Ladies and gentlemen, Welcome to Basilea Pharmaceutica's Full Year Results 2019 conference call and live webcast. I'm Alice, the conference call operator. I would like to remind you that all participants will be in listen only mode, and the conference is being recorded. The presentation will be followed by a Q&A session. You can register for questions at any time by pressing star and one on your telephone. For operator assistance, please press star and zero. The conference must not be recorded for publication or broadcast. At this time, it's my pleasure to hand over to David Veitch, Chief Executive Officer. Please go ahead, sir.
Thank you. Hello, this is David Veitch, CEO of Basilea. I would like to welcome you all to our conference call and webcast reviewing our financial results and key achievements for 2019, discussing our upcoming milestones and financial guidance for 2020. I would like also to mention that this call contains forward-looking statements. Joining me on our call today are Adesh Kaul, our Chief Financial Officer, and Dr. Marc Engelhardt, our Chief Medical Officer. This morning we issued a press release and our financial report on the results for the financial year 2019. These documents are available on our website at basilea.com. For those on the call who are less familiar with Basilea, we focus on the research, development, and commercialization of innovative medicines that address the medical challenges in the therapeutic areas of oncology and infectious diseases.
We have a proven track record of progressing brands from research through clinical development to the market. We have successfully brought two anti-infective brands to the market, our antifungal Cresemba, and ZEVTERA, our broad-spectrum MRSA active antibiotic. We have oncology assets in clinical development, and we have a number of oncology and anti-infective assets in pre-clinical development. I would first like to provide a brief summary of our key achievements in 2019. We had a strong year in 2019. We achieved important financial goals, and we continued to make great progress launching our two marketed brands, Cresemba and ZEVTERA, around the world. During 2019, our commercial partners doubled the number of Cresemba-launched countries. Cresemba is now marketed in more than 40 countries worldwide. Our total revenue increased to CHF 134.4 million. This includes a significant revenue increase from Cresemba and ZEVTERA by 39% year-on-year to CHF 114.3 million.
This largely reflects the in-market sales growth we're seeing around the world. Due to the increase in revenues and management of our operating expenses, we improved our operating result for 2019 by 29% compared to 2018. As a result of increasing cash flow from our revenues and the management of our cost base, we continued to reduce our net cash consumption year-over-year, resulting in a solid year-end cash position of CHF 161 million. This provides us with the necessary flexibility to continue moving forward to our next potential value inflection milestones in 2020 and beyond. We also made significant progress in our clinical stage assets. For our FGFR kinase inhibitor, derazantinib, we initiated a phase I-II study in patients with urothelial cancer, both as a monotherapy and in combination with immunotherapy. This study is called FIDES-02.
Early in 2019, we reported positive interim results from the FIDES-01 study, a potentially registrational phase II study with derazantinib in patients with iCCA or intrahepatic cholangiocarcinoma, and whose tumors harbor FGFR2 gene fusions. We subsequently expanded FIDES-01 into iCCA patients with other FGFR2 genetic aberrations, with the aim of further strengthening the differentiation of the compound. For our tumor checkpoint controller, lisavanbulin, based on data readouts from two early-stage studies with the IV and oral formulation in patients with glioblastoma, we decided to move into a targeted biomarker-driven phase II study with the oral formulation of lisavanbulin in patients with recurrent glioblastoma. We reported positive top-line results from our antibiotic, ceftobiprole, from a phase III study in patients with complicated bacterial skin infections or ABSSSI.
This is the first of two phase III studies required for potentially gaining regulatory approval in the U.S., which is commercially the most important market for ceftobiprole. Adesh is now going to give an update on our commercial progress and present more detailed financial highlights for the financial year 2019, as well as provide our financial guidance for 2020. Then Marc will provide you with more detailed information on the progress of our most advanced development programs. I'll now hand over to Adesh.
Thank you, David. In 2019, together with our partners, we continued to make significant progress in the commercialization of our two hospital anti-infective brands, Cresemba and ZEVTERA. The most current in-market sales numbers available for Cresemba show that in the 12-month period ending September 2019, the global in-market sales of Cresemba grew by 32% year-on-year to approximately $190 million. In the U.S., our partner, Astellas, reported Cresemba sales for January to December 2019 of $146 million, which is a 28% growth year-on-year.
The sales performance is not only driven by a continued strong sales uptake in the U.S., but also by strong sales in markets outside of the U.S. The Cresemba sales in Europe were particularly positive, which was reflected by two sales milestones in the total amount of $12 million being triggered in 2019. The strong growth of Cresemba sales is expected to continue over the coming years, as we expect our partners to launch Cresemba in more and more countries. By year-end 2021, we anticipate that Cresemba will have been launched in about 60 countries. Partnerships continue to play an important role in the execution of our global commercialization strategy and provide a strong basis for future revenue growth. Our partners for Cresemba include Pfizer for most of Europe, China, and Asia-Pacific, and Astellas for the U.S.
We have strong regional partners for ZEVTERA and cresemba in other territories. All in all, our partnerships for Cresemba and ZEVTERA cover more than 100 countries worldwide. In 2019, new important Cresemba launch countries included Canada, for instance, and Singapore as the first country in the Asia-Pacific region. Basilea participates in the commercial success of Cresemba and ZEVTERA through royalties or a transfer price structure. We already received around $250 million in upfront and milestone payments, and could realize up to an additional $1.1 billion in potential future regulatory and sales milestone payments from our partnerships. Moving on to financials. I will highlight some of the key financial figures that were published in today's press release and in more detail in the full year report. I'd like to mention that all the figures that I will refer to are in Swiss francs.
