All right. Well, good afternoon, everybody. I'm Ryan Feeney. I'm Vice President in the healthcare group here at Barclays. And today I have the pleasure of introducing Jack Khattar here, who is the CEO, Founder, President, and Director, excuse me, for Supernus Pharmaceuticals. Jack founded the company in 2005. Prior to that, from 1999 to 2005, Jack served in a variety of roles at Shire Laboratories, and he's done an amazing job of building this company. He had done a tremendous amount in terms of drug development and acquisitions. And with that being said, I wanna turn it over to Jack so he can walk you through where the company is today and what they're looking forward to.
Great. Thank you so much. And, thank you all for attending our presentation. Before I really get started, just to remind everyone, I'll be making forward-looking statements, so please check our SEC filings for all the risk factors associated with the business. So as far as background, for those of you, especially, who may not be familiar with the company, we have 30-plus years of experience in the CNS space, primarily in ADHD, with four products that we have developed over the years. The company, as Ryan said, you know, started in 2005 as a spin-out from Shire. And over the years, we've developed a whole set of products that are on the market today, close to 12 products that are either marketed by us or our partners.
Primarily for Supernus, we developed Oxtellar XR and Trokendi XR initially, and then Qelbree, which is a product I'll talk a little bit more about. We also have a portfolio of Parkinson's drugs that came through a couple of acquisitions that we did back in 2020 and 2021, and a very strong pipeline that I would like to spend some more time on, today. So as far as the history and our execution, you'll see that we launched our first two products, Trokendi XR and Oxtellar XR, in 2013, able to build the business over the years, with peak revenue of around $667 million in 2022. And then last year was the first year where we lost exclusivity on our flagship product, Trokendi XR. And despite losing about $170 million in revenue, we were able to still achieve $608 million in revenue last year.
And then this year, we guided somewhere between $580-$620, despite the continued erosion of Trokendi XR and also the loss of exclusivity on Oxtellar XR. So why are we able to manage that transition smoothly? A couple of reasons. First, we work over the years, try to diversify our portfolio. So the last year, we had actually two products. It was 2019. You'll see that Trokendi XR accounted for about 77% of our net sales. Moving forward, four years later, 2023, that went down to 16%, and you'll see the diversification across different products. Two important products on that pie chart on the right side, the yellow, which is Qelbree, about 24% of the business is in Qelbree, and then the one next to it, the purple, 21%, which is GOCOVRI. A couple of products that I'll highlight today.
They are our current growth products that helped us tremendously achieve, you know, the results that we did back in 2023. So starting with Qelbree, which is a novel non-stimulant in ADHD. There hasn't been a novel therapy in ADHD, especially in adults, as a non-stimulant for over two decades, actually. And the product was launched in May 2021 with a pediatric indication, then followed by the adult indication in 2022. In 2023, we had a very robust growth in the product with 91% in prescription growth, 2023 versus 2022, reaching about 617,000 prescriptions on an annual basis, while the market also grew by only 3% and reached about 93 million prescriptions a year. So as a market, it's a fairly big opportunity for us. Clearly, with our 617,000 prescriptions, we still have a long way for us as far as growth and potential penetration of that market.
On a net sales basis, we hit a high or a record last year of $140 million in net sales. Very pleased with that, which is about 129% growth versus 2022. And why is Qelbree doing really well? One of the key aspects of it, and you'll see that on this chart, where are the patients coming? You'll notice that 29% of the business is coming from new patients. So it's being used as a first-line treatment. Doctors are really looking at Qelbree as one of the first-line treatments and as a good alternative instead of starting with a stimulant. The other part of the business, about 71%, is coming from switch patients who are being switched over from their existing medications or having Qelbree added onto their existing medication.
The interesting thing on the right side of this slide, you'll see that 65% of these switches are actually coming from stimulants. So that truly speaks to the efficacy of Qelbree, speaks to the perception of Qelbree that it is really a non-stimulant that actually works. For physicians to start thinking of Qelbree as an alternative and a good alternative for Adderall XR or Vyvanse and so forth, that's really remarkable. It doesn't happen too often. Over the years, non-stimulants have always had the perception that they really don't work. They don't work, you know, fast, and they don't work as well as stimulants. So we're making major inroads in this category and establishing Qelbree as a non-stimulant that actually works and works fast.
