DURECT Corp (DRRX) 2018 Q2 法說會逐字稿

Greetings, and welcome to DURECT Corporation conference call.

(Operator Instructions) As a reminder, this conference is being recorded.

I would now like to turn the conference over to your host, Matt Hogan.

Please go ahead.

Hey, good afternoon, and welcome to our Second Quarter 2018 Earnings Call.

This call will begin with a brief review of our financial results, and then Jim Brown, our President and CEO will provide an update on the business.

We'll then open it up for a Q&A session.

Before beginning, I'd like to remind you of our safe harbor statement.

During the course of this call, we may make forward-looking statements regarding DURECT's products and development; expected product benefits; our development plans; future clinical trials; our projected financial results.

These forward-looking statements involve risks and uncertainties that can cause actual results to differ materially from those in such forward-looking statements.

Further information regarding these and other risks are included in our SEC filings, including our 10-K and 10-Q's under the heading Risk Factors.

So now turning to our financials.

Total revenue was $3.4 million in the second quarter of 2018, which compares to $4.3 million in the second quarter of 2017.

If one excludes all deferred revenue from upfront fees already received by the company, then revenue from our R&D collaborations was up approximately $200,000 in the second quarter compared to the second quarter of last year.

Product revenue from the sale of ALZET pumps and LACTEL polymers was $2.8 million in the second quarter 2018 as compared to $3.1 million in the second quarter of 2017.

The gross margin for the combined ALZET and LACTEL product lines was 61% in Q2 2018, and these product lines continue to be strongly cash flow positive for us.

R&D expense was $6.1 million in the second quarter of 2018 as compared to $9.1 million in Q2 2017, primarily due to lower R&D costs associated with POSIMIR, partially offset by higher costs for DUR-928.

SG&A expenses were $2.8 million in Q2 2018 as compared to $3.7 million in the second quarter 2017.

In the second quarter 2017, we did have a large onetime kind of payment for some advisory work that was done during that quarter.

At June 30, 2018, we had cash and investments of $42.5 million, which compares to $36.9 million at December 31, 2017.

Jim will elaborate on this in a minute, but the approval of debut PERSERIS drug triggers a $5 million milestone payment to DURECT, which we would expect to receive in the third quarter.

We had an underlying cash burn rate of $6 million in the second quarter of 2018, which compares to $6.7 million in the second quarter last year and $6.9 million in the first quarter of this year.

These figures exclude any licensing deals or equity financing activities or proceeds from option exercises.

And given the milestone payment I just mentioned, our burn rate in Q3 should be quite modest.

In Q2, we raised $3.1 million net of expenses from using our ATM facility that settled in early Q2 by selling around 1.5 million shares at an average price of $2.22.

These sales actually occurred at the very end of Q1, but they settled in Q2.

And I just wanted to briefly mention 2 SEC filings that will take place tomorrow, along with our usual 10-Q: the first is to register the shares underlying the increase to our option plan that was approved at the shareholders meeting in June; and the second is to file a new shelf registration statement, our existing shelf registration statement expires in November because such filings or statements have a 3-year life.

And given that, there are some efficiencies by filing right after filing our 10-Q.

So we view these filings as normal corporate housekeeping, but I thought I'd mention it just so people have an explanation when they see the filings.

And with that, thanks for joining the call, and I'll turn it over to Jim to go through some nonfinancial matters in greater detail.

Thank you, Matt, and hello, everyone.

Let's briefly review what we've accomplished through July of this year.

We started 2 Phase II trials with DUR-928 for 2 different indications, with a third on track to start in this quarter.

Indivior announced this week the approval of PERSERIS, which was formally known as RBP-7000, a once-a-month injectable product for schizophrenia.

I'll now update on the programs in greater detail beginning with PERSERIS.

Over the years, we provided biodegradable polymers for many FDA-approved products.

We also have a partnered veterinary product approved in -- on the market.

That being said, PERSERIS marks the first approval of a product for human use, which is a nice milestone for the company and for our shareholders.

PERSERIS is the first once-monthly subcutaneous, risperidone-containing, long-acting injectable for the treatment of schizophrenia in adults.

