Forging a Path for Early Clinical Development in Sickle Cell Disease

Distinguished Pfizer researchers David Beidler and Debra Pittman share their experience in Sickle Cell Disease including:

Using longitudinal biomarker data from early clinical development decision in sickle cell disease (SCD)

Building the patient perspective into clinical trials: development of the vaso-occlusive crisis (VOC) day endpoint through the use of the electronic patient-reported outcomes (ePRO)

Bring it all together in ELIPSIS: a natural history study to evaluate longitudinal changes in ePRO, actigraphy, and biomarkers in SCD patients

 

Transcript

Speaker 1:
Today. It is my honor to present to you Debbie Pittman and David Beidler from Pfizer who will be presenting, “Forging a path for early clinical development in sickle cell disease.” From biomarkers to clinical endpoints.

Debra Pittman:
Okay. Thank you and thank you for the opportunity today to speak about the work that we’ve been doing on early clinical development in sickle cell disease. David and I are going to be doing a tag team presentation covering a couple of topics. First, the part of the talk, we’ll give a very brief overview of sickle cell disease so we’re all on the same page as we move forward and then we’ll go on to talk about where early clinical development is in sickle cell disease and then proceed to talk about some specific examples where we bridge no-clinical data to clinical data using an example of phosphodiesterase inhibitor, PDE nine and that the majority of the talk at the end will be focused on a phase zero longitudinal study on sickle cell disease.

Debra Pittman:
Please feel free to ask questions as we’re going through the talk. So briefly, sickle cell disease is a rare disorder. It affects approximately 100,000 Patients in the US but it’s really even a larger number of patients outside the US greater than 2 million worldwide. It’s a chronic condition with a substantial morbidity and mortality and a clear unmet medical need. The life expectancy in the developing world is 40 to 50 years. However, in the US it’s between 40 outside the U S it’s between 20 and 25 years. It’s a very complex pathophysiology. They’re currently two approved therapies for sickle cell disease, hydroxyurea, which was actually approved early in the 1990s and it reduces these painful crisis about 45%. But it does require extensive optimization of the therapy. There’s poor compliance in terms of patients taking the hydroxyurea and in addition there’s a black box warning. L-glutamine was recently approved.

Debra Pittman:
It’s well tolerated. I think they’re still out on what the clinical benefit is, but patients, there is some uptake with the patient. But patients have these recurring pain managements and the current therapy for that is primarily hydration and pain medication. So sickle cell disease is caused by a single point amino acid change at glutamic acid, position six and the hemoglobin molecule. And what this does under conditions of hypoxia causes the globin chains to depolymerize. And this forms the long side rolls and what this results in is changes in the cell membrane and that’s what gives this, the typical sickle cell that you see and how it describes the disease. But it’s not just a sickling of the cell that causes problems.

Debra Pittman:
If membranes can flip exposing [inaudible 00:03:09], which can impact the coagulation system. There’s homolysis, which are actually contributes to releases Hayne and that contributes to the inflammation.

Debra Pittman:
The cells also become sticky, they stick to the endothelium, they form heterotopic aggregates, and what you get is a vascular blockage. And it’s this vascular blockage that causes a vaso-occlusion. If patients undergo these painful crises, that sometimes requires hospitalization and you’ll see from some of our later data how patients choose to treat these painful vaso-occlusive crisis. But the vaso-occlusion also causes other aspects of the disease. There’s downstream complications, there’s end organ failure. There can be an acute chest syndrome, there can be stroke, there can be skin ulcers as well as osteonecrosis. So there’s clear on that medical need, a very complex pathophysiology when you’re trying to treat and don’t want to just treat the VOC but you want to get to the root cause of the disease and understand what the patient’s need is in terms of effective therapy.

David Beidler:
So now we’re kind of studying for stage and we’ve been using it actually those Pfizer people, this is Nick’s Clarke idea. So it’s a nice visual for what we feel. The current state of sickle cell clinical trials and recognition by the medical community and treatment really is the tip of the iceberg. There’s a lot more that we’re going to talk about. The clinical end points are about medical utilization. A crisis is only a crisis when you go to the ER or hospital and get admitted. We feel that that’s obviously a just a small fraction of what’s going on. Length of hospitalization stay or IV opioid use have really been used in interventional studies and are really regarded within the established community as meaningful issues with the sickle cell disease as well as some of the end organ damages.

David Beidler:
Whereas what we do know from interviewing patients and from their perspectives is there’s a lot of things going on underneath the water and we have things like at home their fatigue, their quality of life. They can’t get up, they can’t go to school, they can’t do their work, they can’t have their social life be the way they wanted. They’re medicating at home. They have anywhere from massages to relaxation to yoga, anything to try to relieve the pain and reduce the onset of these crisis. So again, their daily life is severely impacted and underneath all this of course is there really are no validated or really qualified biomarkers for sickle cell disease. There are a lot of studies that have looked at before and after and try to understand what is elevated in sickle cell relative to healthies. But nothing really that the regulatory agencies recognized at this time.

David Beidler:
So again, a lot of issues here around building in things at home and in the outpatient that that just aren’t being recognized in the clinical studies or a validated in the medical community. So in addition to knowing that my role a lot in the precision medicine and the translational part is, we try to learn early in a clinical as early as possible. We want to build this value of a confidence index that we call. Just a nice way of saying that we think we’re testing something that is meaningful. And the earlier we learn the better. In the old and traditional system you would put a drug into the clinic, you would do PK and tolerability, is it safe? Is it kind of acting the way you want? And you really only learned maybe when you hit your clinical endpoints, your proof of concept.

