Search

Michael Balter's recent article about French AIDS research (News & Comment, 16 Jan., [p. 312][1]) describes a feudal system with barons, czars, big bosses, and 35-year-old scientists considered as only young wolves. Unfortunately, this is not restricted to the AIDS research field, but is a general phenomenon in science and medicine in France. Thirty-five-year-old medical investigators have the same problems obtaining some independence from their chiefs, called “mandarins” by some. The choice is often between going into private practice or staying in a university position for years without any possibility of developing one's personal ideas. The potential of young French clinical investigators is thus often lost. [1]: /lookup/doi/10.1126/science.279.5349.312

Despite the significant progress in imaging, particularly endoluminal ultrasound for the examination of the pelvis, and CT for the examination of the abdomen, the diagnosis of cancer of the ovary is still usually made at a late stage with a resultant poor prognosis. However, some improvements in treatments have recently been accomplished. These include the introduction of the platinum derivatives and even more recently several other promising drugs (Taxanes...). There have also been advances in surgery which is essentially the first stage of the treatment.

Background: Iniparib (I) is a novel anticancer agent initially thought to be a poly (ADP-ribose) polymerase (PARP) inhibitor. Recent preclinical data suggest that iniparib does not inhibit PARP at pharmacologically-relevant concentrations. The initial Phase 1 did not identify any dose limiting toxicity (DLT) and the dose of 5.6 mg/kg used in further development was based on a PK/PD model for PARP inhibition. This new Phase 1 study was designed to determine a maximum tolerated dose (MTD) based on DLT occurrence for single agent iniparib and in combination with gemcitabine and carboplatin (GC). Patients and Methods: Patients with advanced cancer and ECOG performance status (PS) 0-2 with no standard therapeutic option or for whom GC could be considered an adequate treatment were enrolled in this trial. In the Phase 1 portion, patients received escalating doses of single agent iniparib starting at 15 mg/kg in a 3+3 design. In the Phase 1b, two different regimens of GC and iniparib were evaluated (GC2+I: carboplatin AUC2, gemcitabine 1,000 mg/m2 and iniparib on Days 1 and 8 every 21 days; and GC5+I: carboplatin AUC5 on Day 1, gemcitabine at 1,000 mg/m2 and iniparib both on Days 1 and 8 in a 21-day cycle). At the MTD, the cohort was expanded by 12 patients. Safety, pharmacokinetics, pharmacodynamics, and antitumor effects were also evaluated. Results: Of the 59 patients (M/F= 18/41; ECOG 0/1-2=21/38) enrolled in this study, 29 received single agent iniparib (15 mg/kg=3; 20 mg/kg=16; 26 mg/kg=10); 23 received GC2+I (15 mg/kg=3; 20 mg/kg=20) and 7 were treated in the GC5+I arm at 15 mg/kg. DLT in the phase 1 was G3 transient increase in blood pressure during the iniparib infusion in 4 patients treated at 26 mg/kg. Nine patients at 20 mg/kg (5 in Phase 1 and 4 in Phase 1b) experienced also transient non-DLT hypertension. In most cases HTN was managed by interrupting the infusion; infusions were completed after blood pressure normalization. No patient was discontinued due to it. DLT for GC2+I was G4 thrombocytopenia in 3 patients. GC5+I at 15 mg/kg was considered non-tolerable due to G4 hematological toxicity in 2/5 patients. One ovarian cancer patient treated with iniparib single agent at 26 mg/kg for longer than 18 months had a durable (9 months) partial response (PR). Ten PRs were recorded in combination with GC. Conclusions: The MTD of iniparib single agent and in combination with GC2 is 20 mg/kg weekly. Hypertension during the infusion has been identified as a characteristic DLT. The MTD of GC5+I is below 15mg/kg of iniparib weekly due to hematological toxicity. The GC2+I regimen demonstrated manageable tolerability. ClinicalTrials.gov identifier: NCT01455532 Citation Information: Mol Cancer Ther 2013;12(11 Suppl):C242. Citation Format: Monica Mita, Raffaele Baffa, Eric Charpentier, Craig A. Lockhart, John C. Morris, Olivier Rixe, John Sarantopoulos, Ramesh Ramanathan. A phase 1/1b dose escalation study of iniparib as a single agent and in combination with gemcitabine/carboplatin in patients with advanced solid tumors. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr C242.

Background: SAR566658 (SAR) is a maytansinoid-loaded ADC (huDS6-SPDB-DM4) targeting CA6, a specific glycol-epitope of MUC-1 over-expressed in solid tumors (pancreas 26%, ovary 55%, breast 30%, bladder 60%) and rarely in normal tissues. This FIH study was designed to assess the safety, dose limiting toxicities (DLTs)/recommended dose (RD) and pharmacokinetic following SAR administration in Pts with CA6-expressing STs. Trial is funded by Sanofi. Methods: This Phase I study explored escalating intravenous doses of SAR administered as single agent every 3 weeks (q3w). An accelerated dose escalation scheme was used for the two first dose levels (DL), followed by a standard 3+3 dose escalation scheme. Results: 34 heavily pretreated Pts were enrolled including: 11M/23F, median age 58 years (range, 32-77), ECOG-PS ≤1, with a variety of advanced STs including ovary (13), pancreas (10) and breast (4). A total of 114 cycles (cy), median 2, (range,1-14) of SAR was administered across 9 DLs ranging from 10 to 240 mg/m2. DLTs were observed at the highest DL of 240 mg/m2 and included grade (Gr) 3 diarrhea at cy1 in 1 Pt and Gr3 keratitis at cy2 in 2 Pts. Anticipated toxicity was cornea, peripheral neuropathy, hematological and pulmonary. So far the number of Pts with these toxicities are: keratitis (all Gr: 11 Pts, including 2 Pts with Gr3), peripheral neuropathy (5 Pts, no Gr≥3), neutropenia (Gr3, 2 Pts), interstitial pneumonitis (1 Pt). Other than late occurrence of reversible corneal adverse events (AE) at 150 mg/m2, no dose-dependent AE was observed. Exposure to SAR (Cmax and AUC) increased with no major deviation from dose proportionality over doses of 20 to 240 mg/m2. Clearance was roughly constant over the doses with a low to moderate total variability. SAR 190 mg/m2 fulfills the criteria for RD: no DLT and manageable ocular AE versus highest DL. Clinical benefit was observed at doses ≥120 mg/m2: 1 partial response (breast), 1 PR to be confirmed (ovary), 3 stable disease (SD)>6months and 11 SDs were noted. A significant decrease in tumor marker was noted in 1 Pt. Conclusions: SAR has a favorable safety profile and encouraging antitumor activity. SAR at 190mg/m2 q3w was selected as the RD and is being confirmed in an ongoing extension cohort. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A73. Citation Format: Valentina Boni, Olivier Rixe, Drew Rasco, Carlos Gomez-Roca, Emiliano Calvo, John C. Morris, Anthony W. Tolcher, Sylvie Assadourian, Helene Guillemin, Jean-Pierre Delord. A Phase I first-in-human (FIH) study of SAR566658, an anti CA6-antibody drug conjugate (ADC), in patients (Pts) with CA6-positive advanced solid tumors (STs) (NCT01156870). [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A73.

e13023 Background: Several strategies have been recently developed to target EGFRvIII in Glioblastoma Multiforme (GBM), including vaccines (CDX-110) and antibody-drug conjugates (AMG595) and represents a new challenging therapeutic avenue with potential great clinical benefits. EGFRvIII is a truncated extracellular mutant of the EGF receptor (EGFR) found in about a third of GBMs, confers enhanced tumorigenic behavior and is associated with chemo- and radio-resistance. GBM patients testing positive for EGFRvIII have a bleaker prognosis than those who do not. Methods: We developed a strategy to detect EGFRvIII deletion in the circulating tumor DNA. The purpose of this study is to identify a simple and robust biomarker from the peripheral of patients diagnosed with GBM in order to screen patients for the EGFRvIII deletion. 11 patients have been included in this study. Results: The circulating DNA status for EGFvIII correlates with the analysis performed on the respective tumors samples, and its level seems to be correlated with the extend of the tumor resection. Conclusions: This semi-quantitative blood biomarker may represents a strategy to 1) screen patients for anti-EGFRvIII therapy 2) monitor the therapeutic response to specific targeted therapies.

