Search

10601 Background: This pilot study was conducted to elucidate the barriers and drivers for genetic testing by combining validated measures to inform the content of a clinical trial aimed at increasing uptake of clinically appropriate genetic testing amongst individuals with a personal history of cancer. Previous research shows that the rate of recommendation of genetic testing varies among Healthcare Providers (HCPs) despite national guidelines. Here we evaluate the rate and effect of HCP recommendation on the uptake of clinical cancer genetic testing. Methods: Data for these analyses were obtained from a cross-sectional survey completed by 794 adults (> 18 years) who were diagnosed with cancer in the past 10 years and were seen at an academic medical center within the past two years. Statistical analysis includes logistic regression and crosstabulation with SPSS and R statistical software. Results: Provider recommendation of genetic testing was found to be a very strong indicator of receiving genetic testing across all cancer types (OR = 146.4, p =

10591 Background: Genetic testing allows patients and their families to identify hereditary cancer syndromes. Financial barriers to genetic testing are a major concern for patients offered genetic risk assessment. The cost of germline genetic testing has decreased substantially over the last several years. It is not clear, however, that oncologists are knowledgeable about the cost of genetic testing. The purpose of this study was to investigate oncologists’ knowledge of genetic testing costs. Methods: We deployed a survey to all oncologists who are members of the Michigan Oncology Quality Consortium, a physician-led quality improvement collaborative whose members represent 95% of the oncologists in Michigan. The modified Dillman method was used to achieve the maximum response rate. Responses were collected from December 2020 through May 2021. The responses to the question, “If a patient were to ask you how much it would cost for them to have clinical genetic testing for hereditary cancer syndromes, what would you tell them?” were independently coded by three investigators into one of three categories – correct and helpful, correct and not helpful, or incorrect and not helpful. We investigated associations between the number of years in practice and how important was cost as a barrier to genetic risk assessment. Results: The response rate to the survey was 61.2% (194/317). Only 25% of respondents provided an answer that was deemed to be both correct and helpful to a patient, for example “approximately $250 if out of pocket.” Nearly 40% of respondents gave an answer that was correct but was non-specific and would not help a patient decide about pursuing genetic testing. This category included responses such as deferring to a genetic counselor or that cost varies based on insurance. About 28% gave an incorrect response, such as “several thousand dollars” or responded, “I don’t know.” No associations were found between cost response category, number of years in practice, or reported responses that financial barriers play a role in referral patterns. Conclusions: The majority of oncologists in our statewide sample do not have an accurate understanding of the cost of germline genetic testing. Lack of precise information about the costs of testing may lead patients to believe that genetic testing is either not important or not within reach. Furthermore, providing incorrect information can have a downstream effect and prevent patients and their families from pursuing genetic counseling. Improving oncologists’ knowledge about the cost of testing may decrease barriers to uptake of genetic risk assessment. [Table: see text]

6568 Background: Treatment with AI therapy for at least five years is recommended to reduce risk of cancer recurrence in postmenopausal women with hormone-positive breast cancer, but up to 50% prematurely discontinue their AI. Prior studies have demonstrated higher rates of nonadherence in older patients. Higher rates of comorbidities are also present in older patients, many of which require treatment with prescription medications. However, in prior studies, the impact of polypharmacy on AI persistence in older women has been variable. We evaluated the relationship between use of prescription and over-the-counter (OTC) medications and AI discontinuation in older women in a randomized controlled trial of AI therapy. Methods: In the Exemestane and Letrozole Pharmacogenetics (ELPh) trial, postmenopausal women with stage 0-III hormone-positive breast cancer were randomized to receive exemestane or letrozole and followed for two years. Prescription medication and OTC supplement usage was prospectively collected. Only enrolled participants at least 65 years of age were included in this analysis. Univariate and multivariable Cox proportional hazards models evaluated the association between baseline prescription medications, OTC medications, initial assigned AI medication, and time to discontinuation of initial AI. Results: 131 women at least 65 years old with baseline medication and OTC supplement information were identified. The average age was 70.4 (SD 5). 58 participants (44%) were randomized to exemestane and 73 (56%) to letrozole. 21% had previously received tamoxifen and 18% received taxane-based chemotherapy. The average number of prescription medications at AI initiation was 4.95, and 127 patients reported taking at least one. The average number of OTC medications was 3.23, and 107 patients reported taking at least one. There was no difference in number of prescription medications or OTC medications by study arm. As previously reported, exemestane was associated with premature discontinuation (HR 1.92 [95%CI 1.06-3.50], p = 0.032). Controlling for AI, the risk of discontinuation decreased by 12% for every additional prescription medication (HR 0.88 [95%CI 0.78-0.99], p = 0.038). OTC medication use was not significantly associated with AI duration. Conclusions: Increased prescription medication usage at AI initiation was associated with decreased risk of premature discontinuation of AI. Further research is needed to evaluate this interaction and identify characteristics that impact AI persistence in older women with breast cancer. Clinical trial information: NCT00228956.

