Health News Review

A front-page story in the New York Times today, “Rare Mutation Ignites Race for Cholesterol Drug,” is worthy of further exploration. Any time the NYT puts a health/medical/science story on the front page, it probably warrants further exploration.

 

But I don’t think NYT front-page placement automatically warrants further front-page pickup by media across the country, but it often does get that imitated response.  My local Star Tribune paper put today’s cholesterol story even higher on its front page than the Times did.

When smart, veteran science journalists – like Gina Kolata in this case – get front page space, it’s often because they capture a compelling combination of intriguing science and “Wow, what if?” framing.  This story frames:

  • “a fevered race among three pharmaceutical companies…to test and win approval for a drug that mimics the effects of the mutation, drives LDL levels to new lows and prevents heart attacks.”  Guest commenter Harold DeMonaco reacts to “the race” below.
  •  “All three companies have drugs in clinical trials and report that their results, so far, are exciting.”  Our observation:  we’re not given any details of those results.  And, if they came straight from the pharma developers’ mouths, we should be skeptical anyway.  So why the front page story?
  • “This is our top priority,” said Dr. Andrew Plump, the head of translational medicine at Sanofi. “Nothing else we are doing has the same public health impact.” What public health impact?  The drugs have had no public health impact yet. 
  • “Dr. Gary H. Gibbons, the director of the National Heart, Lung, and Blood Institute, estimates that even if the drugs were expensive and injected as many as two million Americans might be candidates. But if they could eventually be made affordable and in pill form — two very big ifs — they might be used by one in four adults, he said.”  That excerpt swallows three, not just two, very big ifs.  Discussing one in four adults potentially using drugs about which we are given no data at this point is like predicting the 2093 Super Bowl winner.  Fun, but silly.  Dr. Gibbons is entitled to his opinion.  We think journalists should provide a balancing perspective. The closest the story comes to providing data is in the next two bullets:
  • “So far, people with stubbornly high cholesterol levels who are taking the drugs in preliminary studies have seen their LDL levels plunging from levels well over 100 to 50, 40, or even lower.”  All the people?  How many people?  What are their demographics?  Woefully empty.
  • “Pfizer is interrupting treatment when LDL levels reach 25 or lower. The people seemed fine, but the company got nervous.”  All the people?  How many people?  What does “seemed fine” mean?  How do you define that?

These were just a few of the things that caught our eye at first glance.  Intriguing story.  Terribly incomplete.

Journalist Paul John Scott wrote me about the story early today.  When reading it, he reflected:

“Yes, let’s start drugging LDL down to 25; what could possibly go wrong? The piece made me sad, mostly.  I can’t see how a reasonable person doesn’t read this and envision the day the patent runs out and the economy has been drained of another trillion in game changer drug spending and heart disease remains firmly in place.”  He thought the story should have looked at “the side effects of lowering LDL, and the heterogeneity of LDL, and the questionable returns of statins, and the dubious nature of moving surrogates for risk factors.”

Harold DeMonaco, MS, one of our most active story reviewers and guest bloggers, offered his own reaction:

There are numerous quotes from folks from industry and their clinical consultants in the story but the following caught my attention: “This is our top priority,” said Dr. Andrew Plump, the head of translational medicine at Sanofi. “Nothing else we are doing has the same public health impact.”  An interesting perspective to say the least.  While I appreciate the desire on the part of pharma to identify a blockbuster drugs, something it has failed to succeed in for some time, suggesting a dramatic public health impact may be a bit extreme.

Heart disease is a major problem in the United States and a leading cause of both death and disability.  The introduction of the statins in the late 80’s revolutionized the approach to primary prevention.  As the story notes, “Statins, the cholesterol-lowering drugs that went on the market in 1987, were a huge breakthrough, but far from a panacea.”  But why are they less effective than the clinical trials showed? Is it because we need new approaches to lowering LDL cholesterol?  How low is low enough and in whom?  My read of the story is that we do indeed need new drugs for people with difficult to treat elevated levels of LDL cholesterol and if a little reduction is good, greater reductions must be better. And yes it will cost more but isn’t it worth it?

