Kidney International Reports

Dysfunction of the tubulointerstitial compartment is a key driver of chronic kidney disease (CKD) progression. However, tubular function remains largely unaddressed by routine clinical assessment. Here, we show that the urine ammonium-pH index (uAPI), a composite of urinary ammonium and pH, reflects kidney tubular function and predicts kidney function decline. Using acid/base, dietary, and potassium perturbations, segmental disruption of tubular ammonium handling, mathematical modelling, and data from patients with renal tubular acidosis, we identified defective ammoniagenesis as the main uAPI determinant. The uAPI was suppressed across four kidney disease models and dissociated from GFR. Kidney proteomics and single-nucleus RNA sequencing indicated downregulation of ammoniagenesis in proteinuric and diabetic kidney disease. In type 2 diabetes patients with preserved GFR, a reduced uAPI was associated with faster kidney function decline. In three CKD cohorts, low uAPI predicted CKD progression and significantly improved risk prediction. Together, this positions the uAPI as a scalable, non-invasive measure of kidney tubular function.

Background: Chronic kidney disease (CKD) imposes persistent metabolic stress, yet whether metabolic organization can be preserved under sustained renal stress remains unclear. Essential trace elements support redox defense, mitochondrial function, and intermediary metabolism, and their homeostasis is disturbed in CKD; however, single-element supplementation has yielded heterogeneous results. We examined whether a defined multi-element formulation preserves metabolic resilience in CKD. Methods: In a randomized, double-blind, placebo-controlled trial, 80 maintenance dialysis patients received standard care alone (n = 40) or plus a trace element cocktail (TEX, 20 mL/day; n = 40). Longitudinal trajectories were analyzed with linear mixed-effects models; endpoint plasma metabolomics and fecal metagenomics were profiled in subsets. A 5/6 nephrectomy mouse model was used to quantify plasma tryptophan-pathway branches and renal spatial metabolism by MALDI imaging mass spectrometry. Results: TEX was well tolerated, with no treatment-related safety signal, and was associated with modestly more favorable creatinine and eGFR trajectories. Supplementation constrained, rather than erased, CKD-associated metabolic displacement (median |Cohen's d|, 0.25 versus 2.09 for disease effects), selectively realigning circulating metabolism toward the healthy state. The host kynurenine branch responded most strongly, whereas microbial indole metabolism and gut community structure changed little. Renal metabolic remodeling was region-specific, with preservation of selected tubular and antioxidant features. Conclusions: Trace element supplementation preserved metabolic resilience in CKD, limiting metabolic displacement without complete restoration. Preservation of metabolic organization may represent a complementary dimension of therapeutic response in CKD. Trial registration: International Traditional Medicine Clinical Trial Registry, ITMCTR2026000263 (registered 3 February 2026).

Background: Current understanding of acute kidney injury (AKI) risk factors remains largely descriptive, offering limited precision into how specific biomarker values or physiologic thresholds influence susceptibility. We aimed to synthesize knowledge from machine learning models trained across multiple health systems to identify generalizable, value-specific risk drivers and biomarker interactions contributing to AKI risk. Methods: We analyzed electronic health records (EHRs) from 785,497 adult inpatients between 2010 and 2019 across nine U.S. academic medical centers within PCORnet. Interpretable gradient boosting machine models were independently developed at each health system to quantify predictor-outcome associations. Meta-regression was applied to integrate these site-level results, characterize nonlinear value-risk relationships, and identify bivariate interactions between predictors. Results: Meta-analysis revealed consistent, value-specific risk drivers across health systems. An increase in glucose from 100 mg/dL to 140 mg/dL was associated with a 1.46-fold higher risk of AKI. Chloride and anion gap also demonstrated elevated AKI risk with risk increases overlapping portions of their reference ranges, with anion gap showing a 1.14-fold increase across 4-12 mmol/L and chloride a 1.28-fold increase across 96-100 mEq/L. Electrolytes including potassium, calcium, and sodium showed quadratic associations with AKI risk. Bivariate meta-regression identified interactions between key predictors, highlighting pathways that jointly modulate AKI risk. Conclusion: This cross-system meta-analysis synthesizes machine learning-derived evidence into clinically interpretable knowledge, revealing how specific biomarker ranges and interactions modulate AKI risk. By moving beyond surface-level associations to quantitative, generalizable physiologic thresholds, these findings provide actionable insights to enhance risk stratification and personalized prevention in hospital care.

