Journal Logo

In View: Research Highlights

Research Highlights

Lewik, Guido MD1; Issa, Fadi DPhil, FRCS1

Author Information
doi: 10.1097/TP.0000000000004272
  • Free


Long-term success in heart transplantation remains limited by cardiac allograft vasculopathy (CAV), a fibroproliferative disease characterized by luminal narrowing of the macro- and microvasculature.1 To predict CAV development in heart transplant patients, clinical, functional, and macroscopic indicators have been assessed with very limited success.2 More precise individual risk stratification including data on microarchitectural changes, which often precede the other indicators,3 remains unavailable.

In a recent, computational analysis of routine endomyocardial biopsies, Peyster and coworkers have provided a compelling paradigm of supplementing and integrating data from digitalized pathology samples with clinical features to strive for personalized, foresighted treatment regimens.4 The authors first developed a clinical risk factor model predicting CAV onset using data from 302 patients, 249 without early-onset CAV and 53 with pre–early-onset CAV (ie, angiographic CAV diagnosis within 5 y of transplant). Of these, two-thirds were used as training sets for model development, whereas one-third subsequently served as a validation test set. Out of 85 clinical variables, 7 were chosen as clinical risk factors: donor coronary angiography score, donor proteinuria, Quilty lesion (endocardial infiltrate), history of high-grade cellular rejection/treated rejection within the first year and recipient low-density lipoprotein, actively treated diabetes, or body mass index at 1-y posttransplant. Overall, the clinical risk factor prediction model showed mediocre performance, not exceeding an accuracy of 66.3% on the test set.

For histological evaluation, endomyocardial biopsy samples from 1-y posttransplant were grouped into 50 pre–early-onset CAV samples and 82 samples from patients without early-onset CAV (no-CAV). Additionally, 51 disease control biopsies from patients with definitive angiographic CAV diagnosis were assessed. A comparison of no-CAV and (definitive CAV) disease controls revealed 11 variable discriminative features, such as perivascular cellular density and collagen proliferation inside and outside of the myocardial compartment. This diagnosis model reached 86.7% accuracy on the test set. For predictive modeling, no-CAV samples were compared to pre–early-onset CAV samples. 10 variables were included, such as the ratio of microvessels to myocyte staining areas and perivascular cellular density. Accurate prediction of the test set was found to be 81.6% with this model. Lastly and most crucially, the authors incorporated these histological and clinical predictors toward an integrated model, which excelled in performance with a validation test set accuracy of 88.6%.

Despite its limitations as a single-center study evaluating a mid-sized data collection, the thorough approach and promising results of this study are encouraging. Risk stratification with routine histology sample analysis may offer a quick, inexpensive, yet precise prognosis for CAV development in this integrated model. We are likely to see increased integrated clinical-histological analysis in the future as digital pathology techniques evolve in transplantation.


Tissue-resident memory T (TRM) cells are typically found in barrier organs where they provide protective immunity against pathogenic challenges such as viral infection.1 These cells are also known to persist in transplanted tissues long-term,2,3 although their precise role in rejection is not yet clear.

In this study, Tian and coworkers provide unprecedented insight into the role of skin-resident TRM cells in transplant rejection using a novel sequential mouse skin transplantation model.4 In this model, BALB/c mice are immunized with C57Bl/6 (B6) tail skin before transplanting their immunized BALB/c skin onto major histocompatibility complex–matched or syngeneic Severe Combined Immunodeficient, severe combined immunodeficiency disease (SCID) mice. These SCID mice with immunized BALB/c skin, yet devoid of any other lymphocytes, then receive another B6 skin graft to assess TRM cell–mediated alloreactivity. Interestingly, SCID mice receiving such immunized (TRM cell–containing) BALB/c grafts reliably rejected subsequent B6 allografts when placed inside and outside the BALB/c skin area. Mice receiving BALB/c skin 30 d after immunization displayed quick rejection (15 d) irrespective of secondary B6 graft placement, whereas recipients of BALB/c skin transplanted 100 d after immunization showed slower rejection in a site/distance-dependent manner (20 d inside versus 50 d outside of the BALB/c skin area). The authors ascribe this finding to the decreasing number of TRM cells in immunized BALB/c skin over this time period. Even third-party major histocompatibility complex–mismatched skin grafts were rejected within the immunized BALB/c skin area, which may be related to long-lived polyclonal and/or bystander skin-resident T cells or TRM cross-reactivity.

