Laurent Mesnard

Genome and exome sequencing have become central to diagnosing rare hereditary diseases, but each test returns thousands of variants that a clinical scientist must review by hand to find the one responsible for the patient’s condition. This manual interpretation is the main bottleneck in clinical genomics. To reduce it, we developed DiagAI, a machine-learning system that ranks the variants found in a patient and returns a short list of the most likely causal candidates. DiagAI combines three sources of evidence: a pathogenicity score from the Universal Pathogenicity Predictor (UP2), a model we trained to estimate how damaging a variant is on the five-tier scale of the American College of Medical Genetics and Genomics (ACMG); a phenotype-matching score from PhenoGenius, which weighs how well a gene’s known clinical features match the patient’s symptoms (encoded as Human Phenotype Ontology, or HPO, terms); and expert rules covering inheritance pattern and sequencing quality. We evaluated DiagAI on 966 exomes from adults investigated for kidney disease of unknown cause, of which 196 had a confirmed genetic diagnosis. We first tested UP2 on its own by ranking 62 confirmed disease-causing missense variants that were absent from its training data: UP2 placed the causal variant within the top 100 candidates in 87% of cases, compared with 61% for the widely used tool REVEL. Across the 196 diagnosed exomes, the full DiagAI shortlist contained the causal variant in 94.9% of cases when the patient’s symptoms were provided and in 90.8% when they were not, with a typical shortlist of about 10 variants. When symptoms were provided, the single top-ranked variant was the correct diagnosis in 74% of cases, versus 42% without symptoms, exceeding the performance of the established tools Exomiser and AI-MARRVEL on the same cohort. DiagAI produces compact, accurate shortlists that can reduce the manual interpretation workload as diagnostic sequencing volumes continue to grow.

Introduction: Tenofovir‑based preexposure prophylaxis (PrEP) is widely used and proved its efficacy for HIV prevention, and kidney function monitoring relies primarily on serum creatinine-based estimation of the glomerular filtration rate (eGFR). Oral creatine supplementation can increase serum creatinine without reflecting a true reduction in glomerular filtration rate (GFR). As is not widely known it can potentially lead to inappropriate PrEP discontinuation. Case description: We report three clinical cases of healthy men who have sex with men (MSM) receiving tenofovir‑based PrEP who referred to a nephrologist for declining serum creatinine‑based eGFR while taking oral creatine supplements for muscle development associated with sport. Serum creatinine compared to cystatin C-based eGFR was performed and measured GFR (DTPA‑Tc99) when available to assess kidney function. In the first case, a 45‑year‑old man taking daily PrEP and oral creatine presented with elevated serum creatinine (1.39 mg/dl) and reduced creatinine‑based eGFR (61 ml/min/1.73 m2). Cystatin C-based eGFR indicated a preserved kidney function (83 ml/min/1.73 m2). After discontinuing creatine, serum creatinine decreased to 1.17 mg/dl and creatinine‑based eGFR improved to 75 ml/min/1.73 m2, aligning with cystatin C estimates. In the second case, a 38‑year‑old man using on‑demand PrEP and taking oral creatine referred to the nephrologist with a decline in creatinine‑based eGFR from 112 to 78 ml/min/1.73 m2 over four years. Despite this apparent decline, measured GFR was in range (113 ml/min/1.73 m2) aligned with cystatin C-based eGFR (107 ml/min/1.73 m2). In the third case, a reduced creatinine-based eGFR was observed in a 41‑year‑old man who was taking daily PrEP with the supplement. Moreover, a one‑day interruption of creatine prevented this artifact, with both serum creatinine and cystatin C returning to normal ranges. A last serum creatinine measurement performed four days after creatine was reintroduced reduced the eGFR by 20 ml/min/1.73 m2. Discussion: Oral creatine supplementation, frequently used by tenofovir-based PrEP users, can artifactually elevate serum creatinine and lead to underestimation of kidney function. Cystatin C-based eGFR is not affected by creatine and so its use can be an easy tool for PrEP prescribers for an accurate assessment of kidney function. Its use may prevent risky and unjustified PrEP interruptions and reduce nephrological referral. Creatine metabolism permits a withdrawal one day before the PrEP follow-up lab tests.

While transplantation outcomes are closely linked to HLA compatibility, minor histocompatibility antigens (mHAgs)—peptides embedded in HLA molecules bearing amino acid differences within the donor/recipient pair—also contribute to T‐cell alloreactivity, especially in HLA‐matched pairs. Because mHAgs initially arise from non‐synonymous genomic polymorphisms, we and others have focused on genomic data to approximate mHAgs load. However, most teams based their strategies on genotyping data, which only allows the study of polymorphisms that are already known and well described at the population level. Nowadays, no publicly available software exists that can facilitate unbiased identification of mHAgs candidates from whole‐exome or whole‐genome sequencing data. We present AlloPipe, a bioinformatics tool designed for this purpose. AlloPipe operates in two sequential steps: Allo‐Count, which accurately identifies directional amino acid mismatches, and Allo‐Affinity, which reconstructs peptides around these mismatches and returns their affinity for HLA class I molecules. With its ability to process rapidly large datasets in modest‐scale computing infrastructures in a clinically relevant timeframe, AlloPipe makes possible the implementation of class I restricted mHAgs approximation into clinical decisions, such as immunosuppressive therapy optimization or donor selection. This open‐source software offers a flexible choice of tools at each step of the analysis, especially when predicting affinity towards HLA molecules. It is available locally and via a web interface, making potential mHAgs predictions more accessible and actionable in clinical practice.

