Preclinical rationale and current pathways to support the first human clinical trials in cardiac xenotransplantation

Abstract

Recent initiation of the first FDA‐approved cardiac xenotransplantation suggests xenotransplantation could soon become a therapeutic option for patients unable to undergo allotransplantation. Until xenotransplantation is widely applied in clinical practice, consideration of benefit versus risk and approaches to management of clinical xenografts will based at least in part on observations made in experimental xenotransplantation in non‐human primates. Indeed, the decision to proceed with clinical trials reflects significant progress in last few years in experimental solid organ and cellular xenotransplantation. Our laboratory at the NIH and now at University of Maryland contributed to this progress, with heterotopic cardiac xenografts surviving more than two years and life‐supporting cardiac xenografts survival up to 9 months. Here we describe our contributions to the understanding of the mechanism of cardiac xenograft rejection and development of methods to overcome past hurdles, and finally we share our opinion on the remaining barriers to clinical translation. We also discuss how the first in human xenotransplants might be performed, recipients managed, and graft function monitored.

1. Introduction:

During recent years, experimental xenografts have demonstrated long‐term, rejection free survival in preclinical models, something that was eagerly sought for almost 4 decades. Despite this progress, funding for xenotransplantation research from some sources waned, forcing many to leave the field. Still, the National Institute of Allergy and Infectious Diseases of the National Institutes of Health (NIAID/NIH supported some work as did Industry, and these sources helped some make significant advances. Some of the recent work in our laboratory both at the NIH and the University of Maryland School of Medicine has played a critical role in rejuvenating the field. A culmination of this work has resulted in the first FDA‐approved clinical application of cardiac xenotransplantation [1]. In this review we describe what we have learned in last 10 years in our laboratories at two separate institutions and why we are so optimistic about the transition of this life saving procedure to the clinic.

1.1. Current progress in cardiac xenotransplantation

Enduring survival of heterotopic cardiac xenotransplants:

Until 2012, the longest survival of pig cardiac xenograft in an abdominal heterotopic baboon model using a pig xenograft had been 179 days. We modified the immunosuppressive protocol in this model by introducing B‐cell depletion with anti CD20 monoclonal antibody (mAb) and blockade of co‐stimulation by CD40L (CD154) using anti‐CD154 (clone 5C8) mAb at 25 mg/kg. In addition, we used pigs knocked out for alpha 1–3 Galactosyl transferase and expressing human CD46 (GalKO.hCD46) as sources of hearts. The modifications and pig organ sources enabled us to achieve xenograft survival up to 236 days. The use of the CD40/CD40L pathway was demonstrated to be pivotal, due to its role in preventing germinal cell‐mediated elicited antibody formation from activated B‐cells after xenoantigen recognition in draining lymph nodes [2–4]. However, we observed thromboembolic complications we thought were caused by anti‐CD154, as this agent had caused similar complications in clinical trials (unpublished). To avert this problem we began to use anti‐CD40 (clone 2C10R4) mAb, which instead of anti‐CD154 to disrupt the CD40/CD40L pathway. However, our initial results with the dose used in the experiments with anti‐CD154 mAb (25 mg/kg) failed to provide long‐term survival. We therefore made two further changes – the dose of anti‐CD40 was doubled to 50 mg/kg and the pigs used as sources of xenografts were engineered as above but also expressed human thrombomodulin (hTBM) as a transgene. These changes brought dramatic improvement in outcomes, including the longest survival of an experimental heterotopic cardiac xenotransplant (945 days).

We next sought to determine the minimum dose of anti‐CD40 required to maintain graft survival. With a dose of 25 mg/kg a heterotopic xenograft survived 100 days after transplantation (n = 2). However, slowly thereafter all the xenografts underwent rejection. When we delayed the reduction in dose of anti‐CD40 until grafts had survived one year, we found we could decrease the dose to as little as 10 mg/kg without triggering rejection and graft loss. However, anti‐ CD40 was still needed as discontinuation led to failure of all xenografts within next 2 months, a time coinciding with washout of residual anti‐ CD40 mAb from the circulation (n = 3). After cessation of anti‐CD40, antibodies directed against antigens other than Gal were detected in serum of xenograft recipients and the xenograft tissues had histology and immunohistochemistry consistent with antibody mediated rejection (AMR). These results indicated that blockade of CD40‐CD40L pathway is essential to avert rejection and allow for long term survival of pig to non‐human primate xenografts.

Successful translation to orthotopic cardiac xenotransplantation:

Our success in heterotopic transplantation of porcine hearts in baboons indicated that rejection of a cardiac xenograft can be avoided for extended periods of time with the use of co‐stimulation based immunosuppression and minimal genetic modifications of the donor pig (1 carbohydrate knockout and expression of two human transgenes for complement regulation and thromboregulation). The next logical step was to test whether this regimen would enable orthotopic cardiac xenografts to sustain life.