Then the next part of the slide, you'll see that the remaining 35% of these switches are coming from the other products, which are non-stimulants, primarily from atomoxetine or Strattera. On that product, actually, there was a study that was published by two physicians who use Qelbree, use atomoxetine, and many other products. They published the study on their own, independent of Supernus, where they switched people from atomoxetine. They had patients on four-week treatment of atomoxetine and then switched them to Qelbree for four weeks. They measured the ADHD rating scale and the improvement in that scale at baseline after four weeks' treatment with atomoxetine and then after four weeks' treatment with Qelbree. This was published back in July of last year, about 50 patients, a fairly good-sized study.
They were able to show significant improvement, which Qelbree that was able to deliver even versus atomoxetine, with 96% of patients preferring Qelbree over atomoxetine and also 86% reporting response within two weeks. And this is something, you know, we've talked about for many times, even since the time when we had our phase III data published showing the product working as early as one week for pediatric and as early as two weeks for adult, which is a major differentiating factor of Qelbree versus other non-stimulants, primarily atomoxetine. And then finally, on the tolerability, interesting, 4% discontinued Qelbree because of AEs, which also is very consistent with the phase III data that we have generated on the product, versus about 36% on atomoxetine. So clearly, a lot of clear benefits of Qelbree even versus existing non-stimulants.
And that's why you see a lot of switches from atomoxetine to Qelbree. The next product is GOCOVRI in the Parkinson's space. This is actually the only product in Parkinson's that is indicated to treat the off episodes as well as dyskinesia. This is a typical problem in this space where physicians and patients are struggling with the conflict between treating the off episodes, raising the dose of levodopa/carbidopa, and then bumping up against dyskinesia. And the more and the higher they go with levodopa/carbidopa, the more dyskinesia occurs, and therefore, they have that dilemma. Well, with GOCOVRI, you can actually change that treatment paradigm. You can treat both off episodes and dyskinesia. You don't have to sacrifice on the levodopa dosage, daily dosage. You can keep that dose the same, and you can address the dyskinesia with GOCOVRI.
This is a product that we acquired back in November 2021 from Adamas Pharmaceuticals. We've done very well with the product, very happy with the performance for the last couple of years, delivered double-digit growth. On top of the 18% we delivered in 2022, we delivered another 15% growth in 2023, hitting $120 million in sales. You'll see on this data, the efficacy of the product. Notice the 27% decrease in dyskinesia while at the same time also decreasing the off time by 36% and increasing what we call the good on time. Good on time is basically you're not off. You have a good on time, and it is good because there is no dyskinesia associated with that on time. All right.
Moving on to the pipeline where I really would like to spend the remaining portion of this presentation, I'm gonna just highlight the first three projects. The first one, which is SPN-830 in Parkinson's disease, and then talk about the depression molecule and epilepsy molecule that we have. Very exciting two programs that are first-in-class mechanisms of actions in both therapeutic areas. So SPN-830, which is our novel apomorphine subcutaneous injection device. This is a pump that pumps apomorphine continuously, subcutaneously. The only treatment option at this point for people who are advanced in Parkinson's disease, they get to the point where they have to choose between deep brain stimulation or a surgically implanted infusion for levodopa and carbidopa. So with this product, we are potentially creating a new segment in the market, giving patients a whole new novel treatment option before they resort to these invasive procedures.
So very exciting, you know, product potentially for Parkinson's patients. The PDUFA date is pretty much pretty close, a few weeks from now, April 5th of this year. We project peak sales in the range of $200 million-$300 million. This is only U.S. because we don't have rights for the product outside the U.S.. This will show you, you know, where Supernus' presence with the pump on the spectrum of treatment over the Parkinson's disease progress from stage one all the way to stage five. You'll notice that, hopefully, with the launch of SPN-830, which we hope to launch in the second half of this year, we will have four products in the market with very strong presence, starting with GOCOVRI that I mentioned earlier, XADAGO, which is also an adjunctive therapy to levodopa/carbidopa.
We have the APOKYN apomorphine injection, and then SPN-830, which will be more towards the more moderate to severe and more advanced patients on that spectrum of disease. As far as the data on SPN-830, this has been already published a few years back, the phase III study showing on an average a 2.5 hours' improvement or reduction in the off time and corresponding increase in the on time. Fairly, very meaningful clinical improvement in the off time for these patients that are well advanced in their disease. Moving on to SPN-820, which is a first-in-class mechanism of action as a novel treatment in depression. This is a molecule that doesn't work through receptor binding or on the cell, the surface of the cell. This is an intracellular mechanism through the mTORC1, mediating synaptic function, improving synaptic function, and therefore addressing some of the core symptoms of depression.