e13518 Background: The mammalian target of rapamycin (mTOR) is a critical signaling pathway in many tumors including cancers of the breast and colon, glioblastoma multiforme, and hepatocellular carcinoma (HCC). Preclinical studies demonstrated the combination of BEZ235, a competitive dual phosphatidylinositide 3-kinase (PI3K)/mTOR inhibitor, and the mTOR inhibitor, everolimus led to greater regression of a carcinogen-induced HCC than treatment with higher doses of either drug alone. Based on this, we initiated a study of BEZ235 combined with everolimus in patients with advanced solid tumors. Methods: A single institution phase Ib dose-escalation study. Patients with advanced solid malignancies, no available standard of care treatment option, and ECOG performance status 0-2 were eligible. Prior treatment with PI3K inhibitors was not allowed. Sequential cohorts of 3-6 patients were treated. The starting dose was everolimus 2.5 mg and BEZ235 200 mg daily in an oral sachet formulation. Cohort 2 received everolimus 2.5 mg and BEZ235 400 mg daily. Pharmacokinetic and pharmacodynamic studies were performed during the first cycle. The phosphorylation of specific downstream effectors of the mTOR pathway was assessed in peripheral blood mononuclear cells (PBMC). Results: Eleven patients, median age 58 (36-73 years) were treated. Tumors included non-small cell lung cancer, colon cancer, and glioblastoma, hepatocellular carcinoma, pancreatic cancer, esophageal cancer, adenoid cystic carcinoma of the larynx, and appendiceal carcinoma. Four patients were treated on cohort 1. None experienced dose-limiting toxicity. Seven patients were treated on the second dose cohort. One patient withdrew necessitating replacement and another developed grade 3 stomatitis from herpes virus requiring cohort expansion. The most common adverse events were thrombocytopenia, lymphopenia, transaminitis, diarrhea, nausea and fatigue. No tumor responses were noted. PBMC showed a decrease in 4E-BP1S65 phosphorylation on day 28 in 2 of 3 patients in cohort 1. Conclusions: The combination of BEZ235 and everolimus was well tolerated at these doses. The trial remains open to accrual. Clinical trial information: NCT01508104.

2609 Background: MM-121 is a fully human monoclonal antibody targeting the epidermal growth factor receptor family member ErbB3. ErbB3 has been implicated in driving cancer growth and in the development of resistance to conventional chemotherapies across multiple malignancies. Here we present results of an open-label, Phase 1, multicenter, non-randomized, dose-escalation trial which recently completed enrollment evaluating MM-121 in combination with one of the following chemotherapies: Gemcitabine (Arm A, n=11), carboplatin (Arm B, n=11), pemetrexed (Arm C, n=10), or cabazitaxel (Arm D, n=11). Methods: Patients were treated in a dose escalation “3+3” design to assess the safety, tolerability and pharmacokinetics (PK) of MM-121 administered weekly in combination with anticancer therapies in subjects with advanced cancer. Doses were escalated until the maximum tolerated dose (MTD) was identified or the combination was shown to be tolerable at the highest planned doses. Secondary objectives included: Determining the objective response rate, clinical benefit rate, PK and immunogenicity of MM-121. Data summarized are as of 1/17/2013 from a live database. Results: Overall, 43 patients, [22 (51%) female and 21 (49%) male] have been treated with a median treatment duration of 57 days (range 1-302). The median age was 59 years (range 42-84) and patients had received a median of four prior lines of therapy (range 0-13). Common (>20%) adverse events of any grade and causality across all arms included diarrhea (74%), nausea (54%), fatigue (51%), anemia (44%), vomiting (33%), hypokalemia (30%), decreased appetite (26%), thrombocytopenia (26%), peripheral edema (23%), neutropenia (21%), and constipation (21%). Four DLTs were observed: Two in combination with carboplatin (G4 thrombocytopenia and G3 rash), one with gemcitabine (G4 thrombocytopenia), and one with pemetrexed (G4 hyperuricemia). Overall 38 (88%) patients were evaluable for response and the overall clinical benefit rate (PR or SD >18 weeks), is 32% (12/38). Conclusions: MM-121 can be combined at its recommended single agent dose with standard doses of gemcitabine, pemetrexed, and cabazitaxel and adapted doses of carboplatin. Clinical trial information: NCT01447225.

2560 Background: Cbz is approved for treatment of pts with hormone-refractory metastatic prostate cancer after progression on a docetaxel-containing regimen. The dose-escalation part of this Phase I study (NCT00925743) found the maximum tolerated dose (MTD) of Cbz/Cis to be 15/75 mg/m2. No Cbz–Cis PK interactions were seen. As Cbz is mainly metabolized by CYP3A, the study also evaluated the impact of a strong CYP3A inhibitor (K; study part 3) or strong CYP3A inducer (R; study part 4) on the PK of Cbz in combination with Cis. Methods: The study included pts with metastatic or unresectable solid tumors for which Cis-based therapy was considered appropriate. Pts received Cbz/Cis every 3 weeks at MTD, with K 400 mg (part 3) or R 600 mg (part 4) administered orally once daily prior to, on, and after Cycle 2 Day 1 for a total of 10 (K) or 14 (R) days. In part 3, the Cbz/Cis dose was 5/75 mg/m2 in Cycles 1 and 2 to provide a safety margin to the MTD. Effects on Cbz PK were assessed with a linear mixed-effects model. The primary endpoint was Cbz clearance (CL). Safety assessments included adverse events (AEs) and laboratory abnormalities. Results: The PK population included 23 pts (part 3) and 21 pts (part 4). Repeated administration of K resulted in a 20% decrease in Cbz CL (L/h/m2) (geometric mean ratio [GMR]: 0.80; 90% confidence interval [CI]: 0.55, 1.15; n = 18), corresponding to a 25% increase in AUC (ng*h/mL/mg/m2) (GMR: 1.25; 90% CI: 0.86, 1.81; n = 18). Repeated administration of R resulted in a 21% increase in Cbz CL (L/h/m2) (GMR: 1.21; 90% CI: 0.95, 1.53; n = 20), corresponding to a 17% decrease in AUC (ng*h/mL/mg/m2) (GMR: 0.83; 90% CI: 0.65, 1.05; n = 20). The GMR of AUC0–24was 1.09 (90% CI: 0.9, 1.33; n = 21), suggesting a low impact of R during the initial phases of Cbz elimination. The most frequent AEs included nausea (part 3: 68%; part 4: 74%), vomiting (part 3: 68%; part 4: 74%) and fatigue (part 3: 52%; part 4: 70%). Conclusions: The CL of Cbz was decreased by 20% by co-administration with K and increased by 21% by co-administration with R. Safety results were consistent with prior findings for this Cbz/Cis combination. Clinical trial information: NCT00925743.

