Consolidation with high-dose chemotherapy for responding patients to standard chemotherapy in advanced, metastatic soft tissue sarcoma (STS): A randomized trial from FNCLCC-French Sarcoma Group

10505

Background: Whether high dose (HD) chemotherapy improves disease-free (DFS) or overall (OS) survival has been suggested in phase II trial, but never explored in a randomized setting. This randomized, open, phase III study was designed to assess whether or not an HD chemotherapy with peripheral blood stem cells (PBSC) would improve OS in patients with advanced or metastatic STS responding to MAID chemotherapy. Methods: Pts aged 18 to 65 and with advanced STS were enrolled. After 4 courses of MAID, patients in PR or CR, or in whom complete surgical removal of all lesions was performed, were proposed for randomisation between 2 more cycles of MAID (control arm) vs 1 MAID followed by an intensification with MICE, ie: mesna (3.6g/m2, d1–5), ifosfamide (4g/m2, d1–4), carboplatin (UCA5, d2–4) and etoposide (300mg/m2, d1–4), followed by PBSC (HD arm). The primary endpoint was OS and the study was designed to detect by logrank test a 25% difference between the arms. Results: From 03/00 to 06/08, 266 patients were included and 87 were randomised (15 centres); low accrual and new treatment concepts lead to an IDMC in 11/08 who analysed 45 treated in the control arm (41 with full treatment) and 40 in the HD arm [only 21 received MICE, because consent withdrawal (6), insufficient PBSC harvest (5), tumor reprogression (4)]. Baseline characteristics (pts and tumors) were similar between treatment arms. With a 39 months follow-up, 25 pts were alive in the control arm, and 19 in the HD arm. The 3 years OS was 45.5% for control arm versus 35.8 for HD arm (HR = 1.12; 95% CI 0.58, 2.14; p = 0.72 Intention to treat analysis); PFS was 29.9% and 12.1 respectively (HR = 1.48; 95% CI 0.87, 2.53; p = 0.14). Higher rate of grade 3 - 4 hematologic (87% vs 46%), and digestive toxicity (33% vs 0%) were observed in the HD arm. Two treatment-related deaths occurred, both in the HD arm. Conclusions: In this study, HD chemotherapy for STS patients didn't improve OS and DFS. Consolidations with new concepts have to be explored.

[Table: see text]

Effect of adjuvant chemotherapy on survival in FNCLCC grade 3 soft tissue sarcomas: A multivariate analysis of the French Sarcoma Group database

10504

Background: The SMAC meta-analysis failed to demonstrate that adjuvant chemotherapy (AC) significantly improves overall survival (OS) in adult patients with localised resectable soft-tissue sarcoma (STS). We report here the analysis of the impact of AC in the population of STS patients included in the prospective database of the French Sarcoma Group. Methods: Between 1980 and 1999, 2,029 pts with STS were admitted to one of the 20 tertiary cancer centers of the GSF for the management of a first tumoral event and were included prospectively in a comprehensive database. 152 pts were excluded from the study because of metastatic disease at diagnosis. All the cases were reviewed by the pathology subcommittee of the GSF. Tumor grade was assessed according to the FNCLCC system based on tumor differentiation, mitotic count, and necrosis. Results: 283 pts (14.5%) had grade 1, 736 (39.5%) grade 2 and 858 (46%) grade 3 tumors. 1,102 pts (59%) had extremity tumors. The commonest pathological subtypes were MFH 22.5%, liposarcoma 18%, leiomyosarcoma 13%, and synovial sarcoma 10%. 1,122 pts (60%) received adjuvant radiotherapy. AC was delivered in 16 grade 1 pts (6%), 167 grade 2 pts (23%) and 323 grade 3 pts (38%). The majority of patients who received AC had tumors with a deep topography (91%) and/or > 5 cm (75%) and/or located in the limbs (61%). The median follow-up was 9 years. The 5 year-OS was 90% for grade 1 pts, 63% for grade 2 pts and 46% for grade 3 pts. On multivariate analysis ( table 1 ), AC was strongly associated with improved metastasis-free survival (MFS) (5 year MFS: 53% vs 47%, HR 0.7 [0.5–0.9], p=0.003) and overall survival (OS) (5 year OS: 56% vs 44%, HR 0.7 [0.5–0.8], p=0.004) in grade 3 pts. This association was not observed in grade 2 pts (5 year MFS: 73% vs 72%, HR 0.9 [0.6–1.4], p=0.9; 5 year OS: 73% vs 65%, HR 0.7 [0.5–1.1]). Conclusions: This large cohort-based analysis with long-term follow-up indicates that FNCLCC grade 3 pts are likely to benefit from AC. [Table: see text]

[Table: see text]

Frequency of β-catenin heterozygous mutations in extra-abdominal fibromatosis as a surrogate marker for disease management

