Search

11508 Background: An ISG, GEIS, FSG and PSG randomized trial on 3 cycles of neoadjuvant epirubicine+ifosfamide (EI) versus a histology-tailored (HT) regimen in selected localized high-risk STS showed some superiority of EI in all histologies with the exception of Myxoid Liposarcoma (MLPS) where EI and HT regimens seemed equivalent (J Clin Oncol 2020; 38:2178-86). This MLPS cohort was expanded with the aim to assess the non-inferiority of the HT regimen compared to EI. Methods: This was a multicenter European randomized trial comparing EI versus a HT regimen. Patients (pts) had localized high-risk (grade = 3; size >5 cm; deeply seated) undifferentiated pleomorphic sarcoma, leiomyosarcoma, malignant peripheral nerve sheath tumor, synovial sarcoma or MLPS of extremities or trunk wall. Primary end-point was Disease Free Survival (DFS). Secondary end-point was Overall Survival (OS). The MLPS cohort was expanded after the results of the 3rd interim analysis (Lancet Oncol 2017; 18:812-22) in order to reject the hypothesis that HT regimen trabectedin is associated with a HR of relapse = 1.25 with a non-inferiority design. To this aim, a Bayesian monitoring approach was used until the probability that the true HR is higher than 1.25 was greater than 80% or smaller than 5%. Results: From May 2011 to June 2020, 101 pts affected by high-risk MLPS were randomized, 56 to EI and 45 to HT regimen. The median follow-up was 66 months (IQ range 37-89). Median size was 107 mm (IQ range 84-143), 108 mm (IQ range 86-150) in the EI and 106 mm (IQ range75-135) in the HT arm. The DFS and OS probabilities at 60 months were 0.86 and 0.73 (HR:0.60; 95%CI: 0.24-1.46; log rank p = 0.26 for DFS) and 0.88 and 0.90 (HR:1.20; 95%CI:0.37-3.93; log rank p = 0.77 for OS), in the HT and EI arm, respectively. 5-yr observed and Sarculator-predicted OS were 0.89 (95% CI 0.82-0.97) and 0.80 in all patients (p = 0.020), 0.90 (95% CI 0.81-1.00) and 0.79 in the EI arm (p = 0.049) and 0.88 (95% CI 0.77-1.00) and 0.81 in the HT arm (p = 0.204) respectively. Conclusions: In the expanded cohort of MLPS, the HT neoadjuvant therapy trabectedin was not inferior to EI. While survival in both arms was better than predicted by Sarculator, it is left to understand whether this patient population, who had on average a lower Sarculator-predicted risk of death compared with the rest of the trial population, may benefit from a neoadjuvant therapy. Clinical trial information: NCT01710176.

11559 Background: Over the last decade, the category of undifferentiated round cell sarcomas (URCS), defined by the absence of Ewing sarcoma-associated translocations, has emerged. Aim of this study was to assess prevalence of each entity and outcome after genomic classification. Methods: Ewing sarcoma and other URCS diagnosed between 1920 and 2020 were reviewed. All URCS with available material were analyzed with FISH, RT-qPCR and/or Archer FusionPlex Sarcoma Panel. Demographic and treatment were collected. Survival was analyzed in patients with available follow-up. Results: 1995 cases identified, 20 cases lacked material for further genetic analysis and were excluded. 1975 cases were classified as follows: 1925 Ewing sarcomas (97.47%), 25 CIC-rearranged sarcomas (1.27%), 16 BCOR-CCNB3 rearranged sarcomas (0.81%), 2 EWSR1-NFATC2 sarcoma (0.1%), one each as CIC-LEUTX and FUS-NFATC2 rearranged sarcoma (0.05% each), and 5 as unclassified URCS (0.25%). A different presentation according to tumor type was shown in 43/50 ultra-rare tumors (Table). Forty-one/50 cases had available follow-up: 20/41 patients underwent surgery, 14/41 surgery+radiotherapy, 6 radiotherapy only, and no local treatment for 1 patient. Chemotherapy was administrated to 36/41 patients (Ewing sarcoma drugs in 16/22 CIC-DUX-4 and 8/11 BCOR-CCNB3; osteosarcoma drugs in 2/11 BCOR-CCNB3, and doxorubicin/ifosfamide in 2/22 CIC-DUX4 and 2/5 URCS; not specified in 6 cases). The 3-years overall survival (OS) was 32.7%f for CIC-rearranged sarcomas (75% in localize disease, 7,7% for the advanced disease, p 0.0084), 81.8% for BCOR-CCNB3 sarcomas (87.5% localized, 66.7% advanced; p 0.0734), and 60% for URCS (p 0.057). 1 patient with CIC-LEUTX sarcoma presenting with metastases died 13 months from diagnosis, 1 patient with FUS-NFATC2 and 1 with EWSR1-NFATC2 rearrenged sarcomas were alive without disease at 8 and 5 years from onset. Conclusions: Prevalence of URCS characterized by a combination of morphologic observation ad molecular techniques is provided. The majority of the cases underwent surgery or surgery combined with radiotherapy, and Ewing-like chemotherapy. The survival difference among different entities underscores the need of accurate subclassification of round cell sarcomas. Novel drugs for CIC-DUX-4 sarcomas presenting with metastases are needed. [Table: see text]

11532 Background: Limited are data on high-grade osteosarcoma occurring in the spine. Wide resection is recommended, but it is a difficult and high morbidity procedure in the spine. High-energy particle therapy has been recently used. The goal of this study was to examine treatment and outcome of patients with osteosarcoma in the mobile spine. Methods: Spine high grade osteosarcoma patients who underwent surgery at the Rizzoli Institute between 2009 and 2020 were identified. Treatment, outcome, and prognostic factors in patients treated in a single institution were examined. Results: Characteristics of the 20 patients (8 female; 12 male) included: median age, 39.7 years (range, 14-71 years), 5 (25%) with tumors in the cervical spine, 6 (30%) with tumors in the thoracic spine, and 9 with tumors in the lumbar spine (45%); 14 (70%) patients with localized disease and 6 (30%) with metastatic disease at the time of presentation. Nineteen patients (95%) underwent chemotherapy, the majority were treated with MAP (methotrexate, doxorubicin, cisplatin) regimen. In 12 patients undergoing preoperative chemotherapy (n = 11) or chemotherapy and radiotherapy (n = 1), the median tumor % necrosis was 20 (IQR 20 - 40), with none achieving a good histologic response (> 90%). All patients underwent surgery. Adequate surgical margins were achieved in 5 patients (25%). In patients with positive margins, radiotherapy was administered to 8 (40%) patients. Four patients with positive margins after resection received photon neoadjuvant or adjuvant therapy. Four patients received high-energy particles as adjuvant therapy after planned gross total excision. The median overall survival rate was 10.5 months (IQR 5.8 - 15.0) for patients with metastatic disease and 25.5 months (IQR 8.3 - 46.0) for patients with localized disease (P =.0501). Patients treated with planned intralesional gross total resection followed by adjuvant high-energy particle therapy had a significant higher disease-specific survival than patients with positive margins after resection with or without additional conventional radiotherapy (P =.023). Conclusions: Metastatic disease, is a poor prognostic factor for high grade osteosarcoma of the spine. Post-operative high-energy particle therapy improved overall survival in patients undergoing a planned gross total resection compared to intralesional resection in this series. Chemotherapy induced necrosis was low underscoring the need of more aggressive multidisciplinary approaches for these patients.