126 Background: Cabazitaxel (Cbz) is approved in combination with prednisone/prednisolone for the treatment of men with hormone-refractory metastatic prostate cancer previously treated with a docetaxel-containing treatment regimen. In vitro studies showed that Cbz is mainly metabolized through CYP3A, resulting in inhibition of this family of enzymes. Midazolam (Mdz) is primarily metabolized by CYP3A4. We aimed to determine the effect of Cbz on CYP3A activity by comparing the pharmacokinetic (PK) properties of Mdz when administered alone and following co-administration with Cbz. Methods: An ongoing safety and PK study of Cbz in patients with metastatic or locally advanced solid tumors and varying degrees of hepatic impairment (NCT01140607) included a cohort with normal hepatic function to assess the effect of a single Cbz dose on the PK profile of a single dose of Mdz. This was an open-label, two-period, fixed-sequence study in patients aged between 45 and 60 years with advanced solid tumors and normal hepatic function. A single dose of Mdz (2 mg) was administered orally alone (Day –1) and at the end of a 1-hour infusion of Cbz (25 mg/m2) (Day 1), with a 24-hour interval between the two administrations of Mdz. Endpoints included AUC and AUClastof Mdz with and without Cbz administration, and safety evaluations. Results: Of the 13 patients enrolled and treated in the cohort, 11 patients were included in the PK analysis. Exposure (AUC and AUClast) and other PK parameters after a single administration of Mdz alone and in combination with Cbz (Day 1) were similar. The AUC ratio for Mdz administered alone or with Cbz was 0.97 (90% CI: 0.76–1.23). The AUClast ratio for Mdz administered alone or with Cbz was 1.04 (90% CI: 0.81–1.34). All 13 patients had ≥1 adverse event (AE), 11 (84.6%) experienced a Grade 3–4 AE, and 4 (30.8%) experienced a serious AE. The majority of Grade 3–4 AEs were haematological and no new or unexpected safety findings were observed. Conclusions: In this study, Cbz did not increase the plasma exposure of Mdz. This indicates that Cbz is not a CYP3A inhibitor in the clinical setting and can be administered in combination with drugs metabolized by CYP3A. Clinical trial information: NCT01140607.

The aim of the study was to use a virus-free system to transfer the Herpes Simplex Virus-thymidine kinase (HSV-TK) gene in mice bearing melanoma tumours. B16 Fl murine melanoma cells were injected subcutaneously. On days 11 and 14, an intratumoral injection of either naked plasmid containing the HSV-TK gene (PAGO) or PAGO-lipofectamine complexes was given. Ganciclovir (120 mg/kg/day) was given for 5 days starting on day 14. Tumour weight reduction (40-50%) was observed in treated animals versus different control groups. Moreover, histopathological analysis on tumours showed large areas of cavitary necrosis (85%) in treated groups compared to controls (10%). Using a simple and safe method, the results presented here demonstrated that virus-free mediated delivery of the HSV-TK gene is efficient in viva in murine malignant melanoma. Copyright 0 1996 Elsevier Science Ltd

Background In this study, the authors identified seven-year survivors after completion of the French multicenter Phase II trial of fotemustine for the treatment of metastatic malignant melanoma. Methods One hundred sixty-nine patients with metastatic malignant melanoma were included in this Phase II study. One hundred fifty-three patient records were evaluable with an overall response rate of 24.2%. Results Five of these patients are alive and in complete remission. One patient had a complete response after fotemustine administration and then relapsed. One patient had a partial response. Three patients had stable disease. These five patients underwent surgery for relapse or residual disease and subsequently achieved durable complete remission. Conclusions Long term survival may be the outcome after surgical resection of residual metastatic melanoma after chemotherapy.

Rationale Gem + taxane combinations have shown activity in some advanced cancers. Cbz (a novel taxane) + prednisone (P) improved overall survival vs mitoxantrone + P in pts with metastatic castration-resistant prostate cancer previously treated with docetaxel (de Bono 2010). Methods A Phase I study (NCT01001221) was conducted to determine the maximum tolerated dose (MTD) and dose-limiting toxicities (DLTs) of Cbz + Gem (parts 1a and 1b) and to evaluate antitumour activity at the MTD (part 2). Pts with histologically or cytologically confirmed metastatic or unresectable solid tumours without existing standard treatment were eligible. Cohorts of 3–6 pts were treated with Cbz (15–20 mg/m2 1-h infusion) followed by Gem (700–1000 mg/m2) on d1 and Gem alone on d8 (part 1a). The administration sequence on d1 was then reversed (part 1b), based on preliminary PK data. Results Of the 12 pts in part 1a, 5 pts were treated at Cbz 20 mg/m2 + Gem 1000 mg/m2 dose level (DL 20/1000), 5 pts at DL 15/900 and 2 pts at DL 15/700. In part 1b, all 6 pts were treated at the lowest DL (700/15) allowed. At all DLs, ≥ 2 pts experienced a DLT, regardless of administration sequence. DLTs were febrile neutropenia (FN) (4 pts), Grade (Gr) 4 neutropenia (2 pts), Gr 4 thrombocytopenia (2 pts) and Gr 3 AST increase (1 pt). No MTD was established and part 2 was not performed. All pts experienced at least 1 treatment-emergent AE (TEAE); the most frequent all grade non-haematological TEAEs were fatigue 66.7%, decreased appetite 50.0%, diarrhoea 44.4%, nausea 38.9% and weight decrease 33.3%. Gr 3–4 haematological toxicities included neutropenia 66.7%, thrombocytopenia 33.3% and FN 22.2%. Nine pts continued study treatment and received 6–22 cycles; 3 pts had a partial response (melanoma, prostate small cell and appendiceal tumours), while 8 pts experienced stable disease. PK analysis did not reveal a drug–drug interaction between Cbz and Gem. Conclusion The MTD of Cbz + Gem could not be established due to DLTs. Antitumour activity was observed and drug administration sequence did not affect the toxicity profile. Further investigation of alternative dosing regimens is warranted in an effort to establish a tolerable combination. Disclosure P.M. LoRusso: Has received research funding from Sanofi. J. Yin: Is a Sanofi employee and owns Sanofi stocks and shares. S. Doroumian: Is a Sanofi employee (pharmacokineticist) and owns Sanofi stocks and shares. X. Zhi: Is a Sanofi employee (statistician) and owns Sanofi stocks and shares. A.J. Olszanski: Research funding - Sanofi supplies support for the conduct of the study. All other authors have declared no conflicts of interest.

TPS2103 Background: EGFRvIII is a constitutively active tumorigenic deletion mutation of EGFR. It is expressed in ~30% of primary GB where it is linked to poor long-term survival (Pelloski 2007). The investigational vaccine rindopepimut consists of the unique EGFRvIII peptide sequence conjugated to keyhole limpet hemocyanin (KLH), delivered intradermally (500ug with 150ug GM-CSF as an adjuvant). Remarkably consistent and promising results across 3 phase II studies in newly diagnosed, resected EGFRvIII+ GB (Lai 2011) represent a statistically significant improvement over a historical control cohort matched for major eligibility criteria (median overall survival [OS] = 24.4 - 24.6 vs. 15.2 months from diagnosis [m] and median progression-free survival [PFS] = 12.3 - 15.3 vs. 6.4 m). ACT IV, a phase III trial in this population, is ongoing. The immunosuppressive influence of residual/advanced GB presents a challenge to activation of efficacious antitumor immune responses. Anecdotal evidence (compassionate use cases, Sampson 2008) suggests that rindopepimut may induce specific immune responses and regression in multifocal and bulky residual tumors. Rindopepimut with BV, which inhibits VEGF and its immunosuppressive properties (including impaired maturation of dendritic cells and disruption of tumoral T cell infiltration [Johnson 2007, Shrimali 2010]) may further optimize EGFRvIII-specific immune response and antitumor activity. Methods: ReACT is a Phase II study of rindopepimut plus BV in patients (pts) with 1st or 2nd relapse of EGFRvIII+ GB. BV-naïve pts will be enrolled to Group 1 (n=70: randomized 1:1 to BV plus either rindopepimut/GM-CSF or control injection [low-dose KLH]) while BV-refractory patients will enter Group 2 (n=25: to receive BV plus open-label rindopepimut/GM-CSF). Concurrent with BV (10 mg/kg, q 2 wks), blinded treatment or open-label vaccine is given in priming phase (days 1, 15 and 29), then monthly until PD. Tumor response is assessed every 8 weeks, and patients are followed for survival after PD. Objectives are PFS at 6 months (primary), objective response rate, PFS, OS, safety, immunogenicity and elimination of EGFRvIII. ReACT opened to accrual in December 2011 (NCT01498328).

