AI-generated summary. Verify each item against the source before relying on it.
This issue tracks the transplant pathway from the donor organ to long-term survivorship. On the donor side, a US nationwide analysis shows how machine perfusion and normothermic regional perfusion have driven a rapid expansion of donation-after-circulatory-death combined liver-kidney transplantation without a survival penalty. Two studies address access at the front of the queue: a randomised trial of a self-directed digital tool to promote living donation, and a controlled cohort of interleukin-6 blockade as a desensitisation adjunct for the most highly sensitised candidates. A rejection and biomarker thread runs through the middle of the issue — preformed anti-DQα donor-specific antibodies as an under-recognised driver of antibody-mediated rejection, molecular transcriptomics recasting the ambiguous Banff 2022 categories, and deep-learning morphometry extracting prognostic signal from lesion-free protocol biopsies. The issue closes on the recipient: robotic transplantation challenging body-mass-index exclusion, and an English population study quantifying the long shadow of cancer mortality after transplantation. Together they map a field working simultaneously to widen the donor pool, sharpen how rejection is detected, and confront the cumulative morbidity of immunosuppression.
1. Revisiting Simultaneous Liver and Kidney Transplantation from Donors After Circulatory Death in the Era of Machine Perfusion Technologies: A US Nationwide Analysis of 10,687 Cases.
Kusakabe J, Fernandes E, Bhamidimarri KR, Hussein A, Kumar K, Refaai K, Yokoyama M, Arosemena L, O’Brien C, Zervos X, Abdalla A, Saith S, Rhazouani S, Agrawal N, Pinna A. Ann Surg, 2026. PMID: 42393772.
This retrospective national cohort used the UNOS STAR files to characterise how the adoption of machine perfusion and normothermic regional perfusion has reshaped donation-after-circulatory-death (DCD) simultaneous liver-kidney transplantation (SLKT) in the United States. The authors analysed 10,687 adult primary SLKT performed between 2000 and 2025, comparing DCD with donation-after-brain-death (DBD) grafts in the contemporary 2020–2025 window using propensity-score matching and Kaplan-Meier survival estimates. Use of DCD donors climbed sharply after 2018 and accounted for 29.3% of all SLKT by 2024, by which point liver machine perfusion, kidney machine perfusion, and normothermic regional perfusion were applied in 58.1%, 82.9%, and 40.1% of DCD-SLKT cases respectively. Before matching, DBD recipients carried the higher baseline burden — more often dialysis-dependent at transplant (69.1% versus 54.2%), more frequently hospitalised preoperatively (51.0% versus 17.2%), and with higher MELD scores (30 versus 23) — while follow-up was shorter for DCD grafts, reflecting their more recent uptake. Liver graft and patient survival were comparable between DCD and DBD recipients both before and after propensity-score matching. The authors conclude that DCD utilisation for SLKT has grown in step with perfusion technology and yields outcomes equivalent to DBD in appropriately selected recipients. As a registry study with limited follow-up in the DCD arm and residual confounding, it establishes contemporaneous equivalence rather than long-term parity. For UK practice, where DCD donation and normothermic perfusion programmes are well established, it supports extending combined liver-kidney transplantation to selected DCD donors as a means of expanding a scarce multi-organ resource.
2. Randomized Controlled Trial of a Digital Intervention (KidneyTIME) to Promote Live Donor Kidney Transplantation.
Kayler LK, Koizumi N, Nie J, Keller M, Cadzow RB, Noyes K, Von Visger J, Gardiner H, Feeley TH. Clin Transplant, 2026. PMID: 42383298.
This single-centre, individually randomised, parallel-group trial tested whether a patient-led digital tool could increase living-donor kidney transplantation (LDKT) among candidates at a safety-net hospital transplant programme in Buffalo, New York. Between April 2022 and July 2023, 422 adults in the referral, evaluation, or listed phases (mean age 54 years, 57% male, 45% with household income below $30,000, 36% Black) were randomised 1:1 to KidneyTIME — animated educational videos plus shareable outreach content delivered by email or text every three weeks over twelve months — or to an active-control video. The primary intention-to-treat outcome, at least one live-donor inquiry within twelve months, was not improved (sub-distribution hazard ratio 0.84, 95% CI 0.52–1.36; P=0.474) in a competing-risks analysis adjusted for computer ownership and race. Intervention participants did, however, gain more LDKT knowledge (P=0.002, a medium effect that reached large magnitude in low-literacy and no-computer subgroups), enacted more new transplant-listing behaviours (incidence rate ratio 1.62, 95% CI 1.16–2.27; P=0.005), and shared educational videos more widely, while donor-outreach behaviours showed a ceiling effect in both arms. A planned as-treated analysis was more encouraging: the 58% of intervention participants who watched at least one optional video had a markedly higher cumulative donor-inquiry probability than those who did not (22.9% versus 9.8%; sub-distribution hazard ratio 2.52, 95% CI 1.10–5.77; P=0.029). High attrition, single-centre recruitment, and probable contamination between arms limit inference, and the dose effect is susceptible to self-selection. For UK practice, it offers cautious support for low-intensity, scalable digital education to raise knowledge and listing engagement in disadvantaged groups, while underlining that patient engagement — not content — is the rate-limiting step for converting education into living donors.
