Anti-VEGF, A Biography: Part 2

New Retina Radio continues its biography of anti-VEGF. John D. Pitcher III, MD, picks up the story where we left off: in a world where pegaptanib was the only intravitreal injection approved for treating wet AMD. But a deep dive into the literature and a few calculations on the back of a napkin are about to change retina practice forever. Guests in this episode include Bob Avery, MD; Dave Brown, MD; Anne Fung, MD; Jeffrey Heier, MD; Andrew Moshfeghi, MD, MBA; Kirk Packo, MD; and Philip Rosenfeld, MD, PhD.

Speaker 1: On this episode of New Retina Radio.

Moshfeghi: Not only did we see a response, we saw a response that was way better than what we were getting, with current therapies.

Fung: they returned to our clinics for OCT imaging on post-injection days one, two, four, seven, fourteen, then again at day thirty and then at days thirty-one, thirty-two, thirty-four, thirty-seven, forty-four and then again at day sixty.

Heier: patients going from a retinal thickness of 500 microns down to 300. You didn't need highly evolved OCT to understand those differences.

Speaker 1: That, and more, coming up.

Speaker 1: New Retina Radio is an independent podcast supported with advertising by Alcon.

Mike: This is Mike Lee at Horton Plaza Park in San Diego. New Retina Radio is brought to you by Alcon Surgical. Stop by our booth at an upcoming meeting to see how Alcon is advancing vitreoretinal surgery.

Krzywonos: Alright, are you ready?

Jaraha: I am.

Krzywonos: I’m Scott Krzywonos.

Jaraha: I’m Ranna Jaraha.

Krzywonos: This is New Retina Radio. We’re continuing our biography on the history of anti-VEGF, narrated by John Pitcher.

Jaraha: Let’s recap the first part of this story. If you don’t remember, or you never listened in the first place, go back and listen to part 1 of this miniseries.

Krzywonos: Here’s a quick summary of where we are: in 1971, Judah Folkman, who was a young and extremely talented clinician and cancer researcher in Boston, coined the term “angiogenesis” in a New England Journal paper.

Jaraha: In 1989, Napoleon Ferrara and William Henzel at Genentech identified the growth factor that led to angiogenesis, and called it VEGF.

Krzywonos: Eventually, the jump is made to retina, where it is determined that anti-VEGF, an antibody, injected intravitreally could halt the progression of angiogenesis, and thus might be able to affect AMD progression. At the time, other treatments for AMD were destructive and not very effective.

Jaraha: Fast forward to the early 2000s, a company called Eyetech engineered Macugen, or pegaptanib, and got it approved by the FDA in 2004.

Krzywonos: Around the same time, researchers were learning more about VEGF, which, they discovered, came is several varieties, VEGF-A, VEGF-B, C, and D. VEGF-A was the important one here, and that was broken down into several isoforms, called VEGF-A 121, 165, and 189.

Jaraha: Macugen blocked 165 and 189, but didn’t block 121, which was operative in angiogenesis.

Krzywonos: By the time the drug hits the market in 2005, it’s the best option out there, but it’s still not very good.

Jaraha: Macugen is about to be evaluated for treating RVO and diabetic eye disease, when the entire system is disrupted by the arrival of a new drug—or rather, by a new set of drugs.

Krzywonos: And that’s where things start in this episode.

Jaraha: John spoke with us from his home base in Albuquerque. We last left off in 2005, when Macugen was about $1000 per dose and wasn’t as effective as had been hoped. At the same time, there was another anti-VEGF agent in the pipeline. Here’s Phil.

Rosenfeld: Hi. My name is Phillip Rosenfeld. I'm a professor of ophthalmology at the Bascom Palmer Eye Institute of the University of Miami Miller School of Medicine.

Pitcher: Phil described that during the era of photodynamic therapy and Macugen, another therapeutic agent was in the pipeline.

Rosenfeld: Well, it was about that time that Genentech became interested in what was called rhuFab.

