Amgen’s Right, ICER Cherry Picks Who Lives Or Dies

A lot of the health media was surprised that Amgen decided to criticize ICER before it released its draft decision about the price and access to new medicines for multiple myeloma including Amgen’s Kyprolis. I wasn’t.  After all, Amgen had already dealt with ICER’s ‘methodology’ when it recommended that Repatha be sold for $2600 a year and be limited to about 3 percent of patients with statin resistant LDL that could benefit.

The surprise is a function of the fact that the media is buying into to ICER’s well-funded – and extremely effective — attempt to establish itself as the de facto price setting group.   It is not a function of the fact that ICER’s findings were not pre-ordained.

Amgen believes ICER is “using opaque methods to combine multiple, disparate trials to arrive at different estimates of efficacy, or make assumptions to create unrealistic “worst-case” scenarios. Results produced by independent organizations should be informed by experts, made fully transparent and available, and undergo complete and independent peer-review.”

The company is right. ICER defines value as whatever doesn’t exceed an arbitrary cap on drug spending as set by PBMs and insurers. But it is clear that ICER is cooking the numbers ala Breaking Bad to get the desired outcome.

Like NICE, the rationing body in the UK, ICER cherry picks data to achieve its desired conclusion. If anyone needs an alternative to incarceration, they should flip through a NICE guidance to see how it picks and chooses what data to accept from companies and what data it uses to say yes or no. It accepts what it wants and rejects what it wants.

The sloppy, even sleazy, approach ICER takes is on full display in its effort to compare newly approved drugs for multiple myeloma and compare them with the combinations of lenalidomide (Revlimid) plus dexamethasone (Rd) and bortezomib (Velcade) plus dexamethasone (Vd).

Let’s set aside that there is no standard treatment for myeloma patients who relapse at ANY stage of their disease.  In many cases, the combination used is a function of what medicines were used before.  Myeloma is incredibly heterogeneous.   Yet ICER has no problem making comparisons and assuming every patient will respond the same.  The Mayo Clinic’s Dr. Rafael Fonseca, on the country’s experts in treating MM notes: “The value of these interventions can vary significantly by the presence of these various risk factors. For instance, a patient who requires stem cell transplant and is considering maintenance should discuss with the treating physician the various options for treatment based on genetic heterogeneity. Patients with standard genetic factors could very well be treated with lenalidomide versus patients who have a high-risk disease where the use of proteasome inhibitors it is highly recommended.”

Not only does ICER ignore these important variations, but it also uses a statistical magic trick – called network meta-analysis — to turn highly different patients in different clinical trials into carbon copies of each other.   ICER never tells anyone how or why it achieves this transformation.  It never shares its methods or data and it never submits ANY of it’s work to peer review.  Instead, a group of ‘experts’ that also happen to be dues-paying members of ICER pass judgment.

At least NICE has patients on their panels, ICER has none.  NICE has people who actually use the medicines they are evaluating on their panel, ICER has none.   ICER doesn’t publish in academic journals: instead, it issues it’s ‘findings’ by sending around press releases which are then reported and repeated by medical journals and journalists.

I could go on.  The number of scientific offenses that ICER commits could fill a book, a very boring book to be sure.   The most important thing to keep in mind is that for all the statistical mumbo-jumbo, ICER establishes prices and access based upon a GDP+1% cap on the total spending on drugs as a percent of total health spending.  That cap, like ICER’s value measure, is arbitrary and set to, as ICER President Steve Pearson has observed, to “set off alarm bells” about drug prices.   To that point, ICER will only look at single drugs because it’s trying to set the price of each drug so it doesn’t add more than $900 million a year to health spending.  Forget about what combination of treatments works best.

So let me save you the time and effort of reading another ICER report and show you how they crank out their pharmaco-economic fairy tales.  (Trust me, my rough estimates follow the ICER formula without all the footnotes.)

Pick what you want to spend to extend someone’s life from $50K-150K a year (ignoring consensus economics that it’s more like $300K)

2. Multiply the list price of a drug by the number of patients that could benefit.

3. Divide that total by the amount you want to spend per QALY (always use $50K even though that number was pulled out of thin air in 1980 to establish the value of dialysis. If you haven’t figured out by now, the cost and choice of a QALY cost are subjective.  In the case of ICER and Peter Bach’s rationing calculator, it assumes insurers and PBM — and both fund ICER and Bach –will choose the QALY )

4. If your cost per QALY for all patients is above $900 million a year you either reduce the price to meet the cap or restrict access.

So with that in mind let’s look at how ICER acts as judge and jury.

1. There are about 80000 people with MM.   A recent study estimated that 65 percent of patients will relapse each year.  That’s 52000.

2. To spend $50k per each life of each of the 52000 would cost $2.6 billion.

3. Which means we need to cut spending by $1.7 billion one of two ways:

Spend only $17000 per patient.   (By way of comparison, 20 mg Cialis has annual cost (at list price) of $16800.

Treat only 18000 patients, let 34000 people die.

I won’t even discuss combination therapy because ICER won’t, even though it’s the best way to treat relapsed patients.

You can’t blame the media for not taking a closer look — or at least the same close look they have applied to pharma — at how ICER measures value.  But that’s because the producers of medical innovation have failed to systematically explain the social, economic and medical consequences of the rationing ICER proposes.

So while I agree with Amgen and other companies that ICER (and Peter Bach) is shortchanging value, I can admire how well they have defined the conversation about drug price.  ICER has done something pharma, biotech, and MedTech have the resources to do but never done. It has effectively and properly defined its audience as the media, Congress, and consumers.  It has been proactive and at least gives the appearance of being objective and ‘independent.’

The biotech and pharma industry should support a competitor to ICER, one that — in concert with patients and providers – uses a long term measure of value that takes into account individual differences in needs and response and outcomes that matter, like productivity, quality of life, physical and emotional independence, time spent with loved ones, caregiver burden.

I hope the biopharma industry ceases fiddling and responds to ICER’s cherry-picking of data and who lives and dies.

This was authored by Robert Goldberg, Ph.D. and originally published at DrugWonks Blog.

Close Menu