The Long and Winding Road to Clinical Xenotransplantation: A Personal Journey

Since the mid-1980s, my major research interest has been the field of xenotransplantation (cross-species transplantation) with the aim of addressing the critical shortage of organs from deceased human donors by providing an alternative source of organs from genetically engineered pigs. It has been an interesting and exciting, if at many times frustrating, road to follow, but we now seem very close to the initiation of formal clinical trials. The recent clinical pig heart transplant carried out at the University of Maryland in Baltimore [1] was the first step in what I hope will be many more. Having been involved in xenotransplantation research from the beginning, I felt it would be useful to record progress from my own recollections.

In 1985, my colleagues and I began to explore the potential of the pig as a source of organs for humans. Hearts from wild-type (WT, i.e., genetically unmodified) pigs, when perfused with baboon blood or heterotopically transplanted into baboons, were rejected usually within minutes or hours [2, 3] (Fig. 1). When we adsorbed anti-pig antibodies by perfusing the baboon blood through a donor-specific pig kidney before transplanting the heart and administering cyclosporine-based immunosuppressive therapy, graft survival was extended for a few days [3].

It had been ascertained by others that an ABO-incompatible human kidney allograft could be transplanted successfully if the recipient’s anti-AB blood type antibodies were depleted (e.g., by plasmapheresis) before the kidney was implanted [4]. With effective immunosuppressive therapy, this generally resulted in either no return of antibody or a return of antibody that was not associated with rejection of the graft (a phenomenon called “accommodation”).

It occurred to me that if we could identify the target for anti-pig antibodies and if they were carbohydrates (as are ABO antigens), then we may be able to deplete the baboon of anti-pig antibodies, perhaps enabling pig organ transplants to become successful. We perfused human plasma through pig hearts ex vivo, then eluted the antibodies that had become bound to the vascular endothelial cells, and sent the antibodies to collaborators at Chembiomed, a company in Edmonton, Canada, that produced synthetic oligosaccharides. We identified galactose-α1,3-galactose (Gal) as the major xenoantigen (Table 1) [5, 6]. Although anti-Gal antibodies could be successfully removed from the blood and this delayed pig organ rejection, the continued production of these antibodies (despite immunosuppressive therapy) always resulted in eventual failure of a pig graft.

Antibody binding to pig cells activates the complement cascade, but the expression of complement regulatory proteins (CRPs) on the human (or nonhuman primate [NHP]) host cells protects these cells from (autologous) complement injury. CRPs are largely species-specific [7, 8], and so David White’s group in the UK introduced the gene for the human CRP, CD55 (human decay-accelerating factor), into pigs and demonstrated that rejection of these pig organ grafts in NHPs was significantly delayed [9, 10]. This demonstrated that, for the first time since clinical organ transplantation was initiated in the early 1950s, we could modify the donor rather than just treat the recipient.

In an effort to induce a state of immunological tolerance to a pig kidney in baboons by achieving mixed donor hematopoietic cell chimerism, we harvested hematopoietic progenitor cells from WT pigs and infused them into baboons after removal of anti-Gal antibodies. We observed that when a cyclosporine-based regimen was administered, within a few days, there was an adaptive immune response with a 100-fold increase in anti-Gal IgG [11, 12]. In contrast, agents that blocked the CD40/CD154 (CD40L) T-cell costimulation pathway prevented an adaptive immune response [11].

Since then, almost all researchers in pig-to-NHP xenotransplantation have used a regimen based on blockade of the CD40/CD154 costimulation pathway. When anti-CD154 mAbs were found to be thrombogenic [13], an anti-CD40mAb was introduced [14]. This became the standard agent until anti-CD154 agents that had been Fc-modified to prevent thrombogenesis were introduced within the past 2–3 years.

Although deletion of a gene in pigs was not possible in 1993, we suggested that knockout of the gene for the enzyme α1,3-galactosyltransferase (GTKO) would result in a pig that did not express Gal (Table 1) [15]. There would therefore be no targets to which human (or NHP) anti-Gal antibodies could bind, thus greatly reducing the possibility of complement activation and of early antibody-mediated rejection. After the successful cloning of large mammals in 1996, the first GTKO pigs were bred in late 2003 [16].

Molecular incompatibilities in coagulation and anticoagulation factors between pigs and primates had been investigated by several researchers [17], but it was not until the late 1990s that disturbances of coagulation were definitively demonstrated in the pig-to-NHP model [18‒20]. A pig organ xenotransplant might fail through the development of a thrombotic microangiopathy, leading to a consumptive coagulopathy in the recipient [20]. If the graft were excised before this was too advanced, this complication generally resolved.

These studies and those of others [21] led to the development of pigs that expressed one of more human coagulation regulatory proteins, e.g., thrombomodulin, endothelial protein C receptor, tissue factor pathway inhibitor, or CD39 [22]. The expression of one or more of these human coagulation regulatory proteins later became an essential element in pigs being used as organ sources for xenotransplantation.

