Norah Terrault, MD, MPH: Benefits of Avatrombopag for Thrombocytopenia
DECEMBER 05, 2018
MD Magazine Staff
Norah Terrault, MD, MPHFollowing a report at the American College of Gastroenterology meeting in Philadelphia, PA, in October, which showed avatrombopag (Doptelet) was efficacious in treating chronic liver disease and thrombocytopenia, the Dova Pharmaceuticals-based therapy garnered even more discussion at the Association for the Study of Liver Diseases (AASLD; The Liver Meeting) in San Francisco, CA, last month.
In an interview with MD Magazine® while at the Liver Meeting, Norah Terrault, MD, MPH, director of the Viral Hepatitis Center at University of California, San Francisco, detailed recent analysis of avatrombopag for thrombocytopenia and impeding real-world data on the drug approved for the market in May of this year.
MD Mag: Does avatrombopag reduce bleeding?
Terrault: In a study, the number of bleeding events was low, and so, the study wasn’t powered to detect the difference of bleeding rates. I think that’s a decision that was made early on because in general, the bleeding risks depend on many factors, including what procedures you’re doing, etc. It would have taken an enormous-sized study to show a difference of bleeding rates.
Similar to that, what had been done in a prior study, the avatrombopag study for example, where we focused more on the ability to increase platelet counts and safety.
I think where the controversy is [physicians] ask, ‘What is the right platelet count for us where we should be doing measures to prevent bleeding?’
In the study, we chose individuals with a platelet count below 50,000, and that was just based on really looking at literature that talked about doing procedures in patients with cirrhosis and what a threshold should lead individuals to have platelet transfusions. We chose that based on data from literature that implied that clinicians should be increasing platelet counts in individuals with levels below 50,000.
For example, if you’re doing a liver biopsy, or if you’re going to do an invasive surgical procedure. I think what individuals [experience] in difference of opinion is ‘Does this stick to the right number? I talked to my neurosurgeons and they said the [platelet count] should be 70,000.’ And if you talk to other individuals, some will say, ‘I’m perfectly happy with a platelet count under 30,000.’ The controversy is sort of, when is it indicated?
The study did not attempt to address that. What it’s saying is that we recognize that in clinical practice, patients are getting platelet transfusion in the context of invasive procedures, and we’re looking for where that clinical niche is. We’re trying to come up with an alternative—one that is safer and potentially more convenient.
The study shows that it’s safe. As you know, prior studies with avatrombopag suggested there was an increased risk of thrombosis, and so I think it was really important to look at safety in this study. The other is really to look at what kind of increase in platelet count, what can we anticipate? That helps clinicians make decisions about whether a drug makes sense for them.
I think you’re right in the sense that transfusions are not benign. Many patients in the past have gotten infections via transfusions and all those blood supplies that are safe now. There’s always that risk. Transfusion reactions—the hassle of having to go in beforehand to get transfusions. There was meant to be an alternative in terms of providing an alternative to those risks, and maybe one that’s more convenient.
I think the study sort of is, as Dr. Bernstein highlighted, it is really nice to have an alternative where you can just prescribe patients a pill and tell them, ‘Okay, take the pill, give it 5 days, and then stop,’ you don’t have to worry about scheduling platelet transfusions ahead of the procedure.
It is convenient and nice to have an alternative.
Going back to the platelet levels and not yet knowing what’s going to be that minimal threshold, will there be, or do you anticipate there would be, different minimum thresholds based on the procedure the patient is undergoing?
I think this is up to the clinician. If you talked to my cervical colleagues, for example, they have very fixed ideas on what they think the platelet count needs to be. I have patients at my liver transplant center as they’re going for an ERCP, and they’re going to be getting a procedure where there’s a risk for bleeding and the doc absolutely wants a platelet count over 50,000. Surgeons wanted it over 70,000.
What is the right number? If you look in the literature in terms of platelet count, it is hard to find a lot of data. I think it’s really about clinicians’ comfort level. I think everyone’s trying to mitigate any risk of bleeding, and so it’s probably based on that individual’s experience. They maybe had a bleed, and therefore came up with that threshold, or they’re basing it on the literature.
I think it varies with the procedure, I think it varies with the clinicians, it probably also varies with other aspects of the patients’ bleeding risk. Patients with cirrhosis not only have low platelet counts, but some of them may have elevated thrombin times, and they have some other aspects related to coagulation and/or thrombosis that could be playing a role.
It’s not a simple decision to be saying that every patient with a certain platelet level should be treated; it really is that complex decision that needs to be individualized.
Where do the dental procedures fit into this?
Again, most dentists—at least in my practice—if they have a patient with low platelet counts, and I think below is typically anything under 60,000, typically. [They then] send them back to the hepatologist or to the primary care doc and say, ‘Would you clear the patient for the procedure with their given platelet counts?’ or ‘What would you recommend in terms of management?’
We either say, ‘You can get a transfusion of platelets beforehand,’ and now we can say, ‘you can take the drug avatrombopag instead and get a bump in platelet counts.’ Many dentists especially if they’re going to do major extractions, would want the platelet count to be over 50,000. I think that’s a platelet count that is used fairly commonly. Some use 60,000—between 50,000 and 60,000 is probably the number—and I’ve seen patients who have had significant bleeding after dental procedures.
Again, it depends on the procedure, but I think most dentists are going to do something where if they anticipate bleeding to be an issue, they will want the hepatologist to clear the patient for the procedure and make a recommendation about the need for increasing platelet count—either with transfusion and/or a drug therapy.
So that’s another place I think it [avatrombopag] will be used as the word gets out that there is this alternative.
The drug was just approved in May. In terms of next steps, is there real-world trial data?
