Currently, little is known about factors acting as genetic modifiers to ovarian cancer tumorigenesis. While several studies have explored the effect of genetic variants on cancer risk in BRCA mutation carriers [31–33], no distinct genetic variant that strongly predicts cancer risk in these subgroups have been identified.
Studies addressing the impact of the MDM2 SNP309 status on risk of tumor development and age at onset in different types of cancers have provided conflicting results, with a trend for positive associations in Asians but a lack of correlation in Caucasians [23, 25–27]. While the MDM2 SNP309G is an ancient polymorphism detected in all ethnic groups examined so far, its frequency display distinct ethnic variation , accounting for 12.4% of the alleles in African Americans  contrasting about 34% in Caucasians [28, 34] and 50-55% in Asians [30–37]. In contrast, SNP285C is a more recent polymorphism arising on the SNP309G allele and with a distribution so far restricted to Caucasian populations [28, 30]. Previously, we found the SNP285C/309G haplotype to account for 11.7% of all the SNP309G alleles in Western Europeans (British, Dutch and Norwegian populations) but in less than 2% of Finns and absent in healthy Chinese individuals . In general, this distribution has been confirmed by The 1000 Genomes Project Consortium . Thus, we hypothesize that the occurrence of the SNP285C/309G haplotype may have contributed to the conflicting results from studies addressing the impact of SNP309 status on ovarian cancer risk in Caucasian populations.
Only a few studies have addressed the risk of ovarian cancer with respect to SNP309 status. Three studies (two Caucasian populations and one Japanese) reported no association between SNP309G status and elevated risk of ovarian cancer [37, 39, 40]. In contrast, the SNP309G allele was found to be a potential protective factor, associated with a reduced risk of ovarian cancer, in a Chinese population . However, these studies included a limited numbers of patients (302 or less) without defined BRCA mutation status. In our previous study on Caucasians evaluating nearly 2,000 sporadic ovarian cancer cases and > 3,500 healthy controls, we found an elevated risk of ovarian cancer in carriers of the SNP309TG and SNP309GG genotypes . However, we found a reduced cancer risk in carriers of the SNP309TG genotype harboring the SNP285C/309G haplotype.
To the best of our knowledge, only two studies have addressed the impact of SNP309 on ovarian cancer risk in BRCA mutation carriers. Yarden et al.  found the SNP309GG genotype to be significantly associated with risk of BRCA1 related ovarian cancer in Ashkenazi Jews diagnosed before 51 years of age, while Copson et al.  detected a non-significant trend for increased incidence of ovarian cancer in British BRCA1 mutation carriers harboring the SNP309GG genotype. While the distribution of the SNP285C allele in Ashkenazi Jews is unknown, it occurs in the British population , thus, it may have influenced the result reported by Copson et al. , partly masking an effect of the SNP309G status on ovarian cancer risk in this population.
Here, we report an elevated OR for ovarian cancer in BRCA1 mutation carriers harboring a MDM2 SNP309TG or SNP309GG genotype. While the exact OR values presented should be interpreted carefully due to a limited number of observations, the results indicate that SNP309 status may influence ovarian cancer risk with OR to a similar extent in BRCA1 mutation carriers as observed in sporadic ovarian cancer .
While the SNP285C/309G haplotype was observed in 13.1% of the SNP309G allele carriers in healthy controls , the percentage in BRCA1 mutation carriers was only 7.3% (OR 0.50; CI 0.24-1.04). The lower incidence of the SNP285C/309G haplotype found in BRCA1 mutation carriers with ovarian cancer may indicate a risk reducing effect of the SNP285C allele. However, this finding needs confirmation by independent studies.
A key feature in BRCA related breast and ovarian cancer is earlier age at onset as compared to sporadic disease, with the most distinct differences related to BRCA1 mutations. When censoring patients harboring the SNP285C/309G haplotype from our analyses, we found BRCA1 related cancer cases carrying the SNP309TG and SNP309GG genotypes to be on average 3.1 year younger at time of diagnoses as compared to carriers of the SNP309TT genotype. Although this difference was of borderline statistical significance, it supports the finding that the SNP309G allele promotes ovarian tumorigenesis in BRCA1 mutation carriers. If confirmed in other studies, this information may be useful for timing of risk reducing surgery in healthy BRCA1 mutation carriers, as the age at onset in index cancer cases has been found to predict individual risk .