Mesoblast — Update 21 September 2016

Mesoblast has an extensive portfolio of products based on its allogeneic, off-the-shelf adult mesenchymal precursor cell (MPC) and mesenchymal stem cell (MSC) technologies, which are summarised in Exhibits 1 and 2. While many of the products have shown promise in exploratory efficacy studies in Phase II, with Teva’s withdrawal as funding partner Mesoblast needs to carefully target expenditure to maximise chances of success, while minimising the risk of shareholder dilution through capital raises.

The release of FY16 accounts shows that Mesoblast is making good progress in reducing ongoing expenditure. Operating and investing cash outflows were reduced by 15% in FY16 to US$89.7m. Guidance is for further cost-cutting measures, including a headcount reduction in July, to reduce burn by up to US$25m in the current year. However, this reduction in expenditure will be partly offset by expenditure on the Phase III HF trial, which was previously funded by Teva. Guidance is for the trial to cost US$13m to the interim analysis in Q1 CY17 – we assume a further US$5m expenditure over the remainder of the financial year. Cash at 30 June was US$80.9m and the company has put in place an equity funding facility that will provide US$90m over three years, so funding is assured until at least mid-FY18.

Mesoblast has four active programmes in Phase III, one that is Phase III ready and two in Phase II. The company works on a two-tier product structure, with the primary goal of supporting late-stage or top-tier products on their path towards commercialisation – either through own-company financing or partnership agreements (Exhibits 1 and 2). The first tier comprises four products with a significant mid-term revenue opportunity: MPC-150-IM in congestive heart failure, MPC-06-ID in low back pain due to degenerative disc disease, MSC-100-IV in acute graft versus host disease (aGvHD) and MPC-300-IV in a number of chronic inflammatory conditions including biologic-refractory RA and diabetic kidney disease. The remaining tier two products are in Phase II or III and will advance into tier one on the basis of data, market opportunity or partnering capability.

Teva’s withdrawal from the cardiovascular programme and Celgene’s decision to let its right of first refusal for selected products lapse, combined with the increased risk that dilutive capital raisings may be required to fund its ongoing development programme, has seen the stock price fall by 60% over the past year, despite reporting good clinical progress over the period.

Mesoblast announced encouraging top-line results in August from the final cohort in its Phase II trial of MPC-300-IV in rheumatoid arthritis (RA) patients who are refractory to biologics such as TNF alpha. Patients were randomised to receive a single intravenous (iv) infusion of either 1m or 2m MPCs/kg or placebo. The iv infusions of MPCs were well tolerated with no serious adverse events over the 12-week study period.

The exploratory efficacy analysis showed there were dose-related improvements in clinical symptoms, function and disease activity, with the highest dose (2m MPCs/kg) providing the greatest benefit. Exhibit 3 shows that the most clinically meaningful ACR70 (at least 70% improvement in the number of affected joints and other symptoms) was achieved at 12 weeks by significantly more patients receiving the high dose (2m cells/kg) than the control group (27% vs 0%). More patients in the high-dose group also achieved ACR50 response at 12 weeks (31% vs 19% in controls), but the difference was not significant (Exhibit 4).

Responses to MPC-300-IV were consistently higher in the pre-defined subgroup of patients who had been treated with one to two prior biologics; 36% for high dose vs 0% in controls for ACR70, 55% vs 11% for ACR50 and 55% vs 33% for ACR20. However, with the smaller numbers in this subgroup the effects were not statistically significant.

Exhibit 5 shows that by week 12 there was no difference between the treatment and placebo groups in the number of patients achieving ACR20 responses, with the high placebo response meaning that around 50% of subjects in each group achieved this minimal improvement. This result will present some challenges in choosing the appropriate primary endpoint for a Phase III trial because, while the ACR70 and ACR50 responses are much more clinically meaningful, drugs for the treatment of RA have typically been approved on the basis of a greater proportion of patients achieving the minimal ACR20 (20% improvement) response. We expect that in a larger trial we would see a benefit in ACR20 as well as for the more stringent ACR50 and ACR70 endpoints, but the Phase II results add an extra element of uncertainty.

MPC-300-IV treatment also led to dose-related improvements in physical function as measured by the health assessment questionnaire-disability index (HAQ-DI). Exhibit 6 shows that subjects in the high-dose group had significantly greater improvement in the HAQ-DI score from baseline compared to the placebo group at both four and 12 weeks. At 12 weeks significantly more subjects in the high-dose group had achieved the minimum clinically important improvement in physical function (HAQ-DI reduced by at least 0.22), 90% vs 38%, p=0.003.