Purpose: Patients with cancer are an especially vulnerable population to potential drug-drug interactions (DDIs). This makes it important to adequately screen them for DDIs. The objective of this study was to compare the abilities of nine DDI screening tools to detect clinically relevant interactions with oral oncolytics. Methods: Subscription-based tools (ie, PEPID, Micromedex, Lexicomp, Facts & Comparisons) and free tools (ie, Epocrates Free, Medscape, Drugs.com, RxList, WebMD) were compared for their abilities to detect clinically relevant DDIs for 145 drug pairs including an oral oncology agent. Clinical relevance was determined by a pharmacist using Stockley’s Drug Interactions. Descriptive statistics were calculated for each tool, including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), and then compared grouped by free or subscription-based tools for the secondary analysis and analyzed via generalized estimating equations. Results: For individual metrics, PPV had overall higher values (0.88 to 0.97) relative to the low values included for sensitivity (0.65 to 0.96), specificity (0.53 to 0.93) and NPV (0.38 to 0.83). The top-performing subscription and free tools, Lexicomp and Drugs.com, had no statistically significant differences in performance. Overall, subscription tools had a significantly higher sensitivity (0.85 ± 0.017 v 0.78 ± 0.017; P = .0082) and NPV (0.57 ± 0.039 v 0.48 ± 0.032; P = .031) than free tools. No differences were observed between the specificity and PPV. Conclusion: Due to the low performance of some tools for sensitivity, specificity, and NPV, individual performance should be examined and prioritized on the basis of the intended use when selecting a DDI tool. If a strong-performing subscription-based tool is unavailable, a strong-performing free option, like Drugs.com, is available.

Purpose Breast cancer, the most common cancer worldwide, is the leading cause of cancer mortality in Ghanaian women. Previous studies find Ghanaian women are diagnosed at a younger age and at more advanced stages (III and IV), and have tumors with characteristics similar to African American women. We sought to remedy gaps in knowledge about breast cancer survival in Ghana and its relation to demographic and biologic factors of the tumors at diagnosis to assist in cancer control and registration planning. Methods Individuals with a breast cancer diagnosis who sought care at Komfo Anokye Teaching Hospital from 2009 to 2014 were identified via medical records. Follow-up telephone interviews were held to assess survival. Kaplan-Meier plots and Cox proportional hazards models assessed survival associated with clinical and demographic characteristics. Results A total of 223 patients completed follow-up and were analyzed. The median survival was 3.8 years. Approximately 50% of patients were diagnosed with grade 3 tumors, which significantly increased the risk of recurrence or death (hazard ratio [HR] for grade 2 versus 1, 2.98; 95% CI, 1.26 to 7.02; HR grade 3 v 1, 2.56; 95% CI, 1.08 to 6.07; P = .04). No other variables were significantly associated with survival. Conclusion Higher tumor grade was significantly associated with shorter survival, indicating impact of aggressive biology at diagnosis on higher risk of cancer spread and recurrence. Contrary to prevailing notions, telephone numbers were not reliable for follow-up. Collecting additional contact information will likely contribute to improvements in patient care and tracking. A region-wide population-based active registry is important to implement cancer control programs and improve survival in sub-Saharan Africa.