TPS3116 Background: Downregulation of a number of signaling pathways, including the mammalian target of rapamycin (mTOR) pathway, has been demonstrated to be efficacious in a large range of solid tumors such as breast, colon, endometrial, glial and hepatocellular carcinoma (HCC). However, we find that rapamycins lead to suppression of a negative feedback loop from S6 Kinase 1 (S6K1) to Protein Kinase B (PKB), leading to hyperactivation of PKB. In pre-clinical studies using a mouse model of carcinogen-induced HCC, we have demonstrated that combining BEZ235 (a potent and highly selective reversible ATP site competitive inhibitor of PI3K and mTOR) with everolimus (an allosteric inhibitor of mTOR) synergizes to inhibit tumor growth. BEZ235, an orally administered agent, has demonstrated preliminary antitumor activity in a first-in-human phase I study. The current study will evaluate this combination in patients with a variety of solid malignancies that includes glioblastoma multiforme (GBM). Methods: This study is divided into a phase 1b portion designed to determine safety of increasing doses of the combination, with extensive pharmacokinetics, pharmacodynamics and pharmacogenomics analysis; and a phase 2 portion that includes both solid tumors and GBM based on predominance of the mTOR and PI3K deregulation in these tumors, to determine preliminary anti-tumor activity and the recommended dose for phase 2 studies. We will also integrate biomarker assessment for gene expression products of the mTOR downstream pathway such as eukaryotic initiation factor 4E binding protein (4EBP1) and S6 kinase (S6K). The phase 1b portion has started accruing.

7556 Background: Erlotinib is effective in NSCLCs with wild-type EGFR and shows enhanced benefit in EGFR mutation-positive cancers. However, resistance invariably develops, often involving persistent ErbB3 signaling and activation of the PI3K/AKT survival pathway. We present the full results of the Phase 1 study evaluating the safety and tolerability of erlotinib plus MM-121 a fully human IgG2 monoclonal antibody (mAb) to ErbB3. Methods: Eligible patients had advanced stage NSCLC, and ECOG 0-2. Seven cohorts were enrolled, evaluating varying levels of the MM-121 and erlotinib, as well as alternate MM-121 infusion schedules. Tumor response was assessed every 8 weeks. Dose levels were determined by safety and pharmacokinetic (PK) data, and immunogenicity, efficacy endpoints and exploratory biomarker evaluations were performed. Results: From February 2010 – July 2011 32pts entered the study (median age 63y; 45% male; 18% ECOG 0, 82% ECOG 1. 56% had adenocarcinoma and 30% pts received 3 or more lines of prior therapies (range 0-7) with 91% having had prior platinum. 65% had wild-type EGFR status and were never treated or never responded to EGFR-TKIs (EGFRwt) and 28% had acquired resistance to erlotinib treatment (EGFRresist). The most common toxicities (any grade) observed were diarrhea (82%), rash (64%), and fatigue (64%). DLTs observed in different cohorts were diarrhea, mucositis, rash and failure to thrive. Clinical activity was observed including 1 PR (an EGFR TKI naïve EGFR mutant) and 14 SD. Average duration of disease stabilization was 21.6 wks (range 7.1 -89.3 wks). Median PFS is 8.1 weeks and the 16 week PFS rate was 41% in the overall population. For EGFRwt and EGFRresist pts the median PFS were 7.7 weeks and 15.1 weeks, and the 16 week PFS rates were 32% and 44%, respectively. 7/20 EGFRwt pts and 5/9 EGFRresist pts achieved SD. Conclusions: This study suggests that the combination of erlotinib and MM-121 has an acceptable safety profile in heavily pre-treated NSCLC patients with and without EGFR mutations. Early signals of patient benefit were seen and a large randomized phase II study is ongoing.

e16017 Background: Cetuximab is an EGFR-blocking antibody that has been approved for the treatment of patients with head and neck squamous cell carcinoma (HNSCC) and metastatic colorectal cancer, but no predictive biomarkers of cetuximab activity have yet been identified. Methods: We report on a patient with HNSCC who had a complete tumor regression following treatment with cetuximab given as a single agent after initial surgery and radiation therapy. The EGFR protein expression level, gene copy number and sequence were assessed from both normal and tumor tissues from this patient. Results: Besides protein overexpression and gene amplification in the tumor tissue, sequencing of the EGFR gene from the patient revealed the presence of two somatic mutations, one in the kinase domain (R705G) and the other in the ligand binding domain (P546S). Cells that stably express these EGFR mutants were treated with cetuximab and their sensitivity to the drug was compared to cells expressing the wildtype gene. While the P546S mutation sensitized NIH-3T3 cells to cetuximab, R705G had a marginal effect. The double mutant (P546S/R705G) behaved the same as the P546S mutant, indicating that the mutation in the kinase domain does not contribute to the increased sensitivity to cetuximab. No mutations were found in K-RAS or B-RAF genes and there was no indication that the tumor was HPV-positive. Conclusions: Our results support a role for the P546S mutation in cetuximab sensitivity. To our knowledge, this is the first report of a somatic mutation in the EGFR ligand binding domain that may contribute to increased sensitivity to cetuximab. Other factors including EGFR copy number, EGFR over-expression and the immune response, as indicated by a very adverse side effect that correlated with the antitumor activity, may have also contributed to the observed response. It remains to be determined how frequently this mutation occurs in patients with HNSCCs and other cancers. Prospective evaluation of cetuximab anti-tumor activity in patients harboring P546S mutation needs to be clinically evaluated.

Neo-Adjuvant chemotherapy in breast cancer means that chemotherapy is used as the first therapeutic tool and that the locoregional treatment is considered as an adjuvant. The term has been contested and others were proposed, but induction treatment is too specific for the first chemotherapy combination used in leukemia to achieve complete remission. Primary treatment does not encompass the consolidation phase which belongs to our strategy so that even if it is not quite satisfactory we keep the term Neo-Adjuvant which has been coined in the seventies by E. Frei for head and neck cancers.

Background: The benefit of EGFR tyrosine kinase inhibitors (TKIs) is largely restricted to EGFR mutation-positive cancers and resistance invariably develops. A central theme of acquired resistance is persistent ErbB3 signaling, resulting in activation of the PI3K/AKT survival pathway. MM-121 is a fully human IgG1 monoclonal antibody (mAb) to ErbB3 with pre-clinical activity as a single agent and in combination with erlotinib in NSCLC, particularly in cancers with ligand-dependent activation of EGFR. This phase 1 study evaluated the safety and tolerability of MM-121 and erlotinib in NSCLC, as well as PK, immunogenicity, efficacy endpoints and exploratory biomarker evaluation. Methods: Patients with advanced NSCLC, good performance status and adequate organ function were enrolled. Patients were EGFR TKI-naïve, unless they were EGFR mutant, in which case acquired resistance was allowed. MM-121 was administered weekly and erlotinib was administered daily. Seven cohorts were enrolled, evaluating varying dose levels of the combination, as well as alternate MM-121 infusion schedules. Dose levels were determined by safety and pharmacokinetic (PK) data. Results: Between February 2010 and July 2011, 33 patients were enrolled. Median age was 64 years and there were 19 (57.5%) women. Twenty-four patients were erlotinib-naïve and 1 patient was an EGFR mutant. The most frequent adverse events were rash, diarrhea, nausea and fatigue. As of 31 July 2001, 16 patients remain on study. Full results will be presented at the meeting. Conclusions: In this phase 1 dose escalation study, MM-121 plus erlotinib was well tolerated by the majority of patients. A phase 2 study is planned. Reference: NCT00994123 Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr C27.

