The data underlying this article are freely available and have been deposited in the synapse public repository (https://doi.org/10.7303/syn53033407). A sign-in process is required to access the data. Additional information is available from the corresponding author upon request. Detailed methods are provided in the Supplemental Material.The pig model, organ procurement, and xenotransplantation procedures implemented for this experiment received ethical approval and have been described previously.9Briefly, the first xenotransplantation was performed on a 72-year-old man with a history of coronary artery bypass graft surgery and heart failure who received a 10-month-old pig heart. The second recipient was a 64-year-old woman with previous kidney and pancreas allotransplants, maintained on mycophenolate and prednisone, and received an 11-month-old pig heart. Total ischemia times were 4 hours and 21 minutes for the first recipient and 3 hours and 30 minutes for the second. Both recipients received preoperative methylprednisolone and rATG (thymoglobulin), with an additional thymoglobulin dose 24 hours after transplantation for the second recipient. Maintenance immunosuppression included methylprednisolone and eculizumab.The study protocol was approved by the New York University (NYU) Langone Health Research on Decedents Oversight Committee (RDOC 21-003) and the NYU Grossman School of Medicine institutional animal care and use committee (IA18-01515). Written authorization was obtained from the authorized parties of the 2 brain-dead decedents. Given that the study protocol was deemed research on decedents, no institutional review board approval was necessary. Surgical procedures on porcine models were performed in accordance with the Animal Research: Reporting of In Vivo Experiments guidelines,11 the guidelines of the French ministries of agriculture and research, and those of the local institutional committee for the use and care of laboratory animals. Further details regarding organ selection, procurement, and the xenotransplant procedure are shown in Methods S1 and S2.Genetically modified heart xenografts samples included wedge biopsy specimens obtained 66 hours after reperfusion. Left and right ventricles as well as coronary regions were sampled. Control samples included xenografts before implantation, wild-type (WT) nontransplanted pig hearts, WT nontransplanted pig hearts with ischemia injury, WT nontransplanted pig hearts with cold ischemia and reperfusion, and WT pig hearts with brain death, cold ischemia and reperfusion. Further details describing the control model procedure are shown in Methods S3 and summarized in Table S1.
All biopsy specimens from the study were analyzed through a multimodal integrative approach including histological and immunophenotyping assessment, ultrastructural evaluation, multiplex immunophenotyping, automated image analysis, and transcriptomic profiling to elucidate potential early xenoimmune response and rejection processes (Figure 1).
Multimodal phenotyping of the heart xenografts after transplantation. All biopsies included in the study were processed following a multimodal integrative approach, including: (1) morphological evaluation by histological analysis, (2) immunophenotyping of immune deposits and cell infiltrates, (3) ultrastructural assessment, (4) multiplex immunofluorescence and deep learning–based automated image analysis, and (5) and gene expression profiling. Parts of the figure were created with BioRender.
Formalin-fixed paraffin-embedded biopsy specimens of xenografts and controls were stained with hematoxylin-eosin for histological assessment and by immunohistochemistry for immunophenotyping. Anti-C4d, anti-CD3, anti-CD20, anti-CD15, and anti-CD68 antibodies were used to stain C4d, T cells, B cells, neutrophils, monocytes, and macrophages respectively, as detailed in Methods S4 and Table S2.
All specimens included in the study were evaluated by 2 expert cardiac pathologists (P.B. and A.S.) blinded to clinical and molecular data.
Ultrastructural assessment of heart specimens was performed by electron microscopy. After sampling, xenograft biopsy specimens were immediately fixed in 2.5% glutaraldehyde. Once fixed, small pieces of the specimens were embedded in resins. Ultrathin sections (70 nm) were cut and stained with uranyl acetate and lead citrate in a standard manner. They were imaged using a Talos L120C transmission electron microscope (Thermo Fisher Scientific, Hillsboro, OR), and the ultrastructure of the heart xenografts was examined by an expert cardiac pathologist (P.B.). Further details are described in Methods S5.