We are very pleased with our financial performance in 2019. We did beat our guidance both on revenues and operating results. Deferred and non-deferred revenue contributions from our two marketed brands, Cresemba and ZEVTERA, together increased by 39% to CHF 114.3 million in 2019. Most importantly, non-deferred revenue from Cresemba and ZEVTERA increased by 36% to CHF 68.8 million, reflecting the strong in-market sales performance reported by our partners. Other revenue, mainly consisting of R&D reimbursements from BARDA, decreased CHF 6.7 million in line with the lower expenses for the ceftobiprole development program, as the first of two phase III studies was successfully completed in 2019. We completed the Toctino-related deferred revenue recognition in 2018, which is why there is no corresponding revenue contribution reported in 2019.
In spite of the approximately CHF 30 million reduction from lower Toctino-related revenue and BARDA reimbursements, total revenue increased from CHF 132.6 million to CHF 134.4 million in 2019. For 2020, we anticipate that the non-deferred revenue contributions from Cresemba and ZEVTERA will continue to grow at a healthy double-digit rate to CHF 77 million-CHF 87 million in line with the anticipated strong commercialization progress of our partners, especially related to Cresemba. Deferred revenue contributions from Cresemba and ZEVTERA are expected to decrease to CHF 33 million as we complete, in the course of 2020, the deferred revenue recognition of the Pfizer upfront payment and the Astellas upfront development and regulatory milestone payments received in previous years. In other words, our Cresemba and ZEVTERA-related revenue mix is moving increasingly to its non-deferred revenues, which more directly correlate with the in-market sales and provide near-term cash flows.
Non-deferred revenues are expected to reach between 70%-72% of total Cresemba and ZEVTERA-related revenues in 2020. For the breakdown of the deferred revenue in 2020, about CHF 21 million are expected to be booked in product revenue and about CHF 12 million in contract revenue. We continue to carefully manage our expenses. This is reflected in our flat operating expenses. That means R&D and SG&A expenses 2019 versus 2018. We expect to keep our R&D and SG&A expenses at a stable level also in 2020. Cost of products sold are influenced by a number of factors. Generally, they correlate with the volume of product that we deliver to our partners. As we are still in the launch phase, and at the same time transitioning stepwise responsibility for supply to Pfizer for their territory, there are some one-off effects that impact our cost of products sold.
Once the transfer is completed in 2020 or 2021, our Cresemba revenue mix is expected to move more towards royalties and milestones. This and increasing economies of scale will result in increasing gross profit margins forward-looking. Summarizing our guidance for 2020, we anticipate a continued strong double-digit growth of the non-deferred revenue contributions of Cresemba and ZEVTERA to CHF 77 million to CHF 87 million as a result of the expected strong in-market sales growth. The total deferred and non-deferred Cresemba, and ZEVTERA-related revenues are expected at CHF 110 million to CHF 120 million. As we continue to carefully manage our expenses, we expect to further reduce our net cash consumption and to report a strong cash position of CHF 100 million to CHF 110 million at the end of 2020. I will now hand over to Marc for the clinical development update.
Thank you, Adesh. Let me continue further with our antibiotic, ceftobiprole. In Europe and several markets outside of Europe, ceftobiprole is approved for the treatment of community and hospital-acquired pneumonia. It is marketed in most countries under the brand name ZEVTERA. One of our key priorities for ceftobiprole is to gain access to the U.S. market, which is by far the most important country for the commercialization of branded hospital antibiotics, and is estimated to account for up to 90% for anti-MRSA treatments such as [oritavancin and ceftaroline]. Based on special protocol assessment agreements with the U.S. FDA, two successful cross-supportive phase III studies are necessary for registration in the U.S. Our phase III program for ceftobiprole includes one study in Acute Bacterial Skin and Skin Structure Infections, and one study in Staphylococcus aureus bacteremia or bloodstream infections. The program is funded up to approximately 70% by BARDA.
This allows us to advance the development of ceftobiprole for the U.S. market in a cost-effective way. In 2019, we reported positive top-line results from the first of the two studies, the so-called TARGET study, a phase III study in patients with skin infections. The second phase III study in Staphylococcus aureus bacteremia is called ERADICATE. It is well on track and is expected to report top-line results as planned in the second half of 2021. If the bacteremia study is also positive, Basilea plans to submit a new drug application to the U.S. FDA. As ceftobiprole is designated a Qualified Infectious Disease Product by the FDA for these indications, if approved, ceftobiprole will be eligible to receive 10 years of market exclusivity in the U.S. from the date of approval. TARGET was a randomized, double-blind phase III non-inferiority study and enrolled 679 patients.
Patients received either ceftobiprole given intravenously three times daily or the comparator regimen of twice-daily intravenous vancomycin plus aztreonam. In summary, ceftobiprole was non-inferior to vancomycin plus aztreonam in this study, and the key endpoints for the FDA and Europe were both met. Ceftobiprole met the pre-specified primary endpoint of early clinical response at 48-72 hours after start of study drug administration in the intent-to-treat population, which is a key endpoint according to the FDA guidance for the U.S. Ceftobiprole also met the pre-specified secondary endpoints of investigator-assessed clinical success at the test of cure visit 15-22 days after randomization. This is the key endpoint for the EMA in Europe. Moving on to oncology. Our lead oncology drug candidate is derazantinib, which we in-licensed from ArQule, which is now a wholly owned subsidiary of Merck.
