Chelsea Therapeutics International CEO Presents Study 306B Conference Call (Transcript)

| About: Chelsea Therapeutics (CHTP)

Chelsea Therapeutics International Ltd. (NASDAQ:CHTP)

Study 306B Conference Call

December 4, 2012 5:00 pm ET


David Pitts - Investor Relations, Argot Partners

Joseph Oliveto - Interim President, Chief Executive Officer

William Schwieterman - Chief Medical Officer, Vice President

Dr. Robert Hauser - Professor of Neurology, Molecular Pharmacology and Physiology, University of South Florida

Arthur Hewitt - Chief Scientific Officer, Vice President


Robyn Karnauskas - Deutsche Bank

Alan Carr - Needham and Company

Scott Henry - Roth Capital

Juan Sanchez - Ladenburg Thalmann


Good day, ladies and gentlemen, and welcome to the Chelsea Therapeutics Study 306B Conference Call. At this time, all participants are in listen-only mode. Later, we will conduct a question-and-answer session and instructions will follow at that time. As a reminder, this conference call is being recorded.

I would now like to introduce the host for today's conference David Pitts. Please go ahead.

David Pitts

Thank you and thanks to everyone on the call for joining us today to discuss data from Chelsea Therapeutics Study 306B, for which a press release was issued this afternoon. Please note that there is a slide presentation accompanying today's call, which is available at

On the call today are, Joe Oliveto, Interim Chief Executive Officer, Dr. Bill Schwieterman, Chief Medical Officer, Dr. Art Hewitt, Chief Scientific Officer, and Nick Riehle, Chief Financial Officer. In addition, we are joined by Dr. Robert Hauser, Professor of Neurology, Molecular Pharmacology and Physiology at the University of South Florida. Dr. Hauser is also principle investigator for this trial.

Before we get started, I remind you that we will be making forward-looking assertions during today's call that represent the company's intentions, expectations or beliefs concerning future events which constitute forward-looking statements for the purposes of the Safe Harbor provisions under the Private Securities Litigation Reform Act of 1995.

Such statements are subject to factors, risks and uncertainties such as those detailed in today's press release announcing this call and in our filings with the SEC, which may cause actual results to differ materially from those results expressed or implied by such statements. In addition any forward looking statements represent our views only as of the date of this recording and should not be relied upon as representing our views as of any subsequent date. We specifically disclaim any obligation to update any such statements.

I will now turn the call over to Joe.

Joseph Oliveto

Thank you, David. Good evening, everyone, and thank you for your interest in hearing this presentation of the top line results from Chelsea's Study 306B, a Phase III trial, designed to assess the safety and efficacy of Northera in treating patients experiencing symptomatic neurogenic orthostatic hypotension in Parkinson's disease.

Symptomatic NOH is an orphan disease affecting less than 200,000 patients in the United States. Study 306B was conducted exclusively in the United States across 53 centers, and we are very grateful to have Dr. Bob Hauser, principal investigator of the study for joining us on this call. Thank you, Dr. Hauser.

I will provide a brief summary of the high-level and preliminary results and then will be transitioning the presentation to Bill Schwieterman, Chelsea's Chief Medical Officer, who will provide some detail and context behind what was provided in the press release. Following Bill's presentation and a short summary, we will conclude with Dr. Hauser providing his thoughts on the study results before opening up for questions.

Next slide provides the top line results from Study 306B. The primary endpoint of the study was predefined as the mean improvement in symptoms of dizziness and lightheadedness associated with patients' orthostatic hypotension relative to placebo following one week of stable dose of therapy.

The assessment is performed as the first item or question of the orthostatic hypotension questionnaire which is a clinically validated patient reported outcome scale ranging from 0 to 10 units with zero being asymptomatic and 10 being most severe.

Dizziness, lightheadedness is the cardinal symptom of neurogenic OH and is the symptom assessment of greatest interest to the FDA as was presented at the Northera Advisory Committee Meeting earlier this year.

For the primary endpoint Northera treatment resulted in 1.0 unit improvement difference when compared to placebo at Week 1. The 1.0 unit improvement was statistically significant with a P value of 0.018. Additionally, the one unit improvement is consistent with point estimates for the Item 1 dizziness, lightheadedness improvement scores observed in previous studies.

Standing systolic blood pressure is an important secondary measure of drug activity. In this study, Northera increased standing systolic blood pressure by 5.6 millimeters of mercury as compared to placebo at that same one week time point and this result was also statistically significant with a P value of 0.032.

Patient falls is a very important clinical event in this population. The first formal assessment of Northera's impact on the rate of patient falls was initially assessed and reported for Study 306A. In this study, 306B, there were many patient reported falls. The predefined metric for evaluation was the rate of patient falls calculated as a ratio of falls per patient per week.

In Study 306B, Northera-treated patients reported an 80% reduced rate of falls compared to placebo. The result was not statistically significant. Bill will present some sensitivity analysis that were performed around these results during his portion of the presentation. Additionally, 34% fewer Northera-treated patients reported injuries that were prospectively defined as related to falls compared to placebo-treated patients, although this result did not reach statistical significance.

