The Consequences of CATT: Will It Change Your Practice?

Some say it validates Avastin use, while others say it reinforces the importance of on-label indications

The Consequences of CATT: Will It Change Your Practice?

Some say it validates Avastin use, while others say it reinforces the importance of on-label indications.

Frank Celia, Contributing Editor

When results from the CATT trial were made public in May at ARVO, lead investigator Daniel Martin, MD, directly addressed the widely held belief that cost considerations inspired the study. “Nothing could be further from the truth,” he explained to an audience of thousands, because when the protocols were designed “we didn't even know what the cost of Lucentis would be.” True as those words certainly are, it would be impossible to separate human factors from the practice of medicine. Patient convenience and access to care, the divergent financial goals of the drug and insurance industries, and physicians' pride in the autonomy of their clinical decision-making have all played a part in shaping treatment patterns for AMD. Indeed, upon hearing Dr. Martin state his unequivocal belief in first-line bevacizumab therapy's safety and efficacy, the ARVO audience erupted in thunderous applause.

As it stands, the 2010 ASRS Preferences and Trends study found that respondents opt for Avastin over Lucentis by a factor of nearly two to one. When a profession favors an off-label reformulation of a systemic therapy over one expressly designed and approved for ocular use, clinical management will necessarily be fraught with compromises and trade-offs. While CATT offers invaluable data on a disease that until recently was devastating, it's unlikely to impact established practice patterns in a significant way. There's too much ammunition for both sides of its twin debates over product selection and dosing. Moreover, in the time these issues would take to resolve, newer, better therapies could make them obsolete. So perhaps the most noteworthy aspect of CATT is that it exists at all. A prominent retinal physician suggested that, had the financial model introduced for Lucentis been different, perhaps ophthalmology would never have maintained the drive to finish CATT.1

This article endeavors to assay the various CATT-inspired debates, clinical and otherwise, with an emphasis on subjects that have received little or no attention in public forums.


Clinically speaking, the most relevant finding of the Comparison of AMD Treatments Trial was proving bevacizumab to be as effective as ranibizumab at preserving and improving visual acuity.2 Though not powered to assess bevacizumab's systemic safety, the study measured adverse events as a secondary endpoint and found serious risks failed to emerge statistically among 1,200 patients treated for one year.

Divided into four treatment groups, the randomized patients received ranibizumab monthly or PRN, and bevacizumab monthly or PRN. In terms of visual acuity, monthly injections achieved parity at 52 weeks (bevacizumab yielding a +8.0 letter gain and ranibizumab a +8.5 letter gain). Also, monthly ranibizumab and PRN ranibizumab achieved parity (+8.5 letters and +6.8 letters, respectively), as did monthly bevacizumab and PRN ranibizumab (Figure 1). However, comparisons between monthly bevacizumab and PRN bevacizumab (+8.0 letters and +5.9 letters, respectively) proved inconclusive, as did comparisons between monthly ranibizumab and PRN bevacizumab. In these last two comparisons, results overlapped what the study deemed “non-inferiority” (no difference) and “inferiority” (five-letter loss) and thus no firm conclusions could be drawn. Despite these mild statistical caveats, the study authors conclude, “The mean number of letters gained, the proportion of patients in whom visual acuity was maintained (<15 letters lost), and the proportion of those who had a gain of at least 15 letters were nearly the same for each drug when the regimen was the same.”2

Aside from visual acuity, OCT outcomes have garnered the most attention. Retinal physicians have expressed surprise at the extent to which Lucentis outperformed Avastin in drying and thinning retinas. Maculas receiving Lucentis were thinner by a margin of about 32 µm, a statistically significant difference. Monthly Lucentis also produced the highest percentage of retinas with no trace of fluid at one year, as well as the highest percentage of retinas with no leakage seen on fluorescein angiography. Though none of this impacted visual acuity, such discrepancies could threaten vision over the long term, physicians say. Great interest surrounds whether two-year acuity scores will be affected. Expect those results at next year's ARVO conference.

The bevacizumab safety issue also remains unsettled. Though CATT did not link off-label use of the drug with any thromboembolic events, the Avastin group showed a higher incidence of “serious adverse events,” particularly hospitalizations (bevacizumab, 24%; ranibizumab, 19%). Genentech highlighted this point as evidence of ranibizumab's superiority, releasing a statement warning of an “emerging body of evidence” indicating Avastin patients may be at risk.

The company also pointed to a study it sponsored at Johns Hopkins University, unveiled at the same ARVO meeting, in which a retrospective review of 77,000 Medicare patients who were treated with either Lucentis or Avastin found Avastin patients showed an 11% higher all-cause death rate, and a 47% higher risk of hemorrhagic stroke.3 However, that study did not control for confounding variables in the systemic health profiles of patients. “I really view this study as hypothesis-generating,” said Emily Gower, MD, in her ARVO presentation, “as opposed to formulating final opinions on this analysis.”

