Basilea Pharmaceutica — Update 23 March 2016

Basilea is a Switzerland-based biopharmaceutical company with a focus on developing innovative antibiotics, antifungals and oncology drugs that target drug resistance and non-responsiveness to available drug treatments. Basilea has two approved hospital products, Cresemba and Zevtera, for the treatment of severe mold and bacterial infections respectively. Basilea commercialises both drugs in the major European markets and seeks partnerships outside these markets. Basilea was spun out of Roche in 2000 and to date has raised around CHF925m net proceeds (including initial funding from Roche, a private placement, IPO, a rights offer and convertible bond issue). Basilea is headquartered in Basel and employs c 250 people.

We value Basilea at CHF1,114m or CHF103.2/share, which is based on a risk-adjusted NPV (rNPV) analysis, including Cresemba and Zevtera, in addition to indicative valuations for the earlier-stage pipeline (BAL101553 and BAL3833). Our valuation suggests the current share price is more than underpinned by the potential for both Cresemba and Zevtera where approvals have already been granted, with the market ascribing seemingly limited value to Zevtera in the US or the earlier-stage oncology pipeline.

The key sensitivities for Basilea, in our view, relate to successful European commercialisation of both Cresemba and Zevtera, progress with partnering Zevtera in the US and crystallising value from the earlier-stage pipeline. There should be natural commercial synergies between Cresemba and Zevtera, which should provide operational leverage. While both are targeting significant markets, we believe there are two main challenges to uptake: (1) related to Zevtera antimicrobial stewardship (described later in this report); and (2) the availability of cheaper generic treatments. In the US, success with Zevtera, and Cresemba is largely in the hands of existing, or prospective, partners where we have limited visibility.

We believe that Basilea’s gross cash of CHF364.7m (which includes the CHF194.7m net of transaction costs raised from the CHF200m convertible bond placement) should be sufficient to fund operations beyond our forecast operational profitability in 2019 even excluding future potential milestone income for either Zevtera in the US or the earlier-stage assets. Profitability will be driven by European sales of Cresemba and Zevtera, which should overtake and offset the expected deferred revenue decline in the coming years. We also assume an uptick in R&D spend owing to Basilea’s potential contribution towards the Zevtera US Phase III clinical development; we assume a contribution of around CHF40-50m (with a partner funding the deficit). We also include S&M spend for Cresemba and Zevtera in the five major European markets, and assume that this will be largely in place in the next six to 12 months.

Cresemba (isavuconazole) is a broad-spectrum antifungal for the treatment of severe, life-threatening fungal infections. It was approved in the US in March 2015 for the treatment of both invasive aspergillosis and invasive mucormycosis and has been launched by US partner Astellas. Cresemba was approved in Europe in October 2015. Basilea plans to commercialise Cresemba itself in the major European markets (launched in Germany and the UK in Q116) and to seek partners in other territories. With broad-spectrum activity against fungal moulds, in addition to safety and other benefits (including fewer drug interactions and once-daily dosing) over current standard treatments, we believe Cresemba could have a unique position in this growing market, even in the face of increasing genericisation. We forecast global peak sales potential of $600m, with Cresemba contributing c 60% to our valuation.

Cresemba is approved in the US and EU to treat both invasive aspergillosis, caused by Aspergillus moulds, in addition to invasive mucormycosis (also known as zygomycosis). While these are uncommon infections, they have high mortality rates; according to the CDC, the mucormycosis all-cause mortality rate is 54%, with 76% for pulmonary infections. For invasive aspergillosis mortality rates vary between 25% for stem cell transplants to 59% for organ transplants. Although precise patient numbers for each infection are not available, increasing numbers of immunocompromised patients (for example from more transplants and more cancer patients undergoing intensive chemotherapy) has likely led to increasing incidence, especially of mucormycosis, in the last decade.

Current standard of care for invasive aspergillosis is with Pfizer’s Vfend (voriconazole), also a triazole antifungal, available in both oral and IV formulations, which was approved in the US in 2002. The intravenous formulation of Vfend cannot be used to treat patients with moderate to severe renal impairment and the dose has to be lowered in patients with liver impairment. It also has a number of drug interactions and is associated with visual disturbances. Vfend is not approved to treat invasive mucormycosis, where until the Cresemba approval, treatment was limited to amphotericin B (generically available), and its newer liposomal formulation Astellas/Gilead AmBisome which is associated with infusion-related events and nephrotoxicity. Other antifungals include echinocandins such as Merck’s Canidas (caspofungin), used for yeast (Candida) infections and in invasive mold infections patients who cannot tolerate other treatments.