Background Pre-implantation biopsy utility for deceased-donor kidney acceptance is controversial due to processing limitations and interpretive variability. We evaluated whether systematic histological assessment, including expert evaluation and quantification with the BanffNET automated deep learning system (17 lesions), predicts post-transplant failure under optimal laboratory conditions. Methods Biopsies from deceased donor kidneys (N=733) underwent optimized laboratory processing, whole slide imaging, expert Banff 2024 pathologic assessment, and BanffNET lesion quantification. We assessed their incremental predictive value beyond clinical characteristics for early graft dysfunction, longitudinal eGFR/UPCR trajectories, need for indication biopsies, and death-censored graft failure. Results Histopathology improved clinical model discrimination by 1-4% and increased explained variance for longitudinal trajectories by 1-2%. Continuous bivariate surface analysis, however, revealed that while histology adds marginal overall predictive accuracy, mapping continuous composite and individual BanffNET scores against the Kidney Donor Risk Index (KDRI) uncovers distinct topological risk gradients for early graft dysfunction and long-term graft failure. Conclusion While pre-implantation biopsies add minimal overall prognostic value, continuous bivariate mapping of KDRI and BanffNET scores uncovers distinct synergistic risk gradients for graft failure. These continuous clinical-histological surfaces could optimize organ salvage by precisely identifying viable high-KDRI kidneys lacking critical compounded damage and vice versa.

Autosomal dominant tubulointerstitial kidney disease (ADTKD) is a genetically heterogeneous group of diseases that regularly lead to kidney failure in mid adulthood. In ADTKD-MUC1 a distinct neoprotein generated by a frameshift mutation, MUC1-fs, is pathogenic and accumulates in distal renal tubular cells. However, the pathomechanism of the disease and the regulation of the MUC1-fs are poorly understood. Primary tubular cells from the urine of several probands were established and immortalized. The length of the repeat region of both alleles of MUC1, as well as the mutated repeat localization were precisely defined by long range sequencing. The pathogenic MUC1-fs is readily detectable in patient-derived cells and has a considerably longer half-life than the wild-type mucin 1. Targeting the secretory pathway with BRD4780, a compound previously reported as putatively therapeutic, leads to downregulation of MUC1-fs in each of our patient-derived cell model. However, wild-type mucin 1 and numerous other transcriptional pathways were also affected by BRD4780, thus being rather unselective. Tubular cells derived from the urine appear to be a suitable model to analyze the pathogenic MUC1-fs of an individual patient, enabling pharmacological studies of novel therapies.

Autosomal dominant polycystic kidney disease (ADPKD) is characterized by cysts from different nephron segments, yet their origin remains inaccessible to in vivo imaging. Building on ex vivo studies that cyst sodium concentration ([Na+]) may reflect tubular origin, we developed a non-invasive 23Na magnetic resonance imaging (MRI) approach at 7 Tesla to phenotype ADPKD cysts in vivo based on cyst sodium. In nephrectomized ADPKD kidneys, cyst fluid [Na+] closely matched 23Na MRI signal intensities, revealing two distinct cyst phenotypes: serum-like high-[Na+] cysts and low-[Na+] cysts. High-[Na+] cysts expressed the proximal tubular marker sodium-glucose cotransporter-2, whereas low-[Na+] cysts expressed markers of distal tubules and collecting ducts. The vasopressin V2 receptor (V2R), the therapeutic target of tolvaptan, was detected exclusively in low-[Na+] cysts, linking sodium phenotype to therapeutic target expression. In vivo, 23Na MRI enabled classification of 2,299 cysts in 20 ADPKD patients. High-[Na+] cysts were 1.8-fold more frequent overall. Anatomical cyst localization did not predict sodium phenotype, underscoring the need for functional imaging rather than structural inference. Patients exhibited marked interindividual variation in cyst composition, with low-[Na+] cysts comprising 4.6% to 83.9% of all cysts. Given the exclusive expression of V2R in low-[Na+] cysts, this heterogeneity may influence disease progression and therapeutic responsiveness. These findings establish 23Na MRI as a non-invasive method to phenotype ADPKD cysts in vivo and provide a functional imaging approach with potential relevance for stratifying patients and guiding future therapeutic interventions.

