Development of specific inhibitors targeting the enzyme aromatase represents a highly successful example of rational drug design in hormone-responsive cancer. Aromatase inhibitors (AI) are approved as a first line of therapy in both primary and metastatic estrogen-receptor positive (ER+) breast cancer, and have been shown to be superior to tamoxifen at preventing local and distant recurrence. In spite of the improved clinical efficacy, approximately 20% of women taking AI relapse within 10 years of diagnosis and mechanisms of resistance vary widely. Pre-clinical and small clinical studies have found differences in biochemical efficacy among commonly used third-generation AI that suggest the potential for incomplete estrogen suppression in a proportion of women on AI. However, the relationship between sub-optimal estrogen suppression during initial or adjuvant AI and disease outcomes has never been systemically explored. Liquid-chromatography / tandem mass spectrometry (LC-MS/MS) and liquid chromatography - high resolution mass spectrometry (LC-HRMS) assays provide robust, versatile and highly sensitive platforms for reliable quantification of sex steroids at trace levels present in circulation in post-menopausal women and in women on AI. In this dissertation, rapid and highly sensitive LC-HRMS assays were applied in vitro to conduct a comprehensive analysis of biochemical efficacy of third-generation AI in cell models of AI therapy and resistance. These studies demonstrated that among the third generation AI, letrozole has the greatest efficacy for suppressing estradiol and estrone formation from testosterone and Δ4-androstendione in vitro. Biological response and metabolism of the circulating steroid precursor dehydroepiandrosterone (DHEA) and its sulfate conjugate was assessed in in vitro models of ER+ breast cancer (MCF-7 variant cell lines +/- aromatase). These studies demonstrated estrogen formation from physiological levels of DHEA and DHEA-S in MCF-7 variants after growth in steroid-depleted conditions only in the presence of aromatase. An exploratory clinical study was undertaken for a first look into the variety of estrogen and androgen levels present in serum of women taking adjuvant AI who go on to relapse. A wide variety of serum estrogen levels were quantified from women in this cohort, demonstrating incomplete serum estrogen suppression in approximately half of the women analyzed. The applicability of this work is to shed light on the potential for variable individual response to AIs and the need for larger clinical studies examining the relationship between sub-optimal suppression of estrogens in the initial and adjuvant settings and disease outcomes.