Acquired immunologic tolerance: with particular reference to transplantation

Abstract

The first unequivocally successful bone marrow cell transplantation in humans was recorded in 1968 by the University of Minnesota team of Robert A. Good (Gatti et al. Lancet 2: 1366-1369, 1968). This achievement was a direct extension of mouse models of acquired immunologic tolerance that were established 15 years earlier. In contrast, organ (i.e. kidney) transplantation was accomplished precociously in humans (in 1959) before demonstrating its feasibility in any experimental model and in the absence of a defensible immunologic rationale. Due to the striking differences between the outcomes with the two kinds of procedure, the mechanisms of organ engraftment were long thought to differ from the leukocyte chimerism-associated ones of bone marrow transplantation. This and other concepts of alloengraftment and acquired tolerance have changed over time. Current concepts and their clinical implications can be understood and discussed best from the perspective provided by the life and times of Bob Good.

Fig. 1

An historical shift in policy of immunosuppression for organ transplant recipients. (A) During 1962–1964: azathioprine monotherapy was given before and after kidney transplantation at the University of Colorado, adding prednisone postoperatively only to control and reverse breakthrough rejections. Surrogate immune monitoring was done with serum creatinine tests. (B) Mid 1964-onward: In most centers, pretreatment was omitted and heavy prophylactic therapy with prednisone (plus additional drugs as these became available) was started at the time of transplantation. More than a third of a century later, it was recognized that these management changes essentially eliminated the possibility of achieving drug-free tolerance. Tx = organ transplantation

Fig. 2

Drug-free tolerance in long-surviving kidney allograft recipients of 1962–1963 whose immunosuppression was administered as shown in Fig. 1A. Light portions of transverse bars: off immunosuppression. By permission of Starzl et a1 [5]

Fig. 3

Failure of the immune system to recognize and reject skin allografts whose mobile antigen (i.e. passenger leukocytes) is prevented from reaching host lymphoid organs. By permission of Barker and Billingham [40]

Fig. 4

This illustration and caption (in quotes) were published in 1969 to explain organ engraftment. “Hypothetical mechanisms by which non-specific immunosuppression may lead to selective abrogation of the host immune response. Special susceptibility to these agents of a fraction of the lymphoid population could lead to exhaustion of a clone and, hence, tolerance. Since maintenance of such cell lines [clones] even in adult life is apparently thymic dependent in experimental animals, thymectomy would be expectedto aid the process; this appears to be true in rodents, but such an effect of thymus removal has not been detected in dogs or humans”. The concept proposed in 1969 was for the most part correct, but was not considered credible because of lack of scientific support. By permission of Starzl [44]

Fig. 5

Prolonged transfer from donors to recipients of positive tuberculin, coccidioidin, or other delayed sensitivity skin tests in 1962–63 cases of kidney transplantation at the University of Colorado. Although inexplicable at the time, these observations of adoptive transfer were consistent with donor leukocyte migration and relocation in the recipient. By permission of Starzl et al. [52]

Fig.6

Basis of a paradigm shift (see text). (A, B) Historical perception of organ and bone marrow cell recipients. (C, D) revised view of transplantation recipients

Fig. 7

Contemporaneous immune responses following allotransplantation: host versus graft (HVG, upright curves) and graft versus host (GVH, inverted curves). In contrast to the usually dominant HVG reaction of organ transplantation shown here, the GVH reaction usually is dominant after bone marrow cell transplantation to the irradiated or otherwise immune-compromised recipient. Therapeutic failure with either type of transplantation implies the inability of immunosuppression to control one, the other, or both of the responses. By permission of Starzl and Zinkemagel [67]

Fig. 8

The relation of organ engraftment to classical models of spontaneous donor leukocyte chimerism-associated tolerance. By permission of Starzl et a1 [6]

Fig. 9

The pace of development of a donor-specific response following organ transplantation in different extant species and the subsequent development of donor-specific non-reactivity. These events closely match the kinetics of donor leukocyte migration and relocation. Note that immunosuppression may not be required for liver engraftment in 3 of the 5 species shown (lightly shaded portion of transverse bars). Tx = transplantation

