Search

e21601 Background: MBD4 encodes a glycosylase implicated in repair of lesions induced by DNA deamination and its inactivation in tumors is associated with a hypermutated phenotype. Uveal melanoma (UM) is a rare but aggressive form of melanoma carrying an extremely low mutation burden. Although metastatic UM (mUM) are usually resistant to immune checkpoint inhibitors (ICI), the first reported MBD4-mutated (MBD4m) patient responded to ICI, suggesting that MBD4 mutation may predict response to ICI. Since then, other MBD4-inactivated UMs, acute myeloid leukemias, colorectal adenocarcinomas, gliomas and spiradenocarcinoma cases have been reported. Methods: Retrospective cohort of mUM patients treated with ICI. MBD4 was sequenced in a subset of these patients. Results: Three hundred mUM patients were analyzed. Median follow-up was 17.3 months. Ten objective responses and 20 stable disease for > 12 months were observed, corresponding to an objective response rate of 3.3% and a clinical benefit (CB; i.e. responder patients and stable disease) rate of 10%. Median progression-free survivals (PFS) in patients without and with CB were 3.1 and 16.3 months, respectively (p < 0.0001). Of the 134 tumors sequenced for MBD4, five (3.7%) were mutated. MBD4 inactivation was associated with higher mutation burden (179 to 643 variants versus a median of 16 in MBD4 wild-type), better objective response rate as 60% of MBD4m versus 4% of MBD4-wild type patients responded (Fisher’s exact p-test = 0.0009). Median PFS was 4.0 months in MBD4-wild type and 22.3 months in MBD4m patients (HR = 0.22; p = 0.01). Median overall survival was 16.6 months in MBD4-wild type and unreached in MBD4m patients (HR = 0.11; p = 0.004). Conclusions: In mUM patients, MBD4 mutation is highly predictive for response to ICI, PFS and overall survival benefit. MBD4 could be a tissue-agnostic biomarker and should be sequenced in mUM and other tumor types where MBD4 mutations are reported.

90 Background: Darolutamide, a structurally distinct and highly potent androgen receptor inhibitor, had a consistently favorable safety and tolerability profile in patients with nonmetastatic castration-resistant prostate cancer in ARAMIS, and discontinuation rates due to adverse events (AEs) remained consistently low after longer follow-up. In previously reported phase 1/2 studies in patients with metastatic castration-resistant prostate cancer (mCRPC), darolutamide was well tolerated for up to 25 months. We report the safety and tolerability of extended treatment with darolutamide in these mCRPC studies. Methods: Data from three phase 1/2 studies (NCT02363855, NCT01317641/NCT01429064, and NCT01784757) in patients with mCRPC were pooled for analysis of long-term safety. Results are summarized descriptively. Results: Of 173 patients with mCRPC evaluable for safety in the 3 studies, 13 patients received > 2 years of darolutamide treatment; median duration of treatment was 38 months (range 24–90). Six patients received > 4 years of darolutamide treatment. Nine of 13 patients with > 2 years of darolutamide therapy initially received 600 mg BID; others received 200 mg (n = 1), 300 mg (n = 1), or 700 mg (n = 2) BID. Median age was 68 years in the > 2-years group, and most patients were white. Patients were enrolled from France (n = 4), Latvia (n = 3), United Kingdom (n = 3), Finland (n = 2), and Japan (n = 1). Most patients had normal renal function (n = 10) and hepatic function (n = 10), 12 patients had Eastern Cooperative Oncology Group performance status of 0, and 9 patients had a Gleason score ≥7. Prior systemic anticancer treatments taken by more than 1 patient were bicalutamide (n = 10); cyproterone (n = 3); and triptorelin and leuprorelin (n = 2 each). No patient received prior chemotherapy. All 13 patients reported AEs, most grade 1 or 2; 6 reported grade 3 AEs. Treatment-related AEs were reported in 5 patients, and all were grade 1 in severity. Serious AEs were reported in 6 patients, and all were judged not related to darolutamide. One patient treated with darolutamide for > 2 years had an AE (new neoplasm; not related to darolutamide) that led to treatment discontinuation. Conclusions: Extended darolutamide treatment > 2 years in this small group of patients with mCRPC was generally well tolerated, with safety and tolerability consistent with that previously reported. Clinical trial information: NCT02363855, NCT01317641/NCT01429064, and NCT01784757.

