Amgen 4Q15 – Repatha (evolocumab) approved in Japan in January; CV outcomes data expected in 2H16; Revenues not broken out but product “off to a strong competitive start” – January 29, 2016

Amgen provided its 4Q15 update late last week in a call led by CEO Mr. Bob Bradway. Management announced that Amgen’s PCSK9 inhibitor Repatha (evolocumab) was approved in January in Japan for patients with familial hypercholesterolemia or patients at high risk for cardiovascular events who do not adequately respond to statins (similar to the US indication). In addition, the company reiterated that the product’s phase 3 cardiovascular imaging (intravascular ultrasound [IVUS]) and outcomes data are expected in 2H16, which we first heard at JP Morgan – this is an earlier timeline than we might have expected and we share Amgen’s optimism for the results given the impressive LDL reductions seen with Repatha. While the company continues to not break out Repatha sales, management highlighted that the product is “off to a strong competitive start”, emphasizing that brand recognition is strong and that more than 80% of US commercial lives currently have access to Repatha. However, Amgen noted that strict payer utilization management criteria are limiting the uptake, which is reflected in the IMS scripts. The company continues to work with payers on this issue, and we are very eager to know what the impact has been on access for patients with diabetes.

Outside the US, management highlighted that reimbursement negotiations are ongoing in Europe, with the expectation that reimbursement will be added in many countries throughout 2016. Specifically, the call pointed to national reimbursement in Spain – occurring earlier than expected – as notable progress. Looking forward, Amgen predicted that Repatha growth will remain steady in the near term, with “breakaway potential” once the outcomes data is in the label – we’d be excited to see a type 2 diabetes indication brought into greater discussion at this time point. Overall, the PCSK9 inhibitor class has received an impressive amount of enthusiasm with the launch of Repatha and Sanofi’s Praluent (alirocumab) (see the latest from the two companies from our JP Morgan coverage). Reimbursement positions are looking relatively favorable for the class, although this patient eligibility issue will be an important roadblock to overcome – for more details on this, see below for Q&A.

Questions and Answers

Q: On Repatha's uptake, you noted the happiness on your part with the share gains. But are you disappointed overall with the size of the pie at this stage? I know you're seeing reimbursement headwinds. But nonetheless, I guess is this on a more shallow trajectory than you thought? And assuming we do get the positive outcomes data toward the second half of the year, should we see an immediate benefit from that, or would you think it would still take some time to work with payers to work through these headwinds?

A: I'm glad you see the relative performance in the marketplace. The hub we put together was clearly done because we understood there was going to be a little bit of time before the payers made a decision around formulary approval. And to me, the hub has therefore been a surrogate to the level of prescriptions that cardiologists and endos and primary care physicians are prepared to prescribe. We are seeing a really robust level of prescriptions coming through the hub. So to me, that continues to give me great confidence in terms of physician willingness to prescribe this drug for patients who fit inside the label. The prescriptions themselves in terms of both the NBRxs and the TRxs are continuing. It's clear that the utilization management criteria in place is restricting the number of prescriptions that get dispensed. And we are working with payers at the moment to make sure that patients who are eligible actually get access to these drugs.

Q: On Repatha, when you look at the subtleties, either reimbursement or populations or clinical practice, can you compare the EU and Japanese markets to the US, assuming that you do get outcomes data this year?

A: So one has to remember that outside the United States, once access is granted, physicians are not making a decision on anything other than a clinical decision around the value for patients. So there is no economic decision once you have access in Europe and in Japan. So the negotiation is to get access as quickly as possible, and then to move patients onto the drug as physicians prescribe. From an inside the US perspective, yeah, it's the number of prescriptions we're seeing across the range of physicians who have been prescribing that is encouraging.

Q: On the US market, what can you tell us in terms of leading indicators of demand? In other words, like physician prescribers or visits to your hub or things like that? Just want to get some demand metrics beyond TRx.