Regarding the safety and tolerability, supine hypertension is always an assessment of interest given the labile blood pressures in this population. Study 306B, the incidence of supine hypertension was low with 7% of Northera-treated patients and 5% of placebo-treated patients having measurements of supine systolic blood pressure in excess of 180 millimeters of mercury at some point during the study.

With regard to adverse events, the most common AEs reported in Northera patients were headache, dizziness events, nausea, hypertension and fatigue, which were predominately mild-to-moderate in severity and generally consistent with previous studies.

Before I turn the presentation over to Bill, I do want to remind everyone that while Study 306B represents a very large study relative to other studies in this orphan drug population it is not considered by the FDA as a study that can be used as the pivotal confirmatory trial to support a Northera registration. Chelsea is abiding by that guidance and present these top line data as important and informative data that help us better characterize Northera's safety and efficacy profile.

With that, I will turn it over to Bill.

Bill Schwieterman

Thank you, Joe. Let me first explain Study 306, and how it was divided into two parts, 306A and 306B. As Joe just mentioned, Study 306 is an entirely U.S.-based study from Northera for the treatment NOH from Parkinson's disease.

After we performed an interim analysis on the first 51 patients, we modified the statistical plan while being fully blinded to data from ongoing patients and leaving the protocol intact. After the interim analysis, we continued the study. The first group of patients, those studied in the interim analysis, we called 306A.

The second group of patients, those in the continued blinded portion of the study, were called 306B. Overall counting both, parts A and B, there were 225 patients enrolled in the meta-analysis with a 197 patients included in the modified intent-to-treat.

As I mentioned, there were 51 patients enrolled in study 306A, while 174 patients were enrolled in 306B. With 147 in the modified intent-to-treat analysis, which was defined as those patients making it through the randomized dose optimization period and having a week one efficacy assessment.

On slide six is a schematic of Study 306. Study 306 is a randomized placebo-controlled study that includes a blinded dose optimization phase, which lasts for up to two weeks followed by an eight-week maintenance phase, where patients receive a stable dose of Northera.

Patients were treated in the dose optimization phase with incremental of 100 milligram dosages until it became asymptomatic or nearly asymptomatic or experience an adverse event or safety issue. In other words, this was a post-titration to maximize the dose of Northera tolerated for each patient.

As you can see on the slide as well as the follow-up visits during the maintenance phase that patients were evaluated for efficacy assessments following weeks one, two, four and eight weeks of stable dosing.

Slide seven shows the results of the primary endpoint, which was pre-specified as a change in the dizziness, lightheadedness score, a week one of stable dosing compared to baseline. Northera is shown in blue on this slide, placebo in grey. The y-axis shows the improvement in dizziness, lightheadedness as measured by the OHSA 1 score.

Recall from our previous studies that the OHSA 1 score is completed using a questionnaire given the patients when they were asked to rate their dizziness on a scale of zero to 10 as they proceeded over the past week. The data on this slide shows that there was a one point greater improvement on dizziness at week one compared to placebo, which was statistically significant at P=0.018. You can also see on this slide that there was a relatively large placebo effect, something we observed in all of our studies.

Slide eight shows the dizziness, lightheadedness data at week one according to a responder analysis. Here, we evaluate the number of patients who experienced 1, 2, 3 or 4 unit improvement in the OHSA 1 scale compared to baseline as well as those who experienced a 25% or 50% improvement in the OHSA scale from baseline.

You can see that regardless of the cutoff we use to define a response whether it's a one through four-unit improvement or 25% or 50% improvement, patients receiving Northera shown in blue experienced more improvement in dizziness, lightheadedness than did placebo patients shown in grey. Except for the analysis of one unit response, all of these were statistically significant at P less than 0.05.

Slide nine shows the dizziness, lightheadedness scores over time. Again Northera is shown in blue and placebo in grey. On the x-axis, are the data per time point at weeks one, two, four and eight with the amount of improvement in dizziness shown on the y-axis.

At week one, the primary endpoint which was shown earlier, there was a statistically significant improvement in patients receiving Northera compared to placebo. Data at weeks two, four and eight do not show statistically significant differences. However, they do show trends favoring Northera over placebo that continued out through week eight. Again, we see the relatively high placebo response as maintained throughout the study as well as the relatively large amount of variability reflected in the wide arrow bars for each value.

Slide 10 shows the blood pressure values for 306B, change in systolic blood pressure upon standing during the orthostatic standing test as compared to baseline was an important secondary outcome for study 306B, although we recognize the limitations of collecting blood pressure at a single time point given the high variability that occurs in these patients. Northera ratings are shown again in blue, placebo in grey with change in blood pressure shown on the y-axis. What we see here is that the change in systolic blood pressure upon standing is statistically significant favoring Northera at week one, but not at later time points although again there are trends favoring Northera at both weeks two and eight.

As we anticipated, there was a great deal of variability both, at any given time point as reflected in the wide error bars and the variability in the point estimates over time. In general, these data overall support the blood pressure benefits of Northera, both acutely and over the eight week time course.

Slide 11 shows the falls data. The rate of falls as measured by the number of falls per patient per week was an important secondary endpoint for Study 306B. We collected falls by using an electronic diary, where patients was asked each evening to record the number of falls as well as the conditions that existed when they fell. These results are remarkable and show that there was an 80% reduction in the rate of patient falls favoring Northera shown in blue, compared to placebo shown in grey. This value was not statistically significant however despite the large magnitude of the effect. This is because fall benefits are difficult to capture statistically given the wide variability in the number of falls between patients.