Achieving an ACR70 response in 27% of subjects vs 0% on placebo represents impressive efficacy in refractory patients. This result is comparable to response rates seen to TNF alpha biologics in first-line therapy, and better than the approved JAK inhibitor Xeljanz in comparable biologic-refractory patients (ACR70 in 10-14% of patients). If benefits of this magnitude are confirmed in subsequent clinical trials, MPC-300-IV could potentially be used as a first-line treatment option for patients who have failed to respond to anti-TNF or another biologic agent.

The approval of Entresto (sacubitril/valsartan, Novartis) in the US and Europe in H215 provided another option for treating patients with symptomatic (NYHA class II-IV) HF and a reduced ejection fraction (HFREF), the type of HF that Mesoblast is targeting in its development programme. Entresto is a combination of sacubitril, a neprilysin inhibitor, and valsartan, an angiotensin II receptor blocker. The list price for Entresto in the US is about US$4,500 a year before discounts. Entresto reduced the risk of cardiovascular (CV) death or HF hospitalisation by 20% versus the ACE-inhibitor enalapril in the Paradigm-HF Phase III trial in 8,442 patients (76% had NYHA Class I or II HF). As Exhibit 7 shows, there was a trend towards greater benefit in patients with milder class II HF than in those with more severe Class III symptoms. In Class III patients there was only minimal 7% improvement in the primary endpoint (a composite of CV death or first hospitalisation for worsening HF, HR=0.93), although there was a more meaningful 18% reduction in the risk of death from cardiac causes (HR=0.82). For the small number of Class IV HF patients (n=29) the situation was reversed, with a small 10% reduction in cardiac deaths but a larger 25% reduction in the composite primary endpoint.

Mesoblast’s Phase II trial of MPC-150-IM randomised a total of 60 patients with Class II or III HFREF and normal ventricular rhythm (QRS < 120ms). A post hoc analysis of HF-related major adverse cardiac events (HF-MACE), defined as cardiac death, resuscitated ventricular fibrillation or HF hospitalisation, found that HF-MACE incidence was significantly lower in the high dose group (150m cells) than the controls (0% vs 33%, p=0.025) over three years of follow up. Exhibit 7 compares the efficacy of Entresto seen in the Paradigm-HF pivotal trial and the results of post hoc analysis of HF-MACE and death from cardiac causes in Mesoblast’s phase II trial. HF MACE is similar to the composite endpoint in the Entresto trial, but with the addition of successfully resuscitated ventricular fibrillation and exclusion of deaths due to stroke.

As a post hoc analysis in a small study that was not powered to detect clinical endpoints, it is not clear to what extent these benefits of MPC therapy will be repeated in large pivotal trials. However, we note that the incidence of HF-MACE events and death from cardiovascular causes in the control patients in the MPC trial were similar to those seen in the control arm of the Paradigm-HF trial, which suggests that the difference between the MPC-treated and control patients were not due to an unusually high event rate in the control arm. We note that even when looking at the less dramatic improvements seen when the data for all 45 patients in the 3 MPC dose groups were pooled, the incidence of HF-MACE events was still 40% lower and incidence of cardiac death 78% lower than in the control group.

In September 2015 Mesoblast released additional analysis of the Phase II trial of MPC-150-IM which found that the cell therapy had the greatest cardioprotective effect in the subset of patients with more advanced heart failure, defined by substantial baseline left ventricular (LV) contractile abnormality (LV end systolic volume (LVESV) >100ml), as shown in Exhibit 8. The bottom rows of Exhibit 7 shows that the HF-MACE and cardiac death event rates were also much higher in the patients with advanced HF in the control group but no events occurred in the high MPC dose group.

Mesoblast concluded from this analysis that the optimum target population for MPC therapy is likely to be patients with advanced heart failure with a high rate of progression.

In May the American College of Cardiology Foundation, the American Heart Association (AHA) and the European Society of Cardiology issued updated HF treatment guidelines which strongly recommended that Entresto be used to treat patients with Class III HF due to reduced ejection fraction, as well as those with milder Class II HF symptoms. This is a stronger endorsement that we had expected given the modest benefit of Entresto on the composite primary endpoint in Class III patients in the pivotal study.

