The REAL scandal of drug pricing

Healthcare provision, at least in the Western world, is in a permanent state of turmoil, as costs rise inexorably and politicians search for ways to cut spending without angering voters.

But why do costs keep on rising so steeply?

High priced drugs often get the blame.  After all, every time new medicines come on line that treat something that was previously untreatable that can only ratchet costs up (albeit taking the health of the population to new levels).  The spotlight has been firmly on this upward cost pressure with the launch of orphan drugs Gattex™ and Kalydeco™, with price tags around $300,000 per year for each patient.

But drug prices do not always rise.  As atorvastatin became generic at the end of 2011, a $10billion per annum market for Pfizer’s Lipitor™ (the only available source of atorvastatin during the period of market exclusivity) almost disappeared overnight – translating virtually directly into savings for healthcare providers.

So why do these competing trends always seem to result in rising drug bills? Are the high-priced drugs really to blame?

Only by tackling the REAL scandal in drug pricing – the unearned premium – do we stand any chance of stemming the haemorraging healthcare budget

DrugBaron examines the real reason why drug prices remain stubbornly high, despite a decade of relatively poor productivity in pharmaceutical R&D.  The focus should be on prices paid for high-volume proprietary drugs that offer little incremental value over generics, or – most scandalous of all – in replacing branded drugs with generic equivalents wherever the opportunity exists.

Paying a premium for something that’s scarcely superior (or even identical) – what DrugBaron calls the “unearned premium” – is the REAL scandal in the drugs bill, and not the high, but justifiable, prices for drugs like Kalydeco™.  Politicians, driven on by a baying public, need to be careful not to jump on the bandwagon and cut the highest prices, threatening to once again ‘orphan’ those with rare diseases, while leaving untackled the real flaws in the system.

Efficient pricing can only come from efficient markets.  While multiple providers offer comparable services, then the competition between them will drive prices down until each provider earns only a slim margin over his costs.  Such markets also provide a strong incentive to lower costs by seeking efficiencies through the product or process.

Even if the offerings from different sellers are not strictly comparable, the market can still set prices efficiently, awarding a premium to any market participant whose product or service is clearly superior to the competition, with the degree of the premium proportional to the perceived degree of superiority.

Such markets work very well for generic drugs, keeping costs down to a fraction above the costs of manufacture and distribution.

Doctors and patients must get behind the mantra of “much better or else much cheaper”

But as every economist knows, the introduction of a monopoly breaks the market model.  How do you set a price for a product where there is no competition at all?

Proprietary drugs enjoy a state-sanctioned monopoly for a defined period as a “reward” to encourage investment in innovation.  Fortunately, however, this monopoly is usually very narrow – the precise composition of the drug may enjoy a monopoly but there is often nothing to stop others developing products with similar composition, or with the same mechanism of action, or even a completely different approach that achieves the same outcomes in practice.

While the monopoly is narrow, a limited market can still form as the participants compete to win market share for “their” monopoly.

This is the situation for the vast majority of marketed drugs.

Just occasionally, though, a drug comes along that does something quite unique – treats or prevents a disease that was otherwise poorly treated.  In these circumstances, the effective monopoly is broad (even if the actual intellectual property claims are still narrowly directed to the specific composition).

For these drugs, no kind of market is available to be the arbiter on price.

Such drugs are much more likely to occur in rare diseases for two reasons: (a) rare diseases tend to be more homogeneous and have better defined mechanisms, facilitating the development of new interventions that are step change better than early treatment – not least because for many rare diseases there simply is no pre-existing treatment; and (b) the economics are unattractive to “fast followers” – there is typically only room for one product in a rare disease space anyway, and late arrivals to the party rarely have the appetite to stay in the kitchen.  As a result, fewer drugs will get developed in each rare disease space.

The lack of market “arbitration” on price is then exacerbated by the small number of candidate patients to be treated with a drug targeting a rare disease.  To be at all attractive, the drug has to earn back the price of its development plus a premium to account for the risk.  We all know the risk in drug development (even in a rare disease) is high, so this risk premium is high.

Look at the equation that sets the minimum acceptable price: (C * P) / N where C is the total development cost, P is the risk premium and N is the number of patients that can be treated.  When N is small, the minimum price will always be very large.

As DrugBaron discussed previously, then, the choice is between paying such a price (without arbitration from any kind of market), thereby allowing a proper risk-adjusted return on capital invested in rare disease drug R&D, or else refusing to pay (or arguing for a lower price) and making development of future drugs in such rare diseases uneconomic.

That is clearly a choice for society, but the nature of the choice should be clearly communicated.  At present, there is an almost hysterical backlash against high-priced orphan drugs, such as Gattex™ and Kalydeco™, simply because the public do not understand the underlying economics.  It is easy to blame the greed of the drug companies, but that is, quite simply, the wrong way to look at the problem.

