Currency in EUR
Last close As at 17/03/2023
EUR8.80
▲ −0.18 (−2.00%)
Market capitalisation
EUR203m
Research: Healthcare
In January, Carmat announced data from the initial 10 patients included in the first leg of its EU pivotal study investigating the surgical implantation of the Carmat heart in patients suffering from end-stage biventricular heart failure (HF). In total, 70% of these patients achieved the primary endpoint, which is survival at six months post implant. Also, improvements to the device manufacturing process starting in Q418 slightly delayed timelines. The trial is expected to resume with the improved Carmat heart in Q319.
Written by
Maxim Jacobs
Carmat |
Half-year results and clinical updates |
2019 outlook |
Healthcare equipment |
27 September 2019 |
Share price performance
Business description
Next events
Analysts
Carmat is a research client of Edison Investment Research Limited |
In January, Carmat announced data from the initial 10 patients included in the first leg of its EU pivotal study investigating the surgical implantation of the Carmat heart in patients suffering from end-stage biventricular heart failure (HF). In total, 70% of these patients achieved the primary endpoint, which is survival at six months post implant. Also, improvements to the device manufacturing process starting in Q418 slightly delayed timelines. The trial is expected to resume with the improved Carmat heart in Q319.
Year end |
Revenue (€m) |
PBT* |
EPS* |
DPS* |
P/E |
Yield |
12/17 |
0.03 |
(31.5) |
(3.24) |
0.0 |
N/A |
N/A |
12/18 |
0.72 |
(43.7) |
(4.54) |
0.0 |
N/A |
N/A |
12/19e |
0.70 |
(49.5) |
(4.47) |
0.0 |
N/A |
N/A |
12/20e |
7.50 |
(49.0) |
(3.84) |
0.0 |
N/A |
N/A |
Note: *PBT and EPS are normalised, excluding amortisation of acquired intangibles, exceptional items and share-based payments.
Total artificial heart aims to imitate human function
Carmat has developed a biocompatible, biventricular mechanical heart that is designed to replicate the functionality and morphology of the human heart as closely as possible using self-regulatory mechanisms and biocompatible materials. It aims to provide a long-term solution to patients with advanced biventricular HF and potentially terminal acute myocardial infarctions (MI) for whom no human transplant or remaining treatment possibilities are available.
70% survival rate at six months post implant
Carmat announced the first cohort of 10 patients surgically implanted with the device achieved a six-month survival rate of 70% compared to a six-month survival rate of 50% in the feasibility study in four patients. Importantly, there were no incidents of stroke, gastrointestinal bleeding or driveline infection (common adverse events with similar devices). To date, 11 patients have been implanted with the device and, according to the company, data from all 20 patients in both cohorts should be sufficient to obtain CE marking (submission expected in H120).
Process improvements caused delay
Starting in Q418, Carmat halted device production following analysis from over 20 years of cumulative operating between the clinical study and reliability bench tests that revealed areas for improvement to the manufacturing process. Production has been resumed and the company expects to begin implanting the device in Q319.
Valuation: €856.0m or €68.01 per share
We have adjusted our valuation to €856.0m or €68.01 per share from €773m or €83.89 per share. The increase in total value is primarily attributed to rolling forward our NPVs and higher net cash due to an offering. The per-share value fell due to the dilution associated with the offering. We have also delayed our expectations for enrolment completion to H120 (previously mid-2019) and launch in Europe by the end of 2020 (previously H120) due to a temporary halt in device manufacturing.
Investment summary
Company description: Bioprosthetic heart development
Carmat is developing the first biocompatible, biventricular mechanical heart intended to treat end-stage HF and potentially terminal MI and to help compensate for the significant shortfall in the availability of human donor hearts. Carmat completed a four-patient feasibility study (first implantation in late 2013) in January 2016 and began an EU pivotal trial in July 2016, which was suspended in late November 2016 after the death of the first patient (implanted with the device in August 2016). The study was resumed in May 2017 following analyses that made it clear the cause of death was not due to device malfunction. As of September 2018, 11 patients have been surgically implanted with the Carmat total artificial heart (TAH). However, in Q418 production was halted following analysis that exposed potential room for improvement specifically regarding device manufacturing. Production has since been resumed and the company expects to begin implanting the revamped device in Q319.
Valuation: €856.0m or €68.01 per share
We have adjusted our valuation to €856.0m or €68.01 per share from €773m or €83.89 per share. The increase in total value is primarily attributed to rolling forward our NPVs and higher net cash due to an offering. The per-share value fell due to the dilution associated with the offering. We have also delayed our expectations for enrolment completion to H120 (previously mid-2019) and launch in Europe by the end of 2020 (previously H120) due to a temporary halt in device manufacturing.
Financials: €24.0m post-tax loss in H119
Carmat’s H119 post-tax loss was €24.0m, up roughly 27% from H118 (€18.9m) and is attributable to increases in both R&D expenditure and SG&A as the company continues product development. We expect this increased R&D spending to persist with €33.5m in R&D expenses in 2019 and an additional €30m in 2020. The company had €15.7m in cash and equivalents and €15.5m in debt at 30 June 2019. In December 2018, Carmat engaged in a €30m non-dilutive loan agreement with the European Investment Bank (EIB). Carmat drew down the first of three available tranches of €10m in January 2019 and has an additional €20m remaining under the facility. Subsequent to the quarter, Carmat raised an additional €60m in gross proceeds in an equity offering. Following this raise, we assume an additional financing requirement of €40m (half of which can be covered by the EIB facility). As per our usual methodology, we assign these additional financings to long-term debt.
Sensitivities: Building confidence in the Carmat heart
The main investment sensitivity and driver would be the upcoming results from the Carmat pivotal trial, including safety and tolerability. The company achieved a six-month survival rate of 70% in the first cohort of 10 patients surgically implanted with the device. New implantations were placed on hold for manufacturing improvements and are expected to resume in Q319. Any residual concerns involving battery safety, hemocompatibility and mechanics efficacy will need to be demonstrated throughout the second half of the pivotal trial for commercial approval. Risks inherent in the Carmat device include the range of different technologies, including biomaterials, micro-mechanics and self-regulatory electronics, which add to the complexity of the device. If the Carmat heart obtains regulatory approval, the company will need to build confidence among physicians for them to select the Carmat implant over alternative mechanical circulatory support devices such as the SynCardia TAH device and ventricular assistance devices (VADs). Carmat will require further funding to support the completion of the pivotal trial and commercial launch in CE mark territories. We highlight that our commercialisation timing forecast assumes near-perfect execution.
