Seroquel Lessons: Raised Eyebrows and Furrowed Brows

The very interesting Washington Post front page story on March 4 about an unfavorable Seroquel study that may have been “buried” during clinical development ends with a comment from the principal investigator of the CATIE trial, the large NIH comparative trial of antipsychotics.

Post writer Shankar Vedantam says “Jeffrey Lieberman, a Columbia University psychiatrist who led the federal study, said doctors missed clues in evaluating antipsychotics such as Seroquel. If a doctor had known about Study 15, he added, ‘it would raise your eyebrows.’"

That’s an intriguing observation, especially since Lieberman was not only the lead investigator for the NIH study but also for a smaller comparative study commissioned by Seroquel manufacturer AstraZeneca which was conducted during the course of the large NIH comparative project.

As we wrote at the time of the first CATIE results in 2005, AstraZeneca was one of the most aggressive participants in the atypical antipsychotic field in its preparations for the release of comparative information from CATIE.

During 2003, while CATIE was already underway, AstraZeneca commissioned a 52-week comparative trial of Seroquel, Lilly's Zyprexa and J&J's Risperdal in first-episode psychosis, a trial that closely paralleled CATIE but compared a response in a category of patients that was excluded from the NIH trial.

Called CAFÉ (Comparison of Atypicals in First Episode Psychosis), AstraZeneca’s trial even echoed CATIE’s name. The two studies shared about 14 of the sites included in CATIE.

CAFÉ may have added information about another group of patients but it also improved the results for Seroquel noticeably and gave AstraZeneca something positive to say when the CATIE results were released.

The recent Post story raises issues about studies that might not have been reported to keep unfavorable information from the light of day.

The real world is even more complicated. As the CATIE/CAFÉ situation reminds, studies can be designed and run by similar teams that make it difficult to parse a useful meaning from the research. And that's an important message for the future.

During a period while the policy and budget communities in Washington are rallying around comparative effectiveness as a way to align health care spending with the most sensible and cost effective treatments, the CATIE experience is a useful warning of a tough road ahead for that effort.

Previous experience presages a particularly difficult path ahead for the advocates of comparative research as an educational tool – what could be called the Baucus school of comparative effectiveness.

The way comparative trials have been run to this point suggests there will be plenty of furrowed, confused brows in the future as well as raised eyebrows.

Apixaban: Bad News Highlights Good Strategy

The delay announced yesterday to Pfizer and BMS's apixaban blood clot prevention therapy is certainly a setback for the two pharmas, though they presented the news--and the market reacted--relatively smoothly.

Apixaban--pitted here in a head-to-head against current standard Lovenox, from Sanofi-Aventis, for the prevention of venous thromboembolism in knee replacement surgery patients--just barely failed to show non-inferiority, thanks to a much better than expected result for Sanofi's drug (see the companies' release for the statistical details). This was the first of eight Phase III trials planned for the compound, and ongoing studies aren't affected by yesterdays news, BMS apixaban program head Jack Lawrence told Reuters.

But still, the failure will add at least several months or more to the drug's development program and with rivaroxaban from Bayer/J&J already a half-step ahead, every little bit helps. An NDA for the compound, an oral Factor Xa inhibitor, won't be filed in 2009 as planned.

All of which validates Bristol's decision to go halvesies on the apixaban program with Pfizer in the first place, in a deal we've been fans of since it was announced back in April 2007.

To reacquaint you with the terms: in exchange for $250 million upfront cash and up to $750 million in development and regulatory milestones Pfizer gets an equal share of profits and foots an equal share of commercialization expenses, and Pfizer will fund 60% of any development costs from January 1, 2007 onward. (BMS's similarly risk-sharing and lucrative pact with AZ in the diabetes space provoked our admiration only a few months earlier.)

So with each setback, Bristol's strategy--sharing risk on important projects and increasingly biotech-like--seems cannier. That said, it's fair to ask just how many snafus Bristol's reputation can take before those nine-figure upfront payments dry up: after all, its other big risk-hedging adventure was on the now-dead muraglitazar PPAR deal with Merck & Co. back in 2004, which brought in $100 million upfront and at least $55 million in milestones before stumbling at FDA.

And it's not only Bristol among Big Pharma that's spreading the risk--and possibly the massive reward--on late stage development projects. We'll have more to say in the next IN VIVO about Eli Lilly's $300 million financing deal with TPG-Axon and Quintiles' NovaQuest on its two lead (Phase III) Alzheimer's disease treatments.