Derazantinib is a targeted, orally available small molecule inhibitor of the fibroblast growth factor receptor, or FGFR, family of kinases. FGFR genetic aberrations, for example, gene fusions, mutations, or amplifications, have been identified as potentially important therapeutic targets for various cancers, including intrahepatic cholangiocarcinoma, or iCCA, urothelial, gastric, breast, and lung cancers. Derazantinib also inhibits the colony-stimulating factor one receptor, or CSF1R, and the vascular endothelial growth factor receptor two, or VEGFR2 kinases. CSF1R is an important target in the modulation of the tumor immune microenvironment. The inhibition of CSF1R by derazantinib seems to be a unique feature for derazantinib compared to other FGFR inhibitors. VEGFR2 is known as a therapeutic target in the anti-angiogenic treatment in multiple cancers, including gastric cancer. Our development strategy focuses on achieving differentiation over other FGFR kinase inhibitors by leveraging the unique properties of derazantinib.
Key differentiation factors for derazantinib include its unique kinase inhibition profile and its clinical safety profile. Basilea's clinical development program for derazantinib currently comprises three studies. Two of them, FIDES-01 and FIDES-02, are ongoing in iCCA and in urothelial cancer. We are planning to start the third study, FIDES-03, in gastric cancer in the third quarter of 2020. Crystal structures indicate that the improved CSF1R inhibition activity of derazantinib versus other FGFR kinase inhibitors can be explained by a better fit of derazantinib into the active site of CSF1R. Other FGFR kinase inhibitors, such as the approved FGFR kinase inhibitor, erdafitinib, have different chemical structures and may not fit as well into the active site. The CSF1R inhibition may be important in the treatment of urothelial cancer, but may also have a broader utility to support combination studies with immunotherapy in other cancer types.
Preclinical data has shown that tumor macrophage modulations through CSF1R blockade may render tumors more responsive to T-cell checkpoint immunotherapy, including approaches targeting PD-L1 and PD-1. CSF1R kinase inhibition may thereby improve the susceptibility of tumors to immunotherapy. Experiments with mouse bone marrow-derived macrophages support that derazantinib modulates CSF1R kinase activity at clinically achievable concentrations in vivo. Basilea has entered into a clinical supply agreement with Roche for atezolizumab, or TECENTRIQ, a PD-L1 checkpoint inhibitor, to explore this immunotherapy combination in patients with urothelial cancer and gastric cancer. In urothelial cancer, patients' FGFR genomic abnormalities frequently show low PD-L1 expression, which has been associated with reduced response to immunotherapy. Therefore, derazantinib as a single agent and combined with PD-L1 inhibitors may address several oncogenic mechanisms and provide a new treatment paradigm.
Another pillar in our differentiation strategy of derazantinib to other FGFR kinase inhibitors is the safety profile, where derazantinib shows some of the FGFR kinase inhibitor class effects, but also clear differences with a lower occurrence of retinal events, nail toxicity, hand-foot syndrome, and stomatitis. Such differences are relevant for patients, as these side effects may compromise the patient's quality of life and have been shown to result in treatment discontinuations. In January 2019, we reported encouraging interim results from the registration phase II study called FIDES-01 in the second-line treatment of FGFR2 fusion-positive iCCA. Top-line results for this cohort are expected in the second half of 2020. In addition, we have expanded the FIDES-01 study in June 2019 with a new cohort of iCCA patients with FGFR2 gene mutations or amplifications in their tumor.
Through this new cohort, we intend to further define the full therapeutic potential of derazantinib in patients with iCCA. Interim data from the second cohort are also expected in the second half of 2020. In the third quarter of 2019, we have started FIDES-01, a phase II study with derazantinib as monotherapy and in combination with Roche's PD-L1 checkpoint inhibitor, atezolizumab. This is a biomarker-driven multi-cohort clinical study in patients with advanced urothelial cancer expressing FGFR genetic aberrations, and first interim results are anticipated for the second half of 2020. In the third quarter of 2020, we are planning to start FIDES-03. This study will explore derazantinib in patients with gastric cancer and FGFR genetic aberrations and will also include a cohort in which derazantinib is combined with atezolizumab.
We've decided to investigate derazantinib in gastric cancer based on derazantinib's unique kinase inhibition profile, convincing preclinical in vivo data, and the high medical need in this indication. Moving to our tumor checkpoint controller, lisavanbulin, or BAL101553. We are focusing our clinical development activities with lisavanbulin on glioblastoma, the most common and aggressive form of primary malignant brain tumors, in an area of high unmet medical need with very few treatment options available. Lisavanbulin is a novel microtubule-targeting small molecule. It can be administered oral and IV, crosses the blood-brain barrier, and has shown potent activity in brain tumor models in monotherapy and combination therapy. In line with our approach to involve biomarkers early in clinical development, we have been evaluating a panel of biomarkers. One of those is end binding protein one, or EB1, which was previously identified in preclinical models as a response-predictive biomarker for glioblastoma.