e23506 Background: Metastatic osteosarcoma (MOS) with synchronous metastases accounts for 20-25% of all new cases of osteosarcoma. Lungs are the most common site of metastases at presentation (80%) followed by Bone (10%). 5- years Overall Survival (OS) of MOS ranges from 11 to 40%. In a previous study from our institution (Bacci G 2006) on 57 pts < 40 years old (1995-2000) the 2- and 5-year OS were 55% and 18%. Methods: Data of patients with pathologic and radiologic confirmed MOS with adequate follow up were reviewed (EC Approval N 916/2020/Oss/IOR). Time-to-event outcomes were estimate with Kaplan-Meier method and compared between groups with log-rank test and Cox model. Results: From August 2000 to October 2018, 92 patients had a diagnosis of MOS: median age 16.5 yrs (6-73, twelve pts > 40 ), gender rate was M 51/F 41, axial primary tumor in 15 cases, extremity in 77. Lung only metastases were described in 66 (71.7%). In 75/90 cases primary tumor was surgically removed, 43 (46%) cases had at least one surgical metastasectomy. All patients received chemotherapy: preoperative only in 6 cases, postoperative in 6, and pre and postoperative in 66 patients. The 1st line chemotherapy was a combination of drugs: Adriamycin in 91/92 pts, Cisplatin in 89/92, Ifosfamide in 88/92, Methotrexate in 83/92; 59 patients received a 2nd line chemotherapy, 34 pts received a 3rd line; most employed regimen were Gemcitabine-Docetaxel, Ifosfamide 15 gr/m2, Cyclophosphamyde-Etoposide, TKI (Pazopanib, Sorafenib), and a few received experimental drugs. Complete remission (CR) was obtained in 26/92 (28%), in 19 cases after surgical metastasectomy.In 30 pts the information of PGP (P- glycoprotein) was available; patients with positive PGP (19/30) had a worst overall survival as compared to those PGP negative (P = 0.038). Of those in Complete Remission 14/26 relapsed. At December 2020 with a median follow-up of 95 ms (IQR 34-159): 65/92 (70%) died , 12 are alive and free from disease , 6 are alive with disease, 9 were lost . The 2-yrs OS for all 92 pts from diagnosis was 66% (95%CI 55-75) and 5-yrs OS was 26% (95% CI 16-37). From the end of treatment, for those who reached a CR the 5-year OS was 57% vs 9% for those who did not (P < 0.001). At univariate analysis, primary tumor site (2-y OS 48% axial vs 72% extremity, P < 0.001), site of metastases (2-y OS 74% only lung vs 48% other, P = 0.004) and number of lung nodules (P = 0.007), were significantly associated to OS. At multivariate analysis, only site of metastases (other vs. only lung HR = 2.26, 95%CI: 1.21-4.22) and number of lung nodules (≥10 nodules vs ≤3 HR = 2.44, 95%CI: 1.24-4.81) were confirmed as significant for OS. Conclusions: Compared to our previous report from 20 yrs ago, 2-years and 5 yrs OS of MOS has improved but it remain unsatisfactory (66% vs 55% and 26% vs 18%).

e22010 Background: Osteosarcoma is a malignant primitive bone tumor whose prognosis is not changed since 4 decades, after the introduction of neoadjuvant chemotherapy with Methotrexate, Cisplatin, Doxorubicine and Ifosfamide. Histologic response to preoperative chemotherapy is a significant prognostic factor. Huvos I (necrosis ≤ 50%) has worst prognosis . Previous studies reported a 3 years EFS of this Huvos I patients around 25% (Tsuda Y,2020). In order to evaluate if survival has changed in recent years in this unfavourable prognostic group we evaluated the outcome of osteosarcoma patients with Huvos I. Methods: from our Pathology archieves we retrieved all cases of localized osteosarcoma treated at Rizzoli with neoadjuvant chemotherapy who reported an histologic necrosis below or equal to 50% (Huvos I grade) after preoperative chemotherapy MAP (Ethical C. Approval 917/2020/Oss/IOR). Results: from 2003 to 2019 we had 70 cases of localized osteosarcoma with Huvos I necrosis after neoadjuvant chemotherapy ( MAP in 66 and MAPI in 4) evaluable. Median age 21,5 (3-70); M:F = 44:26. 10/70 had axial localization vs extremity(60), subhistotype distribution:46 osteoblastic,11 chondroblastic, 7 fibroblastic, 5 teleangectatic, one not classified. In 24 cases PgP was available(14 PgP positive). With a median follow up of 86.7 ms (IQR 41-136) 43/70 had already relapsed. The median EFS was 25 ms (95% CI 15-42) and the 3 yrs EFS was 40.6% (95% CI 29-52). The 3 yrs overall survival was 80% (95%CI 68-88) and median OS was not reached. Axial tumor site was associated with significant inferior EFS (P = .004). Conclusions: these data confirm the poor prognosis of patients with necrosis ≤50% and the need of new drugs to improve their survival in this sub-group.

11511 Background: Pazopanib (P) was assessed prospectively in a phase 2 study in SFT resulting in a longer progression free survival (PFS) and overall survival (OS) compared to historical controls treated with chemotherapy. No statistical correlation was found between angiogenic factors and P in its pivotal phase III sarcoma trial. In the last two years, a soaring interest on the prognostic and predictive value of inflammatory indexes such as neutrophil/lymphocyte (NLR) and platelet/lymphocyte (PLR) is emerging in sarcomas. A retrospective analysis of inflammatory indexes of patients who entered the GEIS-32 (NCT02066285) trial was performed. In that trial advanced SFT patients were treated with P from front-line. Methods: All eligible patients who entered in the typical- and malignant-SFT cohort of the GEIS-32 trial were included in this analysis. To determine NLR and PLR, baseline values of platelets (10e9/L), neutrophils (10e9/L) and lymphocytes (10e9/L) were obtained from complete blood count tests. Additionally, RDW (standardized as 1 = upper value of normal range) values at baseline were also determined. The impact of NLR, PLR and RDW on OS, PFS and Choi response were analyzed by univariate and multivariate analysis. MAXSTAT was used to determine optimal cut-off points for overall survival. Metastasis free interval (MFI), mitotic count and ECOG were also analyzed, among others. Results: Sixty-seven out of 70 enrolled SFT patients, median age 63-y and 57% female, were considered for this analysis. The median follow-up from treatment initiation was 20.0 months. High standardized RDW value at baseline (cut-off 1.03) was significantly associated with worse OS [10.7 months (95% CI 3.8-17.5) vs 49.8 months (95% CI 9.4-90.2), p < 0.001] and worse PFS [8.8 months (95% CI 0.9-7.0) vs 9.8 months (7.4-12.3), p = 0.001]. High PLR (cut-off 242) significantly correlated with worse OS [10.7 months (95% CI 5.2-16.2) vs 49.8 months (95% CI 14.6-85.0), p < 0.001] and worse PFS [4.5 months (95% CI 2.0-7.0) vs 10.1 months (95% CI 6.3-13.9), p = 0.005], and high NLR (cut-off 3.78) was significantly associated with worse OS [11.7 months (95% CI 3.5-19.8) vs NA, p < 0.001] and worse PFS [4.5 months (95% CI 1.9-7.0) vs 10.8 months (95% CI 8.7-12.9), p = 0.010]. Independent variables in multivariate analysis were NLR, RDW, MFI and mitosis for PFS; while RDW and ECOG for OS (see table). Further, NLR and mitosis were independent factors for Choi progressive disease (as best response). Conclusions: High NLR and RDW values were independent biomarkers of worse outcome in advanced SFT patients treated with pazopanib.[Table: see text]

