Although many factors influence the effects of

Although many factors influence the effects of medications, during recent years it has become evident that genetic factors could explain the differences between individuals in drug response. These differences are due to variants in genes encoding drug targets [8]. The challenge for pharmacogenetics is to establish the relation between gene variant and medication response, to develop diagnostic tests that can predict drug action and adjust therapy accordingly [8]. Therefore, genetics could be used to predict reproductive lifespan and to select the best treatment to obtain the best outcome. The AR gene has been associated with ovarian failure in both animals [9] and humans [10]. The role of testosterone as an estradiol precursor in females is well understood [11]. However, the direct involvement of androgens in female reproductive physiology remains controversial [12]. Recent studies appear to support the importance of proper androgen levels to female fertility [13] and some strategies have been developed with the ultimate objective to increase ovarian response [14]. The human AR (androgen receptor) gene contains a highly polymorphic CAG repeat sequence within crizotinib receptor 1. Recently, variations of the AR-CAG tract, while still within the normal range (11–38), have been linked to an increase in the severity of different diseases [15]. Recent researches reported an association between CAG repeats in AR gene with premature ovarian failure patients [10,16]. The aim of this study is to investigate whether CAG polymorphism on AR gene have predictive value for POR and if it could be used to select patients that could benefit from the use of androgens in previous cycles. From our knowledge, nothing is known about the potential effect of CAG repeats in AR gene and COS in POR patients. In order to find a possible correlation between CAG on AR gene polymorphism and POR we proposed to evaluate the difference of biallelic mean of CAG-tract between controls and POR. To evaluate the effect of AR polymorphism in the COS where androgens were used, we compared the cycles with and without androgens as a pretreatment performed by the same patient.
Materials and methods
Results
Discussion The present study showed that CAG repeat length in exon 1 of the AR gene in POR patients trends to be longer than that observed in women belonging to the general population. To our knowledge this is the first investigation about the involvement of AR gene polymorphism in POR. Moreover, our data suggest that the AR genotype could clarify the effectiveness of the androgens pretreatment. AR genotyping could help us to identify POR patients that will be actually benefited from transdermal testosterone pretreatment. For the first time we have a consensus about definition of POR. According with the Bologna criteria we can determine a homogeneous population [20,21]. In our study we decided to consider only patients under 39 years and no known cause or risk factor for POR. It means that all our patients were previously under COS with low response and all of them had biomarkers of POR. We included in the study group only young patients trying to avoid physiologic low ovarian reserve assuming that it’s no useful to predict a low fertility potential in women over 38 years [22]. While this is the first study to analyse the distribution of this polymorphism in patients with POR, its relationship with premature ovarian insufficiency (POI) has been investigated with discordant results. A study reported no significant differences in allele distribution of the CAG-tract between controls and POI in Caucasian [23]. Another investigation in Japanese women found shorter CAG repeat compared to control [16]. Finally, Chatterjee et al found longer CAG repeat in exon 1 of the AR gene in Indian patients with POI [10]. Heterogeneity of patients with POI and different ethnic backgrounds may explain the conflicting results. Moreover, in normal ovarian reserve patients, it has been shown that AFC is associated with CAG polymorphism of the AR gene [24].