Optimistic results were obtained in the treatment of 39 patients with surgically incurable metastatic malignant melanoma using a regimen including 2 to 3 monthly induction cycles of cis-diamminedichloroplatinum (CDDP), recombinant interleukin-2 (rIL-2) and interferon alpha-2a (IFN alpha-2a). 33 of 39 patients were pretreated with chemotherapy (dacarbazine and/or fotemustine:31, CDDP:6) and 17 of 39 with IFN alpha-2a. Overall response rate was 54% with 13% achieving a complete response for up to 59+ weeks. Moderate to severe side-effects were reversible on rIL-2 cessation and toxicity was manageable in a routine inpatient setting. These results are especially encouraging as they were seen in previously treated patients, classically low responders, including 3 who were resistant to cisplatin or other platinum complexes. The question remains if this regimen bypasses traditional mechanisms of drug resistance.

e15065 Background: A retrospective study was performed to determine whether the site of metastasis affects the response rate (RR), progression-free survival (PFS) and overall survival (OS) of patie...

Background: Axitinib (AG-013736) is an oral, potent, and selective inhibitor of vascular endothelial growth factor receptors 1, 2, 3. The objectives of this analysis were to assess the impact of food, gender, race, weight, age, and disease status on axitinib PK and to explore the relationship between PK, diastolic blood pressure (dBP), and clinical efficacy in sorafenib-refractory mRCC pts. A relationship has been previously reported in refractory mRCC pts. Methods: Pooled PK data were analyzed using nonlinear mixed-effects modeling to estimate population PK parameters (mean and inter-individual variability; Rixe et al. ASCO Annual Meeting 2009; abstract 5045). Efficacy data from sorafenib-refractory mRCC pts (n=59) were included in the PK/pharmacodynamic analysis. Mean steady-state area under the plasma concentration-time curve (AUC) at the end of cycle 1 and dBP during axitinib therapy were utilized as predictors of clinical efficacy in the mRCC pts using logistic regression and Kaplan-Meier survival analyses. Results: Median overall survival (OS) for sorafenib-refractory mRCC pts with an AUC below the median (662 ng·hr/mL; n=29) was 42 weeks vs 84 weeks (P Conclusions: Increased exposure to axitinib and dBP ≥90 mmHg during axitinib therapy are associated with greater probability of a PR and longer median OS in sorafenib-refractory mRCC pts. These data support an ongoing, pivotal, phase III trial in previously-treated mRCC that incorporates a dosetitration scheme based on patient tolerance and BP. Citation Information: Mol Cancer Ther 2009;8(12 Suppl):B213.

5045 Background: Axitinib (AG-013736) is an oral, potent, and selective inhibitor of VEGF receptors 1, 2, 3. An association between dBP elevation and clinical outcome has been previously reported (Rini, ASCO. 2008). The objective of this pooled analysis of two phase II mRCC studies was to explore the relationship between PK, dBP, and clinical efficacy. Methods: PK data from two phase II studies in cytokine-refractory mRCC patients (pts) and 8 single-dose healthy volunteer (HV) studies were included (n = 109 mRCC pts and 240 HV) in the population PK analysis; the efficacy analysis included mRCC pts only. PK data was analyzed using nonlinear mixed-effects modeling to estimate population PK parameters (mean and inter-individual variability). Mean steady-state area under the plasma concentration-time curve (AUC) at the end of cycle 1 and the dBP during axitinib therapy were utilized as predictors of clinical efficacy in the mRCC pts using logistic regression and Kaplan-Meier analyses. Results: The median overall survival (mOS) for mRCC pts with at least 1 dBP measurement ≥90 mmHg (n = 59) during axitinib therapy was 130 weeks vs. 42 weeks (p < 0.01) for pts without any dBP ≥90 mmHg (n = 50). The mOS of pts with an AUC below the median (605 ng.hr/ml; n = 54) was 69 weeks vs. 88 weeks (p > 0.05) for pts with an AUC above the median (n = 55). Among pts with dBP ≥90 mmHg, mOS was 120 weeks and 131 weeks (p > 0.05) for pts with AUC below and above the median (n = 23 and 36), respectively. Among pts without dBP ≥90 mmHg, mOS was 42 weeks and 43 weeks (p > 0.05) for pts with AUC below and above the median (n = 31 and 19), respectively. An 82% increase in probability of a partial response was predicted for a 10 mmHg higher dBP during therapy. There was no apparent correlation between the AUC and maximum dBP during axitinib therapy. Conclusions: dBP ≥90 mmHg during axitinib therapy is a strong predictor of clinical efficacy in patients with mRCC, and is not merely a reflection of higher axitinib drug levels. These data support an ongoing pivotal phase III trial in previously treated mRCC that incorporates a dose-titration scheme based on patient tolerance and BP. [Table: see text]

e16137 Background: Pancreas represents an uncommon site of metastasis in renal cell carcinoma (RCC). The natural history of pancreatic metastases is largely unknown, and efficacy of targeted agents has never been assessed. We examined the outcome in a series of RCC patients (pts) with pancreatic metastases, treated by either surgery or antiangiogenic therapy. Methods: We reviewed the charts of RCC pts treated for pancreatic metastases, between 2001 and 2008, in Institut Gustave Roussy and Pitié-Salpetrière Hospital. Data base was reviewed to determine presentation, clinical symptoms, pancreatic metastases treatment, use of systemic therapy, disease-free survival (DFS) or progression-free survival (PFS) and overall survival (OS) after treatment. Results: A total of 40 pts with pancreatic metastases from RCC have been analyzed. Median age was 63 years (range: 48–81), sex ratio M/F was 29/11, and metastases occurred initially (synchroneous) in 9 pts, while they were metachronous in 31 pts. Most of the metastases were asymptomatic (87%), detected on routine CT scans, while only 12% were symptomatic (pain, cholestasis). Surgical resection of pancreatic metastases was performed in 8 (20%) patients, and DFS and OS were 45 and 66 months. In patients with multiple metastatic sites (lung : 20 (50%); liver : 8 (20%); thyroid : 3(7%)), therapeutic approaches have included antiangiogenic therapies (75%): sorafenib (20 pts), sunitinib (7 pts), bevacizumab (2 pts), Torisel (1 pt). Best response was partial response in 30% of pts and stable disease in 50% of pts. Median PFS and OS were 21 months and 30 months. Conclusions: Patients with RCC pancreatic metastases seem to have a longer survival than the usual metastatic RCC pts. Surgical resection must be discussed in pauci-metastatic disease, as it can be associated with prolonged survival. Antiangiogenic therapies are active in these patients with high response rate and prolonged survival. No significant financial relationships to disclose.

16146 Background: Sunitinib (SU) is an angiogenesis inhibitor (AI) recently approved and widely used for the treatment of mRCC. This oral multi-targeted kinase inhibitor blocks the VEGF and PDGF re...

9006 Background: No treatment has extended survival in stage IV melanoma. Melanomas are highly angiogenic and metastatic, express high levels of vascular endothelial growth factor receptors (VEGFRs...

3557 Background: Aflibercept (AF), a recombinant fusion molecule of the human VEGF receptor extracellular domains and the Fc portion of human IgG1, is a potent inhibitor of vascular endothelial growth factor. This study assessed the safety, dose limiting toxicities (DLTs)/recommended dose (RD) and PK of AF+I-LV5FU2. Methods: DE of AF followed by I-LV5FU2 (fixed standard dose) every 2 weeks (q2w) were explored. Results: 38 pts were enrolled (M/F: 18/20, median age 53 [19–77], ECOG PS 0/1: 24/13) including colorectal (23) and ovarian (5) cancers. Thirty six pts received prior chemotherapy (CT): median 3 [1–5]. A total of 404 cycles (median 10 [1–26]) of AF+I- LV5FU2 were administered across 4 AF dose levels (DL) ranging from 2.0 to 6.0 mg/kg. DLTs were: 2/12 pts at 4.0 (Grade (Gr) 3 proteinuria (PrU)>2wks including 1 pt with thrombotic microangiopathy), 2/10 pts at 5.0 (Gr 3 stomatitis and esophageal reflux) and 2/12 pts at 6.0 mg/kg (febrile neutropenia, Gr 3 stomatitis). The main AF-related Gr3–4 toxiciti...