Regarding serious adverse event data, CATT researchers cited several mitigating factors. First, patients in the Avastin arm were older and had higher rates of hypertension, diabetes and smoking. Second, the PRN Avastin group suffered more adverse events than the monthly group, seeming to argue against a causal link. Finally, most of the events had nothing do to with the sort of vascular problems associated with the higher doses of Avastin used in cancer treatment. Nevertheless, all parties agree that the safety question remains open. Unfortunately, to resolve it completely would take a lengthy, prospective trial involving tens of thousands of patients, researchers say. And with the arrival of next-generation therapies imminent (VEGF Trap-Eye could be approved in August), no plans exist for such a study.


In short, ophthalmology's most visible and heated drug rivalry—an internecine one at that—will not soon end. The Lucentis and Avastin camps can each draw support from the CATT data, as can advocates of either dosing regimen studied. In addition to ongoing safety concerns, ranibizumab's greater drying ability will no doubt influence decisions.

But to consider these drugs interchangeable would also be false, according to retinal subspecialists. Anecdotal reports abound of patients achieving an initial response with one drug, which then wears off, only to be revived by substituting the other drug. “The significance of anecdotal reports of Avastin nonresponders—defined as those with persistent fluid—responding when switched to Lucentis, and vice-versa, is unclear,” says David Williams, MD, MBA, former president of the American Society of Retina Specialists. “I've certainly seen this myself in both directions, but I've also seen no incremental response when a nonresponder to one drug is switched to another. I've also seen apparent nonresponder eyes with unchanging chronic fluid over many months unexpectedly show a decrease in residual fluid with ongoing same-drug treatment.”

One theory posits that nonresponders are not failing to respond, but rather failing to respond long enough. “I believe real resistance is very rare,” says Michael Singer, MD, of San Antonio. “The problem is that we look at these patients at four weeks. Maybe if we saw them at week two, there wouldn't be as much edema.” VEGF is a positive feedback loop, he observes. “The more we suppress it, the more the body produces.” Often such patients respond favorably to a larger dose of either drug, Dr. Singer notes.

Another factor little remarked upon concerns pharmaceutical compounding. The freshly made bevacizumab prescribed in CATT arrived in glass vials under nitrogen, concealed in a box that protected against light degradation—high standards not always achievable in everyday clinical practice. “My bevacizumab comes from the ‘low price guy,’ in a plastic syringe in a zip-top plastic bag delivered in Houston's 100-degree heat—in all likelihood on the dashboard of a Buick,” grouses David Brown, MD, of Houston, in a recent article.4 He speculates that this might explain the variability he sees in many of his Avastin patients. One recent study found compounding discrepancies that could affect the drug's efficacy and safety; some compounding pharmacies deliver product with up to 50% less active drug.5

Though many expect CATT to foster greater emphasis on cost in decision-making, a countervailing force exists in the Medicare Modernization Act of 2003, which permits a 6% markup on the average sales price of Medicare Part B drugs. “There is a profit made by prescribing Lucentis,” notes Rahul Khurana, MD, of Northern California Retina Associates. “Although not large, with the drastic cuts in reimbursement this year for injections and OCT, it is playing a larger role than physicians would care to admit.”

Many physicians will continue to rely on a case-by-case approach, taking financial and patient history factors into account. Says Daniel Kiernan, MD, a vitreoretinal fellow who will be joining Ophthalmic Consultants of Long Island this July, “If one of my patients had a recent heart attack or a stroke and their vision was at risk due to active CNV, I would prefer to use Lucentis unless the financial burden was serious.”

It remains unclear how the insurance industry will respond to CATT. Carriers could use the study as grounds to discontinue covering Lucentis. In markets outside the United States, where physicians enjoy less prescribing freedom and Avastin is sometimes prohibited, CATT is expected to yield a more potent pro-Avastin impact.6


The CATT data confirm what the phase 3 studies of ranibizumab concluded: a monthly dosing schedule provides better visual outcomes than PRN. In CATT, this difference meant about two additional letters of vision at one year—admittedly, not large on a 60-letter vision scale. However, the worse (though not statistically significant) performance of as-needed Avastin when compared with the monthly regimens of both drugs worries some physicians, especially in light of higher incidences of fluid in these patients and small but statistically significant growth in lesion sizes among PRN patients. A slight divergence of the PRN visual acuity curves from the monthly groups' at 36 weeks has caused concern about how PRN vision will fare at the two-year mark.

“People who are prescribing PRN Avastin should take a long, hard look at the CATT data,” cautions Dr. Brown. As-needed dosing never made sense to him, he says, because it permits too many recurrences of retinal swelling. The repeated bouts of edema inherent in PRN regimens amount to tissue insults that over time can atrophy retina layers. “That is why we often see a very thin retina after deturgescence in DME or CRVO. It is never healthy to get repeated occurrences of edema. I call PRN dosage ‘progressive retinal neglect.’”