Treatment adherence is a big challenge for schizophrenic patients, which is why the long-acting injectable market in the U.S. has grown to over $2 billion annually and is growing still.

Risperidone is a well-established treatment for schizophrenia.

PERSERIS provides sustained levels of risperidone for over a month.

The initial peak of risperidone occurs within 4 to 6 hours of dosing and is due to an initial release of the drug during the depo formation process.

PERSERIS may have product attributes that will be attractive to patients and to physicians and could enable the product to become a valuable addition to the large and growing long-acting injectable antipsychotic market.

For example, the fact you don't need a loading dose or any supplemental oral risperidone when the patient is first dosed with PERSERIS.

The approval was a result of our robust clinical development program.

Efficacy was evaluated in a pivotal Phase III randomized double-blind placebo-controlled trial of 8 weeks' duration in 354 patients.

An efficacy was demonstrated by an improvement in the primary clinical endpoint of the positive and negative syndrome scale.

This score was done on day 57.

There was also an improvement in the Clinical Global Impression Severity of Illness Scale, which was also statistically significant at day 57.

The clinical trials demonstrate that PERSERIS can be initiated without a loading dose or any supplemental risperidone.

Safety was evaluated in 814 adults with schizophrenia, who received at least one dose of PERSERIS during the clinical program.

322 patients were treated with PERSERIS for at least 6 months, with 234 of these having been treated for at least 12 months.

The systemic safety profile of PERSERIS was consistent with a known safety profile of oral risperidone.

We would encourage you to review the PERSERIS prescription prescribing information that can be found on indiviormedia.com.

Last year, DURECT received an upfront payment of $12.5 million, and we will receive a $5 million milestone based on the NDA approval of -- this week.

We will also receive quarterly earn-out payments that are based on a single-digit percentage of U.S. product net sales.

These payments extend until the expiration date of the patents covered by this agreement, which is until 2026.

Indivior stated that they commit to providing the launch timing of PERSERIS as soon as it is reasonably practical but no later than when it's third quarter earnings results are announced, which is currently scheduled for November 1, 2018.

Indivior have stated in the past that their peak sales projection for PERSERIS is between $200 million and $300 million annually.

Now to DUR-928.

DUR-928 is a lead product candidate of our Epigenetic Regulator Program.

DUR-928 plays a regulatory role in lipid metabolism, inflammation and inflammatory responses and cell survival.

It's an endogenous, first-in-class small molecule, which may have broad applicability in a number of disease states, including several liver and kidney diseases, such as alcoholic steatohepatitis or NASH and other disorders of the liver, including primary sclerosing cholangitis or PSC and in acute injury, such as alcoholic hepatitis and acute kidney injury as well as potentially in inflammatory skin disorders, such as psoriasis and atopic dermatitis.

Quite a list.

DUR-928 has demonstrated positive results in more than a dozen different animal models and achieved reductions in important biomarkers or other signals of potential clinical activity in Phase Ib single-dose studies in NASH, psoriasis and chronic kidney disease patients.

We have 2 Phase IIa studies underway in the United States that we are reinvestigating DUR-928 in PSC and acute alcoholic hepatitis.

The Phase IIa trial of PSC is a randomized open label study with 2 cohorts that are dosing at 10 and 50 milligrams orally.

Each of these cohorts will dose approximately 15 to 20 patients for 4 weeks with a follow-up period of an additional 4 weeks.

The objection of this study includes safety, pharmacokinetics and pharmacodynamic markers, including the percent change from baseline of serum alkaline phosphatase or ALP and other biomarkers.

Because this is an open-label study, we expect to generate interim data in 2018.

PSC is a chronic liver disease characterized by a progression of cholestasis that is a decrease in bile flow with inflammation and fibrosis of the bile ducts.

DUR-928 has also been granted orphan drug designation to treat patients with PSC.

The Phase IIa trial in alcoholic hepatitis, or AH is an open-label dose escalation study that's been conducted in 2 parts.

Part A includes patients with moderate alcoholic hepatitis as determined by the Model of End-Stage Liver Disease or MELD score of between 11 and 20.