Background: Genomic reassessment may reveal clinically relevant diagnoses in patients with young-onset kidney failure despite established clinical diagnostic labels. We evaluated this approach after excluding clinically apparent polycystic kidney disease (PKD). Methods: Among 1,139 patients receiving maintenance dialysis at participating Japanese facilities, 355 had initiated dialysis at age [≤]50 years. After excluding 17 patients with clinically apparent PKD, 108 of 338 eligible patients underwent whole-genome sequencing through voluntary, nonrandom enrollment. Analysis included 465 nuclear candidate genes, mitochondrial DNA, and copy-number/structural variants. Clinically concordant diagnoses were adjudicated using integrated clinical and genetic evidence, with a separate pathogenic/likely pathogenic (P/LP)-supported subset. Results: Diagnoses were identified in 10/108 participants (9.3%; 95% confidence interval, 5.1-16.2), comprising seven renal and three systemic/metabolic diagnoses. Nine diagnoses were P/LP-supported (8.3%; 95% confidence interval, 4.4-15.1). The remaining diagnosis was phenotype-based clinical Dent disease associated with a CLCN5 variant of uncertain significance. Diagnoses occurred across diabetic kidney disease, nephrosclerosis, chronic glomerulonephritis, unknown-cause, and other clinical categories. In a post hoc analysis excluding three participants with low target coverage, the integrated and P/LP-supported proportions were 9/105 (8.6%) and 8/105 (7.6%). Six exploratory findings were excluded from both diagnostic numerators. Gitelman syndrome was included as a clinically manifest renal disorder, although its contribution to kidney failure remained uncertain. Conclusions: Integrated genomic assessment can uncover renal and systemic/metabolic diagnoses across conventional labels in young-onset kidney failure after exclusion of clinically apparent PKD. Nonrandom enrollment and variable sequencing coverage limit generalizability and interpretation of negative results.

Introduction: Autosomal-dominant tubulointerstitial kidney disease (ADTKD) is characterized by chronic kidney disease (CKD) with an average age of end-stage renal disease (ESRD) of approximately 45 years, bland urinary sediment, the absence of proteinuria and autosomal dominant inheritance. While several causative genes have been found, there remain families in whom no molecular diagnosis has been identified (ADTKD-NMD). Methods: We identified BICC1 truncating variants in several families with ADTKD-NMD in the Wake Forest Rare Inherited Kidney Disease Registry and then screened families in our database and other referred families for BICC1 truncating variants. We performed segregation analysis and characterized affected individuals for clinical and histopathologic phenotypes. We analyzed oligomer formation of BICC1 mutants with wild-type BICC1-, ANKS3- and ANKS6 proteins through co-immunoprecipitation and Western blotting, and we tested for posttranscriptional regulation of the BICC1 target mRNA, Dand5, in a Luciferase reporter assay. Results: We found 6 heterozygous truncating mutations in BICC1 segregating with the ADTKD phenotype in 8 independent pedigrees worldwide. Affected individuals developed kidney failure in the 6th to 7th decade of life that was characterized pathologically by tubular atrophy and interstitial fibrosis. The truncated gene products localized to cytoplasmic bodies and demonstrated various degrees of self-association or binding to the known interaction partners, ANKS3 and ANKS6. While the wild-type BICC1 gene product acts as a posttranscriptional repressor of target mRNAs, all truncation variants exhibited increased expression of substrate mRNA. Conclusions: Truncating variants in BICC1 are a novel cause of ADTKD, segregating with the disease phenotype and upregulating BICC1 target gene expression through a dominant-negative- or a gain-of-function mode of action.

Banff acute lesion scores underpin histologic classification of kidney allograft biopsies; however, biomarker studies rely on second-order associations with diagnostic categories that introduce confounding. We quantified the first-order relationships between Banff acute lesion scores and the validated urinary cell three-gene rejection signature. In 354 biopsy-urine pairs, three-gene signature scores computed using a locked regression equation incorporating absolute copy numbers of CD3E mRNA, CXCL10 mRNA, and 18S rRNA in urinary cell RNA-were related to glomerulitis (g), peritubular capillaritis (ptc), interstitial inflammation (i), and tubulitis (t). Signature scores rose monotonically with Banff acute lesion severity, with 1.5 to 1.8-fold higher odds of more severed g, ptc, i, and t (all P<0.0001), and showed good calibration. Associations remained robust for composite microvascular (g+ptc) and tubulointerstitial (i+t) indices and were strongest for severe g and t, supporting this signature as a noninvasive, quantitative readout of acute rejection pathology with immediate diagnostic applicability.