12068 Background: Aromatase inhibitors (AI) are recommended for at least five years to reduce risk of breast cancer recurrence in postmenopausal women with hormone receptor-positive disease. Despite the potential benefit, up to 50% of patients discontinue AI early, primarily because of musculoskeletal side effects. Some patients, including those with fibromyalgia and other chronic pain conditions, have preexisting pain that arises from altered nociception without clear evidence of tissue or peripheral nerve damage, called nociplastic or centralized pain. We hypothesized that preexisting nociplastic pain is associated with early discontinuation of AI therapy due to toxicity. Methods: Patients diagnosed with breast cancer between 2012-19 and who enrolled in the Michigan Genomics Initiative (MGI) were identified. Patients who were female, had hormone receptor-positive breast cancer, had not received neoadjuvant therapy, and who initiated adjuvant AI therapy were included in the analysis. Prior to breast cancer surgery patients completed surveys about pain in the affected breast and overall worst pain (Brief Pain Inventory [BPI]), nociplastic pain (2011 Fibromyalgia Survey [FS], which assesses widespread pain plus somatic symptom severity), and anxiety (PROMIS Anxiety 4a v.1). Demographics, cancer characteristics, and treatment history, including date and reason for discontinuation of AI therapy, were abstracted from the medical record. Patients were censored at date of last oncology follow-up. Univariate and multivariable analyses were conducted to assess relationship between age, BMI, surgical site pain, overall worst pain, FS score, anxiety, and time to discontinuation of initial AI medication. Results: Of 207 analyzed patients, the average age was 61 years, average BMI was 30.3 kg/m2, 181 (87%) were postmenopausal at the time of breast cancer diagnosis, and 31% received chemotherapy. The majority were prescribed anastrozole (n=196); 6 took letrozole and 5 took exemestane. Average scores prior to surgery were 1.5/10 for worst pain, 0.2/10 for surgical site pain, 4.2/31 for FS score, and 54.1/100 for anxiety. On univariate analysis, only use of exemestane (HR 5.55, p

538 Background: Letrozole is a non-steroidal aromatase inhibitor (AI) used to treat hormone receptor positive (HR+) breast cancer. Variability in letrozole efficacy and toxicity may be partially attributable to variable systemic drug exposure, which may be influenced by germline variants in the enzymes responsible for letrozole metabolism, including CYP2A6. The objective of this genome-wide association study (GWAS) was to identify genetic variants that affect steady state letrozole concentrations. Methods: The Exemestane and Letrozole Pharmacogenetics (ELPh) Study randomized 503 post-menopausal patients with HR+ non-metastatic breast cancer to exemestane or letrozole treatment. Germline DNA was collected pre-treatment and blood samples were collected after 1 or 3 months of treatment to measure steady-state letrozole concentration via HPLC/MS. Genome-wide genotyping was conducted on the Infinium Global Screening Array to the Haplotype Reference Panel ( > 2 million variants). The association of each polymorphism with square-root transformed letrozole concentration was tested in self-reported white patients via linear regression using the standard alpha for genome-wide significance (α = 5x10−8) assuming an additive genetic model and correcting for age and body mass index. Results: 228 patients met inclusion criteria and had all necessary data. Each variant allele of rs7937 a patient carried increased their letrozole concentration ~22.9 ng/mL (standard error = 4.01, p = 3.51x10−8, Table) and this variant explained 13% of the variability in letrozole concentrations. rs7937 is located ~50 kB upstream of CYP2A6, and has previously been identified in GWAS of CYP2A6-related phenotypes, including nicotine metabolism and lung cancer. Conclusions: This GWAS confirmed that steady-state letrozole concentrations are partially determined by germline polymorphisms affecting CYP2A6 activity. If letrozole concentrations affect treatment efficacy or toxicity, CYP2A6 genetics may be useful to individualize letrozole dosing to improve clinical outcomes in patients with HR+ breast cancer. [Table: see text]