Pitcher: RhuFAB.

Rosenfeld: R-H-U capital F-A-B, rhuFab

Pitcher: The second version was called RhuFAB V2, and you might hear a few doctors call it that in the episode. Nowadays, it’s just known as Lucentis, or ranibizumab. The gears around RhuFAB were already well in motion by the time Macugen was approved by the FDA.

Rosenfeld: they were moving forward into a phase one clinical trial to investigate rhuFab for a neovascular age related macular degeneration. And we had a meeting. I believe it was ARVO 2000 to organize this phase one clinical trial.

Pitcher: A phase 1 trail was indeed conducted.

Rosenfeld: Phase one clinical trials are really safety trials. So, we looked for efficacy but our goal really was to determine the dose limiting toxicity and the maximum tolerated dose.

Pitcher: The results came in…

Rosenfeld: we published the phase one trial in ophthalmology, and the results were very impressive.

Pitcher: In order to know why they’re impressive, you’d have to know a bit about the study. The phase 1 researchers performed a dose escalation trial.

Rosenfeld: we had cohorts of six patients and we'd treat them with the lowest dose

Pitcher: And then they would escalate the dose slightly…

Rosenfeld: and then we would escalate the dose further

Pitcher: They were trying to find the maximum dose a patient could tolerate.

Rosenfeld: we found that when we got to the .5 milligram dose, patients tolerated that dose very well but when we got to the 1 milligram dose, there were two patients that were injected, and they both developed marked inflammation in the eye.

Pitcher: Therefore, they determined that…

Rosenfeld: the dose limiting toxicity was inflammation and the maximum tolerated dose was 0.5 milligrams.

Pitcher: That would be reassessed in a study to follow, showing that the maximum dose was 2 milligrams. Anyway, Phil and his colleagues noticed something in the phase 1 study.

Rosenfeld: In the Phase I, when we looked at visual acuity outcomes, we were very surprised that in this cross section of patients that received these low doses of Lucentis, up to .5 milligrams, we saw vision improvement.

Pitcher: That’s interesting to note in a phase 1 trial, but again, the important focus of phase 1 is safety and tolerability. That’s it. Phase 2, however, looks at safety and efficacy. And after phase 1 results for Lucentis showed no safety risk after .5 milligrams, phase 2 initiated.

Rosenfeld: there were two different Phase II studies.

Pitcher: One was a trial with monthly dosing…

Rosenfeld: and Jeff Heier was the principal investigator that led that trial

Pitcher: The other phase 2 trial was a dose escalation study…

Rosenfeld: in which I was the principal investigator.

Pitcher: You probably remember Jeff. We heard from him in part 1 of this series.

Heier: Hi, I'm Jeffery Heier, I'm the co-president and medical director of Ophthalmic Consultants of Boston and for the last 18 years or so I've served as the director of retinal research for Ophthalmic Consultants of Boston.

Pitcher: Phil and Jeff were the two major investigators for Lucentis in its earliest days.

Heier: I was one of the lead investigators along with Phil Rosenfeld in the Genentech Phase 1/2 studies for ranibizumab. In fact, the first five patients who were multi-dosed with ranibizumab were all at our center.

Pitcher: Jeff saw those same inflammation issues that Phil saw during the phase 1 trial.

Heier: There was a percentage of patients early on who were getting fairly significant inflammation with ranibizumab.

Pitcher: Inflammation was present in a not insignificant number of patients.

Heier: 0 to 20% of patients

Pitcher: Jeff said that inflammation was due to the formulation of the drug, not the molecule itself.

Heier: the formulation at that time was a lyophilized powder that was mixed up and it turns out there was, some component of that formulation that led to inflammation.

Pitcher: Inflammation didn’t get in the way of vision, however.

Heier: those patients still did well, despite the inflammation. They still ultimately had nice visual responses and nice anatomic responses but it was still something we had to be concerned about

Pitcher: Those issues would be fixed later.