Our initial efforts to translate our success from heterotopic to orthotopic cardiac xenografts was not successful [5]. Few hearts (n = 5) survived > 48 h and only one survived as long as 57 days. The orthotopic porcine heart transplants that failed within 48 h revealed no pathologic evidence of rejection and instead manifest primary graft dysfunction, termed perioperative cardiac xenograft dysfunction (PCXD). This phenomenon first described by McGregor, et al. and is characterized by progressive dysfunction without the evidence of intravascular thrombosis, interstitial hemorrhage, edema or antibody binding consistent with hyperacute xenograft rejection (n = 5) [6,7]. The Munich group led by Bruno Reichart had observed a similar condition in orthotopic cardiac xenografts, and Stig Steen overcame PCXD by applying extracorporeal perfusion with continuous perfusion (NICP, non‐ischemic cold perfusion) preservation, containing hyperoncotic perfusate with catecholamines, hormones and erythrocytes [8]. When PCXD was averted in this way however, orthotopic cardiac xenografts exhibited excessive post‐operative growth and physiologic dysfunction within 30 days [9]. Langin, et al. postulated physiologic mismatch led to aberrant growth and accordingly tested whether Temsirolimus (an inhibitor of the mechanistic target of Rapamycin, mTOR) could suppress this process [10]. Consistent with this hypothesis, they found that administration of temsirolimus with other pharmacological agents to reduce blood pressure and heart rate enabled orthotopic cardiac xenografts to function for up to 6 months [9]. Early steroid weaning was an additional strategy to reduce this growth after transplantation in this context. However, temsirolimus was found to be toxic and discontinuation led to excessive growth and failure of grafts.

We took a different approach to this problem. Source pigs were engineered as described above but with further modifications, including knockout of growth hormone receptor (GHRKO), hoping to control intrinsic graft growth after transplantation. Other genetic modifications included knockout of multiple carbohydrates and multiple human thromboregulatory, complement regulatory and anti‐ inflammatory proteins (Table 1). Hearts harvested from these pigs were conditioned using an XVIVO heart box and XVIVO heart solution (perfusate based on Steen solution) and then transplanted orthotopically into baboons treated with the anti‐CD40‐based regimen described above. These orthotopic xenografts functioned and survived up to 9 months the longest reported life‐supporting xenograft survival reported to date (reported at the 2021 IXA meeting) [11]. 7‐gene cardiac xenografts demonstrated the longest survival in a non‐human primate model.

Table 1.

Available Combinations of Donor Xenografts for Xenotransplantation.

Experiments	Gene Combinations	Explanation
NHP
3-gene	GTKO.CD46.TBM	Basic model
5-gene	GTKO.CD46.TBM.CD55.EPCR.CD39	Additional Complement and Coagulation regulatory proteins (CD55 and CD39)
6-gene	GTKO.CD46.CD55. EPCR. TFPI.CD47	Additional Coagulation regulatory protein(EPCR) and anti-inflammatory gene (CD47)
6-gene	GTKO.CD46.TBM.EPCR. CD47.HO1	Additional protection from oxidative stress by HO1 gene
6-gene	GTKO.CD46.TBM.A20.EPCR.CD55	Additional protection from oxidative stress by A20 gene
7-gene	GTKO.B4GALNT2KO.CD46.TBM.EPCR.CD47.HO1	Additional gene B4GALNT2 knock out for reducing antibody mediated rejection
8-gene	GTKO.B4GALNT2KO. CMAHKO.CD46.TBM.EPCR. CD47.HO1	Additional deletion of CMAH gene for reducing AMR
9-gene	GTKO.B4GALNT2KO. CMAHKO.GHRKO. CD46. TBM.EPCR.CD47.HO1	GHR gene knockout to control growth
Humans
10-gene	GTKO.B4GALNT2KO.CMAHKO. GHRKO. CD46.TBM.EPCR.CD47.HO1	Same as last NHP group but with additional knockout of CMAH gene to reduce antibody mediated rejection. Same 10-gene pig as used in NHP model

1.2. Our current understanding of cardiac xenograft rejection

Antibodies specific for porcine saccharides underlie a major immune obstacle:

The immune response to transplantation including xenotransplantation was initially (for the first 60 years) thought to consist entirely of antibodies. In the 1920–1930 s, some of these antibodies were identified as natural antibodies, (e.g. anti‐blood group antibodies in allotransplantation and other pre‐formed antibodies in xenotransplantation). Ab were shown to inflict injury via complement both in allotransplantation and in xenotransplantation and the most dramatic manifestation was hyperacute rejection (HAR). Until the mid 1980 s, xenografts between disparate species were thought unavoidably to undergo HAR and until 1990 HAR was considered the main impediment to the progression in the field of xenotransplantation. However, by 1990 it was demonstrated that HAR could be reliably averted by depleting Ab or inhibiting complement [12–14]. However, when early AMR/AVR was averted, grafts inevitably underwent AVR/AMR at later times, generally 2–3 weeks [15]. Elicited antibody responses to foreign proteins begin at 14–21 days and increase in amount and avidity thereafter, unless controlled by immunosuppression. Blockade of CD40‐CD40L is especially effective in exerting such control.

