GW Pharmaceuticals plc
Q4 2016 Earnings Call Transcript

Published: 5 Dec 2016

[Operator:]

Greetings. And welcome to the GW Pharmaceuticals fourth quarter and year-end 2016 financial results. At this time, all participants are in a listen-only mode and a brief question-and-answer session will follow the formal presentation. [Operator Instructions] As a reminder, this conference is being recorded. I would now like to turn the conference over to your host, Steve Schultz, VP, Investor Relations. Please go ahead.

[Stephen Schultz:]

Welcome all of you and thank you for joining us today. I’m Steve Schultz, Vice President of Investor Relations at GW Pharmaceuticals. Today I'm joined by Justin Gover, GW's Chief Executive Officer; Dr. Stephen Wright, our Chief Medical Officer; Adam George, our Chief Financial Officer; Chris Tovey, our Chief Operating Officer; and Julian Gangolli, President of North America. I hope you’ve had a chance to review our press release from earlier today. This document will be supplemented by an additional 20-F filing later today. As a reminder, during today’s call, we’ll be making certain forward-looking statements. These statements reflect GW’s current expectations regarding future events, including, but not limited to statements regarding financial performance, clinical and regulatory activities, timing of product launches and statements relating to market acceptance and commercial potential. Forward-looking statements involve risks and uncertainties and actual events could differ materially from those projected herein. A list and description of risks and uncertainties associated within investment in GW can be found in the company’s filings with the U.S. Securities and Exchange Commission. These forward-looking statements speak only as of today’s date, December 5th, 2016. Finally, an archive of today’s call will be posted to the GW Web site in the Investor Relations section. I’ll now turn the call over to Justin Gover, GW’s Chief Executive Officer.

[Justin Gover:]

Thank you, Steve. And welcome to all those who are able to join us. We are pleased to be conducting this call from Houston, Texas where the American Epilepsy Society’s annual meeting is being held. On today’s call, reflecting the importance of the AES meeting, we will be focusing our remarks on the Epidiolex clinical and regulatory update from Dr. Stephen Wright. In addition, Adam George would discuss our financial results. At the conclusion of our prepared remarks, we will open the line for questions. Looking back, 2016 has been an exceptional year for GW, after initiating the Phase III pivotal program last year, this year we reported three successful pivotal trials; one in Dravet Syndrome and two subsequently in Lennox-Gastaut Syndrome. Combined, these trials will make up the principal safety and efficacy data set for the NDA. We have also held two pre-NDA meetings with the FDA and our strategy as agreed with FDA is to submit a single NDA for both Dravet Syndrome and LGS indication. We are making excellent progress towards submission of this NDA. Looking forward, the company has a robust balance sheet and is making excellent progress on plans for the future commercialization with all aspects of these preparations on track, including having significant manufacturing capacity, building a talented U.S. commercial team that will deliver a first class U.S. launch. We are well aware that the epilepsy community has been eagerly anticipating our Epidiolex data and the response has been very enthusiastic. Dravet syndrome and LGS represents some of the most difficult types of epilepsy to treat with nearly all patients continuing to have uncontrolled seizures and other medical needs throughout their lifetime. We are excited that our data showed Epidiolex to have the potential to offer much needed hope for patients and their family. Let me now hand the call to Dr. Stephen Wright to provide the Epidiolex’s clinical and regulatory update.

[Stephen Wright:]