MELD is a common scoring system to assess the severity and prognosis of AH patients.

Part B will include patients with severe alcoholic hepatitis and MELD scores that range between 21 and 30.

The study is being conducted using 3 IV doses of DUR-928.

For Part A, there will be 30, 90 and 150 milligrams.

The sequential dose escalation following review of the safety and pharmacokinetic results of the prior level.

The dose may be adjusted in Part B based on the findings from Part A.

Patients will be enrolled in multiple clinical sites in the United States, and we are targeting between 24 to 36 participants to complete the study.

The objectives of this study are: safety, pharmacokinetics and pharmacodynamic signals as determined by improvement in liver biochemistry, MELD and Lille scores and other biomarkers.

Because this is also an open-label study, we expect to generate interim data in this year, in 2018.

Alcoholic hepatitis is a syndrome of progressive inflammatory liver injury associated with long-term heavy intake of alcohol and involves a spectrum that ranges from mild to moderate to severe life-threatening liver damage.

The prevalence of AH has not been accurately determined, but it's believed to occur in between 10% to 35% of heavy drinkers, and they were over 320,000 hospitalizations related to AH in 2010, resulting in hospitalization costs of approximately $50,000 per patient.

Pharmacodynamic data from the use of DUR-928 in animal models of acute organ injury support the potential to approve the outcomes of AH patients and data generated from this trial has the potential to be relevant to other liver and acute organ injuries.

The final DUR-928 program that I will discuss today is for the topical treatment of psoriasis.

We developed a topical formulation of DUR-928 because of the promising results we achieved in an exploratory Phase Ib trial in psoriasis patients, utilizing intralesional injections of DUR-928.

We've completed the last nonclinical study requested by the FDA prior to the IND submission.

We expect to initiate this study this quarter.

There's a sizable market opportunity for new topical drugs in inflammatory skin diseases, such as psoriasis and atopic dermatitis.

Now to REMOXY ER.

REMOXY ER is a unique long-acting formulation of oxycodone that's designed to be dosed twice a day and to discourage common methods of tampering associated with opioid misuse and abuse.

On June 26, the FDA held an advisory committee meeting for this product candidate, and the committee voted against approval.

The PDUFA date is set for August 7.

The final product I'll mention today is POSIMIR.

In May, we amended our U.S. licensing agreement with Sandoz, pursuant to which DURECT is now eligible for up to $30 million in milestone payments based on NDA approval and remains eligible for up to additional $230 million in sales-based milestones.

With this amendment, each party is also permitted to develop or commercialize competing products.

The amendment also includes modification to DURECT's development obligations and to both parties termination provisions, including a right for DURECT to terminate for convenience prior to NDA approval.

There is also a new termination fee payable to DURECT in the event that Sandoz terminates the agreement for convenience.

The agreement between the 2 companies remains in full force and effect, except as expressly covered in this amendment.

We continue to evaluate and consider potential next steps for this program, and we will provide an update when the decision on the future of the program has been made.

In summary, PERSERIS offers differentiating features that may benefit patients and the healthcare system in a large and important market.

The approval of PERSERIS allows us to add this new revenue stream to our existing ALZET and LACTEL product lines.

Of course, all of PERSERIS payments flow straight to the bottom line as there are no internal costs associated with their generation.

It's nice to have unlocked the value from our extensive patent portfolio that's been built up over the years.

Of course, the most important component of DURECT's future potential is DUR-928, an endogenous small molecule that has demonstrated profound effects in various animal models and in single dose studies in human patients with what appears to be a wide safety margin.

We are dosing in 2 Phase II trials, with the third Phase II trial lined up for initiation this quarter.

During the remainder of 2018, we look forward to generating preliminary data from the ongoing DUR-928 Phase II studies and dosing patients in the third Phase II study.

A positive signal from anyone of these Phase II studies could be transformational for DURECT.

With that, we'd like to now take any questions you might have.

(Operator Instructions) Our first question comes from the line of [Ed Arce].

This is Matt McLaughlin on for Ed.