In our phase I glioblastoma clinical study with daily oral dosing of lisavanbulin, we have observed a profound and exceptional objective response in a glioblastoma patient whose tumor tissue was EB1 positive. This patient continues on treatment with lisavanbulin for more than 20 months now and shows a more than 80% area reduction of the brain tumor. Non-responding patients did not show this pattern of strong EB1 expression, and based on additional biomarker work, we are therefore assessing the potential utility of EB1 in a biomarker-driven clinical study in glioblastoma and are planning to start this phase II study mid-2020. We also continue with our phase I study in newly diagnosed glioblastoma patients of lisavanbulin in combination with standard radiotherapy in the U.S., which is conducted by the Adult Brain Tumor Consortium. I will now turn over to David.
Thank you, Marc. In summary, we are on track with the execution of our strategy in terms of both significantly increasing cash flows from our marketed brands, Cresemba and ZEVTERA, and advancing our R&D portfolio towards the next milestones. In particular, the clinical milestones in 2020 and beyond are as follows. We will progress the phase III ceftobiprole study in Staphylococcus aureus bacteremia with the aim of having top-line results in the second half of 2021 in order to file in the important U.S. market at the end of 2021.
In the first half of 2020, we anticipate to complete enrollment into the phase II registrational study in derazantinib in iCCA patients, FIDES-01, and expect then top-line results for the cohort with FGFR2 fusions to be available in the second half of this year. In the second half of this year, we also expect interim data from the iCCA FIDES-01 cohort with other FGFR2 genetic aberrations. Finally, on derazantinib, we also expect the first interim data from the FIDES-02 study in urothelial cancer in the second half of 2020. We have already amended the clinical supply agreement with Roche for their PD-L1 checkpoint inhibitor, TECENTRIQ, and our plan to explore the combination with derazantinib as well as derazantinib as monotherapy in gastric cancer. This gastric cancer phase I/II FIDES-03 study is anticipated to start in Q3 this year.
For lisavanbulin, we are planning to start the biomarker-driven phase II study in glioblastoma mid-2020. Finally, we expect completion of patient enrollment into the ongoing phase I study with lisavanbulin in patients with newly diagnosed glioblastoma by mid-2020. We'll now open the line to any of your questions.
We will now begin the question and answer session. Anyone who wishes to ask a question may press star and one on their touch-tone telephone. You will hear a tone to confirm that you've entered a queue. If you wish to remove yourself from the question queue, you may press star and two. Participants are requested to use only handsets for asking a question. Only three questions might be asked in a row, then the line will be open to others. You can get back in line again for any follow-up. Anyone who has a question may press star and one at this time. First question comes from the line of Louise Chen from Cantor. Please go ahead.
Hi. Thanks for taking my questions here. My three questions are as follows. First question is, just curious if you could provide more color on your competitive advantage for derazantinib in urothelial cancer and gastric cancer. Second question is, what supports the use of derazantinib in gastric cancer? I know you talked a little bit about it on the call, if you could elaborate more, that would be helpful. The last question I have for you is on lisavanbulin. Can you provide more color on your biomarker and the type of patients that can be best treated with your drug? Thank you.
Okay. Thank you, Louise. David here. Actually, they're probably best, all those questions are best answered by probably Marc. Marc, why don't you kick off and then Adesh or I can jump in. In terms of the competitive advantage of derazantinib in urothelial and gastric, and then the data supporting gastric cancer and why gastric cancer, and then the EB1 lisavanbulin.
Louise, thanks a lot. For the urothelial cancer, the key differentiation is, as we've outlined also in our presentation today, the activity of derazantinib against colony-stimulating factor one receptor, which the kinase on macrophages and is involved in the modulation of the tumor immune microenvironment. We have done a comparative kinase screen versus all competitors in clinical development, and that's a unique feature of derazantinib, and we believe that this might contribute to potential improved efficacy when combined with an immune checkpoint inhibitor. I think that this is clearly the differentiation we have, and we have done some more work into this. We have looked at the crystal structures, which explain why derazantinib and, for example, not the erdafitinib fits into the binding pocket of CSF1R, and we've also done in vivo experiments now in actual macrophages to confirm that CSF1R inhibition really happens.
That's, I think, the key differentiation for urothelial. In addition, I think for any combination, it looks like that derazantinib is differentiated by the safety profile I mentioned during the call, that this is related to retinal toxicity, nail and foot syndrome, and stomatitis, so it may be easier to combine it. For gastric, I guess where we were coming from was a large screen in patient-derived xenograft models, where we looked across a large number of tumors and different tumor types to just see where we have the most consistent, best efficacy, and that was gastric. From a kind of biology perspective, we also have seen that in addition to the CSF1R inhibition, derazantinib inhibits vascular endothelial growth factor receptor 2. There are compounds approved for this, or is a compound approved for this target in gastric cancer.
This may contribute or may really be an underlying biologic rationale, but the primary kind of goal for gastric cancer came from convincing the data in a series of non-clinical models.
The lisavanbulin and the EB1, the type of patients.
For lisavanbulin, I think we would initially, the phase II study that we're planning to start mid of this year would be in a recurrent setting. This is where we've seen this exceptional responder in the study in the U.K. with daily oral lisavanbulin. We have done quite substantial non-clinical work and also looked into tissue banks and have a relatively good idea on how to do the biomarker selection. These data will be published later this year, I think it's premature to detail on them. We basically have prevalence estimates and also have, I think, identified how to do the patient selection in that trial.