For example almost 40% of patients in Study 306B regardless of treatment arm did not fall at all. Although remaining patients who fell, many fell only a few times during the course of the study. A small minority of patients fell more often dozens of times in some cases and a small number of patients fell many times over a 100 times or more.

While we anticipated this variability in 306A and modeled our statistical plan accordingly, we were never left unable, with this study's size, to overcome this variability and show a statistically significant benefit on reduction in fall. Still these data replicate what we observed in Study 306A, where we showed that Northera patients fell approximately over 60% fewer times than placebo as well as other data from our earlier randomized controlled trials studies 301 and 302, which also showed evidence of falls benefit for Northera.

To further evaluate the robustness of the suggested benefit on falls and to test the extent to which the falls benefit we observed for Northera was driven by those few fallers who fell many times during the study, we performed sensitivity analyses, same as we did when we observed the fall effect favoring Northera in Study 306A. These data are shown on slide 12. The sensitivity analyses selectively removed the data on the top fallers from each treatment arm to assess whether the treatment practice still observed.

Turning the slide, we observed that whether remove the data from the top two fallers from each treatment arm, the top five fallers or the top 10 fallers, there still is a difference in the rate of falls favoring Northera compared to placebo. We conclude from these sensitivity analyses that the suggested falls benefit observed for Northera is not nearly driven by few patient outliers who fall very often, but rather is evident across the broad subgroup of patients supporting a greater likelihood that it is a real and valid observation despite the absence of statistical significance.

Even more important than the data from the sensitivity analyses, we also observed that not only did patients fall fewer times in receiving Northera compared to placebo, but they also experienced fewer fall related injuries as shown on slide 13. This also validates the suggestion that Northera has a clinically meaningful effect on falls.

Let me walk you through this slide. After data from Study 306A showed the falls dented, we prospectively defined in the blinded safety database, those adverse events we believe to be possibly associated with a fall. These are listed on the left under AE Preferred Term. In addition, we stated prospectively that we would define one of these pre-specified adverse events as related to a fall only if we could show that one of these adverse events were reported at the same time the patient fell, that is on the day of the fall or a maximum of one day later.

As shown on this slide, the 306B data showed that both the number of patients with fall related adverse events and the overall number of fall related adverse events were reduced in patients treated with Northera compared to placebo. 15.9% of patients treated with Northera experienced a fall related adverse event while 25.6% of placebo patients showed an event, a 34% reduction. There were 24 told injuries on Northera versus 35 on placebo. When treated those list of injuries includes contusions, excoriations and lacerations as well as some bone fractures.

Slide 14 shows a summary of the safety data sets for Study 306B. There were an equal number of treatment related adverse events, or serious adverse events between the treatment arms, while there were some who were discontinuations 12% versus 6% on Northera versus placebo. Most common adverse events that occurred more in the Northera treated than in the placebo treated patients in Study 306B are shown on the bottom half of the slide.

The adverse events included many of the same ones identified in Study 306A and in our previous studies. Headaches, dizziness, hypertension and nausea were the most common AEs associated with Northera. In general, these adverse events were mild to moderate in severity. The cardiovascular and neurological safety profile also showed Northera to be well tolerated.

Let me talk about the dizziness adverse event before leaving the slide. It can be confusing to see increased adverse of dizziness in a trial that also showed a reduction in dizziness as the measure of efficacy. It's important to remember that dizziness captured in the OHSA item number one score is different than dizziness that is captured as an adverse event.

The OHSA-1 score is a measure of a patient's dizziness and lightheadedness when standing as measured over a week's time and is ultimately reported on a questionnaire. Dizziness adverse events are spontaneous patient reported outcomes that might occur at any time during the study.

We speculate that the reason for the slight increase in dizziness adverse event on Northera compared to placebo might be because many Northera-treated patients are better able to function once the therapy is initiated and begin engaging in activities that then secondarily result in a dizziness episode. For example, few patients are dizzy when lying down but if after taking Northera they are able to rise from a bed and begin again performing daily activities around the house, they are more likely to experience and then report an episode of dizziness. In any event, the increase in incidents of dizziness episodes are small, only about 5% greater than placebo.

Slide 15 shows the incidence of supine hypertension in Study 306B. Like that has been observed in previous studies, patients treated with Northera have a remarkably low rate of supine hypertension about 200 millimeters of mercury defined as three consecutive measures on the Orthostatic Standing Test, 3% on Northera treated patients and 1% for placebo. As the cutoff gets lower, you can see that the incidence of supine hypertension goes up in each treatment arm but the increase in supine hypertension in the Northera-treated patients is only a few percentage points higher than in placebo treated patients.

I would now like to turn to the meta-analysis data where we combine the results from 306A and 306B. Remember, the patients in both parts of 306 were treated under the same protocol and that only the analytic plan for the two portions of the study differed. First on slide 16, I will show the dizziness, lightheadedness data over time for the entire meta analysis.