Do we in fact need new drugs?  Perhaps but we should probably do a slightly better job with the ones we have.  At the moment there are about 24 million adults in the US taking a statin drug.  Many do not take the drug regularly.  At least one simulation study (J Epidemiol Community Health. 2010;64(2):109-113) suggests that increasing compliance and adherence from the existing rate of 50% to 75% would dramatically reduce cardiovascular deaths.  Improving communication between patients and their healthcare providers might be a reasonable step in improving adherence and compliance to therapy.  Although it would seem to be in the industry’s  interest to improve compliance (and sell more drugs), at least three major players seem more interested in identifying the next blockbuster.  But the problem simply does not lie with the patient.  Failure on the part of the prescriber to escalate what appear to be inadequate statin dosing has also been shown in numerous studies.  Again, it would seem to make more sense to improve how we use our existing treatments than developing new and far more expensive treatments.

All current guidelines for cholesterol management include stipulations for relative risk but none to my knowledge suggest lowering LDL cholesterol to the levels that are described in the story.  Are there data to support lowering of LDL cholesterol (regardless of risk of cardiovascular disease) to, “well below 50” as was described by one of the clinician consultants for Sanofi?  Shouldn’t we know the answer to that first?

And finally, let’s think about costs.  How much do you suppose a monthly injection of a monoclonal antibody will cost the consumer?  If the current crop of injectable monoclonals used to treat rheumatoid arthritis are an indication, we are talking about $2-3,000 monthly.  The cost of a generic statin is less than 10% of that.  The investment of $70 million for a plant seems like a good one since it would only take about 3000 patients to recoup the cost of the plant.

As the story suggests, the race is on.  But do we really need to play?

My reactions, and those of Scott and DeMonaco are quick, first-read reactions.  We didn’t know the NYT story was in the pipeline.  Kolata and the Times had a lot of time to prepare their story and far more horsepower to put their story together than we three individuals did in giving quick reactions to something about which we had no advance notice.  But our reactions, our off-the-back-of-the-envelope questions about framing, about evidence, about balance should be helpful to other journalists and to consumers (news consumers and health care consumers) as they read or hear news stories about “fevered races…public health impact…exciting results…blockbuster…might be used by one in four adults.”

——————————–

Follow us on Facebook, and on Twitter:

https://twitter.com/garyschwitzer

https://twitter.com/healthnewsrevu

 

Comments

Marilyn Mann posted on July 10, 2013 at 2:38 pm

Hi

I’ve been following these drugs, which *if* they pan out could be useful for (1) people who are statin intolerant, (2) secondary prevention, (3) very high risk primary prevention (e.g., certain people with genetic lipid disorders).

However, I groaned at certain aspects of the article. The individuals with very low LDL (e.g., 15 mg) are not very useful as examples because only a few such people have so far been found. A recent article that was published in Atherosclerosis in May 2013 said that three such people had been found. Not much information can be gleaned from three people.

So far what is known about these drugs is that they lower LDL and that based on the short-term trials that have been done so far, no major safety problems have been found. The first outcomes trial is just getting underway, and results won’t be available for several years. Until those results are available we just don’t know if PCSK9 inhibitors will prevent heart attacks and strokes, either on their own or when added to a statin.

So far as I am aware the FDA has not said anything publicly about whether the drugs will be approved based on LDL-lowering. My strong preference would be for the FDA not to approve the first PCSK9 inhibitor until information on clinical outcomes is available.

People are excited about the drugs because of the epidemiological studies by Helen Hobbs and colleagues that show that people who carry one loss of function gene variant, who have a lifelong reduction in LDL, have a reduction in heart disease. However, we just have to wait and see if the drugs are shown to be safe and effective.

If the drugs don’t pan out, journalists like Gina Kolata will be long gone, on to the next exciting breakthrough.

The discussion of the people with heterozygous familial hypercholesterolemia is somewhat misleading (heFH). If the disease is identified early, and the person starts taking a statin early, their risk of suffering an early heart attack is greatly reduced. The people who run into problems are the ones who are statin intolerant or who are not diagnosed until middle age. Also, people with homozygous FH, whose LDL is 500-1000 mg, can’t be managed with a statin alone and usually have to have LDL apheresis.

I have never before seen a person with heFH say that they would not have a biological child for fear of passing on the disease. I’m sure there are people who feel that way, but they are the exception to the rule. Most people with heFH don’t fear having children. They just make sure their children are tested early and started on a statin. The exact age to start is somewhat controversial.

The article reminded me of the one she did on Alzheimer’s Disease drugs, which highlighted people with a mutation that causes very early AD (not the typical AD that occurs in the elderly).

Marilyn

David Grainger posted on July 11, 2013 at 2:46 am

The “first reaction” responses you quote to the NYT story are, in my opinion, too negative.