Heier: But once we got to the phase three studies, Genentech had gone to a different formulation that pretty much eliminated this type of inflammation.

Pitcher: Before we get to phase 3, though, let’s back up. There’s still phase 2 to hammer out. Rather than get bogged down in the data, here’s Phil to summarize what the researchers learned.

Rosenfeld: what we learned from the Phase II clinical trial is that the response to RhuFab V2 was immediate and dramatic. you could see a response within one to two days, definitely after two weeks, in patients getting anti-VEGF therapy.

Pitcher: He doesn’t mince words.

Rosenfeld: It was revolutionary. It was totally unexpected and, uh, we had never seen these kind of results before in the treatment of neovascular AMD.

Pitcher: Phil’s team in Miami conducted OCT imaging of patients in the phase 2 trial. It was not required…

Rosenfeld: OCT was never an official part of the study protocol.

Pitcher: …but they performed some for kicks.

Rosenfeld: However, we routinely imaged patients with the Stratus OCT.

Pitcher: Once again, the results were stunning.

Rosenfeld: And what we observed, was dramatic improvement in the macular profile. The fluid resolved and the macular remained dry for the duration of the Phase 2 study. Both the dose escalation, as well as the monthly dosing arms.

Pitcher: Jeff said that the anatomic differences observed were based on images from equipment that was not sophisticated, but that it didn’t matter.

Heier: it wasn't anywhere near what it is now, but it was still reliable enough to appreciate big differences

Pitcher: Which is to say, the differences were HUGE.

Heier: patients going from a retinal thickness of 500 microns down to 300. You didn't need highly evolved OCT to understand those differences.

Pitcher: Still, Phil wanted to know more about the relationship between OCT observations and Lucentis treatment for AMD.

Heier: And that set the stage for what’s been known as the PrONTO study.

Pitcher: That’s capital P, lowercase r, capital O, N, T, O. PrONTO. The study’s full name is Prospective OCT imaging of patients with neovascular AMD treated with intraocular ranibizumab. One of those study names where the acronym is contorted a bit, but you get the picture.

Fung: the PRONTO Study was the brainchild of Phil Rosenfeld.

Pitcher: That’s Anne.

Fung: Hi, this is Anne Fung. I'm the associate group medical director in US medical affairs, ophthalmology, at Genentech, and also a practicing retina specialist with Pacific Eye Associates in San Francisco.

Pitcher: At the time, she was a fellow under Phil at Bascom Palmer in Miami.

Fung: I came in as a fellow just about the time that PRONTO was started

Pitcher: Pronto was an IST

Fung: An Investigator-Sponsored Trial

Pitcher: IST’s allow doctors to execute their own study designs with industry support.

Fung: an investigator, a clinician essentially, can come up with his or her own curiosity and question and think of a potential study design of how to answer it and then approach a company to ask for financial and drug support.

Pitcher: This all arose from Phil’s observations in the phase 1 and phase 2 Lucentis trials.

Rosenfeld: based on the results of our OTC findings in the early phase Lucentis trials, we approached Genentech with the idea of an investigator sponsored trial to look at the role of OCT in determining when re-treatment could be offered to patients after 3 monthly doses of Lucentis.

Pitcher: Genentech agreed to support the study, and enrollment started.

Fung: Phil allowed me to help him to recruit the patients and then follow them, and over a course of a two-year study I think we collected about 36 to 40 data points for each of these 40 patients.

Pitcher: There was a massive amount of data to collect in PrONTO.

Fung: we asked patients to first come in for a screening visit and then the baseline visit at which they would get their first injection but then they returned to our clinics for OCT imaging on post-injection days one, two, four, seven, fourteen, then again at day thirty and then at days thirty-one, thirty-two, thirty-four, thirty-seven, forty-four and then again at day sixty and it was monthly after that.

Pitcher: PrONTO was a 2-year study. The protocol in year 1 required loading doses and then treatment based on anatomic findings.