Cresemba was approved on the basis of two Phase III trials in mould infections; SECURE, a 516-patient trial comparing Cresemba to Vfend in patients with invasive aspergillosis, in addition to the VITAL open-label trial in 156 patients including rare moulds and patients with invasive aspergillosis with renal impairment. Cresemba was found to be non-inferior to Vfend in terms of efficacy (as measured by all-cause mortality at day 42), with a more favourable safety profile, reporting statistically fewer drug-related adverse events; in particular, significantly fewer liver, eye and skin disorders (all known effects of Vfend). Importantly according to the Cresemba product label there is no need for dose modification in mild, moderate, severe renal disease nor in mild or moderate liver impairment. This directly sets Cresemba apart from Vfend, The main data are summarised in Exhibit 3.

A third trial investigating Cresemba for the treatment of invasive Candida infections was recently completed. The ACTIVE trial in 450 patients compared Cresemba to IV caspofungin followed by oral Vfend. The trial did not meet the primary endpoint of non-inferiority to the comparator at the end of IV treatment, measured by overall response (ORR), with Cresemba ORR of 60.3% compared to 71.1%. This was a treatment difference of -10.8% (95% CI: -19.9% to +1.8%), with the lower bound of the 95% confidence interval (-19.9%) below the pre-specified, non-inferiority margin of -15% and hence Cresemba could not be considered non-inferior to the comparator.

However Cresemba was found to be comparable on key secondary endpoints, including overall treatment response two weeks after the end of treatment (54.8% versus 57.2%; -2.7% difference; 95% CI: -12.2% to +6.8%; the original primary endpoint of the study) in addition to comparable all-cause mortality at days 14 and 56. These data did not affect the current approvals, which remain the main target markets for Cresemba. Basilea will now seek to discuss these data with key opinion leaders (KOLs) to establish any future regulatory and/or development strategy.

Following FDA approval in March 2015, partner Astellas is now marketing Cresemba in the US. Astellas already has an anti-infectives commercial presence in the US, marketing AmBisome and Mycamine (an echinocandin), with Cresemba therefore a complementary addition to this portfolio. According to Astellas’ most recent strategic update, Cresemba is one of its “key new products” expected to drive both mid-term and longer-term growth. Basilea is entitled to a tiered royalty on Cresemba sales starting in the mid-teens and ramping up to mid-20s and up to CHF290m of sales milestones.

In Europe, formal European Medicines Agency (EMA) approval was granted in October 2015. Basilea plans to essentially commercialise Cresemba alone in key European markets (including UK, Germany, France, and Italy) through a fee-for-service contract salesforce via an agreement with Quintiles. This agreement provides dedicated sales representatives, who are trained by Basilea, in addition to market access and administrative support. Basilea could potentially seek to bring the commercial infrastructure in house in the future. Cresemba launched in the UK and Germany in Q116. We expect further European launches during 2016 (pricing and reimbursement need to be agreed in each country, which takes time).

Pfizer reported Vfend global sales of $682m in 2015, down from the 2010 peak of $825m following the availability of oral generics in 2011, in addition to an IV generic in 2012 in the US market; full genericisation is expected from 2016. AmBisome, which is used to treat invasive mucormycosis, among others, had global sales of c $470m; the US patent expires in 2016. IMS health reported total voriconazole (branded plus available generic) sales of $890m in 2014. Together, this suggests a combined market of $2.4bn based on sales of approved leading antifungals. Based on average pricing of c $4,500 per patient with Vfend and nearly $9,500 for AmBisome (pricing will vary by country and by treatment duration, which is specific to the underlying cause of the infection and patient response, so these are approximate estimates), this suggests around 220k patients are treated with these two products alone annually.

Given that Cresemba offers safety advantages over current treatment and has a broad spectrum of activity including mucormycosis, it is well suited to empiric use when the exact underlying microbial cause of infection is unknown. This is particularly important with severe, life-threatening invasive fungal infections, where timely initiation of treatment is important. Diagnosing the causative pathogen is not always straightforward; for example, mucormycosis can be difficult to differentiate from other filamentous fungi such as aspergillosis. These features should therefore differentiate Cresemba from the competition, potentially providing physicians with an option that overcomes the limitations of current therapy.