The first biocompatible, biventricular mechanical heart
Carmat was founded in 2008 as the by-product of a collaboration between Matra Défense (now part of Airbus Group), a technology company, and Professor Alain Carpentier, a cardiac surgeon. Carmat is developing an innovative, biventricular bioprosthetic heart, differentiated from previous generations of artificial hearts to make the prosthesis function as closely as possible to a human heart, by applying technology involving self-regulatory mechanisms and biocompatible materials. Carmat’s bioprosthetic heart aims to provide a long-term (permanent) therapeutic solution, or destination therapy (DT), to patients suffering from advanced biventricular HF or acute MI (commonly referred to as heart attack) for whom no human transplant is available and who have exhausted all remaining treatment possibilities.
After the completion of a four-patient feasibility trial in early 2016, Carmat started a 20-patient CE mark-enabling pivotal trial in July 2016, for which enrolment was paused in November 2016 after the death of the first recruited patient. The study was resumed in May 2017 following analyses that made it clear the cause of death was related to poor battery handling by the patient and not due to device malfunction. In July 2018, the device was successfully surgically implanted into the 10th patient, marking the completion of the first half of the CE mark-enabling trial. As of September 2018, 11 patients were surgically implanted with the Carmat TAH. However, towards the end of 2018, production of the device was halted following analysis that exposed potential room for improvement specifically regarding device integrity and cleanliness of the technical component. Production has since been resumed and the company expects to re-start implantations in Q319. We have therefore delayed our expectations for enrolment completion to H120 (previously mid-2019) and launch in Europe by the end of 2020 (previously H120) to meet the significant clinical need for a permanent alternative for the patients on heart transplant waiting lists. The lack of any accepted biventricular implant for permanent use (DT) in either the EU or the US is potentially a significant commercial advantage.
Carmat TAH and the cardiac transplant indication
The Carmat artificial heart is being developed as a permanent replacement, or DT, for biventricular HF or acute MI patients who do not have access to viable or compatible human donor hearts. It is designed to replicate as closely as possible the functionality and morphology of the human heart. Its key potential benefits compared to prior artificial implantable hearts are:
■
It is intended to function similarly to a natural heart and as a self-regulating device, thus delivering blood flow at differing rates in response to differing physiological needs. This is unlike earlier mechanical hearts, such as the SynCardia TAH, which focused on restoring blood flow and tended to pump blood at a fixed rate.
■
The surfaces of the prosthesis (the only portions that come into contact with the blood) are made from biocompatible, haemocompatible and non-thrombogenic materials (bovine pericardium treated to become chemically inert and medical-grade ePTFE), which allow continuous proteinic surface coverage, thereby reducing the risk of complications including blood clotting or haemolysis. Carmat suggests the metals and polymers used in existing competitors’ mechanical circulatory support (MCS) products generally do not cause continuous surface proteinisation (as do the Carmat surface materials described above) and are thus thrombogenic (ie activate the coagulation system and therefore create blood clots, which can either block a pump/motor or migrate elsewhere and cause strokes or pulmonary embolisms).
The prosthesis consists of a prosthetic heart and an electrical connection to a power supply (battery). The device reproduces the operation of the natural heart by using hydraulic actuation, with an activation liquid used as an intermediary to push the blood. The natural cardiac rhythm consists of two periods: systole (where blood is pumped out of the heart) and diastole (when the cardiac ventricles fill up with blood) (Exhibit 1).
Exhibit 1: The Carmat TAH |
Source: Carmat |
Like the human heart, the Carmat TAH comprises two ventricular cavities separated into two volumes (one for blood, one for the actuation liquid) by a flexible hybrid membrane. This membrane aims to reproduce the viscoelastic nature of cardiac muscle and pumps blood as it contracts. A motor-pump group (consisting of two miniature pumps) moves the actuation liquid to the ventricles, thus generating systole or, by working in reverse, it moves this fluid towards an external pouch (reproducing diastole). This mechanism aims to provide a pulsatile heartbeat to eject and admit blood and mimic the natural flow action of the heart.
The blood flow rate is guided by an internal electronic device and microprocessor (autoregulation), controlled by embedded sensors and software algorithms that adjust flow rate in accordance with the patient’s physiological and metabolic requirements, varying both the rate and the strength at which the blood is pumped. The heart is accompanied by a portable system (weighing 3kg) with lithium-ion batteries, which provides over five hours of independent operation (the patient can carry additional batteries to extend independent utility or can also connect to a power outlet). Carmat is also working on implementing a second-generation power supply developed by PaxiTech that employs fuel cell technology and can potentially provide more than 12 hours of independence while weighing below 3kg. We believe the company intends to introduce this power source after the Carmat TAH obtains the CE mark.
Manufacturing and sourcing initiatives
Carmat sources most of the many individual components of the heart from external suppliers and assembles them onsite. It has an agreement with Edwards Lifesciences to use Carpentier-Edwards valves. In late August 2018, Carmat announced the certification of its new manufacturing site located in Bois-d’Archy, France. According to the company, the automated manufacturing site will enable the production of up to 800 Carmat biventricular TAH units per year at full capacity, which should support the recent enrolment ramp-up for the pivotal trial and meet the demands of industrial manufacturing. The company announced the transfer of all production from the Vélizy site to the Bois-d’Archy site in May 2019.
HF and MI explained
MI occurs when there is insufficient blood flow to a portion of the heart, leading to damage and cellular death to the affected heart muscle. The US Centers for Disease Control and Prevention (CDC) estimates that 735,000 Americans have an MI each year1 and the European Heart Network estimated there are at least 750,000 cases in the EU (>2.0 per 1,000 for males and >1.0 per 1,000 for females in a 500 million population).2
CDC.
European Heart Network.
