The case for a descriptive generic nomenclature: clarification of immunostaining criteria for PCA‐1, ANNA‐1, and ANNA‐2 autoantibodies

VA Lennon - Neurology, 1994 - AAN Enterprises
VA Lennon
Neurology, 1994AAN Enterprises
To the Editor: My colleagues and 11-4 have previously addressed several of the points
raised by Drs. Dalmau and P~ sner.~ Some of these points clearly merit reiteration and
further clarification. References l to 4 and 6 to 9 give more detail. The principal point missed
by our Sloan-Kettering colleagues is that the descriptive generic nomenclature we favor is
based on rigorous immunofluorescence staining criteria in a standardized assay performed
on a panel of tissue substrates. These criteria exclude antibodies that yield simply a …
To the Editor: My colleagues and 11-4 have previously addressed several of the points raised by Drs. Dalmau and P~ sner.~ Some of these points clearly merit reiteration and further clarification. References l to 4 and 6 to 9 give more detail. The principal point missed by our Sloan-Kettering colleagues is that the descriptive generic nomenclature we favor is based on rigorous immunofluorescence staining criteria in a standardized assay performed on a panel of tissue substrates. These criteria exclude antibodies that yield simply a superficial resemblance to PCA-1, ANNA-1, or ANNA-2, the three paraneoplastic autoantibody specificities under discussion6 and alternatively named “Yo,”“Hu,” and “Ri.” 5 The table and figure 1 summarize the immunostaining criteria that our laboratory uses to define antibodies of PCA-1, ANNA-1, and ANNA-2 specificities. Figure 2 demonstrates Western blot characteristics of some of the proteins identified by autoantibodies defined by these strict immunofluorescence characteristics. At the outset, it should be noted that in the past 10 years, my own laboratory at Mayo Clinic and Dr. Posner’s laboratory at the Sloan-Kettering Institute have both studied thousands of patients’ sera by methodologies preferred by our individual laboratories. The practicing clinician can be reassured that publications from these two independent groups are in substantial agreement about conclusions to be drawn concerning the clinical/oncologic significance of paraneoplastic autoantibodies defined by our individual laboratories’ riter ria.^,"'^ A second point is that rhetoric cannot resolve questions about the relative sensitivity and specificity of an optimized immunofluorescence technique compared with the combination of recombinant protein Western blots and immunoperoxidase staining of neural tissues favored by the Sloan-Kettering investigators. Each methodology has merits and limitations. For example, a mixture of high-titered, non-organ-specific autoantibodies in occasional patients’ sera (eg, antinuclear and antimitochondrial) will preclude definitive immunofluorescence detection of a neuron-restricted IgG specificity. In those infrequent cases, Western blot analysis is indicated. On the other hand, we have detected PCA-1 and ANNA-2 antibodies by indirect immunofluorescence, and the predicted tumors were found, in several patients whose sera were negative by commercial Western blot tests with recombinant Yo and Ri proteins. Determination of whether one serologic approach is superior to another for purposes of clinical testing will require a side-byside comparison of results obtained for coded sera by independent laboratories in a trial performed under the auspices of an impartial panel, such as could be convened by the American Academy of Neurology or the Clinical Immunology Society. 2 Drs. Dalmau and Posner are correct in noting that the methodologies employed at Mayo Clinic have been continuously refined in 8 years of serologic testing for paraneoplastic autoantibodies on a prospective clinical basis. 3, 4, 6-9.’4 We have attempted to maintain the eneric terminologies originally only slightly in this period of evolution. For the two antineuronal nuclear spe~ ificities,’~.’~ we assigned the numerals 1 and 2 when we recognized that ANNA-1 is distinguishable from ANNA-2 immunohistochemically by the dual reactivity of ANNA-1 with both peripheral (myenteric) and central neuassigned to each~ pecificity’~. f by modifying the nomenclature
American Academy of Neurology