2579 Background: In preclinical studies, the combination of anti-VEGF/Ang2 and anti-PD-1 therapy has been shown to promote an immunopermissive state, which is supportive of T-cell-mediated tumor cell destruction. BI 836880 is a humanized bispecific nanobody that targets VEGF and Ang2, and ezabenlimab (BI 754091) is an anti-PD-1 antibody. Phase I studies investigating each as monotherapies have reported safety and preliminary antitumor activity. This ongoing Phase Ib study is evaluating the combination of BI 836880 and ezabenlimab in pts with advanced solid tumors. In Part 1 (dose escalation), the combination was feasible in pts with advanced NSCLC, with a recommended Phase II dose (RP2D) of BI 836880 720 mg + ezabenlimab 240 mg IV q3w. Here, we report updated results from Part 2 (expansion phase), which is assessing the antitumor activity and safety of the RP2D. Methods: Seven cohorts are currently recruiting pts in Part 2: metastatic (m) NSCLC after checkpoint inhibitor (CPI) monotherapy (Cohort A); mNSCLC after chemotherapy (CT) + CPI (Cohort B); mSCLC after CT ± CPI (Cohort C); 1st and 2nd recurrences of glioblastoma (GBM; Cohort D); immunotherapy-resistant m-melanoma (Cohort E); hepatocellular carcinoma (HCC) after prior sorafenib or lenvatinib ± CPI (Cohort F); and previously untreated/unresectable HCC (Cohort G). Primary endpoint is objective response rate (complete response + partial response [PR]). Results: As of January 2021, 196 pts have received BI 836880 plus ezabenlimab (14 in Part 1, 182 in Part 2 [Cohort A, 26; B, 30; C, 19; D, 31; E, 32; F, 28; G, 16]). 134 (68%) pts were male, median age was 63 years and 102 (52%) had prior CPI use. Any grade and ≥G3 adverse events (AEs; any cause) were reported by 160 (82%) and 62 (32%) pts, most commonly (all%/≥G3%) hypertension (20/8), asthenia (20/3), diarrhea, decreased appetite, and nausea (all 11/1). 95 (48%) pts had a drug-related AE, most commonly hypertension and asthenia (both 11%). 6 pts had a G4 AE (non-related: hyperkalemia + cardiac arrest, laryngospasm, gastrointestinal perforation; drug-related: anaphylactic reaction, acute pancreatitis, transaminases increased); 8 pts had a G5 AE (non-related: general physical health deterioration, epilepsy, hemoptysis, cardio-respiratory arrest, hepatic failure, intracranial hemorrhage, COVID-19 pneumonia; drug-related tracheal hemorrhage). 30 (15%) pts had immune-related AEs (3% ≥G3), including hypothyroidism (3%). 11 (6%) pts had an AE leading to discontinuation. Overall, 145 pts were evaluable for response: 9 pts achieved confirmed PR (2 pts in Part 1 and 7 in Part 2 [NSCLC, n = 3; SCLC, n = 1; GBM, n = 1; melanoma, n = 1; and 2nd-line HCC, n = 1]), 87 pts had stable disease and 49 pts had progressive disease. 111 pts remain on treatment. Conclusions: BI 836880 plus ezabenlimab had a manageable safety profile. The combination showed preliminary antitumor activity in a range of tumor types. Clinical trial information: NCT03468426.

