Introduction:
The molecular mechanisms underlying amyloid deposition in the kidney are poorly understood. 70% of patients have renal dysfunction when diagnosed, rapidly progress, and are on dialysis within 1-2 years. To identify potentially prognostic transcriptional signatures of amyloid light-chain (AL) amyloidosis, we analyzed renal biopsies of the of the Renal AL Amyloid Involvement and NEOD00 (RAIN) trial (NCT03168906) and utilized a novel scoring method.
Methods:
Each biopsy was scored by a composite of scarring injury and amyloid score (AS). Biopsy tissue was microdissected into glomerular (G) and tubulointerstitial (TI) tissue and sequenced separately. Unsupervised clustering was performed on each compartment. Differentially expressed genes were calculated between identified subgroups and to healthy living donors (LD) and focal segmental glomerulosclerosis (FSGS) profiles obtained from the NEPTUNE cohort.
Results:
Clustering of the bulk RNAseq data identified two subgroups (G1 and G2). In the TI compartment, AS was significantly higher in G2 (P<0.05). Pathway analysis of the glomerular compartment highlighted activation of fibrosis as well as elevated signaling of LPS/IL-1. Ingenuity pathway analysis (IPA) for the TI compartment detected enrichment of TNF activation, GADD45 signaling and the Wnt/Cab pathway. Comparing amyloidosis to healthy living donors highlighted PI3K/Akt pathway signaling. In both the TI and glomerular compartment, the LD cluster predominantly with the G1 group. Additionally, the amyloid samples display most similarity with FSGS samples.
Conclusions:
While limited by a small samples size, the molecular data of AL patients could robustly be separated in two groups suggesting presence of separate molecular mechanisms. Increasing sample size and analysis on single cell resolution can help further identify the role of individual cell types in AL and their response to amyloid deposits. The similarity between AL and FSGS samples may help identify effective therapies for AL.
I have no potential conflict of interest to disclose.
I did not use generative AI and AI-assisted technologies in the writing process.