While ultra-rare missense variants in UMOD cause highly penetrant autosomal dominant tubulointerstitial kidney disease, a more frequent UMOD T62P variant conveys intermediate risk with variable penetrance. To determine whether age or polygenic risk contributes to the variable penetrance of T62P, we combined genotype and phenotype data from 882,306 individuals across the UK Biobank (discovery cohort) and the All of Us and MyCode biobanks (validation cohorts). We also analyzed the impact of aging on uromodulin processing and cellular stress in stably transfected kidney tubular cells expressing wild-type or mutant UMOD. We compared the effects of the GPS on risk of CKD between T62P carriers and non-carriers and tested for the GPS-by-T62P interaction. The UMOD T62P variant was reproducibly associated with increased risk of CKD in an age-dependent manner. Compared to wild-type, clones of T62P-expressing cells exhibited a defective uromodulin maturation profile, causing endoplasmic reticulum retention and stress. We also observed significant T62P-by-GPS interaction, with T62P carriers in the top quintile of the GPS having over 5-fold higher risk of CKD compared to population average (OR 5.17, 95%CI: 2.94-9.08, P=1.0E-08). In summary, we demonstrate that the penetrance of kidney disease in T62P carriers is strongly modified by both age and polygenic risk.

Abstract Background. Enteric hyperoxaluria (EH) causes calcium oxalate kidney stones, oxalate nephropathy, and chronic kidney disease (CKD), and can progress to end-stage kidney disease. It occurs in Crohn's disease (CD), cystic fibrosis (CF), chronic pancreatitis, celiac disease, short bowel syndrome, and after bariatric surgery. Its prevalence in these populations has not been synthesized. Methods. PROSPERO CRD420261364424. PubMed and Embase were searched to 24 August 2026 for studies reporting 24-hour urinary oxalate. Prevalence was pooled using the Freeman-Tukey double arcsine transformation with a DerSimonian-Laird random-effects model; unweighted crude prevalence is also reported. Because heterogeneity was high, the narrative synthesis stratified by testing condition is the primary result. Prespecified subgroups addressed oxalate load before collection, diagnostic threshold, ileal resection, age, and urine collection format. Results. Forty-seven studies were included. In CD (17 studies, 471 patients; 204 with EH), crude prevalence was 43.3% and pooled prevalence 49.3% (95% CI 38.9-59.8; I^2 = 95.3%). With an oxalate load, prevalence was 64.7% (pooled 74.0%, 55.9-88.6) versus 38.6% without (pooled 39.1%, 28.7-50.0). Among CD patients, EH was more frequent with ileal resection (48.8% vs 31.7%; OR 1.62, 0.91-2.89) and with steatorrhea (62.5% vs 29.3%; OR 9.88, 3.81-25.61). In CF (7 studies, 348 patients; 179 with EH), pooled prevalence was 49.8% (40.5-59.1; I^2 = 91.9%). After bariatric surgery (15 studies, 849 patients; 474 with EH), pooled prevalence was 54.7% (45.4-63.9; I^2 = 96.7%). Conclusions. Roughly half of these patients have EH, but the proportion depends heavily on whether an oxalate load precedes urine collection (39% unprovoked vs 65%-74% provoked); because most studies do not state this condition, published prevalence estimates are not directly comparable. Ileal resection and steatorrhea identify those who should be screened.