10501

Background: Fibromatosis, a rare soft-tissue and locally invasive neoplasm, consist of distinct clinical entities including sporadic extra-abdominal fibromatosis (EAF). The potential morbidity of loco-regional treatments combined with the high local recurrence rate have led investigators to assess the possibility to design a “wait and see” policy in a subset of patients (pts). Up to date, molecular biomarkers of local recurrence of EAF remain unknown. As dysregulation of β-catenin (βC) expression hallmarks fibromatosis, herein we searched for βC mutations in a large retrospective European multi-centric tumors series to relate with clinicopathologic parameters. Methods: From 1987 to 2007, 155 pts with fibromatosis from all sites were gathered by the Conticanet Network were analysed. Direct sequencing of exon 3 βC gene was performed for 155 frozen tissues. All mutated cases were confirmed by an independent analysis. From this cohort, 101 pts with EAF surgically treated with curative intent and follow-up allowed us to relate βC mutational status to disease outcome. Results: Mutations of βC were detected in 84% of all cases. Among 101 pts with EAF surgically treated, similar mutations types and rates (87%) were observed, namely T41A (39.5%), S45P (9%), S45F (36.5%) and deletion (2%). There were 36 men and 65 women. Majority of pts were treated for primary EAF (57%). Most of surgical margin were RO (38%) or R1 (34%). Few pts were treated after surgery with radiation therapy (18%) and medical treatment (8%). With a median follow-up of 62 months, 51 pts (51%) relapsed. Neither clinicopathologic parameters were found to be significantly associated with βC mutational status. The five-year recurrence free-survival was significantly worse in βC mutated tumors regardless of a specific genotype, than in wild-type tumors (49 versus 93%, respectively, p=0.02). No significant difference was observed in considering either T41A, or S45F or S45P mutants, even if 45Fmutation appears to show a weak trend of significance for a dismal outcome. Conclusions: A high frequency (87%) of βC mutation hallmarks EAF from a large multicentric retrospective study. Moreover, wild type βC appears as a good prognosis marker that can be useful for therapeutic management of the EAF.

[Table: see text]

Molecular epidemiology of GISTs: Incidence of PDGFRA and KIT exon 9 mutations in the large French population-based study molecGIST

10535

Background: KIT exon 11, KIT exon 9, and PDGFRA exon 18 mutations are present in 67%, 10% and 5% of gastrointestinal stromal tumors (GIST). However most of these data are based on analyses of patients with overtly metastatic tumors. The MolecGIST study was design to evaluated prospectively the histological, molecular and clinical characteristics of all GISTs diagnosed at all stages. Methods: All patients leaving in France, whose diagnosis of GIST was achieved between June 1, 2006 and May 31, 2008 were includable in the study. MolecGIST was approved by French ethical comities. Samples were collected in 11 reference centers for histology review, immunohistochemistry and detection of mutations within exon 9, 11, 13, and 17 of KIT and 12, 14, and 18 of PDGFRA. Results: In December 2008, 738 patients were referred to MolecGIST, corresponding to 5.95/million/year incidence of GIST en France (62 millions inhabitants). The median age at diagnosis 63.1 years (range [15–96]), and female/male ratio 0.98. GIST originated in the stomach in 56.3% and small bowel in 29.3% of the cases. Tumors were KIT+ in 90,4% of cases. CD34 was expressed by more than 20% of tumor cells in 83.8% of gastric GIST and 39.7% of small bowel GIST respectively (P<0.0001). The risk of metastatic relapse was moderate or high in 62.7% of the patients, according to Fletcher classification, while only in 44% according to Miettinen's. Mutation analyses were achieved in 438 cases (to be updated), and mutations of either KIT or PDGFRA were detected in 66.2% and 12.6% of the patients, and absent in 21.2%. The most frequent mutations were KIT exon 11 (59.1%), PDGFRA exon 18 (9.8%) and KIT exon 9 (4.3%). The incidence of PDGFRA exon 18 and KIT exon 9 were significantly different than the published data (P=0.0008 and P<0.0001 respectively). Among GIST of the small bowel, the mutations were KIT exon 11 (69.1%) and KIT exon 9 (13.1%). Conclusions: The PDGFRAexon18/KITexon9 mutation ratio is 4.5 times higher in our population based study than in patients included in phase III trials. The incidence of GIST with PDGFRA mutations and intestinal GIST exon KIT exon 11 mutations is underestimated in published series as compared to ours, probably because of the better prognosis.

[Table: see text]

Incidence of gastrointestinal stromal tumors (GISTs) in France: Results of the PROGIST survey conducted among pathologists