These data show that there was a highly statistically significant benefit on the dizziness, lightheadedness score at week one favoring Northera compared to placebo, a 1.2 unit difference between arms was observed with a P value of 0.008. For the remaining time points of weeks 2, 4 and 8 the improvement in dizziness was not statistically significant between treatment arms but did continue to favor Northera at each time point. The P value for these at eight week endpoint, for example, showed a strong trend for improved dizziness favoring Northera with a P value of 0.077.

Slide 17 shows the falls data from 306A and 306B meta analysis as well as the same sensitivity analyses we used for Study 306B. We want to focus on this slide for one to show that these meta analysis showed approximately over the 60% reduction in falls although these data were statistically significant. Like for Study 306B, the suggested benefit from falls could not be erased with removal of top 2, 5 or 10 fallers from each treatment group showing that the results were not driven be a few outliers experiencing a relatively large number of falls.

The last slide I will show before turning the presentation back over to Joe is slide 18 which is a forest plot of all our randomized controlled trials conducted to-date. We divided it into more traditional induction designs on the top and the withdrawal designs on the bottom. The data for the induction study shown in this slide is change in dizziness, lightheadedness after week one of therapy but week two data on this endpoint are shown for the withdrawal studies.

As you can see, the point estimates for the improvement in dizziness for the induction studies will remarkably similar. We now have two positive randomized controlled studies, Study 301, and the Study 306B where we prospectively defined the primary endpoint and showed statistically significant symptomatic benefit for Northera as measured by the primary endpoint.

The point estimate for Study 306A, a 51 patients study, was also consistent with data from these two studies showing benefits for Northera as with the meta-analysis of 306A and B which was already discussed showed statistically significant benefits for this parameter favoring Northera.

Conversely, neither of our withdrawal studies, studies 302 and Study 303 showed statistically significant benefits from the perspective of dizziness endpoints but both did show that point estimates were to the left of the zero bar that is also favoring Northera.

As we have discussed in the past, we believe that because of persistent carryover effects, withdrawal studies are less sensitive than induction designs in determining the treatment benefits of catecholamine prodrugs like Northera. Overall, these data favoring Northera along with other data showing that patients treated with Northera consistently experience pure fall related injuries, have higher blood pressure values, and generally tolerate treatment with Northera show a consistent treatment effect for Northera and demonstrate that Northera is safe and effective for the treatment of NOH.

With that, I will now turn the presentation back of the Joe.

Joseph Oliveto

Thank you, Bill. While clearly 306B enriches our database and knowledge about Northera in NOH, and for this study, importantly, we had a number of key endpoints. For the primary endpoint, the mean improvement in the symptoms of dizziness and lightheadedness following one week of stable dose of therapy, we reached statistical significance, standing systolic blood pressure and important secondary measure of drug activity was also statistically significant at one week and the associate reduction in patient falls and fall related injuries are certainly worthy of further investigation.

Lastly, we saw consistency in Northera's favorable safety and tolerability profile, where results were consistent with our past studies including a low incidence of supine hypertension. But really this is just a start of our evaluation as these data are a part of our preliminary top line assessments, we look forward to digging deeper into the data set.

With 306B, we now have studied over 650 NOH patients constituting an extensive safety and efficacy database in this orphan indication. Understanding these data in total and interacting with key opinion leaders and with the FDA will inform us of the most appropriate next step toward marketing approval for Northera. We look forward to providing you with guidance on what these steps are once they become clear.

Before we turn to Q&A, we would like to invite our principal investigator, Dr. Bob Hauser to share his thoughts on the study including the questions provided on the next slide. Dr. Hauser?

Dr. Robert Hauser

Well, thank you very much for inviting me to be on the call and thank you for that presentation. Let me just start with a few words on NOH. NOH in Parkinson's disease can be disabling. Of course, patients may experience dizziness or lightheadedness upon standing and that can be an annoyance but it is really important to understand the range of symptomatology and the range of difficulty that it can cause for patients.

Patients may pass out or fall and injure themselves. They may be fearful of such things and really limit their mobility and function, really on a voluntary basis. The symptom of NOH may limit my ability to use Parkinson's disease medications to help the actual Parkinson's disease signs of slowness, stiffness and tremor and in addition orthostatic hypotension may be associated with other symptomatology like fatigue or cognitive dysfunction. So this is really an important issue in Parkinson's disease.

So with regard to the 306B study, I am really pleased to see this result. What we saw was this both statistically significant and clinically significant results with the 1.0 unit improvement regarding the dizziness, lightheadedness item and as was pointed out that amplitude of improvement is considered in the clinically relevant range. So I was very pleased to see that and I think that is the most important result in this study and that was backed up by the improvement in standing systolic blood pressure. To see that supportive information, I think, is a quite powerful as well.

The other half of this is the safety and tolerability data and that were quite good as well. The overall percent of patients experiencing treatment emergent adverse events was really quite similar in the placebo and Northera groups. There was about 6% more patients discontinuing in the Northera group which is a pretty darn good. We did see a little bit more headache, dizziness and nausea in the Northera group, which is to be expected and in the presentation we saw that supine hypertension incidence was really quite well with about 2% more in the Northera at that 180 millimeter of mercury level. So that is really outstanding.

So, overall, the efficacy, safety and tolerability looked quite good in this study. It is especially good when you think about this population. This is an older population with Parkinson's disease and orthostatic hypotension. So this is normally considered a relatively brittle population. So I think that is great.

