Search

258 Background: We reported initial findings on the phenotypic differences between de novo versus primary progressive metastatic prostate cancer (Finianos et al., ASCO GU; abstr 285). We sought to determine the differences in the phenotypic characteristics of these 2 cohorts of patients and update the data with overall survival and patterns of response to androgen deprivation therapy (ADT) in patients presenting with de novo versus primary progressive prostate cancer. Methods: A retrospective chart review in a single-institution center for a period of 2 years was undertaken. Phenotypic characteristics included age at diagnosis, race, overall survival, treatment patterns, and response to ADT. Analysis was by t-test, Mann-Whitney U test, and Fisher’s Exact test. Results: A total of 90 patients were included in this cohort with de novo, dn (n = 38) and primary progressive, pp (n = 52) patients. There were no significant differences between the 2 cohorts with regard to the median age at diagnosis (dn = 66, pp = 61, p = 0.11), alkaline phosphatase level (dn = 135.5, pp = 86, p = 0.27), BMI (dn = 28.47, pp = 27, p = 0.78), creatinine (dn = 1.02, pp = 0.99, p = 0.34), LDH (dn = 188, pp = 166, p = 0.34), and testosterone on metastasis (dn = 276, pp = 31, p = 0.16). However, de novo cancers were diagnosed with higher mean gleason scores (dn = 8.36, pp = 7.7, p = 0.004), had higher median PSA upon diagnosis (dn = 63.1, pp = 8.8, p < .0001) and higher PSA on metastasis (dn = 61.7, pp = 12.5, p = .0002), and had a statistically significant decreased duration of hormone sensitivity (dn = 642 days, pp = 1783 days, p = < .0001). Patients with d e novo cancers also had a shorter median survival than primary progressive cancers (dn = 2257 days, pp = 4217 days, p = 0.02). Conclusions: Patients who present with de novo metastatic prostate cancer appear to develop early castration resistance and have worse overall survival than those who present with primary progressive disease. We are exploring the molecular differences in terms of androgen receptor expression as a potential etiology for development of early castration resistance.

414 Background: Bladder cancer is the sixth most common malignancy in the United States. Urothelial carcinoma makes up 90% of bladder cancer histology, which may be pure or mixed. It includes adenocarcinoma (Ad), squamous (SqC), glandular, sarcomatoid (St), micropapillary, small cell (SC) and plasmacytoid variants. Pure non-urothelial cancers have worse overall survival compared to urothelial cancer with mixed histologic features. However, little research has been done to characterize pure non-urothelial histologies, and there are no randomized clinical trials evaluating treatment modalities for non-urothelial cancers. This retrospective study characterizes pure non-urothelial cancers and their treatments. Methods: A retrospective chart review of the last 10 years was performed using data from the George Washington University Cancer Center Tumor Registry Data. Statistical analysis was done using the Fisher’s test and Kaplan-Meier survival curves. Results: Out of 449 consecutive patients with bladder cancers, 19 patients had pure non-urothelial carcinoma (4.2%): 7 SqC, 6 Ad, 3 SC, 2 lymphoma (Ly), and 1 St. SqC and Ad were more likely than SC to be diagnosed at an advanced stage (p = 0.04), with median age of diagnosis at 53.5 years for Ad, 68 years for SC and 69 years for Sq. None of the SC metastasized. Primary treatment for 94% of patients was a surgical intervention (9 TURBT, 2 partial cystectomy, 2 radical cystectomy, and 2 nephroureterectomy); 1 received neoadjuvant therapy. 11 patients received adjuvant chemotherapy – 7 with gemcitabine-based regimens and 10 with platinum-based regimens. While not statistically significant, median overall survival varied – 404 days for Ad, 213 days for SqC, and 1567 days for SC. Conclusions: SC was a more favorable histology when compared to SqC or Ad, presenting at an earlier stage with lower incidence of metastasis that perhaps reflected the improved overall survival. Identifying patients with more aggressive disease earlier allows for the potential role for more aggressive therapies that may result in improved outcomes. While the sample size is small, it identifies characteristics and potential differences between bladder cancer with rare histologies.

68 Background: Distress is a non-stigmatic description of emotional, physical, spiritual or psychiatric stressors experienced by patients (pts) diagnosed with cancer. We sought to determine the prevalence of distress in different cancer population of pts seen in our cancer center as they are commencing chemotherapy. Methods: We retrospectively examined data using the Distress Thermometer (DT) based on the National Comprehensive Cancer Network (NCCN) and assessed a single encounter on 240 consecutive patients undergoing their first chemotherapy session. Univariate associations were examined between specific problems and overall distress levels with a 2-tailed between-group t-test. Problem area scores were computed for each subject by taking the mean number of problems rated positive within each area, and associations between each problem area score and distress was examined using Spearman correlations. Results: Among the 240 patients in the sample, mean age was 60 ± 14, 61% were female, and 82% had solid tumors. The overall mean distress, based on the DT reading, was 3.6 ± 3.0. Specific problems reported by the largest number of patients included worry (n = 85), nervousness (n = 79), fatigue (n = 70), sleep (n = 66), and fears (n = 57). Of these, all but fatigue were significantly associated with global distress in univariate analysis. When mean problems per area were calculated, and correlated with global distress, each problem area (practical, emotional, family, physical) had a significant univariate association with global distress, with emotional problems having the highest correlation (r = .52, p < .0001). The only predictors with significant independent associations to predict global distress in the general linear model were emotional problems (p = .0001) and family problems (p = .0062), independent of age, sex and tumor types. Conclusions: In cancer patients undergoing distress screening as they receive their first chemotherapy, emotional and family problems appear to have the highest correlation with distress. Improvement of supportive care services geared towards the betterment of these symptoms is of paramount importance in improving outcomes.

285 Background: In 2014 alone, 29,480 men are projected to die of PCa (Cancer Statistics, 2014).Approximately 4% are metastatic on presentation (de novo, dn) while a subset of men progress on to metastatic disease during the course of their disease (primary progressive, pp). We sought to characterize and compare differences in disease characteristics and outcomes between those who present as dn vs pp metastatic disease in our institution. Methods: A retrospective chart review of dn vs pp metastatic patients for the past 2 years was undertaken. Characteristics such as demographics, Gleason score, duration of hormone sensitivity, and treatment, were obtained. T-test and Fisher’s exact test were used to test differences in patient and disease characteristics between dn and pp metastatic patients. Results: Preliminary results on 10 de novo and 13 primary progressive patients are presented herein. Statistically significant differences were noted for median age at diagnosis: de novo 66.5 years of age (range=48-71) vs primary progressive 58 years of age (range 42-77), p=.05; Median duration of hormone sensitivity: de novo 149.5 days (range=91-440) vs primary progressive 444.5 days (range 91-3041), p = 0.0002; Median Gleason Score: de novo 9 (range 6-10) vs primary progressive 8 (range 6-9), p = .002. Although there was a tendency towards higher mean and median of PSA levels for the de novo (225.16 and 50 ng/ml, respectively) vs primary progressive (16.97 and 28.2 ng/ml, respectively), the difference was not statistically significant (p=.15). There was also non-significant trends for the de novo group to be more likely to have a lymph node metastasis (70%, p=.40) and Caucasian (62%, p=.16) compared to the primary progressive group (46% lymph node metastasis, 23% Caucasian). Conclusions: While the reported samples are small, our data reveals a potential difference in disease aggressiveness in those presenting with de novo metastatic cancer with higher-risk disease and shorter time to castration resistance. Further data collection is underway and a prospective trial assessing the difference molecularly between both groups is planned to confirm these findings.