In order to start the study in Q3 this year, which is what I think we said during the presentation. Does that answer your questions, Louise?
Yes. Thank you very much.
Okay. Thanks.
Next question comes from the line of Raghuram Selvaraju, H.C. Wainwright. Please go ahead.
Hello, this is Edward Marks standing in for Raghuram. I appreciate you guys taking the questions. I have two financial questions and one clinical question. Just a little clarification. I was wondering how you plan to get to the projected year-end 2020 cash flows or cash position of CHF 100 million-CHF 110 million if you had CHF 161 million at the end of 2019, and the operating loss is only projected to be CHF 20 million-CHF 30 million. I was wondering if there are some non-operating items contributing to this cash decrease.
Adesh, do you want to take that?
Sure. Thanks for the question. As you have seen probably in our guidance, we are looking at CHF 30 million basically deferred revenue contribution to the top line. These are non-cash items. This is non-cash revenue. Hence that's sort of the difference. There are also non-cash items with regard to expenses. In essence, really the big difference comes from deferred revenues, which are all non-cash relevant.
Okay.
Does this answer your question?
Yes, absolutely. Might profitability or maybe at least cash flow break-even be attainable by the end of next year?
Yeah, that's a good question. The way we would answer that is that obviously clearly our model in the past has been we take compounds to the end of phase II and then we partner the phase III like we're doing currently with ceftobiprole with BARDA and we did with Astellas with isavuconazole. The caveat I would say is that, assuming that we would partner one or both of the compounds in development in phase II, which were lisavanbulin and derazantinib. If we partner either of those compounds during the course of 2021, and we keep on maintaining our stable cost base that Adesh talked about, and our cash generation revenues keep going in the direction they're going in, then we could foresee the possibility of us breaking even in 2021.
Excellent. That's good to know. Finally, on the clinical side, just when you outlined a lot of your timelines there, which really appreciate. Looking at top-line data, I was just wondering when we might see the top-line data for the phase II study of lisavanbulin in glioblastoma and would the phase I study in the newly diagnosed patients yield data before the end of this year?
Yeah. We expect for the phase II study, in recurrent glioblastoma, that will be biomarker-driven to have data available that are meaningful in the first half of 2021. I think you asked for the phase I study in the U.S. with the ABTC. This is basically, from a readout perspective, requires survivor data because it's a newly diagnosed GBM study. These patients, per definition, they do not respond. They are kind of followed for progression-free and overall survival. These data we would expect somewhat later, more towards 2022, the readout.
Okay. Thank you. I appreciate all the details.
Thank you.
The next question comes from the line of Victor Floc'h, Bryan Garnier. Please go ahead.
Hi. Thanks for taking my question. I have a couple of questions regarding derazantinib. First one, just wanted to understand, why did you prefer to start a study in gastric cancer rather than in the breast? I understand that you have great confidence about derazantinib in gastric cancer. In the meantime, the frequency of FGFR elaboration seems to be quite high in breast. Just wanted to understand what drove you towards gastric. My second question is about the competition from the monoclonal antibody targeting FGFR, such as bemarituzumab and bemarituzumab. Just wanted to hear your views about those two assets in terms of competition. Thanks a lot.
Marc, you're probably best placed to comment on why gastric rather than breast.
Well, as said, we've done a quite comprehensive screen through various models. The gastric cancer models were quite consistently responding to derazantinib, so that was our main rationale. I also mentioned before that biologically we'd seen the VEGFR2 activity of derazantinib, and that for gastric cancer population maybe really indicate utility. I think from a differentiation perspective, this to us just seemed the best move. Also, gastric cancer is not really explored by other small molecule FGFR inhibitors. The high medical need would probably also allow a quite rapid access to market if the interim results turn out to be positive.
The comment about the antibodies in development, how do we see those vis-a-vis derazantinib, the antibodies targeting FGFR?
I think this is to be seen. As in other areas, FGFR inhibitors, I think there's utility for antibodies, but there's also utility for small molecules. I think the differentiation with derazantinib really is that we have profiled this molecule in terms of its kinase inhibition profile, and it affects more than just the FGFR 1, 2, 3 kinase. It also inhibits CSF1R, which provides the differentiation for combining derazantinib with immunotherapy. The FGFR 2 may be another differentiator for our approach in gastric cancer, whilst the antibodies, they will just work against one very specific target.
Okay. Thanks for that.
Thanks for the question.
The next question comes from the line of Brian White with Cantor. Please go ahead.
Yes. Good afternoon. Thanks for taking my questions. I'm going to ask actually, another variant on a question that's just been asked, actually, on derazantinib. Just thinking about the comprehensive preclinical model experiments, I just wondered, did actually derazantinib show activity in other cancer settings in addition to gastric, or was it only gastric? Then secondly, thinking about the combination with TECENTRIQ, can I get the differentiation and the mechanism behind why it might work well with checkpoint inhibition? I wondered if you'd thought about other classes, for example, the PARP inhibitors or DNA damage repair pathways in particular. Then just thinking about more generally on oncology assets and bringing some of these programs into Basilea. I guess that these could be quite expensive if they were in later stage oncology assets and there would be a reasonable degree of competition for these programs also.
I wondered if you'd thought about co-development structures that could perhaps facilitate Basilea's involvement while still preserving cash for these programs. Thank you.