Formalin-fixed paraffin-embedded biopsy specimens of xenografts and controls were stained for CD31, CD68, CD15, CD3, and CD20 in a sequential way using antibody retrieval, antibody blocking solution, primary antibodies (references and conditions are detailed in Table S2), secondary polymers, and Opal fluorochromes. Finally, slides were counterstained with spectral 4’,6-diamidino-2-phenylindole (DAPI; Akoya Biosciences, MA). Stained slides were digitized under fluorescence conditions and further analyzed using the digital pathology software HALO (Indica Labs).
We developed an end-to-end image analysis pipeline using built-in segmentation and classification tools based on convolutional neural networks (CNNs). A first CNN was trained to segment whole-slide images of all samples and automatically detect interstitial capillaries. Cells were then detected based on the DAPI signal using a CNN-based nucleus segmentation tool. Finally, positivity thresholds were set for CD68, CD15, CD3, and CD20 markers, attributing a phenotype to each cell. Cell densities for extravascular and intravascular areas were exported and processed with R. Further details are shown in the Methods S6.
The total RNA was isolated from each biopsy specimen and subsequently underwent hybridization with the Banff Human Organ Transplant gene panel (NanoString Technologies, Seattle, WA) adapted for porcine-human cross-species analysis10,12 and sequenced on the NanoString nCounter MAX Analysis System (NanoString Technologies, Seattle, WA). Raw gene count data were analyzed for quality control metrics to ensure the accuracy and reliability of the sequencing data. Normalization was performed using the “remove unwanted variation” approach,13 as detailed in Methods S7 and S8.
Afterward, differential expression analysis was conducted using the DESeq function (DESeq2 package)14 using a Wald test to compare gene expression changes between conditions of interest and adjusted for multiple comparisons using the Benjamini-Hochberg method, filtering only significant genes with a false discovery rate (FDR) P value lower than 0.05. To assess the significance of the overlap between differentially expressed genes (DEGs) across all comparisons, we applied a permutation test. Finally, to elucidate potential pathophysiological mechanisms in the heart xenografts, a pathway analysis was conducted using ReactomePA.15 The overall workflow for transcriptome profiling is depicted in Figure S1.
Top-ranked genes were annotated using the UniProt and Gene Cards databases,16,17 and the molecular signatures were defined according to the pathogenesis-based transcripts and the Banff molecular report identifying transcripts associated with rejection and immune response in solid organ transplantation.18 All statistical and bioinformatic analyses were performed using R (version 4.0.5, R Foundation for Statistical Computing). Detailed methods for transcriptome analyses are shown in Methods S9.
Descriptive data are reported as mean±SD unless otherwise indicated. For transcriptomic data, differential gene expression was performed using the DESeq2 package with Wald tests, and P values were adjusted for multiple comparisons using the Benjamini-Hochberg method, considering genes with an FDR <0.05 as statistically significant. DEGs were identified from the linear fitting after calculating fold change for each comparison of interest. Overlap of DEGs across experimental conditions was evaluated using permutation testing and significance assessed as an FDR <0.05. Quantitative cell density data from multiplex immunophenotyping were compared between groups using 2-tailed Mann-Whitney U tests or Kruskal-Wallis tests where appropriate. Correlations between histological, transcriptomic, and image-based features were assessed using Spearman’s rank correlation coefficient. Statistical significance was defined as a 2-tailed P<0.05. Statistical analyses were performed using R (version 4.0.5, R Foundation for Statistical Computing).
Results
The clinical evolution of the 2 xenografts has been reported previously.9 In summary, both xenografts demonstrated good cardiac function immediately after transplantation. Xenograft 1 showed a progressive decline in cardiac function postoperatively, attributed to a size mismatch between the recipient’s native heart and the donor porcine organ or perioperative cardiac xenograft dysfunction. Before transplantation, flow cytometry and complement-dependent cytotoxicity crossmatches using cells from the 10-gene–edited donor pigs confirmed that both recipients had low levels of anti-pig immunoglobulin G and M with negative cytotoxic crossmatch results. Anti-pig antibody levels were reassessed at the time of explantation and showed no significant changes compared with pretransplantation values.9
Sixty-six hours after reperfusion, both xenografts presented mild microvascular inflammation (Figure 1E1 through 1H1) and focal cardiac myocyte ischemia with some contraction band necrosis, primarily in xenograft 1 (Figure S2). Microvascular thrombosis was not observed. Coronary samples of both xenografts showed a pattern of endothelial cell swelling and detachment, more pronounced in xenograft 1 (Figure S3).
