The liver is arguably the most complex immune regulatory organ in humans.1 An overview is useful when trying to understand the relative immunologic inertness of a transplanted liver compared with other organ allografts. Multiple factors interact resulting in a system that generally is nonresponsive to molecules with low potential for harm but is able to mount a rapid, effective response to pathogens. The reason for this immune hyporesponsiveness is unclear;1 one hypothesis is that it protects the hepatic parenchyma from “bystander” damage associated with destruction of harmless microbes and thus interfere with the organ’s life sustaining synthetic functions.
The bulk of this immune surveillance function falls to the various macrophage types cells resident in the liver. The main hepatic macrophage is the Kupffer cell (KC) which comprises ~90% of its resident macrophages.2 KCs possess a number of high affinity receptors that allow them to bind and phagocytize potential toxins or pathogens opsonized with complement and/or antibody in the high blood flow milieu of the liver sinusoids. Once bound, the molecules are internalized and catabolized.
In the quiescent liver, KCs and other liver-resident macrophages and macrophage-like cells (liver sinusoidal endothelial cells, dendritic cells) and hepatocytes have been shown to suppress immune responses within the organ. The process(es) by which this occurs is unclear but absence of costimulatory molecules, reduced/absent HLA molecules, and constitutive expression of anti-inflammatory cytokines have all been implicated. A type of “sterile activation” in which T cells are activated and proliferate, but cannot engage specific antigens, has also been described.1,3 The liver hosts the largest number of natural killer (NK) and invariant NK cells in humans as well as some memory cells and transient populations of T cells, B cells, and neutrophils. The situation changes rapidly in the presence of traditional antigen presentation, infection, or inflammation; resident lymphocytes are activated, additional ones recruited as necessary and the liver immune repertoire transforms into pathogen elimination rather than tolerance mode.
Liver allografts start out in a state of immunologic inertness similar to that seen in a native liver and distinct from other solid organ transplants. Its remarkable regenerative capacity can repair smaller injuries and its large size in relation to hearts or kidneys requires a more substantial immune assault to cause lasting damage. Additional examples include low expression of Class I HLA molecules and minimal or no Class II expression; secretion of soluble HLA antigens which then combine with and inactivate any circulating donor-specifc antibody (DSA); the DSA-HLA complexes phagocytized by the liver’s KCs, and other macrophage populations are rendered ineffective; and finally the KCs which start out as donor cells, are replaced over time and to varying degrees, with cells of recipient origin which then may foster a tolerogenic environment.4-6
This relatively protected status can change quickly as a result of common clinical scenarios including prolonged ischemia/reperfusion times, bacterial or viral infections, and relatively mild rejection episodes. These events are associated with production of inflammatory cytokines, which results in significant upregulation of HLA Class I and Class II antigen expression in the areas affected by the inflammatory process. This provides/increases appropriate molecular targets and renders the previously “safe” organ susceptible to cell- and antibody-mediated rejection.7
There is agreement that liver transplant recipients may develop acute cellular rejection (ACR) and that when present, ACR requires treatment, usually in the form of increased immunosuppression. For the reasons outlined above, transplanted livers are not susceptible to hyperacute rejection. Other than that, there is little agreement concerning the relevance of HLA antibodies to the outcome of liver transplantation.
A review of the literature from the 1980s until today reveals approximately equal numbers of reports suggesting there are no lasting effects from DSAs on liver transplant outcomes and those attributing inferior survival or function to DSAs. A critical review of the findings suggest earlier studies are more likely to find no effect whereas later ones suggest that DSAs are associated with poor allograft function or increased rates of graft loss. The time period of the study is probably important—the earlier studies occurred when immunosuppression was generally heavier, opportunistic infections common and the 5-year survival was <50%. More recent studies are positively impacted by refined surgical techniques and improved immunosuppression and infectious disease protocols, resulting in longer survival times, fewer infectious, and technical issues to complicate the analysis of DSA effects.8
In this issue of Transplantation, Cousin et al9 offer a detailed analysis of a cohort of pediatric liver transplant recipients with clinically well-functioning grafts. They report retrospectively on 44 recipients transplanted over 15 years who underwent protocol liver biopsies. The histologic analysis identified 4 patterns: (1) normal histology; (2) fibrosis alone; (3) inflammation alone; and (4) fibrosis and inflammation. They correlated immunophenotype of the cellular infiltrate in the portal tracts and lobules with the histology and the presence or absence of preexisting and de novo HLA antibodies in the patients’ serum. They found that CD3+ T cells and CD68+ macrophages were the most common phenotypes overall but that in the presence of inflammation, CD68+ cells were reduced, CD3+ T Cells and CD 20+ B cells predominated, and CD8+ cells were associated with lobular fibrosis.