10031 Background: Obesity is an established risk factor for several cancers and higher body mass index (BMI) is associated with poor prognosis. These data are still debated in melanoma. Furthermore, recently the concept of “obesity paradox” has emerged. In a large cohort published by McQuade JL et al, higher BMI was associated with better survival in patients with metastatic melanoma (MM) especially for those treated with targeted therapy (TT) and immune checkpoint inhibitors (ICI). We studied the association between BMI and progression-free survival (PFS) and overall survival (OS) in patients with MM treated with systemic therapies. Methods: This study was conducted from the prospective MelBase cohort (NCT02828202). Patients with MM treated with first-line ICI, TT, or CT were included. BMI was categorized by WHO criteria. Underweight patients were excluded. The co-primary outcomes were the associations between BMI and PFS or OS, stratified by treatment, sex and age. Multivariate analyses were performed. Results: A total of 1214 patients were analyzed. The majority of them were treated with ICI, followed by TT. Obese patients represented 22% of cohort (Table). Median follow-up was 13.5 months. The patients who were overweight or obese did not have different PFS (p = 0.88) or OS (p = 0.25) than patients with normal BMI. Stratifying this cohort by treatment received, age, sex and others parameters (such as LDH, number of metastatic site) did not revealed any difference. Multivariate analysis did not change the results. Conclusions: BMI was not associated with clinical outcomes in our cohort, especially in ICI and TT groups. Thus, we did not confirm the results presented by McQuade Jl et al. with a cohort quite similar in term of size. Because BMI is too simplistic and then an imperfect measure of body composition, the published data are not reproducible. We caution the oncologists, about BMI as valuable predictive marker of survival for melanoma patients. [Table: see text]

9545 Background: Among mechanisms of resistance to BRAF inhibitors (BRAFi), cell cycle effectors including CDK4 have been involved in ERK reactivation. In this phase I-II open label study, we aimed to establish the Maximum Tolerated Dose (MTD) of PD0332991, an inhibitor of CDK4/6, when added to vemurafenib (VM) in metastatic melanoma patients. Methods: Patients with BRAFV600E/K mutated metastatic melanoma harbouring CDKN2A loss and RB1 expression were included. Patients were treated with a 14 days followed by 7 days rest daily dosing schedule of PD0332991 + continuous BID dosing of VM, and stratified into 2 groups according to previous BRAFi treatment (no group 1, yes group 2). Dose levels (PD0332991 (mg/QD)/VM (mg/BID) ranged from 25/720 to 200/960. The primary endpoint was the occurrence of a DLT within the first 2 cycles of therapy. Secondary endpoints included best response (RECIST), OS, PFS, pharmacokinetics parameters, tumour molecular profiling on baseline lesions using transcriptomic and NGS analysis. Results: Nineteen patients were enrolled, among them 16 (84%) in group 2, with 18.5 months median follow-up. Characteristics at baseline were: male 11 (58%), median age 54.4 years, unresectable stage IIIC 2 (11%), stage IV 17 (89%), M1C 12 (67%), high LDH 9 (47%), median time from advanced melanoma diagnosis to inclusion 26.8 months, ≥ 2 lines therapy 13 (68%). A DLT was observed for 1 and 5 patients in group 1 and 2 respectively, defining the MTD at PD0332991 25mg and VM 960mg in group 2. No significant evidence for drug-drug interaction between PD0332991 and VM was highlighted. In group 2, ORR was estimated to 4 (25%), SD to 8 (50%), median PFS to 9.3 months and median OS to 13.2 months. Baseline transcriptomic analysis revealed high alteration rate associated with clinical response and enrichment in genes related to MAPK, cell cycle and apoptosis pathways. Conclusions: While combination of fixed dose of PD0332991 + VM did not allow us to increase PD0332991 dosage above 25mg, significant clinical benefit was achieved in heavily pretreated patients; baseline molecular analysis revealed an association between transcriptomic data and clinical response. Clinical trial information: NCT02202200.