Although there are many patients who receive treatment for invasive fungal infections with generically available products, we believe it will be the current branded market described above (c 220k patients globally) that is likely to be the main target for Cresemba. We assume that Cresemba could be used in around a third of these patients. In the US Astellas has priced IV Cresemba roughly in line with IV Vfend (around $240/day), but with the oral priced at a premium ($140/day, which we estimate is more than double the oral Vfend price). Based on the average treatment duration in the Phase III trials of 47 days, including 8-9 days with IV, this suggests an average Cresemba price per patient in the US of c $7,430. Treatment duration could be longer in the real-life setting compared to the clinical trials; for example according to research in Belgium,1 Vfend mean treatment duration was 63 days.

In Europe pricing will be agreed on a country-by-country basis once formal approval is granted. In contrast to the usual pricing trend where most European drugs are priced at a discount to the US, in the antifungal space the reverse is often true. For example, in Europe Vfend IV is priced at around €325-400 per day and the oral at €60-90 per day, giving an overall treatment cost per patient of €5,500-6,900 ($6,200-7,700 at current FX rates, compared to around $4,500 in the US). Hence, we believe a premium in Europe is possible. We assume an average treatment cost of €7,675 per patient, which is based on an IV price of €450/day, which would place Cresemba at a premium to Vfend, but at a discount to AmBisome (priced at around €545-630), and an oral price of €90-100/day, also at a premium to Vfend (AmBisome is not available orally). This could be justified given Cresemba’s efficacy in mucormycosis, in addition to safety advantages.

Based on these pricing assumptions and our 31% penetration of the current invasive fungal infections branded antifungal patient market, we arrive at global peak Cresemba sales of $600m, which we assume will be reached in around eight to nine years post launch, so in 2024. As previously described, Basilea will be selling Cresemba independently in the major European markets, with royalties on US sales from partner Astellas. For the RoW territories, which could include other European markets, in addition to Asia, South America etc, we assume that Basilea will seek additional partners and for simplicity we model a single royalty, which we estimate at 35%. This takes into account distribution deals, which typically lead to an effective economic interest of around 40%, in addition to more traditional out-licensing agreements, such as Japan, where the royalty rate could be around 30%.

We believe the main competition for Cresemba will be from increasing genericisation of the current invasive fungal infections therapy market, where Vfend is in the process of going generic. However, despite the availability of oral generics since 2011, in addition to the first IV generic in 2012, global sales have only been marginally affected to date, declining 8% from the 2010 peak. Vfend also illustrates that products with improved profiles such as safety can gain and maintain a strong foothold in markets, despite the availability of alternative generics.

Cresemba has been awarded orphan drug status in both the US and Europe, providing 10 years of exclusivity in Europe and seven years in the US. In addition, it has been awarded qualified infectious disease product (QIDP), which provides an additional five years of exclusivity in the US. In Europe further 2 years exclusivity could be granted on the basis of paediatric approval.

Basilea has successfully navigated a number of challenges along ceftobiprole’s path to market and Zevtera/Mabelio, a broad spectrum cephalosporin antibiotic for severe bacterial infections including against MRSA, is now approved in Europe for hospital treatment of pneumonia. Basilea is commercialising Zevtera alone in core European markets through a fee-for service contract salesforce in place through Quintiles. In the US further clinical studies will be needed to secure approval (Basilea is currently preparing phase 3 study protocols seeking Special Protocol Assessments (SPAs) for submission to the FDA in the first quarter of 2016). Furthermore additional US clinical trials will commence once an US partnership has been announced, which the company has guided for 2016.

The discovery of penicillin in 1928 was one of the major health advances of the 20th century, allowing for effective treatment of previously life-threatening infections. However, antibiotic overuse, rapid evolution of bacterial strains with resistance to existing antibiotics, and a lack of new antibiotics has led to the current situation of emerging superbugs (eg Methicillin-resistant Staphylococcus aureus, carbapenem resistant Enterobactericiae spp), which are leading to increasing deaths from previously treatable infections.