And who can forget our multiple pleas for your input on Amgen's potential future deal for denosumab? Certainly the project is now less risky/more expensive since Phase III results in its post-menopausal osteoporosis study were positive, but as Amgen's most important pipeline product in years perhaps (some day) the biggest sign that at least a few Big Pharma (and Big Biotech) now think differently about clinical and commercial risk.

TauRx’s Alzheimer’s Drug: A Web of Tangles

By the end of last Tuesday’s midday press briefing at the Alzheimer’s Association’s International Conference on Alzheimer’s Disease, it was clear that the next news cycle would be chock full of stories around mid-Phase trial results of Singapore-based TauRx’s drug rember, a formulation of methylthionium chloride (MTC—the well-known reagent methylene blue), which has been used to treat urinary tract infections and some hemoglobin disorders since before there was a US Food & Drug Administration.

For one thing, it was an otherwise slow news day, at least to that point: the full Phase II results of Elan-Wyeth’s bapineuzumab wouldn’t be until later in the afternoon, and in any event nothing about that drug had been included in the advance press briefing. TauRx had also put out an unusually aggressive press release, claiming among other things “a new treatment that appears to slow the progress of the disorder by 81% over a year…an unprecedented result in the treatment of Alzheimer's disease” and that the Phase II results “strongly suggest that it is possible to halt progression in mild and moderate Alzheimer's.”

So there was the expected headline grabbing: London’s The Independent ran with “Alzheimer’s drug hailed as a ‘major’ development.” The Daily Mail heralded the "biggest breakthrough against disease in 100 years." News of the rember data even reached Kandahar, Afghanistan, TauRx chairman Claude Wischik told me the next day. But gladly, responsible news outlets seemed mostly to grasp that the work was early, and framed it as such.

What may not be apparent, however, is just how early, at least in terms of a US regulatory timeline. Although Wischik said he expects a Phase III trial to begin in the US, EU, Asia, and the Middle East within a year, there’s been no US IND filing yet. (The Phase II was conducted in the UK and Singapore.) Nor have any large species toxicology studies been completed. Wischik expressed some hope that these could be done in parallel with an IND review, and if not, he said, almost as an aside, it’s likely that the US trials will start with a different formulation than the one in the reported Phase II. Presumably, that means at least a bridging study.

In fact, the extent to which rember's safety profile has been vetted is unclear. Plus, treatment at the highest dosage form showed minimal efficacy in the Phase II due to cross-linking of the gelatin capsule shell by the MTC, leading to impaired dissolution/absorption of active drug. Those data were excluded altogether from the results presented at the meeting.

That said, the data are tantalizing because of the approach. In addition to plaques, the presence in the brain of tangles—abnormal fibers of aggregated tau protein--appears to correlate with dementia. Rember is a tau aggregation inhibitor; it appears to work by breaking up the bonds between tau molecules in tangles, allowing them to be cleared by normal cell pathways. The drug produced significant improvement relative to placebo at 24 weeks in moderate AD patients, and stabilized the progression of AD over 50 weeks in both mild and moderate patients. Plus, the effects were supported by SPECT and PET imaging showing evidence of efficacy in brain regions strongly affected by tau pathology.

As a proof of concept of the mechanism of tau inhibition, the TauRx study is significant. Showing that breaking up tangles has an effect on cognition adds credence to the belief that targeting them is a useful approach to treating AD. It also provides an alternative to drug strategies based on the amyloid beta (Abeta) hypothesis--trying to stop synthesis of amyloid beta or dissolving Abeta once plaques form, the basis for passive immunotherapies using antibodies, such as bapineuzumab. Indeed, Abeta immunotherapy appears to also lead to tau clearance in the brain.

Check this space and the upcoming September IN VIVO for more from the ICAD meeting, including putting what happened with Elan-Wyeth’s bapineuzumab in context now that the dust has settled, as well as a look at Baxter’s IVIg and the anti-histamine dimebon, both set to enter Phase III.

Is Partial Agonism the Key to PPAR Success?

In case you haven't had your fill of TZD news this week, we bring you word that Dr Reddy's Labs and Rheoscience have dosed their first patient in what will be the first of several large Phase III studies of their partial PPAR gamma agonist balaglitazone.

Say what?

The timing of the trial's launch missed the Avandia circus by only a day and suggests there are companies out there willing to put in the time and money and Herculean effort to make their case for potentially differentiated products in a problematic class. Some senior industry executives think the odds are stacked against success.

Among those skeptics is Wyeth R&D chief Bob Ruffolo, who when interviewed by the IN VIVO Blog only six weeks or so ago, asked (rhetorically): "Who is going to bring a new, improved version of Avandia to market" given the regulatory mountain the FDA will surely make such a company climb?