Genetic engineering of pigs to disrupt binding of antibodies, improve regulation of complement and hinder coagulation and thrombosis has the greatest impact on the early types of rejection, namely hyperacute and early antibody‐mediated rejection. Genetic changes include expression human thrombomodulin (TBM) [16–18], endothelial protein C receptor (EPCR) and complement regulatory proteins decay accelerating factor (DAF) [19] and CD46 [20]. Much experience indicates the pathogenic importance of these pathways in xenotransplantation [21]. The targets of natural antibody responses include recognition of three carbohydrate porcine antigens. Disruption of synthesis of these antigens on the porcine xenograft to varying degrees (by knocking out glycosyltransferases) helps prevent early rejection [22]. However, preventing synthesis of antigenic saccharides combined with expression of inhibitors of inflammation, coagulation and thrombosis cannot interrupt immunity elicited by porcine proteins which leads to much higher levels and avidity of Ab (unless deleted entirely). All pig proteins potentially evoke immunity. Highly immunogenic proteins expressed on the surface of pig endothelial cells, such as swine leukocyte antigen (SLA) encoded proteins are likely to have a role, if only minor [23].

The importance of selecting baboon recipients with low basal levels of anti non‐Gal antibodies (which can form against either carbohydrate or protein antigens) cannot be stressed enough—again, this helps prevent or limit early AMR and allows accommodation to occur [24,25]. This will likely be a strategy required for human transplantation as well, as this is akin to a PRA (panel reactive antibodies) in allotransplantation. Typically, our recipient baboons and pigs have been procured from a specific pathogen free facility, free of potential zoonotic infections (Table 3) and have low levels of anti‐non‐Gal antibodies. Baboons with higher levels of these antibodies to begin with rejected grafts at an earlier stage. The rejection, irrespective of the duration of survival, has been dominated by signs of an antibody‐mediated injury leading to microvascular thrombosis, myocyte damage and intravascular hemorrhage. Long term grafts that have been explanted at different timepoints demonstrate some evidence of patchy vasculopathy, similar to cardiac allograft vasculopathy commonly seen in human hearts. The etiology of this finding is still unclear, but allograft vasculopathy is largely antibody‐mediated.

Table 3.

Potential Infections Agents of Concern in Pigs.

Viruses that might affect human recipients	Additional viruses important for pig health
Herpes virus-gamma	Porcine circovirus (PCV)
Influenza virus, swine (SIV)	Porcine delta coronavirus (PDCoV)
Porcine cytomegalovirus (CMV)	Porcine epidemic diarrhea virus (PEDv)
Hepatitis E (HEV)	Porcine reproductive and respiratory virus (PRRS)
Porcine endogenous retrovirus (PERV)1	Transmissible gastroenteritis virus (TGE)
Bacteria that might affect human recipients
N/A	Additional bacteria important for pig health
Examination for Helminths	Mycoplasma hyopneumonia (M.hyo)
Parasitology

Courtesy: David Ayares, Revivicor Inc.

Factors contributing to xenograft rejection besides the antibodies to pig antigen:

Various incompatibilities exist between the pig and NHP/human that leads to activation of complement, coagulation and platelets [17]. Coagulation activation occurs with endothelial activation upon perfusion of xenogeneic blood on xenograft endothelium, leading to thrombin activation and microthrombus formation within the graft. Endogenous porcine thrombomodulin, expressed from porcine endothelial cells, is deficient in the ability to activate the anticoagulant protein C (APC) of recipients [17,18]. Expression of human thrombomodulin as a transgene in pigs has been shown to aid in the prevention of microvascular thrombosis associated with humoral xenograft rejection in pig‐to‐baboon xenotransplantation models (discussed further below) [17,18,26]. Due to these incompatibilities, the xenograft undergoes rapid injury and failure.