To assess complement cascade activation, we evaluated C4d deposition along capillary endothelium, a characteristic marker of antibody-mediated rejection (AMR).19 Immunochemistry revealed no detectable deposition of C4d in interstitial capillaries for either xenograft (Figure 2E2 through 2H2), whereas positive cytoplasmic C4d staining was observed in cardiac myocytes within previously identified necrotic areas (Figure S2). Left and right ventricles of both xenografts showed mild intravascular inflammation mainly composed of innate immune cells (CD68-positive and CD15-positive cells). Rare adaptive immune cells were observed, including focal T lymphocytes (CD3-positive cells) and occasional B lymphocytes (CD20-positive cells; Figure 2E3 through 2E6, 2F3 through 2F6, 2G3 through 2G6, and 2H3 through 2H6, respectively).
Histological and immunological changes occurring in heart xenografts after transplantation. Shown is histological and immunological evaluation of a wild-type nontransplanted pig heart (A); a wild-type nontransplanted pig heart with ischemia injury (B); a wild-type nontransplanted pig heart with brain death, cold, and ischemia and reperfusion (C); a xenograft before implantation (D); the left (E) and right (F) ventricle of xenograft 1 66 hours after reperfusion; and the left (G) and right (H) ventricle of xenograft 2 66 hours after reperfusion. Fields show hematoxylin-eosin-saffron staining (section 1) and C4d (section 2), CD68 (section 3), CD15 (section 4), CD3 (section 5), and CD20 (section 6) immunostaining. Black arrows show infiltrated interstitial capillaries. Scale bars=20 µm. HES indicates hematoxylin-eosin-saffron.
Xenografts before implantation and WT nontransplanted porcine controls did not present microvascular thrombosis or tissue damage (Figure 2A1 and 2D1). WT controls with ischemia revealed focal cardiac myocyte necrosis, including areas of calcified necrotic myocytes (Figure 2B1). WT controls with brain death, cold ischemia, and reperfusion presented areas of cardiac myocytes with contraction bands necrosis (Figure 2D1) and rare microthrombosis. Immunostaining revealed that previously identified necrotic areas were associated with C4d deposits in cardiac myocytes, supporting ischemia-related damage. However, no deposition of C4d in interstitial capillaries, indicative of immune-mediated complement activation, was detected. Microvascular infiltration was absent in all controls with the exception of rare CD3-positive cells (Figure 2A3 through 2A6, 2B3 through 2B6, 2C3 through 2C6, and 2D3 through 2D6).
Overall, histological assessment of xenografts revealed a pattern of mild microvascular infiltration and myocyte necrosis usually associated with perioperative ischemia/reperfusion.
In xenografts 1 and 2, capillary endothelial cells showed cytoplasmic processes protruding within the lumen, evocative of endothelial activation. Additionally, xenograft 1 presented plump endothelial cells and occasional circulating leukocytes. Assessment of the xenografts before implantation revealed normal endothelial cells (Figure 3).
Ultrastructural assessment in heart xenografts before and after transplantation. Shown is an ultrastructural assessment of a heart xenograft, showing endothelial swelling in myocardium (green arrow) as well as intravascular leukocyte cytoplasm (red arrow). Fields show xenografts before implantation (A), xenograft 1 (B) and xenograft 2 (C). Scale bars=2.5 µm for all except for A (bottom; 5 µm).