There are other potential treatments for orthostatic hypotension but they have definite limitations and side effects. So additional therapies would definitely be welcome, especially those that are effective and very well tolerated. So I am looking forward to seeing more studies with the Northera. I have to see if this can be moved forward toward the marketplace.

In addition, in the presentation, we saw this very intriguing results regarding falls. Of course falls are a major source of morbidity. It’s the major cause leading to hospitalization in Parkinson's disease patients. I think this really deserves careful explanation and exploration with regard to what is causing falls in Parkinson's disease, how Northera might improve and reduce falls, what kind of falls it may have an effect on and how that affects. I think this is really a very intriguing area that deserves further investigation.

Joseph Oliveto

Great, Dr. Hauser. Thank you for those comments. Much appreciated and we will open up the call for questions of the management team as well as Dr. Hauser.

Question-and-Answer Session


(Operator Instructions) Our first question comes from Robyn Karnauskas from DB. Your line is open.

Robyn Karnauskas - Deutsche Bank

Hi, guys, thanks for taking my questions and thank you for putting up the slides. I think that was very helpful for us all, and I think that was great and congratulations on the data. I have several questions. I guess the first one is, big picture, what does this mean? I know originally the primary endpoint was falling, so you did change your primary endpoint. What do you think this means as far as the FDA's view of Northera? Do you have any preliminary sense?

Second, along those lines, the data was not statistically significant over time. So how does that affect your ability to develop the drug, and what are your thoughts on those two things? Thanks.

Joseph Oliveto

Thank you very much, Robyn. I will just open up but really transition to Bill, as well as Art to comment on those items. First of all, let's start by focusing on our intended indication which is for neurogenic orthostatic hypotension and you are absolutely right this Study 306B was changed from a fall study to try to reposition it back at this point time to support the NOH claim with the primary endpoint. The FDA's view on fall is that obviously they are very clinically important endpoint but they didn’t necessarily associate falls with NOH. So hence the change back of the primary endpoint to Item 1 dizziness, lightheadedness.

With regard to, and I will allow Bill to comment on that but first I will just mention with regard to how we would position this relative to interactions with the FDA. The basic stance we see now is you have to have enough data for really a very short amount of time and being able to give you the top line. Our thoughts are pretty much digestive and spend a little bit more time digesting it and then see if we can learn from this data relative to what an ideal plan would be and to, quite honestly, educate our colleagues at the FDA on that as well. They view this as a very important study, a very sizable study and we have every expectation that they will be looking forward to hearing the data.

So let me turn it over to Bill.

William Schwieterman

Well, thanks, Robyn. Actually, I think you are hitting on a lot of important questions and I will just give my opinion here quickly. A, I think these data show consistency with that not just between 306A and 306B but across all of our endpoints. That includes both efficacy and safety. On the two, I think it is important to remember that this is the second time we have shown symptomatic benefits in this study which what the FDA is asking for. That's also important.

The question on durability is one that we really can't answer. WE have had some initial discussions with the FDA about what it would take to show a durability response over time. Throughout the course of the year, we have met with the FDA had these preliminary discussions and they have appeared at least in some instances to be open to the idea of shorter-term studies with trends supporting long-term benefit. At least that was some of the early discussions at some of our meetings in the spring.

We just can't say exactly how the FDA is going to view this until we actually present these data and have guidance from them on what they want to do. They are fully aware and actually have made us aware from the very beginning of the variability associated with the patients in this condition, of the difficulty with the metrics, of the difficulty with study designs overall and we had productive scientific discussions about how to get there and I think that they are going to continue.

My last point, I will simply make is, just speaking of having been at the FDA, like Joe said, we are not considering this to be a confirmatory study but certainly, they are supportive of many of the things that we have already presented and they have reviews in our original NDA. I think they will continue to review that.

Robyn Karnauskas - Deutsche Bank

That's helpful. I just have a quick follow-up. So do you have a sense of why, what about studies or the way these patients take the medication would cause it to be not statistically over time, given the size of the trial?

Second, just out of curiosity, just thinking from a marketing standpoint, if you do come to market and you have a benefit in falls, that would be great. What is the randomization of heavy fallers evenly distributed between the placebo and the treatment arm?

William Schwieterman

Those are great questions. There is a whole graduate school course on that. The variability associated with patient reported outcomes, but I think it's an important thing to remember that with any PRO whether its depression, whether its pain scales, whether its NOH, you get a great deal of variability simply due to the subjective nature of the scale and that’s what we are seeing here. We are also confounded somewhat by the high placebo effects associated with that, and also that these patients are older with cognitive defects that could also influence how they go about answering these questions.

My own anecdote about this and we have sat many hours looking over the data to figure this out but talking to the patients actually verifies what we already deduced. One of our star patients who went from basically housebound to then picking up gas and doing other sorts theatrical productions and so forth, told me that she sat in the parking lot before going into the doctor's office because she didn’t know how to answer the OHSA-1 scale because she couldn't understand exactly how to correlate her findings on the number on this particular questionnaire because she didn’t know what to integrated on what days during the week.

