Mereo BioPharma: OI potential confirmed

The Phase II ASTEROID data confirms setrusumab’s potential to be the first approved treatment for osteogenesis imperfecta (OI), even though the primary endpoint was missed. Overall, the results of the trial indicated that there were impressive increases in BMD throughout the body, which should result in a significant reduction in fracture risk being detected in future Phase III studies. The failure to meet the primary endpoint, an improvement in trabecular BMD at the radius appears to be the result of unexpected heterogeneity, but there were significant increases in total BMD in all bone sites. Mereo is advancing setrusumab into Phase III; a European Phase III trial in paediatrics has already been agreed with the EMA and will be initiated once financing has been addressed. The requirements of the US FDA will be discussed in early 2020. We value Mereo at $25.29/ADS or 506p/share.

Exhibit 1: Mereo BioPharma pipeline

Source: Mereo BioPharma, Trinity Delta  Note: NE – neutrophil elastase; HH – Hypogonadotropic Hypogonadism

Osteogenesis imperfecta and setrusumab

OI (osteogenesis imperfecta), better known as brittle bone disease, is an orphan disease with different genotypes characterized by varying degrees of skeletal fragility. Most types (usually classified into eight categories) are caused by disruption or mutations of one or both of two genes (COL1A1 or COL1A2) that carry instructions for the production of type 1 collagen. Collagen is the major protein of bone and connective tissue including the skin, tendons, and sclera. Patients with OI suffer bone fractures with only minimal to moderate trauma. Type I is the least severe form of the disease and is associated with relatively few fractures, whereas Type II is the most severe and causes babies to be born with multiple fractures and die within a few weeks of birth. Breaks can occur in any bone, but breakage of the lower limbs (often the femur) is the most common. In general, the earlier the fractures occur in life, the more severe the disease is.

Setrusumab (BPS804) is a fully humanised monoclonal antibody targeting sclerostin (SOST). Sclerostin is a protein secreted by the osteocyte cell (which is the cornerstone cell of bone structure) and that plays a pivotal role in bone homeostasis. Sclerostin was discovered to be the underlying mechanism of the high bone mass seen in two rare autosomal recessive diseases (sclerosteosis and Van Buchem disease). It is the deficiency in sclerostin that is associated with increased bone formation and decreased bone resorption (Exhibit 2)[1]. This led to the inhibition of sclerostin being explored as a novel therapeutic target[2] for a variety of bone disorders, including large commercial indications such as osteoporosis and rare diseases like OI.

Exhibit 2: The mechanism of action of setrusumab for osteogenesis imperfecta

Source: Mereo BioPharma

The appeal of sclerostin inhibition lay in how it could not simply improve bone mass through reducing bone resorption (antiresorptive) but, importantly, through the activation of bone formation (osteoanabolic). Other than parathyroid hormone (PTH) analogues, all current therapies for osteoporosis are antiresorptive. Reducing bone resorption, though helpful for the maintenance or improvement of bone strength, cannot replace the bone already lost and restore skeletal architecture. The promise of sclerostin inhibition meant that the optimal strategy of an initial anabolic treatment, followed by antiresorptive maintenance therapy, could be potentially achieved.

The approach has been validated by the approval of Amgen/UCB’s romosozumab (Evenity), which has been approved for the treatment of severe osteoporosis at high risk for fracture in the US and Japan, and is expected to be approved shortly in Europe following a positive CHMP decision in October. The improvement from baseline in lumber spine BMD (measured using DXA) in the Phase III ARCH study was 13.7% with romosozumab vs 5.0% with alendronate at 12 months, and corresponded with a reduction in the incidence rate of new vertebral fractures of 2.3% (4.0% vs 6.3%, respectively) at 12 months, and 5.7% (6.2% with romosozumab for 12 months followed by alendronate for 12 months vs 11.9% with only alendronate) at 24 months.

There are currently no approved treatments for OI in Europe or the US. Therapy is aimed at preventing and treating fractures, maintaining individual mobility, and currently striving to strengthen bones and the supporting muscles. Surgical procedures, such as rodding (the insertion of metal rods in the bones), have shown benefit but carry the risks associated with such major surgery.

The main medications used off-label to treat OI are bisphosphonates, such as alendronate and risedronate, which are anti-resorptives. They have been shown to ameliorate bone mass; however, as the 2016 Cochrane review concluded from a meta-analysis of 14 trials, it is not clear that they reduce fracture rates.