HF occurs when the myocardium (cardiac muscle) is not fully capable of performing its essential function as a blood ‘pump’ to provide a sufficient cardiac output and oxygenated blood to meet an individual’s metabolic needs. HF can occur in either the left (targeting the primary systemic circulation) or right ventricle (which pumps bloods to the lungs and pulmonary circulation), or in both (biventricular HF). HF is primarily caused by coronary disease or other cardiovascular disorders including hypertension3 and is one of the possible consequences of an MI.
High blood pressure causes an increased workload to cardiac muscle, leading to enlargement and less efficient functioning.
HF is estimated to affect nearly 38 million people worldwide, including 20 million across the US and Europe, and total medical care costs in the US have been forecast to rise from $21bn in 2012 to over $53bn by 2030.4 The US CDC estimates that 5.7 million Americans have HF. HF can be classified using the New York Heart Association (NYHA) scale, show below.
Ziaeian B, Fonarow GC. Nat Rev Cardiol. 2016 Jun;13(6):368-78.
Exhibit 2: NYHA heart failure grading system
Class I |
Class II |
Class III |
Class IV |
|
Symptoms |
No symptoms |
Tiredness, palpitations, |
Symptoms or discomfort on |
Symptomatic even at rest |
Activity |
No limitations |
Modest limitations |
Marked reduction |
Inability to perform nearly all |
Source: Company reports
Current treatment approaches for HF
Class I and II stages of HF are often treated with medical therapy (including anticoagulant/anti-platelet aggregation medications, angiotensin-converting enzyme inhibitors, beta-blockers, diuretics, etc).
Moving between Class II and Class III marks a significant shift for a patient’s quality of life, reflecting the transition between a virtually normal life and one with considerably reduced activity, potentially involving a loss of independence. Class IV patients represent about 2.3% of heart failures5 or approximately 500,000 people in the US and EU.
Jhund PS et al. Circulation. 2009; 119:515-523
Starting in Class III, surgical options and the implantation of supportive medical devices are considered, such as mono or biventricular pacemakers, implantable defibrillators, intra-aortic balloon pump procedures and stents. Generally, these procedures can restore or increase blood flow to the heart or slow the progression of the disease. Further advanced cases (within Class III or Class IV) can be treated with human heart transplantation (HHT) or with MCS devices.
HHT traditionally reserved for most severe cases but supply very limited
For the most severe cases of HF, HHT is considered the best treatment for patients who are refractory to management with medical therapy, or less invasive options. The cost of HHT, including surgery, assessment, admission costs, medication (eg immunosuppressants) and postoperative care has been estimated by the Milliman Risk Institute at over $1m per patient.6 Extensive screening is done before surgery to exclude those with comorbidities that can increase complications to ensure the patients chosen for surgery are those who are likely to survive the longest after transplantation.
Milliman Report 2014 - Table 2: Estimated US Average 2014 Billed Charges Per Transplant. www.milliman.com/uploadedFiles/insight/Research/health-rr/1938HDP_20141230.pdf
It has been estimated that up to 20,000 US patients could benefit from heart transplantation each year,7 but due to a shortage in donor organs, recently (2013–2017) only about 2,500–3,200 heart transplants have been performed each year in the US.8 These are generally reserved for younger candidates with fewer comorbidities. A similar supply/demand mismatch exists in Europe (across France, Germany and the UK there are only c 900 transplants per year).9 Reasons for the shortage of donors relative to need include the strict criteria for donors (eg aged under 61 years, not suffering from certain infectious diseases, etc) and a reduction in motor vehicle accident-related fatality rates. The limited number of human hearts available provides a need for alternative approaches including MCS devices to restore cardiac function to maximise advanced HF patient survival.
Khush KK, Zaroff JG, Nguyen J, et al. Am J Transplant. 2015 Mar;15(3):642-9. doi: 10.1111/ajt.13055. Epub 2015 Feb 10.
Statistics from Organ Procurement and Transplantation Network.
Using statistics from Agence de la biomédecine (for France), Eurotransplant (for Germany), and NHS Organ Donation and Transplantation (for UK).
Types of MCS device and usage patterns
MCS devices considered for patients with advanced HF include VADs and TAHs. The type of device used depends on the stage of HF and whether one or both ventricles are affected. Currently, the SynCardia device is the only approved TAH on the market (the Carmat heart, if approved, could be the second). VADs, also called implantable heart pumps, are more commonly used than TAHs, are implanted in parallel to the native heart and assist the existing ventricles to pump blood, reducing the cardiac workload in patients with HF. The left ventricle (which pumps oxygenated blood to the general circulation) is most susceptible to failure as it has around five times the pressure workload of the right. Left VADs (LVADs) are most commonly implanted in patients with monoventricular failure.
MCS device use categories: BTR, BTT, DT
The intended usage of MCS devices falls into three categories: bridge-to-recovery (BTR), bridge-to-transplantation (BTT) or DT. BTR refers to scenarios when the HF scenario is temporary (eg a recovery from heart surgery) and an MCS can be implanted for a few weeks or months to assist the heart during its recovery period. In the vast majority of advanced HF cases where MCS is indicated, the damage is permanent and BTT or DT treatment may be required.
BTT (or pending transplantation) refers to the intent to implant the device temporarily until an organ transplant is available, or until the patient’s condition improves sufficiently to tolerate such surgery. The patient’s MCS device (often an LVAD) may remain in place for several years until a donor heart becomes available for transplant. DT refers to the MCS being implanted permanently or for patients who are not expected to be eligible for or compatible with a heart transplant. DT aims for an improvement of at least two classes on the NYHA scale.10
The NYHA Functional Classification scale for HF places patients in four symptomatic (I-IV) and objective categories (A-D) classifying the severity of the disease.
VAD technology improvements have improved its market penetration
VAD technology has also improved since the first VADs were approved in the 1970s, leading to devices with improved flow rates and with lower risks of infection and thrombosis, making them capable of long-term or permanent circulatory support (rather than BTRs, which were the first-generation VADs). Frazier et al.11 showed a 34% increase in survival to transplantation in patients supported with a VAD compared with those treated with medical therapy and several other studies suggest LVADs provide excellent outcomes in advanced HF compared to medical therapy.12 Between 2007 and 2013, the number of LVADs implanted in the US alone each year increased over 600% to more than 2,500.
