Is off-label prescribing good for your health?

If we ever needed a reminder of the potential health benefits of off-label prescribing, it just arrived on-line in the journal Eye. While every prescription drug on the market has an official indication (or condition) that it can be prescribed for, doctors are allowed to prescribe pharmaceuticals "off-label" for conditions that the drug has not been approved for, as long as there is some evidence or rationale for that decision.

As with any health care practice there are pros and cons to off-label prescribing. Among the pros: It speeds the natural evolution of scientific knowledge. When a drug approved for a certain condition is found to have other beneficial health effects doctors can begin using it without waiting many years for the drug to complete the slow and costly approval process necessary to expand an indication. Among the cons: Doctors may begin using drugs before there's adequate evidence that it is any better than existing treatments, and possibly without fully understanding the safety profile of its use for unapproved conditions.

For many years, age-related macular degeneration (AMD) has offered an apparent case study in the wisdom of off-label prescribing. AMD is the leading cause of blindness in the developed world for those over the age of 50, and the gold-standard treatment for the more serious "wet" version consists of injecting Genentech's AMD-approved drug Lucentis into the back of the afflicted eye-ball. At $2,000 per injection, and requiring multiple injections per year, it is a costly treatment.

To save costs, many ophthalmologists have advocated for the use of another drug from Genentech called Avastin. Though approved for colorectal cancer and not AMD, Avastin is essentially a molecular twin to Lucentis. Independent studies of each drug on its own haven't found either to be clinically superior or more risky. Yet Avastin costs 1/40 as much as Lucentis: $50 per injection.

Now the first double-blind head-to-head trial of the two drugs has failed to find any difference in their effectiveness or side-effect profile. It is a small study, 15 patients received Avastin and 7 Lucentis, all were followed for a year. And those on Avastin received on average twice as many injections, 8 over a 12-month span compared to 4 in the Lucentis group.

In addition to the potential to reduce AMD treatment costs 20-40 fold, the study also has two additional points of interest: Generally health studies with "negative" findings (no differences reported) are considered ho-hum and may have a hard time being published. In this case, the "no-difference" finding is the good news. The second point is that testing treatments against each other, especially a new drug against a gold-standard, is often viewed as better than placebo-controlled trials. A placebo is essentially "no-treatment," and practically anything is better than no-treatment. To truly evaluate the value of a new health intervention, it can be argued that we should be testing it against existing treatments known to work. Some go so far as to argue that the approval process should take value into account as well as effectiveness: Does a new drug offer substantial enough advances over existing treatments to warrant its likely higher cost, and its less-known safety profile?

In the case of Avastin, that answer increasingly appears to be "yes," and at a lower cost and no apparent increased risk.