2. Methods to overcome cardiac xenograft rejection

2.1. Immunosuppression

Conventional immunosuppression used in allotransplantation has proven unsuccessful in prolonging rejection‐free xenograft survival for extended periods. Removing the preexisting antibodies by either immune adsorption or by using synthetic conjugates like galactose alpha 1–3 galactose trisaccharide polyethylene glycol conjugate (TPC) provides only limited extension of graft survival [27]. Blocking the CD40‐CD40L co‐stimulation pathway controls T cell‐dependent antibody production, enabling prolongation of xenograft survival. B‐ lymphocytes depletion during induction immunosuppression has also prolonged cardiac xenograft survival, but that treatment alone does not suffice [4,25,28]. However, depletion of T cells and B cells by induction at the time of transplantation and disruption of CD40‐ CD40L with anti‐CD40 thereafter allows maintenance immunosuppression with an anti‐metabolite (mycophenolate mofetil or MMF) to support long term survival and function of porcine organ xenografts in non‐human primates, as summarized in Table 2 [29].

Table 2.

Standard Immunosuppressive Regimen for Cardiac Xenotransplantation.

Induction
Agent	Dose	Timing	Route	Pre-Treatment	Purpose
Anti-CD20	19 mg/kg	Day −7, 0, 7	IV infusion	Solu-Medrol, Benadryl, H2 blocker	To deplete B-cells
ATG	5 mg/kg	Day −2, −1	IV infusion	Solu-Medrol, Benadryl, H2 blocker	To reduce number of T-cells
Anti-CD40 (clone 2C10R4)	50 mg/kg	Day −1 and 0	Slow IV infusion	None	Co-stimulation blockade. Suppression of both B- and T-cell response.
CVF or Berinert	50–100U/kg 17.5 mg/Kg	Day −1, 0 and 1	IV	None	To inhibit complement activity
Tocilizumab	8 mg/Kg	Day 0	IV	None	Anti-Inflammatory
Etanercept	0.7 mg/Kg	Day 0	SC	None	Anti-inflammatory
Maintenance
Anti-CD40 (clone 2C10R4)	50 mg/kg	Days 3,7,10,14, 19 then weekly	Slow IV infusion	None	Co-stimulation blockade. Suppression of both B-and T-cell response.
MMF	20 mg/kg/2hr	BID, daily	IV infusion	None	B and T cell suppression
Tocilizumab	8 mg/Kg	Weekly until day 90	IV	None	Anti-Inflammatory
Etanercept	0.7 mg/Kg	Weekly until day 90	SC	None	Anti-inflammatory
Solu-Medrol	2 mg/kg	BID tapered off in 7 weeks	IV	None	Suppress inflammation
Aspirin	40 mg		Oral	None	Prevent platelet aggregation
Heparin	50–400U/hr	Continuous	IV infusion	None	Maintain ACT 2X normal and prevent inflammation
Supportive
Ganciclovir	5 mg/kg/day	Daily	IV infusion		For CMV prophylaxis
Cefazolin	250 mg	Daily for 7 days and whenever needed	IV	None	Antibiotic cover
Epogen	200U/kg	Day −7 to 7 then weekly	IM or IV	None	To increase hematocrit

2.2. Genetic modification of donor pigs avoids rejection

Several genetic modifications are used to overcome rejection (Table 4), with specific genetic combinations used so far in experimental models listed in Table 1. The advent of xenograft genetic backgrounds with KO of the dominant carbohydrate epitope (GTKO) along with expression other anti‐inflammatory and complementary regulatory transgenes (human CD46, CD47, DAF, TBM), have significantly improved xenograft survival of kidney, lung and heart [30–32]. GTKO xenografts have avoided hyperacute rejection, but still AMR has been seen against other carbohydrate antigens (i.e. B4Gal and Neu5Gc antigen). Therefore, B4GalKO and CMAHKO along with GTKO animals (i.e., triple knockout or TKO swine) were developed and have showed decreased antibody binding in vitro [22,33]. More complex genetic constructs with human transgenes have mitigated early AMR/HAR and reduce a dominant antibody response toward the xenograft and lead to prolonged survival [30,34]. Genetic engineering to prevent synthesis of other carbohydrates has presented challenges in kidney transplantation and result in early graft failure [35]. We have transplanted cardiac xenografts with TKO backbones, hTBM and multiple combinations of complement regulatory proteins, thromboregulatory proteins and anti‐inflammatory genes with survival up to 264 days in a life‐supporting model of cardiac xenotransplantation [11]. Specific “multi‐gene” constructs such as these appear to provide durable, long‐term xenograft function.

Table 4.

Genetic Modifications Available for Cardiac Xenotransplantation and Associated Mechanisms.