11501 Background: The role of dose intensification of chemotherapy in Ewing sarcoma (ES) is under evaluation in prospective trials. This is a controlled, randomized phase III study evaluating the impact on event-free survival (EFS) of two arms at different intensity of induction therapy in localized ES at onset. Methods: Newly diagnosed localized ES patients aged 2-40 were eligible. They were randomized to receive 4-courses induction therapy - 1 every 21 days - either with a standard arm (arm A) as per ISG/SSGIII protocol (Ferrari S, et at, Ann Oncol. 2011;22(5):1221) or with an intense arm B, consisting of vincristine 1,5mg/sqm+ doxorubicin 80mg/sqm+ifosfamide 9g/sqm for each course. After induction, patients underwent surgery and/or radiotherapy,followed by an adaptive treatment. Good responders received standard courses chemotherapy: arm A pts received 9 courses, while arm B pts received 5 courses. Poor responders in both arms received 4 courses followed by high-dose busulfan/melphalan+autologous stem cell rescue. The primary outcome measure was EFS for the 2 arms in the intention-to-treat population. Kaplan-Meier curves compared with log-rank test and Cox model were performed to assess differences between study arms. A secondary outcome was toxicity differences, assessed by means of the Fisher’s exact test. Initial sample size was 230 pts, type I error rate 5%, power 80%. Results: Between 2009 and 2019, 234 patients were randomized (arm A-115; arm B-119). M:F ratio was 1.8; median age 14 years (range 2-40); tumour site extremity in 55%, axial/pelvis in 45%; tumour volume < 200ml in 31% and ≥200ml in 69%. A good response was obtained in 56% in arm A and 60% in arm B. Median follow-up was 68 months. EFS was not significantly different between arms; HR: 0.85; 95% CI: 0,51-1,41, 5-year EFS (95% CI) was 73% (64-82%) in arm A and 75% (67-83%) in arm B ( p = 0.526). Good responders in arm A and in arm B and poor responders in arm B had comparable results: 5-year EFS (95% CI) was 80% (71-91%), 77% (67-88%), and 72% (59-86%), respectively, while poor responders in arm A showed a worse, not statistically significant (p = 0.164) performance (63%; 50-78%). Subgroup analyses showed similar outcome for age, tumour site and volume in both arms. Hematological, gastrointestinal, and cardiovascular grade ≥3 toxicities were more pronounced in arm B (p < 0.05). Conclusions: Intense induction therapy with arm B did not improve 5-year EFS when compared with the standard arm A. The higher toxicity observed in arm B than in arm A was counterbalanced, in good responders, by a similar outcome with a shorter treatment plan. For poor responders, with almost 30 patients per arm event-free and with < 48-month FUP, better 5-year EFS in arm B than in arm A was observed but needs further observation. Clinical trial information: NCT02063022.

11527 Background: Overexpression of ABCB1/P-glycoprotein (Pgp) predicts poor outcome in retrospective osteosarcoma series.Two prospective trials with Pgp expression and post-induction histologic response as stratification factors were activated in Italy (ISG/OS-2) and Spain (GEIS-33). Methods: Patients ≤ 40 years with extremity non-metastatic high-grade osteosarcoma were eligible. Analysisi of Pgp expression from diagnostic biopsy was centralized. Preoperatively, all patients received methotrexate, adriamycin, cisplatinum (MAP). Surgery was performed at week 8. All patients received a dose of adriamycin following surgery. In case of Pgp overexpression (Pgp+), mifamurtide (2 mg/m2 twice/week for 3 months then weekly for 6 months) was added after surgery, with 4 consecutive cycles of ifosfamide 3 gr/m2/day, day 1-5 (HDIFO) in case of poor histologic response (necrosis < 90%) to MAP. Patients without overexpression of Pgp (Pgp-) received MAP postoperatively, regardless the pathological response. From March 2013, an amendment increased high dose methotrexate cumulative dose from 60 g/m2 (5 cycles) to 120 mg/m2 (10 cycles). The post-amendment regimen was adopted in the observational prospective study by GEIS. Here we present the merged analysis of ISG/OS-2 patients treated post-amendment and GEIS-33. Results: From March 2013 to April 2018, 274 patients were included. Median age was 14 years (range 4-38), male/female: 163/111; 90 were Pgp-, 164 were Pgp+, 20 not evaluable. With a median follow-up of 48 months (1.3-78.5 months), the 3-year EFS and OS were 71.9% (95%CI 66-76.9) and 88% (95%CI: 83.2-91.5) respectively, with no inferior survival for Pgp positive patients and improved survival for good responders (Table). Conclusions: In this prospective uncontrolled study with a risk-adapted strategy for non-metastatic osteosarcoma, survival is superior to that of all ISG/GEIS previous series. The 3-year EFS of 71.9% compares favorably with other reports. Pgp+ patients performed well in this study, in which mifamurtide and HDIFO were added after a poor response to MAP. Clinical trial information: NCT01459484; NCT04383288. [Table: see text]

11572 Background: The characteristics of immune contexture and its prognostic and predictive value in STS is left to be understood. This planned analysis of the ISG-STS-1001 trial, which compared neoadjuvant anthracycline + ifosfamide (AI) vs a histology-tailored (HT) chemotherapy (ChT), was aimed at characterizing the immune contexture after neoadjuvant ChT and investigating any association with the risk of recurrence. Methods: Patients registered in the ISG-STS-1001 study (ID: NCT01710176) were included if they had tumor tissue available for Tissue MicroArray (TMA), which was performed in the area of the surgical specimen with the highest lymphocyte infiltrate. The following markers were analyzed with IHC and measured quantitatively: CD3, CD8, PD1, GranzymeB, Foxp3, CD20, CD163, and PDL1. The T-Distributed Stochastic Neighboring Entities (t-SNE) analysis was used to account for the co-expression of IHC markers in each tumor. The prognostic value of each marker for disease-free survival (DFS) was assessed. Results: This analysis was conducted in 256 of 435 study patients. AI and HT neoadjuvant ChT did not result in any different distribution of immune contexture. Conversely, differences were observed between ‘complex’ karyotype STS (ck-STS: LMS, MPNST, UPS, MFS, pleomorphic liposarcoma, and pleomorphic rhabdomyosarcoma) and ‘simple’ karyotype STS (sk-STS: MLPS and SS). Ck-STS were enriched in both CD3+ and CD8+ cells compared to sk-STS. These cells displayed an heterogeneous distribution and were dispersed inside the tumor nest, keeping direct contact with sarcoma cells. Ck-STS also displayed an enrichment in Granzyme B+, and CD163+ cells. PDL1+ cells were occasionally identified and were more frequent in ck-STS, suggesting an immune-related expression. Most STS were negative for CD20+ cells, however, when present these cells were highly represented and organized in tertiary lymphoid-like structure. The t-SNE generated plot clustered tumors, the ‘cold’ mainly including sk-STS and the ‘hot’ mainly composed by ck-STS. In the ‘hot’ group, a cluster of tumors displayed an immune infiltrate enriched with a high number of CD3, CD8, GranzymeB, PD-1, and PDL-1+ cells. When the prognostic value of the immune markers was investigated, the presence of CD20+ cells was the only independent prognostic factor for DFS (HR=0.68, 95%CI 0.52-0.91) in a histology-stratified estimate adjusting for tumor size in cm (HR=1.07, 95%CI 1.03-1.12) and patient age (HR=1.0, 95%CI 0.97-1.02). Conclusions: Immune contexture differed across sarcoma histologies after neoadjuvant ChT, rather than across the two study arms, with ck-STS being marked by a rich immune contexture. While a CD20+ infiltrate was found to be an independent prognostic factor for a better outcome, further analyses are in progress on the prognosis of patients with the richest immune contexture. Clinical trial information: NCT01710176.

