Performance of clinical laboratories for DNA analyses to detect thrombophilia mutations.

To the Editor: Genetic tests for risk factors for venous thromboembolism (1 ) are increasingly available. We report the results of a survey that included a search for 3 thrombophilia mutations. The survey was organized by the Subcommittee on Hemostasis of the CISMEL (Italian Committee for Standardization of Laboratory Tests). The scheme enrolls 250 participants, of which 52 are also registered for DNA testing. After receipt of informed consent, DNA samples were prepared at the organizing center from 4 patients whose genotype had been identified previously, on the occasion of thrombophilia screening. The genotypes were as follows: 3 homozygous for factor V Leiden (FV-Leiden) G1691A, prothrombin G20210A, or methylenetetrahydrofolate reductase (MTHFR) C677T mutations and 1 triple heterozygous for the FV-Leiden, prothrombin, and MTHFR mutations. DNA samples were extracted (2 ), coded, aliquoted, and stored at 4 °C until shipment. The participants were sent 10 L of DNA (400 ng/ L) and asked to detect the mutations with their methods. Results had to be returned to the organizing center for statistical analysis. Forty-one of 52 participants returned results; 28 used commercial and 13 used in-house methods. Details on result interpretation are reported in Table 1. Misclassifications consisted of false-negative (FV-Leiden, prothrombin, and MTHFR), heterozygous vs homozygous (FV-Leiden and prothrombin), homozygous vs heterozygous (FV-Leiden, prothrombin, and MTHFR), or false-positive (prothrombin and MTHFR). A few participants gave no interpretation (Table 1) because of insufficient sample. This is a sequel to previous reports from our and other schemes for DNA analyses in thrombophilia (3, 4) showing that standardization and regular quality-control programs aimed at identifying causes of failure are warranted. The false-positive detections suggest that genotypes for individual patients should be confirmed in a second laboratory investigation. An additional consideration stemming from this study concerns MTHFR, for which there is consensus among experts about its limited clinical value. This mutation (heterozygous or homozygous) is a relatively frequent finding in the general population (5) and is not associated with thrombotic risk (5–7); however, a search for MTHFR mutations is included in many thrombophilia testing panels and is frequently requested by clinicians. One of the issues that clinicians should consider when making decisions on what to request in laboratory investigations is the benefit vs the risk of related anxiety that knowledge of being a carrier of a given mutation may bring to the patient. In the case of MTHFR, the risk clearly outweighs the benefit. The false-positive detections add further arguments against use of MTHFR in the evaluation of thrombotic risk in individual patients.