With pigs provided by Revivicor (Blacksburg, VA, USA), we carried out the first study of xenotransplantation using organs from GTKO pigs. Hearts from GTKO pigs (with no additional genetic manipulations) were transplanted heterotopically (into the abdomen) of baboons immunosuppressed with an anti-CD154 mAb-based regimen [23]. Graft survival was significantly longer than of WT pig hearts (median 78 days, mean 99 days). Graft survival of 179 days (6 months) was the longest survival of any pig organ in an NHP recorded to that date. Importantly, there were few infectious complications, suggesting that the immunosuppressive regimen (anti-CD154mAb, mycophenolate mofetil, corticosteroids) was not excessive.

At graftectomy or necropsy, the predominant histopathological features were of a thrombotic microangiopathy, with surrounding ischemic injury, with focal interstitial hemorrhage and edema. The thrombogenicity of the anti-CD154 mAb was not thought to be a factor as a thrombotic microangiopathy also developed when an anti-CD40mAb was administered.

Following this study, GTKO became an essential feature in the genetically engineered pigs considered suitable for organ xenotransplantation. Subsequent studies by our group combined GTKO with transgenic expression of the human CRP, CD46, in the organ-source pig and demonstrated a further reduction in the incidence of early rejection [24]. This was followed by the transplantation of GTKO kidneys that expressed both human CRP- and coagulation regulatory proteins, which was associated with even more prolonged graft survival (of up to 260 days) [22].

An inflammatory response had been predicted by early in vitro studies and in rodent models, and this proved to be the case in the GTKO pig-to-baboon model [25, 26]. The systemic inflammatory response was remarkably sustained, with prolonged increases, particularly in C-reactive protein and interleukin-6. There is evidence that the insertion of a human apoptotic or “anti-inflammatory” gene, e.g., hemeoxygenase-1 or A20 [27], into the pig may reduce the effect of inflammation on the pig graft [28]. When the organ is taken from a pig expressing human complement, coagulation, and anti-inflammatory proteins, all of which have anti-inflammatory effects, it is uncertain whether any prophylactic systemic anti-inflammatory therapy is required.

It had been known for many years that pigs express the sialic acid N-glycolylneuraminic acid (Neu5Gc) (Table 1), but humans do not (and therefore make antibodies against Neu5Gc) [29, 30]. In 2015, Byrne identified a third carbohydrate antigen against which humans have natural antibodies (Table 1) [31].

Steps were taken initially by Estrada and his colleagues [32] to delete these xenoantigens in addition to Gal, a process made easier, quicker, and less expensive by the introduction of the CRISPR technology. The natural antibody that binds to these “triple-knockout” (TKO) pig cells is absent in many humans [33], suggesting that, after a TKO pig graft is inserted, the likelihood of antibody-mediated rejection is greatly reduced, if not totally eliminated (there are other xenoantigens, e.g., swine leukocyte antigens [protein antigens] against which antibodies can develop if the adaptive immune response is not sufficiently suppressed [34, 35]).

An unexpected observation was that although many humans do not have antibodies against TKO pig cells, all Old World NHPs, including baboons, do have antibodies against TKO pig cells (Fig. 2a) [32, 36, 37]. This provided a new barrier for testing TKO pig organ transplants in NHPs [38] and appears to be related, at least in part, to the fact that they, unlike humans, express Neu5Gc. It has been hypothesized that deletion of expression of Neu5Gc in the pig exposes another xenoantigen, a carbohydrate often termed the “4th xenoantigen,” against which NHPs, but not humans, have natural antibodies. Importantly, therefore, TKO pig organ transplants are being carried out into sensitized NHP recipients.

However, there are other possible contributing factors, e.g., relating to complement, that may be playing a role in the vigorous serum cytotoxicity of NHPs to TKO pig organs (Fig. 2b) [36, 37]. To date, this problem has prevented consistent success of organ transplantation in the TKO pig-to-NHP model, although therapy with an anti-CD154 agent appears to overcome this barrier in some cases [39]. The pig-to-NHP model is therefore no longer representative of clinical pig organ xenotransplantation and has led some to advocate for the initiation of limited exploratory clinical trials [37].

By 2019, many different transgenes could be (and had been) introduced into potential organ-source pigs, but it was uncertain which combination of these would be preferable to enable clinical organ xenotransplantation to be explored. We suggested that the optimal genetically engineered pig at that time would be one with nine gene edits [28] (Table 2). These would include deletion of expression of the three known glycan xenoantigens (i.e., TKO pigs), the introduction of two human CRPs (for CD46 and CD55), two human coagulation regulatory proteins (thrombomodulin and endothelial cell protein C receptor), and one anti-inflammatory (apoptotic) protein hemeoxygenase-1. Finally, in part because it was hoped that eventually tolerance might be achieved by concomitant hematopoietic cell transplantation, the macrophage inhibitor, CD47, was introduced (Table 2).