Thank you, Justin, and good day, everyone. We are very pleased with the amount of Epidiolex data that’s been presented at this year’s annual meeting at the American Epilepsy Society. AES is the premier meeting for epilepsy specialists each year, and Epidiolex is featured in a variety of poster and podium presentations spanning both GW’s clinical program results and a number of our independent physicians involved in the Epidiolex expanded access program, including several posters associated with the Alabama and the Georgia state programs. I’m very pleased to say that when looking at the data presented overall, it remains very consistent with results that we have seen in previous presentations and supportive of the positive data we’ve seen from our pivotal program. I’m not going to cover all of these data and note that the abstracts are available on the AES Web site for your overview. But I do wish to highlight a couple of the new safety and efficacy abstracts from two open label, state sponsored expanded access programs. In the Alabama state expanded access program, the largest one of its kind involving 81 patients, data presented by the independent investigators showed in a real world setting, that at two consecutive visit 58% of patients had a greater than 50% reduction in seizure frequency, 36% had a greater than 75% reduction, and nine patients were seizure free. These results were maintained at three and six months. Interestingly, the patented response was generally similar between children and adults. A separate abstract on these same patients found that Epidiolex also improved seizure severity with 67% of patients experiencing more than 50% decrease in seizure severity after three months, and for those followed for six months 64% had a greater than 50% decrease in seizure severity. In the Georgia State expanded access program, which involves 30 patients, there was a mean reduction in seizure frequency of 69% after 27 weeks treatment including six patients that were seizure free. Today though, I would like to review in detail the GW sponsored Phase III Dravet syndrome and LGS trial posters. These poster highlight primary and a number of key secondary endpoints along with additional safety data. For your reference, these posters are available through links in a press release that GW issued earlier today and in the IR section of our website. In these remarks, I’ll look first to efficacy for each of the two studies and then address safety. Looking first to the LGS trial. This trial involved 171 patients in whom Epidiolex or placebo was added to current anti-epileptic drugs. Despite extensive background treatment, including an average of 10 other AEDs, the median baseline drop seizure frequency was 74 per month, emphasizing the high unmet medical need in this group of patients. The trial message primary efficacy endpoint with patients taking Epidiolex as an add on to standard of care achieving a median reduction over the entire 14 weeks treatment period and monthly drop seizure of 44% compared with a reduction on placebo of 22%, this difference being statistically significant with a p-value of 0.0135. The secondary efficacy endpoint data from the LGS poster included the following. In the 12-week, stable dose maintenance period, this is the period of stable dosing which follows two-week dose escalation period, Epidiolex treatment effect increased further achieving a median reduction and monthly drop seizures to 49% compared with the reduction on placebo of 20%, which was again highly statistically significant with the p-value of 0.0096. While no patients in either treatment group were dropped seizure free during the treatment period, three patients on Epidiolex were entirely drop seizure free during the maintenance period compared with no placebo patients. Caregivers of patients on Epidiolex were significantly more likely to report an improvement in overall condition on the caregiver global impression of change scale with the p-value of 0.0008. For total seizures, during the 14-week treatment period, patients taking Epidiolex achieved a median reduction on monthly total seizures of 41% compared with a reduction of 14% in patients receiving placebo, statistically significant with a p-value of 0.0005. When looking at the 12-week maintenance period, Epidiolex treatment effects increased further achieving a median reduction in total monthly seizures of 45% compared with the reduction on placebo of 15%, again highly statistically significant with a p-value of 0.0004. Median non-drop seizure frequency reduced by 49% in the CBD group, compared with 23% in the placebo group during the entire treatment period, again significant with the p-value of 0.0043. And finally looking at the responder analysis, a consistent separation between Epidiolex and placebo was seen across all drop seizures response rates. 64% of Epidiolex patients versus 44% placebo patients experienced a 25% or more reduction in drop seizures, which was significant with the p-value of 0.0081. 44% experienced at least a 50% reduction versus 24% on placebo, again significant as a p-value of 0.0043. 20% of patients experienced at least 75% reduction on Epidiolex compared with 8% on placebo, again statistically significant with the p-value of 0.0273. Turning now to the Dravet syndrome poster. This trial randomized 120 patients and achieved its primary endpoint. Patients taking Epidiolex achieved a median reduction over the entire 14-week treatment period in monthly convulsive seizures of 39% compared with the reduction on placebo of 13%, a difference which was highly significant with the p-value of 0.0123. The secondary efficacy endpoint data from the Dravet syndrome poster increase the following. In the 12-week, maintenance period Epidiolex treatment effect increased further, achieving a medium reduction in monthly convulsive seizures of 41% compared with the reduction on placebo of 16%, which was highly significant with the p-value of 0.0052. Three CBD patients achieved total seizure freedom during the entire treatment period, with one additional Epidiolex patient achieving seizure freedom during the entire maintenance period. No placebo patients achieved seizure freedom. Caregivers of patients on Epidiolex were significantly more likely to report an improvement in overall condition on the caregiver global impression of change with a p-value of 0.0155. For total seizures, during the 14 week treatment period, patient taking Epidiolex achieved a median reduction in monthly total seizures of 29% compared with the reduction of 9% in patients receiving placebo, again significant with a p-value of 0.0335. If we look at the 12 week maintenance period, Epidiolex treatment effect increased further achieving a medium reduction in monthly total seizures of 37% compared with the reduction on placebo of 10%, also statistically significant with a p-value of 0.0234. And finally looking at the responder analysis, a consistent separation is seen between Epidiolex and placebo across all response rates. 