Yeah. Okay. Thanks for the questions, Brian. We'll come back to the in-licensing. In terms of the derazantinib questions, Marc.
It was a continuation to the previous question. Why did we choose gastric? What have we seen?
That wasn't the question, sorry. The question was, I get why you chose gastric. Was there any activity in any other cancers, or is that it?
Yes. I think, kind of considered this implied. Yes, we have looked across a large range of different tumor types, and we have seen signals beyond iCCA, UC and gastric cancer. I think as said, we explained why we selected gastric. I think the data we've seen and the signals we've seen in the other cancer types, they will need some further consolidation. It's too early at this point in time to disclose them and make a firm statement about our plans for additional clinical studies in cancer types beyond iCCA, UC cancer, gastric cancer. There's clearly potential there, and I guess it's somehow we considered our pipeline and drug in this context, where we have the ability to expand into various indications. The first three we have disclosed, we may move on to additional indications once we've consolidated the present signals.
The other question, I think, was about the combination, and currently, I think we're focusing on combination with immunotherapy. We are also looking into combination approaches with other compounds, including chemotherapy, anti-androgenic treatments, a number of other combination partners, which could also include PARP inhibitors. I think from realizing this clinically, we're there with the atezolizumab combination or PD-L1 inhibitor combination. We'll be disclosing in the first half of 2020 or later in 2020, our detailed combination approach in the gastric cancer study which is atezolizumab, but also will include other combination approaches.
Okay, thank you.
Then the in-licensing strategy in terms of.
Co-development in general. I think we have proven that we can do all kind of different structures around collaboration. With regard to development, we have the clinical supply agreement, for instance, with Roche for TECENTRIQ, for the gastric study, and for the esophageal cancer study. That has, of course, implications with regard to the cost, as you were mentioning costs or reducing costs for running combination trials. Historically, we have had co-development agreements as we did with Astellas, for instance, on a global basis, or more on a regional basis or local basis as we have with Dozan. There are all kinds of different structures, and the question really is: what are we trying to get out of it?
Is it that we need to have or that we like to get access to an asset as such, or do we want to have financial, let's say, participation? We are exploring all kind of different partnerships.
Just to build on that one additional point, for myself, would be the fact that clearly, our sweet spot in terms of in-licensing compounds per se, obviously, and this is what we've shown is from the sort of preclinical pre-IND to early clinical phase I, II, but obviously not any later than that. With our structure, with our financial means, that is the sort of sweet spot for us, and then the concept being that we add value through our development and our science that we can apply to it, and then we can partner or, in the future, maybe keep some ourselves. That's the sort of concept, in terms of in-licensing that we apply.
We also are very clear, we have clear in our head, we haven't disclosed this fully, but the types of target assets we're looking at in oncology, and where we're clear where we want to play and where we don't want to play.
Right. Okay. Thank you.
The next question comes from the line of Christopher Redhead, Goetzpartners. Please go ahead.
Hi, gentlemen. Just a quick question on lisavanbulin. Given the indication and given the high medical need, the nature of the patients there, do you see there's a strong possibility for getting some kind of accelerated approval, going straight from a phase I into a phase II pivotal? Is that a possibility, do you think, or not?
Certainly, this depends on the observed result. We would be really looking for patient clinical benefit, which includes response, but also durability of response. If these data are convincing, certainly GBM is one of the indication where there is an avenue for accelerated approval.
Yeah. There's a possibility that it could move faster than the other products in the end, right? You could be seeing within a relatively short period of time, that product, if you get the results, that product moving faster than the other products. Is that fair to say if you get the right results?
That's fair to say. It's data-driven. Also for the derazantinib data, the way the studies are designed, and the size of individual cohorts also allow, if we see really promising activity in these studies, to probably move them onto a track, that may not require necessary in all cases, a full-on phase III development. It's true for lisavanbulin. It may also be true for derazantinib.
Okay. That's great to know. Thanks a lot.
Next question comes from the line of Paul Verbraeken, Research Partners. Please go ahead.
Yes, good afternoon. I also have three questions. The first one is my recurring question about the progress of Cresemba in Japan and China. Is there any update there on the clinical and regulatory path? The second one is on derazantinib. With ArQule being acquired by Merck, do you notice any impact, or do you think that might change your collaboration on this compound? My last one is financial. I noticed that in your 2020 guidance, you expect an increase in cost of goods, which I find very surprising as Pfizer will take over the manufacturing in Europe of Cresemba. I was actually expecting a decline. Can you maybe give some color on that development? Thanks.
Okay. Thank you, Paul. I'll answer the first one, Japan and China, and then Adesh will pick up on your next two questions. In terms of Japan and China, there's not much of an update, probably from what we would have said previously, apart from saying that the, as you're aware with and by the way, clearly you're asking the question because these are two very important markets. For Cresemba and Japan, there is the ongoing phase III study, which is still on track, ongoing. We anticipate that the projection is that top-line results of the study would be in the second half of 2021, and then an approval subsequent to that. That's obviously pending the data being positive. That our partner, Asahi Kasei is on track with the phase III.
That is an example where we did need to do a phase III, and it's ongoing. With China, we don't yet know if we have to do a study. Our partner there is Pfizer. All I would say there is that we're anticipating definitely this year to hear the news about whether or not we need to do studies or not, whether we've got a waiver or not. We don't know the answer to that question just yet. Obviously, as soon as we do, then I'm sure you would see that because we would almost definitely issue a press release to the news about the China submission and timings and strategy.