11543 Background: The ceiling-drug effect seen for most active drugs in STS could be related, partially, to multidrug resistance mechanisms (MDRM). We previously reported the independent prognostic role for RFS and OS of MRP1 in high-risk localized STS of limbs and trunk-wall treated with epirubicin and ifosfamide (Mol Cancer Ther.2014 13(1):249-59). A translational study was carried out within the randomized phase III trial of epirubicin plus ifosfamide vs histotype-tailored neoadjuvant chemotherapy (NCT01710176), to investigate MRP1 prognostic value using the trial population as validation set. Methods: Patients enrolled in the trial were invited to participate, through the informed consent, to this analysis. IHC used QCRL-1 (Santa Cruz biotechnology) MRP1 monoclonal antibody. TMAs were built on the highest-grade area of each tumor, being the procedure blinded for clinical data. MRP1 expression was grouped as low (≤ 25% positive cells) vs high ( > 25% positive cells) expression. For data analysis, patients were grouped as A) epirubicin plus ifosfamide control arm and B) histotype-tailored experimental arm. Drugs used in group B were: gemcitabine-docetaxel (UPS), gemcitabine-DTIC (LMS), trabectedin (High-grade (HG) myxoid LPS), ifosfamide-etoposide (MPNST) and high-dose ifosfamide (SS). Prognostic value of MRP1’s extension was analyzed using Cox’s proportional hazard regression. A p-value < 0.05 was considered statistically significant. Results: 175 patients were analyzed (median age 49; males 61%) with median follow-up of 4.66 y. Group A (n = 88) included HG-myxoid LPS (27%), SS (25%), UPS (24%), LMS (12%) MPNST (10%) and others (2%); group B (n = 87) included UPS (38%), SS (24%), HG-myxoid LPS (20%), LMS (10%) and MPNST (8%). MRP1 high extension was distributed as follows: 48% (A) and 57% (B). High MRP-1 expression showed significantly worse prognosis for disease-free survival (DFS) (HR 2.71 (1.31-5.62) p = 0.007) and a trend towards worse OS (HR = 2.75 (0.97-7.81) p = 0.058) in group A. No correlation was seen between MRP-1 expression and DFS (p = 0.384) or OS (p = 0.665), in group B. Conclusions: MRP1 overexpression was related to significant worse prognosis in 2 prospective randomized series of high-risk, localized, STS treated with neoadjuvant epirubicin and ifosfamide. These agents are both substrate of MRP1; this could add rationale for a possible predictive role, as MDRM, for the two most active drugs in STS. A trial combining epirubicin, ifosfamide and MRP1 inhibitor is currently under design.

e23564 Background: Activity of temozolomide (TEM) and irinotecan (IRI) in recurrent Ewing sarcoma (EWS) was demonstrated. Few data are available on TEMIRI use upfront. Biological predictive factors are lacking. Methods: This multi-institutional retrospective study (NCT03542097) included 59 patients with EWS. 8 patients with very high risk (HR) EWS (multivisceral ± bone marrow) received TEMIRI (TEM 100 mg/m2/day oral, and IRI 40 mg/m2/day intravenous, days 1-5, every 21 days) upfront, 51 patients after relapse (28% in 1st line, 72% ≥ 2nd line). Overall response rate (ORR: CR+PR), SD, and PD, progression-free (PFS) and overall survival (OS) were assessed. The relationship between pre-treatment expression of MGMT and ABCG2 (when FFPE tissue available) with ORR, PFS and OS was evaluated. Results: Median age was 27 years (range 4-62 years): 47 patients (80%) were adults (≥18 years), 35 (61%) had ECOG 0 and 42 (71%) presented with multivisceral disease (+ bone marrow in 5). MGMT was positive in 16/30 (53%), ABCG2 in 4/33 (12%). ORR for upfront TEMIRI (n = 8) was 50% (CR + PR = 1 + 3), with SD 50%, while in recurrent EWS (n = 49, 2 patients no measurable by RECIST) ORR was 31% (CR + PR = 4 + 11), SD 38%, and PD 31%. A better ORR was observed in adult (p = 0.008) & ECOG 0 (p = 0.001) patients; MGMT & ABCG2 expression did not influence ORR. 6-mos PFS was 87% after TEMIRI upfront, 43% at recurrence (p = 0.06), and 65% vs 28% (p = 0.02) for ECOG 0 vs ECOG 1-2, respectively. 6-mos PFS was 62% in MGMT+ vs 33% in MGMT- (p = 0.4) and 75% in ABCG2+ vs 50% in ABCG2- (p = 0.7). Median time to progression (TTP) with upfront TEMIRI was 9 months (range 5-28 months), with 1 patient with ongoing CR at 56 months; median TTP at relapse was 3 months (1-29 months). MGMT and ABCG2 expression did not influence 1-yr OS, whereas ECOG (0 vs 1-2) did (88% vs 31% p < 0.001). Grade 3-4 diarrhea, neutropenia, and thrombocytopenia incidence was < 5%, 1 patient with recurrent kidney EWS died of acute liver failure. Conclusions: TEMIRI showed promising activity in a very unfavorable cohort of EWS patients, with manageable toxicity. Performance status correlated with 6-month PFS & OS whereas MGMT & ABCG2 did not. Clinical trial information: NCT03542097 .

11558 Background: We investigated the prognostic relevance of % change in LTD in patients (pts) with localized high-risk STS treated with neoadjuvant chemotherapy in a phase 3 randomized trial (NCT01710176), aimed at comparing 3 cycles of a neoadjuvant histology-tailored (HT) over 3 cycles of standard anthracycline + ifosfamide chemotherapy (S). Methods: Pts with localized high-risk STS of extremities or trunk wall, and a diagnosis of myxoid liposarcoma, leiomyosarcoma, synovial sarcoma, malignant peripheral nerve sheath tumors, undifferentiated pleomorphic sarcoma were randomly assigned to receive 3 cycles of S or HT. Pts affected by myxofibrosarcoma, pleomorphic liposarcoma, pleomorphic rhabomyosarcoma unclassified spindle cell sarcoma were prospectively registered and treated by S. Change of LTD was assessed comparing baseline dimension with that measured after 3 cycles of S or HT, before surgery. Only pts treated with neodjuvant chemotherapy alone were selected for the analysis. We first investigated Overall Survival (OS) from surgery of the groups identified by “any % reduction”, “no-change” or “increase” in LTD by Kaplan-Meier estimates and log-rank tests. Then we searched for cutoffs able to separate prognosis among pts with a LTD reduction applying the change-point method proposed by Contal - O’Quigley. Results: Of 325 pts who entered the study and evaluable for response, 181 received neoadjuvant chemotherapy alone (92 S and 89 HT group respectively) and were analyzed, while 144 received concurrent chemo-radiotherapy and were excluded. In the whole population, % changes in LTD were significantly associated (log rank p = 0.032) to OS. “Any % reduction in LTD (101/181pts) displayed a better prognosis compared to “no-change” (28/181 pts) or “any % increase” (52/181). The change-point analysis was applied to all, S and HT groups separately; a cutoff of = / > 18.75% decrease in LTD was the optimal predictor of outcome for the S group (p = 0.031), while no size cut-off could be identified for the HT group. Conclusions: In our study, % change in LTD of pts treated with neoadjuvant chemotherapy for localized high-risk STS correlated with OS. However, a % decrease in LTD cut-off able to predict the best outcome could be identified only for pts treated in the S group, while no differences in outcome were found by any % LTD change in the HT group. Interestingly, the LTD cut-off identified in the S group was lower than the one selected to define a response by RECIST ( = / > 18.75% decrease in LTD vs = / > 30%). Clinical trial information: NCT01710176 .