I think we’ll gather more real-world experience as more individuals gain more experience with the drug. I believe there are some plans to do additional studies in specific subgroups of patients, including those going through surgical procedures that are higher risk. Within the avatrombopag study, we have a range of procedures, but many of them were among the procedures that would be considered a lower risk for bleeding.
There were patients getting treated for HCC, and patients were undergoing cardiac catheterization, so clearly they had some patients who had high-risk procedures, but I think everybody would want to have additional information around subgroups.
When the prescriber prescribes the drug, how many days is the patient taking it before the patient undergoes the procedure, and is there daily bloodwork done to test their platelet counts?
They usually get the drug or 5 days, so what you do is you prescribe avatrombopag for the individual, they take it home with them, and once I know the procedure date, I tell them to start taking the drug 10-13 days before the procedure.
They’re going to take it for 5 days. Then, there’s a gap when they finish the drug and when they have the procedure, and during that time, the platelet count is on the rise. In my patients, what I do is I have a platelet count on the day of the procedure. That’s to help the endoscopist or help the surgeon—whoever is going to be doing the procedure—to let them know what increase in platelet count they have had.
Is that any more testing than would be required for transfusion?
Sometimes patients do get a platelet transfusion and then the surgeon or the procedurist wants to know the platelet count increase that has been obtained with that transfusion. So, it’s sort of the same principal—give a treatment and then evaluate the platelet count. That would be particularly true for patients with very low platelet count.
I would say that equally often, if someone has a low platelet count and is getting a transfusion, what happens is they just get the transfusion ordered, and then they go straight into the procedure and they assume the platelet count has gone up. They don’t need to repeat the platelet counts after the transfusions. It’s not mandated that they check the platelet count after a transfusion.
In terms of cost, when you factor in the additional testing, is it still going to cost less than a transfusion?
I think the cost of doing a platelet count is trivial. I think the real cost, when you’re looking at a cost comparison here, is related to the time, the people that are involved in the transfusion, the facility fee to do the transfusion, and then the cost in terms of if there were any adverse events. You have to factor those in versus the cost of the pill.
That is really the cost driver, and not the platelet count. That’s a very small cost in contrast to getting set up and having a facility to do a platelet transfusion.
Just as an example where I work, if we have a patient that we need to have a platelet transfusion before they are going to get their endoscopies, that we’re going to band their varices or something, and the endoscopist wants the platelet count over 50,000. In our institution, I have an assistant beforehand fill out the paperwork, send it over to the transfusion center, the center has to order the medication, the patient has to go to another facility to get the transfusion before they can have the procedure—do you see the complexity with that?
When you talk about the costs of the manpower to set up transfusions and the actual doing it, that is sort of an inconvenience for the patient as well. For patients, it’s much easier to just take a pill for 5 days before you come in and just walk in for your procedure.
There is a convenience element to this that is quite attractive. In institutions where the procedures and transfusions are done separately, it can be quite clumsy and complicated.
There are also very subtle areas where it’s really, really relevant. For example, I work in a transplant center where we often have patients that are waiting for their transplants and they’re having recurrent procedures at the hospital. The more platelets they get, the more refractory they get to the platelet transfusions because they develop antibodies to the platelets, and so then they don’t even get an effective bump in their platelet count. Then, they’re really are using a lot of platelets.
There are some situations where having to take transfusions of platelet can actually diminish their effectiveness over time. Those aren’t common patients, but when you have those patients, they can be very challenging to manage. Having an alternative to having to give transfusions is helpful in these challenging patients.
Is there a “catch” with this treatment?
It’s a new drug, and with any new drug, you write the prescription and there’s an authorization process to go through. So, like any new drug that comes out, you write the prescription, and then there seems to be some challenges in getting the prescriptions filled. With new medications, there’s a cost—it’s not pennies per pill.
I think that can be a barrier to new prescribers—going through the authorization to get the drug—but there is support to do that.
The other aspect is, if you look at the study itself, not every patient gets a bump in his or her platelet count; although the vast majority of patients had an increase in platelet count, and it’s a pretty uniform increase of about 20,000. If you look at the data, there are some patients who didn’t get that bump. I think clinicians will want to know, going forward, how to identify upfront if they are going to have a patient who responds or doesn’t get the same response.
There are 2 doses that have been approved, and maybe we need more flexibility with doses in order to ensure every patient gets the appropriate rise in their platelet counts. I think what clinicians will want is more information about how to predict patients that are going to respond and is there room for more dosing options because right now there are 2 doses approved.
The question is, ‘Could we potentially have higher doses to have bigger bumps?’ and that sort of thing. Clinicians will appropriately say, ‘Well, this is really good, but I want more,” and ‘I’d love to see more real-world data regarding the proportion of patients who get this nice bump of 20,000 platelets,’ and ‘Are there any patient factors that influence the response,’ and things like that.
That’s the kind of data providers will want as they use the drug and it becomes more available to them. Of course, it doesn’t work perfectly in every patient.
A lot of what we’ve learned about many of our drugs in liver disease is first the drug is approved, and then with use, we learn more about the drug in both its strengths and weaknesses and where we want to see more data.
Any additional comments?
There are other product out there— eltrombopag and lusutrombopag. If anything to me, that just points to me the need for such a drug in this space. The fact that we have 2 drugs that have been developed specifically for this purpose of managing thrombocytopenia in the setting of patients with chronic liver disease going through procedures.
There clearly was a need there, and I think that’s why you hear this excitement about people being happy there is something to offer. It’s a common clinical issue, and it’s nice to have some options now.
Predictive Analytics: How Big Data Will Improve Outcomes and Efficiencies in Diagnosing and Treating Patients