Session: Session 2

Characterization of AA amyloidosis of unknown cause by laser microdissection and mass spectrometry

Martin MATHURIN², Solenne CHARDONNET¹, Mathilde SCARTON³, Soumia HAMADA¹, Léa SAVEY⁴, Gilles GRATEAU⁴, Hélène FRANCOIS⁵, Sophie GEORGIN LAVIALLE ⁴, David BUOB²

¹Sorbonne Université, Inserm, UMS PASS (Production et Analyse des données en Sciences de la vie et en Santé), Plateforme P3S (Post-génomique de la Pitié-Salpêtrière), Paris, France
²Service d’Anatomie et Cytologie Pathologiques, Hôpital Tenon, Assistance Publique – Hôpitaux de Paris, Paris , France
³Service de Néphrologie, Hôpital Tenon, Assistance Publique – Hôpitaux de Paris, Paris, France
⁴Service de Médecine Interne, Hôpital Tenon, Assistance Publique – Hôpitaux de Paris, Paris, France
⁵Service de Transplantation Rénale, Hôpital La Pitié Salpêtrière, Assistance Publique – Hôpitaux de Paris, Paris, France

Introduction

AA amyloidosis consists in the accumulation of insoluble Serum Amyloid Protein (SAA) deposits during chronic inflammation. In some cases, the etiology of AA amyloidosis remains unknown (AAx), making a precise characterization of AA amyloidosis essential. Here, we have used laser microdissection (LMD) coupled to mass spectrometry to analyze biopsy samples from 16 patients in order to identify potential false positives of immunohistochemistry (IHC)-based typing. We also aimed to evaluate the expression of the different variants of SAA, SAA1 and 2.

Methodology

We selected patients with a diagnosis of AAx amyloidosis attending Tenon Hospital. The samples were formalin-fixed and paraffin-embedded. Tissue sections were deposited on Polyethylene Naphthalate membranes and stained with Hematein-Eosin. After LMD, protein extraction and tryptic digestion, we analyzed the protein composition of the deposits by mass spectrometry using the coupling nanoElute – timsTOF Pro in DDA mode.

Results

LMD combined to nLC-MS/MS allowed the identification of the type of amyloidosis in 100% of cases. Proteomic analysis confirmed the initial diagnosis of AA amyloidosis in 15/16 cases. Identification of SAA1 and 2 variants was carried out in 14/15 patients, showing the predominance of SAA1.1 in 85% of the samples.

For the last patient LMD-MS results concluded in a non-AA amyloidosis, i.e. AL lambda amyloidosis. Retrospective analysis of the clinic-pathological characteristics showed ambiguous IHC typing with both anti-SAA and anti-lambda staining. Furthermore, the patient had a monoclonal gammopathy, with detection of urinary lambda light chains

Conclusion

LMD coupled to nLC-MS/MS is efficient to classify different types of amyloidosis and to study the prevalence of SAA variants. This approach confirmed the diagnostic for 15/16 patients and corrected an erroneous IHC-based classification for one patient.