First thing to note is that approving drugs on surrogate end-points or risk factor reductions is almost always wrong, and in fact something regulators consider only exceptionally. So in that sense, the concern about the hype surrounding PCSK9 inhibitors (which, so far, have only data showing reduction in such a surrogate / risk factor – LDL cholesterol – in 12 week studies) is justified.

BUT LDL cholesterol is an exception. Even to a skeptic on the use of surrogates, such as myself, LDL cholesterol is proven to be a causative risk factor and therefore suitable for use in this way. If high LDL cholesterol (unlike low HDL cholesterol) is considered a fully validated surrogate, then the hype is more justified.

Further, clinical evidence from the use of statin suggests there is no floor to how low LDL cholesterol should be taken – each further lowering leads to incremental benefit (AfCAPS and TexCAPS were the first studies to show that in the late 1990s). The only question about treating already low LDL cholesterol is that such people might already be at sufficiently low risk to benefit enough to overcome any safety signals or the health economic question to justify treatment.

And we also know that the mechanism by which LDL cholesterol is lowered or raised makes little difference (on efficacy – side-effects are a different matter). So again, the confidence that PCSK9 inhibitors will lower heart attacks is justified.

If you want reasons to be negative, they are very different ones to the use of what many consider a sufficiently validated surrogate end-point:

(1) There is only 12 week safety data for a product that is slated for chronic use. There has to remain a risk that side-effects that could kill these antibodies will emerge in longer Ph3 studies. Again, though, I dont see that as a risk as big is typically seen for small molecules in Ph3. The reason? These are antibodies, whose specificity is virtually garunteed, and individuals with PCSK9 mutations lack exactly the interaction that these antibodies block – without side effects in the long term.

(2) As Marilyn Mann noted, these drugs will only be useful in those individuals who are statin intolerant or for whom statins are in effective at achieving goal LDL cholesterol. True. But thats a massive population globally, Particularly among diabetics, whose risk of CV events is much higher anyway, achieving a normal lipidemic profile with existing drugs is often impossible. Across the wider population, 20% or so of individuals ‘at risk’ of CV disease either cannot tolerate statins or do not achieve goal LDL cholesterol with them. So there are many millions who would benefit from a new approach to lowering LDL cholesterol – and for whom payers will pay for an expensive new drug on top of or instead of generic statins.

(3) Dose and cost. A closer look at the Ph2 data suggests that as much as 470mg of antibody every 2 weeks will be required. With cost of goods high for biologic drugs, that means the cost per dose will have to be very high to achieve a premium margin justifying the research costs. In the end, health economic considerations could limit the use of PCSK9 antibodies more than efficacy or safety, or the inconvenience of bi-weekly trips to the doctors for an injection.

(4) Most intriguingly, the size of the Ph3 program Sanofi have designed, with 20,000 people enrolling, is a concern: good drugs that induce a step-change in clinical benefit can be registered quickly in smaller studies than those that show only incremental benefit. It one really believed in the power of the anti-PCSK9′s would we not have seen a smaller trial to achieve early registration? Some industry commentators might see the huge Ph3 programme as a sign of confidence – I, though, see it exactly the opposite way.

So four very different reasons to view the ‘hype’ of the NYT article as unjustified. The first two are not real concerns. Even #3 and #4 are probably insufficient to question the potential for this medicine.

In reading the article, I was left thinking: can I think of any other medicine, recently approved or in Ph3 that offers this kind of “game changer” potential for improving public health? And the answer I came up with was ‘no’. For sure, its still all potential, and it could easily all go wrong, but that alone is enough to justoify front page NYT news. The public deserve to know about the potential advance thats in the pipeline, and overall the degree of hype relative to the potential of the drug is nearer to being justified here than with almost any other health story I have read in mainstream journalism over the last few years.

So it s a ‘well done’ to Gina Kolata from me.

    Gary Schwitzer posted on July 11, 2013 at 10:59 am

    David,

    Thanks for your note.

    We can agree to disagree.

    But it is worth noting that we view the story from different perspectives. You represent an Index Ventures company. I’m just an old lifelong health care journalist. You see game-changing potential. I see overstatement and imbalance. You talk about Phase 2 data and Phase 3 study design; I criticize the new story for failing to reveal any of those details – only a couple of patient anecdotes.

    We stand by our critique of the journalism.