Politicians must not fan this outcry to provide a lever for them to argue for lower prices: when rare disease sufferers are once again orphaned that will prove to have been the wrong decision for all concerned

In reality, the problem with drug pricing lies not with these high-price orphan drugs, nor with the generic drugs whose price is efficiently set by the market, but with the high-volume proprietary drugs in the middle of the spectrum.  The orphan drugs may attract all the attention because, like the tallest man in a crowd, they are the most visible.  But the real focus should be on the value for money provided by drugs that offer only incremental improvement over existing medications, particularly in diseases that are very common.  Even a small mismatch between price and value will translate into vast and unwarranted inflation of the healthcare budget.

A good example right now is Crestor™ rosuvastatin from AstraZeneca.  Here is a drug that continues to be a blockbuster (selling more than $1billion a year worldwide) despite having no appreciable advantage over the most widely used statin, atorvastatin.  Indeed, AstraZeneca tried every trick in the book to generate clinical data to demonstrate the benefit of Crestor™ over atorvastatin in order to secure continued sales when branded atorvastatin (Pfizer’s Lipitor™) went over the patent cliff.  But they failed.

Hence the comparative evidence is there for all to see.  So presumably no-one gets Crestor™ now, with generic atorvatstatin available for just a few pence per day? Surely it would be hard to find a simpler way to save money than to move patients from Crestor™ to the generic drug?

Sales of Crestor™ have fallen, but not below blockbuster proportions.  There may be, in a few patients, specific reasons why Crestor™ is better, but for the vast majority no reason that could possibly justify the difference in price between a branded pharmaceutical and a generic.

Stories of the demise of incremental innovation turned out to be premature

And its not the only example.  The pharmaceutical industry is now in full swing promoting the new generation of anti-coagulants, direct thrombin inhibitors and Factor Xa inhibitors such as Pradaxa™ and Eliquis™.  Again, head to head comparisons show marginal benefit (if any at all) versus the entrenched generic drug, warfarin.  Yet these newly launched drugs are attracting sales in the hundreds of millions of dollars already, and see growth that may yet turn them into blockbusters (even if the growth is ‘disappointing’ compared to the over-optimistic marketing models used to justify developing these drugs in the first place).

Even where there is some incremental benefit, its unclear whether its sufficient to justify the price hike compared to a generic incumbent.  Lilly’s Effient™ prasugrel is clearly not superior to generic clopidogrel, and attracted virtually no sales as a consequence.  But AstraZeneca’s Brillanta™ ticagrelor does have advantages that stem from its non-covalent inhibition of P2Y12, together with the absence of a requirement for metabolic activation, which ensures more even exposure irrespective of P450 genotype.  Do these differences merit a more than 15-fold difference in price (generic clopidogrel at £3.50 a month in the UK and Brillique™ ticagrelor at £55 per month)?

The NICE methodology suggests it may be – just.  DrugBaron is less convinced.  The cost effectiveness model used by the manufacturer, and accepted by NICE, does not fully incorporate the extra burden due to increased adverse events.  The most likely explanation for the data in the PLATO trial (that compared ticagrelor and clopidogrel head-to-head) is that the modest reduction in major cardiovascular end-points (from 11.7% to 9.8%) was due to the slightly higher anti-coagulant effect of the selected dose, which in turn also increased major bleeding events (including fatal bleeds).  One could achieve that same effect at less than a tenth of the cost by slightly increasing the dose of clopidogrel.

Don’t brand the people who make these assessments as mere “bureaucrats” and dismiss their conclusions as penny-pinching

The list goes on: is Lyrica™ pregabalin (with $3+ billion annual sales for Pfizer) better than generic gabapentin?  A significant proportion of Lyrica™ sales are for off-label usage, for example in neuropathic pain and migraine, where even its efficacy versus placebo has been repeatedly questioned by some analysts.  Certainly, for the vast majority of Lyrica™ scripts there is no evidence whatsoever that it is better than the earlier generation GABA analogue, gabapentin, that’s now generic.

Questions like this abound in almost every major indication.  Quantifying benefit is really difficult, but at least with the introduction of NICE in the UK and IQWiG in Germany, we are seeing attempts at such quantification.  Unfortunately, these bodies attract a lot of criticism by attempting to quantify absolute cost-effectiveness (using for example dubious measures such as cost-per-QUALY).  This can lead to decisions on orphan drugs, for example, where there is often no real alternative (particularly in cancer) that are at odds with the public perception of benefit.  But applying these quantitative methodologies to comparative assessments of different treatments for the same indication should yield, in the end, better pricing of incremental innovation.

In the US, HMOs should perform exactly the same role, with a model based more closely on a market that the attempt at rational economics employed by European government healthcare organizations.

So why isn’t it working? How can Crestor™ or Lyrica™ still be prescribed on the NHS in the UK, when many other drugs have been effectively “banned” by NICE because the benefits they show were considered uneconomic?

Often times, the answer lies in the lack of genuinely comparable clinical data.  Sometimes its because head-to-head comparisons with the gold standard treatment were never required to register the drug.  More often its because of “label creep” – in common diseases, big selling drugs are often used in patient populations well outside those where any comparative studies were conducted (as is the case with Lyrica™).   And, as any statistician or epidemiologist will tell you, its really difficult to combine data from different studies, with different inclusion and exclusion criteria, and come to any kind of definitive answer as to which drug is “better”.