e23531 Background: The Complexity INdex in SARComas (CINSARC) is a gene expression signature related to mitosis and chromosome integrity that stratifies risk for recurrence of soft tissue sarcoma (STS) patients. The aim of this study was to validate the prognostic value of CINSARC in patients enrolled in a randomised trial that compared histotype-tailored neoadjuvant chemotherapy with standard chemotherapy in patients with high-risk STS (ISG-STS 1001). Methods: CINSARC is 67-gene-expression-based signature that has been previously tested in retrospective series. The ISG-STS 1001 was a phase 3 RCT comparing histotype-tailored and anthracycline-based chemotherapy in localised, high-risk STS of the extremities or trunk wall, with one of five histological STS subtypes: high-grade myxoid liposarcoma, leiomyosarcoma, synovial sarcoma, malignant peripheral nerve sheath tumour, and undifferentiated pleomorphic sarcoma. Outcome variables were disease-free survival (DFS), overall survival (OS) and tumour response according to RECIST v1.1. Results: CINSARC was assessed in pre-treatment biopsies of 87 in 435 patients participating in the study. Thirty and 57 patients segregated in the lower (C1) and higher (C2) CINSARC risk group, respectively. Incidence of local recurrences (LR) and distant metastasis (DM) did not differ between C1 and C2 CINSARC groups [2 (6.6%) and 11 (19.3%) patients had a LR, respectively, and 10 (33.3%) and 14 (24.5%) patients had DM, respectively, P = 0.800]. Consistently, we did not observe statistically significant differences for DFS and OS between patients in the CINSARC C1 and C2 groups (log-rank test, P = 0.522 and P = 0.480, respectively). RECIST tumour response was analysed in a subset of patients (N = 39), showing that a RECIST SD was more likely in C1 (N = 12/14, 85.6%) compared to C2 (N = 18/25, 72%) group, while both RECIST PD and PR were more commonly detected in C2 [3/25 (12%) and 4/25 (16%), respectively] compared to C1 [0/14 (0%) and 1/14 (7.1%), respectively] group. Conclusions: In high-risk STS patients treated with preoperative chemotherapy within a RCT, CINSARC did not correlate with different DFS and OS. While this may well be due to a failure of this gene signature in this patient population, an alternative hypothesis is that preoperative chemotherapy may improve the prognosis of higher-risk patients. Clinical trial information: NCT01710176 .

e18236 Background: The double-blind, randomized, placebo-controlled phase 3 ENLIVEN study in TGCT demonstrated a significant tumor response at week 25 by RECIST for pexidartinib (39% vs. 0% for placebo) and improvement in joint function and symptoms. The aim of this analysis was to identify a threshold score for responder definitions for PROMIS-PF scale and Worst Stiffness Numerical Rating Scale (WS-NRS) and compared responder rates for pexidartinib versus placebo. Methods: Anchor- and distribution-based estimates were calculated, and cumulative distribution function (CDF) plots were generated to derive responder definition threshold estimates. Anchor- and distribution-based results and CDFs were evaluated through triangulation, following FDA PRO Guidance, to determine a single responder definition threshold (i.e., meaningful change) for each instrument. The proportion of responders at Week 25 between treatments was compared with Fisher’s Exact Test (2-sided). Results: 120 patients were randomized to pexidartinib (n = 61) and placebo (n = 59) and assessed through 25 weeks of treatment. Anchor-based analysis showed one-level improvement on the patient global rating of PF item was associated with a mean change of 4.0 on PROMIS-PF. Distribution-based estimates (0.5 SD and 1 SEM) for PROMIS-PF were 2.8 and 2.5, respectively. For WS-NRS, a response of “A little improved” by patients on the perception of stiffness item was associated with a mean change of 1.1. The distribution-based estimates for the WS-NRS item were 0.9 and 0.5, respectively. This resulted in the following responder definition thresholds: ≥3 points for PROMIS-PF and ≥1 for WS-NRS improvement. A greater proportion of pexidartinib as compared to placebo patients were responders by PROMIS-PF (30% vs. 5%, p < 0.001) and WS-NRS (39% vs. 19%, p = 0.02) at week 25, respectively. Conclusions: Triangulation yielded responder definitions of ≥3 points for PROMIS-PF and ≥1 for WS-NRS. With these definitions, a greater proportion of patients treated with pexidartinib compared to placebo had meaningful improvement in physical function and stiffness.

11042 Background: TGCT is a rare, locally aggressive neoplasm of the joint/tendon sheath linked to colony-stimulating factor 1 (CSF1) overexpression. Pexidartinib (pex), a selective inhibitor of CSF1 receptor, KIT, and FLT3-ITD, had a compelling tumor response rate in the TGCT cohort of a phase 1 study (NCT01004861) and significant tumor response vs placebo by RECIST v1.1 (39% vs 0%, P< 0.0001) and tumor volume score (TVS) (56% vs 0%, P< 0.0001) in the randomized, 2-part, crossover phase 3 ENLIVEN study (NCT02371369). Updated efficacy and safety with longer treatment are reported. Methods: Patients (pts) were ≥18 y with TGCT that was inoperable or for which surgery would likely be associated with worsening functional limitation or severe morbidity. Best overall response (complete or partial [CR/PR]) and duration of response (DOR) by RECIST and TVS were assessed by independent central review. Data cutoff was Jan 31, 2018, 16-67 mo after pts’ first dose. Results: In both studies 130 pts received pex, 61 ongoing at data cutoff. Median treatment duration was 17 mo (1, 60+). CR/PR rates were high and consistent and, together with DOR, improved with prolongation of treatment (Table). Most frequent adverse events were hair color change (75%), fatigue (60%), nausea (45%), arthralgia (38%), AST increase (30%), and diarrhea (30%). In ENLIVEN part 1, 3 of 61 (5%) pts had reversible ALT and AST ≥3 × ULN with TBil and ALP ≥2 × ULN; all started in the first 8 weeks of treatment, and no new cases emerged with continuation of treatment. Conclusions: Tumor response rate increased with continuation of pex treatment. The safety profile remained similar, with no new mixed or cholestatic hepatotoxicity. Clinical trial information: NCT01004861 and NCT02371369. [Table: see text]

11046 Background: Angiosarcoma of bone (B-AS) is an exceedingly rare malignant tumor of vascular origin. The aim of this EMSOS retrospective study is to report on natural history, type of treatment and prognosis of B-AS. Methods: Patient data were collected according to the national rules for observational studies through electronic dataset available on EMSOS WEBsite. Results: 80 patients (51 male and 29 female, median age 54 years, range 17 to 92 years, 56% with localized disease, 44% metastatic) from 8 EMSOS Centers were collected. Surgery of the primary tumor was performed in 76% of patients (amputation in 30%), with intralesional margins in 26%. A surgical complete remission status (SCR) was achieved in 47% of the patients. Radiotherapy (RT) was delivered in 41% of the patients (in 15 patients as definitive local treatment), chemotherapy (CT) in 47% (56% in metastatic and 41% in localized cases). With a median follow-up of 31 months (range 40 to 309 months), 68% of patients died, 16% were disease-free, 12% were alive with disease and 4% dead for other causes. The 5-year overall survival (OS) was 27% (95%CI 16-30): 41% (95%CI 25-56) for localized patients (45% SCR, 17% no SCR, p = 0.03) and 8% (95%CI 0-20) for metastatic patients (p = 0.002). Among metastatic patients, 29/35 have died, with a median time to death of 6 months (1-45), while 6 patients were alive with a median follow-up of 22 months (8-106). Improved survival was observed for male patients (30% vs 8%, p = 0.04), while type of treatment (surgery, chemotherapy, radiation), pattern of metastases and age did not affect outcome. In 18 metastatic patients who underwent chemotherapy partial respo nse (PR) was documented in 1/3 patients after paclitaxel (RR 33%), and stable disease (SD) in 5 patients (2/2 with gemcitabine, 2/8 with doxorubicin/ifosfamide, and 1/4 with osteosarcoma-like chemotherapy). In localized patients, the 5-year OS was significantly better for age ≤ 50 years (68% vs 34%, p = 0.02) and extremity and central tumor as compared with pelvis/sacral location (50% vs 58% vs 0%, p = 0.008), with improved disease-free survival (DFS) after chemotherapy (49% vs 33%, p = 0.04). Conclusions: Metastatic B-AS is a fatal disease and inclusion in experimental trials is warranted. In localized patients, a better probability of survival is expected in younger and surgically treated patients. The use of chemotherapy was associated with improved DFS.

