Quality Control and Laboratory Issues

The study of Eng et al.31 highlights many of the quality control issues facing laboratories performing analysis of BRCA1 and BRCA2. For the laboratory performing the SSCP technique, a false-positive result was reported that occurred due to switching of samples; this also accounted for one of the false-negative results. There appeared to be variability in the ability to detect subtle changes by SSCP. For the laboratory performing CSGE, false-negative results were in part due to erroneous interpretation of CSGE results, and in part due to failure to confirm abnormal gel mobility results using sequence analysis. Clerical errors led to three additional false-negative results. However, CSGE is readily applied to the analysis of known mutations at a low cost. A caveat regarding SSCP and CSGE is that these test methods can systematically fail to identify single-nucleotide substitutions and specific mutations such as the common BRCA1 5385insC mutation.

Although TDGS is appealing as a particularly high-throughput method, five mutations were missed due to incorrect interpretation of the two-dimensional gel, and two mutations were not successfully confirmed using sequence analysis. In addition, three false-positive results were reported, two of which might have been introduced by the long-distance PCR method. The TDGS method is prone to preferential amplification of the wild-type allele. DHPLC, which showed the best performance among the DNA scanning methods, may be further improved by the use of proofreading DNA polymerases to increase the fidelity of the PCR.61

The high-throughput DNA platform developed by Myriad Genetic Laboratories, Inc (Salt Lake City, UT), addresses many of the quality control issues that surfaced in the above study (personal communication, T. Scholl, December 2002). The test requisition and blood specimen are matched by a bar code that is scanned into a database. Pre-PCR procedures are carried out in positive pressure clean laboratories where genomic DNA is isolated from mononuclear cells. Next, 82 test PCR inoculations are performed, in conjunction with two reactions that detect M13 priming sites to control for contamination by amplified products. The PCR amplification products are thermal cycle-sequenced and the sequencing reaction products analyzed on slab gels using high-throughput sequencers (ABI 377). Data analysis also is automated, and a proprietary data review application analyzes the sequence for mutations and polymorphisms. Algorithms are applied to the data to assess signal-to-noise ratios, alignment scores, and signal intensities. A data reviewer who reviews electropherograms from sense and antisense DNA strands confirms sequence variants; antisense DNA information is reverse color complemented to facilitate review. Background subtraction of the composite wildtype sequence also can be done. The involved amplicon then is confirmed by repeat sequence analysis. The reported sensitivity of this platform is 98%, with a specificity of 99%, for mutations detectable by sequence analysis.

0 0

Post a comment