David Beidler:
But between there is a very low confidence and if you’re relying on hope. And you’re kind of putting it off saying we hope to hit those POC end points and we could have said it’s tolerable, but you may get it in, several years in and be disappointed. And Pfizer and other companies that come up with a lot of strategies to build in early confidence around understanding natural history, the patient perspective and a lot about biomarker criteria. And anywhere from that you’re engaging the target to modulating a pathway or a mechanism all the way to is it really relevant to a disease? And that’s why we have some milestones around either proof of mechanism or something analogous to a surrogate signal of efficacy. The earlier in the clinic that you can learn, even in healthy volunteers, that your pathway is engaged, the more confidence and more value you’re providing to your asset.

David Beidler:
And that can be very important in the competitive environment, both within your company and outside of your company. So, the patient perspective, the need for this type of early value really kind of drove some early questions. And again, the schematic here that we’re looking at is, in addition to the PK by using in vitro assays I think that translated into the early pre-clinical models, are we engaging our targets? Are we hitting a pathway? Is that pathway then feed into disease? I hope it does. Your confidence and rationale is your pathway should have some relevance to the disease. Hopefully those overlap quite a bit. And of course then how does that relate to the clinical endpoints? And especially in the rare disease area, there’s usually a very little data available linking all this together. So we try to build as much of this in place prospectively so that we actually have not just, let’s do some biomarkers, let’s see what happens and hope for the best.

David Beidler:
We try to actually create some criteria and make it a little bit of a push so that our management or competition sees we’re starting to do this, this is important to us. So, very early on in the phosphodiesterase here I’m going to have Debbie introduced this asset.

Debra Pittman:
Okay, so in the next few slides I’m going to focus on the non-clinical data around the PDE9 inhibitor. As I mentioned earlier, hydroxyurea is used in the clinic and its primary mechanism of action is to increase fetal hemoglobin to shift the polymerization during that hypoxic state and therefore reduce the sickly. But hydroxyurea probably has other activities. It’s anti-inflammatory. It can also act as a nano donor and it can impact the soluble guanylate cyclic pathway. Within this pathway there’s also an enzyme called phosphodiesterase nine or PDE9, which is responsible for the degradation of cyclic GMP to GMP and it’s normally expressed in the brain but also in sickle cell patients, it’s elevated both in neutrophils as well as erythrocytes. And it’s thought that this elevation leads to a degradation of the cyclic GMP and contributes to downstream activation such as an increase in the selectins through the inflammatory process on the endothelial cell, but also an increase in the cellular aggregates through the white blood cells.

Debra Pittman:
And this would lead to an inflamed situation within sickle cell patients. So our working hypothesis was that inhibiting phosphodiesterase nine could mediate some of these downstream events in terms of cellular aggregates as well as inflammatory markers. We routinely use two non-clinical mouse models in our studies. One is a TNF induced inflammatory model in wild type mice and that’s really just induced to general inflammatory response. But we also use a model that’s a sickle cell mouse model, the town’s model of sickle cell disease, and this is a transgenic mouse model in which the globin genes of mice have been removed and a human mutated globin genes have been introduced into the mouse and this mouse model actually exhibits much of the pathophysiology that’s observed in human sickle cell patients. So in part of our studies with the PDE9 we wanted to look at the effect of both short term administration and chronic administration in both of these mouse models and look for effects on pharmacodynamic markers that might actually be translated to be used in early clinical trials.

Debra Pittman:
If you look on the left hand side following a chronic dosing scheme here we can see compared to the vehicle are the dosing of the PDE9 alone. We do see a decrease in plasma soluble E-selectin and in combination with hydroxyurea we actually also see an additive effect. We’ve also looked at this in terms of looking at neutrophil platelet aggregates as well as cellular aggregates, and that’s shown on the right and I’ll go into a little bit more detail in the next two slides on how we assess aggregates in the mouse model. We use intravital microscopy to look at cellular aggregates. In addition, after we’ve taken blood afterwards, we take blood for analysis of markers, but in this what we do is we expose the cremaster muscle and we have specific antibodies that recognize neutrophils as well as specific antibodies that recognize platelets.

Debra Pittman:
So we administer these antibodies to visualize cellular aggregates in the occasion, and then we can do quantitative analysis of this. You can see on the left hand when you look at vehicles, you can see that in this sickle cell mouse model there’s poor flowing of the blood. You see large aggregates flowing as well as adhesion to the end of the helium. As you add, in this case the PDE9 inhibitor, you can see that the blood flow has increased as well as its cellular aggregates have decreased and we also see that in combination if we use the PDE9 inhibitor with hydroxyurea. We can then go on to quantitate this and we generally looked at 25 to 35 venules within a mouse and then we do a quantitative analysis where we can see the administration of a PDE9 inhibitor, we see a decrease in neutrophils adhere to the endothelium, but we also see a decrease in neutrophil platelet aggregates with the PDE9 inhibitor alone as well as in combination with hydroxyurea.

Debra Pittman:
So based on these studies we would look at neutrophil or monocyte-platelet aggregates in an early clinical study as well as monitor some of that soluble factors such as a soluble E-selectin, P-selectin based on the data that we’ve gotten from the non-clinical model to try to identify potential from the Chi-dynamic models in the clinic.