Microvascularization modifications should precede tumor size‐changes during anti‐angiogenic therapy. We applied contrast functional ultrasound imaging (fUSI) to detect changes in Wilms tumors with anti‐angiogenic treatment (Bevacizumab). Human Wilms tumor cells was grafted in left kidney of 32 mice. Once tumors had >5mm diameter, mice received : placebo, N=14; Bevacizumab for 21days, N=11; and Bevacizumab for 10days followed by placebo for 11days, N = 7. On days ‐1, +1, +9, +14 and +21 with respect to treatment start, fUSI was performed (CPS mode, SonoVue). Linear time intensity curves were obtained from regions in kidney cortex and matched‐depth of tumor for first bolus passage and 50s following acoustic destruction of contrast. Excised tumor weight decreased with increased treatment duration: 3.7+/‐1.8 g (placebo), 2.3+/‐1.9 g (Bevacizumab‐10days, placebo‐11days), 1.4+/‐0.7 g (Bevacizumab‐21 days) [p

4551 Background: The current standard of care for patients (pts) with advanced pancreatic cancer (APC) is gemcitabine (GEM)- based chemotherapy. Axitinib (AG) is a potent inhibitor of vascular endothelial growth factor receptors. A phase I study of AG in solid tumors identified 5 mg BID as the therapeutic starting dose. The main objectives of this trial are to determine whether the overall survival (OS) of the combination of AG and GEM is superior to that of GEM alone as first-line therapy in pts with APC, and to determine doses of AG and GEM that can be safely administered together. Methods: 8 pts were treated on the phase 1 (P1) portion of this trial. For the phase 2 (P2) portion of the trial, 103 pts with locally advanced or metastatic disease, no prior GEM or VEGF/VEGFR inhibitors, PS 0–2 were randomized (2:1) to GEM 1,000 mg/m2 over 30 minutes on days (D) 1, 8, 15 every 28 D with or without AG at a starting dose of 5 mg po BID between Jan 06 and Aug 06. CT scans were performed every 2 cycles. Results: Data from the evaluable P1 pts (n=5) indicated that PK of GEM and AG appeared to be unchanged in combination. The GEM mean plasma exposure (AUC) was 11,348 alone vs 12,840 ng.h/mL with AG; similarly mean steady-state AG AUC was 250 alone vs 270 ng.h/mL with GEM. 3 of 8 pts had a PR. The median number of days on study drug was 158 days (range: 57–330 days) with 2 pts ongoing. Data from the randomized P2 pts: male 48%; median age 62 (36- 81); PS 0/1 vs 2: 91% vs 9%; 41% with locally advanced disease; 59% with metastatic disease. The most commonly reported adverse events are anemia (48%), alk phos elevations (48%), leukopenia (45%), neutropenia (42%), LFT elevations (39%), and thrombocytopenia (27%). The most common non-hematologic adverse events are nausea (24%), vomiting (20%), fatigue (19%), diarrhea (18%), anorexia (18%), constipation (13%), dyspnea (12%), and pyrexia (12%). 55 events have been reported in the 103 P2 pts. The pooled median OS is 203 days with a 95% CI of (158, not estimable). The median follow-up time is currently 223 days. Conclusions: AG can be safely administered at a starting dose of 5 mg BID in combination with standard dose GEM in patients with APC. Final safety and OS results by treatment arm will be presented. [Table: see text]

5024 Background: In a randomized phase III trial of patients (pts) with mRCC, sunitinib demonstrated a significant improvement in progression-free survival (PFS) and objective response rate (ORR) compared to IFN-a as first-line therapy (Proc ASCO 2006;24:2s [Abstract LBA3]). We present the most recent data from this trial and an analysis of prognostic factors. Methods: Untreated pts with clear-cell mRCC were randomized 1:1 to receive either sunitinib (repeated 6-week cycles of 50 mg/day orally for 4 weeks, followed by 2 weeks off treatment) or IFN-a (9 MU given subcutaneously three times weekly). The primary endpoint was PFS. Results: A total of 750 pts were randomized: 375 to sunitinib, 375 to IFN-a. The median duration of treatment is 11 months (range: No significant financial relationships to disclose.

5566 Background: High EGFR expression occurs in 35–70% of primary ovarian tumors and is often associated with poor prognosis. This Phase II, open-label, non-comparative multicenter study investigated the efficacy and tolerability of gefitinib (Iressa) in combination with paclitaxel (P) and carboplatin (C) for second-line treatment of patients (pts) with ovarian, tubal or peritoneal adenocarcinoma. Methods: Women (>18 years) with platinum-resistant/refractory (relapsed 6 months) disease were enrolled. Pts received gefitinib (500 mg/day), P (175 mg/m2) and C (AUC 5) every 3 weeks for 6–8 cycles, after which pts could continue to receive gefitinib. The primary endpoint was objective response rate (ORR) assessed using RECIST or Rustin criteria. Results: Sixty-eight pts (26 resistant/refractory and 42 sensitive) were enrolled (median age [range]: 57 [34–72] years; ECOG performance status 0/1/2: 41/26/1). ORR and disease control rates were 19.2% and 69.2%, respectively, for resistant/refractory; and 61.9% and 81.0%, respectively, for sensitive pts (see table ). Grade 3/4 toxicities (in =10% pts) were neutropenia (59%), diarrhea (25%), leukopenia (22%), anemia (13%), and acne (13%). Two myelodysplastic syndromes (MDS) and one acute biphenotypic leukemia were observed during treatment. Another pt developed MDS 34 months after study treatment discontinuation. Conclusions: Gefitinib (Iressa) in combination with P and C has promising activity as second-line treatment for ovarian, tubal or peritoneal adenocarcinoma and is generally well tolerated. The hemopathies are under further investigation. [Table: see text] No significant financial relationships to disclose.

9121 Background: The IRMA study reported the high prevalence of renal insufficiency (RI) in 4684 solid tumour patients, with a glomerular filtration rate (GFR) 110 μmol/L was assessed. GFR was estimated with Cockcroft-Gault (CG) and abbreviated MDRD (aMDRD) formulae. Drugs necessitating dosage adjustment and those potentially nephrotoxic were identified. Chi-square test was used to compare the prevalence of RI between patients with BM and patients without, for all patients and for breast cancer (BC) ones. Results: 1,000 patients (BC 577) with BM were included: median age 60, mean 59.8, weight 66 kg, 659 women. The prevalence of SCR>110 μmol/L was 8.3%. That of GFR [Table: see text] No significant financial relationships to disclose.

LBA3 Background: Two multicenter phase II trials of 2nd line monotherapy with sunitinib (SU11248) in patients (pts) with mRCC showed a response rate of approximately 40% (JCO 2006;24:16–24; Proc ASCO 23, 380s). This international, randomized phase III trial compared the efficacy and safety of sunitinib to IFN-α in treatment naïve pts with mRCC. Methods: Untreated pts with clear-cell mRCC were randomized 1:1 to receive sunitinib (6-week cycles: 50 mg orally once daily for 4 weeks, followed by 2 weeks off) or IFN-α (6-week cycles: subcutaneous injection 9 MU given three times weekly). The primary endpoint was progression-free survival (PFS). Secondary endpoints included objective response rate, overall survival, and adverse events. Based on a planned sample size of 690 patients, the trial was designed to have 90% power to detect a 35% improvement in median PFS from 20 weeks to 27 weeks (4.6 months to 6.2 months; 2-sided unstratified log-rank test; significance level 0.05). Results of a planned analysis on the primary endpoint, PFS, are presented in this report. Results: From Aug 2004 to Oct 2005, 750 patients were randomized: 375 to sunitinib, 375 to IFN-α. Baseline characteristics were well balanced, and included pooled median age = 60 and prior nephrectomy = 90%. Median PFS assessed by third-party independent review was 47.3 weeks (95% CI 40.9, not yet reached) for sunitinib vs. 24.9 weeks (95% CI 21.9, 37.1) for IFN-α [hazard ratio 0.394 (95% CI 0.297, 0.521) (p < 0.000001)]. The objective response rate by third-party independent review was 24.8% (95% CI 19.7, 30.5) for sunitinib vs. 4.9% (95% CI 2.7, 8.1) for IFN-α (p < 0.000001). The objective response rate by investigator assessment was 35.7% (95% CI 30.9, 40.8) for sunitinib vs. 8.8% (95% CI 6.1, 12.1) for IFN-α (p < 0.000001). 632 pts (85%) are alive, with 49 deaths on sunitinib arm and 65 deaths on IFN-α arm. 8% withdrew from the study due to adverse event on sunitinib arm vs. 13% on IFN-α arm. Conclusions: These results demonstrate a statistically significant improvement in PFS and objective response rate for sunitinib over IFN-α in first-line treatment of pts with mRCC. [Table: see text]