The vast and experienced marketing departments of large pharmaceutical companies then exploit this “data haze” to sustain sales of drugs whose incremental benefit is dubious at best

It is always possible to find individual studies, individual end-points, even individual patients, whose story supports the continued use of a more expensive alternative – if doing so feeds directly into your bottom line.

Its time for payors, whether in government or quasi-governmental agencies, or market-driven insurers, to fight back.

The first place where battle must be joined is on direct substitution of generics.  Since there is no medical argument to sustain a premium once a generic version of the same active substance is available, then the argument for the cheaper option ought, at least in principle, to be more easily sustained here.  But, just because the rationale is obvious to all, you should not assume this battle is already won.  In some countries, the vast majority of possible generic substitutions are still not being made – in Ireland that figure may be up to 90% according to the Sunday Business Post.

That’s the same healthcare system that considers Kalydeco™ too expensive to give to its citizens with cystic fibrosis.

In the UK, this particular problem is (now) less widespread, with most systems operating on an ‘argue otherwise’ basis – even the IT systems automatically substitute generic alternatives and leave the physician to actively over-ride the normal workflow to secure a pricier alternative for the patient.

But the real battleground lies with the drugs where the generic alternative is not exactly the same.  Crestor™ is not atorvastatin; Lyrica™ is not gabapentin; Effient™ is not clopidogrel; Eliquis™ is not warfarin.

Many would argue it is not appropriate in these cases to have an IT system switching products behind the scenes.  Ostensibly there is a medical as well as an economic decision to be made.

The question, then, is whether such a decision needs to be made again and again by every doctor, simply because every patient is different – or whether a single decision based on the summation of the evidence is sufficient.  After all, despite the much-trumpeted age of “evidence-based medicine” there is rarely enough evidence to treat a single individual patient differently from the optimum average identified from clinical trials, at least not when considering “first line” therapy (that is, the first attempt to treat a patient with a particular disease).  On what possible grounds could a physician decide to prescribe Crestor™ or Lyrica™ before knowing whether there is an issue with atorvastatin or gabapentin?

The biggest problem, as is so often the case, is therefore one of culture.  Doctors and patients alike scream blue murder at the possibility of “bureaucrats” handing down global decisions that tie the hands of the treating physician.  For some reason, we all believe that the particular doctor sat in front of us knows better how to treat us than the summation of all the medical knowledge in the system.  Too often, we believe that the motivating factor behind such a global decision is “penny pinching”.

But if we are not allowed to penny-pinch, is there any wonder that drug bills, and healthcare costs generally, are rising so fast that they have already reached levels that are economically unsustainable for the longer term?

Penny-pinching is a good motive.  As long as people are not being denied effective medicine, surely we ought to insist our politicians achieve lower cost delivery for our healthcare system?

We cannot keep equating lowering costs with lowering standards – we already know that the reverse is not true: spending more does not automatically yield better health outcomes.

Maybe the mis-directed public ire against high-price orphan drugs can be deflected into a more useful direction.  Lowering the price of Kalydeco™ and Gattex™ will scarcely scratch the surface of the drugs bill, but it risks killing the incentive for developing drugs for ‘orphan’ indications.  But switching scripts from Crestor™ to generic atorvastatin? Lyrica™ to gabapentin? That has the reverse effects: it will make a meaningful dent in the drugs budget, AND it will send a clear message to the pharmaceutical industry: incremental innovation does not pay.

DrugBaron optimistically predicted the death of incremental innovation two-and-half years ago.  But such stories of its demise turned out to be exaggerated – or at least premature.   Many of the class of 2012 approvals by the FDA were just such incremental innovations, and early sales data from Eliquis™ suggest that sales are being booked.  Even reformulations are still making money, as the story of Ritalin shows.  Such tiny innovations remain sufficiently expensive to sustain a huge industry.

The only way, then to take back control of healthcare budgets and pay a premium only where it is merited by real innovation is for doctors, patients and payors to unite.  The doctors and the public must get behind the mantra of “much better or else much cheaper”.  Don’t brand the people who make these assessments as mere “bureaucrats” (a term loaded with deep disdain in the 21st Century), but recognize that they are experts who understand medicine at least as well, and quite often very much better, than most doctors who see patients every day.  Don’t decry their motive to buy you cheaper healthcare.

Only by tackling the REAL scandal in drug pricing – the unearned premium – can we stand any chance of stemming the haemorraging healthcare budget.  If you are worried that too much of your taxes are going to healthcare companies, you are right.  If you think its by paying for orphan drugs at very high prices, making development of drugs for rare diseases economically viable, though, you would be well wide of the mark.  The real problem is much more insidious – by paying too high a premium for incremental innovation you are not just paying too much for your healthcare today, but you are incentivizing yet more marginal improvements in the future.