Unreliability of Immunohistochemistry for Typing Amyloid Deposits

To the Editor.—The article by Satoskar et al1 addresses a seminal issue; namely, the therapeutic necessity of identifying the protein that forms renal amyloid deposits. These authors state that “immunofluorescence staining for immunoglobulin light chains on renal biopsy, as the first step to differentiate between AL primary or light chain–associated and AA secondary or serum amyloid A protein–associated amyloidosis, may sometimes be inconclusive or even misleading.” Notably, their commercially obtained anti–light chain antibodies lacked specificity, that is, they cross-reacted with AA fibrils and did not necessarily reveal a predominant immunostaining of one light chain type (κ vs λ). In an accompanying editorial, Picken and Herrera2 discuss the numerous limitations and pitfalls that can occur when using immunohistochemistry to establish the type of amyloid present in biopsy specimens and emphasize that this method requires special expertise and reagents; furthermore, “misinterpretation may have profound consequences.”These reports indicate the danger of relying on immunohistochemical methods, as well as ancillary laboratory or clinical data, to ascertain if an individual has AL, AA, or some other kind of amyloid disease. To obtain an accurate diagnosis, the amyloid contained in biopsy-derived specimens are best extracted and analyzed chemically.3 In this regard, we have used tandem mass spectrometry to gain this information from formalin-fixed, paraffin-embedded tissue biopsies, as well as from subcutaneous fat aspirates,4 and perform these studies on request. Depending on the size of the biopsy specimen and amount of amyloid present, as few as four to six 4-μm-thick sections may suffice (even if there is no tissue remaining in the paraffin blocks, it is possible to use sections that have been subjected to Congo red, hematoxylin-eosin, or other histochemical stains). Further, material scraped from 1 or 2 amyloid-containing fat aspirate smears have also yielded sufficient fibrils for tandem mass spectrometry analysis.Because the treatment as well as prognoses of patients with amyloidosis is dependent on the amyloid type, it is crucial that the nature of the fibrillar protein be established unequivocally to avoid inappropriate and costly therapy that can have dire and possible legal consequences.