Background: Traditional real-world studies comparing SGLT2 and DPP4 inhibitors on renal outcomes rely on propensity score matching, which causes high-dimensional data loss. We used causal machine learning (Causal ML) to unmask heterogeneous treatment effects in diabetic kidney disease (DKD). Methods: Using data from 4,588 patients within the Japanese J-CKD-DB-Ex registry, we implemented a doubly robust (DR) learning framework (Linear DR-learner with XGBoost) to compare SGLT2 and DPP4 inhibitors. Outcomes included the chronic eGFR slope and a composite renal endpoint ([&ge;] 50% eGFR decline or end-stage kidney disease). Heterogeneity was explored via causal SHAP and decision trees. Results: At the population level, SGLT2 inhibitors modestly slowed chronic eGFR decline (average treatment effect [ATE] = 0.14 [95% CI: -0.86, 1.15] mL/min/1.73m^2/year) and reduced composite endpoint risk by 9% (ATE: -0.09 [-0.11, -0.08]) versus DPP4 inhibitors. However, individual-level counterfactual analysis suggested that for the chronic eGFR slope, non-glinide users with stable pre-treatment trajectories who were also taking ACE inhibitors had a greater benefit from SGLT2 inhibitors (ATE: 2.95 [-0.68, 6.58]). Conversely, glinide users with steep pre-treatment decline had a greater benefit from DPP4 inhibitors (ATE: -8.98 [-16.11, -1.85]). For composite renal events, SGLT2 inhibitors had a 28% absolute risk reduction within the algorithmically identified high-risk subgroup (eGFR [&le;] 28.1 mL/min/1.73 m^2 and positive proteinuria; ATE: -0.28 [-0.33, -0.23]). Even non-proteinuric decliners demonstrated a 8% risk reduction with SGLT2 inhibitors (ATE: -0.08 [-0.10, -0.06]). Conclusion: Causal ML advances precision medicine in DKD, shifting from uniform prescribing to individualized, data-driven therapy targeting distinct intrarenal pathways.

Background: Whether benchmark performance reflects robust clinical reasoning rather than surface-level pattern recognition remains uncertain. We evaluated the robustness of state-of-the-art large language models (LLMs) on nephrology board renewal questions using "None of the other answers" (NOTA) substitution. Methods: From 210 Japanese Society of Nephrology board renewal questions (2014-2023), two nephrologists independently reviewed all items. Questions in which NOTA became the sole correct answer after replacement were included, yielding 145 validated questions. GPT-5, GPT-4o, Gemini 2.5 Pro, and Gemini 2.0 Flash were evaluated via application programming interfaces under default settings. The primary endpoint was accuracy, and paired differences were assessed using the exact two-sided McNemar test. Results: Accuracy was significantly lower after NOTA substitution for all models: GPT-4o, 66.21% to 19.31% (drop, 46.90 percentage points [pp]); GPT-5, 87.59% to 73.10% (14.48 pp); Gemini 2.0 Flash, 58.62% to 31.03% (27.59 pp); and Gemini 2.5 Pro, 86.90% to 55.86% (31.03 pp); all P < .001. GPT-5 showed the smallest decline and the highest accuracy in both versions. Conclusions: All evaluated LLMs showed a significant robustness gap after NOTA replacement. Newer models may be more robust, but multiple-choice accuracy remains an incomplete measure of clinical reasoning robustness.

Background IgA vasculitis (IgAV) is the most common systemic vasculitis in children, and its prognosis is largely determined by renal involvement (IgAV nephritis). No routine blood test identifies IgAV or stratifies the risk of nephritis, although aberrant O-glycosylation of the IgA1 hinge region is central to its pathogenesis. We developed a mass-spectrometry assay to profile IgA1 hinge O-glycoforms and define signatures of disease activity and renal involvement. Methods IgA was affinity-purified from 5 uL of plasma from 91 children (27 with acute IgAV, 26 in remission, and 38 age-matched healthy controls; 24 with and 29 without nephritis), trypsin-digested, and hinge-region O-glycopeptides were quantified by LC-MS. Sixty-nine glycoforms were normalized to a total-IgA1 tryptic peptide. Duplicate measurements showed good analytical repeatability, with a median coefficient of variation of 6%. Groups were compared using Mann-Whitney and Kruskal-Wallis tests (Benjamini-Hochberg FDR). Discrimination was assessed by ROC analysis and cross-validated logistic regression panels. Results Acute IgAV showed broad remodeling of the hinge glycoform profile (35 glycoforms differed with excellent discrimination (AUC 0.93-0.95) for the best ones), with an increase in low-sialylated, agalactosylated species and a decrease in complex sialylated species. The profile was normalized in remission (no glycoform differed from the controls). Two distinct renal patterns emerged: disease-associated glycoforms already altered without nephritis and renal-specific glycoforms altered only in nephritis (H2N2S1, H3N3S5, H3N4S4, and H4N4S3). A four-marker panel discriminated nephritis among IgAV children with a cross-validated AUC of 0.86 (IC95 % 0.75-0.94). Conclusions A single mass-spectrometry assay, from a small blood volume, captures an IgAV-associated IgA1 hinge O-glycoform signature that normalizes in remission, together with a distinct renal involvement associated signature. These findings identify candidate IgA1 O-glycoform signatures associated with IgAV activity and documented renal involvement. Prospective longitudinal studies are required to determine whether the renal-associated panel can predict subsequent nephritis.