David Beidler:
That’s where I pick up. So as the PM lead, my job is to try to take a lot of the findings from Debbie’s group and try to make them feasible for going into the clinic. So we had a little bit of data from a pan-selectin inhibitor GMI-1070 Rivipansel. They had published a pilot model, a study where they actually dosed the stable sickle cell subjects. They weren’t undergoing a crisis, but their basal levels of activity were there. And when they dose their inhibitors and then collected samples at the 4, 8, 24 and 48 hours, they did see a movement in a panel of biomarkers that were relevant to the pathophysiology that Debbie just described to you. And with the all are included here, what was notable for us was that with this addition of this inhibitor in the stable subjects, they did see a reduction in the soluble E-selectin and P-selectin as well as the ICAM.

David Beidler:
So again, a more anti [inaudible 00:13:47] property was chronically produced with administration and well they didn’t look at the neutrophil in their monocyte platelet aggregates, they in fact did see a reduction as well. So I think you’ve seen where I’m heading with this. We’re starting to see some nice translation from some of Debbie’s groups work in the town’s model to some early pilot studies with a drug that has a similar mechanism of action. And so the big question that would have came for our phase 1b study, which I’m going to mention in a minute, is can we generate a similar biomarker in the stable state in just again, this is by getting that mechanism in place, can we create a less adhesive, less inflammatory state that might then have an impact on these VOC crisis? So this is where Sanguine comes in, so that’s why we’re here today.

David Beidler:
So we need patient samples and we just didn’t need a bunch of samples one at a time. We needed longitudinal patient samples because we don’t do drug studies with just one dose at one day and then we have them go. You don’t understand the longitudal nature of the changes. So we were able to have, Sanguine was able to recruit 43 subjects through their mobile formatable groups and we were able to bleed them about every two weeks or so apart. So that had, as you can see below what we had a panel of longitudinal data from, I think it was about 36 patients, at least two to five or so apart and now you’re giving your [inaudible 00:15:02] really good longitudinal data. They can do a lot of work with that to understand things around, do you in fact have a longitudinal variability that’s acceptable? And can you now predict and catch your estimate a cohort size? Moving forward into the phase 1b that it will at least have statistical significance.

David Beidler:
And we will use some of the data from the pilot study and from Debbie’s work to understand what is biologic significance. If it’s going to take 80 or 90 subjects because of high variability that may not be feasible. The good news was when our statisticians looked at some of our longitudinal data, they were able to in fact show us that we didn’t need large cohorts, that we could get some answers about significant movement in some of these biomarkers with some fairly small under 20 cohort sizes. So just very briefly with the setup of the study design, we did five mg BID and 25 mg BID. And again that’s twice daily dosing. We had four weeks of treatment and then we had a four week follow up. So I’m going to show you some biomarker collections kind of before, during and after that active treatment period for these two doses.

David Beidler:
So what I wanted to, wait again we did a very similar panel of biomarkers that have been done pre-clinically and in that pilot study and what you see here is that, for example, with the monocyte platelet aggregates at the end of treatment a day 29 we achieved some statistically significant reductions in the aggregates. Now there is a lot of noise, but again, we were able to achieve the significance and we saw roughly about a 40% reduction from where there were pre-treatment. Fewer aggregates and think of those videos, if they were a human video that you have fewer of these aggregates floating around and flowing through the cells, would that potentially be a more conducive to avoiding these occlusions and the microvasculature causing the crisis. Same thing with a neutral for platelet aggregates. We did see again roughly 30 to 40% again, a little bit of a dose response here, very small.

David Beidler:
So again, small study, quick answers. We’re trying to get with a phase 1b study that we’re getting our mechanism engaged. And even here with E-selectin, we’re reducing solubility selecting levels. Again, endothelial activation is a little reduced. So reduced activation, reduced aggregation, is that now a more conducive state to avoid these VOC crisis? So that really was kind of the goal of this study was that chronic administration at least even in the stable state because we’re pointing for prophylactic use, they would take these drugs continuously and hopefully this will then result in fewer prices in a given time period. So this is just to kind of sum up this stage of it. So what we’ve done with the translational part is we’ve put together from the pre-clinical models, the selectin and the aggregate data. We had some guidance from the clinical pilot study and we’re now were able to fill in that bridging gap to say, yeah with some chronic administration we do see that similar type of profile that was seen pre-clinically and suggested clinically.

David Beidler:
So this really gave us guidance as we now take on a larger study where we are going to look at crisis endpoints, which is much larger numbers of patients and more involved. We at least have some confidence that our dosing will at least be testing that mechanism. And that’s really what we’re trying to achieve with some of the early translational strategies. So that kind of ends chapter two hopefully that there are any questions, let us know. But again, this is now setting up for taking this into the clinic and about clinical registration endpoints. Getting back that iceberg, remember we still haven’t solved the iceberg problem. We still only have, hospitalization and opioid use as a criteria. And so again, this was the definition. So the sickle cell pain crisis has no medical determined cause other than a vaso-occlusive event.

David Beidler:
Fortunately, we learn from interviews with our patients. They know when they’re having a crisis and it’s separate from their average daily pain. So they feel confident they can say that. But here, this is about actual medical utilization and actually the duration here, depending on the clinical study that you looked at here, some required at least four hours, some were two, some were regardless, but again this was all about medical facility visit that was required to kind of ring the bell, but yes you are having a crisis. You also then had some criteria around perennial narcotic agents [inaudible 00:19:09] agents while you’re in the clinic, so again this is all above the iceberg very much. Again, it’s about a care provider or caregiver mediated caregiver centric environment, which is, it has some advantages. It’s a controlled setting. You’re in the hospital, you hope it’s controlled and that it’s being reported to other trials it’s familiar to the FDA.

