Search

5520 Background: The DESKTOP III trial has demonstrated a significant survival benefit in AGO-score positive patients who underwent complete cytoreduction at 1st relapse compared to those treated with chemotherapy alone. The question whether eligible patients who missed the opportunity of potentially life prolonging surgery at 1st relapse would benefit from surgery at the time of their second relapse, remains open. Methods: Patients randomized in the standard, non-surgical arm of the DESKTOP III trial who underwent cytoreductive surgery at a subsequent relapse at investigator’s discretion were separately analyzed. Results: The median progression-free survival (PFS) counted from randomization of 201 patients in the control arm of DESKTOP III was 14.0 months. 171 (85%) had progressive or relapsing disease and 32 of 171 (19%) underwent cytoreductive surgery. Patients’ median age at this subsequent surgery was 63 years (range: 46 – 78). Complete tumor resection was achieved in 19 patients (60%), while 5 (16%) had postoperative residual disease (n = 8 missing data). Sixteen patients (50%) commenced systemic treatment within 90 days from surgery, as documented. Thirty- and 90-day surgical mortality rates were 1 (3%) and 2 (6%), respectively. Within a postoperative median follow-up time of 43.8 months, 12 (38%) deaths were reported. Median overall survival after surgery (OS) was 54.0 months. One- and 2-year OS rates were 91% and 84%, respectively. Conclusions: Cytoreductive surgery for subsequent ovarian cancer relapse appears feasible and with low mortality in selected patients who received non-surgical treatment at 1st relapse despite a positive AGO-score. Surgery should be considered as an option in carefully selected patients also later in their journey within a specialized gynecological cancer setting. Clinical trial information: NCT01166737.

5564 Background: This is a prospective international Survey to evaluate the impact of the COVID-19 Pandemic on the management of patients with gynecological malignancies from the multidisciplinary physicians' perspective, with particular focus on clincial infrastructures, and trial participation. Methods: The anonymous online survey consisted of 53 COVID-related questions. It was sent to all healthcare professionals in gynaecological oncology centres across Europe and the Pan-Arabian region from April 2020 to October 2020. All healthcare professionals treating women with gynecological cancers were able to participate in the survey. Results: A total of 243 answers were collected from 30 different countries. The majority (73%) of participants were gynecological oncologists from university hospitals(71%) with at least an Intensive care unit with cardiopulmonary support available at their institutions. Most institutions continued to perform elective surgeries only for oncological cases (98%). Patients had to wait on average 2 weeks longer for their surgery appointments compared to previous years(range 0-12 weeks). Cases that were prioritised for surgical intervention across all tumors (Ovarian, Endometrium, Cervical) were early stage disease (74%), primary situation (61%), and good ECOG status (63%). The radicality of surgery did not change in the majority of cases (78%) across all tumor types. During the pandemic, only 38% of clinicians stated they would start a new clinical trial. 45% stated the pandemic has negatively impacted the financial structure and support for clinical trials. 79% do not routinely screen patients included in trials for SARS CoV2. Overall, approx. 20% of clinicians did not feel well informed regarding clinical pathways for COVID-19 patients throughout the pandemic. The majority preferred regular updates and training via Webinars (75%), followed by tumorboards and interdisciplinary conferences (45%). 30% of clinicians stated that they are currently experiencing difficulties in providing adequate medical care due to staff shortage. Conclusions: Despite well-established guidelines for patient care and performing clinical trials in gynecological oncology, the COVID-19 pandemic has impacted clinical research, and financial structures. Longer waiting times for operative interventions, less support for clinical trials and concerns regarding provision of adequate medical care and triaging patients are very real. This survey underlines the necessity for building robust emergency algorithms tailored to gynecological oncology patients in the future.