Frazier OH, Rose EA, Oz MC et al. J. Thorac. Cardiovasc. Surg. 122, 1186–1195 (2001).
Makdisi G,Makdisi PB, Bittner HB et al. J Thorac Dis. 2017 Apr; 9(4): 932–935.
The primary MCS devices for advanced HF patients are intracorporeal13 VADs (ie those placed using highly invasive open-heart surgery with implantation in the surgery suite) or TAHs. The market intracorporeal LVAD leader are Abbott’s (ABT: NYSE) HeartMate line14 and Medtronic’s HVAD line.15
Intracorporeal VADs differ from percutaneous VADs; percutaneous assist devices (examples include Abiomed’s Impella line and Abbott’s HeartMate PHP) are inserted through a small skin puncture in the leg and allow for much less invasive placement and removal. However, percutaneous VADs are indicated to support the heart for short period (ie generally up to six days) for high-risk percutaneous coronary intervention (PCI) procedures and thus are not indicated as potential long-term treatments for HF.
Abbott recently acquired St. Jude Medical, which obtained the HeartMate line as part of its 2015 acquisition of Thoratec Corp.
Medtronic acquired HeartWave International, in August 2016; HeartWave developed and marketed the HVAD System at the time of the acquisition.
Thrombosis and right ventricle heart failure a risk in LVADs
LVADs are not without risk, as up to c 20% of patients implanted with them generate right ventricle HF (RVHF)16 and may require right VADs (RVADs), although the more recent LVADs, such as those with continuous flow (eg HeartMate II) or centrifugal/electromagnetic-based mechanisms (eg HeartMate III or HVAD), have lower RVHF and mechanical failure risks. Although RVADs can be used in some cases in RHVF, in addition to surgical risks there can be technical complications involved with using non-integrated VADs for different ventricles (such as ensuring proper synchronisation or communication between sides, as imbalance of flows can lead to thrombosis or pulmonary oedema). In cases of biventricular failure, biventricular VADs (BiVADs)17 can be implanted (in place of LVAD or RVAD) to provide biventricular support, but these are more complicated to manage18 and are less commonly used than LVADs (a TAH or HHT could be more suitable for such patients).
Argiriou M, Kolokotron SM, Sakellaridis T, et al. J Thorac Dis. 2014 Mar; 6(Suppl 1): S52–S59.
Abbott/Thoratec’s pVAD, an earlier-generation VAD, offers biventricular support.
Sen A, Larson JS, Kashani KB. Crit Care. 2016 Jun 25;20(1):153.
VADs have also been associated with an increased risk of thrombosis and blood clots and patients generally require long-term anticoagulant therapy. In August 2015, the FDA issued an alert indicating the HeartMate II was associated with an increased rate of pump thrombosis (blood clots inside the pump) and that patients implanted with the HeartWare HVAD had a markedly higher rate of stroke (>28%) at two years than those implanted with the HeartMate II (<13%) in the ENDURANCE study (n=446). In the case of the HeartMate II, some studies have shown a higher pump thrombosis rate than was observed in the pivotal studies conducted to support its approval in BTT and DT in 2008 and 2010, respectively. Starling et al. (2013) found a pump thrombosis rate of 8.4% at three months and Kirklin et al. (2014) found a 6% rate at six months; this compares to a 1.6% rate at one year in the BTT clinical trial and 3.8% at two years during the DT clinical trial. The US FDA approved Abbot’s HeartMate 3 for BTT (August 2017) and DT (October 2018), a follow-on device to the HeartMate II, designed to lower thrombosis risk. The device previously received a CE mark in late 2015. Two-year analysis from the MOMENTUM 3 trial (n=366), which compared the HeartMate 3 (n=190) with the HeartMate II (n=176) in patients with advanced HF confirmed both noninferiority and superiority of the follow-on device.19 Rates of reoperation or device malfunction were significantly lower in the HeartMate 3 group compared to the HeartMate II group. Moreover, there were significantly fewer (p<0.001) suspected events of pump thrombosis in patients who received the Heartmate 3 device (two patients) versus patients who received the HeartMate II device (27 patients).
Mehra, M. R., et al. (2018). Two-Year Outcomes with a Magnetically Levitated Cardiac Pump in Heart Failure. New England Journal of Medicine, 378(15), 1386–1395.
The Carmat TAH could have a potential advantage versus existing VADs in reducing thrombosis risk given it was designed such that only biocompatible or bioinert materials would come into contact with a patient’s blood, to reduce thromboembolic risks.
TAH, such as SynCardia, are potential alternatives in biventricular failure
TAH are alternative MCS treatments to VADs for advanced HF patients. A TAH comprises two ventricular volumes and is designed to fully replace an existing heart. The only TAH on market in the US and Europe is produced by privately held SynCardia. TAH are more likely to be employed in cases of biventricular failure (where LVADs would not be sufficient as both ventricles require support) and could be used instead of BiVADs. Although no head-to-head randomised trials have compared the SynCardia TAH to BiVAD mechanical circulatory support, one retrospective study (Kirsch et al, 2012; n=383, including 90 TAH implants) showed no difference in mortality for patients implanted with a TAH compared with BiVADs.20 It also found that TAH patients had a substantially reduced rate of stroke and that among patients who experienced prolonged support (≥90 days), those with the SynCardia TAH showed a trend towards improved survival. Kirsch et al. (2013)21 reviewed data on the 90 SynCardia TAH implantations performed at Hôpital Universitaire de la Pitié Salpêtrière (Paris, France) between 2000 and 2010 with a BTT intent on patients with cardiogenic shock22 and determined that actuarial survival on the device was 74% ± 5%, 63% ± 6% and 47% ± 8% at 30, 60 and 180 days after implantation.
Kirsch M, Mazzucotelli JP, Roussel JC, et al. J Heart Lung Transplant 2012;31:501–8.
Kirsch ME, Nguyen A, Mastroianni C, et al. Ann Thorac Surg. 2013 May;95(5):1640–6.
A potentially fatal condition, often resulting from a severe MI that leads to sustained low blood pressure and poor tissue perfusion. The overall in-hospital mortality rate was estimated by Kolte et al. (2014) at approximately 40%.