David Beidler:
Again, it’s well by precedent and so this is what previous HU and other studies have used and this is what, and some of the cell G1 and other things. They’d still were used. Some of the medical utilization. The disadvantages on average and even in the cell G1 they had a placebo, had roughly three VOCs a year at most generally one to three VOCs a year. So it’s infrequent. A lot of it could be dependent on the patient’s ability or willingness to go to the hospital setting. There is a stigma around sickle cell that we don’t want to touch on too much but they feel hesitant to go to the hospital cause they’re treated as drug abusers cause they’re coming in demanding IV narcotics. So there’s that whole stigma, are you really getting the true patient perspective and it’s really reflective of a meaningful clinical benefit.

David Beidler:
And again, that’s what we’re getting into. You really have to, if you’re going to use medical utilization, how do you normalize access to medical facilities across all your investigators sites? That could be a challenge. Sure. You have a question.

Speaker 4:
Did you consider transfusion dependance?

David Beidler:
So for this type of control study, we would actually probably exclude the transfusion dependent phenotype because that changes the whole blood chemistry for us. But that’s part of the equation is all their other standards of care that they could confound your study here. So typically we would have that as an exclusion criteria. Great. Okay. So we did have some background information. Wally Smith at Virginia had done a study in the early 2000s where again, the background was that it was, healthcare was a proxy for pain and the underlying crisis and it really doesn’t reflect the amounts, the true number of VOC that were being reported.

David Beidler:
So his group did a very nice study, again, following patients longitudinally and looked at the pain crisis. So it was basically 232 patients and they did a paper diary, a daily paper. This is earliest before all this, I mean smart phones were out, but they weren’t where they are today. So we still had the old fashioned stuff. So 31,000 days were analyzed by a paper diary. They did see 56% of total patient days were filled with pain. When they patients reported a crisis, roughly 13% of the time that was managed at home or has only about three and a half percent of the time, it actually was being caused him to go to the hospital and seek medical utilization. So again, this bears out what we’ve been saying is that three times as many, you’re three times more likely to stay at home and manage your VOC.

David Beidler:
So we’re missing a large portion of an unmet need. These people are dealing with a crisis, but an interventional study that would not be captured. So this was a problem to us and we’ll talk about that in more. So obviously more endpoints, quicker studies, smaller studies as well. So there’s not just for the patient experience, but for the developer’s experience. It helps us try to be more efficient and more effective in intervening and capturing that we’re in fact creating clinical benefit. So we had it, we have a whole outcomes group and we’ve been interviewing patients for a number of years now. And this really is about defining the patient experience. So, they really want us to reduce the recurring VOC. I think that’s pretty obvious, but it’s good to hear them talk about this. They’d love to be able to self administer.

David Beidler:
They really have an issue with the fatigue components. They would really love to address the daily fatigue. Again, avoiding the stigma of medical seeking, they’d like to be able to treat at home again.Again all this preventing the longterm complications, improving their quality of life that can be socially active, go to work everyday, go to school every day. And then really it’s about the complication of the disease and actually being able to treat the root cause of the disease itself and not just the symptoms. So again we’re thinking of this disease now as a spectrum and how do we cover activities at home, in the clinic, at the physician’s office, at the ER, or actually in inpatient. So we’re trying to capture the new endpoint that we’re creating, a new concept that’s called the VOC day. And so we’ve kind of mapped out the conventional pain crisis on the left here and then our VOC day on the right.

David Beidler:
So, both share the common thing. It’s a sickle cell related pain crisis and if no, nothing medical admitted to get determined causes other than evasive occlusive events. So that’s common. But we now have, instead of relying on medical utilization or perennial narcotics, we have daily reporting through e-diaries. So they self report that they’re having a vaso-occlusive pain episode. They also then have some secondary end points around daily pain, daily fatigue, their daily functionality, what medications are they taking and did they do they infact seek medical care. But this new VOC day really doesn’t require that they go to the hospital to be registered as a VOC day at or receive the IV narcotics and duration really is, it’s agnostic to duration at this point. This just kind of shows the idea of that the VOC day is still capturing the conventional date day pain crisis that has been used in the previous studies.

David Beidler:
But in addition we are trying to bring in these more patient-centric end points and create a tool that can capture these in a future interventional study. So as far as the overall patient experience, here’s kind of the style we use for the, here’s the concept of the patient experience. The various domains that we have are the actual VOC painful crisis, pain, fatigue and functional impairment. The end points that we’re using mostly through e-diary, we asked them, there’s a flatness we’ll see in a later screenshot that Debbie’s going to show you, are you having a pain crisis? And that do you feel that all out on a screen on your e-diary. And, but then every day regardless of whether they are having a crisis or not, we’re asking them about their pain, how fatigued they are and what is their overall functioning.

David Beidler:
Each of these components, these were previously validated patient reported outcome endpoint. So this is kind of a Frankenstein model. It’s called a brief pain inventory, there’s a fatigue scale, it’s been validated. And this is from a group called return to normal activity questionnaire that asks about can you do your self care? Are you socially active as normal? Can you do your normal physical activities and can you go to workers school? So this is again trying to get multiple concepts of the patient experience because they told us it’s not just about the crisis. They’d like to reduce the fatigue, they’d like to be able to be normally functioning every day. So how do we do it? [inaudible 00:25:45] question?

