Light chain amyloidosis (AL) is the most common subtype of systemic amyloidosis in the United States, with 4,500 new cased diagnosed annually. Symptoms of AL often exist for years prior to diagnosis and initiation of treatment. In AL dysfunctional proteostasis leads to misfolded immunoglobulin light chains (IgLC) that form fibrils that deposit into tissues throughout the body leading to organ failure. With amyloidogenic clones toxic abnormal light chains accumulate and enhance endoplasmic reticulum (ER) stress without properly activating the protein degradation pathways. There is little understanding of the genetic and epigenetic predispositions that lead to amyloidogenicity. The need is to develop molecular biomarkers for early detection of AL to prevent significant organ damage that leads to the current morbidity and mortality.

We analyzed the mutational burden of a combined cohort of 113 bone marrow samples from AL patients, making this one of the largest cohorts. This included 14 internal treatment-naïve bone marrow aspirate samples with the clonal plasma cell burden ranging from 5% to 50%. The remaining 99 AL bone marrow samples with whole exome sequencing data are a pooled database comprised of 3 previously reported studies: 27 patients in Cuenca et al., Leukemia 2020, 24 in Boyle et al., Blood 2018, and 48 in Huang et al., Amyloid 2020. For the 14 in-house AL samples, after plasma cell sorting, we utilized whole-exome next-generation RNA sequencing. The RNASeq data were aligned to the GRCh38 reference genome using STAR with Salmon alignment after FASTQC. The expression matrix and differential expression were analyzedby edgeR. The variant calling results were obtained using Strelka, annotated with ANNOVAR and filtered by damaging polymorphism phenotyping (Polyphen>0.85) and sorting intolerant from tolerant scores (SIFT<0.05) from ensemble variant effect predictors. The data from all 113 AL patients were further analyzed to identify unique clonal biomarkers within AL. Clinical annotations were obtained regarding patient demographics, treatment history and response.

Our cohort of 14 AL patients had a median age of 74 (53-82), predominantly male (79%) with all having tissue biopsy proven AL. Nine of the 14 patients were enrolled in the CAEL-101 study with standard of care Daratumumab+Cyclophosphamide+Borezomib+Dexamethasone plus 1:1 randomized 1:1 to also receive CAEL-101, an amyloid-eating antibody, vs placebo. The remaining patients received a combination of standard of care therapies. All but one patient produced lambda as the amyloidogenic light chain.

The next generation RNA sequencing identified a total of 37 common, non-immunoglobulin and nonsynonymous mutations with a variant allelic fraction (VAF) varies <0.02 at 100X depth. The VAF increased with plasma cell clonality and disease burden. Interestingly, 13 out of 37 mutated genes were altered more than once in our cohort of 14 AL patients. Notwithstanding that, 13 of the 14 cases presented with at least one mutation in one of the 37 genes. Prior studies indicate the high genomic variability of AL (Cuenca et al., Leukemia 2020), yet we found several potential hotspots of gene mutations across the combined cohort of 113 AL patients, like HERC1 (HECT and RCC1-like domain-containing protein 1) which encodes the E3 ubiquitin ligase engaging in ER proteostasis during stress and intracellular trafficking. More than a third of the patients had mutant HERC1. AL patients harboring a mutation in the c-terminal HECT-domain (Homologous to E6AP) resulting in loss of ubiquitin transfer to substrate proteins resulting in its escape.

Another key gene in proteostasis in AL, IGLL5 (Immunoglobulin Lambda-Like Polypeptide 5) with 11.6% occurrence. This encodes the protein structure of immunoglobulin lambda light chains and is involved in early plasma cell development and receptor signaling. This finding is consistent with previous report that IGLL5 could be a potential signature mutation in AL.Here we report that loss of function mutation in E3 ubiquitin ligase HERC1 HECT domain may allow the exit of toxic IgLC from the plasma cells and hence created an aberrant proteasomal proteostasis. This mutation causing decoupling of proteasome from ER-stress providing new insights for early-stage biomarkers and therapeutic targets in AL.

This content is only available as a PDF.
Sign in via your Institution