Hospitals can be breeding grounds for opportunistic bacterial infections, owing to dense populations of ill people who are generally less able to fight infection, combined with movement of healthcare workers, with bacteria able to spread through direct contact. According to a CDC prevalence survey, hospital acquired pneumonia (HAP) is one of the most common healthcare-associated infections (HAIs), responsible for around 22% of HAIs. In contrast to HAP, community-acquired pneumonia (CAP) is contracted outside healthcare facilities and is one of the leading causes of hospitalisation in the US. Pneumonia can be caused by a number of different pathogens, including Staphylococcus aureus, Streptococcus pneumoniae, Pseudomonas aeruginosa and Enterococcus faecalis.

The implementation of simple techniques in hospital and healthcare settings, such as hand sanitisation and better hygiene practices, has in general led to a decrease in the rate of MRSA (and other drug-resistant infections) in recent years. However, latest reports suggest that this has not come anywhere close to fully addressing the issues; the recently commissioned Review on Antimicrobial Resistance estimates that around 700k people currently die each year globally from drug-resistant infections, with this projected to increase to 10 million by 2050. Importantly, the WHO has reported that antibiotic resistance is a serious, worldwide threat to public health, and estimates that “people with MRSA are estimated to be 64% more likely to die than people with a non-resistant form of the infection”.

Zevtera was approved in Europe for hospital-acquired and hospitalised community-acquired pneumonia towards the end of 2013. Obtaining approval was a lengthy process owing to a number of regulatory challenges along the way. An extensive ceftobiprole Phase III programme was initially commenced in 2004, consisting of two Phase III trials in complicated skin and skin structure infections (STRAUSS I and STRAUSS II), which included a total of c 1,600 patients, in addition to trials in both HAP (including VAP) and CAP in nearly 1,500 patients, which started later, in 2005 and 2006. In 2005 a development and commercialisation partnership was agreed with J&J.

Initial positive data were reported in 2006 from STRAUSS I where ceftobiprole was found to be non-inferior to vancomycin (93.3% of patients cured on ceftobiprole compared to 93.5% with vancomycin), with positive data from STRAUSS II in early 2007 (ceftobiprole was found to be non-inferior to the comparator of vancomycin + ceftazidime, with 91% cured compared to 90%). These two studies formed the basis of the initial filings in complicated skin infections (cSSSI, which was the typical route to market for antibiotics for serious bacterial infections) in both the US (May 2007) and in Europe (June 2007). In 2008, the approvals were hindered by issues identified during the review process. The FDA highlighted issues with data integrity relating to clinical trial site issues, with regulators concluding that the studies had not been conducted in compliance with good clinical practice (GCP), rendering the data unreliable (audits executed by Basilea/J&J concluded that even with exclusion of any “unreliable” data, Zevtera’s reported efficacy was maintained). However, these issues precluded approvals in both the US and Europe. J&J returned ceftobiprole rights to Basilea in 2010. Meanwhile, Basilea successfully went into arbitration with J&J, seeking damages for the approval delays, with damages of $130m awarded to Basilea at the end of 2010.

Since these issues, Basilea has held meetings with regulators to determine the best path forward for ceftobiprole given available data, including positive pneumonia data (summarised in Exhibit 5). These trials demonstrated that Zevtera was non-inferior to typical combination therapy for the treatment of severe pneumonia. In Europe, Basilea filed the pneumonia data via the decentralised procedure in 2012, with approval granted in October 2013. Since then, national launches have occurred in Germany, France, Italy, Austria and Switzerland and the UK, with a launch in Spain expected in 2016. Sales build up for novel antibiotics takes time to materialise, as regional access often needs to be granted, and then hospitals generally test new antibiotics to ensure effectiveness against specific pathogens found in the hospital (antibiotic stewardship). Basilea has disclosed CHF0.5m reported product sales for Zevtera in 2015.

In the US, the FDA has advised Basilea that new Phase III studies will be needed to be eligible for review. Ceftobiprole has been awarded Qualified Infectious Disease Product (QIDP) for both CAP and acute bacterial skin and skin structure infections and we believe that these indications are likely to form the basis for any future development strategy. Following recent discussions with the FDA, Basilea will consider conducting cross-supportive clinical studies for the indications of acute bacterial skin and skin structure infections, community-acquired bacterial pneumonia and/or Staphylococcus aureus bloodstream infections, or bacteraemia (including bacterial endocarditis). Staphylococcus aureus bacteraemia (presence of bacteria in the blood), is an indication where few antibiotics are currently approved and which the FDA considers an area of unmet need. Thus Basilea intends to conduct a bacteraemia study to further differentiate Zevtera from available cephalosporins. The company is in the process of preparing phase 3 study protocols seeking Special Protocol Assessments (SPAs) for submission to the FDA in the first quarter of 2016. An SPA provides written guidance by the FDA that the agreed clinical trial design would support a regulatory submission for drug approval, should the study meet its objectives.

