Biological Time Travel: Tracking the Journey of Alloreactive T Cells through Longitudinal Biobanking

Summary:

A significant risk for transplant recipients is the development of tumors. In general, some but not all malignancies are more frequent in transplant hosts due to chronic immunosuppression caused by a compromised immune surveillance. Of additional relevance, checkpoint blockade therapies (CBT) to treat malignancies can also drive transplant rejection. In a recent study published in Nature Communications, Dunlap et al. reported a case study of a patient who experienced kidney allograft rejection following CBT for melanoma. The foresight of longitudinally preserving donor splenocytes, blood samples, and graft biopsies in addition to tumor and metastatic lymph nodes enabled paired single-cell RNA-seq (scRNA-seq) and TCR-sequencing (TCR-seq) and subsequent tracking of alloreactive T cells before and after CBT. This revealed an enrichment of alloreactive TCRs in the kidney transplant post-CBT but not the tumor. In addition, this approach helped identify an alloreactive CD8⁺ T cell subset with a unique transcriptional profile. This study illustrates possible advances in personalized medicine and highlights a transcriptional signature that may serve as a prospective biomarker of rejection.

The emergence of tumors is a great risk to transplant recipients. Moreover, utilizing checkpoint blockade therapies (CBT) may result in rejection of the transplanted organ¹. In a recent issue of Nature Communications, Dunlap and co-workers² reported a case study of a patient who, over ten years post-kidney transplantation for chronic glomerulonephritis, experienced episodes of allograft rejection following the use of CBT (pembrolizumab, 200mg every 3 weeks) for metastatic melanoma. The authors hypothesized that graft rejection was due to the expansion, post-CBT, of pre-existing alloreactive T cells. The authors evaluated donor-specific alloreactivity using a mixed-lymphocyte reaction (MLR) in which recipient PBMCs collected post-CBT were carboxyfluoroscein succinimidyl ester (CFSE) labeled and stimulated with banked donor splenocytes. Proliferating T cells were then sorted and underwent subsequent paired scRNA-seq/TCR-seq to identify a population of bona fide alloreactive CD8⁺ T cells, defined by their proliferative transcriptional signature and clonal expansion. Next, systemic T cells were evaluated by bulk TCR-seq longitudinally pre-CBT in addition to post-first and second rounds of pembrolizumab. Similar studies were performed in biopsies from the kidney transplant, from a pre-CBT metastatic lymph node and from post-CBT metastatic skin samples. The authors confirmed the MLR-identified alloreactive clones in post- but not pre-CBT PBMCs and observed them infiltrating the kidney transplant but not the tumor sites. This tissue restriction to the graft, perhaps unexpected, is significant and novel. Others have previously shown that activated alloreactive T cells and irrelevant effector CD8⁺ T cells enter murine allografts³ and that bystander (non-tumor-reactive) effector CD8⁺ T cells enter human malignancies⁴. Although with limitations as a case report, the current study demonstrates feasibility of tracking alloreactive T cell phenotypes over time and helps understanding CD8⁺ T cell expansion and activation as a threat to the transplanted allograft following CBT.

Previous cutting-edge work had already taken advantage of MLR assays to identify and subsequently track donor-specific T cell clones. A manuscript published in Science Translational Medicine in 2015 by Morris et al.⁵ tracked alloreactive T cells in patients receiving dual donor bone marrow and kidney grafts, helping the authors establish clonal deletion as a mechanism of human transplantation tolerance. To this end, the authors defined a pre-transplant fingerprint of the donor-specific TCR repertoire by leveraging high-throughput TCRβ CDR3 sequencing of T cells having expanded in an anti-donor MLR. CDR3 areas were amplified for all 54 Vβ and all 13 Jβ regions prior to high throughput sequencing. Post-transplant T cell tracking in PBMCs was then verified by repeated MLR reactions and evaluation of TCR repertoires. The findings by Dunlap et al. take this method a step further: by analyzing MLR-expanded anti-donor T cells with paired scRNA-seq/TCR-seq, the authors are not only able to track the identified alloreactive clones directly ex vivo, but also examine their transcriptional signature, leading to the identification of a ZNF683⁺ (Hobit) subset of CD8⁺ T cells that represents proliferating alloreactive T cells. Their approach has thus revealed a potential tractable biomarker of rejection. Moreover, access to biopsies and the capacity to sequence graft- and tumor-infiltrating T cells and determine where these circulating proliferating alloreactive T cells go after CBT added unprecedented depth. Previous efforts to track alloreactive T cells have been limited to evaluation of PBMCs, whereas the direct comparison of alloreactive T cells in the PBMC compartment with alloreactive T cells in tissue biopsies provides a more complete understanding of these cells’ function. Whether this molecular method can be used to predict risk of CBT-associated rejection prior to treatment with a low clonal frequency of alloreactive PBMCs prior to CBT remains to be determined. However, Morris et al. were able to use MLRs to identify donor-specific clones pre-transplantation, suggesting that the approach of Dunlap et al. may succeed prior to CBT to identify alloreactive clones and use the expansion of the ZNF683⁺ subset post-CBT to flag patients for careful graft function monitoring.

Paired evaluation of scRNA-seq/TCR-seq from MLR-expanded T cells remains a highly skilled and onerous technique that is not yet widely accessible. Moreover, subsequent tracking of alloreactive T cells requires diligent, repeated sampling and biobanking from the recipient’s blood and graft. Additionally, access to donor cells as stimulators for the MLR is a prerequisite. For deceased donors, donor spleens as used by Dunlap and co-workers are an excellent and large source of donor antigen that is not currently utilized, though some investigators have proposed injecting donor splenocytes into transplant recipients for inducing chimerism⁶, T cell dysfunction⁷, or as a desensitization strategy in sensitized patients (NCT04827186)⁸. Alternative sources of antigens would be necessary for live donor transplants with PBMCs representing promising candidates.

Expansion of alloreactive T cells may also represent a potential biomarker of rejection. CBT, although important to treat tumors, poses a rejection risk and Dunlap et al. demonstrate that alloreactive CD8⁺ T cell clones expand post-CBT. The authors also identified a transcriptionally unique subset of expanded alloreactive CD8⁺ T cells characterized by their proliferative phenotype and expression of ZNF683. ZNF683 is a transcription factor implicated in migration, memory formation and tissue residency of CD8⁺ T cells. Whether ZNF683⁺ CD8⁺ T cells that expand post-CBT and infiltrate the kidney allograft represent key drivers of rejection and whether expansion of this subset in the blood can be used as an early biomarker of rejection will require additional studies. If confirmed, tracking of this subset prior to and following CBT may inform ongoing clinical trials (NCT03816332, NCT04339062)^9,10 aimed at preventing rejection following CBT.

Overall, the study by Dunlap et al. illustrates new possibilities in personalized medicine using the currently available molecular technologies in an innovative way while demonstrating the usefulness of longitudinal biobanking in clinical transplantation.

Figure. CBT can promote both tumor control and transplant rejection.

CBT blocks negative interactions between T cells and APCs or between T cells and target cells, unleashing restrained T cells. Those effects can result in beneficial anti-tumor immunity but may also activate alloreactive T cells leading to transplant rejection.

Abbreviations

CBT

checkpoint blockade therapy

CFSE

Carboxyfluoroscein succinimidyl ester

MLR

mixed lymphocyte reaction