532 Background: Chemotherapy induced peripheral neuropathy (CIPN) is a common, debilitating paclitaxel side effect that has been primarily attributed to cumulative systemic paclitaxel exposure. Paclitaxel dose-adjustment to achieve target systemic exposure decreases but does not eliminate severe CIPN, suggesting some patients are inherently CIPN-sensitive. Ephrin ( EPHA) polymorphisms have been reported to increase CIPN occurrence (Baldwin 2012, Leandro-Garcia 2013, Boora 2016) but replication has been challenging, perhaps due to the inability to isolate CIPN-sensitivity by accounting for systemic exposure differences. The study purpose was to determine if EPHA genetic variants previously associated with CIPN occurrence are associated with CIPN-sensitivity. Methods: PN was assessed at baseline and weekly using the 8-item sensory subscale (CIPN8) of the patient-reported EORTC CIPN20 in patients receiving paclitaxel 80 mg/m2 weekly x 12. EPHA4, EPHA5, EPHA6, and EPHA8 were sequenced in germline DNA. Associations with higher PN-sensitivity were tested for three genetic models (total variants, coding variants, and rs7349683) by incorporating genetics into previously published CIPN8 predictive models that included measured paclitaxel exposure. Significant associations were tested for association with higher risk of PN-related treatment disruption (i.e. dose decrease, delay, or discontinuation). Results: In the 59 included patients, carrying EPHA5 rs7349683 was associated with greater CIPN sensitivity (beta coefficient: 0.40, standard error: 0.14, p = 0.005), indicating these patients had a greater increase in CIPN8 during treatment than would be predicted based on cumulative paclitaxel exposure. Rs7349683 was not associated with increased PN-induced treatment disruption, perhaps due to the low number of events (n = 19). Conclusions: Using a novel approach that isolates CIPN-sensitivity by accounting for measured paclitaxel exposure, our results provide further evidence that EPHA5 rs7349683 may be a promising marker for CIPN. If additional validation studies confirm this association, genetic testing could enable personalized treatment strategies to prevent CIPN in patients with breast cancer.

e23018 Background: As the treatment for early-stage breast cancer continually evolves, there has been a pragmatic shift with regard to standard treatment placing a high-value on multidisciplinary management. Recent studies have shown that sentinel lymph node biopsy (SLNB) and adjuvant radiation (RT) can be omitted in a subset of older women with early-stage hormone receptor positive breast cancer who plan to take adjuvant endocrine therapy (ET). Prior analyses estimate a wide range of discontinuation rates, with studies citing 31-73%. The primary aim of this analysis is to estimate the discontinuation rate of endocrine therapy in women ages 70 and older with early stage breast cancer who omit radiation therapy at our institution. Methods: We performed a retrospective review of institutional cancer registry and EMR data to identify all women with breast cancer, ages 70 or older, AJCC 7 stage I or II, hormone receptor positive, HER2 negative, who underwent lumpectomy at the University of Michigan inclusive of years 2014 through 2017. Multiple variables of interest were collected and analyzed. Results: 174 women met initial inclusion criteria for ER/PR positive, HER2 negative, Stage I/II disease, who underwent lumpectomy. Of these, N = 10 pursued adjuvant radiation therapy with omission of ET, and N = 19 declined both ET and RT upfront. The remaining 145 women chose to pursue ET alone (N = 78) or ET+RT (N = 67). 30/145 (21%) discontinued endocrine therapy, 22% (N = 15) in the ET+RT, 19% (N = 15) in the ET alone groups. There was no statistically significant difference between groups with regard to the discontinuation of ET upon comparison, log-rank p-value 0.4. The primary reason for ET discontinuation, N = 27 (90%) was secondary to side effects. Conclusions: This review demonstrated that a significant number (54%) of older women opt to omit radiation in favor of ET alone after multidisciplinary discussion, in accordance with NCCN guidelines. Endocrine therapy discontinuation rates among older women at our institution were comparable to rates in the literature for women of all ages, and the reason for discontinuation was attributed to side effects. The pivotal clinical trials (PRIME II and CALGB 9343) exploring ET alone versus ET+RT did not report on ET discontinuation rates. In our analysis, almost 20% of women receiving ET alone discontinued ET therapy, calling into question a missed radiation treatment opportunity in a curative setting.