Basilea estimates that each trial will cost around $40-50m, with all three trials therefore costing $120-150m. Although Basilea does currently have sufficient cash resources to fund these trials, a partnership is being sought. This is to preserve cash and to focus on the European commercialisation of both Zevtera and Cresemba. Basilea has indicated that it is in active discussions for the US opportunity, which could potentially lead to an agreement in 2016. Furthermore Basilea expects that it would not initiate phase 3 studies until a US collaboration deal is finalised. On the basis that a partner is announced in 2016, phase III trials could commence in 2017 (we anticipate 3 year duration), implying that a US 2021 launch date could be feasible.

There are a number of branded hospital antibiotics for use in severe bacterial infections. These include Pfizer’s Zyvox and Merck’s Cubicin. Zyvox which is approved for HAP, CAP and cSSSI (complicated skin and skin structure infections) including diabetic foot infections reported 2014 sales of $1.35bn. Cubicin (originally sold by Cubist in the US, which was acquired by Merck in early 2015) last reported US sales of $908m by Cubist in 2013 were Cubicin is approved for cSSSI and bloodstream infections including right-sided endocarditis, but is not approved for pneumonia, having failed in clinical trials. Cubicin had a 14% share of the US market for IV antibiotics to treat serious gram-positive infections, based on days of therapy (according to Cubist’s 2013 Annual SEC filing). Other branded hospital antibiotics include, but are not limited to, Tygacil (Wyeth/Pfizer), Vibativ (Theravance) and Teflaro (Forest/Actavis/Allergan). In addition, vancomycin is available generically and has around a 70% share of the US market (based on days of therapy).

Given the size of the patient market, significant peak sales could be possible. However, despite Cubicin’s success in the US, demonstrating the willingness of US physicians to embrace new antibiotics despite the availability of cheaper generic options, in Europe we believe uptake of new antibiotics has been slower. Although Novartis (commercial partner for Cubicin in Europe) has not specifically disclosed in-market sales to our knowledge, Cubist’s royalty payment from Novartis in 2013 was $58.7m, which would equate to sales of $235-391m (based on a 15-25% royalty). We believe this reflects a general caution in Europe around use of new antibiotics unless absolutely necessary, to prevent future antimicrobial resistance (also referred to as “antimicrobial stewardship”). However, with Zevtera as a single agent shown to be as effective as combination therapy in pneumonia, this could present an attractive option for physicians, reducing the need for two antibiotics. Furthermore, despite advances in establishing the underlying causative pathogen, diagnosis still takes time (around 48 hours) and can be inconclusive; hence broad spectrum products, such as Zevtera, are attractive empiric options. Zevtera’s broad spectrum bacterial activity includes MRSA and Pseudomonas spp as outlined by its product label.

$100m peak sales in Europe could be achieved with only 3.2% penetration of pneumonia patients (22% of the c 4-5 million patients that receive treatment annually for serious bacterial infections) and an average treatment price per patient of €2,160 based on the midpoint price and treatment duration (Zevtera is priced in Europe at €160-198/day with average treatment duration of 10 days). We forecast peak around six years from 2016. In the US, if Basilea can secure a partner for future development, we believe Zevtera could generate around $317m of peak sales. This assumes use in bacteremia, pneumonia and skin indications and is based on a 2.5% penetration of the 4-5 million patient market with pricing of $2,000 per patient. This price is based on a premium to Cubicin (priced at around $1,400 for 14 days of treatment for an average 70kg person), but a discount to Zyvox (priced at around $2,500 for 10 days or intravenous treatment). Given the US market has generally been more willing to accept new antibiotics, this penetration could be overly conservative. A 5% penetration would equate to $650m peak sales in the US.

For the RoW territories, which could include other European markets, in addition to Asia, South America etc, as for Cresemba, we assume that Basilea will seek additional partners and, for simplicity, we model a royalty that we estimate at 35%, reflecting a blend on distribution and more traditional out-licensing agreements. A distribution agreement was signed with Hikma in October 2015 for MENA countries (Middle East and North Africa).