Just a quick question regarding the PSC and AH trials.

We see that the alkaline phosphate is the main biomarker.

What other biomarkers might you be using, such as like bilirubin or others?

Yes, we'll be looking at a number of other things.

We have -- [they're too long to break them all right now], but they'll be -- you're right to look at the NASH Phase Ib trial and look at some of the biomarkers that moved there, and we saw it respond to bilirubin, and we also saw a full-length, cleaved CK-18's move, things like that.

So we'll be having a look at various biomarkers and see what we see.

There's been a lot of work recently done by a consortium of physicians who study acute alcoholic hepatitis, and that's led by Craig McClain, and there is a paper that's due to come out, maybe by the end of the year, that relates the CK-18's to survival of the patients.

And so those data could be quite interesting as well.

Right.

And in terms of minimum improvement levels, what might you need to see to consider it a success and proceed with a definitive POS Phase IIb study?

I think we'll have to -- it will have to depend on the data, right, and the circumstances.

Each patient is different, I'm thinking now in particularly of the alcoholic hepatitis patients, in that case, each patient has their own course and everything else.

So we'll have to look at how they're doing clinically, what their blood chemistries look like and what these various biomarkers and scores come out to be.

In PSC, it will be certainly looking at alkaline phosphatase and how the patients are doing generally.

So each one, I think, you have to -- especially in these small trials, you've got to look at the whole of it all, each patient and then determine next steps.

Just one more.

With regard to PERSERIS.

I think you'd mentioned peak annual sales in the range of $200 million to $300 million, but when might you expect to see an earn-out from revenues there?

Well, it depends obviously when they launch.

We get paid on a quarterly basis, right?

So if one looks at the U.S. market for these long-acting injectables to treat schizophrenia, they're quite impressive, I think, just the 2 J&J products.

I think about INVEGA SUSTENNA, I think, is close to $1.6 billion and Risperdal Consta is over $360 million.

So those plus everything else adds up to more than $2 billion worth of sales for the U.S. So it's a big market opportunity.

They projected $200 million to $300 million.

They'll have the first -- they have the first subcutaneous injectable risperidone product out there, they have a product that starts right away within hours, which is great because compliance can be a problem with these patients.

And if you can keep them on their meds, it's a much better life for them and their families.

So we're hoping that [they] can get out there and make a difference.

Anyway, long-winded answer, but as soon as they start selling the product within the next quarter, we'll get a check.

Our next question comes from the line of Adam Walsh.

Just a couple of them here.

The first one, can you just remind us, Jim, a little bit more about what you saw in the earlier DUR-928 studies and how you think those results could transfer to the Phase II PSC and AH studies?

That's the first question.

And then the second one, Jim, is can you outline some of the thresholds in your own mind for releasing interim open label data in the ongoing studies that you said could come this year?

What are you thinking about in terms of data releases?

What do you need to see?

What are the hurdle rates that you have to clear before you make those kinds of announcements?

Very good questions.

The first -- I'll take them in order so that they started -- although they only started just a month or 2 apart.

So [I'll] start with PSC.

The main -- we have a lot of chronic liver studies that we've done in animal models where we've seen damage from various circumstances, right, primarily they were metabolically- damaged animals vis-a-vis diet or sometimes assault on the liver function, like the study done over in Japan where the mice are made diabetic at a very young age and then fed a high-fat diet.

So it's kind of a 2-trigger point there but basically, kind of a metabolic assault in the animals, and then there's the physical assault on the chronic form in the way of the bile duct ligation model.

And we also look at models where they was genetic damage in Leptin-deficient animals.

And what we saw across all of these was generally an improvement in the animals' outcome.

We saw an improvement histologically.

If you look at the livers, we saw reductions in various biomarkers depending on the model structure and everything and the duration, but we saw things like reduction in lipids, both in the circulation and in the liver.

So we saw reductions in cholesterol and triglycerides and the like, as you would expect if you look at the genes that DUR-928 influences.

We saw reducing inflammation, and we saw an improvement in reduction in hepatocyte ballooning and fibrosis, depending on various models.