That's the status with China and Japan. Adesh, do you want to comment on the Merck deal with ArQule and the cost of goods?
For the acquisition of ArQule by Merck, for the time being, there is not an immediate impact. The transaction only closed four weeks ago. This is not a co-development, as such, we are developing the drug. From our perspective, nothing has immediately changed. I suppose Merck also has to really step into the program and have to understand the full program and the concept behind derazantinib. I think generally speaking, we could say it's probably not a disadvantage to have a partner like Merck on the oncology side. We will see how this will evolve. For the time being, no immediate impact. With regards to our cost of product sold, you're right about the handover to Pfizer, which is actually still happening step by step. It is not completed. In 2020, we're still supplying certain material to Pfizer.
This may actually even extend into 2021. Just generally speaking, I would say that, if you look at our guidance, our product sales are actually going up. In spite of Pfizer eventually taking over the supply for their own territories, we also have our distribution partners that are selling more. The actual amount of products being sold, it increases. The way that you could look at COGS is however, or at cost of products sold, is we're currently probably around 30%, give or take. If you look at the 2018 and 2019 numbers and at the 2020 numbers and put them into relation to the non-deferred revenues, they add up to about 30%.
I think it is fair to say that in, let's say 2021 and going forward, we would expect that margin to improve because more of our revenues will be coming from royalties and milestones, which flow straight to the bottom line. The other thing is that we expect to realize economies of scale. Currently, I think cost of products sold about give or take 30% of non-deferred revenues, depending on whether you're at the lower or the higher end of our guidance for non-deferred revenues. Forward looking, that will improve.
Okay. Thanks a lot.
Thanks a lot.
Next question comes from Bob Pooler, valuationLAB. Please go ahead.
Good afternoon, gentlemen. Two questions, if I may. First, if you look at now, the threats of infectious disease have been quite prominent in the news lately. Currently one is the threat of coronavirus, but also in November, excuse me. There was also a CDC report highlighting the antibiotic threats in the U.S. Very important threats there. Do you expect with all the momentum behind infectious diseases, that we will finally see some triggers increase in action into research in infectious diseases? What is needed to make antibiotics attractive again?
Yeah.
Yeah.
Hi, Bob. Yeah, it's a good question. Obviously, I'll try and keep the answer very brief. In essence, I guess unfortunately, news such as the coronavirus but also with bacterial outbreaks and things that are resistant, our belief is that this only helps sort of spike interest and raise awareness of the issues, including AMR. Actually, we've seen during the course of the last 12 months, particularly in the U.S., changes to, for example, reinvestment of antibiotics in the U.S. market, which are helping to provide these so-called sort of pull incentives, which is the bit that's really lacking at the moment. I think there are a lot of incentives, obviously we benefit from that through the BARDA funding, through the QIDP designation. We have a series of benefits in the sort of push incentive area that we benefit from, very significantly.
Yeah, our view is that in the so-called pull incentive area, where there hasn't been too many meaningful commercial incentives in place. We've seen signs that things are getting better. I think the external environment, it's only increasing and it's in the news. Our belief is it's just a matter of time before real pull incentives are in place. There's a couple of initiatives, again, in the U.S., DISARM and REVAMP, which were talked about, but they haven't become law yet. I believe strongly is that at some point, there will be more meaningful pull incentives, and then that would change the whole NPV of an antibiotic in the future versus what it is today. We don't know when, but we firmly believe that it has to get better.
Okay. Thank you. Just a small question on ZEVTERA. You have the positive TARGET results in skin infections. Is there any potential that you'll file for this indication in Europe?
Actually, what we're currently doing with regard to the Yes, just to build on your question, slightly, is that obviously for the U.S., we need both. We can't file the skin infection in the U.S. then subsequently file the bacteremia. We have to actually wait for both to file the two together because they're cross-supportive of each other. In terms of the usefulness, the utility of the skin infection study around the globe, clearly from a medical affairs point of view, and obviously we're planning on publishing in a scientific meeting the detailed results of the skin infection study this year. We can do that, and we're planning to do that. That would be then disseminated through our commercial partners so they could utilize from a medical affairs perspective in their respective organizations.
In terms of the utility and the label in other parts of the world, my understanding is that once we've shared the data, we'll then have to find out if we can use it, if we can get any label enhancements in terms of indications. We're currently going through that process with our partners around the world to see whether or not we can actually file for the indication. Immediately, that doesn't stop us from a medical affairs point of view, using the data in that arena.
Okay. Basically, once the data is published, then there could be some potential off-label use next to lung infection.
Exactly. As I said-
Okay.
You're probably aware of this, that actually Marc, you could comment, but antibiotics are used pretty much routinely off-label anyway.
Mm-hmm. Yeah.
Obviously, companies don't promote off label, but they're used off label. I'm sure a lot of ZEVTERA usage probably already now is used in things other than HAP and CAP. Marc, you.
Of course, we would not do any proactive activities there, but there is data out from years ago on the activity in skin infections. This will confirm that this is an active compound in skin infection. We know that ceftobiprole is used outside of pneumonia. I agree with David's statement, but it doesn't give us any, unless we really file it, to get into the SmPC, we're not going to be able to actually promote it.