2581 Background: MUC1 is a glycoprotein often over-expressed and underglycosylated in renal cell carcinoma (RCC) making it an attractive antigenic target for tumor-specific immunotherapy. TG4010 is a cancer vaccine based on a modified vaccinia virus, strain MVA, expressing both MUC1 and Interleukin 2 (MVA-MUC1-IL2). The objective of this phase II, non-randomized study was to determine the efficacy of TG4010 alone and in combination with cytokines. Methods: Thirty seven patients (pts) with progressive metastatic RCC expressing MUC1 in at least 50% of the tumour cells were treated by subcutaneous injections of TG4010, 108 pfu/inj weekly, for 6 weeks then every three weeks until progression. At progression, TG4010 was continued in combination with Interferon α2a (INF) and Interleukin 2 (IL2). Results: Treatment efficacy and toxicities were previously presented at ASCO 2005 (abstr 4653). No objective responses have been observed, however, 7 pts (19%) remained stable for more than 6 months with TG4010 alone, 3 of them more than 22 months. After progression on TG4010 alone 22 pts received TG4010 in combination with cytokines. Six pts (27%) have been stabilized more than 6 months. The median TTP were 2.6 months (95% CI, 2.4–2.9 months) for TG4010 alone and 3.5 months (95% CI, 0.2–6.7 months) for the combined treatment. There were 24 deaths, with a median OS of 19 months (95% CI, 10–27.9) for the whole population. Seven pts were treated by sorafenib after immunotherapy failure. After censoring pts at the introduction of sorafenib, the median OS was 16 months (95% CI, 6–26), with 41% of pts alive at 2-years. The most frequent adverse effects related to TG4010 were minor to moderate injection site reactions, fatigue and flu-like symptoms. Twelve out of 24 pts evaluable for MUC1 ELISpot show evidence for MUC1-specific CD8+ T cell response while 14 out of 21 evaluable for MUC1 specific T cell proliferation were responsive. Conclusions: The cancer vaccine TG4010 alone and in combination with IL2 and INF induces some disease stabilizations in pts with progressive metastatic RCC and can improve survival in a population selected for MUC1 positivity, which is a factor of poor prognosis. [Table: see text]

6078 Background: The EDIFICE study aimed to allow better understanding of population’s adhesion to the tests available for the 4 most frequent cancers: breast, colorectal, prostate and lung. In 1998 the French National Consensus Conference advocated for mass cancer screening (CS) using Hemoccult II. The departments are divided according to the existence or not of an organized program: this screening was organized in 22 departments (3 “scout” started in 1998, 9 “first wave” in 2003 and 12 “second wave” in 2004). Results are reported hereunder. Methods: This first nationwide observational study was carried out in France from January 18th to February 2nd, 2005 among a representative sample of 1504 subjects aged between 40 and 75 years and a representative sample of 600 general practitioners (GPs). Information about participating subjects included socio-demographic characteristics, attitude towards CS, and about GPs’ medical practice regarding CS. Results: Only 25% of the 970 subjects aged between 50 and 74 years had undergone a colorectal screening test. In the organized departments (OD), the rate of persons who self-report any colorectal cancer screening was 34% vs 20% in unorganized departments (UD) (OR=1.99, CI95% 1.47- 2.69, p95% 1.91- 5.88, p95% 1.11–3.49). The rate of GPs who advocated systematically for screening was 40%, 29%, 26% and 13% for scout, first wave, second wave OD and baseline UD respectively. Organization reduced the rate of screening based on colonoscopy alone from 69% persons in UD to 35% in the OD. Conclusions: The main result of this survey comparing data in the same frame of time, in the same country, is that organized programs for colorectal cancer impact of health outcomes. No significant financial relationships to disclose.

18004 Background: The presence of metastases is a strong indicator of poor survival in many types of cancer. It has recently been shown that vascular endothelial growth factor-C (VEGF-C), and its receptor Flt-4 (VEGF-R3) are increased in a variety of tumours and may play a pivotal role in the promotion of metastasis. We previously demonstrated that high soluble VEGF-C level was correlated to high tumor burden. Objectives: This study was designed to i) detect and evaluate whether soluble VEGF-R3 play a role in metastatic malignant melanoma patients ii) to determine if they have any relationship with clinicobiological parameters, clinical response and survivals. Methods: Using a sensitive enzyme-linked immunosorbent assays, VEGF-R3 was retrospectively measured in sera of 60 patients with a fully documented history of melanoma disease in comparison with 30 healthy controls. Disease free survival (DFS) and overall survival (OS) were calculated from the beginning of treatment until either the progression (DFS) or death (OS) and analyzed using the Kaplan-Meier method. Results: Pretreatment circulating VEGF-R3 was detectable in all samples from either melanoma patients or healthy donors. Furthermore, median level of sVEGF-R3 was significantly higher (p = 0.000015) in melanoma patients as compared to healthy donors. (38890 vs 30773 pg/ml respectively). No significant association was noted between sVEGF-R3 levels and gender, age or LDH level. Median serum VEGF-R3 level was significantly higher in patients with high tumor burden as compared to patients with low tumor burden (p = 0.045). The pretreatment sVEGF-R3 level was significantly different (p = 0.025) between responder (n = 27) and non-responding patients (n = 33). Lastly, the relapse-free survival was higher in the group with low sVEGF-R3 concentration compared to the high one’s (14.1 vs 11,9 months) as well as for OS (14.3 Vs 12.6 months) but theses differences were not significant (χ2 = 2,30, p = 0.12 & χ2 = 0.74, p = 0.37 respectively). Conclusion: these results suggest that high pretreatment sVEGF-R3 level is related to bad prognosis in melanoma patients. No significant financial relationships to disclose.

8049 Background: Vascular endothelial growth factor (VEGF-A) and its circulating form (sVEGF-A) play a major role in tumor progression and angiogenesis. However, the clinical implications of circulating VEGF-A in metastatic malignant melanoma patients and its role in tumor-associated lymphangiogenesis and lymphatic metastasis remained unclear. Objectives: The aim of this study is to evaluate the potential role of circulating VEGF-A levels in lymphangiogenesis and its usefulness in metastatic malignant melanoma patients. Methods and Patients: using a sensitive enzyme-linked immunosorbent assays, VEGF-A level was measured in sera of 65 patients treated by biochemotherapy with a fully documented history of disease in comparison with 30 healthy controls. Results: A wide range (0- 566 pg/ml) of pretreatment sVEGF-A level was detected in the serum of MMM patients and the median level (40.5 pg/ml) was significantly higher (p=0.0007) than that of healthy donors (7.5 pg/ml). Age, gender and LDH were not associated with sVEGF-A levels. Regarding tumor burden, median pretreatment sVEGF-A level was higher in patients with high tumor burden (n= 40) as compared to patients with low tumor burden (n=25). Furthermore, the most surprising finding of our study was that patients with only lymph node metastasis (n=23) had a significantly (p= 0.021) higher median sVEGF-A levels (45.7 pg/ml) as compared to the 42 patients with different metastatic sites (16.6 pg/ml). Conclusion: these results suggest that sVEGF-A levels may predict lymph node metastasis in metastatic malignant melanoma patients. In addition, it may be promising targets for new therapeutic strategies in melanoma disease. No significant financial relationships to disclose.