Being reserved for more severe cases, TAH use rates well below those of LVADs
Although well over 28,000 HeartMate II LVAD implants have been performed worldwide since its launch, only about 1,700 SynCardia TAH implantations have been made thus far, in part due to the more invasive nature of TAH implantation compared to an LVAD (or even BiVAD) and a higher rate of surgical complications such as infections with the SynCardia TAH. Further, although a DT study is underway, the SynCardia device is only approved for BTT in the EU and US. The design of the SynCardia TAH is over 40 years old and, like some earlier-generation LVADs, its mechanics are driven by pneumatic (compressed air) actuation and as such it requires the constant use of an external compressor or driver (weighing about 6kg), which is itself powered electrically with lithium-ion batteries. Newer-generation LVADs use continuous flow or centrifugal/magnetic designs for pumping blood, and the Carmat device uses hydraulic actuation; these approaches use power sources that involve fewer mobility encumbrances than the external driver required by the SynCardia device. With its biocompatible materials, a more convenient power source and a differing mechanical approach, the Carmat heart has several potential advantages over the SynCardia device.
Exhibit 3: Comparison of selected mechanically assisted circulatory support devices
Device |
Manufacturer |
Approval status |
Characteristics |
Data |
Artificial heart |
||||
Bioprosthetic artificial heart |
Carmat |
In-development (trial temporarily halted during Q418 and expected to re-start in Q319) |
Self-regulating electro-hydraulic pulsatile flow contained within the body; uses external (lithium ion, potentially fuel-cell in 2nd gen) batteries. Algorithms mimic reactions of cardiac muscle to BP and postural changes. All blood-facing surfaces biocompatible. |
EU feasibility study successful (n=4) with 75% surviving more than 30 days; CE mark study ongoing. Interim analysis of first cohort (n=10) achieved a survival rate of 70% at 180 days post-implant. Enrolment completion of second cohort expected H120. |
Total artificial heart |
SynCardia |
Approved: BTT in US/Europe; |
Ventricles adjust to increase blood flow during exercise; blood can contact non-biocompatible surfaces including Medtronic Hall valves (titanium and pyrolytic carbon); anticoagulant therapy required post-implantation. Offered in two sizes. |
In BTT pivotal study, one-year survival (n=81) was 70% vs 31% in control arm (n=35). Over 1,700 implants since approval. |
Intracorporeal ventricular assistance devices |
||||
HeartMate II LVAS |
Thoratec (Abbott) |
Approved for BTR/BTT and DT in US and Europe |
LVAD with continuous flow, rotary pumps with axial flow, generating a net pressure rise across the pump. External system driver connected by a percutaneous lead to lithium battery providing up to 10 hours of autonomy. Requires anticoagulant therapy. |
In a DT pivotal study (n=200) in advanced HF patients, survival of 68% and 58% at one and two years, respectively. Over 28,000 implants worldwide. |
HVAD |
HeartWare (Medtronic) |
Approved for BTR/BTT in US and Europe; approved for DT in US. |
Centrifugal LVAD that uses hydrodynamic and magnetic forces to hold the impeller in place. The impeller spins at high speed to create suction that pulls blood into the pump, changes its flow direction and then pushes it out of the pump. Portable power unit with |
Non-inferiority vs Heartmate II shown in the DT ENDURANCE study (n=446), where HVAD arm had 55.0% stroke-free survival at two years, vs 57.4% for HeartMate II. |
HeartMate 3 LVAS |
Thoratec (Abbott) |
Approved for BTR/BTT and DT in US and Europe. |
Centrifugal LVAD that uses magnetic levitation to hold the impeller in place. Designed to reduce thrombosis risk (vs HeartMate II). |
CE mark-enabling study (n=50) had 92% six-month survival and 80% one-year survival; with no occurrences of pump thrombosis, malfunctions, or haemolysis. MOMENTUM 3 (n=366) US IDE study showed non-inferiority and superiority vs HeartMate II. Suspected events of pump thrombosis occurred less in HeartMate 3 vs HeartMate II. |
Source: Edison Investment Research, Medscape, company reports. Notes: DT: destination therapy; BTT: bridge-to-transplant; BTR: bridge-to-recovery; HUD: humanitarian use device; BP: blood pressure; IDE: investigational device exemption.
Bivacor, Cleveland Heart developing TAH competitors
Besides Carmat, at least two private companies have their own proprietary TAH devices in the development stages. Texas-based Bivacor is developing a centrifugal rotary pump-based TAH using a single moving part that applies magnetic levitation (to reduce mechanical wear). Like the Carmat device, it adapts the pump’s operation to changes in activity levels. Similar to the Bivacor device, Cleveland Heart’s SmartHeart TAH uses centrifugal pumps with a single moving part. Both the Bivacor and Cleveland Heart devices are in preclinical testing (having both been implanted in calves) and they are both intended to be small enough to be implantable in men, women and children.
INTERMACS classification system for MCS implantations
The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) register, established in 2009, intends to record all MCS implantations in North America across over 150 participating hospitals. INTERMACS uses a seven-level profile scale reflecting the general clinical description of patients receiving MCS devices (generally primary LVAD or TAH implants).
Exhibit 4: INTERMACS classification system
NYHA c lass |
Profile level |
Patient profile |
Description |
IV |
1 |
Critical cardiogenic shock |
Life-threatening hypotension |
2 |
Progressive decline |
Dependent on ionotropic support but shows signs of deterioration (eg renal function, fluid retention, etc.) |
|
3 |
Stable but inotrope dependent |
Stable but requires mild-moderate doses of intravenous inotropes (or has temporary circulatory support device) |
|
4 |
Resting symptoms |
Patient who is at home on oral therapy but has frequent symptoms of congestion at rest or with daily living activities |
|
5 |
Exertion Intolerant |
Patient who is comfortable at rest but unable to engage in activities, and is predominantly housebound |
|
6 |
Exertion Limited |
Patient can do some mild activity |
|
III |
7 |
Advanced NYHA Class III |
Patient is clinically stable with reasonable level of comfortable activity (ie can walk for more than a block) |
Source: INTERMACS
Generally, patients with mono-ventricular failure in profile 3 and above are more likely to be implanted with a VAD than a TAH. We estimate that c 33% of the c 500,000 US and EU Class IV HF patients have biventricular failure and could be potential candidates for a BiVAD or a TAH (if approved for DT and if all other eligibility criteria are met). Because the SynCardia TAH is not yet approved for DT, we believe it is primarily oriented as a BTT for patients with biventricular failure in INTERMACS profile 1 or 2 where no human donor heart is available.