Fig. 10

How the spread and localization of antigen determines both the induction at host lymphoid organs of adaptive immunity, and then the target of this immunity. (A) hepatitis virus (B) migratory passenger leukocytes of a transplanted organ (here a liver) (C) cytomegalovirus (CMV) localized to the lung

Fig. 11

Analogies between infection by non-cytopathic microparasites and the scenarios of transplantation. The outcomes are governed by the migration and localization of the respective antigens (refer also to Fig. 10). The analogies have been obscured by the presence of contemporaneous host-versus-graft (HVG) and graft-versus-host (GVH) responses after transplantation and the additional factor of therapeutic immunosuppression. (A) The presence of a pathogen that fails to reach organized host lymphoid tissue is not recognized (immune ignorance). Consequently, there is no immune response. The relation of immune ignorance to clonal exhaustion-deletion must be understood to comprehend alloengraftment (see text). (B) A highly infectious but asymptomatic and stable carrier state may be reached when a rampant non-cytopathic microorganism exhausts and deletes the antigen-specific immune response, (e.g. viral hepatitis). The transplant analogy is ‘complete’ repopulation of an immunodeficient or cytoablated bone marrow recipient without the penalty of GVHD. (C) Complete elimination of a non-cytopathic pathogen by an immune response that then subsides without memory. The transplant analogy is rejection of an organ’s passenger leukocytes and the outlying source graft (refer to Fig. 10). (D) Instead of the outcome in Panel C, persistence of small numbers of microorganisms may maintain cellular plus antibody ‘memory’ krotective immunity). In the transplantation analogy, residual microchimerism may result in a “presensitized” state that renders a transplant candidate “crossmatch positive” with most donors [83, 84]. (E) Disease carrier states in which a balance is established that favors pathogen load over pathogen-specific T cells. The corresponding spectrum of transplantation analogues is delineated by simply substituting “donor leukocyte” for pathogen. For organ transplantation purposes, an umbrella of immunosuppression usually is needed to keep the various balances in stable equilibrium (see also Fig. 12). (F) Refractory infection syndromes in which neither adaptive immunity nor therapy with antimicrobial agents can achieve the kind of disease control shown in Panels C and D or the asymptomatic carrier state sometimes seen in Panels B and E. The treatment failure analogues of transplantation occur when the opposite objective (sustained dominance of the migratory alloantigen over an aggressive antidonor T cell response) is not achievable with immunosuppression or other means

Fig. 12

Balance between antigen with access to host organized lymphoid tissue, and the antigen-specific cytolytic T cells (CTL) induced at these lymphoid sites. A balance favoring the migratory antigen over the antigen-reactive CTL (Upper teeter totter) may be achieved for transplant purposes in some models by simply adding adjunct donor leukocytes. In standard clinical practice, however, the desired balance is almost always maintained by reducing the specific CTL response with immunosuppression (discussed in text). Tilting the balance in the opposite direction results in immunity (lower teeter totter)

Fig. 13

Weakening or elimination of the clonal response by excessive post-transplant immunosuppression to the extent that efficient exhaustion and deletion of the clonal response is prevented. Subsequent graft survival is permanently dependent on immunosuppression

Fig. 14

Principles of tolerogenic immunosuppression. (A) Experimental organ transplant models of spontaneous tolerance (no immunosuppression needed). In these unusual models, the host versus graft immune response induced acutely by the migratory donor leukocytes is too weak to eliminate the donor cells and is exhausted and deleted. The deletional state induced at the outset is then maintained by microchimerism. (B) Organ transplant models in which the recipient response that normally would cause rejection (dashed line) is reduced into a deletable range (continuous thin line) with a short course of early post-transplant immunosuppression. Neither the mechanisms nor the ultimate result are different than in the spontaneous tolerance models of panel A. (C) Models in which the global recipient immune responsiveness is weakened in advance, thereby making deletion easier of the subsequently induced donor-specific T cell clone. This pretreatment (conditioning) principle is the essential basis of bone marrow transplantation. It has not been systematically exploited in organ recipients. Tx = Transplantation