The potential of denosumab (Amgen’s Prolia, RANK-L antibody) and teriparatide (Lilly’s Forteo, PTH analogue) have also being investigated, but both have severe limitations. The use of denosumab is associated with the “rebound effect” (an increased risk of fractures), hypercalcaemia and hypercalciuria once treatment is stopped or even between doses. It is not recommended that teriparatide is used in children or in adults for more than two years in total, because to the risk of osteosarcoma.

Mereo is the only company developing an anti-sclerostin antibody for IO. Amgen are focused on osteoporosis, and Eli Lily deprioritised development of blosozumab in 2015 despite positive Phase II results because of material formulation difficulties.

Look beyond the primary endpoint

The data from the Phase IIb ASTEROID trial indicates that setrusumab increases BMD throughout the body, and to a similar degree in all three subtypes of OI (Types I, III, and IV) assessed. There was also a trend of a reduction in fractures, (trial not powered-statistically to show a difference). Inconsistent with the rest of the data, the primary endpoint of change in trabecular volumetric BMD (Tb vBMD) of the wrist (radius) using HRpQCT (high resolution peripheral quantitative computed tomography) failed to show an improvement; there was unexpected heterogeneity in the data, including some patients with greater vBMD than seen in normal individuals.

HRpQCT is a novel method of measuring BMD and was selected as the primary endpoint as it was thought to be more predictive of fracture risk. However, there was exceptional heterogeneity in the Tb vBMD (Exhibit 3); a previous study by Kocijan et al[3] had suggested that the range would be 59-113mgHA/cm³. In this study it was 18-279mgHA/cm³, with some OI patients with higher baseline vBMD than was recorded in normal individuals in the Kocijan study (range: 149-190 mgHA/cm³). This heterogeneity resulted in modest mean changes in Tb vBMD with very high standard deviations.

Exhibit 3: Changes in trabecular and total vBMD in the radius, measured using HRpQCT

Source: Mereo BioPharma  Note: SD: Standard deviation

In marked contrast to the limited changes in Tb vBMD, there were significant changes in Total vBMD in the radius with the medium and high doses, which were consistent with those seen in other bones measured with DXA (see below). The precise dosing of setrusumab during the trial has not being disclosed. This suggests that setrusumab has a greater impact on the outer cortical bone than it does on the inner trabecular bone tissue. However, we would be cautious about drawing any conclusions using the Tb vBMD data, given the very high recorded baseline Tb vBMD in some OI patients.

The secondary endpoints for the trial included changes in areal BMD (aBMD) measured using the widely-used DXA (dual-energy X-ray absorptiometry) at the lumbar spine, femoral neck and total hip.

In the lumbar spine, a dose response was observed with highly significant increases in aBMD in the high and medium dose arms (Exhibit 4). The effects on lumber spine aBMD seen in this trial are consistent with that seen in the 14-patient Phase IIa, in which there was a 4% increase on day 141, and also the open-label arm of the current trial in which there was a 3.5% increase at six months (n=12). There was also a similar increase at the lumbar spine aBMD across the different subtypes of OI.

Exhibit 4: Change in areal BMD in the lumber spine at 6 and 12 months with different dose cohorts, measured using DXA

Source: Mereo BioPharma  Note: aBMD – areal bone mineral density

Exhibit 5: Change in areal BMD at the lumber spine at 6 and 12 months with the high dose with different subtypes, measured using DXA

Source: Mereo BioPharma  Note: aBMD – areal bone mineral density

A similar result was seen at both the femoral head and total hip; there was a significant improvement in aBMD with the highest dose of setrusumab.

Exhibit 6: Change in areal BMD at the femoral and the total hip at 12 months, measured using DXA

Source: Mereo BioPharma

The changes observed with the high dose of setrusumab in the current trial compare very favourably with those seen in trials in adults with both teriparatide and bisphosphonates. As with all meta-analysis, it is important to recognise the limitations of the approach, this is particularly the case with OI as all the trials are small in size (Mereo’s Phase IIb ASTEROID trial in the largest ever trial conducted in OI). Having said that, it is reassuring that the analysis consistently suggests that setrusumab is better than teriparatide and bisphosphonates at improving aBMD in OI, in a manner similar to that seen with the anti-sclerostin antibody romosozumab in osteoporosis.

Exhibit 7: Comparison between the increases in areal BMD observed in the ASTEROID Phase IIb with high dose setrusumab with those seen with teriparatide and bisphosphonates at 12 months, measured using DXA

Source: Mereo BioPharma, Trinity Delta

The key clinical benefit, which will need to be demonstrated for any drug to be approved for the treatment of OI, is a reduction in fracture risk. This was a secondary endpoint in the ASTEROID trial. The study was not powered to demonstrate a significant change in fracture risk; however, a trend was observed which bodes well for the Phase III studies. In the high dose arm, the occurrence rate of fractures was less than half that seen in the other arms (Exhibit 8).