62% of Epidiolex patients versus 44% of placebo patients experienced at least a 25% reduction in convulsive seizures, a p-value of 0.051. 43% experienced at least a 50% reduction in convulsive seizures versus 27% on placebo, a p-value of 0.078 and 23% of CBD patients experienced a 75% reduction on Epidiolex compared with 12% on placebo, a p-value of 0.11. It’s important to put both of these trial results into their clinical context. The patients included in both studies are essentially coming towards the end of their treatment path. In all cases, there were few, if any, treatment options remaining. Where you to choose a patient population in whom it is difficult to show efficacy, this is the patient population you would select. In the Dravet study, 98% of patients had developmental delay and in half of them it was defined as severe or profound. There are no approved drug treatments in the United States for Dravet Syndrome. As has been reported previously, Epidiolex was generally well tolerated in these trials. The most common adverse events were consistent with what had generally been seen in expanded access studies. You will recall from previous expanded assets posters and the Lancet Neurology Paper that overall the most common adverse events seen in the expanded access program were also those seen in the Phase 3 trials. In our view, it’s important to look across the entire safety data set when seeking to understand the adverse event profile of a drug since there will inevitably be differences in each single study. For example in the Dravet syndrome trial, the list of common AEs and the overall incidents of adverse events was greater than in the Lennox-Gastaut study trial. Looking at the more serious adverse events, there was no difference in episodes of status epilepticus between Epidiolex and placebo for both trials and none of the episodes of status epilepticus were deemed to be treatment related and none of them led to withdraw from the study. The type of serious adverse events that we see has also being consistent with the common -- with the pattern of common AEs as you might expect. We are not seeing unexpected adverse events and in particular we are not seeing unexpected serious adverse events. When looking at the distribution of response and non-response there is no evidence that CBD might cause seizures to increase. In fact when looking at the proportionate of patients who deteriorated in these trials the numbers were very much in favor of Epidiolex whichever way you look at it. As one example in the Dravet study over the three month period 11% of Epidiolex patients deteriorated by 25% or more, compared with 20% of placebo patients. The posters did note increases in the liver enzymes ALT and AST to greater than three times the upper limit of normal and noted those had occurred in a proportion of Epidiolex patients, not all of these occurrences were reported an adverse events, and most of the patients experiencing these increases at ALT and AST were on concomitant valproic acid, a drug known to cause elevated liver enzymes, all elevations of the enzymes result during treatment period and while most on Epidiolex. There is a set of accepted expected criteria for identifying whether these transient elevations of liver enzymes are a signal for serious liver toxicity known as Hy's Law. None of the cases that we have seen on Epidiolex meet these criteria. These increases in liver enzymes were noted in the previously published Lancet article on the expanded access programs. So this is not a new phenomenon and it is one with which epileptologist are familiar because it is seen with a number of other anti-epileptic drugs including carbamazepine, oxcarbazepine, valproic acid, levetiracetam among others. Indeed its customary practice when introducing a new anti-epileptic drug for clinicians to monitor patients closely including having routine liver tests. As I have often commented, my view is that the key to understanding the safety profile of a drug is to evaluate whether adverse events lead to withdraw from treatment. In our Phase III trials the discontinuation rates are well within the range generally seen with other anti-epileptic drugs and lower than the clobazam trials for example. Even more insightful is the evidence from the real world use of Epidiolex that now exists, where 98% of patients who completed the pivotal Phase III trials have entered an optional long term extension study. The long term withdrawal rates from both the expanded access program and the open label extension remain low, approximately 20% over the long term. Perhaps close to 80% retention and long term usage of this patient population is a powerful comment on the long term tolerability of Epidiolex. Let me also comment on another topic of much discussion, that of drug-drug interaction. We now had extensive real world experience with approximately 1,000 patients currently on treatment. We do not consider there to be any clinical relevant drug-drug interaction except for that with clobazam, which I will discuss further shortly. Drug-drug interaction in epilepsy are common place and clinicians routinely monitor blood levels of concomitant AEDs. The adequacy of the scope of our drug-drug interaction studies to be submitted in our NDA was again confirmed at our recent pre-NDA meeting, making us comfortable with our approach to this important issue. We view this package as helping to guide labeling on how Epidiolex should be used as adjunct therapy. To-date, the only drug-drug interaction, we have any evidence for is the pharmacokinetic interaction with clobazam, which results in increases in its first metabolite nor does methyl clobazam, but not in raised levels of clobazam itself. This is seen in the poster presentation -- in the poster presented on the pharmacokinetic part A of the Dravet Phase III study. We have positive evidence of a lack of a drug-drug interaction with several of the most commonly used anti-epileptic drugs including valproate, stiripentol, leviteracetam and topirimate for example. We are aware of some market commentary on this observation. The regulatory question we would be required to address with respect to the clobazam observation is whether concomitant clobazam is required for efficacy of CBD. Although, I should point out that at our pre-NDA meeting this issue was not raised by FDA. So mindful of investor interest in this topic of whether clobazam is required for Epidiolex efficacy, let me make the following observations with some additional new data. First, we know that CBD applied on its owned is anticonvulsant in vitro and in vivo models of seizure. Second, in the