11030 Background: TMZ+IRI has significant activity in metastatic ES. Epigenetic silencing of the MGMT DNA gene by promoter methylation has been associated with response to TMZ in glioblastoma. Our aim was to assess if MGMT methylation 1) has a role in ES progression and 2) is predictive of response to TMZ. Methods: 1) In 10 ES cell lines presence of MGMT gene (Real-time PCR), methylation of its promoter (methylation-specific PCR) and protein expression (western blot) were assessed. MGMT protein (IHC) and methylation of its promoter was searched in 97 ES pts samples (74 localized; 23 metastatic). 2) In metastatic ES pts treated with TMZ+IRI, with pre-treatment FFPE tissue and measurable disease, the relation between RECIST response, PFS and MGMT expression (IHC) was assessed. Results: 1) The expression of MGMT gene and its protein was detected and concordant (p = 0.02) in all ES cell lines evaluated; methylation was a rare event. In ES tissue samples the methylation of the MGMT gene was found at a low intensity as compared with the unmethylated gene, but the protein expression was relatively low: 36% in localized, 65% in metastatic pts (p 0.03). 2) 24 pts (median age 19 years, range 3-50 years; F/M: 7/17) treated with TMZ + IRI from 2010 to 2015 were identified. Line of treatment: 8 patients were in 1st line; 16 ≥ 2nd line. Median n of cycles was 6 (range: 2-31). Pattern of metastases: 16 multiple sites, 4 lungs, 3 multiple sites + bone marrow, 1 bone. MGMT was positive in 63% of cases. ORR: 16.5% (1 CR , 3 PR); SD: 50% (13 pts); PD: 33.5% (7 pts). According to MGMT expression the ORR was 11% in negative and 20% in MGMT positive patients (p = 0.8). 6-mos PFS rate was 59% (38-80 %IC), no differece according to MGMT expression (pos 61% vs neg 56%, p 0.7). Conclusions: Whereas in cell lines the MGMT gene and its protein expression is a generalized event, in tissue samples MGMT protein is present in a minority of localized pts, and might be associate with tumor progression. Methylation of MGMT gene does not seem responsible for its regulation in ES, and post-transcriptional mechanisms are more likely to be involved. The presence of MGMT protein does not predict the response to TMZ + IRI in this small series.

11062 Background: Extraskeletal myxoid chondrosarcoma (EMC) is an exceedingly rare sarcoma, marked by a specific translocation involving the gene NR4A3 that can be rearranged with different partners. Preliminary retrospective data suggest that sunitinib is active, but no formal prospective studies are available. We report on a multicentric European prospective, investigator-driven, Phase 2 study on pazopanib (P) in NR4A3+ advanced EMC patients (pts), carried out by the Spanish, Italian and French Sarcoma groups. Methods: From June 2014 to November 2016, 24 advanced EMC pts entered this study (median age: 64 yrs - disease extent: metastatic 77%, locally advanced 23% - prior medical treatment: 18 (86%) naive; 2 (9%) 1 line, 1 (5%) > 1 line). Path diagnosis and NR4A3 rearrangement (FISH and/or real-time PCR analysis) were centrally confirmed. Pts received P 800 mg/day (relative dose intensity = 0,82%, 658 mg/day), until progression or toxicity. The primary study end-point was response rate as per RECIST 1.1. Secondary end-points were overall survival, progression-free survival (PFS), clinical benefit rate (CBR) (RECIST CR+PR+SD≥6mos). An exploratory evaluation of the correlation between the rearrangement subtype and the outcome is ongoing. Results: 20/24 pts were evaluable for response (1 early death; 3 too early). One patient (5%) had a partial response, 17 (75%) stable disease, 2 (10%) progression as their best RECIST responses. At the time of this analysis, 12 pts were still under treatment, while 12 interrupted P (10 progression, 1 toxicity, 1 other). At a 13-month median follow-up, the median PFS was 13 months (range 1.6-25.1), with 29% pts progression-free at 18 months and a 65% CBR. Median OS was not reached. Conclusions: This Phase 2 study is formally negative since the target of at least 3/21 RECIST responses was not reached. However, looking at PFS, P was associated with a prolonged disease stabilization in a significant proportion of pts. This suggests to further explore the use of P in EMC. Clinical trial information: NCT02066285.

11003 Background: SFT is a rare soft tissue tumor. In advanced SFT chemotherapy has only limited activity. With the rationale of a rich vascular network & VEGF (tumor cells and endothelium) and VEGFR1/2 (endothelial cells) expression in SFT, we designed an international, single-arm phase II trial to test pazopanib (P) in advanced SFT. Clinical and preclinical evidence suggesting that antiangiogenics was less effective in more aggressive compared with less aggressive SFT (Stacchiotti et al), led us to conduct the trial on two different cohorts: typical and malignant (M)/dedifferentiated (DD) SFT. Here we present the outcome of the latter cohort. Methods: Most relevant inclusion criteria were: unresectable or metastatic, M/DD SFT confirmed by central pathologic review with evidence of STAT6 (IHC and /or FISH or RT-PCR), ≥ 18 years, ECOG 0-2, progressive and measurable disease. Main endpoint was response rate (RR) according Choi criteria. Central radiological assessment was mandatory. P was administered at 800 mg/d continuously until progression or toxicity. Results: From June 2014 to November 2016, 34 patients (pts) were enrolled with a median age of 61 y (23-87). Median tumor size and mitosis at diagnosis were 77 mm and 8x10 HPF. Most relevant grade 3-4 toxicity were neutropenia (9%) and hypertension (12%). At the time of the present analysis, 31 pts are evaluable for response. RR according to Choi and RECIST were: PR 16 (52%), SD 7 (22%), PD 8 (26%) and PR 1 (3%), SD 19 (61%), PD 11 (35%) respectively. With a median follow-up of 15 months, the median PFS was 5.53 months (4.24-6.82), while 72% survived at 18 months. Size > 5 cm, mitosis > 8 and DD subtype showed significantly worse PFS. The 18-month OS was 90% for those with SD and PR and 25% for PD according to Choi (p < 0.001), while 94% for SD and PR and 45% for PD according RECIST (p = 0.002). In multivariate analysis, only Choi was an independent prognostic factor for OS with PD showing a HR of 11.9 (2.3-63.1), p = 0.003 for the risk of death. Conclusions: Pazopanib showed activity in malignant SFT. Choi criteria exhibited a more accurate assessment of response than RECIST. Clinical trial information: NCT02066285.