Fig. 15

Combination of the two principles shown in Fig. 14B and C in a “tolerance friendly” immunosuppression protocol routinely used since 2001 at the University of Pittsburgh Medical Center [96]. Lymphoid depletion was done before organ allograft revascularization. Weaning from post-transplant monotherapy was systematically attempted (see text). The usually silent graft versus host (GVH) reaction depicted in Fig. 7 is not shown here. Tx = Transplantation

Fig. 16

The patient and graft survival of the first 89 intestinal or multivisceral allograft recipients treated with the tolerogenic immunosuppression shown schematically in Fig. 15. Note that only 8% of the surviving patients with functioning grafts are on more than a single immunosuppressant, and that 40% are on spaced weaning

Fig. 17

Course of an intestinal recipient who was pretreated with ATG and managed postoperatively with tacrolimus monotherapy. She has been on full oral alimentation for nearly 5 years and has been treated with two doses of tacrolimus per week for the last 4 years. The xxx’s at the bottom indicate intestinal biopsies. Note the stable creatinine (i.e. avoidance of tacrolimus nephrotoxicity) with this regimen

Fig. 18

Patient and kidney graft survival using alemtuzumab (campath^R) lymphoid depletion and minimalistic post-transplant immunosuppression, versus historical experience. The pie shows the current immunosuppression of survivors with functioning grafts

Fig. 19

Height (A) and weight (B) for age and gender Z scores for 16 kidney transplant recipients who have born functioning grafts for 2 years or longer under tolerogenic immunosuppression. Reference population data is from the 2000 Center for Disease Control, growth data available at www.cdc.gov/growthcharts. By permission of Shapiro et al. [99]

Fig. 20

Patient and cadaveric liver allograft survival of HCV-negative adults who were lymphoid depleted with ATG or alemtuzumab and treated after transplantation with minimal post-transplant immunosuppression. The pies indicate the maintenance immunosuppression at 3 years

Fig. 21

Course of an alemtuzumab-conditioned cadaver liver recipient who was completely weaned from immunosuppression without incident

Fig. 22

Results in HCV-infected liver recipients who were lymphoid depleted with ATG or alemtuzumab and treated with tacrolimus monotherapy from which spaced weaning was attempted. The drastic decline in survival was due almost exclusively to accelerated recurrence of hepatitis [6, 102]

Fig. 23

Explanation for the loss of poor results shown in Fig. 22 (see text and Refs. [6, 102)

Fig. 24

The “house of transplantation”, viewed in hindsight (see text)

Fig. 25

Tolerance protocol applied in 2005–2006 based on the concepts summarized from the historical perspective of Fig. 24

Fig. 26

Course of the first patient treated by the protocol shown in Fig. 25. Complete drug discontinuance has not been possible (see text). TAC = tacrolimus

Fig. 27

Second patient treated by the protocol in Fig. 25 (see text). The patient has been drug-free for a half Year

Fig. 28

Course of a patient whose post-transplant course was complicated by a major biliary fistula that was corrected at reoperation. Note that little or no immunosuppression was given from the third postoperative week onward

Fig. 29

Balances of mobile donor antigen (donor leukocytes) and donor-reactive T cells after organ transplantation. (A) With organ transplantation alone under the immunosuppression shown in Fig. 15, a balance favoring drug-free antigen supremacy over the T cell response induced by passenger leukocytes is theoretically most likely if there is a large quantity of persisting donor cells (macrochimerism). A positive antigen balance also is possible with the microchimerism but this usually requires continuous immunosuppression to weaken the CTL arm. (B) With leukocyte infusion in advance of organ transplantation (see Fig. 25), the patient is partially tolerized by the time the second load of donor cells (the passenger leukocytes) arrive and boost the tolerogenic process while theoretically increasing the risk of GVHD (see text). Preliminary experience with this protocol in liver and kidney recipients has been encouraging