Automated whole-slide cell quantification after multiplex staining for CD31, CD68, CD15, CD3, and CD20 (Figure 4A) highlighted higher densities of CD15-positive cells (mean=8.06 cell/mm², SD=5.73) in the ventricular regions of both xenografts compared with CD68-positive cells (mean=2.18 cell/mm², SD=2.13). Right ventricles presented higher infiltration levels of innate immune cells compared with left ventricles. Low densities of CD3-positive T lymphocytes (mean=9.57×10−2 cell/mm², SD=7.88×10−2) were detected across xenografts, and no CD20-positive cell was identified. Automated detection of intravascular and extravascular cells using deep learning–based algorithms demonstrated that, on average, 51.3% of CD15-positive cells and 44.2% and CD68-positive cells were located in the interstitial capillaries, supporting the histological observations (Figure 4B).
Multiplex immunofluorescence and automated detection of intravascular inflammation. A, Whole-slide image of a xenograft biopsy specimen 66 hours after reperfusion stained with anti-CD31, anti-CD68, anti-CD15, anti-CD3, and anti-CD20 by multiplex immunofluorescence and counterstained with DAPI (top). The high-magnification field shows positive staining for CD31, CD15, and CD68, with each marker displayed on its own (middle). The same field is shown with merged markers, and with overlays indicating tissue segmentation derived intravascular (yellow) and extravascular (red) areas, as well as the detected immune cell distribution in these regions (bottom). B, Bar plot showing the densities of CD15-positive, CD68-positive, and CD3-positive cells detected in the left and right ventricles of both xenografts 66 hours after reperfusion. Densities of intravascular (yellow) and extravascular (red) cells are shown. The proportion of intravascular cells is indicated above each bar plot. Scale bars=100 µm. LV indicates left ventricle; and RV, right ventricle.
Analysis of the top significant DEGs between heart xenografts before and after reperfusion revealed increased expression of transcripts related to neutrophil activation (S100A8 and S100A9), interferon-gamma response (B2M, IFNGR1, and HLA-A), natural killer cells (HLA-E and FCGR3A/B), apoptosis (TIMP1 and ANXA1), and injury (CTSS) in xenografts after transplantation (Figure 5A).
Molecular profiling of heart xenografts after transplantation. Volcano plots represent the differential gene expression analysis of xenografts 66 hours after reperfusion compared with controls of interest: a xenograft before transplantation (A), a wild-type (WT) nontransplanted pig heart (B), a WT nontransplanted pig heart with ischemia injury (C), a WT nontransplanted pig heart with cold ischemia and reperfusion (D), and a WT nontransplanted pig heart with brain death, cold, and ischemia and reperfusion (E). Dots represent individual transcripts, and significant genes are labeled (adjusted P<0.05). Relative fold changes of expression are represented on the x axis and compared with the relative P values of significance (y axis). Heart xenografts are compared with relative controls: genetically modified xenografts before implantation, wild-type nontransplanted pig heart tissues with ischemia, and a wild-type heart in a brain death model.
When compared with WT pig hearts, heart xenografts showed an enrichment of transcripts associated with neutrophil activation (S100A8 and S100A8), monocyte and macrophage activation (CD163), innate immune response (FCGR2A and PTPRC), endothelium (THBD), apoptosis (TIMP1, ANXA1, and ARHGDIB), injury (CTSS), interferon-gamma response (B2M and HLA-A), natural killer cells (HLA-E and FCGR3A/B), interleukin related (IL1R2), and antibody-mediated injury (HLA-B) (Figure 5B).
When compared with WT pig hearts with ischemia injury, heart xenografts presented increased expression of transcripts related to injury (FOS, SERPINE1, NFKBIZ, and STAT3), neutrophil activation (S100A8), complement cascade (CD46), monocyte and macrophage activation (CD163), and B cells (BCL6) (Figure 5C).
When compared with WT pig hearts with cold ischemia and reperfusion, heart xenografts 66 hours after reperfusion were enriched with transcripts related to neutrophil activation (S100A8 and S100A9), antibody-mediated injury (HLA-B), monocyte and macrophage activation (CD163), interferon-gamma response (B2M and HLA-A), endothelium (THBD), injury (CTSS), natural killer cells (HLA-E and FCGR3A/B), and apoptosis (TIMP1, ANXA1, and ARHGDIB) (Figure 5D).