    Gary Schwitzer
    Publisher

Karan Chhabra posted on July 11, 2013 at 11:59 am

One of the quotes mentioned what might happen when they come off-patent. In fact there’s a strange legal quirk that means biologics may never go generic: http://theincidentaleconomist.com/wordpress/the-monoclonal-antibody-revolution/

This ought to support your point that cost will prevent these from ever taking off like statins. If they make it to market, I’d imagine they’d only be used in very severe cases.

Linda Williams posted on July 11, 2013 at 2:23 pm

Kolata has been a problem at NYT for a long time:
http://www.mindfully.org/Reform/Gina-Kolata-Dowie6jul98.htm

Paul Scott posted on July 11, 2013 at 2:59 pm

Hi David

I hope my reaction is indeed too negative and the drug turns out to be all that it’s made out to be in this story, a story that is very annoying to anyone who approaches it skeptical of drug development motives thus far and unconvinced of the mechanistic relationship between LDL and heart disease that you describe.

Like Gary I’m also only a scribe, so please forgive the biology novice, but having spent time with Ronald Krauss I am not certain how one can say that LDL is a causative risk factor in heart disease when LDL is not even a homogenous entity. There are what, 11 subtypes that act in different relationships to heart disease. I’m sure I don’t have to tell you that when particle types are taken into consideration large bouyant LDL has no relationship to heart disease and total LDL has only a small relationship. Small dense LDL do, of course, but as part of phenotype expressed with elevated HDL.

I don’t see how LDL can be a determinative causative agent when people live long lives with high LDL and 75 percent of people hospitalized for a heart attack in a 2009 UCLA study had healthy cholesterol. The effectiveness of statins such as it is — they confer benefits o heart disease patients with normal cholesterol — could be due in part to their anti-inflammatory properties. Surely we are drugging down a biomarker that is sloppy shorthand and plays no causative role in heart disease if the diversity of the marker is to be taken into consideration.

Backing out of my shaky knowledge of lipid science, it is healthy one hopes to be skeptical of the idea that a natural substance that is important to the body for a host of functions can be drugged nearly out of the blood stream with no cost to the patient. That’s not even getting into long-standing concerns about low cholesterol and elevated all cause mortality, cancer, cognitive problems and muscle pain. It’s also presumably only rational to be skeptical that a drug such as this would not find its use expanded over time, as has been the case with statins and nearly every other blockbuster drug. The past is prologue, right?

Beverly Richards-Smith posted on July 17, 2013 at 3:51 pm

First, I agree with Paul Scott’s skepticism regarding the relationship between LDL cholesterol and heart disease. Also, as a biologist, I cannot help suspecting that a class of drugs (the statins) that interferes with an essential reaction in mitochondrial energy production is causing long-term harm to muscles, including cardiac muscle, even in individuals who do not experience muscle pain while taking them.

With regard to the anti-PCSK9 monoclonal antibodies, I would not assume their long-term safety without knowing 1) the precise mechanism(s) whereby reducing PCSK9 lowers LDL cholesterol; and 2) any other functions of PCSK9. Inactivating the orthologous gene, Psk9, in mouse models has been shown to reduce LDL cholesterol to nearly undetectable levels and to increase the rate of clearance of labeled cholesterol; however, it also was found to interfere with liver tissue regeneration, to cause hypersensitivity to statin drugs, and to impair glucose tolerance and produce abnormatlities of the insulin-producing pancreatic beta cells, among other effects. (For more about the mouse models, see the reference lists at http://www.informatics.jax.org/reference/allele/MGI:3577021 and http://www.informatics.jax.org/reference/allele/MGI:4943496.) Gina Kolata could have used these publications as sources of balancing information in her article; their abstracts and, in some cases, links to the papers themselves are accessible via PubMed.

ed posted on August 4, 2013 at 7:55 pm

I’m a practicing doctor and have been earning CME (required educational activity) on PCKS-9 blocking antibodies for many months now. It’s available free on-line and touts the manufacturers point of view that this is the next big thing. I’m also getting promotional videos posing as education for zapping renal nerves with microwave catheters, the next multibillion dollar treatment for high blood pressure.

It’s interesting that the investment community considers LDL lowering by any means proven. Xetia (ezietimibe) makes money and lowers LDL without preventing heart attacks. If anti-PKSC-9 ends up costing $2000 a month that’s not 10x more, it’s a hundred times more expensive than a $20 statin. Statins are the best proven of all preventive meds (even though Cochrane found the benefit to be 1 in 1000 per year in primary prevention).

Ed