We had previously assumed that uptake would be 3% of eligible patients. With the strong guideline endorsement of Entresto we have reduced forecast uptake in the US and Europe to 2.5% and 2.0% of eligible patients respectively, as we expect use of MPC-150-IM to be more tightly focused on patients with advanced heart failure. Entresto is a twice-daily tablet, whereas intracardiac injection of MPC-100-IM is an invasive procedure that is administered in a specialist cardiac catheter lab, so it makes sense that use of MPC therapy would be concentrated in the optimum target patient population.

The Phase III trial of MPC-150-IM is recruiting patients with New York Heart Association (NYHA) Class II or III HF with an ejection fraction ≤40%, with additional criteria including high baseline NT-proBNP level and a history of a heart failure hospitalization in the past 9 months to select for patients with advanced heart failure. In Exhibit 9 we identify patients as eligible for MPC-150-IM therapy (the addressable market) as those who are in Class II-III HF with a reduced ejection fraction (LVEF<40%). As Exhibit 9 shows, this includes 36% of all HF patients, which is equivalent to 2,150,000 patients in the US in 2016. Our revised uptake rate in the US of 2.5% of eligible patients would represent 53,700 patients receiving MPC therapy each year.

As Exhibit 9 shows, we consider all patients with Class II or III HF and reduced ejection fraction as eligible for treatment. However, as previously stated, we expect use to be concentrated in patients with more advanced heart failure. While it is not clear what proportion of eligible patients could be considered prime candidates for treatment, in Exhibit 10 we have taken as a proxy those patients with HF with a reduced ejection fraction and Class III symptoms. This category represents 12% of all HF patients, equivalent to 726,000 patients in the US. Our peak utilisation of 53,700 patients per year would represent 6% of these prime candidates for treatment.

We note that if we assume that the average survival of these prime candidates patients is four years and that each patient receives MPC therapy only once, this 6% annual uptake would be equivalent to 24% of prime candidates receiving MPC therapy over the course of their treatment.

The eventual uptake of MPC-150-IM, if approved, will be strongly influenced by its relative efficacy when compared to Entresto. We will reassess our uptake assumptions when we have more information on the efficacy of MPC-150-IM. This may come when the results of the interim analysis in Q117, or we may have to wait for the top line results of the first pivotal trial, potentially in mid-2018.

We also note that we assume that MPC-150-IM is priced at US$20,000 per treatment in the US (US$15,000 per treatment in Europe). This is intermediate between the US$4,500 annual cost of treatment with Entresto and ~US$35,000 for an implantable defibrillator device (ICD). We think this pricing is reasonable given that multiple clinical trials have shown long-lasting reductions in mortality of 20-50% with the use of these devices in this patient group. If MPC-150-IM therapy is shown to provide a long-lasting benefit, this may justify pricing in line with ICD devices, which represents potential upside to our forecasts.

Mesoblast has suspended recruitment in the Phase III trial of MPC-CBE for the expansion of haematopoietic stem cells (HSC) within cord blood for HSC transplant recipients. Existing and future cash resources will be primarily directed towards tier one programmes for the foreseeable future and the company was unable to say when recruitment will restart. The company impaired the full carrying amount of the intangible assets relating to the programme (and for a separate MPC-Micro-IO programme for age-related macular degeneration) and recognised a US$61.9m non-cash impairment charge in the FY16 accounts. As a result of the indefinite suspension of patient enrolment on the Phase III trial we have removed MPC-CBE from our valuation model (previously valued at A$46.4m or A$0.12 per share).

We reduce our valuation for Mesoblast to A$1.5bn from A$1.8bn (A$3.84 per basic share from A$4.67, or to A$3.82/share diluted for options that would be in the money if shares traded in line with the NPV, from A$4.62). The breakdown of contribution to the rNPV is shown below (Exhibit 11). Note that these product NPVs do not account for R&D costs, which are grouped together as a single line item.

Most notably, our NPV for MPC-150-IM in HF decreases from A$2.79/share to A$2.08/share due to reduced peak uptake among eligible HF patients from 3% to 2.5% and 2.0% in the US and Europe, respectively, deferred launch of MSC-100-IV for aGvHD from FY17 to FY18 with recruitment in the pivotal trial now expected in Q117 vs Q416 previously, and the removal of the MPC-CBE cord blood expansion programme from our model as outlined above (previously valued at A$46.4m or A$0.12 per share).