individual expanded access sides that first reported the potential interaction, the publish paper shows that for patients taking Epidiolex and clobazam, reductions in concomitant clobazam do not result in reduce efficacy. In the larger pool analysis of the expanded access program presented that last year AES seizure reduction did appear to be enhanced in patients overall on concomitant clobazam. However, in the subset of patients with LGS and Dravet, concomitant clobazam had not effect on reported seizure reduction responder rates observed on CBD. Within both the expanded access program and the pivotal trials there have been individual patients who achieved meaningful seizure reductions even becoming seizure-free who were not taking concomitant clobazam. In addition to these important high level observations, we’ve chosen today to provide a little more data to investors from one of our Phase III trials to further highlight the type of analysis that helps reinforce this issue. This data aims to address the discussion from a known pharmacokinetic perspective. The literature indicates patients taking stiripentol have maximum inhibition an enzyme called CYP2C19. Inhibition of this enzyme means the patients taking stiripentol and clobazam have no further potential for increasing the levels of nor-clobazam, the key active metabolite in questions. A good example of this is provided in our Dravet part A poster where there was no increase in the nor-clobazam concentrations for the patients taking clobazam, who were also taking stiripentol. Clearly, if the efficacy of CBD depended on raising nor-clobazam levels than patient on concomitant CLB clobazam and stiripentol would see no clinical effect since such patients would not experience any raised nor-clobazam levels. In our Dravet Phase III study the median reduction from baseline in convulsive seizure frequency in patients taking CBD plus clobazam and stiripentol as statistically significantly superior to patients taking placebo plus clobazam and stiripentol. The significant difference cannot be explained by increases in nor-clobazam levels. Therefore it’s reasonable to conclude that CBD efficacy is independent of elevations of nor-clobazam. We hope that this additional data is helpful to address this point. Turning now to our NDA preparations, we have in recent months held two pre-NDA meetings with the FDA. In the first meeting in July, the proposed preclinical and clinical content of the NDA submission was confirmed. We also confirmed that we would be filling a single NDA for both the Dravet Syndrome and LGS indications incorporating a single Dravet trial that has already reported results and the two LGS trials. In addition to the pivotal trial, we will be submitting approximately 10 Phase I and Phase II trials including an agreed abuse liability package and the drug-drug interaction studies. The NDA will also include a very robust safety database that will include over 1,500 patients from both the expanded access and pivotal programs and including over 400 patients who had at least one year of continues exposure to Epidiolex. More recently we held a CMC pre-NDA meeting last month, as with the clinical pre-NDA we found the discussion constructive and we are clear on the proposed CMC content on our NDA submission. We will be including large scale commercial manufacturing facilities in our NDA, and believe we are on track to be ready for FDA pre-approval inspection. In the light of these pre-NDA meetings, we are confident that we would be submitting a robust package of data in line with FDA expectations to support approval. Investors will be aware that the IND for Epidiolex was opened in May of 2014, and hence this comprehensive NDA is being put together in a rapid timeframe. We expect the submission to be made at the end of the first half of 2017, which will represent just three years from IND to submission. Outside the U.S. we are also now progressing plans for European regulatory submission. As a reminder, all of our Phase III trials have included European sites. We expect to hold pre-submission meetings with the EMA in the first quarter of 2017, and our current plans are to submit a marketing authorization application in the second half of 2017. Investors will be aware that the primary endpoint in our studies of a median reduction in seizure frequency is the FDA preferred endpoint and this is also recognized by the European Authorities. We know that the EMA will also be interested in the 50% responder analysis, but we understand that the EMA had previously given scientific advice that the FDA preferred endpoint of percent reduction in seizure frequency is acceptable for purposes of the EMA in cases of orphan indication. The 50% responder analysis showed a 44% response rate and was highly significant in our LGS trial. And in the Dravet trial, the response rate was 43% and was very close to statistical significance. For all these reasons, we are therefore confident in a success of regulatory application for both indications in Europe. Beyond the first two indications of Dravet Syndrome and LGS, we continue to advance plans to expand the label. We’re currently recruiting a 200 plus patient Phase III trial for our third target indications of Epidiolex, Tuberous Sclerosis Complex. We have also just commenced the two part Phase III study in our fourth target indication, infantile spasms. The first part of the trial will recruit 10 patients and is expected to complete in the first half of 2017. Subject to review by an independent panel, the second pivotal phase of the trial will then commence. Beyond Epidiolex, our part time programs continue to progress. We’ve commenced a Phase I trial of intervenes CBD, a formulation intended for use in Neonatal Hypoxic-Ischemic Encephalopathy, an orphan indication for which we also have fast track designation. With respect to CBDV, we expect that results in the Phase II study with an adult partial onset epilepsy in mid-2017. We’re also progressing plans to study CBDV in clinical trials within the field of autism spectrum disorder. We have in fact just received orphan designation from the FDA for CBDV in the treatment of Rett Syndrome, a condition which affects about one in 10,000 females -- about 10,000 females in the U.S. Rett Syndrome is caused by genetic mutation in the MECP2 gene found on the x-chromosome. Most patients are non-verbal and have limited use of their hands. Seizure control is also a common problem in Rett Syndrome. This is one of the conditions for which we will be advancing CBDV into phase II clinical trials in 2017. We have completed Phase 2 placebo controlled studies in glioma and in cerebral palsy specificity. Topline data is expected from both studies in the first quarter of 2017. Thank you. And let me now ask Adam George to provide a financial review.