9520Background: Brain metastasis commonly occur in patients with metastatic melanoma (MM) and are managed with surgery or stereotactic radiosurgery (SRS) and/or systemic treatment based on BRAF sta...

e21536Background: Melanoma’s incidence is increasing, and elderly people could be significantly impacted since more than 50% of melanoma arise in > 65 years-old (yo). Combined BRAF and MEK targeted...

Purpose Thymosin α 1 (Tα1) is an immunomodulatory polypeptide that enhances effector T-cell responses. In this large randomized study, we evaluated the efficacy and safety of combining Tα1 with dacarbazine (DTIC) and interferon alfa (IFN-α) in patients with metastatic melanoma. Patients and Methods Four hundred eighty-eight patients were randomly assigned to five treatment groups: DTIC+IFN-α+Tα1 (1.6 mg); DTIC+IFN-α+Tα1 (3.2 mg); DTIC+IFN-α+Tα1 (6.4 mg); DTIC+Tα1 (3.2 mg); DTIC+IFN-α (control group). The primary end point was best overall response at study end (12 months). Secondary end points included duration of response, overall survival (OS), and progression-free survival (PFS). Patients were observed for up to 24 months. Results Ten and 12 tumor responses were observed in the DTIC+IFN-α+Tα1 (3.2 mg) and DTIC+Tα1 (3.2 mg) groups, respectively, versus four in the control group, which was sufficient to reject the null hypothesis that P0 ≤ .05 (expected response rate of standard therapy) in these two arms. Duration of response ranged from 1.9 to 23.2 months in patients given Tα1 and from 4.4 to 8.4 months in the control group. Median OS was 9.4 months in patients given Tα1 versus 6.6 months in the control group (hazard ratio = 0.80; 9% CI, 0.63 to 1.02; P = .08). An increase in PFS was observed in patients given Tα1 versus the control group (hazard ratio = 0.80; 95% CI, 0.63 to 1.01; P = .06). Addition of Tα1 to DTIC and IFN-α did not lead to any additional toxicity. Conclusion These results suggest Tα1 has activity in patients with metastatic melanoma and provide rationale for further clinical evaluation of this agent.

Purpose: To evaluate the efficacy and toxicity of novel chemotherapy combinations including cisplatin with gemcitabine (GC) or irinotecan (IC) for patients with carcinomas of an unknown primary site. Patients and Methods: Eighty patients were randomly assigned to receive GC or IC. In the GC arm, chemotherapy consisted of cycles combining gemcitabine 1,250 mg/m2 intravenously (IV) on days 1 and 8, and cisplatin 100 mg/m2 IV on day 1 at 3-week intervals. Patients in the IC arm originally received 3-week cycles of irinotecan 200 mg/m2 IV on day 1 and cisplatin 80 mg/m2 IV on day 1. After the inclusion of 15 patients in that arm, the toxicity profile required the irinotecan doses to be reduced to 150 mg/m2 per cycle. Independent histologic and radiologic reviews were done. Results: A total of 78 patients were assessable for efficacy and toxicity. The median number of cycles was four in each arm. Objective responses were observed in 21 patients (55%) in the GC arm (95% CI, 34% to 66%) and in 15 patients (38%) in the IC arm (95% CI, 23% to 54%). Treatment had to be stopped because of toxicity in seven patients in the GC arm and in eight patients in the IC arm. With a median follow-up of 22 months, the median survivals were 8 and 6 months in the GC and IC arms, respectively. Conclusion: This study demonstrates the activity of both the GC and IC regimens. There was toxicity associated with both regimens. Additional studies of combination chemotherapy regimens are required.

