THE JOURNAL OF UROLOGYâ
evaluated. Kaplan-Meier curves estimated 5, 10, and 15-year overall (OS) and cancer-speciﬁc survival (CSS). Men receiving frontline radiation therapy were identiﬁed, and Cox proportional hazards models estimated impact on OS and CSS. Log-binomial regression estimated risk of secondary malignancy after radiation therapy. Causes of death were quantiﬁed, and excess deaths from secondary malignancies associated with radiation therapy were identiﬁed. RESULTS: A total of 17830 men (median age 37, 76% Caucasian, 15% Hispanic) with testicular seminomas were included for survival analysis of whom 16969 (95%) had data on radiation therapy and no prior history of malignancy. Over 50% of men with stage 1A, 1B, and 2A seminoma underwent radiation therapy. Survival rates were generally excellent (10-year CSS 1A (99.5%), 1B (99.5%), 2A (98.2%), 2B (97.3%), 2C (96.8%), 1S (99.1%)); improvement in OS was observed with radiation therapy for stage 1 (HR 0.59 (95%CI 0.50-0.70), p<0.001)) and stage 2A (HR 0.29 (95%CI 0.13-0.67), p<0.004) disease. No beneﬁt was observed for stage 2B (p¼0.70) and 2C (p¼0.91) disease. The most common causes of death were cardiovascular/peripheral vascular disease (N¼131), testicular cancer (N¼116), other cancer (N¼115), and accident/suicide/homicide (N¼107). Radiation increased risk of secondary malignancy (RR 1.78 (1.56-2.04), p<0.001) and relative proportion of deaths from other cancers (0.90% vs. 0.41% for no radiation) with most notable differences for respiratory (prevalence ratio (PR) 2.5), gastrointestinal (PR 1.7), lymphoma/leukemia (PR 1.5), and other visceral cancers. For stage 1A (radiation vs. no radiation), the increase in absolute proportion of deaths from other cancers (0.98% vs. 0.28%, p<0.001) was not offset by any reduction in cancer deaths (0.55% vs. 0.46%, p¼0.502). CONCLUSIONS: In a national sample of men, survival rates for testicular seminoma were excellent with increased incidence of secondary malignancies associated with radiation therapy. While deaths due to other cancers was increased, the relative survival beneﬁt of radiation therapy for stage 1B, 1S, and 2A testicular seminoma appeared to outweigh the risks but not for stage 1A. Source of Funding: none
PD53-06 HISTOLOGIC FINDINGS OF LATE RELAPSED SEMINOMA MANAGED WITH RETROPERITONEAL LYMPH NODE DISSECTION Clint Cary*, Joseph Jacob, Richard Foster, Indianapolis, IN INTRODUCTION AND OBJECTIVES: The overall survival for men diagnosed with pure seminoma is excellent. We sought to detail the retroperitoneal pathologic ﬁndings and treatment of men with pure seminoma who experienced a late relapse after their initial therapy. METHODS: The Indiana University Testis Cancer Database was queried for patients with a diagnosis of pure seminoma from 1987 to 2016. Patients with a late relapse who underwent a retroperitoneal lymph node dissection (RPLND) were included. Late relapse was deﬁned as the dates between orchiectomy and RPLND being > 3years. This timeframe was used to account for treatment in the initial year for those who presented with metastatic disease. Patient charts were reviewed to identify patient demographics, tumor, and treatment characteristics. Categorical variables were assessed using Fisher’s exact test. RESULTS: A total of 16 patients met inclusion criteria. Twelve patients initially presented with clinical stage I disease. Of these, 6 (50%) were treated with adjuvant radiotherapy, 5 (42%) were managed with surveillance, and 1 (8%) received 2 cycles of adjuvant carboplatin. The remaining 4 patients initially presented with metastatic disease and were treated with combination platinum-based chemotherapy. The median time between radical orchiectomy and RPLND was 64 months (range 32-415). The location of recurrent disease was the retroperitoneum in 14 patients (88%) and the pelvis in