[Adam George:]

Thank you Steve. I’d just like to provide some high level comments on GW’s results for the financial year ending September 30, 2016. Additional detail can be found in the 20-F filing that will be made later today. We prepare our results in accordance with international finance reporting standards in British pounds sterling. But for convenience purposes on this call, I propose to refer to the U.S. dollar equivalent of certain key numbers. Today’s results for our fourth quarterly earnings release for the three months and the year ended September 30, 2016. As usual, I’ll start with the most important line of our income statement which is our expenditure on research and development. Q4 GW funded research and development expenditure increased to $31.4 million taking R&D spend for the year to $128.9 million, driven primarily by the Dravet and Lennox-Gastaut trials program, plus the cost of executing our TSC study and second costs for the infantile spasm study. Continuing supply of Epidiolex to around 1,000 patients in our long-term extension study and the various state studies and investigator INDs continues to be a driver for spend together with our Epidiolex lifestyle cum management activity and cost of the other in-progress clinical studies, including CBDV in partial seizures, glioma and NHIE. Moving to SG&A, Q4 SG&A expenses increased to $9.6 million taking 2016 spend to $25.7 million. This reflects steadily increasing investment in building our U.S. commercial operation and commercial planning projects. Consistent with previous quarters, further strengthening of the U.S. dollar against sterling has resulted in the recognition of a net foreign exchange gain in Q4 of $8.1 million, taking the cumulative FX gain for the year at $33 million. This is all resulted in a loss before tax for Q4 of $31.5 million and the loss before tax for the year of $111.3 million offset by a $29.1 million tax credit to give a 2016 loss after-tax of $82.2 million. Turning to cash flow, in 2016 we spent $109.2 million on operating activities and invested $11.2 million in capital projects associated with increasing growing and manufacturing capacity as we scale up the launch. Having successfully raised new equity proceeds of $273 million in July, we ended the year with a strong closing cash position equivalent to $483.4 million. Looking ahead, I expect total cash outflows for 2017 in the range of $130 million to $150 million. This includes $30 million of capital expenditure as we continue to expand our manufacturing capacity, plus operating spend of between $100 million and $120 million. We expect R&D spend to continue at current levels, although the spend on Phase III clinical trials is expected to reduce we will be investing about $30 million to $40 million into prelaunch inventory build all of which will be expensed by the R&D line in 2017. SG&A spend for 2017 is likely to increase by about 30%, mainly in the second half of the year once our NDA is filed and we’re on track towards approval. Thank you. I’ll now hand the call back to Justin.