And it's a fair question, particularly in light of the past few months' events. Let's review.

• GSK's Avandia, subject of everyone's favorite meta-analysis, though allowed to remain on the market for now, will get slapped with some stern patient- and doctor-scaring warnings. Takeda's Actos has avoided the whirlpool so far.


• The PPAR class hasn't exactly had smooth safety sailing prior to Avandia's snafu. Warner-Lambert's Rezulin kicked off the parade back in 2000 and so-called next-generation dual gamma/alpha PPAR agonists like AZ's Galida and Merck's MK-767 and Merck/BMS' Pargluva eventually followed.
  
• Clinicians and patients may be moving beyond the PPARs in any case. The RPM Report has reviewed the winners/losers: Actos may have seen a bump in the wake of the initial Avandia news, but that hasn't lasted. Merck's first-in-class DPP4 inhibitor Januvia is enjoying wild success in its first year on the market and observers are salivating over potential newcomers like Amylin's once-weekly Byetta and Novo Nordisk's liraglutide.

Nevertheless, Rheoscience presses on. If Actos is now the TZD safety standard bearer, it makes sense that to get to market now companies will have to prove their drugs as safe or safer than Takeda's (in fact, European drug regulators require such an active comparator to prove non-inferiority). Rheoscience is aiming to do just that. The first balaglitazone Phase III will be a European "six-month, double-blinded, randomized, placebo-controlled multicenter trial in which type 2 diabetes patients will be given daily doses of either 10 or 20 mg of balaglitazone versus the active comparator Actos (45mg/day) as an add on to stable insulin treatment," according to Rheoscience.

Our interest piqued, we got in touch with Rheoscience CEO Philip Just Larsen, MD, PhD. (For background on Rheoscience, check out our 2004 profile of the company.) Rheoscience's September 2005 deal with Dr. Reddy's got the Danish biopharma company European and Chinese rights to balaglitazone in exchange for taking responsibility for execution and costs associated with the drug's US/EU Phase III program and US regulatory submission (Rheoscience will also receive an undisclosed milestone payment if an NDA is approved and undisclosed royalties on US sales). Novo Nordisk once held rights to the drug, but returned them to Dr Reddy's in 2004, back when dual PPARs were still expected to dominate the class, says Larsen.

Larsen stresses the benefits of balaglitazone's partial gamma agonist properties. "It doesn't take full agonism for full glycemic control," he argues, pointing out that Phase II studies of the drug as well as preclinical models suggest that 20mg of balaglitazone is comparable to 45mg of Actos in that regard.

What's more, he says, balaglitazone shouldn't raise the same safety concerns as other TZDs, since in vitro models and preclinical trials suggest it causes less fluid retention than Avandia and Actos and is less adipogenic than those drugs. The second Phase III trial for the drug is an international long-ranging study to generate a safety database; again that trial will use Actos as a comparator, but Larsen says he doesn't expect either drug to generate the same cardiac ischemia safety signals that have plagued Avandia.

"It's worth emphasizing that it's not the class as such that is under suspicion for inducing cardiac ischemia," he says. "Actos is not associated with cardiac ischemia," a notion backed by FDA's David Graham during Avandia's recent panel meeting, he says.

Nevertheless the series of trials required to get balaglitazone onto the European and US markets will be very costly (Larsen declines to say exactly how costly). Patients and clinicians may be wary of TZDs and hamper trial enrollment (a suggestion Larsen disputes, particularly in terms of European patient accrual). The future of diabetes treatment may indeed by combination therapy, but whether or not TZDs are a mainstay of those combos remains to be seen as newer drugs like the GLP-1s make progress. The regulatory environment--for all drugs, it sometimes seems--is increasingly tricky. Larsen remains optimistic.

"I'm not worried about the regulatory environment. The advisory panel was in favor of drugs in this class, they're a necessary tool in the toolbox," he says. "Treating diabetics is a challenge and by limiting the options you put yourself in a difficult position, and these [TZD] products are quite efficacious."

As for the class' past failures, he points out that "the entire concept of pursuing the dual agonists was a blind alley, and it took those Phase III trials to show that is the case. That is how the selective PPAR agonists like balaglitazone were revitalized," and now shown that pursuing partial agonism could control the side effects related to full agonists like Avandia.

Rheoscience and Dr. Reddy's have competition. A second partial-PPAR gamma agonist in late stage trials belongs to Metabolex and Johnson & Johnson (we wrote about their innovative deal last summer). But don't expect any winners for quite some time. Balaglitazone's NDA won't be filed before late 2009 at the earliest.