We have also removed the MSC-100-IV programme in biologic-refractory Crohn’s disease from our valuation model until we have better visibility as to future development plans (previously valued at A$8.2m or A$0.02 per share). The Phase III trial of MSC-100-IV in biologic-refractory Crohn’s disease patients completed recruitment in June 2014, according to the entry on ClinicalTrials.gov (trial NCT00482092). However, with no news to date as to the outcome of the trial, and in light of the fact that an additional pivotal Phase III programme will be required for approval, it is uncertain whether this programme will progress any further.

Note that our model retains similar revenue share assumptions for the HF and acute myocardial infarction programmes to those that we used under the lapsed Teva collaboration, on the assumption that if the development programme is successful it could negotiate similar terms with a new partner given the more advanced stage of development. These assumptions include a revenue share that ranges from the twenties to reach 40% when combined sales for HF and myocardial infarction exceed US$2bn, and COGS that range from 14% down to 10% as sales increase. We assume combined milestone payments for approval for HF in the US and Europe total US$125m, risked to 50% likelihood.

We highlight key upcoming potential catalysts that could allow for share price inflections: interim analysis of MPC-06-ID in low back pain in Q117, futility analysis of MPC-150-IM in HF in Q117, interim results from MSC-100-IV Phase III in paediatric aGvHD in Q416 and full results from the same trial in Q217.

Mesoblast changed its reporting currency to US dollars ahead of its IPO in the US in 2015.

Mesoblast cut its cash burn in FY16 (year end 30 June) by 15% to US$89.7m. Further cost-cutting measures, including a headcount reduction in July, reductions in laboratory and office space and completion of recruitment in the MPC-25-IC heart attack study, are expected to reduce burn by up to US$25m in the current year. However, this reduction will be partly offset by expenditure on the Phase III HF trial that was previously funded by Teva. Guidance is for the trial to cost US$13m up to the interim analysis in Q1 CY17; we expect the full year cost to be around US$20m. We have increased forecast operating cash outflow by 2% and 1% in FY17 and FY18 respectively, and now forecast operating cash outflow to be US$80.3m in both FY17 and FY18 and US$82.4m in FY19.

The reported pre-tax loss for FY16 was US$90.8m, a 6% reduction on the prior year. The reported loss of US$4.1m was affected by a number of non-cash items, including: recognition of the remaining US$22.5m of deferred revenue under the Teva collaboration; a US$61.9m non-cash impairment charge for the full carrying value of intangible assets relating to MPC-CBE for the expansion of HSC, as well as for the separate MPC-Micro-IO programme for age-related macular degeneration; a US$37.4m benefit from revaluing the contingent consideration payable to Osiris for MSC assets including Temcell; and a US$86.7m tax benefit relating primarily to the recognition of deferred tax assets (US$60m) and reversal of deferred tax liabilities (US$21.8m).

In the absence of additional financing and/or partnerships, we estimate that Mesoblast has sufficient cash to fund operations including key late-stage trials into late FY17, while faster uptake of MSC-100-IV (aGvHD) in Japan could extend this runway. Thereafter, its late-stage pipeline will require funds – ideally through partnerships, although the timing of partnering agreement is uncertain. We estimate that an additional US$85m will be required to fund operations until end FY18, most of which could be supplied by the US$90m discretionary equity facility with Kentgrove Capital.

Management has indicated that it does not intend to take on debt to fund its development programmes; however, for purely illustrative purposes we follow our standard practice and add long-term debt of US$85m in FY18 in our forecasts. We note that a scenario where US$85m was raised through an equity issue at the current share price of A$1.13/share would see our diluted valuation fall from A$3.82/share to A$3.27/share.

Mesoblast is subject to the risks typically associated with biotech company drug development, including the possibility of unfavourable outcomes in clinical trials, regulatory changes, success of competitors and commercial decisions by partners or potential partners. With Teva’s withdrawal from the cardiovascular programme, funding of Mesoblast’s annual cash burn of ~US$70-80m is a significant source of uncertainty. The company has entered an equity finance facility that will provide up to A$120m (~US$90m) to fund the HF Phase III programme, but with a current market capitalisation of A$530m (~US$400m) that could result in significant dilution of existing shareholders. We expect Mesoblast to seek non-dilutive funding from partnering deals, which could reduce the equity funding requirement. Potential partnering catalysts could include filing a US BLA for paediatric aGvHD in 2017, interim futility analysis for MPC-06-ID degenerative disc disease Phase III in Q117 and the second interim futility analysis in the HF Phase III in Q117.