Pharma’s Cutting Edge

It was bound to happen sooner rather than later, and Pfizer or GSK was likely to be the first to do it: create exclusive contracts with investigators I mean.  That would be your best enrolling sites.  Scared?  You should be. 

As reported by ClinPage from the annual DIA meeting last week, Pfizer’s Andy Lee let on that Pfizer has been signing exclusive deals with Pfizer’s best enrolling sites.  Either the sites will contract for studies only with Pfizer or they will not contract with Pfizer for any studies.  It takes chutzpah to make a move like that, but more than chutzpah it takes research volume, lots and lots of money devoted to clinical research.  Without the money to keep a site at capacity, or on the sidelines getting paid to be at less then full capacity, no company can hope to craft such deals.  Only the very largest research companies can afford this, and even the largest CROs will not be able to wrest a site from big pharma’s grip for the benefit of its more modestly sized clients.  Talk about a game changing situation.

As I say, it wasn’t hard to predict this move, I’m only surprised that it’s been going on for a while, and I hadn’t heard about it.  These things tend not to stay secret.  Why am I not surprised?  Because pharma development productivity, particularly in the later stages of clinical development, must improve to compensate for environmental pressures on sales and volume growth.  It must, or the industry’s entire business model is in jeopardy, and I don’t see the industry allowing its model to collapse without a good fight.

Despite some propaganda to the contrary, the workflow inefficiencies constraining overall development productivity are not widely appreciated, and no company has demonstrated that it knows how to improve productivity, however it is (reasonably) defined.  That said, it is likely that local operational inefficiencies, such as those confounding clinical study enrollment, are disrupting local process flows sufficiently to negatively affect cumulative study completion rate.  In other words, know one knows for sure, but it’s likely that decreasing the average time from protocol approval to final subject enrolled will ultimately decrease total development cycle time without diminishing the quality of development (i.e. the learning that accrues from doing studies), thus improving R&D productivity.  It’s also pretty clear that if you can find a way to contract with better performing (in terms of enrollment) sites more frequently, you are more likely to decrease this cycle time.  As Pfizer’s Lee noted in his DIA appearance, and is generally appreciated now, some investigative sites consistently enroll better than others, and it’s believed that they do so without diminishing the quality of research.  If you can lock these high-performing sites up, with exclusive contracts say, you’re more likely to increase enrollment speed than if you must compete for sites and risk losing high-performer to your competitors.

As far as I know, Pfizer is the first to try this all-or-none strategy.  I think other firms will wait to see how it shakes out before following their lead.  Although, as I say, exclusivity is the surest way of locking up the best performing sites, it’s also probably the riskiest.  A company the size of Pfizer doesn’t have to worry about not doing enough studies to keep a site busy–thereby wasting the premium it must pay to maintain exclusivity–but it does have to worry about frightening or angering the doctors that prescribe its drugs.  It’s easy to imagine that an investigator or research institution might wish not to engage in an exclusive relationship with a single sponsor:  Will it create a perception among patients, insurers, regulators, or professional associations that she is in the sponsor’s back pocket?  Maybe such thinking is reasonable, especially for academic sites.  Harmful perceptions aside, I would imagine that many high-performing sites will find such strong-arm negotiating tactics ridiculous.  After all, from the investigator’s perspective, it’s not as though research studies are hard to find.  It’s a supplier’s (investigator’s) market in developed countries by all accounts.  Why not hold out from such a deal, and let competition decide who gets the best sites?  If supply remains regionally and situationally constrained, as it is presumed to be, investigator grant payments will continue to increase.  Exclusive deals, unless very lucrative, will not benefit sites in these regions.  And perhaps no amount of money will satisfy some site’s need to work on the most promising medicines, regardless of who is developing them. 

This enrollment strategy absolutely is worth keeping an eye on.  If any investigators have signed such a deal with Pfizer or another firm, I’d love to hear from you.

In my last post, I  discussed what I believed would be the basis for debate and concern among the panelists at Wednesday’s FDA DMEDP Advisory Committee–the benefit to risk balance of the drug in light of modest efficacy and risk of suicidality.  This was indeed the topic of much of the committee’s deliberations.  As you now know, the committee voted unanimously (14-0) against approval of rimonabant due to a perceived unfavorable benefit to risk balance.  They also voted unanimously that Sanofi-Aventis hadn’t sufficiently characterized the safety profile of the drug.