11502Background: TGCT is a rare neoplasm of the joint/tendon sheath associated with colony-stimulating factor 1 (CSF-1) overexpression. No approved systemic therapy is available. Pexidartinib (Pex), a selective inhibitor of CSF-1 receptor, KIT, and FLT3-ITD, showed promising Phase 1 activity in TGCT. Two severe liver toxicity cases (1 required liver transplant, 1 associated with death) have been observed with Pex across the non-TGCT development program. Methods: Patients (pts) ≥18 yr with symptomatic TGCT, for whom surgery would be associated with potentially worse function or severe morbidity, were randomized (1:1) to Pex or placebo (Pbo) 1000 mg/d x 2 wks then 800 mg/d PO x 22 wks (Part 1). Pts completing Part 1 could continue into an open-label Pex extension (Part 2). The primary endpoint was centrally reviewed overall response rate (ORR) by RECIST at Week 25 for which the study was powered to detect a 25% difference (35% vs 10%). Results: 120 pts were treated, 61/59 on Pex/Pbo. Due to 2 cases of nonfa...

e23501Background: Osteosarcoma is a rare disease recurring mainly to the lungs. In unoperable cases high dose ifosfamide, gemcitabine/docetaxel, /etoposide are employed. Very few studies are availa...

e23502Background: Trabectedin (T) is approved for patients (pts) with STS after failure of anthracyclines (A) and ifosfamide (I), or pts unsuited to receive AI. ISG performed a retrospective study ...

e14543Background: The multi-targeted receptor tyrosine kinase inhibitors (TKIs) sunitinib and nivolumab (anti-PD-1) have shown remarkable antitumor activity. Preliminar clinical data on TKIs and an...

11018Background: Given that trabectedin (T) causes single- and double-strand breaks along DNA, we hypothesized to combine trabectedin with the selective PARP1-inhibitor olaparib (O) to induce irrep...

11021Background: Giant cell tumours of bone (GCTB’s) are RANK-ligand (RANK-L) positive, aggressive and progressive osteolytic tumors. Denosumab, a RANK-L inhibitor, was FDA-approved for adults and ...

11033Background: Treatment of patients with metastatic Ewing sarcoma (ES) remains a challenge. Pre-clinical studies and small retrospective series, mainly including pediatric patients, have demonst...

10527 Background: Although there are many reports on post-relapse survival for osteosarcoma,studies on validated clinical outcome measures (progression-free survival at 6 months [6-mo PFS], complet...

10571 Background: Synovial sarcoma (SS) is a rare and aggressive soft tissue tumor. In a first study (Palmerini E, Cancer 2009) we analyzed clinical prognostic factors in a retrospective series of 250 patients (pts) who were treated at Rizzoli Institute between 1976 and 2006. Stage, age, size, histology and use of radiotherapy influenced survival, whereas the role of chemotherapy was unproven. In the present study on the same patient population, in order to identify new prognostic factors and potential therapeutic targets, a panel of markers involved in chemotaxis and tumor growth were assessed. Methods: The expression of the chemokine receptor CXCR4 (a marker for chemotaxis which plays a critical roles in cancer progression) and the insulin-like growth factor receptor-1 (IGFR1) (a marker of growth activation ) were evaluated by IHC staining. Results: Tissue samples were available for the analysis in 88 patients (45 female and 43 male); median age was 37 years (range 11-63); size of the lesion was > 5 cm in 60 patients (71%); histology was biphasic in 30 (34%) of patients. All pts underwent surgery, 56% of pts underwent adjuvant radiotherapy (RT) and 68% adjuvant chemotherapy. With a median follow-up of 6 years (1-30 years), the 5-year overall survival (OS) was 70% (60-81). A positive stain for IGFR1 was detected in 55 pts (62.5%), with expression in the nucleus in 21 pts. CXCR4 was expressed in 74 pts (84%), nuclear pattern in 31 pts. No relation between IGFR1 and CXCR4 expression and clinical variables was found. The 5-year OS was 63% (95%CI 41-85%) for pts with positive IGFR1/nuclear expression and 73% (95%CI 61-85%) p = 0.05, in pts with negative IGFR1/nuclear staining. Similarly, the 5-year OS was 47% (95%CI 27-66%) in patients with positive CXCR4/nuclear expression and 86% (95%CI 76-96%), p = 0.003, in negative cases. In a multivariate analysis including age, histology, size and use of RT, nuclear expression of IGFR1 and CXCR4 were confirmed statistically significant independent factor for OS. Conclusions: The nuclear expression of CXCR4 or IGFR1 negatively influences the survival in patients with SS. Further studies addressing the role of CXCR4 as a potential target in this high risk subgroup of SS patients are needed.

10060 Background: CS is a rare and heterogeneous disease in which, after failure of surgery and radiotherapy, chemotherapy has a marginal role, if any. Different molecular pathways were shown to be...

10545 Background: Nearly 30–40% of patients with newly diagnosed, non-metastatic Ewing's Sarcoma (EWS) relapse. Post-relapse survival (PRS) in EWS is very poor, with less than 15–20% alive 5-years after recurrence. The role of high-dose chemotherapy (HDCT) with autologous stem cell transplantation (ASCT) is under investigation in metastatic and high risk localized EWS patients. In order to improve PRS, we employed HDCT with ASCT in EWS patients at 1st relapse. Methods: All non-metastatic EWS patients treated in Italian Sarcoma Group centers who relapsed between 1999 and June 2008 were offered HDCT (busulfan 4 mg/kg × 4 days orally and melphalan 140 mg/m2) with ASCT whenever possible (no previous HDCT; stable or responding disease after standard dose chemotherapy; adequate peripheral blood stem cells harvest). Results: 72 EWS patients experienced disease recurrence. Fifty (62%) were male. Twenty-eight (39%) of patients had previously received HDCT. Median relapse free interval (RFI) was 16 months (2.8–64). Pattern of relapse was: lung metastases in 20 (28%) patients, bone metastases in 12 (16%), local recurrence in 11 (15%) and multiple sites in 29 (40%). Treatment at 1st relapse was: standard dose chemotherapy in 31 (43%) patients; HDCT followed by ASCT in 24 (33%); palliative treatment in 12 (17%) and surgery only in 5 (7%). Three patients died of treatment-related toxicity. With a median follow-up of 24 months (1–64), the 3-year post-relapse survival (PRS) was 21% (95%CI 7–35). 3-year PRS was better for patients with a lung only relapse [48%, (95%CI 21–74)] and a RFI > 2 years [51%, (95%CI 27–76)]. 3-year PRS was 33% (95%CI 13–54) for patients treated with HDCT and 22% (95%CI 6–39) for those receiving standard dose chemotherapy. A significant (P 0.02) advantage was observed in the subgroup of patients with a shorter RFI treated with HDCT [3-year PRS 29% (95%CI 5–52)] compared to those treated with standard dose chemotherapy [3-years PRS 13%, (95%CI 2–29)]. Conclusions: Pattern of recurrence and RFI are the main factors influencing PRS in EWS. The use of HDCT with ASCT in recurrent EWS is investigational. Patients with shorter RFI could more likely benefit from HDCT. No significant financial relationships to disclose.