13161 Background: VEGF Trap is a potent angiogenesis inhibitor comprising portions of human VEGF receptor VEGFR1 (Flt-1) and VEGFR2 (Flk-1) extracellular domains fused to the Fc portion of human IgG. VEGF Trap binds VEGF and neutralizes all VEGF-A isoforms plus placental growth factor. I-LV5FU2 is an approved chemotherapy regimen for the first-line treatment of metastatic colorectal cancer. This study was designed to evaluate the safety and pharmacokinetics (PK) of VEGF Trap plus I-LV5FU2 administered intravenously. Methods: Successive cohorts of 3–6 patients (pts) with relapsed or refractory solid tumors received intravenous VEGF Trap plus I-LV5FU2 every 2 weeks. Study endpoints included safety, PK, and immunogenicity. Antitumor activity was assessed by CT scan. Results: Ten pts (3 male/7 female), median age 59 (34–67), ECOG PS 0/1/2: 8/2/0, with a variety of advanced solid tumors, including 5 colorectal and 2 ovarian, have received a total of 50 cycles of VEGF Trap plus I-LV5FU2 across 2 VEGF Trap dose levels (2.0 mg/kg, 4.0 mg/kg) to date. Only 1 dose-limiting toxicity (G3 proteinuria >2 wks with normal plasma creatinine levels) has been encountered so far (4.0 mg/kg, Cycle 2). This pt also had controlled G2 HTN (renal biopsy pending). No other G3/4 VEGF Trap-related AEs have been reported so far. Preliminary free VEGF Trap clearance was 15.4 mL/kg/day. Three pts (synovial sarcoma, ovarian and colon cancers) achieved partial responses. Conclusions: VEGF Trap may be safely combined with I-LV5FU2 at the dose levels studied. Preliminary free VEGF Trap clearance did not differ significantly from that seen with single-agent exposure; chemotherapy PK analysis is pending. The maximum tolerated dose has not yet been reached, and dose escalation continues. Updated safety, pharmacokinetic, and preliminary efficacy results from this ongoing study will be presented. [Table: see text]

6092 Background: The EDIFICE study was the first which aimed to allow better understanding of population’s adhesion in France to the screening tests available for the 4 most frequent cancers: breast, colorectal, prostate and lung. Methods: This nationwide observational study was carried out in France from January 18th to February 2nd, 2005 among a representative sample of 1,504 subjects aged between 40 and 75 years and a representative sample of 600 general practitioners (GPs). Information collected about participating subjects included socio-demographic characteristics, attitude towards cancer screening and actual experience of cancer screening, and about GPs’ medical practice regarding screening. Results: A striking result observed is that screening for colorectal cancer that USPSTF rates as an “A” Recommendation (strongly recommended) deserves less attention: reported rate 25% than prostate cancer screening (either with digital rectal examination and/or PSA): reported rate 36%. In the same way, GPs’ are recommended more often prostate cancer screening (58%) than colorectal cancer screening (18%). Another key result of that EDIFICE national representative survey is the high attendance rate of women at mammography screening. Conclusion: With respect to the cost of such programs, utilization must be monitored and compared among different countries/organization. Additional data will be presented about reasons to do or not cancer screening and about regularity of screening. [Table: see text] No significant financial relationships to disclose.

13092 Background: Axitinib (AG-013736) is a novel small molecule inhibitor of the receptor tyrosine kinases with picomolar potency against VEGFR 1, 2 & 3 and nanomolar potency against PDGFR-beta and KIT. A phase I study in solid tumors identified 5 mg BID as the therapeutic dose; a phase II study in renal cell cancer demonstrated significant efficacy with a response rate (RR) of 46% (Rini et al, ASCO 2005). This study examined the safety, PK and preliminary efficacy of AG-013736 (AG) in combination with gemcitabine (GEM) as first-line therapy for advanced pancreatic cancer. Methods: A randomized phase II study was preceded by a phase I component. All patients (pts) in the phase I portion received 1000 mg/m2 GEM by 30-minute infusion on days 1, 8, and 15 followed by one week of rest from treatment. AG 5 mg p.o. BID was given beginning Cycle 1, Day 3 (C1D3). Eligible pts had no prior chemotherapy for advanced disease, ECOG 0–2, and no previous treatment with VEGF/VEGFR inhibitors, or anti-angiogenesis treatment. Full PK profiles were collected on C1D1 (GEM alone), C1D14 (steady state, AG alone), and C1D15 (GEM + AG). In the phase II trial, pts are randomized to AG or AG plus GEM beginning C1D1. Results: 8 pts were treated on the phase I portion of this trial. Toxicity: The primary Gr. 3/4 toxicity was hematologic: Gr. 4 anemia and Gr.3 thrombocytopenia in 1 pt and Gr. 3 neutropenia in 1 pt requiring a dose reduction for GEM in Cycle 3. Gr. 2 non-hematologic adverse events include pruritus (1 pt), abdominal pain (2 pts), epigastric pain (1 pt), melena (1 pt), and asthenia (2 pts). Gr. 2 hypertension was observed in 3 pts. Efficacy: Radiological assessment suggests 2 pts with partial response and 4 pts with stable disease: response assessments are ongoing. The median number of cycles is 3 [1,6]. Treatment for 4 pts is still ongoing: Cycle 6 (2 pts) and Cycle 2 (2 pts). Conclusions: This combination is safe and appears to be an effective treatment for advanced pancreatic cancer with significant tumor regression observed in 2 pts. Therapy was well tolerated with manageable toxicity. Additional investigation of AG-013736 in combination with GEM in the phase II setting for advanced pancreatic cancer is warranted. [Table: see text]

10701 Background: The EDIFICE study aimed to allow better understanding of population’s adhesion to the screening tests available for the 4 most frequent cancers: breast, colorectal, prostate and lung. Implementation of breast cancer screening (CS) was generalized since 2003 in France: women aged between 50 and 74 years are invited to do a mammography (M) every second year. Methods: This nationwide observational study was the first implemented in France (from January 18th to February 2nd, 2005) among a representative sample of 1504 subjects aged between 40 and 75 years including 773 women and a representative sample of 600 general practitioners (GPs). Information about participating subjects included socio-demographic characteristics, attitude towards CS and actual experience of CS, and about GPs’ medical practice regarding CS. Results: Among the 507 participating women (PW) aged between 50 and 74 years, 93% had done at least one M: 55% underwent this test on their own initiative and 45% of PW had it during a systematic screening plan. Most PW in the systematic screening (89%) had a M dating from less than 2 years vs 74% of the others (X2 = 18.9; p < 0,01). Main reasons for not performing the biennial screening test were: lack of care (18%), “it’s not a priority” (18%) and no advice from the GPs (15%). 79% of the PW had at least one M before 50 years (1st test at the mean age of 37.9±6 years). Among the 600 GPs, 68% systematically recommended M to their patients. GPs’ perceptions of the reasons for women’s avoidance of screening test were their unwillingness to be aware of M results (44%) and their belief in painful symptoms related to M (52%). Conclusions: This first nationwide study has shown the high rate of PW’s attendance at M screening. It pointed out that systematic and organized screening played a major role in regularity of screening over 2 years, and GP is a key actor in heightening public awareness. No significant financial relationships to disclose.

3685 Background: This trial was conducted to establish potential PK interactions between X and O when administrated in combination (XELOX regimen) and to evaluate neurotoxicity when O is administer...