Exhibit 8: Percentage of patients with at least one fracture in 12 months and the occurrence rate per patient year

Source: Mereo BioPharma, Trinity Delta

Setrusumab well tolerated

Setrusumab was safe and well tolerated throughout the trial with adverse events (AEs) distributed across all dose arms. Five (19%), eight (40%) and four (18%) serious AEs were seen in the high, medium and low dose arms, respectively, but none of them were classed as being treatment related (three were initially recorded as being related).

All AEs that were potentially cardiac related (eg. chest pain) were closely investigated due to the cardiovascular concerns regarding romosozumab, which resulted in the black-box warning on the label. Nine events were investigated and discussed due to cardiac concerns by the data monitoring committee (DMC), and none were concluded to represent a cardiovascular safety concern.

Next steps

Patients in the ASTEROID Phase IIb study will be followed up for 12 months to investigate any “off effect”, as is seen with denosumab, as well as providing useful information for drug “holidays” for the potential chronic treatment of OI with setrusumab. Over the next 12 months, patients will have had the option to receive bisphosphonates immediately after treatment with setrusumab was stopped (at month 12) or six months later (at month 18) and they will have the following data collected:

• DXA and HRpQCT scans at months 18 and 24;
• Bone biomarker analysis at months 14 and 18;
• Fracture data collected until month 24.

Mereo will also start preparing for the pivotal Phase IIb/III trial to support paediatric approval of setrusumab in OI in Europe. The study design has been agreed with the EMA and is planned to start in 2020 (Exhibit 9). About 160 children aged 5 to 18 with OI (Type I, III, and IV), who are currently on bisphosphonate therapy, will be treated in a placebo-controlled study and the fracture rate at 12 months will be the primary endpoint. Secondary endpoints will include BMD measured by DXA (HRpQCT for academic purposes only, following the issues in the Phase IIb trial), bone turnover biomarkers, and quality of life scores.

If the data from the paediatric trial is also compelling, it is questionable if it would be ethical to require an additional placebo-controlled Phase III study in adults, as there are no approved treatment for OI. Consequently, setrusumab could be granted conditional approval in adults in Europe on the back of the Phase IIb ASTEROID study and the paediatric Phase IIb/III study, with the requirement for a Phase IV study to confirm the efficacy and safety profile of the antibody.

The requirements for US approval are not yet known. However, Mereo plans to have a discussion with the FDA in early 2020, and we expect the FDA will require a similar trial design to that agreed with the EMA. So, it should be possible essentially to use the same study for the approval of setrusumab in both Europe and the US by adding US sites to the existing trial design, together inevitably with a few additional modifications.

Exhibit 9: Clinical trial design of the pivotal European Phase IIb/III study

Source: Mereo BioPharma

Valuation and Financials

We continue to value Mereo at $25.29/ADS or 506p/share ($704m or £541m) and have made no changes to our estimates. We had been expecting compelling data from the Phase IIb ASTEROID trial, and the results are in line with our expectations. We still consider the likelihood of setrusumab approval in OI to be 50% and value the asset at $431m or £331m. We will continue to review our assumptions as more data from the trial are published and further details on the Phase III programme are disclosed.

We note that there is considerable interest in assets that target bone indications, as was highlighted by the acquisition of Clementia Pharmaceuticals by Ipsen for $1.3bn, including a $1bn upfront payment. At the time of the purchase, Clementia’s only clinical asset was palovarotene, which was in Phase III for  fibrodysplasia ossificans progressive (FOP) and Phase II for multiple osteochondromas (MO), two rare bone disorders.

Exhibit 10: Summary of financials

Source: Company, Trinity Delta; Notes: Our estimates exclude the costs associated with the pivotal Phase III paediatric study with BPS-804, because the timing of the trial is yet to be decided. We include in FY20 long-term debt (other financing cash flow) of £25m, which is indicative of our estimate of the company’s capital requirement; this could be achieved through an equity raise, debt of proceeds from partnering non-rare disease assets.

[1] Role and mechanism of action of sclerostin in bone. J Delgado-Calle et al Bone Vol 96, March 2017, Pages 29-37

[2] Role of sclerostin in bone and cartilage and its potential as a therapeutic target in bone diseases. Lewiecki EM Ther Adv Musculoskelet Dis. 2014 Apr; 6(2): 48–57.

[3] Kocijan R. et al (2015) Bone structure assessed by HR-pQCT, TBS and DXL in adult patients with different types of osteogenesis imperfecta Osteoporosis International Oct;26(10):2431-40