9551 Background: With increasing armamentarium in advanced melanoma management, the impact of various strategies remains to be determined including the importance of time to switch from one treatment to another. We report the impact of time to IPI/APD non-planned switch on APD efficacy in real life patients within MelBase (MB). Methods: MB is a French multicentric biobank dedicated to the prospective follow-up (FU) of unresectable stage III or IV melanoma with 1102 patients included since March 2013. Data were collected (Sept.2016) and analyzed (demography, overall survival (OS), progression-free survival (PFS), response rate, multivariate analysis, safety). Results: 71 patients were treated with IPI/APD sequence. 72% received 4 IPI injections. The median time to switch was 1.7 months (0.36-3). The characteristics at the initiation of APD are: mean age 64 yrs, PS 0-1 80%, elevated LDH 34%, BRAF WT 90%, brain metastasis 25%, ≥ 3 metastatic organ sites (MOS) 49%, median FU 11.9 months, OS 20 months (95%CI:12.6-NR), PFS 3.5 months (95%CI:2.9-6.2). The best overall response was 25%, disease control rate was 54% with a low toxicity profile (17% grade 3/4). In a multivariable analysis, longer time to switch was significantly associated with better OS (adjusted HR 0.38 per 1 more month, 95%CI:0.14-0.93), as well as ECOG 0-1 (aHR 3.11, 95%CI:0.99-9.72) and LDH < ULN (aHR 3.32, 95%CI:1.26-8.75). In addition, the association of time to switch with OS vary significantly according to the number of MOS ( < 3 MOS aHR 0.25, 95%CI:0.10-0.62; ≥3 MOS aHR 0.99 95%CI:0.41-2.39) and AJCC stage (M0/1a/1b aHR 0.06, 95%CI:0.01-0.43 ; M1c aHR 0.77, 95%CI:0.39-1.54). Conclusions: In patients who failed IPI treatment, longer survival after APD was associated to time to switch only in patients with favorable baseline factors. Such results are probably more related to the slow kinetics of the disease than to the delay itself. Our results are different from Blank et al.(ESMO 2016) who tested a planned switch, immediately after 2 IPI perfusions, and showed overall response rate close to IPI+APD association. We are currently conducting a similar study on the reverse sequence (APD/IPI) and the role on IPI efficacy.

9538 Background: In the past decade, the prognosis of advanced melanoma has been greatly improved by new therapeutic agents such as immune and targeted therapies, which are now being evaluated as adjuvant therapies. Although the kinetic of metastatic disease is correlated to patient survival, the unfolding of the dormant disease remains hardly predictable and data from the literature on the topic is controversial (Karakousis, Francken). We hypothesized that the course of the advanced disease could be predicted from time to distant recurrence. Methods: Melbase is a French multicentric cohort dedicated to the prospective follow-up of adults with unresectable stage III or IV melanoma. Date of primary excision, time to recurrence, progression free survival (PFS) and overall survival (OS) of 298 patients treated in first line by immune therapies (IT, n = 148), targeted therapies (TT, n = 68) or within clinical trials (n = 73) were collected on September 5th, 2016. Patients with unknown primary or de novo metastatic melanoma were not included. Time to distant recurrence was analyzed as a continuous variable and as a categorical variable ( < 12 months; 12 to 24 months; ≥ 24 months). Results: Patients’ characteristics at baseline are: mean age 62 years, PS 0-1 84%, elevated LDH 32%, BRAF mutated 39%, brain metastases 18%. Time to recurrence studied as a continuous variable was not correlated to PFS (HR = 0.96; 95%CI: 0.85-1.07) or OS (HR = 0.89; 95%CI: 0.77-1.03). These results remained insignificant after stratification upon treatment or even when time to recurrence was analyzed as a categorized variable. Conclusions: Our data showed no correlation between the time from primary to distant recurrence and PFS or OS in patients treated with TT or IT. Therefore, kinetics of advanced disease cannot be predicted by the history of dormant disease. Dormancy and metastasis proliferation are thus probably driven by different molecular and cellular mechanisms.