Speaker 5:
Are you using any electronic activity monitoring? [inaudible 00:00:25:52].

David Beidler:
Yes. So what we actually were going to show them in the ellipsis because in addition to background we’re going to, we’ve been doing some extra graphy [inaudible 00:25:56] yeah. How this works then is so you, they say they’re having a crisis. We also ask them what are you doing about it? Where are you or how are you then managing this crisis? So one of the questions that Debbie’s going to talk about in a few minutes is, did you go to the hospital? Did you go to an ER or a doctor’s visit? Did you just email our call the doctor? Or did you really strictly manage this at home? So each day, even when they’re in a crisis, we ask them, are you still in the crisis and how are you managing it? So that we can create, when you now string together days, you create a VOC event.

David Beidler:
And so that can last anywhere from one just to several days. And so it’s kind of, what we’ve tried to show is this is a hypothetical schematic of, if it’s, for example, this patient on the very top started going direct healthcare utilization but then went home, maybe they go right to the hospital. So we’re learning more about this or some of the early studies that we did with ellipsis. We then said, okay, we’re classifying event by the highest degree of medical utilization they had. So even if they just spend a day or so in the hospital, we’d still call that a hospital VOC. But you can see now that by monitoring this daily, you get some of that information you were asking about is what is the daily progression of it? It’s not just binary, yes or no in the hospital, out of the hospital.

David Beidler:
So, this hopefully while this, like I said, we still haven’t worked this into a primary end point. You’ve got to kind of walk before you run with regulatory work. But at least when we handle, we can do some analysis like this and see maybe we don’t, potentially you don’t reduce the number of days, but you reduce the number of hospitalizations or does do the pain scores come down. So there’s a lot of permutations you could look at from this type of analysis. So what we have now is, so we have some initial biomarker data with PDE9 and we showed potentially a less adhesive, less inflamed state with E-selectin and the aggregates. We now had some tools in place, at least when you went by talking to patients and building some tools around the VOC day. So we really then designed a study called the ellipsis study that really tied all of that together.

David Beidler:
Getting ready to take PDE9 and other drugs in our portfolio into interventional studies to look at a vaso-occlusive crisis impact. So let me go back to our original iceberg. So now what we have is daily ePROs so that we’re monitoring pain, fatigue, functionality, drug use, the daily activity. We are now introducing actigraphy monitoring if we want to see how they’re moving on good days versus bad days and the transition from those days. And we’re also then doing a blood collections at longitudinal blood collections, which I’m going to talk about in a minute, that span the VOC. We’re trying to get blood before, during and after these crisis to see what is really going on in the circulation that is either causing or as a result of the VOCs.

David Beidler:
So of course, we continue the ePRO monitoring. We have the actigraphy, we do the blood collections regardless of where they are. We did this at one site in Detroit, so we were able to follow these patients through their various levels of medical use. So it was a six months study obviously with patients plus or minus HU. So we gave them this, again, this is where Sanguine comes in again, so they were basically our operations group on the ground in Detroit. They helped, Michael Callahan would recruit them in, we have a lot of us giving them the actigraphy devices as well as the ePRO devices and they would monitor daily pain, fatigue and function and medication used. And then what we had was there was a special cassette which Debbie I think will go on for a VOC crisis.

David Beidler:
It would alert Sanguine and the clinicians that they were starting a crisis. Within 24 hours, we had a phlebotomist out to their house. The next day we had a phlebotomist out to the house. Three days later, we had a … whenever it ended, we had a phlebotomist out to their house if they needed to be, if they were at home, if they were in the hospital, we got it in the hospital. So this is what we’re getting at here is that we were truly trying to get those longitudinal collections that span the VOC event. And so that we had both baselines data from when people were not experiencing a VOC. And then we had these VOC collections so that we could compare VOC versus none. And so we also had the continuous monitoring then with the Philips Actiwatch. Oops, I’m sorry. And then of course we had, if they weren’t experiencing any VOC, we just had collections every three weeks so that we could get a good baseline from them.

David Beidler:
So this just kind of points out what VOC daily actigraphy is going continuously. The black bars are the nod of the stable blood collections night. When this hypothetical one had three or four VOC. We had the day one, day two and then two days after the resolution of the VOC. This is just showing kind of what, our first question, so every given biomarker, every given ePRO is going to have this kind of background noise. And what we wanted to see by looking longitudinally is was there really a threshold where when then we reported a VOC, did you actually see something that was statistically significant and also then clinically meaningful because it was during a VOC. So this is where we were as far as, this is what we’re hoping to see is that was it just going to be noise?

David Beidler:
If you ask a sickle cell patient daily, how is your pain? Our concern was, it just might be chaotic. So hopefully we can convince you that we saw, we got a little bit of information about this that will really help us move on. So I’m going to stop my section here with just saying this was quite an operation because we did this in, Michael Callahan’s group in Detroit, recruited in the 30s five or six so subjects. But, Sanguine did a lot of the operations.