She didn’t know exactly how to frame and so forth and this was a patient that had dramatic response droxidopa. We actually had her at one of our support groups counseling other patients. I use this anecdote only to reinforce the idea that when you put out a questionnaire that actually asks for subjective outcome measured over a week, noise is introduced in to the study and the effects are somewhat variable.

I have to say another anecdote came to mind and then I will turn it back over to Joe. Another nurse also said the scale didn't capture the benefits. So I think what we are seeing here is a lot of variability from a lot of different angles.

Joseph Oliveto

All right. I would just like to remind everybody one thing. While these are large studies in a relative sense they are still quite small in an absolute sense and that's one of the major problems we face in developing Northera in this area that this is an orphan indication. These are patient that are very hard to acquire and with the amount of variability as Bill has just alluded to and the placebo response, we don’t have access to the typical way of dealing with that which is a lot of patients. So it is typical. I just want to caution everybody, again, hanging everything on statistical significance but point estimates of the effects are also important.

And Robyn, I guess your second question was with regard to heavy fallers and randomization of heavy fallers. Again I will ask the guys to comment but we have to remember quiet honestly that this Study 306, before it was turned in to 306A and 306B was originally an NOH study with falls as an important secondary assessment and as such we do not prospectively bring in heavy fallers or anything. We didn’t stratify fallers. Later after 306 was split in to A and B, we saw just a remarkable impact of falls on 306A, we turned it into the analysis of B, but clearly a fall study would not be designed exactly this way where you allow patients and 40% of them are not falling.

So if anything I think what's interesting is we are seeing effects in falls in these patients without the study being formally designed that way prospectively from the start.

Robyn Karnauskas - Deutsche Bank

But did you happen to have evenly randomized?

Joseph Oliveto

It’s a good question. We don't have the falls data. It’s a very good question. We don’t have the baseline demographics on history of falls in each of these patients. And you have even brought this up before, Robyn. I think it’s a great idea when in future fall studies, which this was never primarily designed to be, it was only secondary endpoint and then an exploratory endpoint on falls just to verify the effect. In future fall studies, that’s probably exactly what we will do. Take patients who have a history of falling. I don’t want to say we are necessarily going to do this design but looking to designs like this where we capture heavy fallers, measure them over a running period and then perhaps randomize them.

I think that might let us get a handle on the variability that we see.


Our next question comes from Alan Carr from Needham and Company. Your line is open.

Alan Carr - Needham and Company

Hi, thanks for taking my questions, a couple of them. One of them, I am wondering if you guys have enough resolution here on activity across sites. Is there the same level of effect across different sites?

William Schwieterman

Okay, I think Art, you are in the best position to answer that one.

Arthur Hewitt

Sure, we did analyze the larger enrolling sites to see if there was a slight interaction effect and we did not observe one. So the short answer to your question is no.

Alan Carr - Needham and Company

Okay, and then how about OHQ? I know the FDA has a composite scale, the FDA is not as interested in that, but I am curious, was there some concordance there between Item 1 and the rest of the composite?

William Schwieterman

Yes, Alan, this is Bill. There was concordance. It tended to be the delta tended to be lower. Interestingly, the OHQ was statistically significant at week two in the study, just to show that you get variations and so forth. So the dizziness was not statistically significant at that endpoint but the OHQ was and so they track together throughout the study. We didn’t obviously position the OHQ very high in the analytic plan just because the FDA has made it clear that they like the dizziness endpoint.

Alan Carr - Needham and Company

And then you will have discussions with the FDA about this. I think you have previously said you were aiming for starting another confirmatory trial in the third quarter of '13. Is that still the plan here?

William Schwieterman

Yes, Alan, I think basically we are looking at that as a scenario in the way we always provided that guidance was particularly with regard to financial guidance. People always want to know the runway, et cetera. What we what we typically answer, and what we continue to think is that it will take us a little bit of time to get our hands around the data. It will take, depending on scheduling some time to get this in front of regulatory authorities and have discussions and have meetings coming out of that.

Or having minutes rather coming out of those meetings and with the expectation of clarity we would be in a position to be starting a trial next year was a really kicking in financially in the second half of the year. So that’s how that guidance was provided.


Our next question comes from Scott Henry from Roth Capital. Your line is open.

Scott Henry - Roth Capital

Thank you, and good afternoon. I guess I will start with a big-picture question. A week ago, when we thought about the results of 306B, the thought that it would really define the predictability of the next trial to be run, and when I am looking at the data, clearly to me, the Week 1 efficacy confirms that thesis. It makes me more confident that it will work at Week 1. My question is, from a longer-term durability standpoint, how does this data impact your belief in the ability to reproduce these results in another trial? Does it make you more confident? Less confident? The same? And I guess, what are the levers there? One of them will be, obviously, how much significance the FDA wants to see but I would like to get your big-picture thoughts on that.

William Schwieterman

Scott, that’s an excellent question. You have to keep in mind that we have only had these data for a few days and so forth. I think the data from 306B is supportive of the long-term efficacy of the trial but it raises questions whether you want to use the OHSA-1 score as we have been using it in a two or three month trial over the long term at least with this patient population. Just given the variability that’s involved and the large patient numbers that would be required to do this.