Genetic Modification	Mechanisms	Properties
Alpha-Gal KO (GTKO)	Deletion of immunogenic Gal antigen expression	Anti-Immunogenic
B4GalNT2 KO	Deletion of B4Gal
CMAH KO	Deletion of Neu5Gc
Growth Hormone Receptor Knockout (GHRKO)	Reduction of downstream insulin growth factor-1 (IGF-1) signaling	Reduce intrinsic graft growth
hHO-1	Decreases oxidative products	Anti-Apoptotic
hHLA-E	Protects the graft against human killer cells	Anti-Inflammatory
hCD46	Suppresses human complement activity
hCD55	Suppress human complement activity
hEPCR	Activates Protein C	Anti-Coagulation
hTFPI	Inhibits Factor Xa
hvWF	Reduces platelet sequestration and activation
hTBM	Binds human thrombin, and activates Protein C via activated thrombin

CMAH = cytidine monophospho-N-acetylneuraminic acid hydroxylase; EPCR = Endothelial Protein C Receptor; HO-1: Heme Oxygenase −1; TFPI = Tissue Factor Pathway Inhibitor; HLA = Human Leukocyte Antigen; h = human; vWF = von Willebrand Factor; TBM = Thrombomodulin.

3. Limitations of preclinical models of cardiac xenotransplantation:

3.1. Non-human primates as recipients

Pig to non‐human primate (baboon) model has been commonly used in solid organ xenotransplantation research. Investigators, including us, have faced many issues dealing with post‐operative baboons and their long‐term care [36]. All baboons require central jugular catheter for drugs infusion which is difficult to maintain over long period of time without infections and there is a recurrent risk of recipients pulling the catheter out. Telemetry is also needed to continuously monitor graft function and recipients’ vitals. Since NHPs are stoic, exhibiting only subtle changes in behavior during distress or illness, and since NHPs cannot communicate symptoms, delay in identifying complications is inevitable. Continuous monitoring is needed to identify infection, tether or infusion pump malfunction, unexplained weight loss and bleeding.

3.2. Heterotopic transplantation model

The heterotopic model of cardiac xenotransplantation involves perfusion and drainage of the xenograft by anastomosis of the donor aorta to the recipient abdominal aorta and the donor pulmonary artery to the recipient abdominal IVC, respectively [37]. Proof‐of‐concept rejection free survival has been demonstrated with this model. However, PCXD was either not present or was not pronounced leading to unexpected barriers during the transition from a heterotopic to orthotopic model. Perhaps the non‐load bearing heart in this model does not generate enough stress to cause graft failure. Also, the absence of the transplanted cardiac xenograft to sustain life may allow the heart to recover over a few days without undergoing failure. It was observed in our heterotopic experiments that graft function, while sluggish at first, improved within a few days after transplantation.

While this injury also occurs in allotransplantation, xenograft interactions in the cross‐species recipient is complex and may be susceptible to failure after minimal injury from ischemia‐reperfusion. The mechanism of recovery could be as simple as providing rest to cardiac myocytes to offer time for spontaneous recovery after ischemic injury. The role of various components in different cardiac preservation techniques in this context is not known. Perhaps this model is appropriate to revisit specifically in identifying the role of ischemia reperfusion injury in contributing to PCXD, although the ability to stress the heart in this model is a limitation. But benefits include an intact recipient immune system while affording an opportunity to minimize confounders such as recipient malperfusion, shock and systemic inflammation as a result of a transiently failing life‐supporting heart during PCXD.

Because ischemia‐reperfusion injury and other transient conditions might compromise survival of recipients with life‐supporting xenografts before elicited immunity and consequences of incompatibilities are manifest, non‐functioning xenotransplantation models are likely to remain critical to the development of this field. Heterotopic (non‐ functioning) cardiac xenografts were essential for elucidating and proving the antigenic targets of Ab in HAR and early AVR and the importance of incompatibility of complement, coagulation and thrombosis. Non‐functioning xenografts have also proved important to testing the value of genetic engineering in xenotransplantation.

Role of Heparin:

Administration of heparin in xenotransplantation has been incorporated in the regimen of support of xenograft recipients to address incompatibility of coagulation between donors and recipients [38]. Additionally, long‐term patency of a tunneled jugular catheter requires some heparin administration through the catheter. This catheter is used for continuous injection of drugs throughout the period of xenograft survival. As the heterotopic model uses minimal physiologic load, it was once thought that ongoing administration of heparin might not be needed to support orthotopic xenografts, as blood does not stagnate in ventricular cavities. However, our latest experience with orthotopic cardiac xenografts proved otherwise and we have found continuous administration of heparin essential to prevent intracavitary thrombus formation and resuming heparin therapy slowly resolves these thrombi (Fig. 1). The addition of human thrombomodulin has not mitigated thrombus formation.

Fig. 1.

Formation of intra-ventricular thrombus in panel a-c over approximately-one month. Full resolution of thrombus after resuming heparin within 3 weeks (d–e).