Finally, compared with WT pig hearts with brain death, cold ischemia, and reperfusion, heart xenografts presented a similar pattern of upregulated transcripts related to monocyte and macrophage activation (CD163), endothelium (THBD), injury (CTSS), interferon-gamma response (HLA-A and IFITM2), apoptosis (ANXA1 and ARHGDIB), and natural killer cells (FCGR3A/B). Interestingly, we detected downregulation of transcripts typically related to brain death components (IRF1 and IRF8),20 TNF-alpha (TNFAIP3), and regulation of MAPK (DUSP2)21 in xenografts (Figure 5E). As anticipated, the transgene CD46 is significantly upregulated in xenografts compared with all WT pig heart controls.
The top DEGs in heart xenografts compared with the different control groups are listed in Table S4. Permutation testing showed a highly significant overlap (adjusted P<0.001) between the DEGs from heart xenografts compared with preimplantation biopsies and those from other key comparisons (excepted for comparison with the WT ischemia/reperfusion control), demonstrating strong concordance among these molecular signatures.
To determine whether the molecular signature observed in the xenografts was experiment specific or region specific, differential expression analyses were performed, stratifying the samples based on the xenograft experiments and ventricular regions. When compared individually with all controls, xenografts 1 and 2 showed highly similar molecular signatures, highlighting a consistent molecular profile across experiments (Figure S4). This consistent pattern was further supported by differential expression analysis between xenograft 1 and xenograft 2, which revealed no significant molecular differences (Figure S5).
On the other hand, differential expression analysis comparing the left and right ventricle with all controls revealed that the molecular pattern observed in the 2 xenografts was predominantly detected in the right ventricle, aligning with the previously described immunophenotyping results (Figure S6).
Pathway analysis of differentially expressed genes comparing heart xenografts before and after reperfusion revealed significant enrichment of pathways related to antigen crosspresentation (adjusted P=4.47E−09), MHC I antigen processing and presentation (adjusted P=4.87E−06), neutrophil degranulation (adjusted P=2.38E−04), and interferon-gamma signaling (adjusted P=1.15E−05) (Figure 6A).
Pathway analysis of heart xenografts after transplantation. Dot plots represent the top enriched pathways based on significant differentially expressed genes (false discovery rate <0.05) from xenografts 66 hours after reperfusion compared with controls of interest: a xenograft before transplantation (A), a wild-type (WT) nontransplanted pig heart (B), a WT nontransplanted pig heart with ischemia injury (C), a WT nontransplanted pig heart with cold ischemia and reperfusion (D), and a WT nontransplanted pig heart with brain death, cold, and ischemia and reperfusion (E). The x axis represents different gene categories, and each enrichment result is plotted in accordance with the gene ratio (number of genes associated with the given pathway divided by the total number of genes analyzed). The size of the dots represents the number of genes in the significantly differentially expressed gene list associated with the pathway, and the color intensity represents the P value adjusted for false discovery rate.
Comparisons of xenografts with WT pig hearts confirmed this pattern, showing enrichment in pathways associated with neutrophil degranulation (adjusted P=2.46E−12), interferon signaling (adjusted P=3.65E−10), interferon-gamma signaling (adjusted P=5.10E−11), and antigen crosspresentation (adjusted P=2.39E−11) (Figure 6B).
When comparing xenografts with WT pig hearts with ischemia injury, top upregulated pathways were related to interleukin signaling (IL-3 [interleukin-3] and IL-4 [interleukin-4]) (adjusted P=2.93E−10), Toll-like receptor cascade (adjusted P=1.53E−07), and antigen crosspresentation (adjusted P=2.39E−11) (Figure 6C).
Similarly, when compared with WT pig hearts with cold ischemia and reperfusion, upregulated pathways in xenografts were associated with neutrophil degranulation (adjusted P=3.76E−12), interleukin signaling (adjusted P=2.04E−09), and interferon-gamma response (adjusted P=3.03E−10), and antigen crosspresentation (adjusted P=4.03E−14) (Figure 6D).