3. Desensitization With Tocilizumab and IVIg: A Prospective Controlled Cohort Study in a Predominantly Black American Population.
Amato CJ, Seelam SR, Philogene MC, Paulus A, Gupta G, Moinuddin I. Transplantation, 2026. PMID: 42378336.
This prospective controlled cohort evaluated the interleukin-6 receptor antagonist tocilizumab as an adjunct to intravenous immunoglobulin (IVIg) for reducing HLA antibody burden in highly sensitised kidney transplant candidates, a group for whom desensitisation options remain limited. Thirty-four candidates with a median calculated panel-reactive antibody (cPRA) of 98.0% were studied: 19 received tocilizumab plus IVIg, all after failing at least three months of IVIg-based desensitisation, and 15 did not. Tocilizumab-treated patients were far more likely to show any reduction in cPRA during the study (84.2% versus 33.3%; P=0.002), with a median nadir cPRA change of −0.9% versus 0.0% (P=0.02), although the difference was no longer significant by study end. Treated patients more often achieved removal of high-impact unacceptable antigens (median cPRA of removed antigens 31.7% versus 0.0%; P=0.004), and three were transplanted across antigens previously listed as unacceptable. Transplant rates did not differ significantly (89.5% versus 73.3%; P=0.4), and early post-transplant outcomes — rejection, graft loss, BK viraemia, renal function, de novo donor-specific antibody, and antibody rebound — were similar between groups. The small, non-randomised design and the modest, partly transient antibody reductions temper the findings. For UK practice, where highly sensitised patients can languish despite allocation priority, this adds early support for IL-6 blockade as a desensitisation adjunct while making clear it should remain within trials and specialist protocols rather than routine use.
4. Preformed Anti-DQ Alpha Donor-specific Antibodies and the Risk of Antibody-mediated Rejection After Kidney Transplantation.
Ursule-Dufait C, Aubert O, Matignon M, Anglicheau D, Hertig A, Tortonese S, Usureau C, Lhotte R, Devriese M, Taupin JL, Snanoudj R. Transplantation, 2026. PMID: 42378346.
This multicentre study examined a frequently overlooked category of donor-specific antibody: those directed at the polymorphic α-chain of HLA-DQ, rather than the more commonly assessed β-chain. Across three French centres, 2,041 patients transplanted between 2010 and 2015 underwent four-digit HLA typing and single-antigen-bead mean fluorescence intensity analysis to resolve the precise specificity of anti-DQ antibodies, with Cox models relating preformed anti-DQα antibodies to antibody-mediated rejection (AMR) and a matched comparison of three-month biopsy microvascular inflammation against recipients without donor-specific antibodies. Among 122 recipients carrying both anti-DQα and anti-DQβ antibodies, the α-directed component made up 35.2% of anti-DQ antibodies, with a median peak mean fluorescence intensity of 1,331 (IQR 677–2,896), most frequently targeting the DQA1*05 antigen (83.7%) and consistently recognising the donor α/β heterodimer. In multivariable analysis, preformed anti-DQα antibodies independently predicted AMR after accounting for antibodies at other loci (hazard ratio 2.61, 95% CI 1.31–6.01; P=0.025), and matched recipients with these antibodies had significantly higher microvascular inflammation scores on three-month protocol biopsy (P<0.001). As a retrospective analysis of a historical cohort, it demonstrates association rather than causation. For UK practice it argues for ensuring that HLA-DQ typing and antibody characterisation capture the α-chain — not the β-chain alone — when defining unacceptable antigens and interpreting the virtual crossmatch, so that a real immunological risk is not missed.
5. Biopsy-based Transcriptomics in Banff 2022 Antibody Mediated Rejection Categories.
Hruba P, Girmanova E, Harmacek D, Novotny M, Hanzal V, Jaklova K, Hübel K, Rho E, Lopez KC, Westphal F, Gaspert A, Helmchen BM, von Moos S, Weidmann L, Kment M, Voska L, Schachtner T, Viklicky O. Kidney Int Rep, 2026. PMID: 42254850.
