Schering-Plough recently announced the launch of a major study pitting their pegylated interferon, Peg-Intron, against Roche’s Pegasys. The study is dubbed IDEAL, an inventive acronym for “Individualized Dosing Efficacy vs. flat dosing to Assess optimaL pegylated interferon therapy.” IDEAL will enroll 2,880 adults with hepatitis C at about 100 sites across the United States (see IDEAL web site for study locations). To be eligible for the study, you have to have hepatitis C genotype 1 (the strain most common in the U.S., but least responsive to treatment). Only people who have never been treated before will be included (see inclusion and exclusion criteria).
The study requires that you be “free from substance abuse for the past 2 years” – a more conservative criteria than those recommended in the AASLD clinical practice guidelines [PDF] and the 2002 NIH consensus statement. The study also excludes people co-infected with HIV, who generally do not respond as well overall to hepatitis C treatment.
People enrolled in the study will be randomly assigned to one of three arms, each providing 48 weeks of treatment (study drugs provided for free):
Arm 1: high-dose Peg-Intron (1.5 μg/kg/week) with weight-based ribavirin (Schering’s Rebetol, at 800-1,400 mg/day) – 960 study participants
Arm 2: low-dose Peg-Intron (1.0 μg/kg/week) with weight-based ribavirin (Schering’s Rebetol, at 800-1,400 mg/day) – 960 study participants
Arm 3: Pegasys (180 μg/week) with weight-based ribavirin (Roche’s Copegus, at 1,000-1,200 mg/day) – 960 study participants
The study is designed to see whether there are differences between the three arms in the proportion of study participants achieving a sustained virologic response (SVR), defined as an undetectable hepatitis C viral load 6 months after the end of treatment.
So — assuming that you’re clean and sober and HIV-negative — should you rush off to sign up for this trial? First, it’s worth understanding the reasons for the three-arm design.
IDEAL is billed as a head-to-head comparison of Peg-Intron and Pegasys, but it’s really two studies in one. On the one hand, Schering is comparing Peg-Intron to Pegasys. But at the same time, Schering is using IDEAL to compare low-dose (1.0 μg/kg/week) to high-dose (1.5 μg/kg/week) Peg-Intron. When the Food and Drug Administration (FDA) approved Peg-Intron, they based their decision on study data submitted by Schering-Plough. Based on that data, they approved combination therapy with the dose of Peg-Intron at 1.5 μg/kg/week and the dose of ribavirin at 800 mg/day. But they approved Peg-Intron, used alone, at 1.0 μg/kg/week. At that time, the data suggested that when used without ribavirin, higher-dose Peg-Intron (1.5 μg/kg/week) wasn’t any more effective than lower-dose Peg-Intron (1.0 μg/kg/week).
One study, led by Dr. Steven Flamm of Northwestern University, has been looking at differences in SVR between 246 people receiving either lower-dose and higher-dose Peg-Intron (1.0 vs. 1.5 μg/kg/week) in combination with 800-1,400 mg/day of ribavirin. Preliminary results reported a year ago on a subset of study participants found no differences in early virologic response at 24 weeks. The lower-dose group saw a trend (not statistically significant) towards fewer side effects. If lower-dose Peg-Intron were equally effective with less side effects, that would be good news for patients. But final data has not been reported, so it’s not yet known whether people receiving lower-dose Peg-Intron will as likely to achieve an SVR. It’s also not clear whether the Flamm study will be able to assess differences in SVR rates by hepatitis C genotype—the study includes people with genotypes 2 and 3, which are much more responsive to treatment. In the meantime, doctors using Peg-Intron continue to prescribe it at 1.5 μg/kg/week.
As a condition of FDA approval, Schering agreed to conduct another study comparing low-dose and high-dose Peg-Intron in combination with ribavirin for people with genotype 1 – with targeted dates of full enrollment by June 2003 and completion of the study by December 2004 (see FDA approval letter with postmarketing commitments). IDEAL is that study, and it’s obviously way behind schedule. Indeed, Schering first announced IDEAL in September 2003, and planned to begin enrollment by January 2004. Better late than never?