Even Cancer Ain't Exempt

No drug is exempt from FDA’s ultra-conservative stand, it seems, even those, like GPC Biotech’s satraplatin (Orplatna), that allegedly improve survival among cancer patients with few if any treatment alternatives.

An oncology advisory committee (ODAC) yesterday recommended that FDA delay considering prostate cancer candidate satraplatin for approval until overall survival data from the company’s 950-patient SPARC study is available. This recommendation came despite interim analysis strongly supporting the efficacy and safety of satraplatin, which was granted priority FDA review in April 2007.

Still, it wasn’t a complete surprise, at least since the rather negative FDA briefing documents released last week. These raised five “issues” including questions over the company’s assessment of pain progression, and, more importantly, over the definition and composition of one of GPC’s two key primary end-points used in the trial, progression free survival (PFS).

According to GPC’s “greatly disappointed” CEO Bernd Seizinger, GPC had communicated intensively with FDA throughout the application process and amended various aspects of the trials in accordance with FDA suggestions—including suggestions on what pain score to use. In sum, they’d done everything possible to make this work. (Listen for yourselves here.)

But FDA wasn’t having it. According to one analyst, “GPC walked into a complete stitch-up—FDA primed the advisory committee that they didn’t want this to happen.” Reading the briefing document, we’re inclined to agree.

Now, sources tell IN VIVO Blog that there’s no love lost between GPC and FDA—the ODAC meeting was “hostile and antagonistic,” according to one observer. Combine this with FDA’s overall caution (the last thing they want is to be hauled in front of the Senate for injudicious approvals) and you have a recipe for delay, if only temporary.

Still, GPC isn’t just an innocent victim here. Let’s go back to that issue of end-points. The briefing documents state that FDA had all along warned GPC that a PFS end-point may be problematic. Yet GPC failed to communicate this concern to the investment community.

That’s why analysts are so pissed—many had buy ratings on the stock, even after the review documents were released last Thursday. GPC shares fell 35% back then, and about the same again today. “Trust has been lost,” says one in London. “When it comes to the investment community, GPC are on their own.” (They can sulk in a corner beside Sanofi Aventis, which hasn't won any prizes for transparency among the investment community either, particularly with regard to Zimulti's rocky ride, on which we've reported extensively, including here.)

It’s a worrying time for the German biotech. All of a sudden, the overall survival data for satraplatin—which is the other primary end-point, alongside PFS--has become a make-or-break issue for the company. If it’s positive, the drug will probably still make it, although not in an accelerated fashion. If it’s not, the drug may not get onto the US market at all—which could prove fatal for GPC.

The rest of the sector has cause for concern, too, though (not to mention androgen-independent prostate cancer patients that have failed first-line therapy). This bleak little episode—besides showing up personal clashes between applicants and the agency, which, it seems, do matter—may mean that regulators’ caution is spreading beyond primary-care drugs like Avandia or the Cox-2s, where the main concern is safety.

Today, it seems, the tough hurdles can come from anywhere--including from non-safety-related issues like how pain is measured. Some might call this nit-picking.

Sometimes the Bear Gets You: Idenix Pharmaceuticals Edition

Idenix Pharmaceuticals announced this morning that FDA had placed a clinical hold on its Phase II valopicitabine hepatitis C polymerase inhibitor.

FDA made its decision following an assessment of the overall risk benefit profile observed in clinical rials to date, according to Idenix CEO JP Sommadossi, who addressed analysts on a conference call today. "We were very surprised," he said, adding that along with its partner Novartis, Idenix is evaluating its options. "But i am not optimistic about further development of valopicitabine in the future," Sommadossi added, and later re-emphasized.

So what happened? "FDA looked at all of the Phase IIB data ... and concluded that the GI side effect profile was not commensurate with the antiviral activity we were seeing," said Doug Mayers, CMO. Idenix, the executives said, had seen similar signals such as nausea and vomiting, but were eager to play with the doses of valopicitabine in further trials in an attempt to find a workable risk/benefit profile. FDA had different ideas.

Valopicitabine, previously dubbed NM283, was the furthest-along HCV polymerase inhibitor in clinical development and the first of a next-generation set of targeted HCV therapies. The polymerase class hasn't received as much attention as HCV protease inhibitors, which have been the subject of some pretty sizeable deals--Vertex's alliance with Johnson & Johnson and Intermune's alliance with Roche, for example. But Idenix had maintained that the future of HCV antiviral therapy would be combination, a sentiment shared by many, and so pushed ahead with development despite only moderate clinical success with the compound.