Carmat’s biventricular TAH
Carmat believes its self-regulating artificial biventricular heart could be a superior solution in biventricular failure to BiVADs or LVAD/RVAD combinations, as it can synchronise the pumps between the pulmonary (right ventricle) and the systemic (left ventricle) circulation and instantaneously respond to physiological changes in demand.
Four-patient feasibility study completed in early 2016
Carmat began a four-patient feasibility study in late 2013 and completed it in January 2016 (Exhibit 5). There was a minor delay in recruitment after the second patient’s death in May 2015 as, despite a survival of nine months post-implantation, the device malfunctioned due to a disruption in micro-steering electronics following a micro-leak. Carmat identified the cause of the malfunctioning and then worked on corrective measures, including a software-tool predicting malfunctions and received authorisation from French regulators to resume the study in November 2015. Carmat’s analyses confirmed the malfunctioning was not due to an inherent flaw in product design, but rather issues stemming from the early stage of production. The death of the fourth patient was due to complications unrelated to the performance of the bioprosthetic heart itself.
Exhibit 5: Summary of feasibility study outcomes for Carmat bioprosthetic heart
Patient |
Centre |
Date of implantation |
Primary outcomes |
1 |
Hôpital Européen Georges-Pompidou |
18 December 2013 |
Patient survived until March 2014 (75 days). An electrical component fault caused the bioprosthesis to malfunction. Approval to resume study from the French National Agency for Medicines and Health Products Safety (ANSM). |
2 |
Nantes University Hospital |
5 August 2014 |
Patient survived nine months, of which four months were at home. Malfunction caused by a fault with steering motors led to circulatory insufficiency and hospitalisation on 1 May 2014. Patient re-implanted the next day but died later that same day due to multiple organ failure. |
3 |
Hôpital Européen Georges-Pompidou |
8 April 2015 |
Patient discharged in September 2015 but was hospitalised in November 2015 and died due to respiratory failure following a chronic renal failure. |
4 |
Hôpital Universitaire de La Pitié Salpêtrière |
22 December 2015 |
Patient died on 11 January 2016, but the severity of their underlying condition (patient had biventricular heart failure and required continuous life support) was deemed responsible for their death. The Carmat heart was believed to have functioned optimally during implantation. |
Source: Company reports
Current CE-mark enabling pivotal study
Carmat began a 20-patient EU pivotal trial in July 2016, for which recruitment was suspended by French National Agency for Medicines and Health Products Safety (ANSM) in late November 2016 after the death of the study’s first enrolled patient (implanted in August 2016) in October 2016. Carmat maintained the death was not due to a prosthesis malfunction, but to poor handling of the batteries by the patient. Following a favourable review by ANSM of the actions and analyses taken by Carmat on this incident, ANSM permitted resumption of the trial in May 2017.
In January 2019, Carmat announced interim data from the initial 10 patients comprising the first cohort of its pivotal CE mark-enabling clinical trial investigating the surgical implantation of the Carmat TAH. According to the company, these patients achieved a survival rate of 70% at 180 days post-implant, compared to a six-month survival rate of 50% as demonstrated in the previous feasibility study in four patients. For comparison, 266 patients involved in the INTERMACS database who received the SynCardia TAH achieved a six-month survival rate of 62%.23 This interim analysis of the Carmat TAH supports the biocompatibility of the device and further demonstrates its positive safety profile. These patients were managed with light anticoagulant therapy and did not experience common adverse events such as cerebrovascular accidents, gastrointestinal bleeding or infections related to the percutaneous cable, all relatively common issues with competitive products (see Exhibit 6), which should help increase Carmat TAH adoption.
Arabía, F. A., Cantor, R. S., et al (2018). Interagency registry for mechanically assisted circulatory support report on the total artificial heart. The Journal of Heart and Lung Transplantation, 37(11), 1304-1312.
Exhibit 6: Comparative outcomes at 6 month follow up
Survival rate |
Stroke |
Gastrointestinal bleeding |
Driveline infection |
|
Carmat |
70% |
0% |
0% |
0% |
SynCardia |
54-62% |
23% |
20% |
22% |
BIVAD |
46-68% |
7% |
7% |
7% |
LVAD |
90-92% |
8% |
8% |
10% |
Source: Carmat. American Society of Artificial Internal Organs Annual Conference June 2019.
Enrolment for the second half of the trial, which began in September 2018 and is expected to include an additional 10 patients, is ongoing. Altogether, 11 patients have been surgically implanted with the Carmat TAH so far. According to the company, over 20 years of cumulative operating between the clinical study and reliability bench tests exposed areas for improvement to the manufacturing process specifically regarding device integrity and sterility of the technical component. Production, and therefore surgical implantations, of the Carmat TAH device was halted during Q418. Manufacturing has since resumed and the new prosthesis should be available for surgical transplant in Q319. Consequently, the company now expects to complete patient enrolment in H120. Carmat is validating additional clinical centres in another two countries to quickly complete its enrolment target.
According to the company, data from all 20 patients should be sufficient to achieve a CE mark for the Carmat TAH. The company has also stated that initial commercialisation efforts in the EU will be focused in Germany and France.
Market opportunity and commercial assumptions
The commercial case for the Carmat bioprosthetic heart primarily lies in a general lack of available human heart transplants for patients who need them. There are approximately 0.5 million people in the US and EU with Stage IV heart failure. We view the EU market as the primary opportunity for the Carmat heart, as the product is in EU pivotal studies and Carmat’s US regulatory strategy is still under evaluation, although in September the company announced FDA conditional approval to initiate a feasibility study in five patients in the US. We estimate the target potential market opportunity of the Carmat bioprosthetic heart can be broadly based as falling within two conditions. In both groups, the Carmat heart would only be implantable in patients under age 70, and for anatomical/size compatibility, in about 86% of otherwise eligible men and 14% of such women.
