need to be done to confirm. EN:  Do you think these results will influence your choice of drug to initiate therapy? Definitely. It certainly helps when talking to patients about which drug to start therapy. We often talk about the CATT trial, which answered many of our questions about treatment protocol and efficacy of ranibizumab and bevacizumab.  However, it's hard to interpret the impact of a few letter difference in the clinical trial setting versus the real world.  In our study, I think it was valuable to know that all 3 drugs had similar trends, and that the response a patient has early on (ie. after 2nd or 3rd injection) can correlate with how they'll probably do at 1 year.  But, the limitations obviously need to be kept in mind: vision was not ETDRS based, no OCT characteristics were analyzed, and we were unable to determine the exact treatment protocol based on the current limitations of the IRIS Registry.  Arguably, however, vision is the most important outcome, and this study provides that data. EN: Excellent.  Thank you again for this important study, and congratulations on the Award! Eric Nudleman MD, PhD