Purpose Venous thromboembolism (VTE) has been associated with negative prognosis in cancer patients. Most series reporting on VTE have included different tumor types not differentiating between recurrent or primary disease. Data regarding the actual impact of VTE on primary advanced ovarian cancer (AOC) are limited. Patients and Methods Between 1995 and 2002, the Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian Cancer Study group (AGO–OVAR) recruited 2,743 patients with AOC in three prospectively randomized trials on platinum paclitaxel-based chemotherapy after primary surgery. Pooled data analysis was performed to evaluate incidence, predictors, and prognostic impact of VTE in AOC. Survival curves were calculated for the VTE incidence. Univariate analysis and Cox regression analysis were performed to identify independent predictors of VTE and mortality. Results Seventy-six VTE episodes were identified, which occurred during six to 11 cycles of adjuvant chemotherapy; 50% of them occurred within 2 months postoperatively. Multivariate analysis identified body mass index higher than 30 kg/m2 and increasing age as independent predictors of VTE. International Federation of Gynecology and Obstetrics stage and surgical radicality did not affect incidence. Overall survival was significantly reduced in patients with VTE (median, 29.8 v 36.2 months; P = .03). Multivariate analysis identified pulmonary embolism (PE), but not deep vein thrombosis alone, to be of prognostic significance. In addition, VTE was not identified to significantly affect progression-free survival. Conclusion Patients with AOC have their highest VTE risk within the first 2 months after radical surgery. Only VTE complicated by symptomatic PE have been identified to have a negative impact on survival. Studies evaluating the role of prophylactic anticoagulation during this high risk postoperative period are warranted.

TPS5602 Background: Primary cytoreductive surgery (PDS) followed by chemotherapy has been considered as standard management for advanced ovarian cancer patients (pts) over decades. An alternative approach of interval debulking surgery (IDS) following neoadjuvant chemotherapy (NACT) was subsequently reported by two randomized phase III trials (EORTC‐GCG, CHORUS). Owing to important limitations of these studies, especially regarding surgical quality, optimal timing of surgical therapy in advanced ovarian cancer is still unclear. Methods: TRUST is an international open, randomized, controlled multicenter trial investigating overall survival (OS; primary endpoint) after PDS vs NACT and subsequent IDS in pts with FIGO stage IIIB‐IVB ovarian, tubal, and peritoneal carcinoma. Secondary objectives are safety of complete tumor resection, progression‐free survival and quality of life (QoL) as well as surgical morbidity. In order to guarantee adequate surgical quality, participating centers need to fulfill specific quality assurance criteria (e.g. ≥50% complete resection rate in upfront surgery for FIGO IIIB-IV pts, ≥36 debulking-surgeries/year) and agree to independent audits by TRUST Quality committee delegates. A 1:1 randomization to PDS or NACT followed by IDS stratified by center and age‐ECOG combination (ECOG 0 and age ≤65 years vs ECOG > 0 or age > 65 years) is performed. Pts in the PDS arm will undergo surgery followed by 6 cycles of platinum-based chemotherapy, whereas pts in the IDS arm will be treated with 3 cycles of NACT after histologic confirmation of the disease, followed by IDS and subsequently 3 cycles of platinum-based chemotherapy. Intention of surgery for both groups will be complete tumor resection as per guideline recommendations. Health related QoL will be assessed using the EORTC QLQ‐C30, QLQ‐OV28, and EQ‐5D‐3L questionnaires. For sample size planning, we considered a prolongation of median OS from 45 months in the IDS arm to 60 months in the PDS arm (HR 0.75) as clinically relevant. 380 events are needed to obtain a power of 80% in two‐sided log-rank test with significance level of 0.05. The primary analysis will be done in the ITT‐population of 686 randomized pts. By Feb 3 2017, 46 pts were randomized. Clinical trial information: NCT02828618.

e16569 Background: High-grade serous ovarian cancer is characterized by high degrees of genomic instability and heterogeneity, with the majority of patients eventually acquiring resistance to plati...

5570 Background: Non-epithelial malignancies of the ovary account for

5562 Background: Incomplete surgical staging was shown to be a negative prognostic factor for patients with borderline ovarian tumors (BOT). However, little is known about the prognostic impact of each individual staging procedure. Methods: Clinical parameters of 950 patients with BOT (confirmed by central reference pathology) treated between 1998 and 2008 in 24 German AGO centers were analyzed. Staging quality was evaluated following German guidelines. In patients with adequate surgery regarding their reproductive organs (bilateral salpingo-oophorectomy or intended fertility preserving surgery in patients ≤55 years), other recommended staging procedures (omentectomy, peritoneal biopsies, cytology, and, for mucinous histology, appendectomy) were evaluated applying uni- and multivariate regression models with respect to progression-free survival (PFS), separately for serous and mucinous histology. Results: A total of 559 serous BOT (64.1%) were opposed to 268 mucinous BOT (30.7%). For patients with serous ...