We maintain the activated clotting time at twice basal level by adjusting the heparin dose. In other systems, heparin exerts anti‐inflammatory properties and other benefits, but impact if any in xenotransplantation remains to be proved [39,40].

3.3. Requirements for human translation

We propose human deceased brain dead (DBD) recipient studies for a limited 48–72 h time‐frame in parallel with good‐laboratory practice (GLP) studies with guidance by the Food and Drug Administration (FDA) for long‐term xenograft surveillance (although there may be other pathways to human translation than what is proposed here). The DBD and GLP studies should be with a cohort of animals using the same donor genetics and identical clinically‐translatable immunosuppression, as in our already conducted NHP studies (Fig. 2). As a parallel approach, one might consider a compassionate use/expanded access application with the FDA [41]. Here, a patient that otherwise would not be a candidate for allotransplantation, total artificial heart (TAH) or ventricular assist devices (VAD) may be directed toward an FDA‐directed single use, patient‐specific application of cardiac xenotransplantation. There would have to be careful consideration by the FDA and local ethics and institutional review board (IRB) approval and oversight to ensure informed consent on behalf of the patient.

Fig. 2.

Pathway to Clinical Translation of Cardiac Xenotransplantation.

Based on our experimental results we are confident that with the use of genetically‐modified pig organs and co‐stimulation blockade‐ based immunosuppressive regimen, we could achieve similar graft survival in human xenotransplantation. We believe that management of human recipients will be much easier than the non‐human primate recipients as the clinical staff is well trained to manage human transplant recipients. The need for blood transfusions, antibiotic therapies, postoperative care and surveillance in human recipients will also be standard of care based on years of experience in allotransplantation. That being said, a caveat to clinical translation is that NHPs used for preclinical studies are healthy animals and are different than humans, who require heart transplantation that have multiple comorbidities.

3.4. Translational experiments in deceased brain dead human patients

It has been suggested that brain dead human recipients, who are otherwise disqualified as organ donors, could be recipients of genetically modified pig cardiac xenografts and sustain recipient end‐organ perfusion for 48–72 h to scientifically evaluate pig‐to‐ human cardiac xenotransplantation. Transplants of porcine kidneys into brain dead human recipients has been reported [42,43]. The information gained from these experiments will be limited and include the role of antibodies against Neu5Gc or any other naturally occurring antibodies in humans against pig organs that we have not discovered so far. These antibodies towards unknown antigens may induce hyperacute rejection even to the grafts that are knocked out for CMAH. A secondary endpoint would be the graft performance in a human recipient and the presence of PCXD, as this is another known barrier to clinical translation. A summary of goals for a DBD experiment such as this is summarized in Table 5.

Table 5.

Objectives of a Deceased Brain Dead Recipient Cardiac Xenotransplantation Study.

1) Does hyperacute rejection occur from antibodies that cannot be evaluated in a NHP model?

2) Do further significant incompatibilities remain between pig and human after genetic engineering that may cause rejection?

3) Are novel immunosuppression regimens used in non-human primate model effective in pig-to-human heart transplantation?

4) Can Perioperative Cardiac Xenograft Dysfunction (PCXD) be avoided in pig-to-human cardiac transplantation?

5) Is there transmission of any porcine pathogens to the human recipient?

6) Can ECMO and cold static organ preservation maintain initial cardiac xenograft function?

7) Does hyperacute rejection occur from antibodies that cannot be evaluated in a NHP model?

8) Increase the opportunity for education about xenotransplantation with the layperson and garner acceptance of this procedure as bridge or destination therapy in future therapies

These experiments will also test the surgical techniques and performance of a pig heart in a larger recipient which will be three times the size of baboons we have tested so far. The brain‐dead patients are known to go through a cytokine storm which may induce major inflammatory reaction after transplantation of pig organ. The heart recipient in this experiment may have other co‐morbidities which may affect the outcome of this experiment.

However, brain death is sometimes accompanied by marked hemodynamic instability, refractory to traditional means of resuscitation [44]. At brain death, the vasomotor center (VMC), a part of the medulla, ceases to function or provide vasomotor tone to the peripheral circulation. A compensatory catecholamine surge occurs temporarily, causing hypertension, followed by catecholamine exhaustion from lack of VMC activity, leading to vasoplegia and refractory hypotension. Additionally, the pituitary axis also ceases to function and cannot stimulate the production or secretion of catecholamines, T3, T4, ADH or cortisol from the thyroid and adrenal glands. If there is continued support of a brain dead individual diabetes insipidus and adrenal insufficiency occurs. Therefore, in the context of monitoring PCXD, cardiogenic versus distributive causes of shock should be assessed. Extracorporeal membrane oxygenation (ECMO) could be employed as an adjunct to this study to ensure a complete 48‐hour assessment of the xenograft if this were to occur. However, pharmacologic regimens have been described that readily correct brain‐death vasoplegia for up to 24 h prior to termination of the experiment [45]. This technique could be employed if this phenomenon were to occur, as defined by escalating pressors to maintain MAP > 60 mmHg without an identifiable cause.