What does this mean?  It means that this FDA division and its outside advisers, and probably much of CDER along with them, is taking the position that they want real, sustained evidence of benefits when risks during chronic use of a drug are small in number but potentially fatal.  Stated differently, risk tolerance is low among this regulatory body, and it wants to see relatively more evidence of benefit to counterweigh a risk of a serious adverse event than in the past (I’m thinking 2 to 3 years ago).  With rimonabant, they looked at a drug intended for chronic use that was associated with modest, reversible benefits on weight and markers of cardiovascular disease; no long-term cardiovascular outcome evidence; tolerance issues that promoted drug discontinuation, and an uncommon but potentially fatal adverse effect.  In the current regulatory climate, rimonabant never stood a chance.

Perhaps DMEDP and its advisers would have thought differently had rimonabant been a drug developed to treat a disease without the social baggage of obesity.  In this country, many people, including many health professionals, still consider obesity a condition caused solely by sloth and overindulgence that should be treated by willpower alone.  But I don’t think so.  Rather, I believe that this division and its advisory committee members are simply willing to be more paternalistic than in recent history, in light of missteps made by prescribers, pharmacists, sponsors and others that have led to a series of recent, high-profile drug recalls and new prescribing warnings for marketed drugs.  In essence they’re saying:  “The American public is either unable or unwilling to responsibly manage its use of new drugs.  Until that situation changes, we must intervene by taking away the tough decisions…by limiting the choices that are available to it.”

I haven’t seen FDA place a lot of the blame on itself for the inability or unwillingness of prescribers and users of new drugs to manage their use responsibly, but most senior folks at FDA must know that through their inability to demand creation and dissemination of effective prescribing information, and through their inability to demand creation of and to enforce effective new-drug risk management plans, and through their unwillingness to require the types of large trials needed to define risk before marketing that they shoulder part, perhaps most, of the blame.  None of this has to do with the needed resources for postmarketing surveillance being requested in the new PDUFA negotiations; I’m only speaking of what needs to happen before a drug reaches the market.  For that, FDA has said it has the resources it needs to do the job well.  So, why hasn’t it?

Sponsors, for their part, also shoulder significant blame for the rise of paternalism at FDA.  Perhaps they just missed seeing the writing that has been on FDA’s wall for the past few years, but I think they saw it quite clearly and chose in some cases to ignore it, hoping it would go away, and in others to address it weakly, with a wink to their investors.  Now they must overcome this guilty-until-proven-innocent mentality to market new drugs for chronic disease states, not to mention the formulary and insurance coverage hurdles awaiting new drugs once approved for marketing.  The regulatory landscape won’t change for quite a while.  If anything, we haven’t seen the top of this wave.

Sponsors, if you hadn’t heeded prior warnings, you’ve now been clearly warned:  It’s up to you and you alone to demonstrate that your new chronic-use drugs work as intended, using highly relevant and generally accepted measures of effectiveness; that they can be used safely, where safely means that the risk of use is very clearly outweighed by the benefit, even if that means you must find ways to limit risk by restricting the user base and creating redundant methods of safeguarding their proper use.  New chronic-use drugs must also be priced so as to be fiscally viable to large purchasers, and their value must not be a matter of faith but rather of overwhelming evidence.  Which other companies are in late stages of development with CB1 antagonists?  That would be Pfizer, Merck, and Solvay/BMS.  We’ll see soon if they have been paying attention.

Sanofi-Aventis made many of the right moves with rimonabant.  They misstepped by not gathering safety information in a way that could be used to create a sound risk-management strategy. There is time for them to make amends, in part, by gathering more evidence of both efficacy and safety.  Should S-A be able to generate evidence from the CRESCENDO outcomes study already in progress that very obese users (i.e. BMI >40) experience a meaningful reduction in their risk of cardiovascular events while using rimonabant, they will go a long way towards tipping the scales in favor of benefits.  In the meantime, S-A must collect more evidence from Europe that shows the drug can be used safely in the “real world”.  This, combined with some way of predicting suicidality through the use of simple baseline measures, other than or in addition to weight and history of depression, should suffice to create a viable risk management strategy.  Whether S-A would want to launch in the U.S. with a restrictive prescribing strategy and its attendant black-box warnings is another question. 

In its 2007 renewal of PDUFA, Congress deemed it wise to earmark funds specifically to safety surveillance.  The underlying concern that prompted this action is that prior user fee legislation has resulted in substantial increases in FDA resources applied to review and approval of new medicines but insubstantial increases in resources applied to review of drug safety once medicines have been marketed.  It would be difficult to argue against this view; all evidence indicates it’s accurate. Technology advances—large-scale, real-time electronic capture of prescription fills and health outcomes at the patient level—offer the promise of better post-marketing drug safety surveillance, and the public rightly wants to see its government avail itself of these advances to better define risk and provide early warnings of serious safety issues.