The most clinically relevant finding of the study was the association of circulating de novo Class II DSAs in patients with fibrosis and/or inflammation histology in their liver biopsies. They also determined that ~50% of the DSA were able to bind C1q. This is important because C1q antibodies are strongly implicated in tissue damage noted during antibody-mediated rejection.
The approach to understanding the immunologic status of liver allografts must take into consideration the unique features within the organ including evaluation of all aspects of the immune response elicited by the allograft, assessments of the timing and severity of relevant clinical events and an acceptance that ACR and antibody-mediated rejection are only 2 of multiple possible scenarios that exist patients with DSA/HLA antibodies and/or cellular infiltrates on a liver biopsy.
Evaluation of the immunologic status of liver transplant patients is an evolving process. It will require large numbers of patients, extensive clinical and laboratory investigations, and relatively long follow-up to determine the relevance of all the possible contributors. A clinical trial consortium might be the best way to accomplish this, but absent that, the blueprint provided by Cousin et al9 is a good place to start.
The existing literature largely documents histologic abnormalities, cellular infiltrates, and DSAs as separate unrelated occurrences. The article by Cousin provides evidence, although in nascent form, that these events are likely interrelated. Macrophages—integral to the maintenance of the liver’s immune hyporeactivity—were reduced in biopsies with inflammation and fibrosis, and these patients also had an increased incidence of de novo DSAs. These histologies are also associated with cellular infiltrates that produce inflammatory cytokines that in turn upregulate expression of HLA antigens, the molecules that initiate and are the targets of cellular and antibody rejection. The authors note “How these cellular and humoral factors interact is unclear, but peripheral DSA may be a marker of immune cellular activity in the seemingly quiescent allograft.” They probably also identify a population of recipients appropriate for further study to help answer some of the questions raised by this intriguing article.
1. Kubes P, Jenne C. Immune responses in the liver. Annu Rev Immunol. 2018; 36:247–277
2. Bilzer M, Roggel F, Gerbes AL. Role of Kupffer cells in host defense and liver disease. Liver Int. 2006; 26:1175–1186
3. Limmer A, Ohl J, Wingender G, et al. Cross-presentation of oral antigens by liver sinusoidal endothelial cells leads to CD8 T cell tolerance. Eur J Immunol. 2005; 35:2970–2981
4. Demetris AJ, Bellamy CO, Gandhi CR, et al. Functional immune anatomy of the liver-as an allograft. Am J Transplant. 2016; 16:1653–1680
5. Doreille A, Dieudé M, Cardinal H. The determinants, biomarkers, and consequences of microvascular injury in kidney transplant recipients. Am J Physiol Renal Physiol. 2019; 316:F9–F19
6. Starling RC, Armstrong B, Bridges ND, et al.; CTOT-11 Study InvestigatorsAccelerated allograft vasculopathy with rituximab after cardiac transplantation. J Am Coll Cardiol. 2019; 74:36–51
7. Michelo CM, van Cranenbroek B, Touw P, et al. Human cytomegalovirus infection increases both antibody- and non-antibody-dependent cellular reactivity by natural killer cells. Transplant Direct. 2017; 3:e335
8. Taner T, Stegall MD, Heimbach JK. Antibody-mediated rejection in liver transplantation: current controversies and future directions. Liver Transpl. 2014; 20:514–527
9. Cousin V, Rougemont AL, Rubbia-Brandt L, et al. Peripheral donor specific antibodies are associated with histology and cellular subtypes in protocol liver biopsies of pediatric recipients. Transplantation, this issue