Rosenfeld: we gave three monthly injections and, not surprisingly, the macular fluid resolved after those three monthly injections. And then we would follow patients and our rule in the Pronto study, was that we could re-treat when one hundred microns of fluid re-accumulated in the central macula.

Pitcher: Why 100 microns, you ask?

Rosenfeld: we used one hundred because there was a great deal of skepticism that once a little bit of fluid accumulated, more fluid would follow. Seems obvious now, but back then our colleagues just weren't used to using OCT to follow patients.

Pitcher: Fluid was disappearing at a rate that was quicker than some expected.

Fung: we were really surprised to see that there was statistically significant reductions in retinal edema by day one after an injection and this continued literally reduction day one, day two, day four, day seven and the fluid came down ... First the intraretinal fluid then the subretinal fluid and then the PED fluid was the last to resolve.

Pitcher: During year 2, the study protocol changed.

Rosenfeld: in the second year of the trial, we wanted to make fluid what I call, a never event. The re-treatment criteria were changed so that if any fluid reappeared, re-treatment could be offered. Or, if a pigment epithelial detachment grew in size, re-treatment was offered.

Pitcher: The year 2 results were important, too.

Rosenfeld: We demonstrated in the 2nd year of the trial that we could stabilize the macula once we developed these new rules in which we avoided any fluid from re-accumulating.

Pitcher: Perhaps the most important thing to draw from the PrONTO study was…well, let’s have Anne say it.

Fung: I think PRONTO was a big step forward for appreciating how tight the connection between imaging with OCT could be with anti-VEGF therapy.

Pitcher: At the time, OCT was a technology used for pathologies other than AMD. Its use in PrONTO helped demonstrate its value in AMD treatment.

Fung: OCT technology had existed for quite some time before anti-VEGF therapy did, but prior to anti-VEGF therapies, we mainly used it for understanding the anatomic structures such as epiretinal membrane and Vitreomacular traction.

Pitcher: And even if it was useful for those structural conditions, most retina doctors didn’t really need an OCT device in their offices.

Fung: OCT probably didn't gain as much traction during the pure anatomic times, because retina specialists largely see epiretinal membrane and VMT on examination.

Pitcher: For a disease as prevalent as AMD, and with a therapy coming down the pike that could be this revolutionary, OCT would soon be a tool in every office.

Fung: with anti-VEGF therapy, now that we were able to quantify 20 microns, 10 microns of extra fluid, this was starting to make a difference because as a human eye, it's really hard to quantify whether something has changed in such a granular way. We can see probably 200 or 300 microns of edema, but it, unless you're Don Gass it was probably hard (laughs) to see the really minuscule changes day by day, and OCT unlocked that for us.

Pitcher: So, we’ve got phase 1 finished, phase 2 finished, and an IST finished, all for Lucentis. Retina researchers are starting to understand the relationship between fluid and angiogenesis and treatment—it’s all coming together. But then a new player arrives. Back, after this.

Speaker 1: New Retina Radio is an independent podcast supported with advertising by Alcon.

Lee: This is Mike Lee here at the bowling tournament during the final night of the Retina Fellows’ Forum here in Chicago. New Retina Radio is brought to you by Alcon. Stop by our booth at an upcoming meeting to see how Alcon is taking surgical retina into the future. Alright, let’s get back to the program.

Krzywonos: Awesome! Thank you, Mike.

Lee: We’ll see.

Pitcher: Phil Rosenfeld and Jeff Heier were working on the early Lucentis trials, and Phil was working with a team that included Anne Fung to assess the anatomic effects of Lucentis treatment for AMD. Now, there are two routes to go here: one is the route of the phase 3 Lucentis trials, and the other is that of Avastin. We’ll address phase 3 at the end of the episode. Let’s start with Avastin, and Phil, who, he’ll admit, is a literature junkie.

Rosenfeld: My interest in anti-VEGF therapy drew me to the scientific literature. I wanted to understand how rhuFab V2, Ranibizumab, Lucentis, was developed.