Of course, surveillance is an imperfect method.  Most surveillance relies on passive observation to identify health outcomes, establish relationships to drug use, and define the magnitude of risk.  That is, the observers collect information but don’t intervene in the methods used to generate that information beyond, perhaps, enforcement of coding standards (e.g. ICD, CPT, ATC).  A problem with this sort of surveillance is that substantial noise (adverse events) relative to signal (adverse drug reactions) is generated, leading to significant problems interpreting data.  This is one of the major problems with FDA’s current and proposed surveillance efforts.  More resources can increase the amount of data entering the surveillance funnel (by providing more sources of passive adverse event information), but it won’t necessarily improve the information flowing from the other end.

In the experimental setting such as exists during pre-marketing testing, safety signals may be distinguishable from noise by techniques such as third-party review of adverse events; active intervention to determine causality (e.g. withdraw and careful reintroduction of drug for non-serious events); capture of ancillary history, physical exam and lab tests; and, most usefully, a comparison of event incidence between two experimental treatment groups (ideally including a placebo group).  In the “real world,” where much of the proposed heightened surveillance efforts live, such interventions by observers are impractical to implement.  Nevertheless, there are opportunities for experimental and quasi-experimental observational safety studies to be conducted in the post-marketing setting, and these types of studies may be used by FDA to supplement traditional surveillance techniques.  Examples include nested studies in ongoing, large-scale observational cohorts (e.g. Nurses and Physicians Health Studies) and de novo studies in cohorts whose healthcare is administered by a single institution or network with central administration (e.g. certain large HMOs, VA Health System).  By automatically enrolling participants and assigning contemporaneously matched non-drug users to a control group, some of the noise inherent in observational studies may be reduced.  It’s smarter than passive capture of data with post-hoc sorting of signal from noise, but it’s still imperfect.

Ideally, sponsors and FDA should be able to identify all serious risks of an investigational drug before bringing it to market, while the drug is still in the experimental (i.e. interventional) setting.  Identifying all serious risks before marketing would minimize ‘at-risk’ exposure, establishing those risks at the earliest time practical (assuming that postmarketing surveillance takes much longer than premarketing experimentation to generate unequivocal risk definition, despite much larger exposure to drug in the former).  Sponsors would also reduce their exposure to product liability risk and the potential tangible and intangible costs of a recall.

Despite these benefits, however, both sponsors and the FDA have suggested that identifying all serious adverse effects of a drug prior to market approval is impractical, as the numbers of subjects required is prohibitive. 

The most pressing safety issue FDA and sponsors grapple with, at least for chronic-use drugs, isn’t what it once was.  It’s not identifying rare drug reactions.  Although in the past rare, serious drug reactions eluded detection until after approval, sponsors with guidance from FDA have improved their ability to detect the most frequent types of reactions before marketing approval (probably the best example of this is enhanced premarketing surveillance for torsades de pointe due to drug-induced QT interval prolongation). 

These days the stickiest safety problems are adverse reactions that may be disguised as serious, unrelated conditions.  The voluntary recall of Vioxx is particularly instructive, because it’s a recent example of a serious adverse drug reaction—thrombogenic coronary artery disease—masquerading as a common but serious disease that was missed during premarketing review (some have argued that it wasn’t missed as much as ignored).  Much larger populations are required to detect relative increases in the incidence rate (assessed as the hazard ratio) of such relatively common conditions than are required to detect single occurrences of rare conditions.

Understanding this, regulators have occasionally relied on the power of large numbers to help sort out risks premarketing.  Notable examples include the large, simple safety studies of Vanlev and Ketek.  However, such studies are not demanded routinely.  In these two examples, specific risks that had been suggested by smaller studies were investigated further.  They were not used to detect risks de novo.

Sponsors are unsurprisingly reluctant to add to the cost and time of premarketing safety evaluations beyond what is already required, and FDA, taking increasingly large chunks of funding from the industry, is unsurprisingly reluctant to ask for such sacrifices.  Is there a way out of this conundrum with today’s technology?  I believe that large, simple safety studies are an interim answer for chronic-use drugs aimed at large treatment populations, particularly for conditions where suitable alternative therapies exist.  In such cases, a population risk threshold should be established for common, serious conditions (essentially serving as the excess allowable hazard for study power/sample size calculations).  There is no simple formula for this, as it requires expert assessment of potential population attributable benefits and risks.  Different drugs will require different sized populations.  Such studies can be run in parallel with routine phase 3 studies and should be started when sufficient preclinical information is available, usually after carcinogenicity studies are complete.