e12027 Background: Peripheral neuropathy (PN) is a severe, dose-limiting toxicity of paclitaxel that occurs in up to 25% of patients and can lead to permanent loss of balance and manual dexterity. Due to the lack of effective strategies for PN prevention or treatment, there is a critical need to identify predictive risk factors for paclitaxel-induced PN. Vitamin insufficiencies are known risk factors for PN in other disease states. However, the effect of vitamin insufficiency on paclitaxel-induced PN has not been adequately investigated. Methods: Baseline levels of vitamin D and other nutrients (vitamin B, homocysteine, folate) were measured, and PN was assessed weekly in an observational trial of patients receiving paclitaxel 80 mg/m2 for 12 weeks for non-metastatic breast cancer (NCT0233811). Nutrient levels were measured by Michigan Medicine and insufficiency defined by institutional standards (vitamin D insufficiency < 20 ng/mL). In the primary analysis, the maximum increase from baseline in the 8-item sensory subscale (ΔCIPN8) of the EORTC CIPN20, a validated patient-reported PN assessment tool, was compared in nutrient insufficient and sufficient patients. The effect of vitamin insufficiencies on PN-induced treatment disruptions (dose decrease, delay, or discontinuation) was conducted as a secondary analysis. Results: Only vitamin D insufficiency was identified in enough patients for analysis (15/37 = 41%). Vitamin D insufficient patients reported a greater mean (+/- SD) ΔCIPN8 (36.39 ±22.8) than vitamin D sufficient patients (16.29 ±16.3) (p = 0.003). However, the increase in treatment disruption for vitamin D insufficient patients was not significant (OR = 2.98, 95% CI [0.72, 12.34], p = 0.16). Conclusions: Paclitaxel-treated patients who were vitamin D insufficient at baseline had greater increases in patient-reported PN. If validated in larger studies, vitamin D insufficiency may be a clinically translatable, modifiable risk factor that can be used to prevent paclitaxel-induced PN in patients with non-metastatic breast cancer.

e23096 Background: Cancer survival is at an all-time high; however, patients often face long-term consequences including infertility. Loss of fertility can be devastating and lead to long-lasting emotional distress. Despite consensus guidelines encouraging providers to discuss infertility risk and refer for fertility preservation (FP) consultations, fertility counseling and FP occur in only 10-29% of patients of childbearing age. A few identified barriers include lack of provider knowledge on FP, constraints on time and lack of available resources. Methods: This is a prospective pilot study evaluating the efficacy of an oncofertility program in increasing discussions on infertility risk and FP between providers and patients. The oncofertility program includes medical provider education, an electronic medical record (EMR)-based best practice alert with smart order set, patient education pamphlets and a patient navigator through Reproductive Endocrinology and Infertility (REI). Our primary objective was to increase documentation on fertility in the EMR, while our secondary objective was to increase referrals to REI for females and orders for semen cryopreservation for males. Provider and patient-reported outcomes were also obtained. Results: 19 patients with a new diagnosis of breast or sarcoma (7 pre-intervention and 12 post-intervention) were enrolled and 17 providers participated in the study at the University of Michigan. There was no change in documentation on fertility in the EMR and no increase in fertility referrals placed. However, patients reported a statistically significant change in amount of information provided on infertility risk and FP from "too little" to "the right amount." Providers also reported increased confidence in initiating a conversation about FP and a trend towards increased comfort documenting their conversations on FP in the EMR. Conclusions: With an increase in cancers diagnosed in young adults, there is a need to improve counseling on the risk of infertility with cancer treatment and FP options. Through an easily implemented oncofertility program, we were able to improve medical provider confidence in initiating this important conversation, provide patients with education on infertility risk and FP and streamline the process for FP referrals utilizing the EMR.