Pitcher: So Phil dug into the literature.

Rosenfeld: What I learned during my early stage work with rhuFab was that Genentech was developing another molecule called Bevacizumab

Pitcher: Known today as Avastin, which was developed as a chemotherapy to be administered intravenously.

Rosenfeld: I did my research. I went back and looked at the published literature from Genentech scientists on the development of rhuFab V2 and Bevacizumab and it became abundantly clear that both molecules were derived from a mouse monoclonal antibody against VEGF.

Pitcher: This is a good time to hear from one of Phil’s trainees at Bascom, Andrew.

Moshfeghi: Hi, I'm Andrew Moshfeghi and I'm associate professor of ophthalmology at the USC Roski Eye Institute.

Pitcher: Andrew trained under Phil, alongside Anne, at Bascom.

Moshfeghi: I was a medical retina fellow with Phil Rosenfeld and Carmen Puliafito and others at Bascom Palmer Eye Institute, and Phil was getting ready to plan for a study using intravenous Avastin to treat neovascular age related macular degeneration, and it was in that context that I first got exposed to Avastin.

Pitcher: Andrew worked on a trial called the SANA study.

Moshfeghi: the SANA study was the system Avastin for neovascular AMD study. That's where the acronym comes from

Pitcher: No one knew if administering a monoclonal antibody similar to Lucentis, but delivered intravenously, could have any effect on AMD patients.

Moshfeghi: it was more of a proof of concept that giving a drug with a mechanism of action, of inhibiting VEGF would make sense when injecting intravenously. Normally, we think that drugs that we inject intravenously aren't going to have any significant uptake in the eye due to the blood retinal barrier.

Pitcher: …but

Moshfeghi: From the mechanism of action perspective it made a whole lot of sense because there was already work being done with Lucentis.

Pitcher: It might seem odd to go intravenous, but…

Moshfeghi: we already had a drug that was FDA approved and it was FDA approved for intravenous administration at certain dosages, albeit for other non-ocular conditions. So I think from the perspective of trying to get a study off the ground, the easiest way to do that would be to use it in the way that it was labeled, i.e., intravenously.

Pitcher: Plus, there was the issue of size.

Moshfeghi: it was not considered at the time to give it intravitreally, just because of the large size of the molecule being roughly more than twice the size of the Lucentis molecule.

Pitcher: Researchers were pretty sure that the molecule’s size meant that intravitreal injection was not going to work.

Moshfeghi: It was felt that if it was given intravitreally, that it would not cross the retina to get to the site of action in the outer retina sub-retinal space and choroid.

Pitcher: That would later be proven wrong—we’ll get back to that. For now, let’s focus on the SANA study. It was a small study, only 18 patients. And there were some hurdles to clear.

Moshfeghi: we had to go through IRB obviously, and, get exemption from the FDA to proceed with this

Pitcher: And the dosage was different for each patient.

Moshfeghi: basically patients were treated with intravenous Avastin according to their body weight and et cetera, you know, using the same criteria that the hematologist used when using Avastin for metastatic colorectal cancer.

Pitcher: One thing to note: today’s intravitreal injection of Avastin is quite small—about 1 or 1.25 milligrams per eye. The intravenous dose is huge compared with that, and the dosing frequency—at least in SANA—was rigorous.

Moshfeghi: When we give it intravenously, we're often giving on the order of 400 milligrams and we're giving it every two weeks.

Pitcher: Andrew thought it might not work.

Moshfeghi: we fully expected, perhaps, that this would not work due to the restriction of the blood retinal barrier.

Pitcher: And were they surprised with the results?

Moshfeghi: We were really surprised. Not only did we see a response, we saw a response that was way better than what we were getting, with current therapies

Pitcher: Phil agreed.

Rosenfeld: we were flabbergasted. We never expected such a dramatic result.

Pitcher: The results were immediate in some patients.