Vol. 197, No. 4S, Supplement, Monday, May 15, 2017
2 patients (12%). Recurrent seminoma was identiﬁed at the time of RPLND in 5 patients (31%). Three patients (19%) had necrosis only. The remaining 8 (50%) patients were found to nonseminomatous components in the retroperitoneal specimen. Of the 6 patients with clinical stage I initially treated with radiotherapy, 1 patient had recurrent seminoma, 1 was found to have a Leydig cell tumor, 2 patients demonstrated a malignant transformation to sarcoma and rhadomyosarcoma, and 2 were found to have nonseminoma. Patients initially treated with radiotherapy for clinical stage I disease required more induction and salvage chemotherapy as well as more additional procedures (i.e. nephrectomy, bowel resection, etc) at the time of RPLND (66% vs. 33%, p¼0.09). CONCLUSIONS: Patients who received radiotherapy for clinical stage I disease have an unpredictable retroperitoneal histology and require a higher burden of treatment at the time of relapse compared to those without radiotherapy. Overall, 50% of patients were found to have malignant transformation or nonseminoma at the time of late relapse. Source of Funding: none
PD53-07 CONDITIONAL RISK OF RELAPSE IN 3,601 PATIENTS MANAGED WITH SURVEILLANCE FOR STAGE I TESTICULAR CANCER Madhur Nayan*, Toronto, Canada; Gedske Daugaard, Copenhagen, Denmark; Michael Jewett, Toronto, Canada; Jakob Lauritsen, Mikkel Bandak, Mette Saksoe Mortensen, Maria Gry Gundgaard Kier, Copenhagen, Denmark; Philippe Bedard, Aaron Hansen, Padraig Warde, Peter Chung, Eshetu Atenafu, Robert Hamilton, Toronto, Canada INTRODUCTION AND OBJECTIVES: The baseline risk of relapse following orchiectomy is approximately 15% in patients with clinical stage I (CSI) seminoma and 30% in patients with CSI nonseminoma (NSGCT). Surveillance has been widely adopted for initial management. However, the baseline relapse risk does not reﬂect how prognosis may change the longer a patient has survived without relapse. This dynamic relapse risk is referred to as conditional risk of relapse and can provide important prognostic information for physicians and patients. METHODS: We performed a retrospective review of patients managed with surveillance for CSI testicular cancer in Denmark and the Princess Margaret Cancer Center, Toronto, Canada, between 1980 and 2014. Conditional risk of relapse was estimated using the Kaplan-Meier method. We stratiﬁed patients based on validated risk factors for relapse. We used linear regression to determine trends of conditional risk over time. RESULTS: We identiﬁed 3,601 patients of which 2,462 (68.6%) had seminoma and 1,139 (31.6%) had NSGCT. Median follow-up in those without relapse was 12.1 (interquartile range 8.0 to 19.5) years and 9.4 (interquartile range 5.1 to 17.1) years in seminoma and NSGCT, respectively. At orchiectomy, the baseline risk of relapse at 5 years was 53.2%, 22.1%, 21.2%, and 12.4% in patients with high-risk NSGCT (CSIB with pure embryonal carcinoma), low-risk NSGCT (CSIA without pure embryonal carcinoma), seminoma with tumour size 3cm, and seminoma with tumour size less than 3cm, respectively. The conditional relapse risk decreased over time in all groups (p<0.001), but did so at different rates (Figures 1 and 2). For patients without relapse at 3 years, the corresponding risk of relapse within the next 5 years was 6.6%, 0.9%, 3.5%, and 2.0%, respectively (Figures 1 and 2). CONCLUSIONS: Our pooled study is the largest to date to provide conditional risk of relapse for patients with CSI testicular cancer on surveillance. These results can be used to provide patients with prognostic information and tailor surveillance protocols to reduce the intensity of follow-up in patients with a low risk of future relapse.