5562 Background: Malignant ascites in PROC causes significant impairment in quality of life. The Sequana Medical alfapump System (AP), a remotely controlled device connecting the patients’ peritoneal cavity to their urinary bladder, has been evaluated for the continuous drainage of ascites in liver cirrhosis, but not as yet for malignant ascites. Methods: We implanted the AP in the peritoneal cavity of a 67y old heavily pretreated PROC patient with recurrent malignant ascites requiring 3-5 liters drained 3 times/month. The AP was evaluated for its ability to drain ascites into the urinary bladder using an electronic download of recorded volume pumped, cross correlated with weekly ultrasound, symptomatic scores and QoL evaluation. Early morning urine for evaluation of urinary cytology and tumor-cell molecular analysis was collected weekly. Results: The implantation was performed under general anaesthesia in a 60 minute procedure. Before insertion the patient had 3 liters of malignant ascites; 2 liters were drained during surgery. The pump, draining 350ml ascites per day successfully drained the ascites to dryness after 3 days. The patient underwent weekly sonography and monthly cystoscopies. She did not report pollaki- or dysuria, only an increased micturition volume. Histopathological analysis of the urine revealed rich malignant cell content, used to create FFPE cell blocks weekly for molecular- pathological profiling with sequential Caris Target Now analysis and full exome sequencing. Conclusions: On initial evaluation, the AP represents a tolerable and effective means of diverting peritoneal ascites into the urinary bladder and thus preventing its recumulation in PROC. This innovative approach not only addresses an area of unmet need for the control of malignant ascites but also provides a method of collecting tumor tissue and evaluating longitudinal change in molecular tumor characterization. A EUTROC multicenter European randomized trial (AMAZE) is planned for evaluation of clinical and translational implications of the AP in PROC.

5553 Background: Alopecia is a relevant side effect of chemotherapy. Our aim was to validate and unconfirmed broad patient´s opinion that a low rate of alopecia may predict poor response to chemotherapy and poor overall survival (OS). Methods: Individual patient data analysis of 5114 patients from four prospective randomised phase III trials conducted between 1995 and 2004 to investigate platinum-taxane based chemotherapy regimens in advanced ovarian cancer. Uni- and multivariate analyses were performed adjusted for age, number of cycles, individual trial, residual mass, tumor stage and histology. Results: 5,114 patients with ovarian cancer (OC) were analyzed. Most patients presented with advanced stage FIGO III/IV (87.8%). A median of 6 cycles were applied (range 0-11). Worst alopecia grade was 0 in 2.2%, 1 in 2.7% and 2 in 87.1%. In 8% patients no data about alopecia were documented. Patients with complete alopecia were more likely to achieve remission (OR 2.83, 95% CI 1.68-4.78 for grade 2 compared to grade 0/1) and had favourable progression-free survival of 19.0 months, (95%CI 18,2-19.7) compared to patients with grade 0/1 (13.8, 95%CI 12.2-15.3 and 14.3, 95%CI 10.8-17.8). Median OS was also significantly longer after grade 2 alopecia 48.8 months (95%CI 47.0-50.5), compared to 28.1 months (95%CI 22.3-33.9) and 33.9 months (95%CI24.3-43.6) for grade 0 and 1, respectively. This prognostic impact was not reteined in multivariate analysis. However, onset of alopecia was an independent prognostic factor for OS: patients with complete alopecia after cycle 3 had a favourable outcome compared to patient who experienced alopecia later during therapy (HR 1.22, 95%CI 1.02-1.46) or no alopecia (HR 1.29, 95%CI 0.98-1.70). Conclusions: We could show for the very first time that there is no evidence that the rate of alopecia is associated with the effect of chemotherapy is of any prognostic relevance. The observation that early onset of alopecia is associated with OS should be confirmed.