Certainly in another indication, we see clinically meaningful benefit after Week 8 with the OHSA-1 but because of the variability you would need on the order many hundreds of patients to do this and I don’t that’s in the best interest of the patients and so forth. But I also want to add that we don’t necessarily need to study these exact patients. We haven’t looked at the particular demographics in this population and there maybe subgroups that benefit more in which case we could use the OHSA-1 and I am going to ask Dr. Hewitt here to talk about this in a second but there are also some intriguing other endpoints that we just preliminarily begun to explore that perhaps they combine OHSA-a with blood pressure or other sorts of variables that again increase the power.

I think the name of the game here in NOH is to recognize the variability inherent in this patient population upfront and now that we have a 225 patient database which is by far in a way the largest ever assembled in the world, I think we are going to have the capability to really explore that for various options. I am really encouraged by subpopulation of patients that have the 25% and 50% improvements from baseline over the three or four increases in the improvements in the OHSA-1 from baseline.

That’s what we are seeing in other studies. That’s what we are witnessing from patients and we observed it again here. By taking those particular patients with big responses, we might be able to diminish the placebo effect, decrease the variability, and thereby increase the power in a way that doesn’t require us to do many hundreds of patients.

But again, I don’t want to speculate too far down this line because we have yet to really get into the data. But Art and I talk about this on a regular basis. Art, you want to comment on that?

Arthur Hewitt

I only want to say that I think that’s correct. We have only had the data for a short period of time but now that we have it in our hands, we can certainly investigate it, interrogate the data with a view to, are there characteristics of the patients that we can use our advantage to and magnify the difference between the placebo and treatment arms that would allow us to do longer-term studies with a smaller number of patients. I think it is very likely that there are. But we just need a little more time to really look at that closely.

Scott Henry - Roth Capital

Okay, and I had a couple of specific questions as I look at the slides. On Slide 9, the improvement of dizziness is very strong in Week 1, and also, encouragingly, the improvement seems to separate from Week 2 to Week 8, which is a good thing. What is surprising is the drop-off in Week 2 versus Week 1. I wanted to get your thoughts on what you think is going on there.

William Schwieterman

Scott, I think we are careful about any given time point of point estimate in here although I think you are raising a fair question. Simply because of what we were speaking about this entire conversation about the variability involved. To look at any one particular point estimate in one study, really I think, may mislead you on the overall effect but let's just suppose that this effect is real which it may very well be, that there is a drop off in the effect.

When we examine the literature for patient reported outcomes, we see similar sorts of phenomenon where in patient reported outcomes the initial effect tends to be the highest but then there is diminution of differences over time simply because of the patient recalibrating their baselines. So to put this into plain English, you get older patients and we have a mean age of about 72 in this patient population, many of them with Parkinson's disease and cognitive defects. So after two weeks on therapy, it's difficult for them, at least a reasonable subpopulation of them to recall how they were not a week ago but two weeks ago. Because they are comparing how they were to their baseline scores at the beginning of the study.

If they are confused themselves and compare how they were before that, then they are actually adjusting their baseline and going on. So this is documented in the literature. I hesitate to say more than this because I can't prove that is going on but this is certainly consistent with other data. What I think is most important, though, is the point that you made, Scott, is that it is unlikely that this drug is working for a week and not working for week and then starts working again and then continues to work through Week 8.

It's much more likely that the metric itself is the one responsible for the variation as well as the patient variability and that we see the consistency over the eight weeks and then by eight week, we see that delta. I think this is important. It's in the observation.

Scott Henry - Roth Capital

That's an interesting color on the topic. I mean, is it possible there's any tolerance building up, and perhaps you would need more dose titration? I am just looking at the data.

William Schwieterman

Well, it’s a fair question because there is tolerance built up to other agents that boost blood pressure. When you examine the literature for midodrine, for example, you get that tolerance from those patients who go on drug holidays because of that. That’s because of receptor saturation and down regulation of the receptors and so forth. So it’s a fair question for this drug.

We examined this very carefully before the advisor committee. The Japanese and there are a number of slides in the advisor committee slides I refer you to. I am not going to go over those here except to say that the Japanese have examined this from a number of different angles including looking at receptors over time in patients who were treated with droxidopa and what they found was that there was no down regulation at all.

The second thing I will point out is that our serum levels of norepinephrine are quite low. Below the levels that have been documented in the literature to down regulate the Alpha-1 receptors over time. The last bit of evidence, which I think is the most striking, is that if you give this drug to patients with dopamine beta-hydroxylase deficiency, recall that DBH is a really rare disease where you are missing the enzymes that allows norepinephrine to be formulated from dopamine, you give this drug to them at fairly high doses, they take this drug for decades without any diminution in effect overall.

So the wealth of the evidence combined with our own long-term data actually supports the long term effects and I think reading more into this Week 2 data over time, probably will mislead you more than anything because of our own year, two year long data starting to have relative effect.

Scott Henry - Roth Capital

Okay, and then another question on Slide 10, and this may very well be noise, but I think it's an interesting question to ask. When you look at changes in standing systolic blood pressure which should be a relatively objective data point strong benefit in Week 1, 2, and Week 8, yet in Week 4, active and placebo appear to overlap. What do you think is going on there?

William Schwieterman

Well, I hesitate to sound like a broken record because it’s the same set of arguments. If its anything, that if symptomatic benefits are variable then the blood pressures are ten times as variable. These patients can drop or raise their blood pressure by 30, 40, sometimes 60 to 80 millimeters of mercury in a given day, if they have extreme NOH.

