24-Hour Urinary Steroid Metabolites As Predictors Of Response To Chemotherapy In Metastatic Adrenocortical Carcinoma (Macc)

e15625 Background: Adrenocortical carcinoma is very rare with an estimated incidence of 0.5-2 new cases/million/year. Patients with aggressive disease necessitating chemotherapy often fare poorly, and due to the limited number of cases, there is a paucity of data regarding their treatment outcomes. Recent novel methodology has shown the usefulness of urine metabolomic analysis as a tool for differentiating ACC from adrenocortical adenoma. We hypothesise that the same methodology can be utitlised as a tool to assess response to chemotherapy Methods: We retrospectively reviewed electronic and paper case notes of all patients treated with chemotherapy for mACC at our institution from 2006 to 2013. Results of 24 hour urine metabolomics including pregnenediol (5PD), 5-Pregnenetriol (5PT), and tetrahydro-11-deoxyxortisol (THS) were reviewed where available. Steroid metabolite excretion was compared to results from the reference cohort (n=147) from Arlt et al. 2011, and computed determined likelihoods for ACC were calculated at different times through treatment with chemotherapy. Results: 19 patients were identified (8 female, 11 male) with a median age of 53 (range:25-77). 8 patients were alive at time of analysis. 13 patients (68%) were treated with etoposide/doxorubicin/cisplatin (EDP) as first line chemotherapy, with the remainder receiving streptozocin. The median number of cycles received was 3, with 4 patients (21%) going on to receive 2ndline treatment with Gemcitabine/Capecitabine. The median PFS was 3.5 months (range:0.4-9.9). 3 patients had 24 hour urine metabolomics data available prior to, during and after chemotherapy. In these 3 patients, the 3 compounds 5PD, 5PT and THS were found to correlate with response to chemotherapy and progression, with computer determined likelihood scores for ACC changing clearly in concordance with clinical and radiographic response and progression on chemotherapy. Conclusions: We present a cohort of patients who have received chemotherapy for metastatic ACC, and show a potential utility for 24 hour urine metabolomics as a predictive tool for response. The authors suggest further prospective analyses are required to validate this finding.