Introduction: Kidney volume is associated with renal function, but conventional whole-organ measurements may miss regional changes in kidney substructures. We used magnetic resonance imaging (MRI) to test whether kidney substructure volumes and three-dimensional (3D) thickness phenotypes are associated with disease, adiposity, and incident outcomes. Methods: Automated abdominal MRI segmentations from 70,270 UK Biobank participants were used to quantify left and right kidney cortex, medulla, and parenchyma volumes and mesh-based thickness maps. Mass univariate regression generated statistical parametric maps linking regional thickness to anthropometric traits and disease (CKD, type-2 diabetes [T2D], hypertension). Bidirectional causal mediation analyses were performed using kidney substructure volumetric, thickness PCs and kidney adiposity. Longitudinal changes were assessed in 2,610 participants with repeat imaging. Cox proportional hazards models tested the associations between kidney volumes and principal component (PC) scores of thickness and incident CKD, T2D, hypertension, cardiovascular disease (CVD), and mortality. Results: CKD was associated with thinning of the parenchyma and most pronounced in the medulla (87.7% left, 72.8% right), whereas associations in the cortex were mixed. In contrast, T2D and hypertension were predominantly associated with increased thickness, most extensively in the cortex (T2D: 84.39% left, 96.35% right; hypertension: 94.8% left, 99.9% right). Age-disease interactions were associated with reduced thickness across all conditions. Smaller medulla volume and lower cortical thickness mediated the association between CKD and renal sinus fat. Kidney substructure volume was inversely associated with incident CKD across all substructures and both kidney sides (p<0.05), but positively associated with incident hypertension and mortality (p<0.05). Thickness PC scores also predicted outcomes independently of volume, with PC3 in the left cortex (HR 0.91 [0.84-0.98]), reflecting inferior pole/hilar thickness, associated with incident CKD. Conclusions: MRI-derived kidney substructure volumes and 3D thickness maps provide complementary, spatially-resolved markers of kidney morphology, that are mechanistically linked to renal adiposity and future disease risk.

Background Long-horizon kidney trajectory prediction in type 2 diabetes mellitus (T2DM) is usually reported as a point estimate or event risk, although clinical decision-making also depends on whether an individual prediction is reliable. We developed an uncertainty-aware model for 48-month estimated glomerular filtration rate (eGFR) decline and tested whether conformal interval width provides a clinically structured, patient-level signal of prediction reliability. Methods We performed a secondary prognostic modeling analysis of Action to Control Cardiovascular Risk in Diabetes (ACCORD) participants with baseline and 48-month eGFR (n=6,853). The outcome was annualized eGFR change, calculated as 48-month minus baseline eGFR divided by four years. The primary baseline feature set excluded serum creatinine since eGFR is creatinine-derived, and also excluded urine biomarkers. Random forest, gradient boosting, penalized linear models, and XGBoost were compared using fixed training, calibration, and test partitions. Split and locally adaptive conformal intervals were evaluated by empirical coverage and interval width. Interval-width analyses were repeated after conditioning on baseline eGFR. Results The best primary model was random forest (R2=0.382, MAE=3.271 mL/min/1.73m2). Split 90% conformal intervals achieved empirical coverage of 0.917. Locally adaptive 90% intervals achieved empirical coverage of 0.909 with mean width 13.759 mL/min/1.73m2. In unadjusted analyses, wider intervals were associated with larger errors and more rapid decline. After interval-width quintiles were assigned within baseline-eGFR strata, wider intervals remained associated with realized prediction error (annual adjusted increase, 0.151 mL/min/1.73m2 per quintile). Beyond baseline eGFR, wider intervals were associated with younger age, female sex, higher HbA1c, higher triglycerides, and higher systolic blood pressure. Conclusions Baseline clinical variables predicted 48-month eGFR decline with good long-horizon performance in ACCORD, even after excluding serum creatinine and urine biomarkers from the primary model. Conformal prediction provided calibrated patient-specific intervals, and interval width behaved as an informative reliability phenotype rather than a random modeling artifact. These findings support a novel uncertainty-aware framing of kidney trajectory prediction in which rapid and uncertain decline can be identified from baseline clinical data.