The two conditions we assume the Carmat bioprosthetic heart will be targeting are:
■
Patients with Class IV (end-stage) HF, with biventricular failure; estimated EU market size of about 21,500.
■
Patients suffering from acute MI whose severity or circumstances lead to an expected survival time of under 30 days with conventional management (estimated EU market size of about 60,300).
Altogether, we estimate the EU target treatment population is around 82,000. The American Heart Association assumes the US HF population will rise by c 2.1% pa over the next two decades24 and our model assumes a more conservative 1.5% growth rate.
Heidenreich PA, Albert NM, Allen LA, et al. Circ Heart Fail. 2013 May;6(3):606-19
Given the revision in our timing expectations for completing the pivotal trial, we now assume potential EU launch and commercialisation in Europe by year-end 2020 (previously H120). We continue to assume an initial average per-device market price of €160,000 (in the mid-range of company guidance of €140,000–180,000), rising 2% per year. We estimate that peak market share of 15% of this target market will be realised by 2024, with EU sales of €2.2bn in that year.
For the US market, our base case continues to assume that product introduction will be attempted using the humanitarian use device (HUD) programme, instead of a premarket approval (PMA) process. The HUD approach is less onerous and would shorten the amount of clinical data required for commercialisation but could limit product usage to 8,000 patients per year.25 Under an HUD approach, Carmat would need to provide safety evidence but, unlike a PMA, there would be no necessity to provide rigorous efficacy data. We consider the rationale under HUD would be an orphan subset of severe HF patients reflecting patients in INTERMACS categories 1 and 2, where an HHT is unavailable and where the patient would otherwise be expected to have only days to live.
In late 2016, the FDA increased the population estimate required to qualify under an HUD designation from ‘fewer than 4,000’ to ‘not more than 8,000’.
Exhibit 7: Comparison of PMA and HUD approval processes for medical devices
Process |
Size of clinical trial |
Clinical data requirements |
Timeline |
Addressable market |
PMA |
Over 100 patients |
Safety, efficacy |
Minimum two years |
50,000 |
HUD |
10–20 patients |
Safety, probable evidence of benefit vs risk |
180 days |
8,000 |
Source: Edison Investment Research
Given our expectation that a Humanitarian Device Exemption-enabling study would only start as the EU pivotal programme reaches its final stages, we maintain our US HUD commercialisation launch timeline to 2021 and a US market price of $200,000 per device. We assume peak US sales of $620m by 2025 under this scenario. If Carmat proceeds in the US via the PMA route instead of HUD, we will adjust our forecasts accordingly, but this would entail a higher risk adjustment, a later launch date due to the longer/larger study and additional R&D costs of over €35m (for a >100 patient clinical trial), although the addressable market would be larger. The current regulatory status in the US is that in September the FDA conditionally approved the company’s investigational device exemption application to initiate a US early feasibility study in five transplant-eligible patients. The company awaits institutional review board approval at the trial sites.
Sensitivities
Significant development risk remains within the Carmat TAH. Although the feasibility study and the first half of the CE mark-enabling trial were successful, the study was temporarily halted for the second time for improvements in manufacturing, which subsequently stopped surgical implantations. Manufacturing has since resumed and the new prosthesis should be available for surgical transplant in Q319. Any residual concerns involving battery safety, hemocompatibility and mechanical efficacy will need to be demonstrated through the second half of the pivotal clinical trial for commercial approval.
If the Carmat heart obtains the CE mark, the company will need to generate confidence among physicians and stakeholders for the product to be chosen to be deployed in advanced biventricular HF or at-risk MI patients. The SynCardia TAH device and VADs will be the most immediate competitors to the Carmat heart and we continue to anticipate that LVADs, in part due to requiring a less-invasive surgical procedure, will be the preferred device in most cases of Class IV monoventricular failure. In addition, for optimal sales penetration the company will need to obtain reimbursement coverage with applicable payers in the targeted markets. In the longer term, Carmat may need to compete with the emerging alternative TAH heart implants if they gain approval including, but not limited to, the Bivacor and Cleveland Heart product initiatives.
We do not expect Carmat to start generating sustainable, positive recurring operating cash flows until 2021, once Carmat sales and manufacturing efficiencies start to exceed all projected overhead costs. We highlight that our commercialisation timing forecast assumes near-perfect execution and timing by Carmat of the completing the remaining implantations of the pivotal trial and of CE mark filing requirements, as well as on obtaining an uncomplicated review by EU regulatory authorities.
Valuation
We have adjusted our valuation to €856.0m or €68.01 per share from €773m or €83.89 per share. The increase in total value is primarily attributed to rolling forward our NPVs and higher net cash due to a €60m equity offering. The per-share value fell due to the dilution associated with the offering. We have also delayed our expectations for enrolment completion to H120 (previously mid-2019) and launch in Europe by the end of 2020 (previously H120) due to a temporary halt in device manufacturing.
Exhibit 8: Valuation of Carmat
Product contributions (net of R&D and marketing costs) |
Indication |
Prob. of success |
Launch year |
Launch pricing |
Peak sales (€m) |
rNPV (€m) |
Carmat artificial heart in EU market |
Terminal heart failure and myocardial infarctions |
30% |
2020 |
€160,000 |
2,169 in 2024 |
1114.2 |
Carmat artificial heart in US market (under HUD) |
Terminal heart failure and myocardial infarctions |
20% |
2021 |
$200,000 |
620 in 2025 |
179.0 |
G&A expenses |
(94.1) |
|||||
Net capex, NWC & taxes |
(403.4) |
|||||
Total rNPV |
795.7 |
|||||
Net cash at 30 June 2019 + September offering |
60.3 |
|||||
Total firm value |
856.0 |
|||||
Total shares (m) |
12.6 |
|||||
Value per basic share (€) |
68.01 |
Source: Company reports, Edison Investment Research
Financials
Carmat’s H119 post-tax loss was €24.0m, up roughly 27% from H118 (€18.9m) and is attributable to increases in both R&D expenditure and SG&A as the company continues product development. We expect this increased R&D spending to continue with €33.5m in R&D expenses in 2019 and an additional €30m in 2020. The company had €15.7m in cash and equivalents and €15.5m in debt at 30 June 2019. In December 2018, Carmat engaged in a €30m non-dilutive loan agreement with the EIB. Carmat drew down the first of three available tranches of €10m in January 2019 and has an additional €20m remaining under the facility. It is important to note the drawdowns on the second and third tranches are conditional on technical and financial milestones including the successful execution of the clinical trial or the raising of additional funds. Subsequent to the quarter, Carmat raised an additional €60m in gross proceeds in an equity offering (3.16m shares at €19.00 per share). Following this raise, we assume an additional financing requirement of €40m (half of which can be covered by the EIB facility). As per our usual methodology, we assign these additional financings to long-term debt. We do not expect Carmat to start generating sustainable, positive, recurring operating cash flows until 2021, once its sales and manufacturing efficiencies start to exceed all projected overhead costs.