Okay, thanks. Just maybe one clarification on the partnering plans for your oncology assets. It is true that at the end of phase II, that's when you would be the ideal time to partner?
Obviously, it's a sort of moving feast depending on the discussions you have, the ongoing discussions you have with partners and the sort of.
the offers you get at different points in time. I'm just saying generally, or we were saying that generally, if you look at our history, it's sort of that's the time at which we usually look to partner, that usually makes sense. It doesn't mean every single time it's always exactly at the end of phase II. It depends on the ongoing discussions you're having with a partner at a particular point in time.
Okay, very clear. Thank you.
As a reminder, if you wish to register for question, please press star one. The next question comes from the line of Olav Zilian with Mirabaud. Please go ahead.
Yeah. Thank you for taking my question. I keep it very brief. First question about Cresemba sales in the U.S. Astellas reported for the past quarter that the sales rate accelerated once more, reaching almost 40% year-over-year, while in previous quarters it was in the range of some 20%. What is the particular reason for that? Like a change in the use pattern, prescription pattern in the U.S. Second question would be about EB1. It's obviously a strong predictive biomarker for patients responding to lisavanbulin. What is the frequency we should expect for this biomarker expressed in other solid tumors? Is it in the range of low percentage range, so requiring a tissue-agnostic path to approval, tissue-agnostic studies, or would you need to conduct studies in a very dedicated way, like tissue by tissue, like breast cancer, where EB1 is also relevant?
Thank you.
Okay. Thanks, Olav. Okay, Adesh, why don't you-
I can take the Cresemba question. We are, of course, very pleased with the performance. It's not that we're complaining, but probably it is fair to say that for Cresemba, you have to look at more or less a rolling 12 months performance to really get a feel for where are trends going. What you shouldn't forget is that what is being reported is at the end of the net sales. That there are also accounting impacts and so on. Generally speaking, I think what we can say is there's healthy double-digit growth. Astellas had guided for 20% growth year-over-year within their fiscal year, and they are well on track on achieving that. That's basically what we can say at this point in time. We look forward to further growth.
Our guidance implies that there is continued double-digit growth of e-market sales across the world.
Thank you.
Marc, the EB1 question.
Probably cannot say everything we know about it, but I'll try to at least give some indication on where this is going. We have looked into the prevalence of EB1 in glioblastoma. We've used several independent tissue microarrays. These are arrays where there are a couple of hundred tissues from hundreds of patients mounted onto several arrays. They can be stained and then looked at in a very short time. What we can say, I think that EB1 positive GBM is infrequent, but the prevalence is still high enough to allow for tissue screening on a reasonable scale to conduct clinical trials. That's what I think I can say now, and we will be publishing the prevalence data later this year.
To other cancer types, I know that there are some publications from a couple of years ago from Chinese scientists where, for example, EB1 in breast was, I think, 30%. We believe that it really is quite assay dependent. From what we have seen so far, we believe that we don't expect, for example, if we did a breast cancer prevalence, that we would find 30%. We believe that the published numbers may be due to different assays used. The question whether then beyond GBM, whether this goes to a cancer type by cancer type or an agnostic approach, it's a little too early. I think it really depends on the biology and also the frequency from just a strategy perspective.
If we found that in a relatively large indication, we have 3%-5% EB1, just let's see as an example, that for itself would justify going into that indication, I think from a clinical trial efficiency perspective to go agnostic. If we actually had a significant or by relatively low proportion large tumor type, it's probably not the best approach. This is really to be seen.
Okay. Maybe a follow-up question, if I may. On derazantinib, can we then assume that should you have compelling data at the phase II stage of your program, be it in bladder cancer or then late in gastric cancer, you would be in the position to submit the data for an accelerated or conditional approval?
That's what we would be trying. It's really purely data-driven. These are both high medical need indications and if you look at the erdafitinib FGFR inhibitor from BALVERSA, from Janssen, they received an accelerated approval based on 87 patients. From a patient numbers perspective, non-controlled study. From a patient numbers perspective, if the medical need is really high and unmet, I don't think that necessarily large studies are required, but it will be really data-driven. The design of the study is put in a way that we have the option to expand certain cohorts if we want to, and then we would have to expect, discuss with regulators what they expect to see.
Maybe just to finish, the trigger that would allow you to disclose interim data, is that completion of the phase I stage, or number of patients treated, or the number of events you have observed? Could you give some flavor on this, please?
For the urothelial cancer study, for example, we've just presented the design on the poster on ASCO to you, and you have several things. It's about the optimal dose for the combination of derazantinib and atezolizumab. Most of our cohorts are conducted in a so-called Simon two-stage design. You start, let's say, with a cohort, let's say you have 70 patients in the cohort, just for example purpose. Then you start a cohort with 25 patients looking at the endpoint.
If it's objective response rate, there's a minimum that has to be achieved. The study moves on to enroll another 45. At the end, there has to be a number of responses to say, "This looks promising." This stage approach and the open label nature of the study allows us then to communicate interim data that we are obtaining anyways to define the moving from one station to the next.
Okay, brilliant. Thank you so much.
Gentlemen, there are no more questions at this time.
Okay. Thank you everyone for your questions and your interest in Basilea.
Ladies and gentlemen, the conference is now over. Thank you for choosing Chorus Call, and thank you for participating in the conference. You may now disconnect your lines. Goodbye.