I think the most relevant to all of these as directed by [Frederick Niven], who's an expert in PSC from the University of Leuven.

He was the one that directed us to the bile duct ligation model and there, we've seen some pretty impressive data.

What we did was actually tied off the bile ducts, and we've done this trial, I think, maybe 3 or 4 times now so a lot of different times, different doses.

And we've seen very nice improvement in the animals, and I think most spectacularly, as compared to pretty much anything else I've seen out there, is reduction of the bilirubin and also the bile acids and other things.

So we haven't -- I won't go into some detail because we're hoping maybe to publish some of this.

There's some recent work we've done, and maybe there's a poster or a paper that could come out of that.

But some really nice data that if those things hold up for PSC, that led us to go into that path.

What we don't know about PSC, and you just don't know until you get in there, is what is the inciting cost.

One would think antiinflammatory could help and should help because of fibrosis.

But then if you look, 60% of PSC patients happen to have also concomitantly inflammatory bowel disease.

And so they're being treated with the IBDs with these very significant modulator, the REMICADE type of monoclonal antibodies that inhibit the immune system, and unfortunately, those treatments don't help these patients out seemingly.

And so it's a much more complex disease, but we're hoping DUR-928 will be able to help.

So then to go on to the AH...

[Why don't you mention] the NASH data?

(inaudible)

Oh, yes, I'm sorry, yes, Matt, Matt reminds me, yes, that we do have some data from NASH patients, and that was the single dose patient and there -- single dose study, sorry, there, we dosed 20 different NASH patients to 10 at 50 milligrams and 10 at 200 milligrams, and we saw really impressive data.

We presented these at EASL about 1.5 years ago, and what we saw there was a reduction in inflammatory or markers of inflammation, there was C-reactive protein and IL-18.

And these were all statistically significant, the ones we talked about and which was amazing considering only 10 patients per group.

And then we saw also statistically significant reactions in full and cleaved CK-18 equivalent to or greater than the kind of results of the reductions in these cell death markers that companies had seen after months of dosing with the other molecules that are in the hunt for NASH.

So it looks like DUR-928 might well have a really nice opportunity in NASH.

And then lastly, we saw a reduction that I don't know anyone else has seen unless they improve the liver function over many, many months and that was we saw a drop in bilirubin of about 30% plus within just 12 hours of dosing so -- which shows that hepatocytes are functioning better.

So -- in the liver.

So those 3 areas coincide with the 3 areas of effect that we see from DUR-928.

So those are all giving us.

And then that also leads into the alcoholic hepatitis because those same kind of results relate to there.

In the alcoholic hepatitis, the nonclinical studies that we looked at, we have kind of 3 general areas where we do -- have worked on the effect to improve the outcomes post a chemical assault and then, also, ischemic assault or biological assault, and the chemical assault has been alcohol and acetaminophen and the biological assault has been lipopolysaccharides or endotoxin, and the ischemic [cord] is cutting off the blood supply.

We have a number of different models that we've approached this in, in a number of different organs, including liver, kidney and lungs, actually, and we saw some pretty impressive data in all of these.

In fact, there was an M.D., Ph.

D., who works up in Canada, running a lab up there, where they do a lot of work on organ transplant and in that regard, that physician said that DUR-928 showed the best data he's seen in 30 years in his model.

And that was looking at a rise and protection against a rise in serum creatinine, which, of course, is very important to kidney, but it's also important in the MELD score because if you look at the MELD score, the Model of End-stage Liver Disease score, the 3 things that are used to calculate a patient's MELD score are bilirubin, creatinine and then prothrombin time, which is capacity of the liver to make clotting proteins.

And so we've seen in animals and in humans a reduction in bilirubin.

We have seen in an animal model of kidney disease a reduction in creatinine, and so our hope is that in these alcoholic hepatitis patients in the acute setting, we will be able to strengthen their livers and enhance their potential for survival.

That's the goal.

That would be the objective.

If we can see some improvements in MELD scores, those could possibly lead to a breakthrough designation by the FDA, and then we could be off to [write].