3.5. Regulatory requirement of further NHP studies in the GLP environment:

The GLP studies should focus on developing preclinical efficacy for the support of the first human clinical trial. These GLP studies as required by FDA and other regulatory agencies throughout the world and ensure proper conditions under which non‐clinical health and environmental safety studies are planned, performed, monitored, recorded, archived and reported. That being said, our latest results, along with others, may have already demonstrated appropriate clinical efficacy as defined by the ISHLT [9,11,46,47]. Therefore, the first human clinical trials are at the discretion of the FDA in the United States or other regulatory bodies outside the United States [48]. The donor pig organ is considered a “drug” and therefore must be approved by the Center of Veterinary Medicine (CVM) of the FDA before use in human trials in US. Concern about zoonotic diseases mandates that pig sources be obtained from a “designated pathogen free” (DPF) environment and screened to exclude presence of all known potential pathogens.

3.6. An alternative pathway-compassionate use authorization

The first FDA‐approved xenotransplantation was performed January 7th, 2022 after extensive FDA review of preclinical efficacy, institutional and research expertise, informed consent and IRB and ethics approval [1]. The patient was facing certain death and dependent on maximal mechanical circulatory support to sustain life. The patient was not a candidate for traditional therapeutics including allotransplantation, ventricular assist devices or total artificial hearts because of his non‐compliance, sarcopenic state and non‐ambulatory status. With a cardiac xenotransplantation using a multi‐gene pig and CD‐ 40 blockade‐based immunosuppression, the patient was able to be extubated and weaned from all support. The xenograft functioned for 60 days prior to compassionate withdraw of supportive care. The xenograft underwent rapid diastolic dysfunction with histology demonstrating dissolution of capillaries. This could be from an antibody‐mediated cause, either by iatrogenic administration of IVIG or an elicited antibody response to the xenograft. An alternative explanation is the presence of reactivated, latent porcine cytomegalovirus that was previously not detected prior to transplantation during the postoperative course. Each of these hypotheses is under investigation in our lab, and the latest information on this topic is discussed extensively elsewhere [1].

This pathway was approved by the FDA under expanded access (compassionate use) authorization. It is meant for patients without reasonable alternatives to standard therapies and with a risk of death without a life‐saving therapeutic. While not designed to replace a randomized control trial, it can offer patients an investigational drug or device in instances where a clinical trial is not feasible. Moreover, this mechanism of application is not meant to replace formalized GLP studies and eventual randomized control trials for full FDA approval for broad‐sweeping human use. Therefore, while other compassionate use authorizations may be sought and granted across the United States, it will likely not absolve the need for more formalized and broader pre‐ clinical and clinical investigation.

3.7. Expected manipulations required in human patients

Based on our experience, there is clearly a survival benefit in recipients with xenografts that express human transgenes in addition to just antigenic carbohydrate Kos [11]. “Multi‐gene” cardiac xenografts with multiple complement regulatory proteins, thromboregulatory proteins and antigenic knockouts demonstrate reduced IgM binding and complement dependent cytotoxicity in recipient NHP serum compared to those with less complex genetic modifications. Select multi‐gene cardiac xenografts demonstrate survival up to 9 months. Histologically, there is markedly less complement degradation product staining in those expressing complement regulatory proteins and less thrombotic findings in those with a combination of thromboregulatory and complement regulatory proteins. Xenografts with growth hormone receptor knockout have demonstrated minimal post‐transplantation xenograft growth in the first 6 months after transplantation, without the need for pharmacologic treatment of intrinsic growth, even in the presence of physiologic mismatch. Therefore, growth hormone receptor knockout xenografts with multiple carbohydrate knockouts, thromboregulatory proteins, complement regulatory proteins and anti‐inflammatory proteins are best suited for the first human clinical trials [11].

4. Xenograft and recipient surveillance

4.1. Xenograft surveillance

Post‐transplantation xenograft growth and xenograft medial hyperplasia consistent with transplantation‐related vasculopathy have been documented in cardiac xenotransplantation [9,11,49,50]. Therefore, xenografts should be monitored by echocardiography, right heart catheterizations, endomyocardial biopsies, coronary angiograms and intravascular ultrasounds (if capabilities exist) at regular intervals, which we have described elsewhere in NHPs [37,51]. While we have standardized the approach to surveillance using these adjuncts every 2 months, beginning 30 days after transplantation in NHPs, the approach likely can be liberalized to mirror schedules used in allotransplantation since guidelines have already been developed and human recipients can readily communicate worrisome symptomology in the event of xenograft rejection [52,53]. Clinically, we would recommend this as a standard surveillance measures until we understand the pathophysiology of this medial hyperplasia.