Purpose of Review: In solid organ transplantation, HLA matching between donor and recipient is an important factor in maintaining graft survival. For this reason, national and international transplantation programmes include HLA matching into their algorithms for deceased donor organ allocation and are used in centres as guidance for maintenance immunosuppression. However, HLA matching explains only a short fraction of long term-outcomes. It is likely that assessing genome-wide compatibility could be the next step towards longer graft survival. Recent Findings: Hence in this review, when a donor allele is not present in the recipient, we define this as a genomic mismatch. We describe how the context around a genomic mismatch (i.e. protein expression, surface accessibility, immune recognition, etc.) contributes to an alloimmune risk, which we define as an effective genomic mismatch. As a review focused on the technical aspects, we outline the core components one should consider in analysing the transplant pair through a genome-wide approach, and review current literature based on these core components, contrasting with HLA matching tools. Summary: While several studies have explored the role of the genomic mismatches between a donor and recipient, the biological and immunological context of this mismatch is required. Further advancements in this field will benefit greatly from established guidelines and tools. Concluding this review, we evaluate the clinical relevance of these sequencing-based approaches, and whether this increase in scope is valuable for further insights in alloimmunity.

Crescentic glomerulonephritis (cGN) is a severe kidney pathology characterized by the aberrant proliferation of epithelial cells, leading to crescent formation within the Bowman’s space. The molecular pathways involved in crescent formation remain poorly understood despite its clinical relevance. Given the mechanical stress experienced by podocytes, likely exacerbated in cGN, we hypothesized that activation of the mechanosensor yes-associated protein 1 (YAP), an effector of the Hippo pathway, may contribute to the development of cGN. Here, we demonstrate activation of YAP and its target genes in the nephrotoxic nephritis model, a murine model of cGN. Mechanistically, podocyte-specific hyperactivation of YAP (YAP ^5SA ) in transgenic mice led to cell-autonomous hypertrophy of podocytes and non–cell-autonomous activation and proliferation of parietal epithelial cells (PECs), culminating in crescents. Transcriptomic profiling in a human podocyte cell line expressing the same YAP ^5SA isoform revealed the reactivation of developmental programs within differentiated podocytes and identified the phosphoinositide 3-kinase–protein kinase B–mechanistic target of rapamycin (PI3K-Akt-mTOR) signaling pathway as a candidate involved in YAP-induced podocyte hypertrophy. Furthermore, this analysis identified connective tissue growth factor (CTGF) and heparin-binding epidermal growth factor–like growth factor (HB-EGF) as potential mediators in the cross-talk between YAP-activated podocytes and PECs, driving PEC hyperplasia. Collectively, these findings highlight the pivotal role of YAP in the pathogenesis of cGN and indicate that targeting YAP signaling could be a promising therapeutic strategy for this severe kidney disease.

Genetic investigations in nephrology have long been viewed as the prerogative of paediatricians or restricted to archetypal genetic nephropathies with highly penetrant variants affecting young adults. However, genetic testing has emerged as a pivotal tool in the field of adult nephrology, with the ability to revolutionize the understanding and management of adult kidney diseases. Here, we explore the multifaceted role of genomic testing (such as exome or genome sequencing) in chronic kidney disease, shedding light on current genetic findings for reframing diagnostic paradigms and tailoring treatment strategies. Genomic testing has enhanced our comprehension of kidney diseases of unknown origin by showing that ~20% are attributable to kidney Mendelian genetic disorders with as yet unsuspected phenocopies. Beyond genetic counselling, genetic integration can optimize therapeutic interventions, kidney transplantation and kidney disease prevention, both in index cases and in at-risk family members. Furthermore, the emerging field of rapid nephrogenomics promises streamlined diagnosis and management, with a potential impact on early therapeutic strategy. Importantly, although costs continue to decrease, the integration of genomic technologies in nephrology practice raises several ethical concerns, especially regarding variants of uncertain significance, and incidental or secondary findings. Establishing multidisciplinary frameworks should maximize the potential of nephrogenomics to improve patient outcomes.

Background: Unlike X-linked or autosomal recessive Alport syndrome, no clear genotype/phenotype correlation has yet been demonstrated in patients carrying a single variant of COL4A3 or COL4A4. Methods: We carried out a multicentre retrospective study to assess the risk factors involved in renal survival in patients presenting a single pathogenic variant on COL4A3 or COL4A4. Results: A total of 97 patients presenting a single pathogenic variant of COL4A3 or COL4A4 were included. The prevalence of end-stage kidney disease (ESKD) during follow-up was 28.7% [median age 47.5 years (interquartile range 39.1–55.8). A total of 23 patients carried a ‘severe’ mutation (frameshift, stop gain, extensive deletion, impacting splicing) and 60 patients presented a glycine substitution in a collagenous domain. In patients with glycine missense variants, the location of the mutation in the distal exons was associated with worse renal survival with a more pronounced decline in the estimated glomerular filtration rate compared with variants in proximal exons. Conversely, the presence of a severe mutation did not impact renal survival. Conclusion: Our results confirm that autosomal dominant Alport syndrome (ADAS) can lead to ESKD. We demonstrated that a glycine substitution involving the distal exons had a negative impact on renal survival in ADAS patients, probably due to a trimerization defect. This could help improve personalized follow-up in ADAS patients with glycine substitution and could be integrated into a future prognostic score to accurately predict renal outcomes.