10525 Background: Trabectedin (ET-743, Yondelis), a marine-derived alkaloid has demonstrated significant activity in the treatment of soft tissue sarcomas (STS) in phase III trials, and has recently received EMEA approval in this indication. A subtype that accounts for 10% of STS, MRCL, displays the fusion FUS-CHOP in 95% of all cases. Preliminary results of neoadjuvant trabectedin (T) in advanced MRCL show reduction in the radiological density of the tumor, clinical improvement, and a pathological complete response (pCR) in the resected tumor mass. A phase II study to further determine the response to T in the MRCL population is presented. Methods: In this multicenter Phase II trial, patients (pts) with locally advanced (stage III) or locally recurrent MRCL were treated for 3 - 6 cycles with T (1.5 mg/m2 q3wk) in the neoadjuvant setting. Main endpoints were: pCR rate, objective response rate by RECIST, and correlation of molecular parameters from tissue samples with clinical outcomes. Results: Twenty-five pts with locally advanced MRCL have been recruited, of whom 20 are evaluable. All had the translocation (t12q13, 16p11) which causes the chimeric FUS-CHOP. Median age was 53 (23–75) and male:female ratio was 1. Thirteen pts completed therapy and underwent curative surgery. Pathological assessment was performed in 10 pts: 2 achieved pCR, 1 as per central pathology review and 1 by local pathology assessment. In addition, 1 pt had a very good pathological response. Ten patients remain to be histologically evaluated. Response rate by RECIST from pts who completed therapy was: 6 partial responses (46%) and 7 disease stabilizations. Remarkably, pathological response does not entirely correlate with response by RECIST since the pts with pCR still had radiological disease but no malignant component was found in the excised tumor mass (only connective and reactive tissue). Two serious adverse reactions of severe rhabdomyolysis, and asthenia, nausea and transaminase elevation were reported. Most common toxicities were liver enzyme elevation, neutropenia and thrombocytopenia. Conclusions: These preliminary results in terms of objective and complete pathologic responses, strongly suggest that T may have a role in the neoadjuvant setting in pts with MRCL. [Table: see text]

10578 Background: The best adjuvant treatment for soft tissue sarcomas is still unclear. Soft tissue sarcomas is an eterogenous group of several different hystotypes. The aim of this retrospective study is to evaluate prognostic factors and treatments. Methods: All liposarcoma cases evaluated in our Pathology from Jan 1985 to December 2005 were reviewed: 299 cases of liposarcoma were found of these only 155 were treated at Rizzoli since their first diagnosis. Metastatic since diagnosis and pediatric cases were excluded. Of these 155 pts only 122 had localized liposarcoma of soft tissue and adequate informations and follow up. Results: Male were 71 and female 51, Median age 47 (15–85). Histology was as follows: mixoid 62, round cell 33, pleomorphic 27. Grading according to FNCCL was G1=18, G2=51, G3=53. Median tumor volume was 183 cm3 (8–3,800), mean 401. Surgical margins were adequate in 84 patients (66.4%) unadequate in 34 (27.9%), 4 missing. Treatment received were: 18 no treatment besides surgery, pres...

10506 Background: The optimal treatment for synovial sarcoma (SS) is controversial. The present study examines treatment, outcome and prognostic factors in patients treated in a single institution....

9508 Background: Good cure rate has been reached in osteosarcoma treatment with chemotherapy, but there are late side effects. Methods: We reviewed charts of 755 patients (pts) with localized osteosarcoma of the extremity treated in 6 subsequent protocols (1983–2000) with Doxorubicin (300–480 mg/m2), Cisplatin (300–600 mg/m2), Methotrexate (3.7–75 g/m2), Ifosfamide (30–75 g/m2). Etoposide was employed only in 47 pts. Results: With a median follow up of 8 yrs (5–20), the median 10-yrs EFS was 58%. 13 patients (1.7%), 9 females and 4 males, had a symptomatic cardiopathy with two cardiopathy-related deaths, and 3 patients needed a heart transplant. The median age of these pts was 13 yrs (4–28). Median Doxorubicine dose = 480 mg/m2. Median interval from chemotherapy completion to cardiopathy onset was 3 months; 7 patients are alive, and 6 died: 4 for cardiopathy, 2 for metastatic disease. 17 second malignant neoplasms (SMN) occurred in 16 pts (2.1%) after a median interval of 7 yrs. One patient had both a breast and ovarian cancer. SMN were: 2 AML, 3 ALL, 1 CML, 3 breast cancer, 1 CNS tumor, 1 lung tumor, 1 parotid tumor, 2 soft tissue sarcoma, 1 skin cancer, 1 ovary cancer, 1 Ewing sarcoma. Eight of these 16 pts died of the second tumor. Infertility related to chemotherapy affected mostly males. Amenorrhea affected 69% of postpubertal females during chemotherapy but only two pts had permanent amenorrhea (39 and 43 yrs). No delay in pubertal maturation was seen in both gender. The incidence of infertility in male was related to Ifosfamide and was dose-dependent. Permanent azoospermia was 100% in males who received 60–75g/m2 Ifosfamide. No congenital defects were observed in the offsprings of both gender. The incidence of chronic renal failure was negligible in pts who received 2 Ifosfamide; in those who received >60 g/m2 a subclinical tubular impairment was observed in 48%, but only 1pt required dialysis. In the last protocol, a mild hearing impairment due to Cisplatin was evident in 40% of pts. Conclusions: Late toxicities are relevant and prolonged follow-ups are recommended as well as less toxic protocols. No significant financial relationships to disclose.

9537 Background: Vincristine (V), doxorubicin (A), cyclophosphamide (C), ifosfamide (I), actinomycin-D (Ac) and etoposide (E) are the standard drugs active against ES. Feasibility and efficacy of a six drugs induction treatment were investigated in a monoinstitutional study. Methods: Between March 1998 and May 1999, nonmetastatic ES aged ≤ 50 years were enrolled. Induction treatment: VAC (V 2 mg, A 80 mg/m2, C 1200 mg/m2, weeks 0 and 6), IVAc (I 9 g/m2, V 2 mg, Ac 2 mg, week 3), IE (I 9 g/m2, E 450 mg/m2, week 9). Local treatment, surgery whenever possible, was planned on week 12. Maintenance treatment: alternating courses of VAC-IVAc-IE (three times). In surgically treated patients, chemoinduced necrosis was evaluated and graded: grade III (complete necrosis), II (persistence of microfoci of viable tumor cells) and I (persistence of macrofoci of tumor cells). Results: 34 patients were enrolled; median age was 19 years (6–50); 22 were males and 12 females. Site: extremity 22 (65%), axial location 12 (35%). Despite a large use of G-CSF (94% of cycles), grade IV leukopenia was common (60% of cycles). Nevertheless, febrile neutropenia was observed in only 10.6% of cycles. Grade IV thrombocytopenia occurred in 12.5% of cycles. Platelet and red blood cell transfusions were required in 4% and 11% of cycles, respectively. No toxic deaths were recorded. Local treatment: surgery in 24 patients (70%), followed by post operative radiation (RT) in 6 of them; RT in 10 patients (30%). Chemoinduced necrosis was grade III in 29% of patients, grade II in 34% and grade I in 37%. With a median follow-up of 80 months (1–69) 5 years overall survival (OS) was 62%. 5 years Event free survival (EFS) was 56%. 5 years EFS according to site was: extremity 68%, axial location 33% (p < 0.02); according to local treatment was: surgery 61% (with RT 67%), RT 30% (p = 0.027); according to chemoinduced necrosis: grade III 86%, II 50%, I 55% (p = 0.26). Conclusions: The treatment is feasible. Surgery was possible in 70% of patients with a high cure rate for patients with a grade III chemoinduced necrosis. No significant financial relationships to disclose.

1026 Background: Increase amount of cyclooxygenase 2 (COX-2) is commonly found in malignant and premalignant lesions and overexpression of COX-2 is associated with both human and animal colon cance...