Immunohistochemistry is a widely used technique for characterisation of amyloid fibril type and the presence of normal or aberrant proteins/epitopes and can be used to determine the amyloid fibril protein in a substantial proportion of cases. Antisera to all known amyloidogenic proteins are commercially available, and most are reliable in identifying the fibril type. Once amyloid has been confirmed, we stain the biopsy with a panel of monospecific antibodies against known amyloidogenic proteins (Table 19.1). Morphology of amyloid in certain tissues can give important clues regarding the fibril protein. For example, fibrinogen amyloid is virtually restricted to the glomeruli [12] (Fig. 19.4), and anti-fibrinogen antisera will always be included in the antibody panel when this pattern is identified. AA amyloid deposits are typically extensive in the renal medulla and are always stained in biopsies showing amyloid. In our experience of a small group of Punjabi Indians, ALECT2 amyloid is found throughout renal tissue and can sometimes have a predominance of glomerular deposition. ALECT2 amyloid generally shows bright congophilia with a ‘sparkly glistening’ apple green birefringence when renal tissue is viewed under crossed polarised light (Fig. 19.5). ATTR amyloid also has a distinct appearance and typically displays a ‘honeycomb’ pattern within affected cardiac tissue. ATTR amyloid in the gut is typically identified in submucosal vessels, though it can also present as diffuse amorphous deposition throughout the submucosa (Fig. 19.6). ATTR deposits may be found in other sites such as the bladder, prostate and BMT [13]. Therefore, anti-TTR staining should routinely be carried out on these specimens as part of a panel. In skin biopsies, insulin-induced amyloid is always considered especially in specimens larger than 2 cm², as we have seen several cases over the years. Buccal cavity biopsies are always stained for apolipoprotien A-I (apoA1) amyloid. All biopsies are stained with anti-amyloid P component (AP) so that a comparison can be made with that of a negative Congo red. AP, identical to and directly derived from the normal plasma protein serum amyloid P component (SAP), is present in all human amyloid deposits, though in variable quantities. AP/SAP staining thus corroborates the presence of amyloid deposits of all kinds, though the same protein is also found naturally in basement membranes and some other connective tissue components.

Whilst completely specific antibodies are available to the various proteins from which amyloid is derived, the major conformational transformation from the native soluble proteins to the insoluble β-pleated amyloid fibril forms can result in specific peptide epitopes being lost. Further, protein epitopes can be masked by fixation of the tissue due to the cross-linking of amino acid side groups. In the early days of amyloid immunohistochemistry, it was thought that antigen retrieval was needed to demonstrate the fibril type, and various antigen retrieval methods were used with varying success. However, in our hands, we find that antigen retrieval is of little, if any, use for the detection of amyloid fibril type with the exception of TTR immunohistochemistry where oxidation with 1 % aqueous Na-m-periodate and 0.1 % di-NA borohydride for 10 min each followed by 4 h incubation with high-molarity guanidine treatment [14] is always performed. Without this retrieval, TTR staining is often negative or very weak. This retrieval step is needed due to the β-pleated conformation of amyloid ‘hiding’ some antigenic sites, of which TTR is one. Nowadays retrieval methods are commonplace in most laboratories, and since each antibody differs in the epitope that it recognises, it is important to try the whole range of available antigen recovery methods for each new antiserum.

As well as testing various retrieval methods, the correct way to evaluate the specificity of immunoreactions is to absorb antiserum with its specific antigen [15]. This is routine practice at the NAC.

Although relatively rare, amyloid is a differential diagnosis that ought to be not infrequently considered. Although antisera for AA and AL amyloid are available in most hospitals, if the clinical phenotype is consistent with hereditary systemic amyloidosis (ATTR, AApoAI, AGel, ALys, AFib, etc.), then discussion with or preferably referral to a specialist centre for a full ‘amyloid evaluation’ including detailed immunohistochemistry is advised so that available tissue can be best used to identify the amyloid fibril type.