5593 Background: We conducted an international survey to evaluate the differences in the systemic, radiotherapeutic and operative management of endometrial cancer (EC) in different regions of the world. Methods: In 2009 avalidated 15-item-questionnaire regarding surgical and adjuvant procedures of EC was sent to all German gynaecological clinics and in 2010 the English adapted questionnaire was set online as well as sent per post in most major gynaecological cancer societies. Results: 316 German institutions and 302 Institutions from 24 countries participated. We combined the different countries into regional groups: Central Europe (CE), southern Europe (SE), Asia and USA/UK. In Asian countries and in CE a lymph node dissection (LND) was performed routinely in 72.8% and in 55.6% of the cases, whereas in the USA/UK and in SE a LND was done mainly in selected cases when specific risk factors such as high-grade or non-endometrioid-histology applied (62.8% and 72.5%) than routinely (p < 0.001). A systematic pelvic and paraaortic LND was performed most frequently in CE 91.0%, in SE 76.9%, in Asia 70.9% and in USA/UK 68.6% (p < 0.001). A systematic LND with the intention of both adequate staging and for therapeutic value was performed in countries of central Europe to 74.6% and in Asia to 67.2%. In USA/UK the LND was seen merely as a staging instrument by 53.5% (p < 0.001). The LND was performed up to renal veins in CE in 86.8%, in Asia in 80.8%, in USA/UK in 51.2% and in SE in 45.1 %. A significant difference war found in the treatment for FIGO stage I (high risk factors (high grade, L1,V1)) and stage II disease between the countries: chemotherapy was applied in 84.8% of the participated centers in Asia,42.3% in SE, 21.2% in CE and only 13.6% in USK/UK (p

e19597 Background: Sexuality is a crucial issue influenced by social norms, shame and moral taboos and is generally not discussed in the clinical day and research community. Aim of our study was to systematically evaluate the sexual function and quality of life of cancer pts after operative and systemic treatment in GM. Methods: We assessed in a prospective setting sexual function and quality of life aspects of patients with histological proven GM after completed treatment. Validated questionnaires about sexuality (Female Sexual Function Index 19 items), quality of life (SF-12) as well an additional semi-structured questionnaire consisting of 20 items were answered by all pts at the earliest 3 months after end of treatment during cancer care follow up visit. Results: Overall 55 pts with median age 61 (range, 22-74) were evaluated including 54% ovarian, 26% breast, 13% cervical cancer, 2% endometrial- and 6% vulvar cancer pts. Overall 32.7% (n=18) of the pts stated that they have sexual problems. Based on the FSFI (

5053 Background: Borderline ovarian tumors (BOT) are a low-grade form of ovarian malignancy with significantly less aggressive behavior than epithelial ovarian cancer (EOC). Since there is neither convincing scientific evidence nor a marker capable of predicting BOT’s behavior, we analyzed the role of immune tolerance in differential disease outcome. EOC and BOT were chosen due to similar disease etiology, despite different disease courses. Increased numbers of T-cell subpopulations infiltrate malignant tumors, so we used epigenetic tests to determine the prevalence of regulatory T-cells (Treg) and overall-T-Lymphocytes (oTL) in EOC and BOT. Methods: The ovarian cancer samples were provided by the European multicentric OVCAD study. The BOT samples were obtained from Tumor Bank Ovarian Cancer at Charité Campus Virchow (Germany). Samples and clinical data were prospectively collected and documented using validated SOPs. DNA was bisulphite-converted and forwarded to methylation-specific RT-PCR for CD3 and FOXP3 loci. Results: We evaluated 90 high-grade EOC, 12 BOT with invasive implants and 25 non-invasive BOT samples. Higher oTL-values correlate with mortality (p=0.008) and recurrence (p

