Mesoblast — A significant impact on back pain

Mesoblast recently announced results from the MPC-06-ID trial in 391 patients with chronic low back pain due to degenerative disc disease. Low back pain is an extremely common problem with the Centers for Disease Control and Prevention (CDC) reporting that 28% of US adults (65.8 million people as of the date of the survey) have had low back pain that has lasted a day or more during the prior three months.1 Degenerative disc disease is usually caused by ageing, trauma or repetitive motion, and often results in inflammation and extreme pain. It has been estimated to include about 39% of those with chronic low back pain,2 many of whom are not well served by current treatments.

Treatments include physical therapy, non-steroidal anti-inflammatory medications (NSAIDs) such as acetaminophen or ibuprofen, chiropractic procedures, spinal injections and opioids. Opioid use is especially high, with 59% of opioid prescriptions estimated to go to people with chronic low back pain.3 Data show that the biggest therapeutic impact of opioids is over the short term (less than three months), with that impact dissipating over time. In a review of opioid use for low back pain, short-term opioid use led to placebo-adjusted pain improvements of 10.1 points (out of 100), while intermediate-term use (more than three months but less than 12 months) was associated with a lower 8.1 point placebo-adjusted improvement. There appears to be little evidence of benefit for opioid use beyond 12 months.4

What makes the pain data from the MPC-06-ID trial so remarkable is that rexlemestrocel-L plus HA (which, as mentioned, can improve the targeting of MSCs to the points of inflammation5) was able to demonstrate the same level of pain relief that opioids typically achieve in the short term, over a 24-month time frame with a single injection.

The improvement in pain scores (through VAS) was even better in those who had back pain of shorter duration. This is likely due to the fact that degenerative disc disease is the source of the pain and the longer it goes on, the more intractable the pain might become. In this subset, the difference is even more pronounced, with both 12- and 24-month p-values at p<0.0001.

Rexlemestrocel-L also demonstrated significantly (p<0.05) greater pain reduction in opioid users (n=168) compared to saline, as well as a significant 40% reduction in opioid use versus baseline over 24 months (p=0.03). Patients treated with saline experienced the opposite: an increase in opioid use. Additionally, on the EuroQoL 5-Dimensional (EQ-5D) Index, a measure of function and disability, rexlemestrocel-L with HA showed a statistically significant improvement over saline at both 12 and 24 months (p=0.009 and p=0.02, respectively).

An interesting development following the release of the data was that principals of SurgCenter Development, one of the largest private operators of ambulatory surgical centers in the United States, specializing specifically in spine, orthopedic and joint procedures, led a US$110m financing round for the company. This investment provides some validation for the data from medical specialists familiar with treating back pain.

The company plans to discuss these results with the FDA as well as potential approval pathways. There is an argument to be made for approval based on these results given the significant impact on pain and opioid use, with a generally safe therapy consisting of a single injection over two years. However, our base case is that additional clinical development will be needed, with a launch in FY25.

The company also recently announced the results of the DREAM-HF trial of Revascor (the brand name of rexlemestrocel-L in heart failure) in 537 patients with chronic heart failure. As in the back pain trial, patients received only a single dose of the medicine. The trial missed the primary endpoint of time to recurrent, non-fatal, decompensated heart failure major adverse cardiac events (HF-MACE) that occur prior to the first terminal cardiac event. However, this is a non-standard endpoint that focuses more on worsening symptoms related to heart failure than cardiac-specific events such as heart attacks and strokes. Revascor did demonstrate a benefit on the 3P-MACE endpoint, which focuses on reduction in cardiac death, heart attack or stroke (see Exhibit 3). Revascor was able to reduce these events by a significant 30% (p=0.027), which is a larger effect size than typically seen with heart failure drugs.

Importantly, there was a dramatic 55% effect size (p=0.009) in the 206 New York Heart Association functional classification (NYHA) class II subgroup. The trial enrolled NYHA class II and class III patients, with class II patients being less severe with only a slight limitation in physical activity versus the marked limitation with class III patients.

Revascor was also able to significantly reduce hospitalization rates from non-fatal heart attacks and strokes in all patients by 68% (p=0.0002), with similar effects in both class II and class III patients.

So, while the DREAM-HF trial missed the primary endpoint of HF-MACE (which was negotiated by former partner Teva Pharmaceuticals around eight years ago), there were significant effect sizes across a number of other endpoints. Also, while not the primary endpoint, 3P-MACE has been a standard primary endpoint for cardiovascular outcomes trials involving diabetes drugs.6 Importantly, Novo Nordisk’s Victoza (liraglutide), which had US$3.9bn in peak sales in 2018, received a label expansion to include reduction of risk of major adverse cardiovascular events in 2017 based on data from a trial (LEADER) with a 3P-MACE primary endpoint. Effect sizes were generally small, with the large size of the trial (9,340 versus 537 in the Mesoblast trial) assisting in obtaining statistically significant p-values.