THE JOURNAL OF UROLOGYâ
Vol. 197, No. 4S, Supplement, Monday, May 15, 2017
Source of Funding: none
PD53-08 CLINICAL IMPLICATIONS OF SERUM N-GLYCAN PROFILING AS A DIAGNOSTIC AND PROGNOSTIC BIOMARKER IN GERM-CELL TUMORS Takuma Narita*, Shingo Hatakeyama, Tohru Yoneyama, Hirosaki, Japan; Shintaro Narita, Akita, Japan; Shinichi Yamashita, Koji Mitsuzuka, Sendai, Japan; Toshihiko Sakurai, Yamagata, Japan; Sadafumi Kawamura, Tatsuo Tochigi, Natori, Japan; Ippei Takahashi, Shigeyuki Nakaji, Yuki Tobisawa, Hayato Yamamoto, Takuya Koie, Hirosaki, Japan; Norihiko Tsuchiya, Yamagata, Japan; Tomonori Habuchi, Akita, Japan; Yoichi Arai, Sendai, Japan; Chikara Ohyama, Hirosaki, Japan INTRODUCTION AND OBJECTIVES: Serum biomarker monitoring is essential for management of germ-cell tumors(GCT). However, not all GCT are positive for conventional tumor markers. We examined whether serum N-glycan-based biomarkers can be applied for detection and prognosis in patients with GCT. METHODS: We performed a comprehensive N-glycan structural analysis of sera from 54 untreated GCT patients and 103 ageadjusted healthy volunteers using glycoblotting methods and mass spectrometry. Candidate N-glycans were selected from those with the highest association; cutoff concentration values were established, and an N-glycan score was created based on the number of positive N-glycans present. The validity of this score for diagnosis and prognosis was analyzed using a receiver operating characteristic (ROC) curve. RESULTS: We identiﬁed 5 candidate N-glycans signiﬁcantly associated with GCTpatients (A). The accuracy of the N-glycan score for GCT was signiﬁcant with an area-under-the-curve (AUC) value of 0.87 (B). Diagnostically, the N-glycan score detected 10 of 12 (83%) patients with negative conventional tumor markers (C). Prognostically, the N-glycan score was comprised 4 candidate N-glycans (D). The predictive value of the prognostic N-glycan score was signiﬁcant, with an AUC value of 0.89 (E). A high value prognostic N-glycan score was signiﬁcantly associated with poor prognosis (F). Finally, to identify a potential carrier protein, immunoglobulin (Ig) fractions of sera were subjected to N-glycan analysis and compared to whole sera. Candidate N-glycans in Ig-fractions were signiﬁcantly decreased; therefore, the carrier protein for candidate N-glycans is likely not an immunoglobulin. CONCLUSIONS: Our newly developed N-glycan score seems to be a practical diagnostic and prognostic method for GCT.
Source of Funding: None
PD53-09 MEN WITH TESTICULAR CANCER HAVE LOWER SEMEN QUALITY COMPARED TO THOSE WITH OTHER MALIGNANCIES Daisuke Noro*, Shingo Hatakeyama, Itsuto Hamano, Toshikazu Tanaka, Takuma Nairta, Tohru Yoneyama, Atsushi Imai, Yasuhiro Hashimoto, Takuya Koie, Chikara Ohyama, Hirosaki, Japan INTRODUCTION AND OBJECTIVES: It is suggested that semen quality is already impaired in testicular cancer (TC) patients. However, the impact of type of malignancy on semen quality remains unclear. The aim of this study is to investigate the semen parameters and associated malignancies of men with cancer who cryopreserved sperm before undergoing therapy. METHODS: We reviewed the database from our cryopreservation laboratory between January 1994 and May 2016, and identiﬁed 114 patients who have banked their sperm before undergoing treatment. Age at banking, semen volume, sperm density, percent motile sperm and type of cancer were recorded. Semen parameters were compared between TC and other malignancy (non-TC). RESULTS: A total of 114 semen samples included 25% with testicular cancer (seminoma: 8.8% and non-seminoma: 17%), 25% with lymphoma, 18% with leukemia, 7.0% with myelodysplastic syndrome, 4.4% with solid cancer, 3.5% with urological cancer, 1.8% with immune system diseases, and unspeciﬁed 13%. Median patient age, semen volume, sperm