Autoantibodies to 21-hydroxylase as a diagnostic marker of primary autoimmune adrenal insufficiency, including at its potential and latent stages

Rationale: In Russia, assessment of anti-P450c21 antibodies (AB) in the diagnosis of autoimmune adrenal insufficiency (AAI) has not been commonly used, and the disease screening has not been implemented.

Aims: 1)  To determine the sensitivity and specificity of anti-P450c21 AB determination in the AAI diagnosis; 2)  To estimate the prevalence of anti-P450c21 AB carriage in patients without AAI.

Materials and methods: Anti-P450c21 AB were assessed in 40  patients (group  1) with manifest AAI; 171  patients without established diagnosis of AAI, including 113  subjects with autoimmune thyroid disorders or type 1 diabetes mellitus (AID, group 2); 25 carriers of AB markers of thyroid AID and/or type  1 diabetes mellitus without any target organ dysfunctions (group 3); 33 patients with non-autoimmune endocrine disorders (group  4), and 25 healthy individuals (group 5).

Results: Determination of anti-P450c21 AB for the diagnosis of AAI had 95%  sensitivity, with specificity of 100%, predictive value of a positive result of 100%, and predictive value of a negative result 92.6%. Anti-P450c21 AB were inversely correlated with the duration of glucocorticoid replacement therapy (r=-0.222, p<0.05). High levels of anti-P450c21 AB were found in 4 (3.5%) patients of group  2; based on the results of additional hormonal testing, 50%  cases were diagnosed with the latent stage of the disease and 50% cases with the potential stage.

Discussion: The sensitivity of the anti-P450c21 AB determination for AAI diagnosis in our study was higher, than in the works by other authors. We have confirmed a  time-related reduction of anti-P450c21 AB levels, whereby the strength of the correlation was higher in the subgroups with autoimmune polyendocrine syndrome type  II and, to a greater extent, autoimmune polyendocrine syndrome type I. This might be related to their different pathogenesis, with an abnormality of central immune tolerance in autoimmune polyendocrine syndrome type I and that of peripheral immune tolerance in autoimmune polyendocrine syndrome type II. According to our data, in 50% of cases, the development of AAI was preceded by the manifestation of other AIDs (in 15% of cases being multiple). Among all patients with no AAI diagnosis at the study entry, increased anti-P450c21 AB levels were found exactly in those with pre-existing AID. Thus, we have confirmed the feasibility of AAI screening primarily in a cohort of patients with other AID (especially multiple) belonging to the risk group.

Conclusion: The determination of blood anti-P450c21 AB is a highly sensitive and highly specific method to diagnose AAI. The frequency of anti-P450c21 AB detection might depend on the duration of glucocorticoid treatment. Screening for early AAI stages is relevant primarily in the risk groups with multiple autoimmune disorders.