This two-centre study asked how biopsy-based transcriptomics maps onto the newer and more ambiguous Banff 2022 antibody-mediated rejection (AMR) categories — probable AMR, and microvascular inflammation without donor-specific antibodies or C4d — which have not previously been validated molecularly. The authors analysed 562 kidney allograft biopsies from Prague and Zurich by both histology and the Molecular Microscope Diagnostic System, grouping them into active AMR-spectrum lesions, combined active-chronic lesions with transplant glomerulopathy, chronic AMR, and non-rejection controls. Molecular AMR was detected in a graded fashion across the spectrum — 24% of probable AMR, 43% of donor-specific-antibody-negative/C4d-negative microvascular inflammation, 51% of active AMR, and 63% of chronic-active AMR — but was rare in chronic AMR (6%), and the AMR classifier scores in chronic AMR were indistinguishable from controls. Notably, the active AMR, chronic-active AMR, and antibody-negative microvascular inflammation groups shared comparable AMR scores and all carried elevated T-cell-mediated rejection scores relative to probable AMR and controls. For prognosis, a LASSO-penalised Cox model predicting three-year graft survival performed similarly whether built on molecular or histological features (concordance 0.84 versus 0.82 in Zurich, 0.78 versus 0.75 on external validation in Prague). The authors conclude that transcriptomics chiefly reflects microvascular inflammation rather than antibody status and is most informative in probable AMR and microvascular-inflammation-positive biopsies. For UK practice it supports selective use of molecular diagnostics to clarify equivocal Banff 2022 biopsies, while cautioning that it does not outperform histology for predicting graft loss.
6. Deep Learning Morphometric Analysis on Protocol Biopsies Predicts Future Graft Function.
Ben Haberou M, Bard P, Gibier JB, Pereira De Almeida HG, Bamoulid J, Maanaoui M, Koenig A, Rabeyrin M, Gazeu A, Ezzahid S, Buron F, Picard C, Courivaud C, Felix SA, Paindavoine M, Tinel C, Lenain R, Martin L, Tarris G, Ansart M, Legendre M. Kidney Int Rep, 2026. PMID: 42305258.
This retrospective study tested whether automated deep-learning morphometry could extract prognostic information from protocol kidney transplant biopsies that lack the specific lesions the Banff classification depends on. Eight deep-learning algorithms extracted 23 morphometric parameters from whole-slide images, and ten machine-learning models were trained to predict estimated glomerular filtration rate (eGFR) at three years, with an internal training/test split and a separate external application cohort. Across 367 patients the mean three-year eGFR was 53±23 and 53±22 mL/min per 1.73 m² in the two cohorts. As expected, eGFR correlated negatively with interstitial fibrosis (−0.33), tubular atrophy (−0.39), and arterial luminal stenosis (−0.29), and positively with glomerular density (0.16) and glomerular epithelial (0.33), endothelial (0.30), and mesangial (0.25) cell densities. Kernel Ridge and Bayesian models gave the best predictions (mean absolute error 11±1 mL/min per 1.73 m²), and on external validation the Bayesian model retained a reasonable association with observed eGFR (mean absolute error 13±11, correlation 0.68; P<0.001), with no significant difference between predicted and observed values after Bland-Altman bias correction (P=0.953). The authors frame automated morphometry as a way to derive quantitative prognosis from otherwise unremarkable protocol biopsies. As a retrospective proof of concept with a mean prediction error of roughly 11–13 mL/min — clinically wide at the individual level — it is not yet ready for the clinic. For UK practice it signals that AI-assisted biopsy analysis may eventually add prognostic value beyond semiquantitative scoring, but requires prospective validation before it informs surveillance decisions.
7. Safety in Robotic Kidney Transplant for All-Comers: Does Body Mass Index Need to be Considered?
Kadri H, Maffei R, Hansen K, Yoeli D, Di Napoli M, Yoshida C, Lyons S, Giusti S, Montague T, Cooper J, Choudhury R, Conzen K, Adams M, Kennealey P, Bak T, Schold JD, Nydam T, Pshak T, Abreu P. Kidney360, 2026. PMID: 42384450.
This single-centre retrospective cohort examined whether obesity should continue to bar patients from kidney transplantation when the operation is performed robotically, given that many programmes enforce strict body-mass-index (BMI) cutoffs on surgical grounds. All 104 adults undergoing robotic kidney transplantation between November 2021 and September 2025 were stratified as BMI below 30 (n=26) versus 30 or above (n=78), with a secondary comparison of BMI 30–40 (n=53) against 40 or above (n=25). There were no significant differences between the below-30 and 30-or-above groups in operative time, ischaemia times, estimated blood loss, length of stay, delayed graft function, or 30-day readmission; the lower-BMI group actually had a higher 30-day all-cause reoperation rate (15% versus 3%; P=0.03). In the subgroup analysis, patients with morbid obesity (BMI ≥40) had intraoperative metrics and complication rates comparable to those with BMI 30–40, and graft and patient survival did not differ. The authors argue that robotic transplantation delivers comparable perioperative outcomes across BMI categories and that rigid weight-based exclusion is hard to justify in the robotic era. Single-centre design, small numbers, and follow-up limited to 30 days constrain the conclusions. For UK practice it adds to the case for offering robotic transplantation as a route to transplant obese candidates who would otherwise be declined, while making clear that larger multicentre data with longer follow-up are needed before exclusion thresholds are relaxed.