In early studies, the product had shown an average effect size of 0.6, fairly meaningful from a clinical perspective, only with one single oral dose of 2,400 milligrams, and it was over an initial 72-hour period. This is using the HAM-D6 as a measurement. Also, what's important is that it did show rapid onset of effect. It was measured actually at four hours into the treatment, again, using the HAM-D6. And the effect was sustained over a 72-hour. So not only do we have the rapid onset, we have also sustainable effect, through 72 hours, which was actually the longest measured. We didn't measure more than 72, but that was the last point that it was measured at. The product was very well tolerated in the clinical trials. There is no dissociations or hallucinations.
Again, because of the fact that the product works intracellularly, it doesn't work on any specific receptors that could introduce any abuse liability or any effects such as hallucinations and so forth. It is unlikely to be a controlled substance as well. And we have a phase II-B study that is ongoing right now with about 120 patients enrolled. And we're looking at top-line results in the first half of 2025. So a little bit more flavor on the earlier studies. I mentioned, you know, the rapid and sustained effect. And the reason behind this rapid and sustained effect is the fact that the product actually, or the drug, gets very early and rapid absorption into the brain. And we've been able to measure all that and the brain exposure and through the CSF confirmed by CSF fluid and the drug levels in the CSF fluid.
In addition to that, we have also correlated not just the drug levels but also the biomarker activity in the CSF fluid, which is a surrogate for the mTORC1 activation, which is the mechanism of this drug. So all the data really points in the right direction, the same direction that the drug is absorbed, is absorbed quickly, gets into the CSF fluid, is activating the mTORC1 mechanism as we always thought it would through the biomarker evidence, and it's giving us the rapid onset of effect that I shared with you earlier. In addition to that, the CSF fluids and levels that we have seen actually correlate to the fully effective dose in the animal models, as well. Finally, even if you look at the EEG bands, you'll see statistical significance in these bands, which are typically associated with positive mood states like arousal, alertness, and so forth.
So very consistent early data through the preclinical studies as well as the early human data. So right now, as we speak, as I mentioned, we have the phase II-B study going on in TRD, treatment-resistant depression. It's a double-blind, placebo-controlled study. We're studying 800 and 1,600 milligram, not the 2,400 milligram, because now this study is looking at daily administration rather than administration once every three days. And we got all this through the MAD and SAD studies. We were able to measure the CSF fluids at these doses, which correlated to the effective dose that we saw back in the early studies. Approximately about 268 subjects. As I mentioned, we have about 120 patients already enrolled, looking at a treatment, about five-week treatment, and again, reporting in the first half of 2025. The typical, you know, endpoints you would look at, primary efficacy with MADRS.
We're looking also at the HAM-D6 as a secondary and so forth. In parallel, we initiated recently, you know, around the end of 2023, another study on the same molecule. This is more of an open-label exploratory study for this molecule, looking at the 2,400 single-dose oral dose and looking at it as intermittent dosing. So that's the three-day dosing. And that's related to the 72 hours that I mentioned earlier that potentially you really don't need to hit the mTORC1 activation and do it every single day because of the sustained effect of the drug. We could potentially get by just a dosing, which is every three days. So that's what the study is exploring. The other thing the study is exploring here is the efficacy in MDD, not TRD but MDD.
So again, looking at the other aspect in depression, looking at the rapid onset. We're trying to measure the onset earlier than four hours to see if we can get an effect even earlier than four hours. Okay. Moving on to the last program I would spend a few minutes on, which is 817. Very exciting program looking potentially at getting some data as early as May of this year that we'll be able to share with folks. This is a very unique acetylcholinesterase inhibitor, very selective. It has very low activity on the butyrylcholinesterase. And in all the preclinical models that this campaign has been screened, it looks like it has a broad and potent antiseizure effect. And I'll share with you some of the data, you know, in some of the models that we have seen. So, so far, the product seems to be very potent.
It also has the potential of procognitive effects. Acetylcholinesterase inhibitors as a class of drugs are known to potentially treat Alzheimer's. Some of them are indicated to treat Alzheimer's. So procognitive effects are very potential here for us to be even another differentiating factor versus a fairly crowded market and a lot of products being developed for the treatment of epilepsy. Also, there is a potential for some neuroprotective and anti-inflammatory effects. Right now, as we speak, we have an ongoing phase II-A study in focal seizures in adults. As I mentioned, we're targeting interim data, look, sometime in May about 50% of these patients and presenting the full data in the second half of 2024. Ideally, it would be maybe around August, also the earnings call.