[Justin Gover:]

Thank you, Adam. Given the focus today on AES, we will not cover other updates in detail on this call. Briefly with regard to the manufacturing scale up, we have scaled the key steps in the production process, and with positive feedback from the FDA after our recent CMC pre-NDA meeting we’re confident that we are in good shape with respect to the content of the NDA submission. In addition, in terms of launch preparation, we continue to build an impressive U.S. commercial team under the leadership of Julian Gangolli, GW’s President, North America. This team will continue to evolve in 2017 and is working on all the plans necessary to deliver a first class U.S. launch. You may have noticed in our recent press release that we have introduced a new U.S. identity Greenwich Biosciences, which will be the face of our company to patients and their families, clinicians and the general public in the United States. We are also now starting to put together a European commercial team in anticipation of future European approval and commercialization by GW. This European commercial effort is being led by our COO, Chris Tovey, who has a wealth of epilepsy commercial experience. With regard to scheduling, we continued to have constructive engagement with the DEA and remain confident that our data will support and Schedule IV designation for Epidiolex following FDA approval. Finally, with respect to intellectual property, we continued to prosecute a portfolio of patents relating to the use of CBD and epilepsy, with a view to seeking to extend the exclusivity period beyond the 7.5 years orphan and pediatric exclusivity. This portfolio includes 12 distinct patent families which are either granted or filed. The latest expiry date of these families runs to July 2036. This portfolio includes patent families with claims directed to the use of CBD in the treatment of epilepsy seizure subtypes, particular childhood epilepsy syndrome and formulations. We have two patents granted by the U.S. PTO and numerous patent applications being [technical difficulty] at the U.S. PTO. We anticipate filing additional patent applications claiming the use of Epidiolex in 2017 as new data is generated and expect more of our pending patent applications to be granted by U.S. PTO during 2017. Should the NDA for Epidiolex be approved we do expect a number of these granted patents to be listed in the Orange Book. In addition other patent families provide protection for epilepsy related inventions such as extraction techniques, CBD extras and highly purified CBD. We’re also making progress on plans for formulation lifecycle management. In closing, we feel that all the data disclosed at this year’s AES only reinforces our confidence in the prospects for Epidiolex feedback from the medical community remains very enthusiastic and after a transformative 2016, we enter 2017 with every expectation of its successful pathway to FDA approval and commercialization. Thank you for your time today and for your interest in GW and I would now like to open the call for few questions.

[Operator:]

Thank you. At this time, we’ll be conducting a question-and-answer session. [Operator Instructions]. Our first question is from the line of Paul Matteis with Leerink. Please go ahead with your questions.

[Paul Matteis:]

Hi guys. Thanks very much and thanks for the update with some new information. If you wouldn’t mind, if I can ask a follow-up questions to Stephen about the analysis you did in patients on Epidiolex, clobazam, stiripentol, Steve could you just expand upon the sample size of patients in both of those cohorts and the subgroup analysis you did?

[Stephen Wright:]

There were approximately 30 patients. I need to get precise numbers from our database. But it’s a reasonably sizeable proportion of the patients in 1332 part B in the pivotal Dravet syndrome study, we’re taking stiripentol with clobazam plus or minus other anti-epileptic drugs, it’s a reasonably good sample size.

[Paul Matteis:]

Okay, thanks. And on the poster yesterday for Dravet syndrome, there is a specific comment that says -- that pooling analyses in patients off clobazam will be conducted or presented at a future date. Could you just expand upon when you think the appropriate forum is to share some of those data and when they could be ready?

[Stephen Wright:]

Clearly, there are multiple subgrouping analyses for parts of our NDA submission. So I think that our disclosure of any parts of the NDA submission will be a decision we will have to take as that submission gets put together. I think it will be misleading of me to guide you to a specific meeting or a specific event where those data get presented. But I’m sure we’ll be discussing them during 2017, along with the key information around the submission of our NDA.

[Paul Matteis:]

Okay, fair enough. And then maybe just one quick one for Justin. Justin, I was just curious, if you can talk a little bit about your pre-NDA CMC meeting and what other CMC steps, if any, are limiting to an NDA filing next year. Thanks a lot guys.

[Justin Gover:]

Thanks Paul. So I mean the CMC meeting took place last month, whole series of questions around the various processes and the extent of data required to be included in the NDA, very satisfactory answers provided to all of those questions. And as we’ve previously talked about, the scope of our activities includes additional scale at all the key steps in the process. So, we are expecting all the scaled steps, mainly the enhanced growing extraction, crystallization, and finished product manufacture to be -- the data from those different facilities to be included in the NDA submission and available, should the FDA want to inspect them as well. So, everything is on track, it’s still -- there is a lot of work going on, on that as you can imagine, but it seems to be going very well.

[Paul Matteis:]

Okay. Great. And you [technical difficulty]. Please go ahead.

