Intermediate risk prostate cancer (Pca) is a heterogeneous group with relapse rates ranging from 10-40%. Improved risk stratification using biomarkers would allow clinicians to tailor therapy. The purpose of our research is to determine whether a protein signature exists in patient blood that can differentiate between recurrent and non-recurrent intermediate risk prostate cancer patients. Four hundred thirty-eight Intermediate risk PCa pts were entered on a Phase 3 Trial evaluating optimal sequencing of 6 months ADT in combination with dose escalated XRT (76 Gy). Two cohorts were identified: recurrence < 3 years and those tumor free > 7 yrs. We performed in depth proteomic analysis of pretreatment blood samples to identify biomarkers of recurrence using 2 recurrent and 3 non-recurrent patients. In the discovery work we performed 2D gel electrophoresis on depleted plasma, followed by protein sequencing of tryptic fragments of differentially present proteins by mass spectrometry to identify 57 proteins that were differentially present between the 2 cohorts. A test cohort of 20 patients per group was subsequently analyzed for the levels of these 57 proteins along with an additional 14 candidates selected by literature evidence using selection reaction monitoring tandem mass spectrometry (SRM-MS). An optimization phase was included to ensure detection of 2 peptides per protein with 2 transitions for each peptide. Relative protein expression was determined using SRM transition peak areas derived from MultiQuant software. Logistic regression models were used to develop models that identified potential biomarkers of recurrence, and the area under the receiving operating characteristic (ROC) curve was used to assess the ability of each model to differentiate between recurrent and non-recurrent patients. Of the 71 total proteins interrogated, sufficient signal was obtained for 64 of them. A series of 3 models using different combinations of 6 proteins were found to be predictive of recurrence with good sensitivity and negative predictive value (NPV). Inclusion of diagnostic PSA levels did not enhance the predictive ability of the identified models. Using SRM-MS a series of 3 models using different combinations of 6 proteins were found to be predictive of recurrence. We are currently evaluating whether ELISA techniques are sufficiently sensitive to detect the differences between these protein levels in blood samples from our two cohorts.Oral Scientific Abstract 346; Table 1Receiver Operating Characteristics for the Identified Proteomic Prediction Models.Model (after cross-validation)AUC (95% CI)SensitivityNegative Predictive Value3-feature model 10.778 (0.616-0.895)90.5%85.7%3-feature model 20.804 (0.646-0.914)90.5%80.0%4-feature model 10.772 (0.610-0.891)90.5%80.0% Open table in a new tab

Approximately 5% of prostate cancer patients treated with radiation develops severe proctitis. The development of an in vitro assay for radiation response would allow radiosensitive patients to be identified and considered for alternative therapy. The translational research study is part of a randomized trial evaluating the optimal timing of Dose Escalated Radiation and short course Androgen Deprivation Therapy. 430 patients have entered the phase III clinical trial. 3% of patients (Radiosensitive Cohort (RS)) in long-term follow-up have developed grade 3 proctitis. The RS cohort was compared to patients with no proctitis (Control Cohort). Dose Volume Histograms (DVHs) were reviewed in both cohorts. We examined the in vitro γH2AX response in lymphocytes and lymphocyte subsets. Cytogenetic endpoints were examined to provide information about the mechanisms of radio-sensitivity. Peripheral blood samples were irradiated and examined for γH2AX response. A dose course experiment was conducted with 6 dose points (range 0 - 10 Gy), processed 1 hour after irradiation. The time response of γH2AX was also monitored, with 2 Gy irradiated samples being incubated for 0 to 24 hours before processing. Samples were fixed stained with γH2AX-FITC, CD4-PE, CD8-APC and CD19-PC7 for analysis with flow cytometry. Zero and 6 Gy irradiated blood samples were analyzed for chromosome aberrations and excess fragments per cell. Rectal DVHs were comparable between the two groups. At 6 Gy, the mean number of excess fragments per cell in the RS cohort was significantly higher than in the control cohort with mean values of 2.1 ± 0.4 and 1.7 ± 0.3 respectively (p = 0.005). At 6 Gy, there was a trend toward higher number of aberrations per cell in the RS cohort than in controls with mean values of 3.2 ± 0.3 and 3.0 ± 0.3 respectively (NSS ; p = 0.168). In a subset of the data for the RS population, the γH2AX response was higher in the time course experiment. These preliminary results suggest the existence of potential markers for radio-sensitivity. These biomarkers may be useful in the future for tailoring radiotherapy treatments. Further work is being done to validate the results.

Purpose: Patients presenting with hydronephrosis at the time of bladder cancer diagnosis are considered unsuitable for bladder preservation strategies. We have been treating patients with concurrent intraarterial cisplatin and radiation (IPR) for 15 years. As intrarterial delivery results in higher tumor drug concentration than intravenous adminstration we postulated that IPR might be effective in bladder preservation in the context of hydronephrosis and impaired renal function. This analysis was undertaken to evaluate this premise. Materials and Methods: Between 1986 and 2000 we have treated over 200 patients with IPR of whom 63 had hydronephrosis at presentation (54 unilateral and 9 bilateral). After excluding metastatic disease and performing maximal TURBT, upper tract decompression (stent or nephrostomy) was considered depending on renal function. Treatment consisted of cisplatin on days 1,1 and 42 infused via bilateral internal iliac artery catheterization in a dose of 60-120mg/m2 divided between each artery. Radiation began on day 10 giving 40Gy/20 to the pelvic nodes and bladder with a 4 field megavoltage arrangement and a CT planned whole bladder boost of 20GY/10. 6-8 weeks later cystoscopies and biopsies were done to evaluate response. The 9 patients with bilateral hydronephrosis all died of their cancer with both local and metastatic disease. The 54 patients with unilateral hydronephrosis included 42 males and 12 females with a median age of 68.5 years. Pathology was transitional in 51,adeno in 2 and unspecified in 1. Grade 1=3pts,2=11pts, 3=38pts and unspecified in 2. TURBT was done in 45/54 with gross residual present in 47/54, absent in 3 and uncertain in 4. Pelvic nodes were positive in 7/54(13%)Stage was:Ta=1,T1=1,T2=8,T3A=14,T3B=17,T4A=8,T4B=4 and Tx=1. Results: Among the 54 pts with unilateral hydro 42/54(78%) had a complete response and eventual bladder outcome as follows: tumor free bladder=33/54(61%) with 2/33 requiring BCG, salvage cys for invasive=6(11%), salvage cys for superficial 2(3.7%), unsalvaged invasive=5(9%)and unknown in 8(11%). With a median followup of 32 months 2,5 and 10 year overall and cause specific survival respectively are 55%/65%,42%/65%,and 30%/43%. Current status:ANED=26(48%), DNED=9(17%), dead dist=6(11%), dead loc and dist=7(13%), alive loc=2 and unknown=3(6%). Acute toxicity includes 15% requiring medication for bowel or bladder irritation during radiation and one patient needing vascular repair of arterial catheterization site. No major sepsis or bleeding following chemotherapy. Chronic toxicity has been 1 patient with hemorraghic cystitis and 2 pts with RTOG grade 2 bowel toxicity. 16/54(31%) patients developed a sensory sacral neuropathy which was mild/moderate and severe in 7, 8 and 1 patient respectively. Conclusion: Patients with bilateral hydronephrosis quickly develop metastatic disease and are likely best managed with chemotherapy with or without palliative radiation. Patients with unilateral hydronephrosis treated with IPR have a high complete response rate, durable tumor free bladder preservation in the majority and cure rates equal to that following cystectomy.