ABSTRACT
OBJECTIVE: In evaluation of systemic lupus erythematosus (SLE), anti-double-stranded DNA antibodies (anti-dsDNA) play a significant role in diagnosis, monitoring SLE activity, and assessing prognosis. However, evaluations of the performance and limitations for recently developed methods for anti-dsDNA assessment are sparse. METHODS: Specimens used for antinuclear antibody testing (n = 129) were evaluated for anti-dsDNA assay comparability across 4 medical centers in the United States. The methods compared were Werfen Quanta Lite dsDNA, Zeus Scientific dsDNA Enzyme Immunoassay, Bio-Rad multiplex immunoassay (MIA) dsDNA, ImmunoConcepts Crithidia, and Bio-Rad Laboratories Crithidia. RESULTS: For quantitative anti-dsDNA measurements, Spearman's correlation coefficient was highest between Zeus and Werfen (ρ = 0.86; CI, 0.81-0.90; P < .0001). Comparison of MIA to Werfen or Zeus yielded similar results to each other (ρ = 0.58; CI, 0.44-0.68; P < .0001; and ρ = 0.59; CI, 0.46-0.69; P < .0001, respectively), but lower than the correlation between Zeus and Werfen. Positive concordance between assays ranged from 31.4% to 97.1%, and negative concordance between assays ranged from 58.5% to 100%. The detection of anti-dsDNA in those with SLE diagnosis ranged from 50.9% to 77.4% for quantitative assays and 15.1% to 24.5% for Crithidia assays. CONCLUSION: Current quantitative anti-dsDNA assays are not interchangeable for patient follow-up. Crithidia-based assays demonstrate high negative concordance and lack positive concordance among the methods.
ABSTRACT
Metastasis is the manifestation most directly affecting survival for patients with colorectal carcinoma. Identification of high-risk markers for metastases would allow focused selection of patients for adjuvant chemotherapy. Reports of the relationship between the putative metastasis suppressor NM23 and metastasis and/or survival in colorectal cancer patients are conflicting. The purpose of this study was to separately assess expression of NM23-H1 and NM23-H2 in primary colon cancers and determine whether expression was associated with regional nodal disease and/or liver metastases. Four patient cohorts were selected on the basis of histopathological staging at primary surgery (lymph node status/liver metastasis): -/- (n = 46), +/- (n = 47), -/+ (n = 43), and +/+ (n = 46). Primary tumors were evaluated by semiquantitative immunohistochemical analysis of NM23-H1 and NM23-H2. NM23-H2 expression was not related to survival; however, there was a modest survival advantage with low expression of NM23-H1 (P = 0.027). NM23-H1 expression in the +/+ group was increased compared with the other groups (P < 0.001). The -/+ group had the lowest expression of NM23-H2 (P < 0.001). This analysis distinguishes two high-risk groups of colorectal cancer patients. Prior discrepancies regarding the usefulness of NM23 staining may be explained by the need to evaluate both serotypes in addition to standard histopathological analysis to identify specific "at-risk" groups.