Purpose We tested the hypothesis that Akt-Ser473 phosphorylation (pAkt) predicts benefit from the sequential addition of paclitaxel to adjuvant doxorubicin plus cyclophosphamide (AC) chemotherapy in patients with node-positive breast cancer participating in the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-28 trial. Patients and Methods Primary tumors from the NSABP B-28 trial tissue microarray were available from 1,581 of 3,060 patients who were randomly assigned to receive either four cycles of AC alone or followed by four cycles of paclitaxel. Immunohistochemistry and quantitative analysis of pAkt were performed at the National Cancer Institute blinded to clinical outcome. Association between pAkt and clinical outcome was assessed using multivariate Cox modeling adjusting for age, tumor size, number of positive nodes, tumor grade, estrogen receptor status, and human epidermal growth factor receptor 2 status. Results With a median follow-up of 9.1 years, there were no differences in disease-free survival (adjusted hazard ratio [HR], 1.02; P = .81) or overall survival (HR, 0.97; P = .80) with and without receiving paclitaxel among 975 patients with pAkt-negative tumors. In 606 patients with pAkt-positive tumors, the sequential addition of paclitaxel resulted in a 26% improvement in disease-free survival (HR, 0.74; P = .02) or a 20% improvement in overall survival (HR, 0.80; P = .17). Conclusion pAkt significantly predicts disease-free benefit from the sequential addition of paclitaxel to AC chemotherapy in patients with node-positive breast cancer. Patients with pAkt-negative breast tumors do not appear to benefit from the addition of paclitaxel.

543Background: Three aromatase inhibitors (AI), the steroid exemestane, and the azoles anastrozole and letrozole, are effective in the treatment of estrogen receptor positive (ER+) breast cancer by preventing biosynthesis of estrogens including estradiol (E2), estrone (E1), and estrone-sulfate (E1S) in postmenopausal women. OATP1B1, encoded by SLCO1B1, transports E1S into the liver for desulfation to active E1. Women carrying the low-expression SLCO1B1 rs4149056 single nucleotide polymorphism (SNP) have higher E1-conjugate levels (Dudenkov Breast Cancer Res Treat 2017). We hypothesized that patients carrying this SNP would have increased E1S at baseline, and this E1S reserve could resupply E1 and E2 resulting in detectable estrogen levels during AI treatment. Methods: Five hundred postmenopausal women with ER+ breast cancer were randomized 1:1 to either exemestane 25 mg/day or letrozole 2.5 mg/day. Plasma estrogen concentrations were measured prior to and after 3 months of AI treatment using LC/MS/MS (LLO...

e13004Background: T-DM1 is an antibody drug conjugate with proven efficacy in metastatic breast cancer for progressive disease refractory to trastuzumab. FDA prescribing information cites a pneumon...

1063 Background: Aromatase inhibitors (AI) suppress estrogen biosynthesis and are effective treatments for estrogen receptor (ER)-positive breast cancer. In a prospectively enrolled cohort we observed a subset of post-menopausal women who exhibit high plasma estradiol (E2) concentrations during AI treatment, which could potentially contribute to treatment failure. We tested the hypothesis that incomplete E2 suppression is due to insufficient systemic AI concentrations. Methods: Five hundred post-menopausal women with ER-positive breast cancer were randomized to daily exemestane (Exe) 25 mg or letrozole (Let) 2.5 mg. Plasma E2 was measured using GC/MS/MS (lower limit of quantification (LLOQ) = 1.25 pg/mL) at baseline and after 3 months. Let and Exe plasma concentrations measured after 1 or 3 months were compared with the magnitude of E2 depletion using four complementary statistical procedures to assess associations of drug concentrations with: 1) a binary outcome of E2 suppression below LLOQ (logistic regression), 2) 3-month E2 concentrations (linear regression), 3) absolute change from baseline in E2 concentrations (Spearman correlation), and 4) an ordinal outcome defined by E2: decreased to below LLOQ, decreased but not to LLOQ, stayed the same, or increased from baseline (cumulative logistic regression). Results: 397 patients with E2 and AI concentration measurements were evaluable (Exe n = 199, Let n = 198). Thirty (7.6%) patients (Exe n = 13, Let n = 17) had E2 concentrations above the LLOQ at 3 months (range: 1.42-63.8 pg/mL). Exe and Let concentrations were not associated with achievement of unmeasurable E2 concentrations, on-treatment E2 concentrations, E2 change from baseline, or ordinal groupings of E2 change (all p > 0.05). In a parallel analysis there was no association of estrone-sulfate and drug concentrations (data not shown). Conclusions: Our results suggest that circulating drug concentrations do not explain incomplete E2 suppression in women receiving AI therapy. Additional studies are underway to determine whether age, body mass and genetic variation in the aromatase enzyme influence AI treatment response.