Now, these are Parkinson's patient who tend not to have that kind of a latitude but nevertheless the kinds of fluctuations you see in Parkinson's patients where they have low morning blood pressures and high blood pressures in the afternoon and so forth are operative.

I guess, the bottom line is, we get a dedicated study of blood pressure where we looked over a 24-hour period where the patients had their blood pressures measured in Study 305 every half hour. We documented not only this variability where they were much fluctuations in blood pressure but we also documented there a statistically significant improvements in blood pressure for droxidopa treated patients compared to when they were before droxidopa.

When we speak with cardiologists about this, they all say that the proper way to do this would be with a 24-hour study using a continuous cause and that’s something what we could consider.

Scott Henry - Roth Capital

Okay, thank you. I appreciate that color. The final question, just simply logistically to the best of your knowledge, when would you expect to meet with the FDA, and when do you think we would have timing on what could potentially be the design of the next pivotal trial?

Joseph Oliveto

Scott, again, the guidance after this couple of days of having the data is really no different than the guidance before we had data on this particular point. We feel we owe it to ourselves and to the drug to look into the data set and as well for the next data sets from the previous studies. So the expectation, as previously stated, was to get in front of the agency in the first half of next year and depending on their availability and ability to have meaningful conversations with clear direction, we will report guidance after that.


Our next question comes from Juan Sanchez from Ladenburg Thalmann. Your line is open.

Juan Sanchez - Ladenburg Thalmann

Good evening, guys, a couple of questions. Has the FDA ever explained to you the rationale of demanding different durability response criteria for you guys versus midodrine?

The other question is, what happens when you do responder analyses of 306B on weeks 2 and 8, for example? Is it still as good as Week 1 or not?

William Schwieterman

Okay, I will answer the last question first. We did responder analyses at all of the variables. In all of them, most prominently in weeks four and week eight, we saw benefits across a range of patients. So if you were to plop these on accumulative distribution function curves, just like we did for the advisory committee, they would mimic the spread across these variables.

I believe, off the top of my head, that several of these were statistically significant at the week eight value in particular or close to that, particularly the 50% reduction at the week eight time point. But again these are preliminary estimates.

The question as to why the FDA might require us to have durability effect and not midodrine, no, we have not formally asked that particular question of them in our meetings. I think it’s a fair question. We can only speculate on why they might think that. I don’t know.

I mean, the review team was with a number of issues with that, what they might think as some of the data on the blood pressure and in some of our studies as not being definitive enough. But the answer to your question is, we just haven’t asked that question directly.

Juan Sanchez - Ladenburg Thalmann

Yes, well, final comment, it is not like I am suggesting it could change the endpoint again, but looking through all these data, do you think a responder analysis could be a better endpoint than just a mean improvement analysis?

William Schwieterman

Absolutely. We have done modeling of various data sets and like Art was saying, we haven’t had the time here to do it for this data set. We have only gotten these a few days ago. But responder analysis may in fact be a very good way forward. We just don’t know. It buys you a great deal of power in some of our own calculations where, say for example, you combine some minimum response in OHSA-1 with a minimum response in blood pressure but I don’t want to comment anymore.

We just don’t have enough information at this point to really comment on that.

Arthur Hewitt

I think I said that earlier that, first, and this is a personal opinion, just off the top of my head, but I do think there is an opportunity once we have a chance to look at the data in more detail to create certain responder outcomes that may enhance the power of our analysis and allow us to do longer-term studies with fewer patients.

One of the things to realize here is we have looked at 301 and 302 and found these populations as well. It shifts with signs of those studies. We just didn’t have the confidence given the size of those subpopulations. Now when you start thinking about this study, 225 patients in the meta-analysis, plus 301, plus 302 and you start getting up into the 500 and 600 patients, then when you find these subpopulations, I think you will have a little better, we will have a better chance of having some confidence that that’s something that we can base another study on. So we feel it is the first point in time in Chelsea's history where we are really going to developing a new study plan based on this wealth of data.


This ends our Q&A session. I will turn the call back to Joe Oliveto for closing.

Joseph Oliveto

Okay, well, again, I want to thank everyone for their time today in listening to these results and for the excellent questions that were provided. I also want to thank Dr. Hauser for his contributions, both on the study and for his patience in participating in the study and for his time on the call today. So thank you everyone and I wish everyone a good evening.


Ladies and gentlemen, thanks for participating in today's call. This concludes the call. You may all disconnect.

Copyright policy: All transcripts on this site are the copyright of Seeking Alpha. However, we view them as an important resource for bloggers and journalists, and are excited to contribute to the democratization of financial information on the Internet. (Until now investors have had to pay thousands of dollars in subscription fees for transcripts.) So our reproduction policy is as follows: You may quote up to 400 words of any transcript on the condition that you attribute the transcript to Seeking Alpha and either link to the original transcript or to All other use is prohibited.


If you have any additional questions about our online transcripts, please contact us at: Thank you!

About this article:

Tagged: , Biotechnology,
Error in this transcript? Let us know.
Contact us to add your company to our coverage or use transcripts in your business.
Learn more about Seeking Alpha transcripts here.