Observations on reliability of commercial assay for alpha-fetoprotein

Thirty patients with nonseminomatous germ cell tumors of the testis were monitored with serum levels of alpha-fetoprotein (AFP). Each patient had at least one AFP determination performed in duplicate: one by the National Institutes of Health (NIH) and one by a commercially available assay. With respect to normal and elevated levels, there was agreement in 60 of the 69 duplicate AFP determinations (87 per cent). Six of the 9 discordant AFP determinations were due to increased sensitivity (a lower normal range) of the NIH assay. There was a close correlation between the quantitative levels of AFP when both the NIH and commercial levels were elevated.