Evidence from multiple in vitro studies suggested that a pig with the above genetic modifications (TKO.CD46.CD55.TBM.EPCR.HO-1.CD47) would be the preferred source of organs for human recipients, but not for NHP recipients. However, in view of the reported (by us and others) rapid growth of pig kidneys and hearts in the first few months after transplantation into NHPs (possibly because during this period, growth of the pig organ is still stimulated by pig growth hormone), our colleagues at Revivicor wished to include a further genetic manipulation, namely, knockout of the gene for growth hormone receptor that had been suggested by Hinrichs et al. [40]. This is particularly important in regard to pig orthotopic heart transplantation because of the restricted space within the chest but is less important for kidney transplantation. Knockout of the gene for growth hormone receptor was therefore added to the pig. A pig with this genetic background provided the heart for the recent first clinical xenotransplant in a patient at the University of Maryland at Baltimore (Table 2).

In view of the current barriers to progress provided by the pig 4th xenoantigen in NHPs, the time has surely come when we need to consider moving from the laboratory to the clinic [37]. However, there are still some differences of opinion within the research community on a number of points that need to be collectively and definitively resolved. These include (i) what exact genetic modifications do we need in the organ-source pig? For example, if there are no antigenic targets on TKO pig cells to which antibodies can bind, will the complement cascade be activated? Our data suggest that expression of a CRP provides added protection [41]. (ii) What exact immunosuppressive regimen will we choose? [42]. For example, is an anti-CD154 agent preferable to an anti-CD40 agent? – probably “yes,” and is there a need for any additional conventional immunosuppressive agent? – possibly “no” [43]. (iii) How will we monitor the immune response and diagnose and treat rejection? Today, the evidence is that monitoring of a patient with a pig heart or kidney graft can be identical to that of a patient with an allograft [44], but at present, there is no successful treatment that reverses antibody-mediated rejection of a xenograft. (iv) How do we plan to prevent or treat potential infectious complications? Again, this should follow the approaches used after allotransplantation [45].

Which patients will be offered a pig organ in the first clinical trial? I suggest that pig kidney transplantation should be introduced before pig heart transplantation, if only for the fact that if the graft fails for any reason or if immunosuppressive therapy has to be discontinued, e.g., to help overcome a life-threatening infection, the pig kidney graft can be excised, and the patient returned to chronic dialysis. A similar option is not available to patients with terminal heart failure who receive a pig heart transplant.

Patients who could ethically be considered might be those who are acceptable candidates for allotransplantation but who are unlikely to obtain a kidney from a deceased (or living) human donor for some years [46]. Patients on the waitlist for a deceased human donor kidney in the USA have an approximate 45% chance of being withdrawn from the waitlist within 5 years, either because they die or because they are no longer considered suitable candidates, e.g., from the development of new comorbidities, general physical frailty, etc. Patients in their sixties (with good physiology) for whom it is anticipated that no human donor organ will become available within 5 years may wish to consider kidney xenotransplantation in order to enjoy life without the restrictions imposed by chronic dialysis for a period of time (hitherto uncertain) while they remain waiting for an allograft.

A case can be made for clinical trials of cardiac xenotransplantation in both adults and infants, initially possibly as a bridge to allotransplantation [47]. However, when planning such a trial, it has to be borne in mind that, to date, NHPs with a life-supporting pig kidney (though not a TKO kidney) have survived for >3 years, whereas the maximal survival of an NHP with an orthotopic heart transplant is <9 months [48‒50].

In this brief review, I have concentrated attention on pig kidney or heart transplantation as these procedures have advanced to the point when clinical trials can be considered. The results of transplantation of pig livers and lungs in NHPs lag behind as these two organs appear to generate a more complex immunobiological response. However, ultimately, I am sure we will be carrying out clinical transplants of these organs, as well as cell and tissue transplants for conditions as varied as diabetes (islets), corneal blindness, Parkinson’s disease (dopamine-producing cells), and hemorrhage (red blood cells).

“There is nothing more powerful than an idea whose time has come.”

Victor Hugo (1802–1885)

The author thanks his very many colleagues and collaborators who have contributed to the studies summarized in this review.

David K.C. Cooper is a consultant to eGenesis Bio of Cambridge, MA, USA, but the opinions expressed in this article are his own and do not necessarily reflect those of eGenesis.

Work on xenotransplantation in the author’s laboratory has been supported for many years in part by the NIH NIAID, in part by the Department of Defense and in part by Revivicor, Blacksburg, VA, for which he is most grateful.

The manuscript was written solely by the author.