With regard to heart failure-specific medications, Entresto (sacubitril/valsartan) was approved in 2015 to treat NYHA class II–IV heart failure patients and had sales of US$2.5bn in 2020, with expectations for around $4.5bn in peak sales according to EvaluatePharma. Entresto obtained approval based on a primary endpoint of heart failure hospitalization or cardiovascular death. Interestingly, just as in Mesoblast’s DREAM-HF trial, the data were particularly strong in NYHA class II patients, where there was a 26% risk reduction. NYHA class III patients saw only a 7% reduction in the primary endpoint, which was not significant (see Exhibit 7). It is therefore possible that these patients are too severely afflicted for medical therapy to have much of an impact.

As with back pain, the company plans to discuss these results with the FDA as well as potential approval pathways. The data are complex and the primary endpoint was missed, but some very hard endpoints were hit with both statistical and clinical significance. It is possible that the FDA would accept an accelerated approval pathway, but it is our base case that another trial will be needed, with a launch in FY26.

In December, the DSMB ran the third interim analysis of the remestemcel-L trial in COVID-19 related ARDS patients following the enrolment of 180 patients, and concluded the trial was unlikely to meet the endpoint of a 43% reduction in mortality at 30 days after treatment. This hurdle was based on data from the early part of the pandemic when less was known about how to treat COVID-19 patients. Over the course of the trial, better care of these patients reduced mortality rates, which made the 43% reduction unattainable. So, while this specific endpoint was not met, it is possible that there were trends towards improved mortality. The company has yet to announce any of the secondary endpoints.

As a reminder, in November Novartis signed a partnership with the company to develop remestemcel-L for ARDS, whether or not the ARDS was caused by COVID-19, as well as potentially other conditions. Novartis is expected to make a US$25m upfront payment and an additional US$25m equity investment once the transaction closes. Mesoblast may receive a total of US$505m in development milestones, an additional US$750m in sales milestones and tiered double-digit royalties. Additionally, as part of the agreement, Novartis will fully fund global clinical development for all-cause ARDS and potentially other respiratory indications once the all-cause ARDS Phase III is initiated (however, it is unknown whether the design or timing of this trial will be affected by the DSMB action; the companies are awaiting the 60-day results, expected in Q1). Mesoblast will be responsible for both clinical and commercial manufacturing and Novartis will purchase commercial product from the company. There are also US$50m in manufacturing milestones related to the successful implementation of a next-generation manufacturing process. Novartis will be responsible for any capital expenditure related to increasing capacity requirements for the manufacture of remestemcel-L.

Outside respiratory indications, Novartis also has an option to become the commercial distributor of Ryoncil (the brand name of remestemcel-L in graft versus host diseases). For most non-respiratory indications, Novartis and Mesoblast may fund development and commercialization on a 50/50 profit share basis.

We have decreased our valuation from A$4.9bn or A$8.41 per share (A$7.88 per diluted share) to A$4.1bn or A$6.35 per share (A$5.86 per diluted share). This reduction was driven mainly by the fact that we now expect launch in low back pain in FY25 versus FY22 previously, and launch in CHF in FY26 compared to FY23 previously. We have maintained our 50% probabilities of success for both programs. The fact that the trials did not hit their primary endpoints is balanced by the fact that we have strong data indicating efficacy, while we had very limited data previously. We may adjust our probabilities of success as well as launch timing in the future depending on FDA guidance. With regard to ARDS, we have made no changes as we have never included COVID-19 related ARDS in our valuation and do not believe there is a readthrough to the broader ARDS indication. Our valuation was also affected by the recent private placement, which increased net cash, but also increased the number of shares and options outstanding.

Mesoblast reported total revenue for the second quarter of FY21 (the period ending 31 December 2020) of US$2.2m, slightly higher than the same period a year ago. US$2.1m of this was royalties on sales of Temcell in Japan, which increased from US$2.0m last year. R&D was essentially flat at US$14.2m. Manufacturing expenses increased from US$5.1m to US$6.5m due mainly to the building up inventory for the planned Ryoncil launch (now put on hold following the FDA complete response letter in calendar Q420). Management and administration expenses also increased to US$7.9m from US$6.8m due to higher share-based payments to employees and consultants and increased overheads. The only meaningful change to our estimates was lower finance costs due to a lower than expected run rate.

Also, for the period ending 31 December 2020, Mesoblast reported cash and equivalents of US$77.5m, down from US$108.1m for the period ending 30 September 2020. Also, the company had US$41.6m in short-term debt (although in January 2021, the interest-only period for the loan was extended up to March 2022, subject to achieving certain milestones) and an additional US$49.1m in long-term debt. In March 2021, the company announced a US$110m private placement led by the principals of SurgCenter Development, one of the largest private operators of ambulatory surgical centres in the United States (80 operational facilities across 20 states) specializing specifically in spine, orthopedic and joint procedures. 60.1m shares were issued at A$2.30 per share, with an additional 15.0m warrants issued with an exercise price of A$2.88/share.

We currently forecast a requirement to raise US$75m by the end of FY23 (modelled as illustrative debt), down from US$140m previously. We believe Mesoblast’s financing needs will depend in part on whether Ryoncil sales start sooner than anticipated, potential future partnership royalties/milestones, as well as the company’s aggressiveness in investing in the business. Additionally, we have not yet included the US$25m equity investment from Novartis and will do so once the transaction closes (guidance on timing has not been provided).