I think if you look at all the indications for DUR-928, we talk about these 3 Phase II trials, but quite frankly, there are 3 or 4 behind these, I'd love to start tomorrow if we had the money.

The quickest end market would be acute, where 1 or 2 injections could save someone's life.

And so that -- if you think about it just from a strategic standpoint.

So it's a very long-winded way of tying those ends...

That's really helpful.

That's a great summary of why you're pursuing what you're pursuing.

Just on the open-label data, can you remind us how do we think about when the data might come in terms of what thresholds or hurdles, or what [things we're] going to be looking for between now and the end of the year where we might get a better sense of what could drive that announcement?

Yes, I think we'll need to see enough patients that we get [a crack at that].

And we also -- what we don't want to do and what we actually are guarding against is just looking at patient 1 and then patient 2 and patient 3 say, "Oh, we're so happy, [there's] no more sad, and then we're happy again" kind of thing.

What we need to do -- and we have to be disciplined and gather enough patients.

And so to that extent, we're trying to gather that information before we have a look at it.

And really have a critical mass of patients to be able to say, "Okay, now we think maybe we can see something," as to what that exact number is, we haven't determined that we have some sense of it, but -- so there will be a threshold that we need to achieve as far as number of patients.

And then once we have that, we'll have a look and see what we have.

And then we'll change our direction one way or the other depending on what we're seeing.

Our next question comes from the line of François Brisebois.

That was a great answer, Jim, a lot was answered there on that last one.

But just wanted to touch quickly on SG&A, Matt.

You mentioned there was a onetime thing, I'm seeing it as it seems kind of low.

Is this a rate that -- even with the onetime, it seems a little lower.

Is this something that keeps going down?

Or how should we think about it down the road?

No, this is a more normal number.

Last year, we used an advisory firm to help us get the POSIMIR deal with Sandoz.

And we owe them a fee, which I think was like $750,000.

So that makes the comparison look easier, I guess, for this year's second quarter or the number last year look abnormally high.

But absent that, they're kind of -- it's more or less the same.

Yes, we certainly are (inaudible)

Great.

And then for R&D, with all these Phase IIs going on, how should we -- why is it that it's so inexpensive or that there isn't much of a ramp in expenses with these trials?

Well, they're small studies, the patient numbers that Jim described, and it's still early days in the studies.

So we never thought that they would be very, very expensive.

It's a function of the size of the study, really.

Right.

And it's -- remember, last year, we were wrapping up the PERSIST trial, which is a huge Phase III trial and so that -- the comparison between that, you can fund 3 or 4 of these before you could fund one large Phase III trial or more.

Okay, great.

No, that's very helpful.

And then -- so that -- you talked about the schizophrenia market last year, it was around or in '17, I guess, yes, last year was around $2 billion.

The differentiators here with PERSERIS, the way you guys see it, is it mostly in terms of doctors, [or] is it mostly that it's a once monthly, or is it the fact that it can kick in 5 or 6 hours after injection?

What is it that would make this mostly differentiated versus kind of outpatient?

I would say, yes because it's both of those, and I would add the subcutaneous, which is, if you've ever had an IM injection versus subcu, it's just easier on you.

I still remember when I was a kid, I got an IM injection in the rear end of penicillin, and it was hard to sit down for a while.

So subcu is much nicer.

Okay.

Okay, great.

And then just on the Phase II trials for DUR-928 here.

So you mentioned a little bit what we should be looking at in terms of interim, and you're not going to try to give it away too quick after 1 or 2 patients' kind of thing, even if it's open label, but is it for ALPs and the PSC trial?

Is it more of a normalization versus a reduction or just to make sure what we should be expecting once data kind of reads out.

Well, I think it's important.

It's a -- you make a really good point.

With PSC, yes, the doctors who treat PSC, [Keith Lindor] said they would like to see normalization, absolutely, everybody would for sure.

The reality of it is everything that's been used to try and help these patients hasn't done much of anything, and the best you've got is with some of these bile acid -- modified bile acids, which is like the aerosol products and in there, what you're getting is a percent reduction of 20% to 30%, let's say, and no one really knows clinically whether that -- what that means.

