Two prior chapters have reported their author’s experience of amyloid typing performed on paraffin sections, using either commercial or custom amyloid type-specific antibodies. The former are raised against epitopes of the “native” protein while custom antibodies are raised against the extracted amyloid fibril protein. In contrast to the “native” protein, amyloid fibril proteins are expected to be altered, due to the conformational changes and fragmentation/truncation that occurs during fibrillogenesis. Thus, epitopes that are reactive with antibody raised against the “native” protein may be either altered (as a consequence of conformational changes) or no longer present (if the amyloid fibril protein underwent truncation). Hence, the reactivity of amyloid fibril proteins with commercial antibodies may be correspondingly altered or reduced. This is particularly true in AL, where the fibril protein may be derived from a fragment of the light chain that contains the variable region and a varying portion of the constant region. Since commercial antibodies are raised against the constant region, antibody reactivity largely depends on the extent to which this region is present in the amyloid fibrils. Although, in most cases of AL, there is a mixture of fragments of the light chain that are of variable length, in some cases, the amyloid fibril protein may be derived from the variable region alone and, in such cases, antibody reactivity may be absent altogether. As discussed earlier by Linke, even in the case of custom antibodies raised against amyloid proteins, finding a single antibody against AL-lambda that performs consistently well in all cases has been challenging due to the problem of reduced reactivity in some instances. Thus, in order to circumvent this, Linke routinely uses several of his custom antibodies against AL-lambda in the immunohistochemistry panel. Recently, antibodies to free-light chains have been used in immunohistochemistry with results superior to those obtained with conventional antibodies, which detect both free and bound free light chains [22].

According to a recent worldwide survey of amyloid reference laboratories, commercial antibodies were used by 55 %, and custom antibodies by 45 %, of respondents [1]; in contrast, most renal pathologists use commercially available antibodies [2]. In general, custom antibodies were considered to be more sensitive. However, regulatory issues, e.g., the validation of such antibodies as analyte-specific reagents (ASR), have not been uniformly addressed. Interestingly, however, the extent of the various antibody reactivities with amyloid fibril proteins is variable and appears also to be technique dependent. Thus, as shown in two subsequent chapters, antibody typing using commercial antibodies by immuno-electron microscopy techniques gives better results than typing in paraffin sections; similarly, amyloid typing by the immunoblotting (Western blot) technique has been shown to be more reliable than by immunohistochemistry in paraffin sections.

Rare cases of amyloidoses derived from the immunoglobulin heavy chain, AH, have been reported. Cases derived either from γ or μ- heavy chain, typically, have had deletions in the C_H1 and C_H2 regions [7]. Specific antibodies to the intact heavy chains are routinely used, and hence the fibrils, being derived from a truncated protein, may not be reactive; thus, it is not clear how many cases may be missed using current methods. More recently, Sethi et al. [23] described clinicopathologic findings in four cases of renal immunoglobulin heavy chain amyloidosis, which, in two cases, also showed reactivity for heavy chains together with light chains. Subsequent studies reported additional cases of amyloidosis where both light and heavy chains were detected and the authors coined the term “immunoglobulin amyloidosis” to encompass the apparent spectrum of amyloidoses derived from various immunoglobulin components, including also light+heavy chain [24]. Whether these findings indicate that the fibrils are truly composed of both light+heavy chains is an open question at the present time [25]. Thus far, light/heavy chain amyloidosis has not been officially included in the classification and nomenclature of the amyloidoses [26].

Amyloid deposits do not contain pure fibril proteins but represent a rather heterogeneous complex that also includes, besides the amyloid fibril protein itself, other components such as AP component, Apolipoprotein E, glycosaminoglycans, variable extracellular components, and even lipids [7]. All of these may affect antibody binding. AP component and Apolipoprotein E have been recognized as “amyloid signature” elements.

It has been shown that antigenic sites may be altered during fixation to a variable degree, and, therefore, it is not surprising that many antibodies perform better in frozen than in paraffin sections. This is especially evident in the case of serum protein detection in kidney biopsies, as well as the detection of light chain restriction in general and in AL in particular [27–29]. The use of antigen retrieval in amyloid immunohistochemistry is controversial, and experience, expertise, and validation are critical ([30], please see also Chaps. 18 and 19 by Linke and Gilbertson et al. respectively).

Another important factor to consider is the fact that most amyloid fibril proteins also have their native counterparts present in the serum. Thus, in paraffin sections, the plasma proteins may have been fixed in the tissues and, if they are to be removed, they must be removed by a digestion process. Their incomplete elimination creates a background stain that may compete with the signal from the amyloid protein and result in a low signal-to-noise ratio. In contrast, in frozen sections, plasma proteins can be removed simply by washing [8]. However, to some extent, even in vivo, various serum proteins can be adsorbed to amyloid deposits. As can be seen from the above, in renal pathology, the issues of limited antibody reactivity and background staining appear to be less of an impediment to successful amyloid typing, largely due to the application of immunofluorescence performed on frozen sections.