Feasibility of DNA-Based Diagnosis of Fibrinogen A Alpha-Chain Amyloidosis From Paraffin-Embedded Kidney Tissue

Introduction

Hereditary amyloidoses are rare, accounting for < 5% of cases of renal amyloidosis.¹ Fibrinogen A alpha-chain (AFib) amyloidosis is the most common hereditary form affecting the kidney,² which is more frequent in Europe than in the United States.³

To date, there are 15 reported amyloidogenic pathogenic variants of fibrinogen A alpha-chain (FGA) gene, including single-pair substitutions and frameshift variants.⁴ However, the p.Glu545Val (this nomenclature includes the 19-amino-acid signal peptide, and it is also reported as p.E545V, and c.1634A>T for the coding region) is by far the most common variant, accounting for 86% to 90% of cases; it has been largely reported in Northern Europeans but identified in kindreds throughout the world.^2,3 In contrast, frameshift variants were mostly found in isolated families. Throughout this manuscript, we will refer to our variant of interest as E545V.

Clinical guidelines require DNA sequencing of all amyloidogenic pathogenic variants, which is typically performed on a peripheral blood sample. Here, we report, to the best of our knowledge, for the first time the successful detection of E545V by analysis of DNA extracted from small residual kidney biopsy formalin-fixed and paraffin-embedded (FFPE) tissue, allowing for prompt confirmation of diagnosis of AFib E545V renal amyloidosis.

Results

We identified 115 renal AFib amyloidosis cases by a retrospective review of all specimens typed at the Mayo Clinic Tissue Proteomics Laboratory from January 2008 through December of 2024. The diagnosis of AFib amyloidosis was based on the presence of massive congophilic glomerular deposits and the proteomic findings of abundant spectra corresponding to the following fibrinogen A alpha-chain and amyloid signature proteins: serum amyloid P, apolipoprotein E, and apolipoprotein AIV. In 81 of these cases (70%), liquid chromatography-tandem mass spectrometry detected a pathogenic variant (E545V in 71 [88%] and R573L in 10 [12%]), whereas a pathogenic variant was not detected in the remaining 34 cases. The methods for our proteomics-based amyloid typing are described in the Supplementary Material. Of these 115 renal cases, 13 underwent tissue processing and staining at the Renal Biopsy Laboratory of Mayo Clinic and residual paraffin blocks were available for DNA analysis. liquid chromatography-tandem mass spectrometry detected E545V in all 13 cases (Supplementary Figure S1).

The demographics, clinical, pathologic, and outcome characteristics of these 13 patients (which represent the second largest reported cohort from United States) are presented in detail in Table 1 and Supplementary Tables S1 and S2. The cohort consisted of 10 men and 3 women, all White, with a median age of 61 (range: 47–74) years (Table 1). All patients (100%) presented with proteinuria (median: 4.5 g, range: 1.6–14 g), 12 (92%) had hypertension, 11 (85%) had elevated serum creatinine (median 4 mg/dl, range: 1-15), 6 (50%) had hypoalbuminemia (median serum albumin: 3.4 g/dl, range: 2–4.3), 4 (36%) had peripheral edema, 2 (17%) had hematuria, and only 1 (8%) had full nephrotic syndrome (Supplementary Table S1). At the time of diagnosis, 1 subject had CKD stage 2, 1 had stage 3a, 3 subjects had stage 3b, 4 subjects had stage 4, and 4 subjects had stage 5. Family history of amyloidosis or CKD was present in 5 patients (42%). None of the 9 patients with available data had documented clinical or pathologic evidence of extrarenal involvement by amyloidosis. Fat pad biopsy and bone marrow biopsy, performed in 4 and 3 patients, respectively, were negative for amyloid. Two (15%) had a history of monoclonal gammopathy. Comorbidities and follow-up data are presented in detail in Supplementary Table S1. The 1-, 2-, and 5-year kidney survival rates (from diagnostic kidney biopsy) were 44% (4/9), 33% (3/9), and 11% (1/9), respectively. The 1-, 2-, and 5-year patient survival rates were 92% (11/12), 82% (9/11), and 64% (7/11), respectively.

Table 1.

Demographics, proteomics, and DNA sequencing results

Case #	Age	Sex	Race	Amyloidosis type by LC-MS/MS	Protein variant by LC-MS/MS	FGA pathogenic variant by DNA sequencing on FFPE	FGA pathogenic variant by DNA sequencing on peripheral blood
1	71	F	W	AFib	E545V	c.1634A>T	c.1634A>T
2	73	M	W	AFib	E545V	Unsuccessful	ND
3	72	M	W	AFib	E545V	c.1634A>T	ND
4	59	M	W	AFib	E545V	c.1634A>T	ND
5	54	F	W	AFib	E545V	c.1634A>T	ND
6	60	M	W	AFib	E545V	Unsuccessful	c.1634A>T
7	58	M	W	AFib	E545V	c.1634A>T	ND
8	47	M	W	AFib	E545V	Unsuccessful	ND
9	54	M	W	AFib	E545V	Unsuccessful	ND
10	55	M	W	AFib	E545V	Unsuccessful	ND
11	74	F	W	AFib	E545V	c.1634A>T	ND
12	55	M	W	AFib	E545V	Unsuccessful	ND
13	62	M	W	AFib	E545V	c.1634A>T	c.1634A>T

AFib, fibrinogen A alpha-chain; F, female; FFPE, formalin-fixed and paraffin-embedded tissue; LC-MS/MS, liquid chromatography-tandem mass spectrometry; M, male; ND, not done; W, White.

