532 Background: Metastatic testicular sex cord stromal tumors of the testis (MSCST) comprise an extremely uncommon form of genitourinary malignancy. In a comparative genomic study, we performed comprehensive genomic profiling (CGP) to characterize the genomic alterations (GA) in MSCST and to enable the search for potential therapy targets. Methods: The MSCST were classified as metastatic Leydig Cell Tumors (LCT), Sertoli Cell Tumors (SCT) and Undifferentiated (USCST). In this study, 10 cases of LCT, 6 cases of SCT and 3 cases of USCST underwent hybrid-capture based CGP to evaluate all classes of genomic alterations. Tumor mutational burden (TMB) was determined on 1.1 Mbp of sequenced DNA and microsatellite instability (MSI) was determined on 114 loci. Results: All patients had clinically advanced recurrent and/or metastatic disease. The primary testis tumor was used for sequencing in 6 MSCST (32%) and a metastatic site in 9 (68%) of the MSCST. In 10 (91%) of 11 MSCST positively stained by IHC for inhibin expression. The overall frequencies of GA were similar in all LCT, SCT and UTST ranging from 3.0 to 3.5 GA/tumor. The most frequent untargetable GA found in all MSCST cases included CTNNB1 and CDKN2A/B, both ranging from 20-33% of cases. Targetable GA were uncommon in all MSCST sub-groups but several tumors featured potential for cell cycle inhibitors ( CDK4 in LCT), MTOR inhibitors ( RICTOR, NF2 and PTEN in all 3 tumor types), hedgehog inhibitors ( PTCH1 in LCT) and PARP inhibitors ( BAP1 in SCT). No MSI-High status was identified in any MSCST. The TMB was also low in all MSCST groups. Conclusions: Although the 3 subgroups of testicular MSCST feature defining histopathologic features, these tumors have similar genomic signatures on CGP. The low levels of GA per tumor, infrequent tumor aneuploidy, absence of MSI-High status and low TMB all indicate that testicular MSCST are genetically stable. However, rare cases of testicular MSCST reveal GA linked to potential targeted therapy benefits on CGP linked to dysregulation of multiple biologic pathways. In contrast, the lack of MSI-High status and overall low TMB in testicular MSCST indicates a likely lack of benefit for immunotherapies for these rare forms of malignancy.

514 Background: Metastatic penile (mPSCC) and uterine cervical squamous cell carcinoma (mCSCC) are aggressive forms of malignancy with limited treatment options. We used comprehensive genomic profiling (CGP) to compare the genomic alterations (GA) and their potential impact on targeted and immunotherapy options of mPSCC and mCSCC. Methods: 78 metastatic mPSCC and 604 mCSCC underwent CGP using a hybrid-capture-based next-generation sequencing assay with a mean coverage depth of > 500X. Tumor mutational burden (TMB) was determined on 1.1 Mbp of sequenced DNA and microsatellite instability (MSI) was determined on 114 loci. Results: The median age of mPSCC men was higher than mCSCC women. The HPV+/CDKN2A- status was more frequent in the mCSCC than mPSCC. The GA/tumor were similar for both tumor types (table). TP53 mutations were more common in mPSCC likely linked to loss of original HPV+ status. TERT, NOTCH1 and FAT1 GA were more frequent in mPSCC whereas PIK3CA GA were more common in mCSCC. MTOR pathway targets (GA in STK11, FBXW7 and PTEN) were more common in mCSCC. MSI-High status was extremely rare in all cases, but the relatively high frequencies of TMB ≥ 10/20 mut/Mb in both mPSCC and mCSCC were noteworthy. Examples of patients with mCSCC and mPSCC responding to targeted and immunotherapies will be presented. Conclusions: Although mCSCC and mPSCC share a variety of genomic features, the 2 tumor types can nonetheless be sharply differentiated on CGP. The TP53, CDKN2A and HPV status of the tumor types differ significantly with viral identification much higher in the mCSCC group. There are opportunities for targeted therapies in both groups predominantly identified in the MTOR pathway. Numerous cases with significantly elevated TMB in both mPSCC and mCSCC suggest that immunotherapies might be of benefit in a subset of patients. [Table: see text]

106 Background: Using previously developed prostatectomy incontinence nomogram (PIN) we sought to externally validate the nomogram that predicts probability of incontinence at 6-, 12-, and 24-months after robot assisted radical prostatectomy (RARP). Methods: Prospective data from 663 men with prostate cancer that underwent RARP from 2010 to 2014 at two comprehensive cancer centers and three large group practices was queried. The performance of the previously developed model was evaluated using calibration plots (predicted continence rates versus observed continence rates with 95% CI obtained using Jeffrey’s prior method) and receiver operating curves (ROC). Using Expanded Prostate Cancer Index Composite (EPIC-50) Urinary Function questionnaire, perfect continence was defined as 0 pads, social continence was defined as 1 or 2 pads, and incontinence was defined as ≥ 3 pads used after RARP. Results: The 6-, 12-, and 24- month social continence rates were 77%, 88%, and 93%, respectively. Similar to the 6- and 12-month model development cohort, the external validation cohort has modest predictability with a 6- and 12- month AUC of 0.61, and 0.62, respectively. The 24-month AUC of 0.62 in the external validation cohort is worse than what was reported in the development cohort (AUC 0.80). Conclusions: The externally validated prostatectomy incontinence nomogram is generalizable but has modest 6-, 12-, and 12-month predictability in risk of incontinence after RARP.

who are fit for treatment. We present the experiences of our two supraregional penile cancer referral centres.METHODS: We conducted a retrospective audit of pN3 patients with SCCp at our two centres. Inclusion criteria was any patient whose disease was deemed suitable for adjuvant therapy by the supra-network MDT. Patients who did not start or complete treatment were included in an intention to treat analysis. We recorded dose, site and timing of delivery of radiotherapy AE chemo -sensitisation. Site and timings of recurrence post treatment were established, and both disease specific and overall survival calculated from the date of last lymph node dissection.RESULTS: A total of 151 patients were included. 124 completed radiotherapy AE chemo sensitisation. 23 did not receive treatment and treatment was stopped in 4 cases. Median age was 59 years (range 32-94). Adjuvant (chemo)radiotherapy was administered at a median 75.5 days (range 12 -249) after final nodal surgery. Most commonly used dosing regimen was 45Gy in 20 fractions. Cisplatin was the most commonly used chemo sensitization agent, used in 24% ( 27), 47% ( 53) had no chemotherapy.The mean time to relapse was 8.5 months for the whole cohort, (median 5.5 months). Disease free survival at 5 years was 47.1%. Of the 124 who completed adjuvant treatment 54 relapsed. 50% ( 27) were in field relapses (inguinal or pelvic). In comparison 17 of the 27 who did not have treatment progressed with evidence of inguinal or pelvic involvement in 76.4% (13) of patients after lymphadenectomy.The 5 year cancer specific survival for the whole cohort and adjuvant treatment groups were 44.5% and 47% respectively compared to 31% who did not receive or complete treatment.CONCLUSIONS: Our survival data compares favourably to published data on pN3 survival. This suggests that adjuvant chemoradiotherapy has a role in managing patients with pN3 disease which is associated with a poorer prognosis than pN1 or pN2 patients. The InPACT trial; NCT02305654 is testing the role of chemotherapy vs chemoradiotherapy vs upfront surgery in a randomised control trial.