We'll, we'll see where we land as far as finishing the enrollment, but that's the timeline we're looking at at this point. In parallel, we'll be entering phase II-B also in the second half of this year. This is some of the preclinical data that I referred to. This is a very commonly used 6-hertz model, animal model for seizures. This is a screening that actually happened way back at the NIH looking at huperzine A, which is the drug in SPN-817, comparing it to very well-known anticonvulsants that are on the market approved for seizure activity. It tests different states of seizure-inducing levels, milliamps. You're looking at the 22, 32, and 44. Highlighted here is the comparison between SPN-817 and levetiracetam, for example, which is a very well-known established drug, hit about $1 billion in revenue at peak.
Multiple-fold effect of SPN-817 again speaks for the potency of the drug. Also interesting, you look at the high side of it, the highest seizure-inducing state, which is the 44 milliamps. Actually, SPN-817 is the only drug in this test that showed antiseizure effect. So we believe this drug continues to show, you know, the high potency starting with some of the animal models. But I'll share with you also data from three patients with a very small proof of concept study that was done early, looking at the data. You look at the second bullet, mean reduction in the 28-day seizure from baseline, 71% reduction in seizures through month 14. Going through month 40, went up to 89.8%. And we still have one patient who is seizure-free and has been seizure-free for more than three and a half years, on this drug.
On the right side of the slide, you'll see the bar chart, which shows the data on that patient. Their baseline was more than 150 seizures a month, going down all the way by month three to a very low level, getting to seizure-free after two years, month 24, and still seizure-free till today. So very promising compound, again, very potent, looks like very consistent with the preclinical models that I shared with you earlier. The phase II-A study that is ongoing right now is an open-label. We have about eight sites, approximately 35 subjects. And the key primary endpoint is the safety tolerability as an adjunctive therapy. And clearly, we're going to look at the % change, you know, in seizure frequency, improvement overall. We're also exploring changes in some of the inflammatory biomarkers in the plasma, the cognitive profile.
As I mentioned earlier, this molecule could have a procognitive effect and also looking at some of the sleep patterns as well. The interim data that we had shared back in the R&D Day, last year in October, and that's the data we had at that time, we were able to share some data from seven patients who had completed the titration at that time, you know, through October, during the R&D Day. And they were titrated from 1 milligram b.i.d. all the way up to 4 milligram b.i.d. You'll notice the tolerability, you know, fairly common things that you would notice in a class of drug like this, which is the diarrhea, the nausea, and so forth. But six patients completed the titration, and we had data at that time from these 6 patients. Two of them achieved 63.5%, as a mean reduction in seizures.
Other two patients achieved 100% seizure reduction, and they achieved the seizure reduction, one of them after only eight weeks in titration and the other one after nine weeks. So they didn't even have to finish the complete maintenance period, but they achieved the seizure freedom after the titration and 1 week into the maintenance. And finally, again, at that time, back in October of last year, we had one patient who had completed the whole study and had an average, you know, reduction of 68.3%. So we continue to see very strong seizure reduction data, you know, on this molecule. And hopefully, that will remain throughout the rest of the study.
So finally, to sum it up, as far as our story, I mean, we are looking at a very rich calendar of extremely valuable catalysts behind our story in the pipeline that is really emerging, very, very soon over the next 12-18 months. The first one clearly is the interim data that I talked about in the SPN-817 phase II-A study in May 2024, top-line data for all the patients, again, second half of this year, hopefully around the August timeframe. We also will be initiating the phase II-B study on SPN-817, the epilepsy molecule in the second half as well of this year. Potential launch of SPN-830, the infusion pump that I talked about. If we do get the approval in April, we're looking at potentially launching it in the second half, hopefully as early as possible in the second half.
but we'll, we'll update you in May on that as far as the exact timing of the launch. And then finally, potential getting of top-line data on SPN-820, the phase II-B data and TRD in the first half of 2025. As far as in the last minute, I'll talk very quickly the financial guidance. I did talk about the revenue. Despite the transition we're going through and losing exclusivity, we continue to be profitable on a non-GAAP basis. We guided somewhere between $80 million and $110 million. So all in all, we continue to be positioned for very strong growth in the company behind Qelbree, GOCOVRI, but as importantly, the innovative R&D pipeline that we have that I shared with you with SPN-830, SPN-820, and SPN-817. And some of these molecules are truly first-in-class mechanisms of action in the therapeutic area.
Thank you so much for your attention and for your time.