PURPOSE A phase II trial was designed to determine the efficacy and the tolerance of interleukin-2 (IL-2), interferon alfa-2a (IFNalpha), and fluorouracil (5-FU) in patients with metastatic renal cell carcinoma. PATIENTS AND METHODS One hundred eleven patients were included. Patients received subcutaneous IL-2 9 x 10(6) IU daily for 6 days and IFNalpha 6 x 10(6) IU on days 1, 3, and 5 every other week for 8 weeks. 5-FU was administered through a continuous infusion at 600 mg/m2 for 5 consecutive days for 1 week every 4 weeks. RESULTS The response rate was 1.8% (95% confidence interval [CI], 0% to 4.3%) with only two partial responses (PRs). Toxicity was moderate with 3.6% grade 4 events and two deaths related to treatment. CONCLUSION This regimen of IL-2, IFNalpha, and 5-FU in patients with metastatic renal cell carcinoma was ineffective. The results raise the question of the dose and schedule of subcutaneous cytokines that must be used in metastatic renal carcinoma.

9551Background: Anti-Cytotoxic T-Lymphocyte Antigen-4 (CTLA-4) antibody ipilimumab has been proved efficacy in treatment of metastatic melanoma, but with immune-related adverse events (irAE). Anti-...

9556Background: Melanoma brain metastases (BM) are associated with poor prognosis. Although novel therapies such as ipilimumab (IPI) and anti-PD1 (aPD1: nivolumab and pembrolizumab) have shown thei...

e20004 Background: Uveal melanoma is the most common primary ocular malignancy in adults, accounting for 70 % of all eye malignancy. Despite adequate local treatment, about 30-50 % of patients deve...

9015 Background: Merkel cell carcinoma (MCC) is a rare cutaneous neuroendocrine tumor. The objective was to determine the role of sentinel lymph node biopsy (SNB) in the management of patients with...

e19037 Background: The recent understanding of melanomas’ molecular pathways improved their classification and clinical strategies. Initial studies showed that B-Raf/N-Ras mutations (respectively V600E and Q61) are the most frequent alterations being present in 70 to 80% of melanomas, characterizing non Chronic Sun-induced Damage skins (CSD). These include Superficial Spreading Melanomas (SSM) and Nodular Melanomas (NM). Other studies showed that c-Kit mutations are present at low rate (20-40%) in Acro-Lentiginous Melanomas (ALM), Mucous Melanomas (MM) and in melanomas arising on CSD skins. C-kit mutation pattern is complex with four exons being affected leading to numerous mutations. Because c-Kit targeted therapy is a critical clinical issue, we aimed to identify the most frequent mutations to propose appropriate screening test and adapted therapy. Methods: 186 melanoma samples corresponding to an homogeneous white-caucasian population (Brittany, France) were screened. c-Kit exons 11, 13,17 and 18 were sequenced, c-Kit copy number was quantified by q-PCR and level of c-Kit determined by immunohistochemistry (CD117). Samples were also analyzed for B-Raf mutations (codon 464, 466, 469, 600) and for N-Ras mutations (codon 12, 13, 61) by pyrosequencing. Results: Detectedmutations are shown in the table. Uveal melanomas (n=13) were never mutated. These mutations were never overlaped and comparable profiles were obtained between the primary tumour, nodes or metastatic lesions from a same patient. Conclusions: C-Kit mutations are much less common than in other previously described populations especially for MM or ALM. B-Raf/N-Ras mutations appeared also less frequent than waited in non-CSD , reaching 46% in SSM. Interestingly our population presents a high incidence of N-Ras mutations, especially in SSM and NM. Screening of B-Raf/N-Ras mutations is fundamental not only for SSM but also for NM, CSD and ALM for prescribing targeted therapies such as MEK-inhibitors. [Table: see text]