10047

Background: GISTs are the most common mesenchymal tumors of the gastrointestinal tract but data on the annual incidence of this tumor is scarce due to previous lack of well-defined pathologic criteria. The aim of this study was to investigate the incidence of GISTs in France and describe the distribution according to age, sex, circumstances of discovery, localization and risk categories. Methods: This prospective epidemiologic study was performed among pathologists who were asked to declare all the cases of GIST over a one year period. All pathology centers (330 centers) in France were proposed to participate and asked to report exhaustively the cases diagnosed from 1 December 2004 to 30 November 2005. Results: 591 cases of GISTs were reported, 535 new cases and 56 cases of relapse. However, some large centers were not able to participate. So, the annual estimated incidence was 12 cases/million inhabitants in France. The main characteristics of the new cases of GIST were as follows: mean age 65 (± 13.2) (range: 16–93) years, 48.6% men and 51.4% women, circumstances of discovery: fortuitous 163 (30.5%), symptomatic 249 (46.5%), unknown 123 (23%). The primary tumor locations were: stomach 341 (63.7 %), small intestine 115 (21.5%), mesentery 35 (6.5%), colon/rectum 17 (3.2%), esophagus 4 (0.7%), unknown 23 (4.3%). 95.3% of GISTs were cKIT (CD117) + and 3.7 % were cKIT -. According to the Fletcher consensus criteria, based on tumor size and mitotic rate, among the 490 localized GISTs 14.7% were considered to be at very low prognostic risk, 25.5% at low-risk, 23.1 % at intermediate risk and 23.1 % at high-risk. For 13.6 % data were missing. Conclusions: This study is providing for the first time an estimation of the annual incidence of GISTs and a description of the characteristics of these tumors in France based on pathology registration. The true incidence may be slightly higher as some large centers were not able to reported their cases. These results are comparable to what has been reported in recent studies in other European countries.

No significant financial relationships to disclose.

Tyrosine kinases as possible therapeutic targets in pulmonary artery intimal sarcoma

10055

Background: Pulmonary artery intimal sarcomas (PAIS) are very rare, highly malignant and clinically aggressive tumors, with yet elusive underlying pathogenic mechanisms and histogenesis. Our aim was to gain molecular insight in the oncogenic events leading to PAIS development, and to investigate the possibility of a pharmacological approach for the treatment of PAIS, specifically targeting the receptor tyrosine kinases (RTK) activated in these tumors. Methods: Fourteen cases of advanced high grade poorly differentiated PAIS were analyzed by immunohistochemistry and selectively by a variety of cytogenetic/molecular techniques, e.g. karyotyping, FISH, mRNA in situ hybridization, array-CGH, tyrosine kinases RT-PCR assay and Western blotting analysis. The study was supplemented by sequencing of PDGFRA, PDGFRB, KIT, VEGFR-1, VEGFR-2 and EGFR genes. Ex vivo functional assays were applied to test the sensitivity of PAIS primary tumor cells to different, clinically approved RTK inhibitors. Results: We demonstrate that (1) PAIS do not have mutations within the kinase domains of PDGFRs, KIT, VEGFRs or EGFR, (2) all cases show a similar cytogenetic molecular profile, characterized by amplification of PDGFRA/KIT/VEGFR2 and MDM2/CDK4 genes, while EGFR, PDGFRB and VEGFR1 are frequently polysomic but rarely highly amplified, (3) high level of expression of PDGFRA, VEGFR1/VEGFR2 and EGFR is uniformly present in the tested tumors, (4) PAIS demonstrate activation (phosphorylation) of PDGFRA and downstream PI3K-AKT and MAP-Kinase signaling pathways, (4) in ex vivo assays, a potent effect of BMS354825 on proliferation and survival of primary cells of PAIS is proven. Conclusions: PDGFRA is amplified and activated in PAIS, thus constituting one of many possible targets for the therapy. Tested inhibitors showed differential efficacy for inhibition of PAIS cells by ex vivo assays.

No significant financial relationships to disclose.

Detection of the synovial sarcoma translocation t(X;18) (SYT;SSX) in paraffin-embedded tissues using reverse transcriptase-polymerase chain reaction: a reliable and powerful diagnostic tool for pathologists. A molecular analysis of 221 mesenchymal tumors fixed in different fixatives

Synovial sarcoma (SS) is a relatively rare sarcoma, which may be confused with several other mesenchymal and nonmesenchymal lesions. It bears the t(X;18) (SYT;SSX) translocation, which seems to be specific for this tumor type and can be detected in paraffin-embedded tissue, using reverse transcriptase-polymerase chain reaction (RT-PCR). However, the specificity and sensitivity of this detection method have rarely been examined in a large series. Using RT-PCR, we examined 250 mesenchymal and nonmesenchymal, benign and malignant, paraffin-embedded lesions for the SS t(X;18) (SYT-SSX) translocation. PCR products were obtained from 221 tumors (88.5%). There were 135 non-SS tumors, 22 biphasic, and 64 monophasic spindle/round cell SS, of which 10 were cytogenetically confirmed as t(X;18)-positive. SYT-SSX gene fusion transcripts were detected in the SS tumor category only (100% specificity), including 100% of the biphasic SS and 86% of monophasic spindle/round cell SS. Nine tumors originally diagnosed as SS were t(X;18) (SYT-SSX)-negative. Following reassessment, only 3 of these tumors showed clinicopathologic, immunohistochemical, and/or ultrastructural features consistent with that diagnosis, thus raising the overall detection sensitivity to 96%. With regard to the potential adverse effect of the fixatives used, PCR products were obtained in 100%, 91.5%, 90.5%, and 0% of tumors fixed with AFA, buffered formalin, Holland Bouin, and conventional Bouin's fluid, respectively. This study shows that the detection of the SS t(X;18) (SYT-SSX) in paraffin-embedded tissue is feasible with a 100% specificity and an overall 96% sensitivity, provided non-Bouin's fluid fixation is used.