Background: Diabetic kidney disease (DKD) is the leading cause of chronic kidney disease (CKD). However, most proteomic studies of DKD rely on bulk kidney tissue, which cannot distinguish the contribution or response of individual nephron compartments to injury. Methods: Diabetes was induced in male mice by streptozotocin (STZ) injections. After 16-weeks, the mice and vehicle-treated controls were characterized physiologically, biochemically, and histologically. A deep learning-powered Deep Visual Proteomics (DVP) pipeline, validated against manual annotation, was adapted to isolate proximal tubule (PT) and glomeruli from Megalin stained kidney sections by automated laser microdissection. Bulk kidney, PT, and glomerular proteomes were generated by data independent acquisition mass spectrometry. PT-enriched candidates were prioritized using a composite scoring approach and compared with human tubulointerstitial proteomic data from the Kidney Precision Medicine Project. Results: STZ mice developed sustained hyperglycaemia and albuminuria, alongside elevated markers of tubular injury and interstitial fibrosis. Segmentation models isolated PT and glomeruli with high fidelity (Dice coefficients 0.878 and 0.914; area correlations r=0.993 and r=0.996). Compartment-resolved proteomics determined that PT and glomeruli underwent largely distinct, non-overlapping remodelling: PT exhibited loss of proteostatic, cell cycle, and structural programs with compensatory mitochondrial and lipid metabolic upregulation, whereas glomeruli showed broad loss of oxidative metabolic capacity without any compensatory metabolic program. Fourteen of the top twenty prioritized PT candidates, including LARS2 and ANXA2, changed in the same direction in human CKD tubulointerstitial proteomic data. The STZ PT proteome correlated significantly with this human dataset, while the glomerular comparison did not. Conclusions: Compartment-resolved and deep learning-guided visual proteomics can uncover divergent, biologically coherent PT and glomerular injury programs in DKD that are masked in bulk tissue analysis. A PT injury signature was uncovered that is partially conserved in human CKD, identifying novel candidate mechanisms and biomarkers for future exploration.

Introduction: Mitochondrial DNA (mtDNA) is not routinely analyzed in inherited kidney disease. We evaluated mtDNA variation in families who remained genetically unresolved despite extensive testing. Methods: We reviewed pedigrees from the Wake Forest-Charles University Rare Inherited Kidney Disease Registry to identify genetically unresolved families with suspected maternal inheritance, performed mtDNA genotyping, clinically characterized variant carriers, and functionally evaluated disease-associated mitochondrial variants. Results: Among 33 families with evidence of maternal inheritance, 18 (55%) carried one of seven disease-associated mtDNA variant types, including homoplasmic recurrent single-nucleotide insertions in the second light-strand promoter (LSP2; 9 families), novel MT-TW and MT-TL2 variants (2 and 1 families, respectively), and previously reported MT-TF and heteroplasmic MT-ND5 variant (5 and 1 families, respectively). In 16 families, variants occurred on distinct haplotypes, consistent with independent mutational events and rapid enrichment to homoplasmy across generations. Maternal transmission was strongly supported, with below-normal kidney function observed in 54/60 (90%) offspring of affected mothers versus 1/17 (6%) offspring of affected fathers (p = 1.23 x 10e-11). Pathogenicity was further supported by predicted deleterious structural effects and functional evidence of impaired mitochondrial transcription and translation, respiratory chain deficiency, and CoQ10 depletion. Affected individuals predominantly presented with chronic tubulointerstitial kidney disease, occasionally accompanied by gout and only sporadically with extrarenal manifestations. The rate of kidney disease progression appeared to vary both between and within families. Overall, 109/119 genetically affected individuals or obligate at-risk carriers were clinically affected; most unaffected carriers were younger than 45 years of age. Clinical status was unavailable for an additional 66 obligate at-risk carriers. Conclusions: These findings establish the physiological relevance of the LSP2 promoter, support routine assessment of the mitochondrial genome in inherited kidney disease, and highlight mtDNA variants as an important cause of familial and sporadic tubulointerstitial kidney disease of previously unexplained etiology.