It’s funny.  I write these posts usually twice or more per week not knowing who, if anyone, is reading them besides blog reporters (e.g. FDA News, which has generously featured my posts in its daily drug newsletter on numerous occassions), blog aggregators (e.g. Seeking Alpha…ditto), the PR firms that service big pharma (e.g. Edelman; looks like Pfizer’s an important client) and, of course, the search-engine spiders.  So I probably shouldn’t care about not posting for a week or two, especially when it’s due to competing activities that involve me receiving compensation for my thoughts, but I do.  I’ve missed the ritualistic dispersal of my ego to the ether and have even felt a chill of guilt reading through other editorialists’ musings the past couple of weeks.  This makes me think the act of writing the posts provides some catharsis for me, one that was particularly important when I left my last job at big pharma to start my own businesses.  Who knew?  I thought I was providing a service for others, not for myself.

Anyway, I’m no longer on my own, as I’ve become an employee with an established business called Fast Track Systems.  If you’ve been in the biz for a while, you might remember them as Data Edge, the company that supplies most of the pharmaceutical industry with clinical trial/CRO contracts data that are a must-have for benchmarking trial and related outsourcing costs (for planning, negotiation, and audit purposes).  They’ve re-branded, but they still supply the industry with these data, now provided under the collective name TrialSpace (the PICAS clinical “grants” data are now in TrialSpace Grants Manager and the CRO data are in TrialSpace CROCAS).  These data products and related services remain Fast Track’s core business, but in the last few years they’ve hired a young, visionary CEO to build new business in the eClinical space.  I joined them because of their vision but also because they’re delivering on it…now.

Peruse my back collection of posts and you’ll find numerous references to the urgent need for pharma R&D to come up to speed process-wise with other research-based industries.  Clinical research, in particular, has badly lagged not only other industries but also the rest of pharma’s research enterprise in its adoption of time- and cost-efficient operations.  There’s no direct evidence to explain this lag, but I believe it’s largely because of inertial forces in the form of past industry successes (i.e. the really good years), whereby clinical research costs are relatively unconstrained in the face of high returns on R&D investment.  It’s also likely related to the general inertia to process change that pervades medical practice.  Doctors still write office notes and prescriptions on paper for God sakes.  As medical practice comes out of the paper age, so too will the industries that serve it.

Which leads me to the eClinical (can someone please invent a better word) paradigm shift (can someone please invent a better phrase) now occurring in the life science industries.  The move away from paper-based information flow and its associated processes towards electron-based information flow and its associated processes is absolutely necessary to restrain growth in clinical research expenses.  It will do so by improving productivity (the amount of work output for any unit of resource input), reducing cycle time (the time needed to complete a task), and potentially (no, it’s not yet been proven) by increasing technical success rates of therapies that reach the clinic.

I’ll have much more to say about the eClinical revolution in the months ahead.  For now, though, I’ll simply say that I’m genuinely excited to be part of a visionary enterprise that is working hard to make this shift occur quickly and as painlessly as possible.  The technology is here, and it’s ready to be used now.  Are you ready for it?

The following letter was written by me in response to David Nathan’s February 1st Perspective in the NEJM entitled: “Finding New Treatments for Diabetes: How Many, How Fast…How Good?”  (Subscription req.)  In this essay, Nathan argues that “FDA’s approval process for new antidiabetes medications should take into account their additional and unique contributions, especially when their glucose-lowering efficacy is similar to or less than that of currently available medications.”

In essence Nathan questions the wisdom of allowing approval of new drugs (for chronic diseases, like diabetes) that have not been demonstrated to be better than existing drugs.  My response offers a scientific reason why regulators should not have such discretion; unmade economic arguments are at least as compelling.

_______________________________________________________ 

Dr. Nathan throws light on the subject of hurdles for drug approvability in the U.S. Although his arguments are specific to drugs used to treat Type 2 diabetes, the points raised are germane to a number of chronic disease states, where multiple drug classes are currently in use.  Under current U.S. law, FDA lacks authority to restrict approval of new drugs because they might be deemed unnecessary, no matter in whose judgment.  It is a mistake to believe that putting such authority into the hands of FDA today is wise, for the simple reason that traditional pre-marketing studies of new drugs, such as those cited by Dr. Nathan, are incapable of determining which individual patients will or will not benefit from use of drugs.  Rather, these studies aim to show average effects in populations selected to be representative of users; they ask general safety and efficacy questions and yield general guidance only.  Many hope that future pre-marketing approval studies that incorporate highly predictive markers of risk and benefit (i.e. personalized medicine) will allow regulators to tailor approval decisions to specific subsets of patients, but until then it remains for doctors to perform the “N of 1” experiments needed to determine whether a drug benefits an individual or not.  A system for capturing results emanating from these large number of experiments is sorely needed.  In the meantime, clinicians considering prescribing new drugs for conditions like Type 2 diabetes should demand comprehensive monographs from manufacturers that include all relevant preclinical and clinical information.