It is not a common practice to measure troponin levels for allograft rejection. However, troponin levels in our xenograft experiments have been found to reliably detect onset of rejection. Therefore, troponin levels are likely to be monitored in clinical trials of cardiac xenotransplantation during surveillance as it is readily measured in the clinical setting. Measurement of cell free DNA from porcine heart (xdcfDNA) usefully detects cardiac myocyte injury. We have found that xdcfDNA elevation is an early indicator on myocyte injury, death and rejection in experimental heterotopic cardiac xenotransplantation [54]. However, monitoring of xdcfDNA is more expensive and less accessible than troponin.

4.2. Pathogens tested after transplantation in humans

In allotransplantation, guidelines exist to screen for well‐ established pathogens that pose a risk to transplant recipients undergoing immunosuppression, including those of high risk [55]. Despite 1 in 5 allotransplantations meeting high risk criteria, however, unexpected disease transmission rate after allotransplantation is less than 1 % [56,57]. Similarly, porcine endogenous retroviruses (PERV) have a theoretical risk of transmission to humans, but the true risk is likely very low [58–60]. Moreover, there have been no documented cases of PERV transmission in xenotransplantation pre‐clinical models or in the limited human clinical trials of islet cell transplantation [35,61–66]. The latest pig‐to‐human cardiac xenotransplantation also did not detect transmission of PERV after 60 days from cardiac xenotransplantation [1]. In one study with limited perfusion of a kidney xenograft in a brain‐dead host, PERV transmission was not demonstrated [43]. A list of potential zoonotic infections is listed in Table 6 and guidelines for surveillance are detailed elsewhere [67].

Table 6.

Pathogens Required for Surveillance in Recipients after Xenotransplantation.

HIV	PERV
CMV IgG	pCMV
EBV IgG	pLHV*
HepBc Ab, HepBs Ab	Hepatitis E virus
Hep C Ab	Torque Teno sus virus (TTSuV)
VZV IgG	PCV (circovirus)*
HSV IgG	Porcine microchimerism
Toxo IgG	RPR
Quantiferon gold	HHV 6, HHV 8 PCR (quantitative)
CMV/EBV PCR (quantitative)	Microbial cell free DNA

5. Conclusion

Xenotransplantation appears close to clinical application and early trials are begun. However, the characteristics of source animals and management of recipients that would optimize outcomes remains to be determined and determination could well take decades (as it has in allotransplantation). The experience in pig‐to‐nonhuman primate experimental xenotransplantation suggests the determinants of success and optimization will differ in many respects from determinants of success in allotransplantation. Our experience and the experience of other groups reporting long term survival of pig to primate xenografts indicates that blockade of the CD40‐CD40L pathway to prevent production of elicited antibodies over months or years will be essential. Our experience indicates too that excessive growth and remodeling of porcine cardiac xenografts must be prevented. Whether this problem will afflict porcine kidney or tissue xenografts remains to be determined. Similarly, primary dysfunction of cardiac xenografts (PCXD) poses a solvable challenge in experimental cardiac xenotransplantation and pertinence and solution for clinical xenotransplantation of other organs and tissues is likewise uncertain.

A question looming over the field, but hopefully also of ephemeral significance, is the characteristic of subjects to be entered in the first clinical trials. Initial use of brain‐dead recipients was mentioned above. While experiments in brain dead recipients might offer some assurance about the quality of pig organs it remains unclear whether useful insights can be obtained. Subjects with brain death are highly selected and depart profoundly from normal physiology. For cardiac xenotransplantation, early recipients are likely to be deemed poor candidates for mechanical circulatory support due to biventricular restrictive disease and poor or unlikely recipients of cardiac allografts. Subjects with a high likelihood of disease recurrence would seem most likely to benefit from cardiac xenotransplantation. Subjects that would otherwise die if an organ were not immediately available and also not eligible for an allograft, would be the ideal candidate for the first human xenotransplantation clinical trials. Of course, if cardiac xenotransplantation proves as successful in human subjects as it has in NHP the characteristics of “ideal recipients” will greatly expand. Today the application of cardiac allotransplantation is severely limited by availability of organ and suitability of a given individual for transplantation reflects characteristics not only characteristics of that individual but also characteristics of other potential recipients. Success in cardiac xenotransplantation would allow clinicians to make decision based strictly on what is best for the patient.