Although these experiments underscore the ability of efficient alloimmunity via skin–TRM cells in this otherwise (T and B cell–) immunodeficient mouse model, no detectable steady-state TRM cells were found when using a parabiosis mouse model. However, recirculation and expansion were observed upon secondary allogeneic B6 skin transfer, with spiking numbers of activated/memory CD4+ and CD8+ T cells in the spleens and grafts at 10 d posttransplant. CD4+ T cells outweighed CD8+ T cells both in the spleens (11.2% versus 6.8%) and skin grafts (11.2% versus 3.9%) during rejection. Upon selective CD4+ or CD8+ TRM reconstitution of SCID mice via adoptive transfer, CD8+ TRM cells failed to reject subsequent B6 skin grafts in the absence of CD4+ T cell help. Strikingly, CD4+ TRM cells were able to independently reject B6 skin and B6 heart allografts. Further characterization of the induced alloreactive CD4+ TRM cells via RNA-sequencing, network analysis, and gene set enrichment analysis revealed a T helper cell 17 (TH17)–like profile. In support of their role in this model, TH17-specific interference via inhibition of retinoic acid-related orphan receptor gamma or use of immunized interleukin-17knockout donors resulted in prolonged graft survival.

This study underscores the capacity of induced alloreactive CD4+ TRM to reject both skin and heart allografts independently from other lymphocytes. CD4+ TRM developed a TH17-like phenotype and blockade of retinoic acid-related orphan receptor gamma and interleukin-17 signaling attenuated rejection. Targeting TRM may therefore prove a useful tool in transplantation, and one that may be harnessed using emerging organ perfusion strategies.


1. Tsutsui H, Ziada KM, Schoenhagen P, et al. Lumen loss in transplant coronary artery disease is a biphasic process involving early intimal thickening and late constrictive remodeling: results from a 5-year serial intravascular ultrasound study. Circulation. 2001;104(6):653–657.
2. Moayedi Y, Teuteberg JJ. Predicting where patients will be, rather than just seeing where they are: establishing trajectories of cardiac allograft vasculopathy. Circulation. 2020;141(24):1968–1970.
3. Loupy A, Toquet C, Rouvier P, et al. Late failing heart allografts: pathology of cardiac allograft vasculopathy and association with antibody-mediated rejection. Am J Transplant. 2016;16(1):111–120.
4. Peyster EG, Janowczyk A, Swamidoss A, et al. Computational analysis of routine biopsies improves diagnosis and prediction of cardiac allograft vasculopathy. Circulation. 2022;145(21):1563–1577.


1. Gebhardt T, Wakim LM, Eidsmo L, et al. Memory T cells in nonlymphoid tissue that provide enhanced local immunity during infection with herpes simplex virus. Nat Immunol. 2009;10(5):524–530.
2. Bartolomé-Casado R, Landsverk OJB, Chauhan SK, et al. Resident memory CD8 T cells persist for years in human small intestine. J Exp Med. 2019;216(10):2412–2426.
3. Pallett LJ, Burton AR, Amin OE, et al. Longevity and replenishment of human liver-resident memory T cells and mononuclear phagocytes. J Exp Med. 2020;217(9):e20200050.
4. Tian Q, Zhang Z, Tan L, et al. Skin and heart allograft rejection solely by long-lived alloreactive TRM cells in skin of severe combined immunodeficient mice. Sci Adv. 2022;8(4):eabk0270.
Copyright © 2022 Wolters Kluwer Health, Inc. All rights reserved.