The kidney biopsy findings (Supplementary Figure S2) are presented in detail in Supplementary Table S2 with description.

We performed next generation sequencing on DNA extracted from the residual FFPE tissue blocks of these 13 cases to analyze FGA. Methods for DNA extraction and next generation sequencing are presented in the Supplementary Material. Of 13 cases tested, 7 (54%) yielded a positive result with the identification of E545V in all (Figure 1). The remaining 6 cases did not yield a result because of 1 or more of the following scenarios: low input DNA (n = 4), degraded DNA (n = 3), or low coverage of the DNA sequence of interest (n = 4). Interestingly, only 3 of 13 subjects (23%) underwent next generation sequencing from peripheral blood samples, which confirmed the presence of the FGA E545V pathogenic variant.

Figure 1.

Patient 1’s results of DNA sequencing on FFPE renal biopsy. From the top: FGA is represented as a line with 5 blue exons and 4 yellow introns. Toward the end of exon 5, red square brackets highlight an area of interest. Zooming in in this area (red arrow) at position c.1634 there is an A (green, adenine). However, the nucleotides detected under that position show a T (red, thymine) in approximately half of the reads (forward direction - light green, and reverse - light blue). The codon GAG (translating to glutamine, Glu) becomes GTG (valine, Val). Total reads for the variant = 55; 27 reads (49%, both directions) show A, whereas 28 reads (51%, both directions) show T. FFPE, formalin-fixed and paraffin-embedded tissue.

Discussion

The genetic and phenotypic diversity inherent in hereditary amyloidosis underscores the necessity of genetic testing for an accurate diagnosis and genetic counseling of patients and their relatives, which is typically performed on DNA extracted from a peripheral blood sample. Although genetic testing for ATTR (by far the most common form of hereditary amyloidosis but with uncommon kidney involvement) is widely available, genetic testing for other forms of hereditary amyloidosis (AFib, ALys, AGel, AApoA1, AApoAII, AApoC-II, and AApoC-III) is only available in a few laboratories. It has been our clinical experience in the Mayo Clinic Tissue Proteomics Laboratory, a national referral center for amyloid typing, that genetic testing is not performed on most cases of non-ATTR hereditary amyloidosis that we diagnose by liquid chromatography-tandem mass spectrometry, even though we recommend genetic testing in our liquid chromatography-tandem mass spectrometry reports, as evident in our current cohort of AFib amyloidosis in which genetic testing on peripheral blood was only performed on 3 patients (23%). Archival FFPE tissue could serve as an alternative source for DNA to determine the underlying pathogenic variant responsible for hereditary amyloidosis, as has been previously shown in ATTRv^5,6 and we show in the current report in AFib amyloidosis.

Although FFPE tissue can be used for molecular genetic testing, the DNA is typically fragmented and partially degraded. As a result, DNA quality, quantity, and fragment size extracted from FFPE tissue tends to be lower compared with blood, which can lead to downstream testing challenges. The size of DNA retrieved from FFPE samples depends on the length of interval between procurement of tissue and formalin fixation, nuclease content (which varies among different types of tissue), and length of storage of paraffin block. Furthermore, formalin induces cross-linking of nucleic acids which could lead to the formation of random nucleotide substitutions (artifacts). Despite these limitations and the small size of available residual kidney biopsy tissue after undergoing full light microscopy work-up (sixteen 3 μm levels or more), Congo red staining (10 μm section), and mass spectrometry (10 μm section), this approach was successful in detecting the most common AFib pathogenic variant in over half of cases. We speculate that if molecular testing was performed on larger FFPE tissue samples, tissue or nuclei might have been more abundant, yielding higher DNA concentration and greatly increasing the chances of successful sequencing and variant detection in the cases that failed. Although in this study we performed targeted DNA analysis of the FGA gene, we speculate that pathogenic variants in other genes that are responsible for other forms of hereditary amyloidosis could be detected by this method.

In summary, this proof-of-concept, feasibility study demonstrated that DNA analysis can be successfully performed on readily available residual FFPE kidney biopsy tissue, allowing for prompt confirmation of diagnosis of AFib E545V renal amyloidosis in over half of cases. Further studies are needed to test the performance characteristics of this approach in other forms of hereditary renal amyloidosis. Further discussion and details are presented in the Supplementary Material.

Disclosure

All the authors declared no competing interests.

Funding

This work was supported in part by funding from the Department of Laboratory Medicine and Pathology at the Mayo Clinic, which did not have a role in study design, data collection, analysis, reporting, or the decision to submit for publication.

Author Contributions

The research idea and study design were done by SHN and AB. Data acquisition, curation, and interpretation were done by all the authors. Supervision and mentorship were done by SHN. Each author contributed important intellectual content during manuscript drafting or revision and agrees to be personally accountable for the individual’s own contributions and to ensure that questions pertaining to the accuracy or integrity of any portion of the work, even one in which the author was not directly involved, are appropriately investigated and resolved, including with documentation in the literature if appropriate.

Supplementary Material

Supplementary File (PDF)

Supplementary Methods. Supplementary Discussion. Figure S1. Detection of mutant fibrinogen A alpha-chain in patient’s amyloid deposit. Figure S2. Renal biopsy findings. Table S1. Clinical characteristics and outcomes. Table S2. Pathological characteristics.