5005 Background: Borderline ovarian tumors (BOT) are a rare entity; current standard of care is based on the available data of predominantly small retrospective trials. Therefore we performed a pattern of care study including central pathology review. Methods: All consecutive patients diagnosed with BOT between 1998 and 2008 in 24 German institutions were included. Tumor samples were prospectively sent for central histopathological review to specialized gynecopathologists, clinical data were collected and patient follow-up was prospectively updated. Results: Pathological review was obtained in 1,042 of 1,236 pts resulting in 950 confirmed BOT cases analyzed here. Under- and overdiagnosis occurred in 3.8% and 5.0% of cases, respectively. Median age was 49 years; 84% of patients had FIGO stage I disease; serous type (S-BOT) was diagnosed in 64% and mucinous type (M-BOT) in 31%. Primary/re-staging surgery led to complete debulking in 92.3% of pts (residual disease 1.3%, unknown 6.4%). Adjuvant chemotherapy was given to 33 (3.5%) pts only. 165 (17%) underwent fertility preserving surgery and 31 (19%) of these patients had documented pregnancies thereafter. Overall, 74 (7.8%) pts experienced relapse and 43 (4.5%) died. Disease progression in the form of invasive carcinoma occurred in 30% of the relapses. Inadequate surgical staging, residual tumor, fertility sparing surgery and higher FIGO stage were associated with shorter progression-free survival (PFS). M-BOT showed a non-significant trend to longer PFS compared to S-BOT (p = 0.07). No differences were observed for laparatomy vs. laparoscopy as initial surgical approach or application of adjuvant chemotherapy. Conclusions: To this day, this is the largest data set available for BOT. Prognosis is favorable even without adjuvant therapy if correct surgical staging is performed. Both tumor characteristics and treatment variables had a significant impact on relapse rate and outcome. In contrast to previous studies, disease progression in the form of invasive carcinoma occurred in a significant amount of patients with relapsed disease.

e15531 Background: EpCAM, a well known cancer antigene is currently experiencing a renaissance in its use as a binding site for targeted oncologic therapy and prognostic marker in various epithelial carcinomas such as breast cancer or carcinomas of the oral cavity. Goal of this retrospective study was to identify EpCAM as a potential prognostic marker for patients with primary epithelial ovarian cancer (EOC). Methods: EpCAM expression was assessed in tumor tissue by immunohistochemistry using the Avidin-Biotin-Complex method on paraffin-embedded ovarian cancer tissue samples. EpCAM overexpression was defined as an expression of EpCAM as high as 76-100%. Clinical data as well as tumor tissue samples were collected within the validated “Tumor bank Ovarian Cancer Network (TOC)”. Results: Seventy-four patients with the primary diagnosis of EOC between 01/1994 and 12/2009 were included in this study. Median age at diagnosis was 56 years. The vast majority of the patients (75.4%) presented an advanced stage disease (FIGO III/IV). Forty-one (55.4%) patients were diagnosed with a serous, 19 (25.6%) a endometrioid and 14 (19%) a mucinous histology. EpCAM was overexpressed in 87.7% of the patients. Serous tumors expressed EpCAM in significantly higher rates than mucinous tumors ( p=0.045). EpCAM overexpression correlated in univariate analysis with a significantly better overall survival. Multivariate analysis including histological subtype identified EpCAM expression of 76-100% to be independently associated with a significantly better overall survival compared to a lower EpCAM expression of ≤50% (p=0.008). Conclusions: EpCAM overexpression in EOC appears to be associated with significantly higher overall survival rates. Larger, prospectively assessed multicenter studies are warranted to further evaluate and confirm these novel findings and possibly establish EpCAM as a potent therapeutic target in EOC.

5080 Background: Advanced ovarian cancer is still connected to high mortality rates due to intraperitoneal tumor cells that survive radical cytoreductive surgery as well as adjuvant chemotherapy. These persistent tumor cells need to be targeted in order to improve survival. Catumaxomab has demonstrated the ability to kill EpCAM-positive intraperitoneal tumor cells of ovarian cancer patients in studies aiming to controll malignant ascites in the recurrent setting. This analysis was conducted to investigate the efficacy of intraperitoneal catumaxomab at the timepoint of primary cytoreductive surgery and postoperative period, prior to standard adjuvant chemotherapy. Methods: Ovarian cancer patients undergoing radical surgery received one intraoperative (10 μg) followed by four subsequent intraperitoneal (i.p.) dosages (10, 20, 50 and 150 μg) of catumaxomab on days 7, 10, 13, and 16, respectively. Because of the single arm design of the study, the patients treated with catumaxomab were compared in a matched pair analysis to consecutive patients with primary ovarian cancer who received standard treatment without catumaxomab in a large center, in order to compare survival. The two main prognostic factors of stage and level of tumorreduction were chosen as matching criteria. Results: Of 58 patients screened, 41 were treated with catumaxomab and available for survival evaluation. Median age was 57 years in the catumaxomab group and 59 years in the matched-pair control group. The most comon histology was the serous subtype with 70.7 % in the catumaxomab and 80.5 % of the patients in the control group. The median for overall survival was reached for the historical consecutive matched-pair control collective, but is not yet reached for the catumaxomab group. However, 3-year survival data were available for both groups and showed survival of 85.4% (35) in the catumaxomab group and 63.4% (26) in the matched-pair control group (p-value: 0.041; HR 2.073) Conclusions: There seems to be a trend to beneficial 3-year survival in the catumaxomab group, suggesting that a phase III trial is warranted.

