The term "nuclear speckled" is an overarching competent- HEp-2 IFA pattern that includes both fine speckled (AC-4) and large/coarse speckled (AC-5). In some cases, it can be difficult to define precisely the speckled pattern a sample produces, being either the fine or large/coarse speckled pattern. Obviously, one may use the ICAP competent level terminology AC-4/5. In such case, one may combine the antibody targets and clinical relevance of the individual patterns, i.e., AC-4 and AC-5. Using the competent level AC-4/5 term may also be appropriate in mixed patterns caused by more than one autoantibody specificity in the sample.
Ideally, one should provide the most complete definition of the pattern (expert level), because these provide more specific information on antigenic associations and clinical relevance. For example, as indicated in the "antigen association section" of ICAP website, the nuclear coarse speckled pattern (AC-5) is associated mainly with anti-Sm and anti-U1-RNP antibodies, whereas the nuclear fine speckled pattern (AC-4) has a broader target antigen association. In the ICAP Note #2, AC-4 can be further divided into fine speckled with myriad discrete tiny dots (AC-4a), or simple fine speckled (AC-4b). These are defined patterns and have respective antibody targets, immunological associations, and clinical relevance. The nuclear dense fine speckled pattern (AC-2), on the other hand, should be distinguished from the nuclear speckled patterns AC-4 and AC-5 based on the positive staining of the metaphase plate. With exception of the centromeric pattern (AC-3), the clinical relevance of all patterns is only indicated if the antibody target has been identified.

This is not an uncommon phenomenon and there are several possible explanations for this apparent discrepancy.

It is recognized that each immunoassay platform (IFA, WB, ELISA, immunoprecipitation (IP), etc.) is dependent upon a given set of epitopes of relevant autoantigens (1). Broadly, autoantibodies recognize linear and/or discontinuous (native or conformational) epitopes. In the context of discontinuous epitopes, it is important to appreciate that many target autoantigens are part of macromolecular complexes where unique epitopes might be represented as quaternary structures. Immunoassays where the epitope structures are best preserved include IP of cell lysates and HEp-2 IFA. However, there is an important caveat for HEp-2 IFA because the cells have been fixed in methanol and/or acetone (or other undisclosed ‘trade secret’ fixatives) that likely alter some epitopes as well as well as exposing others. As to the question of native epitope preservation, WB presents the largest challenge because the protocols of boiling proteins under deliberate denaturing conditions most likely alters epitopes significantly. It is true that some may renature/refold after electro-transfer of proteins to nitrocellulose membrane, but many most likely do not. When “antibodies by blot” are “not related to the pattern observed in HEp-2 cells by IFA,” one explanation is that antibodies to epitopes recognized in IFA are not identical to or represent only a proportion of those (or even different ones) recognized in a WB. One example for this is anti-fibrillarin; IFA typically shows an AC-9 pattern and yet negative in immunoblot because a high percentage of anti-fibrillarin antibodies do not recognize denatured fibrillarin in immunoblot.

It is common that a given serum sample (especially in SLE) has more than one autoantibody specificity. Hence, another explanation is that when two (or more) completely different antibody specificities co-exist in the same sera, confounding results can be observed, especially on the HEp-2 IFA. In that case, the HEp-2 IFA pattern may reflect the combination of IFA reactivity of all present autoantibodies and the resultant pattern may be different from the one expected if there were only one autoantibody specificity in the serum.

1. Mahler M, Bluthner M, Pollard KM. Advances in B-cell epitope analysis of autoantigens in connective tissue diseases. Clin Immunol. 2003 May;107:65-79.