The other question posed by IDEAL’s design is why different ribavirin doses are used. When Peg-Intron was approved, there wasn’t enough data to support weight-based dosing of ribavirin. The big combination therapy study submitted to the FDA combined high-dose Peg-Intron with low-dose ribavirin (800 mg/day) and compared it with standard (non-pegylated) interferon combined with weight-based ribavirin (1,000-1,200 mg/day) -- a third arm treated used high-dose Peg-Intron for the first four weeks followed by very low-dose Peg-Intron (0.5 μg/kg/week), both combined with weight-based ribavirin (1,000-1,200 mg/day).
Since then, research has increasingly pointed towards the importance of higher doses of ribavirin in achieving an SVR for people with genotype 1. Doctors are typically prescribing Peg-Intron with weight-based dosing of ribavirin, even though the Peg-Intron label states that the approved dose for combination therapy is 800 mg of ribavirin regardless of weight. People weighing less than 75 kg (about 165 lbs.) are prescribed 1,000 mg ribavirin, while those weighing 75 kg or more get 1,200 mg. This is the same dosing scheme approved and used for Pegasys. In theory, the range of ribavirin dose could extend as low as 800 mg/day (for people weighing less than 65 kg, or about 143 lbs.) and as high as 1,400 mg/day (for people weighing over 105 kg, or about 231 lbs.).
The potential clinical relevance of weight-based ribavirin dosing is being investigated in the Schering-sponsored WIN-R study (Weight-Based Dosing of Interferon and Ribavirin). WIN-R compares low-dose ribavirin (800 mg/day) to weight-based dosing (800-1,400 mg/day), both in combination with high-dose Peg-Intron (1.5 μg/kg/week). With nearly 5,000 enrolled participants, WIN-R is the largest hepatitis C study ever conducted.
WIN-R was developed to fulfill Schering’s postmarketing commitments to the FDA. Schering had requested that the FDA approve Peg-Intron/ribavirin combination therapy with weight-based dosing for ribavirin, but the FDA found that Schering had submitted insufficient data in support of that request. Therefore, the FDA required that Schering conduct an additional study of the safety and efficacy of weight-based ribavirin dosing in combination with Peg-Intron. The study was originally designed for 3,000 patients, scheduled to be completed by May 2003. The study size was expanded to incorporate additional FDA requirements that Schering study the appropriate length of treatment (6 months vs. 12 months) for people with hepatitis C genotype 2 or 3. Schering opted to incorporate the genotype 2/3 study into WIN-R, presumably in order save money by utilizing the existing WIN-R clinical trial apparatus (study sites with standard data collection procedures, trained health care staff, etc.). This has extended the time needed to complete the WIN-R study and conduct the data analysis.
As a result, final WIN-R data has not been released. Preliminary data shows that ribavirin-induced anemia occurs more frequently with higher (weight-based) doses. When IDEAL was designed, the FDA (but not the study investigators) did look at available WIN-R data and apparently concluded that weight-based ribavirin dosing was preferable to low-dose ribavirin – otherwise, the FDA could have required that Schering use the 800 mg dose in the IDEAL study.
Note the difference in the range of ribavirin doses between the Peg-Intron (800-1,400 mg of ribavirin) and Pegasys (1,000-1,200 of ribavirin) arms. Schering's logic apparently was that the IDEAL study could only use the doses of ribavirin that the FDA approved for use with Pegasys. If there’s a benefit to extending the range to 800-1,400 mg, that might put the Pegasys arm at a disadvantage. In theory, very low-weight IDEAL participants could experience more ribavirin-related side effects in the Pegasys arm, while very high-weight IDEAL participants could get better results from the 1,400 mg dose in the Peg-Intron arms. It’s hard to know how much of a difference extending the ribavirin dose range would make. Schering presumably would say that the number of IDEAL participants that would fall into the very low or very high weight categories is likely to be too small to matter.
So why should people sign up for IDEAL? The study offers high quality care, with free drugs. And it could answer a very important and relevant question – which version of pegylated interferon works better? But there’s a downside – what if low-dose Peg-Intron really isn’t as good as high-dose Peg-Intron? One third of people in the study will get low-dose Peg-Intron. And if Schering had separated out the two goals – studying low-dose Peg-Intron vs. high-dose Peg-Intron, and comparing Peg-Intron to Pegasys – they would have needed less people in the low-dose vs. high-dose arm (perhaps about 800 in each). Of course, it’s less expensive for Schering to do one big study than two medium-sized studies. But is it better for people with hepatitis C?