5087 Background: Indication, technique and extent of lymph-node-dissection (LND) in endometrial cancer (EC) remains controversial and is strongly debated in cancer community. We conducted a national- and at a second step an international-survey evaluating the current status-quo of the surgical and medical management of EC. Methods: A validated 15-item-questionnaire regarding surgical and adjuvant procedures of EC was sent to all major gynaecological cancer societies and study groups worldwide. The questionnaire could also be answered online. Results: In a phase-I-national trial, the questionnaire was validated on basis of 316 German institutions. On the phase-II-international survey a total of 302 questionnaires were answered from 24 countries, mainly from Japan (38.7%), Spain (8.3%), Austria (7%), United-Kingdom (6.3%), Italy (6%), USA (4.3%) and Canada (4%). The vast majority of the participating clinics were academic (62.8%), while 75.2% of them belonged to gynaecology. Only 0.7% of the clinics internationally reported never performing LND in EC. 62.3% of the clinics perform both a pelvic and paraaortic lymph node dissection. 59.1% of the participants performed a systematic lymph node dissection with the intention of both adequate staging and for therapeutic value. 15.05% of the clinics perform LND up to the common-iliac-arteries, 9.03% up to the inferior-mesenteric-artery and 70.6% up to the renal-veins. The most common risk-factors to indicate LND were: high-grading (93%), non-endometrioid-histology (90.1%), lymphovascular-invasion (55.3%), blood-vessel invasion (45.4%) and tumor-diameter >2cm (38.4%). For advanced stage III&IV disease the vast majority (60% and 80%, respectively) of the physicians indicated systemic chemotherapy alone. Conclusions: This study presents the large variety in clinical management of EC worldwide, underlining so the high need of future prospective randomised trials which will establish standard and evidence based treatment guidelines for EC- disease.

5070 Background: Mutations in BRCA1/2 genes contribute to increase risk for breast and ovarian cancer (OC).Hypermethylation of CpG islands in gene promoter regions is considered a frequent mechanism associated with inactivation of tumor suppressor genes which contributes to oncogenic transformation. Studies identified methylation changes in OC pts. Most of the studies analyzed a small cohort and combined different histological subtypes. HGSOC is a special entity of OC associated with poorer OS. Aim of this study was to analyze the clinical impact of BRCA1 promoter gene methylation status in HGSOC. Methods: The cohort comprised 127 HGSOC treated by cytoreduction followed by platinum-based chemotherapy at Department of Gynecology, Charité Medical University Berlin, Germany. Fresh frozen OC tissues provided by TOC were examined to access the tumoral tissue quality. Samples presenting at least 50% of tumour area were included within this study. DNA was extracted followed by sodium bisulfite conversion, and assessment of BRCA1 gene promoter methylation rate was determined by PCR. Statistical analisys has been carried out using SPSS software. Results: In our study, 14.2% of the pts presented a hypermethylation of the BRCA1 promoter gene. The hypermethylation of BRCA1 promoter gene was significantly more frequently encountered in the younger pts (

5071 Background: The international scenery around optimal primary treatment of advanced ovarian cancer (AOC) patients (pts) is currently being discussed, with large discrepancies and heterogeneity still existing between the national guidelines worldwide. Aim of the present study was to evaluate the differences in outcome of AOC-pts after primary (PDS) versus interval-debulking-surgery (IDS) based on a prospectively assessed multicenter data set. Methods: Overall outcome was analyzed from the OVCAD database; a prospective, observational, multicenter project. AOC-pts who underwent surgery in five specialized gynecological cancer centers across three European countries between 02/2005 and 12/2008 were evaluated. Overall (OS) and progression free survival (PFS) were calculated by Kaplan-Meier-curves. Univariate and Cox-regression-analysis were applied. Results: Overall, 256 AOC pts (FIGO-stage III/IV) were evaluated. Fifty pts (19.5%) underwent IDS and 206 pts (80.5%) PDS. Despite the non-randomized setting both groups were well balanced in terms of FIGO-stage, grading, histological subtype and presence of ascites. Different selection criteria were however present for each center. PDS pts presented significantly higher rates of intestinal resection (44.2% vs.24%; p=0.01) and lymphonodectomy compared to IDS ones (72.3%vs.48%; p=0.001), by equivalent complete tumor resection rates (67.5% vs.68%; p=0.82). Platinum response was significantly higher in PDS vs. IDS pts (80.6% vs. 54%; p