e12021Background: Tamoxifen ineffectiveness in patients with hormone receptor + breast cancer may be attributed to subtherapeutic concentrations of its active metabolite, endoxifen. Teft et al. previously published an algorithm that integrates clinical and genetic variables to predict endoxifen concentration. The objective of this analysis was to assess the accuracy of this algorithm in two independent cohorts of tamoxifen treated patients. Methods: Two cohorts of patients taking 20 mg tamoxifen daily and not concomitantly receiving strong CYP2D6 inhibitors were enrolled in UNC LCCC 0801. Endoxifen was measured on two different assays, one measured Z- and E-endoxifen (cohort 1, n = 68), the other Z-endoxifen only (cohort 2, n = 293). Actual endoxifen concentration in the validation cohorts was compared with the Teft algorithm prediction using Pearson correlation and parameter estimates from all cohorts were visually compared. Results: Modest correlation was found for cohort 1 (concordance correlation = 0....

158 Background: Tolerance of AI therapy can be poor due to treatment-emergent toxicities and can lead to early discontinuation (non-persistence). Patients often switch from one AI to another when toxicities develop; however, limited prospective data exist on patients who switch AI. Here we describe the effect of switching from E to L or L to E on tolerance of and persistence with therapy. Methods: Postmenopausal women initiating AI therapy were enrolled on the ELPh trial and randomized to E or L. Those that stopped their AI for self-reported intolerance were offered crossover to alternate AI after a 2-6 week washout. Kaplan-Meier estimates of proportions on AI after 1, 3, and 6 months were assessed during 1st and 2nd AI. Associations between time on 2ndAI and clinicopathologic factors were analyzed using univariable Cox proportional hazards model. To evaluate effect of crossover on patient-reported outcomes, multiple questionnaires, including a pain visual analog scale (VAS), were assessed serially. Results: 83 women, mean age 60 years, 45% prior chemotherapy, and 31% with prior tamoxifen use, participated in the crossover protocol. 71% reported improvement in symptoms a mean 4.72 weeks after discontinuing 1st AI therapy. Median time on 1st AI was 6.8 months (95% CI 5.8-9 months), and on 2nd AI was 11.5 months (6.9-24.2). The probability of persistence at 1, 3, and 6 months for the 1st AI was 94%, 76%, and 55% and for the 2nd AI was 89%, 73%, and 62%, respectively. There was no significant association between duration on 2nd AI and 1st AI (L vs. E), duration on 1st AI, age, body mass index, or prior therapies. The change in pain VAS from baseline to 1 or 3 months was not significantly different during treatment with the 1st or 2ndAI. Conclusions: Although all AI medications have similar mechanisms of activity, nearly two-thirds of patients who are intolerant of one AI are able to maintain therapy for at least 6 months following switch to 2nd AI. Switching is a reasonable approach for women who cannot tolerate 1st AI that may improve persistence with therapy. The mechanisms for intrapatient variation in tolerance warrant further study. Clinical trial information: NCT00228956.