Background Delayed graft function (DGF) is among the most used metrics in transplant medicine, yet its prevailing definition-dialysis requirement within the first seven postoperative days lacks granularity and fails to capture the dynamic nature of graft recovery. It reflects a treatment decision rather than graft biology, varies by centre practice, and offers no temporal information about the pace or pattern of recovery. This study proposes a trajectory-based framework using serial serum creatinine measurements across three biologically defined time windows to capture the dynamic nature of early renal recovery. Methods We retrospectively analysed 205 adult kidney transplant recipients at Cleveland Clinic Weston Hospital in 2025. Donor groups included living donors, donation after brain death (DBD), donation after circulatory death with normothermic regional perfusion (DCD-NRP), and DCD without NRP. Creatinine slopes were estimated for Days 1-7, 7-30, and 30-90 using ordinary least-squares regression. Logistic regression assessed the predictive performance of each trajectory window for DGF, adjusting for Kidney Donor Profile Index, cold ischemia time, BMI, and donor type. Results DCD non-NRP kidneys showed negligible early creatinine decline (slope +0.02) compared to DBD (-0.71) and DCD-NRP (-0.57) grafts during Days 1-7. During Days 7-30, DCD-NRP kidneys showed accelerated recovery and converged toward DBD-level trajectories, whereas DCD non-NRP grafts remained sluggish. Predictive modelling identified the Day 7-30 trajectory had the strongest discrimination for DGF, with AUC 0.83, pseudo-R2 0.25, and the lowest AIC (181.73), outperforming Days 1-7 (AUC 0.75) and 30-90 (AUC 0.79). DGF rates were 0% in living donors, 25.5% in DBD, 20.7% in DCD-NRP, and 58.9% in DCD non-NRP recipients. Conclusions A trajectory-based creatinine model provides more informative prognostic value than the conventional binary DGF definition The Day 7-30 window periods offers the highest discriminatory power, and normothermic regional perfusion appears to favourably shift DCD kidney recovery trajectories in a manner the binary definition fails to detect.

Intracranial aneurysms (IA) and their rupture (subarachnoid haemorrhage, SAH) are a rare but serious complication of autosomal dominant polycystic kidney disease (ADPKD). Both genetic and environmental risk factors contribute to the pathogenesis of IA and SAH, but specific information on risk factors in the ADPKD population are limited and international clinical guidelines mainly recommend screening for ADPKD patients with a family history of IA or SAH. We assessed the associations of monogenic variants, polygenic risk, and clinical factors with the diagnosis of IA or SAH in 2,200 adult ADPKD patients from three cohorts: the Irish Kidney Gene Project (IKGP, n=475), FinnGen (n=826), and Genomics England (GEL, n=899). Polygenic risk scores (PRS) for IA, hypertension, and smoking intensity were derived from previously published GWAS summary statistics and additionally conditioned using mtCOJO to account for their genetic correlation. IA or SAH was diagnosed in 158 of 2,200 patients (7.2%; mean age at diagnosis 51.1 years). Female sex (HR 2.21, 95% CI 1.50-3.25, p=6.52x10^-5) and smoking (HR 2.06, 95% CI 1.43-2.96, p=9.49x10^-5) were associated with IA or SAH in univariate Cox proportional hazards models, while diabetes was protective (HR 0.39, 95% CI 0.22-0.70, p=1.37x10^-3). Monogenic variant status was not found to associate with increased risk of IA or SAH. Polygenic score for IA was associated with increased risk of IA or SAH, with a 1 standard deviation increase in PRS associated with a pooled hazard ratio (HR) of 1.30 (95% CI 1.09-1.57, p=4.66x10^-3), and the association of the conditioned IA PRS remained in multivariate Cox proportional hazards models accounting for clinical factors and monogenic variants (HR 1.26, 95% CI 1.03-1.55, p=0.02). The addition of polygenic scores added significant prognostic information for IA or SAH beyond clinical risk factors in FinnGen (p=3.61x10^-3) and GEL (p=8.84x10^-3) but not in IKGP, as assessed by the likelihood ratio test. Overall, our study shows that the strongest risk factors for IA or SAH in ADPKD patients are smoking history, female sex, and polygenic risk for IA. Hypertension was not associated with IA (HR 0.50, 95% CI 0.24-1.00, p=0.051), likely due to the high prevalence of hypertension in this population across cohorts (69.6-85.3%). While family history may be considered a proxy of genetic risk, this study suggests that family history itself may be of limited utility in discriminating individuals with ADPKD at risk of IA or SAH. Keywords: Intracranial Aneurysms, subarachnoid haemorrhage, Autosomal Dominant Polycystic Kidney Disease, Polygenic Risk Scores, Hypertension, Family History