e15520 Background: There is an increasing interest in the management of elderly patients in oncology. Treosulfan is effective as oral (p.o.) and intravenous (i.v.) formultion for recurrent ovarian carcinoma. Primary aim of this study was to explore the individual preference and compliance of elderly patients (≥ 65 years) for p.o. or i.v.-treosulfan. Secondary aims were to evaluate the ADL-score, toxicity, response and survival. We present the final analysis of patient’s characteristics and treatment choice, compliance of the treatment and toxicities for 123 included patients. Methods: Patients with platinum sensitive and resistant ovarian cancer had free choice of treosulfan i.v. (7000 mg/m² d1, q28d) or p.o. (600 mg/m² d1-28, q56d) for a maximum of 12 cycles (i.v.) or 12 months (p.o.). Indecisive patients were randomized. Toxicity was evaluated according to the NCI-CTC version 2.0. Results: 123 recruited patients received therapy and were able for analysis (median age 72 years, range 65-87). 102 patients chose i.v. and 17 p.o., 3 were randomized to i.v and 1 to p.o. Median ECOG was 1 (n=0-2), and median number of prior chemotherapy-regimens was 3 (n=1-6). Most common hematological toxicities (grade 3/4) were thrombopenia (18.7%), leukopenia (15.4%) and anemia (3.6%). Most frequent non-hematological toxicities (grade 3/4) were abdominal pain (7.3%), fatigue (5.7%) and constipation (4.9%). The median progressive-free survival was 116 days, and the median overall-survival was 239 days. There was no difference in survival and activities of daily life (ADL) score between i.v. and p.o. treosulfan. Conclusions: Although heavily pre-treated most elderly patients with recurrent ovarian cancer (85%) preferred the i.v.-treosulfan administration following their individual preference. Treosulfan therapy was generally well tolerated. There were no differences in the patient’s characteristics, survival and ADL-scores in both groups.

5034 Background: Optimal surgical cytoreduction and response to platinum (P) based chemotherapy (ChTh) remain the cornerstones of therapeutic management of primary EOC. Aim of this study was to analyze the predictive role of HE4 and CA125 as biomarkers (BM) for clinical outcome in primary EOC pts at diagnosis and during subsequent follow-up. Methods: In the European OVCAD project 275 pts with primary EOC were enrolled. Pts were eligible if radical cytoreductive surgery and P-based ChTh were applied. Plasma collected at first diagnosis and 6 months after 1st line ChTh in P-sensitive pts was analyzed for HE4 and CA125 levels using ELISA and Luminex technique, respectively. Results: Complete cytoreduction with no residual tumor disease (RTD) was obtained in 69.9% pts. HE4 and CA125 expression in plasma at first diagnosis correlated with RTD, p = 0.002 and p=0.002, respectively. The sensitivity (SE) and specificity (SP) of the combinative use of both BM in predicting RTD was 64.8% and 73.5%, respectively. Pts having over-expression of both BM in plasma had a 6.1 greater risk for RTD (p

e15549 Background: Studies indicates clearly a poorer survival for elderly patients with advanced ovarian cancer (AOC). It is unclear whether tumor biological factors or the clinical management is ...

e15507 Background: Aim of the study was to analyse the expression-profile of IGF1, IGFBP3, sICAM1, sVCAM1, MMP2, MMP9, TIMP2, VEGF, VEGFD, VEGFC and VEGFR in patients with FIGO-stage high-risk Ib-I...

e15556 Background: The value of secondary tumordebulking in ROC remains unclear. Retrospective analyses indicate that patient’s survival is positively affected only when complete macroscopic tumor ...