David Beidler:
We had various vendors, we used Boston Children’s Group for some of our aggregate assays and that’s Alan Michelson and [Larry Freelander 00:00:31:27] really, we’ve worked with them for a number of years. Data management and all this putting together was a lot of work. And it took probably about at least two years or so to complete everything. But if you’re going to put one of these together, it’s a lot, isn’t it? It’s not an easy thing to do, but it was really enjoyable, hopefully what Debbie’s going to show you now was is providing us some really valuable information.

Debra Pittman:
So this is just an example of what the ePRO look like and what one of the modules look like. As David mentioned, it had two modules. There was a daily reporting, a daily diary, which they entered their day. If they had a VOC day, a severity scale was adapted from the BPI. We had functional activity, fatigue as well as medication diary that they entered every day. There was also a button on here that they could push that said if they had a pain crisis and what this did, we really didn’t want to interfere with what their choice of treatment was. It did alert the physician, but it also alerted the mobile phlebotomist that they were having a pain crisis so that they could get blood drawn within 24 hours as well as 48 hours after the onset of VOC. So these are the two modules of how it looked on the first page of the ePRO and then this is just an example of what the patient chose for their treatment.

Debra Pittman:
In this case, they were asked, did you experience a pain crisis in the last 24 hours? If so, how long did that last? Did that last four hours or more? And then they could select what they decided for their treatment. Did they decide to treat themselves at home? They a call or contact a doctor or a nurse via text or email or phone call? As well as they went to the doctor, they went to the emergency room or they stayed in the hospital overnight, so this was all captured over the six months study for each patient. This is the demographics of the study. It was roughly a 50-50 mix of patients on and off hydroxyurea. We started with 37 patient volunteers who did not complete the studies early on for various reasons. We also included the opportunity to include some non sickle cell patients.

Debra Pittman:
We included sickle cell traits from this study and many of these were from the same patients, family members, so we were able to include that and we also had a few normals where we could get baseline data and use it for our comparison. The trait patients and the normals did not have, they only had blood draws for three months and they did not do the ePRO. This is an example of what the schematics look like in terms of the longitudinal pain over at the time and two of the patients. And you can see that the pain scale goes from zero to 10 and in some cases the patients would rate their pain is 10 we actually had some patients that have very high pain and then there were other patients who would consider their daily pain low. So there is a lot of variability.

Debra Pittman:
But the advantage of a longitudinal study is we can see what that baseline pain is and then normalize it through this study. And then if you can see that the VOC is four hours, how long did the VOC last? Was it less than four hours is in blue. And then you can also see your VOC event and you can see from the red that the patients that are having a VOC event, or VOC day are experiencing the worst pain by their own pain scores. Question? Yes.

Speaker 6:
From those two illustrative and now it seems like compliance is pretty good for them. [inaudible 00:34:56]

Debra Pittman:
Yeah, I think we had-

David Beidler:
We did some cherry picking here of course [inaudible 00:35:01]

Debra Pittman:
… but we also had was 63% compliance.

David Beidler:
60 to 70% compliance for daily use. Yeah.

Debra Pittman:
Yeah.

David Beidler:
There’s a whole talk we could have about imputing data [inaudible 00:35:10] and how to fill it in.

Debra Pittman:
Yeah. There was, as a study went on, you could see that compliance got less but did have some learnings from this study. I will note that the ePRO device was a separate device versus not on their phone, so when a patient would go to the hospital or they would not always remember to take the ePRO device because they’ll grab their cell phone. So there are some learnings from, that you could take forward from this. It’s probably best if you can get it on their phone because obviously have that. This is just an example of what the VOC events and durations were from this study. The duration was an average of 3.25 days across the patients as well as the annualized VOC events was between seven to eight events per year. You can see that some patients, there’s one patient that actually had a fair number of events on this study.

Debra Pittman:
What was interesting, and this is really consistent with the PISCES study, is what you asked what the patients chose to do treatment. So if you look at where they were for a VOC day location, almost 72% of the patients chose self treatment. They did not seek any medical utilization. That means if you’re using the standard VOC definition now that requires medical utilization and an input from a medical survivor, I mean a medical supplier, you’re actually missing a majority of the patients and they’re not seeking medical care. If you look at the VOC event location, it’s roughly 65%. So really a lot of patients are, there’s a clear unmet medical need and they’re suffering from their disease trying to manage it through their own home pain management, hydration, massage or whatever they choose.

Speaker 7:
So did all the patients have healthcare or was that [inaudible 00:37:02] ?

Debra Pittman:
These were actually the patients did have health care if they, I mean these were all patients that were under care. If they chose, we didn’t want to interfere with what they, how they chose to treat their VOC.

Speaker 8:
I know it’s a small sample, but did you see any differences in gender in self treatment?Or in self treatment versus going to the hospital location close to home?

Debra Pittman:
I don’t think we saw any difference between gender. These patients were actually all fairly close to the Detroit medical center, so I don’t think that actually played a factor in whether they chose treatment. Is there another question?

Speaker 9:
I was wondering if you could comment on whether, do you have any hypotheses as to why people chose to self treat [inaudible 00:38:01] instead of going to a healthcenter?

Debra Pittman:
I mean I think part of it is because they require, I mean if you go to the emergency room you often sit there for a long time and then for some of the sickle cell patients they’re actually often viewed as drug users because they are seeking opioids and some of it may be that all of the medical centers really don’t understand what the needs are for patients with sickle cell disease, that they really do need to have that pain treatment to reduce their crisis. And I think it’s over time that they just, it’s probably easier for them until it reaches a certain point to treat at home. I mean that’s sort of our hypothesis.