Finally, the comparison of xenografts with WT pig heart with brain death, cold, and ischemia and reperfusion revealed top upregulated pathways related to neutrophil degranulation (adjusted P=1.89E−11), interferon-gamma signaling (adjusted P=3.78E−08), Toll-like receptor cascade (adjusted P=2.18E−07), and antigen crosspresentation (adjusted P=9.61E−09) (Figure 6E).
Discussion
Multimodal deep phenotyping of genetically modified pig heart xenografts transplanted into 2 brain-dead human recipients revealed a mild humoral xenoimmune response 66 hours after reperfusion. This response was characterized by microvascular infiltration dominated by innate immune cells (macrophages, monocytes, and neutrophils), accompanied by endothelial activation without C4d deposition. Molecular analysis further demonstrated upregulation of genes associated with interferon-gamma response, innate immune activation, complement cascade, endothelial activation, and tissue repair. Interestingly, both xenografts presented similar molecular profiles indicative of an early antibody-mediated process predominantly detected in the right ventricle, aligning with the immunophenotyping results.
Our study confirms that the use of genetically modified pig hearts effectively protects xenograft endothelial cells from hyperacute rejection and complement-induced lysis.8,9 The predominance of innate immune infiltration observed in heart xenografts mirrors previous findings from kidney xenograft studies in brain-dead recipients10 with a pattern of milder immune activation and injury. Interestingly, unlike findings from longer-term heart and kidney xenotransplantation experiments, these kidney samples also lacked C4d deposition despite clear histological and molecular features of AMR.8,22,23
The absence of C4d deposition despite microvascular inflammation suggests a mechanism of injury independent of the classical complement pathway. Such phenotype is also described in allotransplantation, classified as pAMR1(H+), and characterized by histologic features of AMR despite negative immunopathology.24 The presence of human complement inhibitor transgenes (CD46 and DAF) in 10-gene–edited pig model and the activation of alternative pathways, such as Fcγ receptor–mediated injury and innate immune activation via “missing self” recognition, might contribute to this phenotype.4
To better understand the origins of this immune phenotype, we considered the influence of early, nonxenogeneic-specific injury mechanisms. Given the experimental design of this study, the observed endothelial injury and immune activation likely resulted from a multifactorial process, including contributions of brain death, ischemia/reperfusion injury, and xenogeneic immune recognition.25–27
To better delineate the xenogeneic component, we included matched ischemia/reperfusion control models (both with or without brain death) that underwent identical ischemic duration and reperfusion conditions as the xenografts. Transcriptomic analysis of heart xenografts against the different control groups revealed a consistent signature of upregulated genes and pathophysiological mechanisms associated with interferon-gamma response, innate immunity, complement cascade, and endothelial activation. This molecular profile aligns with signatures characteristic of antibody-mediated rejection in the heart allograft context.28–30 Combined with the lack of microvascular infiltration in all control groups, these findings support the interpretation that the immune response observed in the xenografts reflects a de novo xenoimmune activation rather than a nonspecific effect of ischemia or brain death.
Additionally, our results highlighted distinct histological features between the 2 cardiac xenografts, reflecting their unique clinical trajectories. Xenograft 1 presented a longer total ischemia time and progressive decline in cardiac function, likely because of a size mismatch between the recipient’s native heart and the xenograft. Moreover, total ischemia time was longer in xenograft 1 compared with xenograft 2. These differences resulted in aggravated ischemic injury of the myocardium of xenograft 1, in which larger necrotic areas and greater endothelial activation were observed. Similar observations have been described by previous studies assessing these xenografts,9,28 including evidence of perioperative cardiac xenograft dysfunction in xenograft 1, highlighting the need for optimized preservation strategies to minimize ischemia-induced injury and reduce the likelihood of primary graft dysfunction in future xenotransplantation studies.31
Finally, previous studies using brain-dead models found no significant differences in tissue-resident immune cells between brain-dead donors and controls,32 reinforcing the interpretation that the immune infiltrate observed in this study is driven by a xenoimmune response rather than brain death–related effects.
