Predictive Determinants of Nephrotic Syndrome Remission in Patients With Focal Segmental Glomerulosclerosis or Minimal Change Disease and At-risk Polymorphism of APOL1 Gene

The APOL1 gene synthesizes Apolipoprotein L1, a transport protein for hydrophobic molecules. When this gene presents bi-allelic "at risk" polymorphisms, it has been associated with an increased risk of end-stage renal disease. These polymorphisms can be frequent on a population scale, affecting up to 32% of patients in some West African countries. The pathogenic nature of these variants is now well established and has been highlighted on several occasions, notably in HIV-associated nephropathy (HIVAN - HIV Associated Nephropathy) or more recently in COVID-19-associated nephropathy (COVAN - COVID Associated Nephropathy), where almost all patients carry APOL1 risk alleles. Irrespective of the causative agent, nephropathy associated with APOL1 risk variants is regularly revealed by a nephrotic syndrome. This syndrome is characterized by generalized edema, high-range proteinuria and acute kidney injury. The course of the disease is particularly difficult to predict and highly variable, with some patients requiring rapid and definitive dialysis while others recover normal renal function without sequelae of proteinuria. In addition to this inter-individual variability, there is also variability in clinical practice with regard to the use of high-dose corticosteroids to induce disease remission. This is largely due to the rarity of the disease in the West and its possible under-diagnosis in developing countries. However, new drugs specifically targeting APOL1 are currently being developed, but their use is currently restricted to studies that exclude the most severe patients. In patients with at-risk APOL1 variants and nephrotic syndromes, it therefore appears critical to identify the determinants associated with remission of nephropathy, to quantify the efficacy of current therapies and to facilitate access to emerging drugs for the most severely ill subjects.

The main aim of this study is to evaluate the predictors (drug, clinical, histological or biological parameters) associated with nephrotic syndrome remission in patients with nephrotic syndrome, biopsy proven focal segmental glomerulosclerosis or minimal change disease and an at-risk variant of the APOL1 gene.

Population involved: eligible patients will be enrolled if they are more than 18 years old, have nephrotic syndrome (serum albumin < 30 g/L and urine protein creatinine ratio > 3 g/g) and biopsy-proven focal segmental glomerulosclerosis or minimal change disease.

Data analysis: the association of each explanatory variable with the variable "remission of nephrotic syndrome" will first be assessed using a semi-parametric univariable Cox model.Then, a multivariable Cox model will be performed with the explanatory variables associated with the variable "remission of nephrotic syndrome" with a p value < 0.2 at the Wald test in univariable Cox regression analysis. Variables independently associated with remission of nephrotic syndrome will be those with a two-sided p value < 0.05 after stepwise backward elimination in the multivariate model.

In order to assess the robustness of our results with the statistical approach, the investigators will also conduct association analyses between the predictor variables and the variable of interest using the machine learning analysis known as "Random Forrest analysis".