Dudbridge et al. Page 262

Rare genetic variants could explain common disease if they are coupled with more common single-nucleotide peptide alleles, creating “synthetic” associations. To address this possibility, Dudbridge and colleagues analyzed 10 loci that have been consistently associated with breast cancer and estimated their allele frequencies and relative risk. Using Bayesian estimation, the authors infer that these causal variants are probably common, supporting the common disease–common variant hypothesis for these loci. Current efforts to resequence these causal variant regions will help validate these quantitative estimates.

Samimi et al. Page 273

The expression of transducin-like enhancer of split 3 (TLE3) was shown to be associated with favorable outcomes in taxane-treated breast cancer patients. In this study, Samimi and colleagues looked at the association of TLE3 and recurrence in ovarian carcinoma patients treated with a taxane. TLE3 immunohistochemistry was performed on ovarian cancer specimens, and disease recurrence was monitored for 5 years. The authors report that TLE3 expression was associated with favorable outcome, but only in patients who had received a taxane as part of their treatment. These results encourage further evaluation of TLE3 as a potential therapeutic response marker for taxane-based chemotherapy in ovarian cancer.

Bodmer et al. Page 280

Use of the diabetes drug metformin has been linked to decreased risk of some cancers, but the association of metformin and colorectal cancer remains unclear. To explore this question, Bodmer and colleagues performed a case–control analysis using the UK General Practice Research Database. The authors report that extensive use of metformin in diabetic men was associated with a slightly increased risk of colorectal cancer, compared with the risk in men who did not use metformin. Extensive use of 2 other drugs for diabetes (sulfonylurea and insulin) was not associated with an increased risk of colorectal cancer. These results indicate that metformin is not associated with a decreased risk of colorectal cancer.

Waterboer et al. Page 287

The costs to prepare, store, and mail sera are high and not practical in many field environments with limited resources. The use of dried blood spot (DBS) biospecimens instead of sera could reduce costs, but the reliability of DBS is uncertain. To address this question, Waterboer and colleagues compared antibody reactivities of sera and DBS from the same donors. They report that quantitative antibody reactivities in DBS and serum were correlated with both high- and low-titer antibodies. The results of this study indicate that DBS can provide a reliable alternative to serum for the detection of antibodies against various pathogens.