This article was originally published in a sponsored newsletter.
For this month’s Retina Minute, I have the pleasure of interviewing my dear friend, Dr. Roger Goldberg, who recently presented intriguing data on a new tyrosine kinase inhibitor (TKI) from the phase 2 ODYSSEY trial for wet AMD.
Dr. Goldberg, tell us about how this TKI differs in terms of chemical format, how it is injected, and how the ODYSSEY trial was set up.
This is both an old and a new approach with a TKI. The active ingredient in CLS-AX is axitinib, which is a highly potent, highly selective pan-VEGF inhibitor that blocks all three VEGF receptors (receptors 1, 2, and 3). These are transmembrane receptors that are normally activated by soluble, extracellular VEGF-A,-B, -C, and -D. Our current agents are proteins that bind to extracellular VEGF-A, and prevent signaling from outside the cell. TKIs like axitinib work by blocking VEGF receptor signaling from the inside of the cell.
Axitinib is being used in other clinical development programs, but what's new here is the delivery via suprachoroidal injection. CLS-AX is being developed by Clearside, which, as a company, is focused on using the suprachoroidal space (SCS) as a means for long-acting drug delivery. Suprachoroidal injection has a learning curve, but we have an FDA-approved therapy already in Xipere, which is triamcinolone suspension developed for the SCS and is now marketed by Bausch + Lomb in the United States.
The ODYSSEY study was a phase 2 clinical trial program with 60 patients total and 40 patients who were randomized to axitinib CLS-AX. The control arm had 20 patients who received 2mg aflibercept (Eylea, Regeneron) on-label every 8 weeks after a series of loading doses. The primary objective was to stabilize best corrected visual acuity (BCVA) at week 36.
Each of the TKI programs have run their early-stage trials in a slightly different patient population. I will highlight two important features about the ODYSSEY trial design. First, this patient population was previously treated, but they had to have reading-center confirmed active disease to be included in the study. The mean duration of diagnosis was approximately 9 to 10 months into their wet AMD treatment journey—earlier than other TKI studies—and, on average, patients had received an annualized 10 injections before enrolling. So this was a high-need, still active patient population.
The second aspect of this study that stands out is that it gave us data on the safety and efficacy of redosing a TKI. In ODYSSEY, if a patient developed disease activity, they could be redosed with CLS-AX as long as the redosing occurred at least 12 weeks after the last injection of CLS-AX. In other TKI studies, aflibercept was used for rescue. In ODYSSEY, if a subject made it to 6 months without needing any rescue or redosing, they all received a second CLS-AX injection. Almost all the patients got redosed with CLS-AX.
CLS-AX had stable BCVA over 36 weeks. At the two key timepoints, which were week 24 and 36, they were within two letters of baseline, comparable to patients who received on-label aflibercept.
As far as the OCT data, looking at the central subfield thickness (CST), we saw a typical sawtooth pattern in the aflibercept control eyes and stable disease control in the CLS-AX eyes. At week 24 vs the aflibercept control eyes, there was approximately a 30 μm difference in CST compared with CLS-AX. By week 36, we saw a 5 μm difference, which was not statistically significantly different.
The whole point of these TKI programs, including ODYSSEY, is determining how many patients can go an extended period without needing re-injection. whether that is rescue with an anti-VEGF, or re-injection with CLS-AX. In ODYSSEY, 100% of the patients made it to 12 weeks without redosing of CLS-AX or rescue with aflibercept. At 6 months, 67% of patients were able to be injection-free, including a couple of patients who did not meet the protocol's definition for disease activity requiring retreatment. If you just look at the patients who met the disease activity assessments—which were either change in vision or change in CST, according to the reading center—almost 80% of patients were able to go out to 24 weeks without needing re-injection.
What was the rescue criteria for CLS-AX?
The rescue criteria were a reduction of five letters with an increase of 75 μm in CST, a reduction in 10 letters, or an increase in 100 μm, or the presence of new or worsening vision-threatening hemorrhage. This is slightly different to other TKI studies. In particular on the 100 μm increase, in ODYSSEY, this criteria only needed to occur once, and triggered redosing whilst in other studies, sometimes this criteria was not included or needed to be present in two consecutive visits before rescue. Every study has its nuances, which is why we know we should never cross trial compare.
Did this study go past 36 weeks? Was there an extension portion or was it terminated then? Was it extended for safety?
The study ended at 36 weeks. Importantly, there were no drug or procedure-related ocular severe adverse events, and no endophthalmitis or retinal vasculitis. There were four cases of intraocular inflammation that were all deemed clinically mild by the Safety Review committee. Two of those cases were thought to be related to the suprachoroidal injection technique–perhaps the result of drug particulates that entered the vitreous–and two cases had minimal clinical signs that resolved within a week. No patients had inflammation by the end of the study.
Were they all handled with topical steroids?
If I recall correctly, one patient was treated with an oral steroid, but it was up to the local investigator to determine how to handle it.
Was the quadrant in which these patients had their injection administered mandated?
I believe it was always inferotemporal in this patient population.
Given the lessons learned from the phase 2 trial, is a phase 3 trial planned and how will that be designed?
Yes, Clearside is planning to move CLS-AX into phase 3 pivotal studies because it showed stable BCVA out to 36 weeks in a difficult-to-treat wet AMD population, along with compelling injection-free rates up to 6 months and a relatively clean safety profile. CLS-AX is the only TKI program that has been able to show efficacy and safety on redosing, and I think all those factors decrease risk in phase three development.
As you mentioned, several TKIs are in clinical trials and development, including another trial using axitinib. In 2 or 3 years, when you have a broader toolbox, how do you foresee CLS-AX playing into your treatment armamentarium vs other TKIs?
That is the million-dollar question. This study had 60 eyes; I will plan to wait until we see more data across hundreds of eyes before daring to dip my toes into the fraught world of cross-trial comparison. That said, ultimately, I think the role for TKIs will be maintenance therapy. I still think we will induce our new wet AMD patients with intravitreal anti-VEGF agents, and then we will use a long-acting TKI for maintenance. Many patients will be able to go 6 months or longer without needing supplementary therapy, but I suspect approximately half will periodically need a booster, so to speak, with an anti-VEGF. I think we will end up mixing and matching to manage our patients with wet AMD. Interestingly, if Clearside mimics in phase 3 what they did in phase 2, CLS-AX would be able to be redosed as soon as every 12 weeks, rather than needing to rescue with a biologic anti-VEGF.
Dr. Goldberg, it is always great to have you on Retina Minute. Given the fact that CLS-AX has shown a good safety profile on redosing so far, I think it is a promising potential solution for our patients. Thank you so much for the time and I am excited to see the next phase in this program.