And so it's uncharted territory as far as how much can you reduce, and what does it mean to the patients.

And also what are the other markets, let's say, you have ex amount reduction in ALP, but then you have some reductions in some of these other markets like CK-18s.

No one knows what that means for these patients, right?

And so there's -- it's [sort of] like landing on a new continent, and you just have to kind of get out there and explore the territory and see what's going to happen.

But you wouldn't expect a normalized and only (inaudible) is too high.

Oh, no, no.

Yes, that -- yes, I think...

We're looking for improvement but...

Yes.

Thank you, Matt.

Normalization in [30] days, don't count on it.

No, yes, that's true.

Very good point.

Okay.

That's very helpful.

And then just switching quickly to alcoholic hepatitis.

In terms of MELD score, is that a -- the improvement there, is there a certain percentage point or number of points sorry, that would make sense, or once again, just how should we think of it?

Well, I think it depends on the patients.

We're starting in the moderate patients, right, which have 11 to 20 and in there, they have a much higher survival.

There the MELD number is almost directly related to the percent mortality expected.

In other words, 11 to 20 leads somewhere between 11% to 20% mortality in the next 60 days.

And so -- but that changes, it goes up almost in a log scale as you go to 30.

So by the time you get to 30, you're looking at 60% plus mortality.

So there's a difference in that scale.

That being said, these changes in MELD score don't change dramatically.

If you take a look at the natural course of history, and we've seen some studies where they look at 150 patients of AH and follow them over time, generally, their MELD scores stay fairly flat to maybe increasing over the first 4 weeks, and it just very general, each patient has a lot of noise, right?

But -- and so, we're looking to see what can we do to these patients, can we reduce, can we improve their MELD scores, and then certainly, as we get into the more severe patients where it's much more dramatic, if we can see some improvements there, then that's where you can start to really think about looking for some kind of response.

We hope that we can actually get enough information to project our -- to be able to design the next trial, which [would] be a much larger trial where [we] looking to show a percent improvement in survival, basically at some amount of time, somewhere between 30 to 60 days I'm hoping, but the FDA still hasn't come to a final agreement on that.

Our next question comes from the line of Len Yaffe.

Two questions for you.

You mentioned in your opening comments, in the acute injury, you mentioned potentially acute kidney injury, I know you had looked at this previously, and I was wondering if you could just update us in terms of what -- where your efforts currently are in AKI?

What your expectations for that disease [they] will be, given that it's a large unmet medical need?

No, no, as you know, it's a huge unmet medical need, and we look forward to, long for the day when we can do that study.

And we're working hard with experts in the field, and we're working hard on mechanisms by which we can figure out a way to fund it in a less dilutive way.

But these are all things we're thinking about, and WeiQi is up to her elbows in that stuff so...

Okay.

And then the second question, less significant, but I just noticed that your net product sales declined year-over-year pretty significantly, and yet the cost of goods in absolute dollars actually went up.

So I was just wondering if in terms of that legacy business, if there was anything that you could comment on to bring us up to speed with how it's doing?

The only comment I would make is, last year, we're a little bit hoisted on the petard of how well the LACTEL product line did last year.

Last year was -- so we set ourselves up for tough comparisons this year.

LACTEL on their side, last year, had their record revenues in the history of the company.

And this year will be a softer year in comparison.

Reason to think, in later years, it will pick back up and grow, but there are just 1 or 2 big customers that had huge orders last year that this year aren't going to recur.

So it's really on that side that the phenomenon occurred.

And I don't think the drop was that dramatic.

I think this quarter, it was $2.8 million versus $3.1 million.

And even with that, it's got a very, very nice gross margin out of those product lines, and they're very cash flow positive for us.

So..

(Operator Instructions) As there are no further questions, I would like to turn the call back over to Matt Hogan for closing remarks.

Okay.

Well, our only closing remark is, as always, if people have questions, feel free to call us, we'd be happy to try to be helpful.

And thank you for participating.

Thank you very much.

This concludes today's teleconference.

You may disconnect your lines at this time.

Thank you for your participation.