In conclusion, this study confirms the presence of a specific xenoimmune response in genetically modified pig heart xenografts transplanted into brain-dead human recipients and highlights the importance of employing multimodal phenotyping and appropriate controls to enhance the assessment of ongoing immune processes. Routine clinical evaluation of heart biopsy specimens relies on standard histology, which lacks molecular and mechanistic insights, and qualitative or semiquantitative assessments of tissue lesions, limiting its ability to capture the spectrum of ongoing pathophysiological processes. To overcome these limitations, we developed an integrative strategy to elucidate the complex interactions between the host immune response and xenografts. In particular, we developed an artificial intelligence–based algorithm for automated detection of microvascular immune infiltration, providing a quantitative and reproducible assessment of a critical diagnostic marker for antibody-mediated rejection.19
Although our findings offer new insights into the heart xenograft immunomolecular landscape, it is important to acknowledge the limitations of this study. Only 2 xenotransplantations were performed and characterized, with observations limited to a 66-hour period. This short timeframe limited our ability to evaluate the adaptive immune response or de novo xenoimmune activation. Additionally, although the study did not employ a state-of-the-art immunosuppressive regimen, the limited postreperfusion window would have limited the impact of most immunosuppressive regimens used in longer protocols.22,33 Extending the duration of pig-to-human xenotransplantation studies would enable more enable more in-depth studies of immune responses.
The application of multimodal diagnostic approaches in upcoming xenotransplantation protocols may help improve graft monitoring and support timely therapeutic decision-making. In longer-term studies, combining longitudinal sampling with integrative phenotyping could contribute to a better understanding of immune response dynamics and treatment effects within timeframes that are compatible with clinical workflows. The importance of developing such diagnostic strategies was emphasized at the 2024 Banff Meeting for Allograft Pathology,34 and this need has gained additional relevance considering the recent US Food and Drug Administration–approved clinical trials of kidney xenotransplantation.35 As the field progresses toward clinical application, continued efforts to adapt precision diagnostic tools to the context of xenotransplantation will be essential.
Article Information
Drs Montgomery and Loupy initiated, designed, and coordinated the study. Drs Stern, Ayares, Khalil, Kim, Metha, Narula, Reyentovitch, Smith, Saraon, Kadosh, DiVita, Goldberg, Pass, Mangiola, Griesemer, Moazami, and Montgomery performed the xenotransplantation experiments. Drs Hauet, Giraud, Kerforne, Allain, and Tissier designed and performed experiments for pig control groups. Drs Giarraputo, Morgand, Mezine, and Certain performed immunohistochemistry and molecular experiments. Drs Giarraputo, Morgand, Mezine, Coutance, Goutaudier, Sannier, and Bruneval performed data analysis and interpretation. Drs Giarraputo, Morgand, Mezine, and Loupy wrote the manuscript. All authors reviewed the draft and provided comments and suggestions for changes. All authors approved the final manuscript.
This study was supported by academic grants from the nonprofit organization OrganX, MSD Avenir (grant No. DS-2023-0001) and the Victor et Erminia Mescle Foundation under the aegis of the Fondation pour la Recherche Médicale (grant No. PME202306017324).
This study was also supported by an academic grant from the French-Speaking Society of Transplantation and the French Foundation for Medical Research (to Drs Morgand and Goutaudier). Research grant support for the decedent studies was provided by Lung Biotechnology PBC, a wholly owned subsidiary of United Therapeutics. The pig hearts were obtained from Revivicor, Inc, a subsidiary of United Therapeutics.
Dr Montgomery is part of the eGenesis advisory board, was a strategic advisor for Recombinetics, and has received funding from Lung Biotechnologies, a subsidiary of United Therapeutics. Dr Griesemer received funding from United Therapeutics for primate research in the past.
Methods S1–S9
Figures S1–S6
Tables S1–S4
SM
Dr. Sapna A. Mehta
Infectious Disease · Transplant/Immunocompromised StatesNYU Langone HealthsapnamNew York City, NY, USA