We believe our relatively modest penetrations should allow for competition in the market, both from existing and new branded products, in addition to generics. Vancomycin, which commands the majority market share in the US, is already generic. A Cubicin generic is anticipated from 2016. Zyvox is in the process of being genericised.

Zevtera has been awarded QIDP in the US for both pneumonia and skin infections, which should provide at least ten years of market exclusivity (includes an additional five years of exclusivity in the US on top of standard data exclusivity that is five years for a new chemical entity) from the date of approval. In the US the ceftobiprole composition of matter patent is due to expire in 2019 (although Basilea will be eligible to a SPC adding up to five years of patent protection, ie through 2024), however generics cannot launch even at risk within the data exclusivity period. In Europe, where no QIDP equivalent exists, Zevtera will benefit from the standard eight years’ data exclusivity and two years of market exclusivity.

To complement the primary focus on anti-infectives, Basilea also has an early-stage pipeline focused on oncology products that target resistance to current traditional chemotherapies. Basilea’s oncology clinical pipeline contains BAL101553 and BAL3833 for drug-resistant cancers. There could be synergies between the anti-infectives and oncology pipelines as a large number of invasive fungal and bacterial infections develop in cancer patients that are immunocompromised owing to treatment with aggressive chemotherapies.

BAL101553 is a highly soluble prodrug of BAL27862, which induces tumour cell death through activation of a checkpoint important for tumour cell division. BAL27872 targets microtubules, but with a binding site and mechanism of action distinct from that of currently approved microtubule-targeting agents (MTA; such as Taxol, Taxotere, Abraxane, Jevtana and the Vinca alkaloids). BAL101553 has shown anticancer activity in a number of treatment-resistant tumour models, including tumours resistant to standard MTAs, as well as other therapeutic approaches including radiotherapy. Furthermore, BAL27862 has been observed to affect tumour blood supply in preclinical models and has also been shown to activate a checkpoint involved in preventing cell proliferation. BAL27862 has also been shown to have penetration into the brain, which could open up possibilities as a treatment for glioblastoma (GBM).

The maximum tolerated dose (MTD) was established in a previous open-label Phase I study of the Intravenous (IV) formulation in patients with advanced solid tumours, which also demonstrated first signs of activity. Of the 15 evaluable patients (from 21 recruited) there was one partial response (PR) and five disease stabilisations (SD). This phase I study was expanded into an open-label phase 2a study (40 evaluable patients) with the primary objective to further evaluate the safety and tolerability of MTD and sub-MTD doses of BAL101553. Patient recruitment in this phase 1/2a study has completed. Of the 72 patients dosed to date in December 2015 52 patients had undergone evaluation for tumour response. One patient with an ampullary pancreatic cancer achieved a partial response and was treated for more than two years with BAL101553 and 12 patients had stable disease lasting between two and eight months. For a two-hour infusion the recommended Phase 2 dose has been defined as 30 mg/m2. The majority of patients with signals of clinical benefit, including the partial response, were dosed at the recommended dose or below. Currently, nine patients are still ongoing with BAL101553 treatment.

A Phase I/II trial of the oral formulation is also ongoing to establish the MTD (Phase I portion), which will then be expanded to a Phase IIa to assess initial efficacy in a variety of tumours. We expect Basilea to seek a partner once Phase IIa data become available (either from the IV study, or later when oral data become available). A collaboration in glioblastoma (GBM) is also possible.

BAL3833 is a panRAF kinase inhibitor in Phase I development in advanced solid tumours. BAL3833 inhibits both BRAF and CRAF, part of the RAF family of kinases, and inhibits SRC, which are involved in cell growth. BRAF mutations are found in certain cancers, most notably melanoma, with BRAF inhibitors such as Zelboraf approved to treat melanoma patients with a BRAF mutation. Preclinical data suggest that BAL3833 has activity in models resistant to current BRAF inhibitors.

Basilea is subject to the usual biotech and drug development risks, including clinical development delays or failures, regulatory risks, competitor successes, partnering setbacks, and financing and commercial risks. The key sensitivities for Basilea relate to successful European commercialisation of both Cresemba and Zevtera, progress with various assets in the US and crystallising value from the earlier-stage pipeline.

