Research Highlights

Stem Cell-derived, Fully Differentiated Islets for Type 1 Diabetes Reichman TW, Markmann JF, Odorico J, et al. N Engl J Med. 2025;393:858–868. Islet transplantation has limited applications in those with type 1 diabetes (T1D) and hypoglycemia unawareness, as it remains the only therapy capable of protecting against serious hypoglycemic events and reducing glycemic variability. However, the lack of suitable deceased donors, substantial islet loss at the time of isolation and transplantation that limits total cell dose, and ongoing requirements for chronic immunosuppression remain major obstacles preventing wider application. Stem cell-derived islets (SC-islets) represents a transformative approach,1 addressing the limitations of donor islet availability by offering a scalable and renewable source of functional islets. Recent advances in differentiation protocols,2 3-dimensional culture systems,3 and encapsulation technologies4 have markedly improved graft survival, glucose responsiveness, and immune evasion. These data show promise for translation as a durable, disease-modifying intervention for T1D. Reichman et al5 now report the phase 1–2 study VX-880-101 FORWARD study conducted in North America and Europe, assessing the safety and efficacy of zimislecel—an allogeneic, stem cell-derived, fully differentiated islet cell product—from the pharmaceutical company Vertex (Figure 1).FIGURE 1.: Stem cell-derived islets for transplantation in T1D. Created with BioRender. F, female; HbA1c, hemoglobin A1C; IEQs, islet equivalents; M, male; SHEs, severe hypoglycemic events; T1D, type 1 diabetes.Eligible participants in the study were adults with T1D, >5 y insulin dependence, impaired hypoglycemia awareness, and at least 2 serious hypoglycemic events in the past year (defined as requiring external assistance). Initial safety studies administrated zimislecel at half dose (0.4 × 10e9, part A) or full dose (parts B, C) via gravity-dependent infusion into the portal vein. Patients in part A could receive a second dose within 2 y. Infusions were staggered with subsequent participants proceeding to infusion only after the preceding one progressed to 29 d. Two participants received a half dose and 12 received a full dose. For the 12 full-dose participants, there was freedom from hypoglycemia from day +90 to 1 y, and hemoglobin A1C was <7.0%. At 90 d, there was universal C-peptide detected, greater blood sugar level stability, and increased proportion of time spent in target range compared with baseline (>70% at day +150, >93% at day +365). Insulin requirement decreased, with independence in 10 participants. The remaining 2 participants had received high-dose steroids at the peri-infusion period, which likely contributed to impaired engraftment and function. Adverse events in n = 6 included diarrhea, vomiting, ulceration, and neutropenia. One patient died at 19.5 mo from cryptococcal meningitis following complex sinus surgery, and another at 30 mo because of dementia. What dose of β cells was actually administered? Islet equivalent (IEQ) is a standardized measure used to quantify the amount of islet tissue in a preparation, representing the volume of a spherical islet that is 150 μm in diameter. Approximately 400–800 β cells are contained in 1 IEQ because of heterogeneous composition of each preparation, and depending on how the cells are counted, the purity, as well as species. The zimislecel product contained 30% ± 5% β cells, with 400 β cells per IEQ; therefore, the infusion in part A was equivalent to nearly 4300 IEQ/kg (0.4 × 10e9 × 30% purity ÷ 400 IEQ ÷ 70 kg). Previous functional and minimal mass studies cited in the accompanying protocol demonstrated comparable function between SC-islets and primary human islets from deceased donors, suggesting that 2.5 × 10e6 equated to 1000–1800 IEQ, suggesting the dose in part A could range from 2200 to 4200 IEQ/kg. Most deceased donor islet transplant programs will release islets for transplantation at 5000 IEQ/kg (with lower IEQ suitable for subsequent transplantation). These data are encouraging, with the SC-islet product recapitulating the efficacy of deceased-donor islets. However, the caveats of ongoing immunosuppression and limited follow-up (1 y) remain considerable barriers for widespread application without further study. Nevertheless, this trial successfully circumvents limits imposed by poor supply of suitable deceased donors, and the stringent isolation procedure that reduces primary islet numbers and compromises viability at infusion. These results also provide an interesting foil to the current work in the field—hypoimmune islets,6 drug reprogramming to generate pluripotent cells,7,8 albeit at different stages of reporting in large animal models,9 and single patients,10,11 respectively. Although zimislecel is being touted in the media as a “cure” for T1D, we need to be circumspect with this taxonomy. Patients are still immunosuppressed—T-cell depletion with the first infusion (and basiliximab—the monoclonal antibody to the α chain of the interleukin-2 receptor—for a subsequent infusion) in conjunction with tacrolimus and sirolimus. The risk of calcineurin-inhibitor-related nephropathy12 and long-term immunosuppression remains. Nevertheless, this study has an enormous impact on the field, with the capacity to expand the application of beta cell replacement and further modifications of cell products enticingly aimed toward immunosuppression-free regimens and the clinical potential of tolerance. Chemotaxis Overrides the Killing Response in Alloreactive CTLs, Providing Vascular Immune Privilege During Cellular Rejection Barba T, Oberbarnscheidt M, Franck G, et al. J Clin Invest. 