The biotech's partner Novartis must have agreed; the companies have a broad alliance whereby Novartis gets first right of refusal on any Idenix compound when it hits Phase II. For valopicitabine, Novartis had to pony up more than half a billion dollars to maintain its share of the drug. Novartis owns upwards of 57% of Idenix.

Today's news is the second clinical hold placed on a Novartis-in-licensed HCV candidate this year. Last summer FDA halted trials of Anadys' ANA975 drug, a TLR-7 agonist in Phase Ib, after a preclinical toxicology study unearthed a potential safety signal.

HCV dealmaking has definitely been a hot space, and will likely continue to be so despite this spate of clinical difficulties. For our roundup of drugs in development for HCV, see this late-2005 IN VIVO story.

UPDATE: Idenix shares are off about two bucks, or 35%, at about 11:30a.

Phase II is the new Phase III

It's what you might call a slow news day for us so we figured we would dip back into our colleague Roger Longman's presentation from our Euro-Biotech conference last week. Roger's on vacation, see, so he can't stop us from pilfering his slides.

One of the points he emphasized during his talk and illustrated with some data from our Strategic Transactions Database is the idea that products that have cleared clincal proof-of-concept, i.e. the Phase II hurdle, are attacting the kind of deal dollars only seen previously for Phase III-stage products. The value of relatively scarce Phase III products these days is another thing altogether.

Taking a look at this phenomenon using upfront payments as a proxy for deal value, above, and you can see what we mean. So while those biotechs that excel at drug discovery can in fact sustain themselves by selling IND-stage candidates, which we pointed out last week, others can play on the next valuation inflection point at proof-of-concept.

Are Phase II deals a remotely new feature of the biopharma dealmaking landscape? Of course not. But the volume of deals for Phase II products has shot up in the past few years, and upfront payments are simply booming. A few recent examples include Novartis and Antisoma's deal around the oncology product AS1404 ($75mm up front), Bayer and Regeneron's deal on ex-US rights to Regeneron's VEGF-trap product in ophthalmology ($75mm u/f), and going back to last summer, Johnson & Johnson's ex-North America, ex-Japan deal for Vertex's hepatitis C protease inhibitor ($165mm u/f). Follow the links to a more detailed analysis of each deal.

Various pharma execs have chalked the rise in Phase II prices up to their own companies' inability to reliably and predicably get products through proof-of-concept themselves. This is excellent news for clinical-stage biotechs. A few, like CNS-focused Synosia, have set themselves up as proof-of-concept specialists, in-licensing compounds in preclinical or IND or even Phase I, shepherding them through a human efficacy trial, and licensing them on for a significant profit.

Sometimes the Bear Gets You: Coley Pharmaceuticals Edition

Coley Pharmaceuticals shares lost more than half their value today after its partner Pfizer discontinued work on its lead lung cancer compound.

The nail in PF-3512676's coffin was an independent data safety monitoring committee's verdict that a mid-trial analysis suggests the compound plus chemotherapy works no better than chemo alone.

PF-3512676, previously known as Promune (and also as CPG7909), was snatched up by Pfizer in 2005 in a deal worth up to $515 million. The compound was supposed to work by triggering the immune system through stimulating toll-like receptor 9 (TLR-9).

This is the second TLR-modulating drug that has crashed out of the clinic after being dealt to a big pharma for top dollar. Novartis bought rights to Anadys's ANA975 drug, a TLR-7 agonist, for hepatitis C only a few months after the Coley/Pfizer deal. That drug's Phase Ib program remains on clinical hold by the FDA after preclinical tox studies unearthed a potential safety signal last summer. Merck also signed a broad TLR deal with Idera in late 2006; that deal is ongoing.

Coley will hold a conference call later this afternoon.

UPDATE: Coley's call provided little in the way of further information specific to '676. Pfizer is leading the data analysis, according to Coley CEO Bob Bratzler, who described the company as "dumbfounded" and "shocked" at the news.

CSO Art Krieg aimed to confine the bad news to these particular trials and drug combinations, saying "this is a setback for this approach to the treatment of lung cancer in these particular combination regimens ... [but] the mechanisms are validated and we don't think the impact of this will be enormous beyond these approaches." Cancer drug development is challenging, he added, and a lot of successful drugs have failed multiple clinical trials before approval.

Perhaps. But the snafu puts Coley in a tight spot. At last glimpse--end of Q1--the firm had $97 million in cash reserves; yesterday's news sunk Coley's market cap to $91 million. Even considering the firm's estimated burn of about $3 million per month, the market is now placing zero value on its technology platform, its partnered projects with GSK and Sanofi-aventis (and the future of the Pfizer collaboration), as well as its various preclinical activities in RA, lupus and oncology. There is blood in the water.