And Speaking of Pharmaceutical Companies...

Not to excuse this behavior (for example Genentech recently ratcheted up the cost of Avastin for cancer patients so much that even oncologists started taking it to task), the writer of the article didn't define what "considerable success" meant in the 7,000 patients supposedly treated this way. Worse, there's not even a single retrospective study as far as I can tell, much less anecdotal work. In other words, it's by no means a slam dunk that Avastin is as wonderful as this article makes it seem. As one the investigator for one recently reported trial stated:

Despite the impressive results seen in this small study, Dr. Pieramici said that many questions remain unanswered. For example, he said, after performing more than 400 procedures himself, he is still unclear as to what the optimum dose of bevacizumab might be. Also, he said the true improvement in visual acuity still needs to be established in a phase 3 trial. It is curious that while some patients exhibited significant improvements in retinal thickening, as seen on optical coherence tomography, this did not translate into comparable improvements in visual acuity, he noted.

In other words, it's not at all clear how much the anatomic improvements noted result in improvements in visual acuity.

If Avastin is as effective in slowing macular degeneration as these preliminary results suggest, it wouldn't take a very large or very prolonged randomized trial to show its value. That trial should be done.

Physicians are free to prescribe Avastin for any off-label indication supported by the clinical literature. However, whether one's insurance company reimburses you or the doc for the off-label use is another question entirely.

My guess would be that Genentech is not entirely trying to be a profiteer, although this does smell bad on the surface. Unlike small organic molecules, there would be a concern about the stability of reconstituted Avastin that is then split and used for multiple patients - the protein can unfold over time and be a potential hazard if injected in this form.

Tried to comment before and it never showed up.

Abel's point is well-taken. Avastin is a humanized mouse monoclonal antibody against VEGF. I do not know what preservative is in it. Splitting a vial is a good way of increasing the risk of bacterial contamination, and injecting bacteria into the eye is something that you don't want to do.

Let's also remember that this is all based on anecdotal evidence and small patient series, as well, with a large component of a bandwagon effect thrown in for good measure. While there is evidence that Avastin can slow the anatomical changes that occur in macular degeneration, unfortunately the evidence that it has much effect on the thing that really matters, namely visual acuity, is not as strong. In the end, if it doesn't save visual acuity, then it doesn't really do what the patient really wants it to do.

If Avastin is as potent as these opthamologists claim, it wouldn't take a very large randomized trial to show it.

In retrospect I see I was a little hasty posting this without putting more thought into it. Obviously, the ideal thing in this situation is to have single-dose vials, and to have at least two good placebo-controlled randomized trials with improved visual acuity as the endpoint.

On the other hand, even if it does not improve acuity, but halts (or even slows) progression, that might be good enough to justify an NDA.