[Justin Gover:]

I think that’s for you Paul.

[Paul Matteis:]

Okay. And you’re comments on DEA scheduling that’s based on human abuse liability data that you’ve generated with Epidiolex that you, I would assume, discussed with the DEA?

[Justin Gover:]

Yes. I should say it is our view -- I don’t think we don’t have a formal view from external agency on our data yet. But, I think we’ve long held this view and obviously as we generate the data needed to support that view, I think we’ve become more comfortable that it’s a very realistic ambition.

[Paul Matteis:]

Okay. Would it be a meaningful difference in kind of your supply chain? And how you can launch in market, this drug, if you were scheduled a pre-incentive schedule for it?

[Julian Gangolli:]

Paul, this is Julian. Not really, where it becomes more restrictive is obviously in Scheduled II as people are aware. I think we’re optimistic having viewed the human abuse liability data and the interaction that we had with the DEA a couple of weeks ago in DC that there is a position that the scheduling is going to be in the zone that we are telegraphing to you all.

[Paul Matteis:]

Okay. Great. Thank you guys.

[Operator:]

Our next question comes from Salveen Richter with Goldman Sachs. Please go ahead with your question.

[Heath Terry:]

Hi. Thanks for taking my questions and congrats on all the progress this year. It’s actually Terry on the line for Salveen. I’ve a couple of questions, first up with regards to manufacturing. How much Epidiolex production capacity do you anticipate to have by the time of launch? And just second question is regarding the AES data presented on liver enzyme elevation, what was the cause in the few patients that were not on valproate and do you see this as Epidiolex related issues? And I just have one follow up question. Thank you.

[Adam George:]

This is Adam George. I’ll answer the capacity question. We’re planning this year, sorry in 2017, we’re planning on growing enough material to produce product for up to 50,000 children per annum. And then we’re aiming to have sufficient launch inventory for about 40,000 children. So, we’re planning ahead to have capacity for up to 60,000 children by the end of 2018.

[Julian Gangolli:]

And then if I can address the liver function. The working hypothesis we’ve got is that these are patients who are receiving multiple anti-epileptic drugs, and most if not all of those drugs place a kind of burden on the liver. They’re all metabolized through the liver, for some of those drugs particularly valproate that burden is higher than it is for others. So one or other maybe more likely to provoke this short-term response from the liver of transient raised liver enzymes. It is a phenomena that one sees relatively commonly for anti-epileptic drugs. So pointing to a kind of specific drug-drug interaction that might be responsible for it I think we probably be wrong, but it is clear that it does occur far more common in patients taking in concomitant valproate. It’s also clear that in the great majority of patients who remain on drug, which of course is a perfectly reasonable response from the investigator when they observe these moderate elevations or mild elevations, in the great majority of those patients, the enzymes resolve completely while they remain on drug and I think that’s one of the reasons why our investigators and clinicians that we speak to are really relatively relaxed about this trend in elevation that we’re seeing.

[Heath Terry:]

Thank you. That was very helpful and just one final question, just regarding your CBDV program for Rett Syndrome, what is the market opportunity here and what are other autism spectrum disorders that could potentially be amendable to treatment?

[Justin Gover:]

Terry, its Justin here. I think the -- we’re obviously interested in this general therapeutic area of behavior and autism spectrum, Rett Syndrome is an example of one of the target we’ve identified. This is an orphan indication within that sphere, specifically they are around 10,000 or so females in the United States with Rett Syndrome and it’s said -- we’re not -- our interest within this area for CBDV is not limited to Rett Syndrome. So I think you should assume that for CBDV we will take example such as Rett and we have some others in mind forward through proof of concept and with a view to selecting some to move forward through to approval. It’s a plan we’ve been continuing to work on and will evolve in 2017 within this are. And of course just to emphasize as Steven mentioned earlier that the CBDV epilepsy program is also moving forward. So we have data from the adult partial onset seizure study is expected next year. So, there is multiple tracks for CBDV that I think you’ll be seeing over the next couple of years.

[Heath Terry:]

Thanks guys. Congrats on all the progress.

[Justin Gover:]

Thank you.

[Operator:]

Your next question is from the line of Tazeen Ahmad with Bank of America Merrill Lynch. Please go ahead with your questions.