Exhibit 9: Financial summary
€000 |
2017 |
2018 |
2019e |
2020e |
||
Year end 31 December |
IFRS |
IFRS |
IFRS |
IFRS |
||
PROFIT & LOSS |
||||||
Revenue |
|
|
28 |
722 |
695 |
7,501 |
Cost of Sales |
0 |
0 |
0 |
(7,501) |
||
General & Administrative |
(8,421) |
(11,897) |
(14,707) |
(15,450) |
||
Research & Development |
(21,890) |
(30,672) |
(33,467) |
(30,000) |
||
EBITDA |
|
|
(30,283) |
(41,847) |
(47,478) |
(45,450) |
Depreciation |
(752) |
(920) |
(1,198) |
(3,671) |
||
Amortization |
0 |
0 |
0 |
0 |
||
Operating Profit (before amort. and except.) |
(31,035) |
(42,766) |
(48,676) |
(49,120) |
||
Exceptionals |
(56) |
(2) |
(9) |
(9) |
||
Other |
0 |
0 |
0 |
0 |
||
Operating Profit |
(31,090) |
(42,768) |
(48,685) |
(49,129) |
||
Net Interest |
(472) |
(945) |
(805) |
165 |
||
Profit Before Tax (norm) |
|
|
(31,507) |
(43,711) |
(49,481) |
(48,956) |
Profit Before Tax (FRS 3) |
|
|
(31,563) |
(43,713) |
(49,490) |
(48,965) |
Tax |
2,335 |
1,984 |
533 |
533 |
||
Profit After Tax and minority interests (norm) |
(29,172) |
(41,727) |
(48,948) |
(48,422) |
||
Profit After Tax and minority interests (FRS 3) |
(29,228) |
(41,729) |
(48,957) |
(48,431) |
||
Average Number of Shares Outstanding (m) |
9.0 |
9.2 |
11.0 |
12.6 |
||
EPS - normalised (€) |
|
|
(3.24) |
(4.54) |
(4.47) |
(3.84) |
EPS - normalised and fully diluted (€) |
|
(3.24) |
(4.54) |
(4.47) |
(3.84) |
|
EPS - (IFRS) (€) |
|
|
(3.24) |
(4.54) |
(4.47) |
(3.84) |
Dividend per share (€) |
0.0 |
0.0 |
0.0 |
0.0 |
||
BALANCE SHEET |
||||||
Fixed Assets |
|
|
4,752 |
6,139 |
15,373 |
26,702 |
Intangible Assets |
72 |
90 |
55 |
55 |
||
Tangible Assets |
4,680 |
6,049 |
15,319 |
26,648 |
||
Current Assets |
|
|
65,098 |
30,691 |
45,813 |
25,528 |
Short-term investments |
0 |
0 |
0 |
0 |
||
Cash |
60,723 |
25,302 |
41,353 |
21,009 |
||
Other |
4,375 |
5,389 |
4,460 |
4,520 |
||
Current Liabilities |
|
|
(7,944) |
(10,601) |
(9,782) |
(9,782) |
Creditors |
(7,944) |
(10,601) |
(9,782) |
(9,782) |
||
Short term borrowings |
0 |
0 |
0 |
0 |
||
Long Term Liabilities |
|
|
(3,714) |
(4,698) |
(15,468) |
(55,468) |
Long term borrowings |
(3,714) |
(4,698) |
(15,468) |
(55,468) |
||
Other long term liabilities |
0 |
0 |
0 |
0 |
||
Net Assets |
|
|
58,191 |
21,530 |
35,936 |
(13,019) |
CASH FLOW |
||||||
Operating Cash Flow |
|
|
(23,806) |
(37,229) |
(46,519) |
(45,510) |
Net Interest |
(472) |
(945) |
(805) |
165 |
||
Tax |
0 |
0 |
0 |
0 |
||
Capex |
(3,559) |
(2,293) |
(10,437) |
(15,000) |
||
Acquisitions/disposals |
0 |
0 |
0 |
0 |
||
Financing |
57,537 |
5,059 |
62,358 |
0 |
||
Net Cash Flow |
29,700 |
(35,408) |
4,596 |
(60,345) |
||
Opening net debt/(cash) |
|
|
(27,951) |
(57,009) |
(20,603) |
(25,885) |
HP finance leases initiated |
0 |
0 |
0 |
0 |
||
Other |
(642) |
(998) |
686 |
0 |
||
Closing net debt/(cash) |
|
|
(57,009) |
(20,603) |
(25,885) |
34,460 |
Source: Company reports, Edison Investment Research
|
|
|
Research: Investment Companies
Aberdeen Diversified Income and Growth Trust (ADIG) continues to progress towards full implementation of its strategy, aiming to achieve attractive capital and income returns (five-year target of Libor +5.5% pa, net of fees) from a widely diversified multi-asset portfolio, including a significant allocation to less liquid, longer-term investments that individual investors would be unlikely or unable to access. Portfolio managers Mike Brooks and Tony Foster at Aberdeen Standard Investments (ASI) say half the funds committed to these long-term strategies have now been drawn, with the pace of investment from underlying managers stepping up over the summer. As the portfolio develops towards its intended shape, there is scope for the current wider-than-average discount to narrow materially. ADIG’s shares have an FY19 prospective yield of 5.0%.
Get access to the very latest content matched to your personal investment style.