Identification of premature ovarian failure patients with underlying autoimmunity.

Abstract

Although known causes of premature ovarian failure (POF) include X chromosome deletions, radiation and chemotherapy, and genetic defects of the gonadotropin hormones or receptors, at least one third to one half of cases remain idiopathic. A significant proportion of patients with apparently idiopathic POF have some evidence for an autoimmune etiology. However, the only gold standard for detecting autoimmune causes of immune ovarian destruction has been invasive ovarian biopsy. Serum antibodies to ovarian and other self-tissue have been described in up to one third of women with POF, but the tests are not well standardized, not well correlated with ovarian histology, and highly variable. Recently, specific defects of expression of cell surface markers on peripheral blood lymphocytes have been shown to identify, in population-based studies, individuals destined to develop autoimmune pancreatic destruction and type I diabetes mellitus, even before any other evidence of autoimmunity. We, therefore, sought to test the ability of cell surface marker expression in women with POF to identify autoimmune defects. Seventeen women with POF, 11 of whom had positive antibody titers to ovary, thyroid, or antinuclear antibody, were studied on at least two occasions and compared in blinded fashion with normal controls and patients with autoimmune type I diabetes mellitus. The most useful marker for identifying autoimmunity was the surface density of conformationally correct HLA class I molecules on macrophages, a structure essential for T cell education. Using this marker, 7 of the 9 patients with autoantibodies and 3 of the 8 patients without autoantibodies were identified as having evidence of a defect in self-antigen presentation similar to that of type I diabetics (chi-square, p = 0.03). Subsequent testing identified antismooth muscle antibodies in 1 of the women with a defect of HLA class I molecules but no previously identified autoimmunity. In addition, there were increased numbers of CD8 T cells in both autoimmune POF and insulin-dependent diabetes mellitus (IDDM) patients. Exclusive to POF patients was a statistically significant increase in CD8 density on T cells. This was most prominent in POF patients with an underlying autoimmune etiology. These data further support a role for autoimmunity in POF patients and suggest that the further development of cell surface markers in combination with other diagnostic tests could result in diagnosis before the development of complete ovarian failure. The possibility for disease-specific therapy to prevent further autoimmune ovarian damage in selected POF patients is also envisioned.