Let’s say that low-dose Peg-Intron isn’t as good in combination therapy – we don’t know that yet, but it’s a reasonable hypothesis. That means that an extra 160 people will receive suboptimal therapy in the IDEAL study. Let’s go further and speculate that low-dose Peg-Intron is 10-20% less effective at achieving an SVR than high-dose Peg-Intron, keeping in mind that these numbers are a wild guess since there’s no data. That translates to maybe 50-100 people who don’t get an SVR because they were assigned to the low-dose Peg-Intron arm of IDEAL instead of the high-dose arm. And that’s 10-20 more people failing treatment than would happen if Schering did a separate study comparing low- and high-dose Peg-Intron, rather than bundling it in with the Pegasys comparison. That's 10-20 people experiencing all the side effects that go along with treatment, without the benefit of an SVR. That’s not a lot of people – less than 1% of the overall IDEAL study participants – but it makes a big difference if you’re one of those people.
So perhaps IDEAL isn’t ideal for everybody. What’s in it for Schering-Plough? Maybe they really think that Peg-Intron is better than Pegasys, and now they want to prove it. In an article published last October on Hepatitis Neighborhood (free registration required), Schering spokesperson Robert Consalvo was quoted saying “I think there's been some perception in the marketplace—and we think it's a misperception—that all things are basically equivalent, whether it's PEG-Intron or Pegasys…. We don't really agree with that.”
That sentiment is certainly understandable, even without clinical data to back it up. Since the introduction of Pegasys, Roche’s share of the hepatitis C market has surpassed that of Schering. Whether this reflects doctor and/or patient preferences or the success of Roche’s marketing department is anyone’s guess. But one thing’s clear – the beleaguered Schering has become increasingly dependent on sales from Peg-Intron, especially since its Claritin franchise (which formerly provided the largest chunk of Schering’s profits) faced a dramatically decline in revenues when it moved from prescription to over-the-counter status.
Schering is still struggling to recover and rebuild after the last few years, having also faced numerous regulatory setbacks from the FDA and SEC, as well as various investigations on sales and marketing practices from a number of state attorney generals. That makes their hepatitis C franchise more important than ever. But they’ve appeared relatively powerless in the face of the market gains of Pegasys. "Roche is eating Schering's lunch and breakfast and dinner and dessert…. I think Schering-Plough got complacent about its former dominance of the interferon market,” said market analyst Michael Krensavage, quoted last September in a Forbes article about the IDEAL study.
To borrow a legal term, Schering’s IDEAL study seems designed to sow “reasonable doubt” among patients and clinicians about the relative equivalence of Peg-Intron and Pegasys – a move that could stem further erosion of Peg-Intron market share. And Schering has guaranteed that during the course of the IDEAL study, leading clinicians across the country will continue to use Peg-Intron. So even if IDEAL eventually concludes that there’s no difference between Peg-Intron and Pegasys – or even that Pegasys is superior – Schering will experience some short-term benefit just from conducting the study.
I don’t have a crystal ball, and final SVR results from IDEAL won’t be available for years to come. But here’s my hypothesis: I’d guess that IDEAL won’t find any overall differences between Pegasys and Peg-Intron. But there might be differences between the drugs for people with both genotype 1 and high hepatitis C viral loads. Available published data could suggest that Pegasys is more effective in people with genotype 1 and high viral loads – but it’s hard to base conclusions on studies that don’t make direct comparisons. Alternately, maybe Schering may be correct by emphasizing the importance of weight-based dosing for both Peg-Intron and ribavirin – unlike Pegasys, Peg-Intron uses a weight-based scheme for dosing. But one thing is clear – the question of which pegylated interferon to use is most pressing for people with genotype 1 and high viral loads.
The problem is, we may never know the answer – IDEAL will stratify its SVR results by baseline hepatitis C viral load (low vs. high), but Schering reportedly doesn’t expect the results to be statistically significant unless the differences in SVR rates between arms are very large. So maybe it doesn’t matter which drug you use if you have a low viral load, but IDEAL probably won’t tell us definitively if it matters for people with high viral loads.
A 'real IDEAL’ would help people with genotype 1 and high viral loads – the hardest to treat – decide which drug to use. From the looks of things, that’s not Schering’s study. Maybe Roche would be interested…?
[My thinking about IDEAL has been informed and enriched by dialogues with Tracy Swan, Brian Klein, and Jules Levin – though I take sole responsibility for the discussion above. But see “The head-to-head that wasn't” written last December by Tracy Swan from the Treatment Action Group for more background and some similar arguments.]