Rosenfeld: We give our first patient the dose and within 24 hours, the OCT results were breathtaking. We noticed a complete resolution of the macular fluid. And patients saw better.

Pitcher: Safety was a concern, and consent forms had to be rewritten during the trial because of a black box warning issued by the FDA.

Rosenfeld: by August of 2004, the FDA issued a black box warning.

So we went back to the IRB and we showed them this new FDA warning and we had to rewrite the consent form, and we had to notify every patient in the study and get them to be re-consented.

Pitcher: And?

Rosenfeld: And every patient stayed in the trial.

Pitcher: The safety profile in SANA was remarkably unremarkable.

Rosenfeld: we did not have any thromboembolic events, we had a mild elevation of blood pressure. On average about 11 millimeters of mercury. And we had two internists monitoring all the patients and by this time the study was over, everyone ended the study with a lower average blood pressure than they started the trial.

Pitcher: The team on the SANA trial shared their data at ARVO 2005.

Moshfeghi: That was all in poster form so we had like six posters that were, you know, within like 20, 30 yards or each other. We could all see each other. And there was a huge amount of buzz at that ARVO meeting when folks would come by and ask all those questions

Pitcher: The results were so encouraging that a group of clinicians gathered to discuss them.

Rosenfeld: we were determined to move forward with a multicenter prospective clinical trial investigating systemic Avastin for wet AMD. So, we organized a meeting at the Marriott Harvard Beach Hotel in Fort Lauderdale on Sunday morning, May 1st, in which we invited 50 respected retina specialists in the hope of kick starting a prospective clinical trial.

Pitcher: Everyone was impressed. However…

Rosenfeld: they were concerned about the potential safety risks from systemic anti-VGF therapy.

Pitcher: You can guess how Phil felt after that session.

Rosenfeld: I was disappointed. I wanted to move forward pretty quickly with the prospective clinical trial because I knew the window of opportunity was closing.

Pitcher: You see, Lucentis was going through its phase 3 trials, and they would likely be presented in a few months at ASRS in Montreal. Phil felt that once Lucentis was approved by the FDA, there would never be a prospective trial for Avastin.

Rosenfeld: So I accepted their advice and I went back that week after ARVO and I was determined to design a clinical trial with lower doses of Avastin to show that it was effective.

Pitcher: And then, as if delivered by the divine, Phil discovered something.

Rosenfeld: So, this is when I had my eureka moment. I was sitting down calculating how I needed to dilute the Avastin a-and what concentrations that should be studied when I had an insight.

Pitcher: And that insight?

Rosenfeld: I realized that if I put Avastin into a syringe and injected it like Lucentis, for $50 I could have a highly effective anti-VEGF therapy.

Pitcher: Phil noticed something about the molarity of the two drugs.

Rosenfeld: the number of VEGF molecules in an Avastin bottle, per unit volume, were identical to the number of anti-VEGF binding molecules in a vile of Lucentis.

Pitcher: The math is pretty straightforward, actually.

Rosenfeld: Avastin was packaged at 25 milligrams per mil and Lucentis was going to be packaged at 10 milligrams per mil. Avastin was about three times larger in terms of molecular weight than Lucentis.

Pitcher: To Phil, the meaning was clear:

Rosenfeld: That meant if you use an equal volume of both drugs right out of the bottle, you were injecting roughly the same number of anti-VEGF binding molecules.

Pitcher: Now, this is where economics came into play.

Rosenfeld: if you injected .05 milliliters of Lucentis, that would be roughly equivalent to .05 milliliters of Avastin when injected into the eye. The big difference was that .05 milliliters of Avastin cost $50 where that .05 milliliters of Lucentis ended up costing about $2000.

Krzywonos: Okay, a quick sidebar here on pricing, and on the question of retinal penetration of Avastin. Ranna and I spoke with Bob Avery about this.

Avery: Hi, I'm Bob Avery. I'm a retina specialist, at California Retina Consultants in Santa Barbara.