[Tazeen Ahmad:]

Hi good morning and thanks for taking my questions. Can you just remind us where you are in enrolling the second Dravet study and when you think you would be able to read that out? In conjunction with that question, can you just give us a better idea of how EMA versus FDA view your responder analysis employed. You gave us some comments about the importance of the primary endpoint, but in the events that EMA has additional questions about Dravet syndrome in particular and the responder analysis, when would you be in a position to provide the additional Dravet syndrome data from the second study? And then I have a couple of other follow-ups.

[Steven Wright:]

Tazeen, Steven here. So I’ll start with the first part of your question about the dose ranging study in Dravet -- of Epidiolex in Dravet syndrome which is still enrolling patients and we anticipate that it will continue to enroll patients for a little while. It’s very difficult to give specifics about enrolment rates because clearly they are future events and so the precise data which that we’ll finish enrolling is certainly not clear to us. We would anticipate that during 2017 most likely to be the case. We don’t believe that study will be available for the submission of the NDA and we have discussed that with FDA and that’s the very position we and they appear very comfortable with. If I move over to the 50% responder analysis, you’re quite right that historically the EMA have preferred a 50% responder analysis in mark contrast to the FDA, who really don’t find that to be a particularly compelling analysis of efficacy. The EMA certainly have shown willing in the past to look at the median or the seizure frequency reduction from baseline as an endpoint that does carry significant meaning, we’re aware that they can revise to others in the past and I think that in the light of all of the positive results that we’ve got, both primary and secondary across the two indications of Dravet and of Gastaut, I think the risks that the EMA will insist on a 50% responder analysis is at pretty small risk. It is clearly nonetheless a risk and we are happy to face that. We regard the Lennox-Gastaut indication of more interest in the European setting honestly and of course we do have the second Dravet syndrome to follow on later in 2017. So if necessary we could make separate submissions in Europe for Lennox-Gastaut and Dravet. So I think that probably summarizes the situation reasonably accurately.

[Tazeen Ahmad:]

Yeah. Thank you for that. And then can you just talk about the importance of the global impression of change outcome and how that might be perhaps even more important than responder analysis?

[Steven Wright:]

I can, yes, there is a general acceptance and it’s been published and discussed fairly widely that the global impression of change is that the method by which you deem that a change in the primary endpoint is clinically relevant or not. So for example if you have a big reduction in seizure frequency and the caregivers are saying my child is no different, it’s very difficult to argue that that reduction in seizure frequency is making a difference to the life of that child. Whereas if you get a relatively modest even reduction in seizure frequency and the parent or the parent or the caregiver is saying my child is substantially improved, then it makes a very much stronger cases that change is clinically relevant. So there are even one or two prescribing information for anti-epileptic drug that specified the global impression of change as a part of the evident that it’s a clinically relevant improvement. So we believe it’s quite important, there is a long literature that I think supports that view and I think it’s one of the major components of the argument about clinical relevant.

[Tazeen Ahmad:]

Okay. Thank you. And then maybe one question for Adam, given the number of trials that you are currently running and do plans to run in calendar 2017. Directionally, I think speaking without being specific should we assume that your R&D expense will continued to increase into next year?

[Adam George:]

As per the guidance that I’ve given Tazeen, I’m expecting it to stay relatively flat as we go into 2017. During ’16 the Dravet and LGS programs has been a big driver for spend, but that spend will tail off, but will be replaced by increasing spend on inventory build which will go through our R&D line. So R&D spend should be relatively flat as we go into next year.

[Tazeen Ahmad:]

Okay. And then concurrently for SG&A, do you expect to be embarking on increased commercial preparation given that you’re planning on filing for approval next year?

[Adam George:]

We do and I guided to expect a 30% increase in 2017, mainly in the second half, post the NDA filing as we start to build the commercial team.

[Tazeen Ahmad:]

And would that SG&A primarily be hiring people or would there be other expenses attached to it?

[Adam George:]

There is all sorts of projects that go into the preparation for launch which are ongoing now to help us position the price of the product optimally. But yes there will be heads to add, systems and processes to put in place just to prepare for launch.

[Tazeen Ahmad:]

Okay. Thank you.

[Operator:]

Thank you. We’ve reached the end of time allotted for today’s question-and-answer session. I’ll turn the floor back to management for closing remarks.

[Justin Gover:]

Thank you. So it just remains to thank everyone for joining today, for those of you at AES we have a scientific exhibit this morning, innovation pavilion later this morning and dinner this evening at a hotel close to the Congress and for everyone else we’ll see you no doubt in the New Year at San Francisco. So thank you again for your attention today.

[Operator:]

Thank you. This concludes today’s teleconference. You may disconnect your lines at this time. Thank you for your participation.