104 Background: The 21-gene recurrence score (RS) assay (OncotypeDX) is prognostic in stage I-II, ER+, HER2-, node negative breast cancer (BC). While RS reflects tumor gene expression its interaction with host factors including metabolic syndrome (MS) is unclear. MS conditions such as obesity have been associated with increased risk of BC and worse prognosis. The aim of this study is to determine correlations between MS conditions and RS group or BC outcomes in both pre- and post-menopausal women. Methods: We retrospectively reviewed women with stage I-II, ER+, HER2- node negative BC evaluated with RS assay treated at University of Michigan since 2005. Demographic, pathologic and treatment data, MS criteria and menopausal status were abstracted from medical records. MS was defined as ≥ 3 of the following within 6 months of diagnosis: BMI ≥ 27.7kg/m2; hypertension (HTN); impaired fasting glucose (IFG); HDL < 50mg/dL; hypertriglyceridemia (trig). RS groups were defined as low ( < 18), intermediate (18-30) and high ( > 31). Results: Of 533 eligible women 74% had stage I disease and 61% were postmenopausal at diagnosis. MS criteria prevalence was 43% BMI, 42% HTN, 34% trig, 14% IFG and 13% HDL. MS was present in 22% and more common in post- vs pre-menopausal women (30% vs 9%; p < 0.0001). RS groups were 55% low, 38% intermediate and 7% high. There was no significant association between RS group and overall MS status or any individual criterion, controlling for stage, and no association after stratification by menopausal status. Postmenopausal status was associated with higher RS group (p = 0.039), independent of stage. Over the 4.2 year median follow-up there were 7 deaths and 20 events, including 5 DCIS, 3 local and 11 distant recurrences and 1 second primary BC. There was insufficient power to detect associations between event rate and MS status. Conclusions: MS and comprising conditions are present in many early stage breast cancer patients, especially postmenopausal women. We were unable to identify associations between RS group and MS criteria. Thus the known worse BC prognosis with MS is not reflected in the RS. Identification of factors underlying this higher risk will be important for optimizing management of patients with breast cancer and MS.

e20635 Background: AI therapy may promote loss of bone mineral density and fractures. We investigated the incidence of falls and fractures in patients (pts) on adjuvant (adj) AI compared to aged-ma...

6566 Background: Sentinel lymph node (SLN) mapping is an integral part of melanoma staging. If regional metastases are identified, patients often undergo lymph node dissection (LND) as definitive t...

1528 Background: Pharmacogenetics requires genetic information from patients’ germline DNA.There is concern that DNA extracted from formalin-fixed paraffin embedded (FFPE) tumors may not accurately...

e20745 Background: The Exemestane and Letrozole Pharmacogenetics (ELPh) trial addressed associations between pharmacogenetics and effects of AI therapy. Here we describe changes in health-related q...

537 Background: We investigated the levels of tumor phospho-Akt(Ser473) (pAkt) and treatment outcome of patients with node-positive breast cancer after adjuvant treatment with doxorubicin/cyclophosphamide (AC) followed by four cycles of paclitaxel (PTX) (AC→PTX) compared with AC chemotherapy alone in the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-28 trial. Methods: The primary tumors on B-28 tissue microarray were available from 1581 of 3060 patients enrolled. pAkt status was examined by immunohistochemistry with the antibody to pAkt(Ser473) (Cell Signaling Technology) at the National Cancer Institute. Levels of pAkt were quantitatively scored with the assistance of an Automated Digital Imaging System blinded to clinical outcome, and categorized by the staining index (intensity X % of staining /100) of > 2 (high) and ≤ 2 (low). The association between tumor pAkt level and clinical outcome at 10 years was assessed using multivariate Cox modeling adjusting for age, tumor size, number of positive nodes, tumor grade, estrogen receptor and HER2 status. Results: Among patients with low tumor pAkt levels (n = 975), there was no DFS difference between those treated with or without PTX (adjusted HR = 1.02, p = 0.81). However, among patients with high tumor pAkt levels (n = 606), those treated with AC→PTX had a 26% reduction in DFS event rate compared to those treated with AC only (adjusted HR = 0.74, p = 0.02). There was no OS difference between treatment groups for those with low pAkt cancer (HR = 0.97, p = 0.80). In patients with high pAkt cancer, those treated with AC→PTX had a 20% reduction in death rate compared to those treated with AC only but this difference did not reach statistical significance (adjusted HR = 0.80, p = 0.17). Conclusions: High levels of Akt phosphorylation at Ser473 independently predict a DFS benefit from the sequential addition of paclitaxel to adjuvant doxorubicin plus cyclophosphamide in node-positive breast cancer. Patients with low levels of pAkt breast cancer may not benefit from the sequential addition of PTX to doxorubicin plus cyclophosphamide. [Table: see text]