Jaraha: Take it away.

Avery: the way Medicare prices things, it's by the unit, or the milligram

Krzywonos: See, Genentech was already selling Avastin, albeit for a different indication.

Avery: Well, they already had Avastinapproved for colon cancer.

Krzywonos: And, because they were already selling it via, among other avenues, CMS…

Avery: there was already pricing in place for that, and, uh, I don't know that they would be able to, charge a whole lot more, uh, per unit, or per milligram, than they were charging for cancer.

Krzywonos: Basically, because a price was already set for Avastin as it was sold—that is, in large units—a pharmacy would pay that amount for the unit, and then, once compounded, charge a commensurate price.

Jaraha: In other words, if one unit was being split into 100 parts, then each new unit would cost about 1% of the entire original unit price.

Krzywonos: Thus, a syringe of Avastin compounded for intravitreal injection, which is a small percentage of the whole unit, would only cost about $50.

Jaraha: There was also a question that Andrew Moshfeghi brought up, which was about Avastin crossing the retina.

Krzywonos: Turns out there was some flawed research on that front. Again, Bob.

Avery: the mantra at the time was that, no, these full-length antibodies don't penetrate through the retina. They're too big.

Jaraha: Like Phil, Bob went back to the literature.

Avery: But if you actually go to the paper, it was published in an obscure journal of pathology and you look at the only published imagine, uh, purporting to show that there was no penetration of full-length antibody though the retina, it did not show what the paper says it shows.

Jaraha: Bob did some studies on rabbit models...

Avery: At the Academy that year I showed a slide of, uh, of the penetration of Avastin into the outer retina.

Jaraha: Which was quite a surprise.

Avery: You could hear a pin drop in the room.

Jaraha: There was one big problem with the original research: They weren’t using Avastin.

Avery: It wasn't Avastin they were using. It was actually herceptin or another full-length antibody.

Krzywonos: Really?

Jaraha: Yeah, and Bob said that using herceptin was understandable.

Avery: it had a different binding site, but it was still the same, you know, antibody type fragments. Same molecular weight, more or less.

Jaraha: A hiccup, in the grand scheme of things, but an important one: without that hiccup, it seems plausible that Genentech may never have engineered Lucentis at all, and instead just researched if Avastin could have treated AMD.

Avery: there was that little quirk back then that seemed to imply that it wasn't going to penetrate, and for whatever reason it, um, I believe was mistaken.

Jaraha: Okay, enough from us. Let’s go back to John Pitcher, who was telling us about when Phil Rosenfeld was trying to figure out if he could compound Avastin for intravitreal injection.

Pitcher: So, after Phil performed his calculations, he visited the pharmacy director at Bascom Palmer Eye Institute.

Rosenfeld: Serafin Gonzalez

Pitcher: The conversation went like this

Rosenfeld: I said, "Could you put Avastin into a syringe so that I could inject it into the eye?" He said, "Let me research it."

Pitcher: Serafin did his research…

Rosenfeld: he went off and he looked at the United States Pharmacopeia Guidelines

Pitcher: And he said, yeah, sure. If Serafin followed chapter 797 of the United States Pharmacopeia and tested for any microbiologic activity, then it was perfectly legal and safe to compound Avastin in a syringe for intravitreal injection.

Rosenfeld: So then I went to my chairman, Carmen Puliafito.

Pitcher: Carmen was the chairman of the Bascom Palmer Eye Institute in May 2005.

Rosenfeld: on the back of an envelope literally I, I wrote out what I want to do. I showed that the molarities were the same and if I inject Avastin it would be roughly the equivalent amount of Lucentis.

Pitcher: Phil asked permission to inject Avastin intravitreally.

Rosenfeld: He said, "Well we inject drugs off-label all the time."

Pitcher: Carmen seemed amenable to the idea.

Rosenfeld: he said, "Yeah, give it a shot." But I told him that I would only use it on patients as salvage therapy.