In Europe, Zevtera and Cresemba are both already approved, hence, few development risks now remain, with the focus on successful commercialisation. This is more or less being pursued independently in the five major European markets, through a contract salesforce via Quintiles, with dedicated sales reps and Basilea responsible for training. We believe there are a number of operational synergies between Zevtera and Cresemba, with both anti-infectives for hospital use, which should provide sales and marketing leverage. However, we believe there are two main challenges to Zevtera and Cresemba uptake: (1) related to Zevtera antimicrobial stewardship, particularly in Europe where novel anti-infectives are often used with caution to try and prevent the emergence of new resistant strains; and (2) the availability of cheap generic options, particularly in an era of pricing pressures and restricted budgets.

In the US, success with Zevtera, and Cresemba is largely in the hands of existing or prospective partners. We believe Astellas is well placed to deliver on Cresemba sales, with Cresemba one of its key new products and an already existing commercial presence in anti-infectives. Basilea is seeking a US partnership for Zevtera, but we have limited visibility on the timing and terms of any deal.

For the earlier-stage pipeline, both clinical development and partnering risks remain. We expect Basilea to develop each oncology asset (BAL101553 and BAL3833) through to Phase II with initial proof-of-concept data before partnering. However, we have limited visibility beyond that on the terms and timing of any potential deal(s).

We value Basilea at CHF1,114m or CHF103.2/share, which is based on a risk-adjusted NPV (rNPV) analysis utilising a 12.5% discount rate for products in development, and a 10% discount rate for approved products. Our valuation includes rNPV contributions for each of the main products described in this report, in addition to last reported net cash. For each product we forecast revenues and income to Basilea (based on our peak sales forecasts and Basilea’s participation in these sales, in addition to any future milestones) and relevant development, commercial and manufacturing costs. We apply a probability of success to each product, with 100% where products have been approved. The breakdown of our valuation is shown in Exhibit 6, with our major assumptions described in more detail below.

For both Cresemba and Zevtera in Europe, our peak sales forecasts include in-market sales in the five major countries (UK, Germany, France, Italy and Spain) that we expect Basilea to generate through its own contract salesforce (supplied via its agreement with Quintiles). As described in the Financials section of this report, we believe the S&M spend in 2015 (of around CHF32-35m) should be sufficient to support commercialisation in these markets, with only marginal increases in future years, and we allocate this pro rata to each product based on peak sales.

In the US Cresemba is partnered with Astellas and our valuation includes tiered future royalties on sales starting from the mid-teens ramping up to mid-20s. Basilea is also entitled to up to CHF290m of sales-related milestones, some of which are included in our valuation. For Zevtera in the US, Basilea is seeking a commercial partnership; our valuation includes an assumed deal with a total value of CHF100m (a base case assumption, which could prove conservative) and a 20% royalty on sales. Our valuation includes an assumed 30% contribution by Basilea towards the cost of Phase III development ($120-150m total cost for three Phase III trials). With efficacy already demonstrated in previous trials for skin and lung infections, we believe the biggest risk to Zevtera in the US is around partnering, where we have limited visibility on the likelihood of securing any deal. Hence we apply a 50% probability of success. For the rest of world (ex-US and ex-EU) we assume a royalty on sales of 35% for both Cresemba and Zevtera; the bulk of our sales for each is in the US and Europe.

We also include indicative valuations for BAL101553 and BAL3833 and for simplicity we assume both will be partnered post completion of Phase II trials, in exchange for a royalty on sales (starting at 15% for both given we have assumed partnering once proof-of-concept data become available). Assessing the potential for each product is challenging in the absence of proof-of-concept data and without knowing the indications that will be pursued in the future. For both we include peak sales of $500m; this will ultimately be determined by which future oncology indications are developed, but we believe it is a reasonable base case assumption for an oncology asset that could have use in a variety of drug-resistant solid tumours.

As previously described in this report, the potential markets for both Cresemba and Zevtera are significant, with our peak sales forecasts based on assumed penetration rates that we believe are achievable. There remains a need for novel antimicrobial products for hospital use, which could drive upside to our forecasts, although increasing genericisation, with the availability of cheaper options, could restrict uptake, which could lead to downside risk. We provide a brief sensitivity analysis below for each product, flexing peak sales (and the associated trajectory to reach those sales) to highlight the impact on valuation, all else being equal.

These assumptions maintain the current split between US/EU and RoW forecasts; there is heightened sensitivity in Europe, where operating margins will be higher than the royalties earned in other regions. Cresemba is therefore more sensitive to peak sales, as we believe the opportunity is larger in Europe than in the US (which is in contrast to Zevtera).