A: What else should we be looking at to see future growth in the marketplace. To me TRx's and MDRx's in terms of new naive patients are the most important thing to measuring in terms of the growth. But with a new launch like this where the plans are put into place at the formularies, one has to remember that IMS only reports dispensed prescriptions; i.e., prescriptions that come to a pharmacy or a specialty pharmacy and the patient actually walks away with the drug. What you have to be able to look at inside that data is how many prescriptions get to the pharmacy and how many are rejected versus how many are abandoned. So we are seeing that the majority of prescriptions getting to the pharmacy at the moment are being rejected, rejected because the prior authorization process has not been properly completed or there's some outstanding information and patients have to go back and get some more data. All patients are seeing the copay at this particular stage as being too high a copay because the product is not properly on formulary yet, and they abandon the prescription and they walk away.

So the data you're seeing is really important, but you have to understand that the majority of prescriptions getting to the pharmacy are either being rejected or abandoned at the moment while the plans complete their process, which is why we spent so much time with the plans at the moment showing them the number of eligible patients who are on label getting to pharmacy and not getting product. It really is a concern. And when you think about the potential Repatha patient, these are patients that are at risk right now for a cardiac event and therefore early intervention is essential, so we're spending quite a bit of urgent work with the payers.

Q: I just wanted to come back to the Repatha management criteria that are limiting uptake. Can you elaborate a little bit on what the primary causes for rejection are? Give us any sense as to the evidence that you are making progress in addressing some of these issues. And then given what those issues are, what gives you the confidence that the cardiovascular trial data may ultimately resolve that as a barrier?

A: Utilization management criteria obviously differ plan by plan, but they include things such as the patient must restate naturally tolerated statin dose or someone requires you to be on one or two statins. Some of them require you to have done statins plus a step-through to Zetia. There's clearly a requirement around your LDL-C level – they have to be at a certain level. But probably the more complex thing is the prior authorization documents. There are five pages of handwritten stuff that physicians have to find out about. And most of the rejections are because the form is not properly completed, and this is the time to collect the data. So as we get the process running a bit more efficiently and hopefully moving from paper to an electronic process, the prior authorizations could go faster. And then as we show the payers the impact of too draconian a utilization management criteria not getting to the right patients, we will see some changes there. The outcomes data of course will dramatically change the value of this particular drug, and we do expect to see some changes in the utilization criteria once we have that turn.

Q: If you don't mind, can you spell out the sales of Repatha last quarter? I know it's a small number, but just a housekeeping question there.

A: As to the Repatha sales, we're not breaking those out by line item at this time.

Q: Just thinking a little bit more about the outcomes data for Repatha expected later this year, where do you think the range of outcomes can fall in terms of how we extrapolate the reduction in LDL capability of the drug to what we've classically seen in terms of CV reduction? And maybe think about what the range should be based on that? How it could deviate from that classical extrapolation, whether it's due to different types of patients or just the trial design? How we can think about that?

A: This is the kind of subject than one can sit around and talk about for many, many hours with experts, which I've done. And I think the best thing I can say is that we have a remarkably linear relationship that we've established, most recently with IMPROVE-IT, extending that line in just a remarkably linear fashion. Extending the line that was created by statins and by other interventions, such as ileal bypass surgery and so on, and the genetics of course. And so when you put it all together, what you have to believe scientifically is that you're going to fall right on that line in the same way as if you achieved that additional LDL lowering with a statin or with ezetimibe, were that possible. Obviously it's not.

Could it deviate from the line? Sure. I mean it is always possible that some of the foibles of the way that the clinical trials are designed and conducted – there is of course, for example, a treatment lag of some sort that occurs when you start therapies. And so when the study reads out very quickly instead of over a longer period of time, that has a bigger influence, etc., that could make it come off the line slightly in one direction. We also know that the agent does have some activities that statins don't have. For example, there is an effect on Lp(a), which is present in some individuals and seems to be a strong prognostic factor that could make the dot come off the line a little bit the other direction.

So there's some variability that you could expect. But from a scientific perspective, based on the human genetics and everything we know, one would expect that you're going to see a reduction that would be proportional. And that would be roughly this ballpark around the one-third level reduction of the risk. So that's the kind of number that many people would expect – keep in mind the 50% reduction in risk that was suggested by the analyses that were published in the New England Journal – there's a very wide confidence interval around those. And while you can't rule out the possibility that you'll see that big of a reduction, that's a bigger reduction than you would expect to get if you were achieving the LDL lowering that we're achieving with a statin. So it requires some other biology like Lp(a) or something to be going on. And I can't tell you that's not happening, but it sure wouldn't be my base case.

-- by Melissa An, Emily Regier, and Kelly Close