e22168 Background: CUP are an heterogeneous family of neoplasms with a dismal prognosis, with empiric chemotherapy as the recommended treatment. The aim of this study was to evaluate the feasibility of a 500-mRNA microarray to identify the tissue of origin in patients with CUP. Methods: Diagnostic biopsy formalin-fixed, paraffin-embedded (FFPE) specimens from 22 patients with CUP were prospectively collected. Gene expression profiling was performed using oligonucleotide microarray that contains 495 genes selected as highly differentially expressed between 49 tumor types (CupPrint). Results: The assay was successfully performed on specimens from 18 of the 22 patients (82%). It could not be performed because of a low RNA preservation in the remaining 4 cases. The median age was 57 years (range: 29–70 years). The median delay from tissue shipping to receipt of CupPrint result was 11 days (range: 1–26 days). The most common tissues of origin identified were lung cancer (22%) and colorectal cancer (17%). Of note, a primary cancer which would not be adequately treated by an empiric chemotherapy regimen currently recommended in CUP (like cisplatin-gemcitabine or carboplatin-paclitaxel) was identified in about half patients: kidney cancer (1), hepatocarcinoma (1), colorectal cancer (3), head and neck cancer (2) and cholangiocarcinoma (1). Conclusions: Gene expression profiling of FFPE biopsy specimens from patients with CUP is feasible in a reasonable delay, making it feasible in clinical practice. A phase III randomized trial is planned to compare therapy based on gene expression-suspected primary cancer versus empiric chemotherapy. No significant financial relationships to disclose.

10710 Background: X (Xeloda) is an oral fluoropyrmidine with consistently high activity in MBC, a good safety profile with little myelosuppression and no alopecia, and the convenience of oral administration. The addition of X to docetaxel in anthracycline-pretreated pts also increases survival. This non-randomized phase II study was conducted to evaluate the efficacy, safety and impact on quality of life (QoL) of X in pts with MBC pretreated with anthracyclines and taxanes. Main findings from this trial have been published previously [Fumoleau et al. Eur J Cancer 2004;40:536–42]. Here we present mature data after a follow-up of 48 months. Methods: Pts with anthracycline- and taxane-pretreated MBC received X 1250 mg/m2 twice daily on days 1–14 every 3 weeks for a median of 6 cycles (range 1–15). Results: Baseline characteristics of the 126 pts enrolled were typical of a pretreated MBC population. X achieved complete/partial response or stable disease in 63% of patients (overall response rate, 28%). Median time to progression was 4.9 months (95% CI: 4.0–6.4).Median duration of response was 5.9 months (95% CI: 4.5–12.7). The only grade 3/4 events occurring in ≥ 10% of patients were diarrhea (10%) and HFS (21%). The most common grade 3/4 laboratory abnormality was granulocytopenia (14%). After a follow-up of 48 months, 8 patients are still alive. Updated median overall survival is 15.9 months (95% CI: 13.5–21.3). 1-, 2- and 3-year survival rates are 63%, 37% and 17%, respectively. QoL assessment showed that X treatment was associated with an increase in mean Global Health Score up to cycle 6, with the increase maintained at subsequent evaluations. Conclusions: X is highly active in patients with anthracycline- and taxane-pretreated MBC, leading to a long median survival of 15.9 months. X is also well tolerated and improves QoL. [Table: see text]

2542 Background: Vaccination with dendritic cells (DC), designed to express tumor antigens, is considered to be a possible strategy to elicit anti-tumour immune responses. Effective vaccines that mediate clinical responses in cancer patients may require generation of broad specific cytotoxic T lymphocytes directed against multiple epitopes. Allogeneic lysates from tumor cell lines are an attractive source of antigens as they contain multiple known and unknown tumor associated antigens (TAA). Methods: This is a multicentre, randomised, blinded phase I/II study in patients (pts) with stage IV (AJCC) metastatic melanoma. Safety was the primary objective; immune response (IR) and clinical response, using RECIST criteria, were secondary objectives. Patients were randomised to receive either non matured or matured DC loaded with lysate from 3 melanoma cell lines. DC were obtained from monocytes cultured with IL-13 and GM-CSF and, as required, matured with a bacterial extract and IFN-γ. Six immunizations of ∼2.0...