In 2015 Basilea issued a CHF200m convertible bond, which is due in 2022. Including the net proceeds of the convertible bond placement, reported gross cash at end Dec 2015 was CHF364.7m; this is a healthy cash balance that should be sufficient to fund operations beyond our forecast profitability in 2019. There are three main levers that will affect achievement of this: sales trajectory of marketed products, operating expenses and partnering activities. These are discussed below.

In the past few years the bulk of Basilea’s revenues have related to deferred recognition of upfront and milestone payments already received under various licensing deals. In 2015 these licence-related revenues were CHF52.2m, consisting of CHF11.8m from Astellas (relating to the original upfront payment of CHF67.5m received in 2010, in addition to the US filing milestone of CHF12m received in 2014, plus CHF30m milestone received for US Cresemba approval in 2015) and CHF36.9m relating to the Stiefel upfront of CHF224m received in 2012; the remaining CHF1.3m relates to additional contract revenue on services provided to Astellas for isavuconazole.

Excluding the CHF1.3m contract revenue we expect similar levels of licence income in both 2016 and 2017 (CHF49.5m), but starting to decline from 2018 as recognition of the Stiefel upfront should be completed during 2018 and the Astellas milestones should be fully recognised during 2020.

We forecast CHF4.8m in product sales in addition to ongoing royalties from Astellas on Cresemba sales in the US. In the absence of a US partner for ceftobiprole, we do not include any potential future royalties on US sales in our financial forecasts (although these are included in our valuation). Hence our future financial forecasts reflect only sales that are possible based on the currently approved products. We believe that from 2018 direct sales of Cresemba and Zevtera in Europe should overtake licence income (which is set to decline from highs of around CHF49.5m in 2017). A summary of our projections towards profitability is shown in Exhibit 7.

R&D expenses in 2015 reported at CHF60.0m. We forecast 2016 R&D spend of CHF62m. We expect R&D spend to increase in future years, particularly owing to Basilea’s potential contribution towards US Phase III Zevtera development. Our forecasts currently assume that Basilea will contribute around 30% of the total $120-150m development costs). Future R&D spend will also depend on partnering activities for the earlier-stage assets, although we do not anticipate any deals for either BAL101553 or BAL3833 until Phase II proof-of-concept data become available.

SG&A in 2015 was CHF54.1m. Although Basilea does not explicitly split out spend on G&A and S&M, we believe underlying G&A costs are around CHF21m (evidenced by SG&A spend of CHF21.3m in 2013 when there were no S&M-related expenses). We forecast a marginal 1-2% increase in underlying G&A. We estimate that S&M spend in 2015 was around CHF33m for the ongoing build-out of a contract salesforce via Quintiles. We believe CHF32-35m of S&M spend should support commercialisation of both Zevtera and Cresemba in the five major European markets, and hence we do not expect any further significant upticks in the near term.

Hence, taken together we forecast total operating expenses of CHF1116.8m in 2016 (+2% compared to 2015 owing to the slight increase in S&M spend), and continued growth in operating expenses, albeit more modestly, in 2017 and 2018 owing to increasing R&D expenses in future years, with spend for Zevtera US Phase III development.

Our future financial forecasts do not include unknown or uncertain partnering milestones. Therefore, if Basilea is able to partner Zevtera in the US, it could lead to an upfront and future milestone income, which is also not included in our financial forecasts, hence any milestones could provide upside to our financial forecasts. Basilea is also aiming to partner the earlier-stage pipeline, in particular the oncology asset BAL101553 once full Phase IIa data become available; we do not include any milestone income in our financial forecasts for any such activity. Hence, it is possible, depending on the timing and terms of any partnering activities, that profitability could be achieved earlier than our current forecast of 2019 if any milestones become due.

Basilea reported cash and equivalents, including liquid assets, of CHF 364.7m at end Dec 2015. This includes the CHF200m convertible bond (which we record on the balance sheet as long-term debt). The convertible bond is due in 2022 and has a conversion price of CHF126 (based on a 30% premium to the volume weighted average share price on 9 December 2015, which was CHF97). The coupon is 2.75%, or CHF5.5m/year, which is paid semi-annually in arrears. Our financial model suggests current cash should be sufficient to fund operations beyond 2019 forecast profitability, even in the absence of any milestone payments.