Abstract

Lipoprotein(a) [Lp(a)] has been measured in numerous clinical and epidemiological studies by a variety of immunochemical methods. However, little, if any, consideration has been given to the confounding effect of the size heterogeneity of apolipoprotein(a) [apo(a)] on the measurement of Lp(a). We developed three direct-binding enzyme-linked immunosorbent assays (ELISAs) with detecting antibodies of different specificities to evaluate the effect of apo(a) size on Lp(a) measurement. The three assays used the same monoclonal antibody to capture the apo(a)-containing particles and were calibrated (in nanomoles per liter) with a serum containing apo(a) with 21 kringle 4 domains. Using all three ELISAs, we measured Lp(a) in a group of 723 subjects selected to have a single apo(a) band, as determined by a high-resolution phenotyping system. Essentially identical results were obtained by the two methods that measured Lp(a) by use of either a polyclonal antibody against apo B or a monoclonal antibody against apo(a) that does not recognize the kringle 4 type 2 repeats. In contrast, the ELISA using a monoclonal antibody specific for apo(a) kringle 4 type 2 repeats overestimated Lp(a) concentration in samples containing apo(a) with more than 21 kringle 4 domains and underestimated Lp(a) samples containing apo(a) with fewer than 21 kringle 4 domains. Thus, these differences in Lp(a) values varied as a function of apo(a) size. We conclude that antibody specificity and apo(a) size heterogeneity can significantly affect Lp(a) measurements.