Pitcher: Context is important here. Patients were receiving either photodynamic therapy or Macugen for AMD, and many were failing. There was no shortage of patients who needed salvage therapy. Now, while all of this is happening, the phase 3 trials for Lucentis were wrapping up. Time to learn more about phase 3.

Brown: I'm Dr. David Brown. I'm a retina specialist in Houston, Texas and a clinical professor of ophthalmology at Baylor College of Medicine in Houston Methodist Hospital.

Pitcher: Dave was involved heavily in ANCHOR and MARINA, the pair of phase 3 clinical trials that assessed Lucentis for neovascular AMD.

Brown: I was the first author of the ANCHOR paper in the New England Journal of Medicine and second author with Phil Rosenfeld on the MARINA Paper.

Pitcher: Dave and his colleagues were familiar with the phase 1 and phase 2 data, and they were confident that the drug would work in phase 3.

Brown: we all thought the Phase 3, I did, thought the Phase 3 would kill it.In other words, meet all expectations because we had had so much experience with the molecule.

Pitcher: But Dave also knew of treatments that had failed before, so he curbed his enthusiasm.

Brown: I really thought the Macugen trials would do better.

Pitcher: However, for anyone involved in the trail, it was clear that Lucentis was working. Kirk Packo.

Packo: Hi, I'm Kirk Packo, resident specialist in Chicago. I'm also the professor and chair of the Department of Ophthalmology at Rush University here in Chicago.

Pitcher: Kirk said everyone knew which patients were randomly assigned Lucentis and which

ones were not.

Packo: all of us around the country knew that something was up here, because it, it didn't take a real lot of, uh, knowledge of whether or not you were given the, the drug when you were in the Phase 3 trial, to know that half your patients were doing fabulously well, and the other half were all rotting their vision away.

Pitcher: Dave said that ANCHOR and MARINA produced some of the most impressive data in medicine.

Brown: when we presented it to the New England Journal, some of the reviewers stated that this was as big of a breakthrough as has been seen in medicinesince the testicular cancer trials or the lymphoma trials showed a 10-fold improvement in survival rates.

Pitcher: In Kirk’s estimation, the entire industry was waiting to hear the data’s presentation.

Packo: We were anxious to see the real data, because none of us, I think, even at that time knew just how the drug was actually working until we saw the graphs. And that was going to be released at the ASRS meeting in Montreal, and that was in July of 2005.

Pitcher: Which was a mere 2 months after the ARVO meeting where the SANA team shared their data. People were pumped.

Packo: And I remember being in the room, uh, which was absolutely packed, and it was packed not only with every doctor who registered, I mean everybody, whether you wanted to go out and enjoy Montreal, but everybody for that session was in the room.

Pitcher: Phil was going to be a big part of this day—and the irony of this arrangement was not lost on him. He explains it from Genentech’s perspective.

Rosenfeld: So here they have their principal investigator for their phase three clinical trial, running the PrONTO study, which uses OCT to determine the best interval for retreatment, running the SANA trial, systemic therapy for Avastin for wet AMD, and here I've gone and injected Avastin into the eye.

Pitcher: That’s four touchpoints for Phil: MARINA, PrONTO, SANA, and intravitreal Avastin—all of which will be presented in Montreal at ASRS 2005.

Krzywonos: Okay, boys and girls, thanks for listening to this episode. We’re going to leave you with a bit of a cliffhanger that’s set north of the border. We’d like to thank our guests: Phil Rosenfeld, Jeff Heier, Anne Fung, Andrew Moshfeghi, Bob Avery, Dave Brown, and Kirk Packo—and, of course, John Pitcher.

Jaraha: We’ve got one more episode in this season of New Retina Radio, and it will be the conclusion of this biography on anti-VEGF.

Krzywonos: I’m Scott Krzywonos.

Jaraha: I’m Ranna Jaraha.

Krzywonos: See you next time.

Jaraha: Bye now.