2025;135:e155191. Rejection remains a central challenge for transplantation medicine, potentially culminating in irreversible graft injury and limiting longevity. Our most elementary understanding of allograft rejection involves recipient recognition of (nonself) major histocompatibility complex antigens resulting in donor-specific antibodies and/or development of effector T cells. Donor-specific antibodies are confined to the systemic circulation, whereas T cells extravasate through the vascular endothelium into the interstitium and specifically target parenchymal cells. Histologically (and not surprisingly), the endothelium is usually spared despite identical antigenic culpability. This compartmental dichotomy is explored by Barba et al1 (Figure 1) who provide an initial exploration of microarray datasets comparing T-cell–mediated rejection (TCMR) with antibody-mediated rejection in renal allografts, demonstrating enriched endothelial signals only in the latter (and epithelial in the former). Endothelial damage results in endothelial to mesenchymal transition, which is characterized by pathological phenotypic changes (cytoskeletal reorganization, eg, fascin expression) and transcriptional modifications (eg, upregulated Snail superfamily, which dictates cellular adhesion, and the collagen-specific chaperones, including heat shock protein 47), with clear overexpression of these factors in renal allografts exhibiting concurrent antibody-mediated rejection. The authors excluded reductionist explanations such as replacement of endothelial cell (EC) by recipient cells (therefore not alloantigenic) and lack of interactions between donor microvasculature and recipient T cells. In an in vitro coculture model comprising human alloreactive cytotoxic T cells with either human glomerular EC or renal tubular epithelial cells, microvascular EC displayed equivalent HLA expression and demonstrated no survival advantage, thereby suggesting the absence of intrinsic defence against T-cell cytotoxicity.FIGURE 1.: Illustration showing a brief overview of the chemotactic gradient mechanism in TCMR proposed by Barba et al1 that confers relative immunity to endothelial cells in graft vasculature compared with epithelial cells in the interstitium because of active migration of the cytotoxic T cells in response to a chemokine gradient. Created with BioRender. AMR, antibody-mediated rejection; TCMR, T-cell–mediated rejection.The authors then used a murine kidney transplantation model using ovalbumin-transgenic C57BL/6 donor kidneys into B6 mice. Adoptive transfer of ovalbumin T-cell receptor 1 CD8+ T cells that recognize ovalbumin peptide presented in the context of the major histocompatibility complex class I molecule H-2Kb (with cotransferred P14 T cell receptor-transgenic T cells as controls) followed by intravital microscopy, demonstrated higher arrest coefficients and interactions with endothelial cells and extravascular compartments (compared with epithelial cells and intravascular spaces, respectively). These altered dynamics were further explored by single-cell RNA sequencing to distinguish the chemokines involved in this process, finding massive interstitial upregulation of cysteine-X-cysteine motif chemokine 12 (CXCL12) compared with other graft compartments. Static and dynamic in vitro assays (to mimic the interstitium and circulation, respectively) were conducted and showed effector T cells reacting to a CXCL12 chemotactic gradient, actively migrating in a dose-dependent manner through EC, which were less damaged as a result. The result was not specific to CXCL12 and could be replicated using other chemokines. Analysis of human renal allograft with TCMR (and the presence or absence of endothelialitis) demonstrated that disruption of this chemokine gradient was associated with severe EC damage. How EC avoids alloimmune injury has been the subject of conjecture and limited proof-of-concept, despite publications reporting contributions from CD95 (also known as Fas)2 and CD313 expression and limited antigenic presentation capability.4 In fact, it is the chemotactic rate of T-cell migration that determine EC (or any cell) fate as changing cytoskeletal responses to the chemotactic gradient makes it difficult for T cells to form active cytotoxic synapses that precipitates cellular damage. Is vascular immune privilege a myth? This study proposes a chemokine-dependent explanation that offers a different perspective to the traditional cell-intrinsic immune privilege mechanisms, but more work is required to further comprehend what other factors might contribute to relative EC-based immunity during TCMR and whether (or if) it can be leveraged for therapeutic benefit. We also presume this phenomenon is not exclusive to transplantation—there is relevance to infection and autoimmunity, where T cells only transit the endothelium to reach other cellular targets despite shared antigenic expression. And the converse may be true—does endothelial damage in sepsis and preeclampsia, for example, partly represent disrupted chemokine gradients?