On the other hand, the benefits of a monthly regimen must be weighed against requiring elderly patients to visit an office every single month with no endpoint in sight. Because family members usually transport these patients, long-term adherence to a strict monthly schedule can be burdensome for working adults. This explains the desire to extend intervals between treatments whenever possible. The PrONTO study, which achieved favorable outcomes among patients receiving about 10 injections over two years, convinced many practitioners that variable dosing regimens could work.7

It is also possible that CATT researchers, misreading Avastin PRN OCT data, failed to green-light enough treatments; independent reading centers identified retinal fluid in 25% of Avastin PRN cases that had been cleared by CATT study sites. But considering Avastin PRN patients already received about eight injections each, additional treatments might undermine the whole purpose of PRN dosing, ie, fewer than 12 injections a year. Complicating matters further, the CATT PRN protocols far surpassed the level of care patients usually receive outside clinical trials. For starters, CATT subjects underwent examinations every 28 days, again seemingly at odds with PRN ideals.

The CATT study might convince some practitioners to switch to monthly schedules or to monitor patients with greater care. But in reality most anti-VEGF patients receive neither PRN nor monthly dosing. Rather, the regimen most favored is the “treat-and-extend” approach: patients receive three or four monthly loading doses, then return at five-week intervals. When stability is achieved, intervals are extended to six weeks. If stable at six weeks, intervals are extended to seven, etc., until edema and fluid appear, which then becomes the de facto treatment interval. CATT did not study treat-and-extend approaches, and with no large-scale trials proposing to do so, it appears the PRN vs. monthly debate will remain for the time being mostly academic.


A complete, behind-the-scenes version of how CATT came to fruition has yet to be told. But the research team published an account of its long, harrowing regulatory and funding struggle last year.8

Many in the retinal community cite the pricing of Lucentis as the foundation of widespread hard feelings that persist to this day. Had the drug's cost been nearer to Macugen, the first anti-VEGF drug brought to market, a predilection for Avastin may have never happened.

After off-label Avastin took off, and Genentech announced plans in 2007 to limit Avastin's availability at compounding pharmacies, tensions escalated further. At a heated AAO conference session that year, ophthalmologists accused the company of depriving lower-income patients of vision-saving drugs. Genentech eventually retracted its decision, but the experience made a lasting impression.

To many, CATT stands as a milestone in eyecare history and perhaps even a clarion call to a medical profession losing independence in the face of growing drug industry power and government reform schemes. New York ophthalmologist Rick Spaide, MD, called this aspect of CATT more important than its clinical findings: “What I find most inspiring was the drive behind the research—the desire by physicians to deliver good medical care at a fair price, to set up a well-constructed trial, to get funding, and finally to devote incredible amounts of time to the study.”1 RP


1. Spaide RS. Commentary: How will CATT affect practice patterns? American Society of Retina Specialists Web site. Accessed May 17, 2011.
2. Martin DF, Maguire MG, Fine SL, Jaffe GJ. Ranibizumab and bevacizumab for neovascular age-related macular degeneration. N Engl J Med. 2011;1897-1908.
3. Gower EW, Cassard S, Chu L, Varma R, Klein R. Adverse event rates following intravitreal injection of Avastin or Lucentis for treating age-related macular degeneration. Paper presented at: Annual meeting of the Association for Research in Vision and Ophthalmology; Fort Lauderdale, FL; April 22, 2011.
4. Brown DM. Commentary: All Avastin might not be created equal. American Society of Retina Specialists Web site. Accessed May 17, 2011.
5. Kahook MY, Liu L, Ruzycki P, et al. High-molecular-weight aggregates in repackaged bevacizumab. Retina. 2010;30:887-892.
6. Rosenfeld PJ. Commentary: CATT results have not influenced my decision making process. American Society of Retina Specialists Web site. Accessed May 17, 2011.
7. Lalwani GA, Rosenfeld PJ, Fung AE, et al. A variable-dosing regimen with intravitreal ranibizumab for neovascular age-related macular degeneration: year 2 of the PrONTO Study. Am J Ophthalmol. 2009;148:43-58.
8. Martin DF, Maguire MG, Fine SL. Identifying and eliminating the roadblocks to comparative-effectiveness research. N Engl J Med. 2010;363:105-107.

CATT's Impact on Other Diseases
Since the inception of the anti-VEGF era, these agents have been used to treat other retinal disorders besides AMD, most notably retinal vein occlusion and diabetic macular edema. Because both RVO and DME involve higher levels of VEGF release than AMD (Figure 2), some theorize Avastin's slightly weaker ability to dry the retina seen in CATT might translate into greater visual deficiencies among RVO and DME patients treated with the off-label drug.

“CATT did show a statistically significant difference between the two drugs in AMD. This difference may be magnified in RVO and DME, where VEGF levels are higher,” says Michael Singer, MD. Retrospective studies have shown that patients whose retinas remain dry progress to better visual outcomes when compared with those who experience recurrent edema and fluid throughout treatment, he added. “I would stick with Lucentis for RVO and DME. Plus, diabetic patients tend to be sicker, so they might have a higher risk of adverse events with Avastin.”
Still, all three are different diseases, and the CATT data cannot be extrapolated comprehensively to the other two. “Although VEGF is involved in both DME and AMD, there are other molecules and cytokines likely involved in the pathogenesis of DME,” explains Diana Do, MD, assistant professor of ophthalmology at the Wilmer Eye Institute. Both drugs have proved effective at fighting DME and RVO in smaller scale trials, but short of head-to-head comparisons, neither can be considered superior to the other.