David Beidler:
Well Debbie but there were some that went right to the hospital. So there is no one trend. And again, this group, Michael Callahan somewhat, he recruited them himself. So these were actively seeking medical regular checkups. So these weren’t even on the XQ kind of the edges of treatment. Yeah. It’s hard to say for whatever reason this group, maybe it’s cultural against anxiety, whatever, but maybe the self treatment just is more effective.

Debra Pittman:
But as you’re thinking about clinical trials and bringing new agents forward, it’s really important to think about where the patient is seeking treatment and can you really start to help these patients that really choose not to go to the hospital or the emergency room to treat their VOC. This is looking at some of the secondary points where we looked at increased pain, decreased functioning and fatigue. And here we’re looking at a normalized data from each patient to their baseline. And you can see and we were looking at non-VOC days which is in the blue or the solid and then the strike bars are all VOC days and we tried to break it down to those patients that were at home or chose to go to the hospital and I think in all cases you can see that in terms of increased pain, decreased functioning and fatigue.

Debra Pittman:
When a patient is reporting that they’re experiencing VOC day, they are seeing an increased pain, decreased functioning and fatigue and we really didn’t see any significant differences between whether chose to treat at home or whether they chose to go to the hospital depending on this analysis. We then also, as we mentioned, we had actigraphy, we use the Phillips Actiwatch and what this allowed us to do was actually the patient wore the watch all the time and they could monitor their total daytime awake activity as well as their sleep activity. In the top panels you can see what type of data that’s collected from the individual points to kind of get into more normalized data. So you can look at this data in a lot of different ways. I would say that it’s a large amount of data to analyze, but you can look at activity counts per day time activity. How sedentary they were.

Debra Pittman:
Were there minutes of interrupted sleep? What the sleep fragment index was as well as total sleep time. I would comment on the watch that the patients were also pretty compliant, but sometimes a lot of the patients were younger and sometimes on the weekends you can see that the watch was somewhat, I would say the not attractive and they would take it off on the weekends. So that’s something else. As we move forward, we need to think about the type of actigraphy watch that we give to patients that it’s not something that they look at and say I just can’t wear this on the weekend or whatever.

Debra Pittman:
So this is where we look at changes in actigraphy and we did see changes in actigraphy between patients that were reporting a self treated VOC versus a non-VOC. You can see the black line versus the blue line. And not surprisingly, if you look at the healthcare utilization, you can also see that there’s changes in terms of activity based on this analysis and if you start to look at the percent change from baseline, you can see that there’s a change in the total rest activity as well as average activity daily as well as you can see that there’s differences in sleep when patients are reporting the self reported VOC.

Debra Pittman:
We also had a large number of biomarker panels that we tried to incorporate into this study, so as David mentioned we had regular blood draws from the whole time in the study, TF baselines, but then we also had 24 hours, 48 hours after the onset of VOC and 72 hours after resolution. We included a lot of samples that required either processing on site such as the leukocyte platelet aggregates and microparticles which required special processing and analysis within a certain timeframe as well as we had a microfluidic-based assay which is a physiological flow based essay and that was done by functional fluidics to look immediately or within a certain period after the blood draw the same day to see if there’s any addition in adhesion to substrates such as VCAM-1 and P-selectin. We had protein panels and included serum and plasma, cell adhesion markers, inflammatory markers, growth factors as well as coagulation markers and we’ve also banked samples.

Debra Pittman:
We’ve banked all the plasma serum samples as well as we’ve banked samples for RNA analysis in the future as well as DNA samples if we want to do profiling. We haven’t touched these yet, but these are all samples that we’d hope to now analyze and that we would make available to others for analysis as well.

Speaker 10:
Can I ask who are the visiting nurses? So we know that there’s a lack of trust in the sickle cell community. How is that received by the patients?

Debra Pittman:
And that was actually received very well. The patients, like the mobile phlebotomist coming to their house and drawing blood and I think they developed a level of comfort having the same person come to their home to draw the blood as opposed to having to go into the clinic to have their blood drawn. Do you want to?

David Beidler:
Yeah. I think we heard back from our MDs. They felt they were really engaged in wanting to learn about the study and were very at ease with those people coming to visit.

Debra Pittman:
We had one patient that asked the phlebotomist to come in the hospital to take their blood cause they got. [inaudible 00:44:24] I wasn’t, we couldn’t do that, but it was. So this is looking at the cellular aggregates and what we noticed, this is consistent with what we’ve seen in the animal models as well as what we’ve seen in the PDE9 study, that when a patient is reporting a VOC, we actually see an increase in cellular aggregates in terms of the monocyte platelet aggregates as well as the neutrophil platelet aggregates when you look at baseline comparison. In addition, we did see some changes in adhesion to P-selectin under flow conditions shown in the bottom. So both of these we do see changes in aggregates when a patient is self-reporting a VOC. So in addition to the other biomarkers that we said, we also took clinical laboratories and this is looking at the standard hemoglobin, hemoglobin F, nucleated red blood cells, eosinophiles as well as C-reactive protein. Interestingly, eosinophiles actually decreased during the VOC, but one thing that we also consistently saw a difference in was the C-reactive protein. And this is actually considered consistent with a study that was done a number of years ago by Ballas who actually did show some changes in C-reactive protein in sickle cell patients and these would be easily monitored in a clinical setting because these are part of the clinical lab monitoring.

Speaker 11:
Excuse me, were those changes consistent if it was a hospital based VOC versus a self-reported VOC?

Debra Pittman:
We haven’t seen differences between the hospital associated and a self reported VOC with the number of patients that we’ve looked at.

David Beidler:
Yet again, very small numbers. Our strategies tried to phase them out. There’s just not enough data. [inaudible 00:46:04]

Debra Pittman:
And then we looked at some cytokine and hemostatic markers and it’s not unexpected, but we did see changes in IL6 from baseline as well as interleukin 10, TNF-alpha as well as some coagulation markers such as TAT and a slight decrease in tissue factor. So based on the study, this electronic patient reported outcomes we believe provided a more comprehensive accounting of the patient experience in sickle cell disease. The majority of the patients had at least one to three VOC events during the six month period. Over 63% of these reported events patients did not seek medical utilization. We did see significant differences in pain, functioning and fatigue levels reported during the VOC compared to non-VOC and the VOC reported days had higher pain scores decreases in measures of functionality as well as an increase in fatigue. We also saw differences in actigraphy, with reduction in daytime activity as well as disturbance in sleep at night during the VOC.

Debra Pittman:
In addition with the biomarker analysis, we did see changes during VOC days that were observed in a substantial number of clinical laboratory and biomarker measures. This included cytokines, hemostatic markers, C-reactive protein as well as nucleated red blood cells and we saw changes in the monocyte neutrophil platelet aggregates. This innovative at home longitudinal study demonstrated the feasibility of employee on using mobile phlebotomy, electronically patient reported outcomes as well as actigraphy and the seasonability of monitoring pain outside the hospital when a patient was experiencing a VOC. So the way we’re looking at sickle cell disease is that you have on the the improved sickle cell disease where you have minimal endothelial activation, minimal numbers of cellular aggregates and a decrease in that endothelial cell activation. As they go to from sickle cell steady state to a VOC, you see the homolysis of the red blood cell.

Debra Pittman:
You see inflammation triggers as well as leukocyte adhesion. You also see an activation of the endothelial cell increase in coagulation and thrombotic markers and the cellular aggregates increase that vaso-occlusion. What we tried to do with our natural history study in sickle cell patients as well as our pre-clinical models is to take different markers of this such as a clinical labs, plasma markers, physiological flow, coagulation markers, and then using ePRO and actigraphy to really try to get a better understanding of what might be going from the improved sickle cell state to when patients are experiencing a VOC. And David, turn it up.

David Beidler:
For our final slide, so what does all this mean for us? And of course we have to pay the bills, we have to create innovative ways of early clinic, a means to get through early clinical development. And remember back to that slide providing a confidence in the value early. So the thought is new. The current state is reducing hospitalization rates, which happened maybe one to three times a year or looking at the duration of hospitalization, acute chest injury or other organ failures with no really validated biomarkers took to teach you anything along the way. You’re just waiting, you’re doing either prophylactic and then you’re hitting the hospital. What we’re really trying to do in the future state again, is patient-centered endpoint. That we have this composite endpoint of the VOC day which pounce and accounts for their disease state and their decisions regardless of what degree of medical utilization is being used as well as then if we do collections of biomarkers that potentially, and also changes in actigraphy that correlate with VOC end points.

David Beidler:
The goal is we accumulated updated, could we actually be somewhat predictive of future VOC endpoints. The actigraphy data potentially with large jump data, could you actually create a profile that says you’re at a higher probability of a future VOC state. Those are things we’re always striving for. So the point is by getting to this future state, we can then really look at the chronic therapy because we’re now capturing a larger number of events, seven to eight a year versus one to three. So again, that’s really important about using fewer subjects and shorter treatment periods. Because if your alpha needs to be 45% so you do the math around cohort size. So you maybe, what needed to be a 180 may only be to be 90 or 60 or so. So we have, that’s one thing that’s really driving this and the biomarkers would better understand the nature of the VOC as escalation over time and potentially for new targets for intervention and and then again the predictive nature of either the ePRO or the biomarker changes and again registration timeline so that could we work with the registry to say, this is now a co-primary, we’re going to get seven or eight of these a year.

David Beidler:
We feel these are clinically meaningful, they’re going to provide benefit and so we can get the early pivotal studies faster so that it’s not such a daunting task to try to innovate and procreate new medicines for this patient population. In addition, by having a lot of those secondary things around fatigue, daily function, is there room here for differentiation between therapies? It isn’t simply about the crisis itself, but it may be there are others are about more of the pain or end organ damage. Could either a biomarker or actigraphy or ePRO profile help provide that level of differentiation. So our group is actively working with ,our outcomes group with working with developing a briefing document to go to the regulatory agencies, to bring this forward as a potential for co-primary. So as we progress hopefully you’ll be seeing this in future studies and for any group out there looking to look at end points.

David Beidler:
This is not a proprietary tool. We’d love to have other groups be using this type of a VOC day in point. And it’s about building that content validity, more data is better. So it’s always a challenge to get that in place. But hopefully this has been informative and a little bit about drug development, a little bit about going from point A to point B and getting ready for early clinical development. But thank you for attending. And we welcome any questions. Before I finish though, I always forget to do this and this is important. A lot of people, both within Pfizer, Rare Disease, all of our collaborators on the studies, Boston Children’s, the folks at Sanguine were very helpful as well. So again, this was a lot of good work by a lot of people. And so thank you everyone for attending.