Joaquim Bellmunt

E15104 Background: Patients with progressive prostate cancer treated with androgen blockade, might achieve prostate-specific antigen (PSA) decline or stabilization after discontinuation of antiandrogen treatment. This phenomenon is called antiandrogen withdrawal response and is often associated with clinical improvement. Methods: PSA response of 170 patients from two different centers was retrospectively analyzed. Eligible patients had progressive prostate cancer during treatment with androgen blockade (orchiectomy or LHRH agonist plus antiandrogen). Patients were stratified according to Gleason grade (≤6 vs >6), stage at diagnosis (II, III, IV), type of antiandrogen (bicalutamide, flutamide or steroidal antiandrogens) and duration of treatment blockade (≤12 or >12 months). Clinical benefit (CB) was defined as a confirmed PSA decline ≥50% or stabilization for ≥3 months. Results: Median age was 71 years (range 48 - 91). One hundred forty-one patients had a Karnofsky performance status (K PS) ≥ 80%. Of the 170 patients, 108 (63.5%) had PSA progression after antiandrogen withdrawal and 62 (36.5%) had PSA stabilization or decline. No significant correlation between stage, Gleason grade or K PS and CB was found. Ninety-eight (57.6%) patients had previously received bicalutamide, 52 (30.6%) flutamide and 20 (11.8%) steroidal antiandrogens. CB after antiandrogen withdrawal was observed in 37.8% (37/98), 44.2% (23/52) and 10% (2/20) of patients in the group of bicalutamide, flutamide and steroidal antiandrogens respectively (p=0.024). Longer duration of antiandrogen blockade was associated with CB: 40.5% (53/131) of patients who had received more than 12 months of antiandrogen blockade vs 23.1% (9/39) among patients who had received ≤ 12 months (p=0.0478). Median duration of response was 7 months (range 3 – 52). Conclusions: In patients with progressive prostate cancer, antiandrogen withdrawal may be a useful maneuver resulting, in some cases, in prolonged responses. According to this study, duration of androgen blockade treatment and use of non steroidal antiandrogen may predict a favorable response to antiandrogen withdrawal.

E15095 Background: Gemcitabine (G) plus Capecitabine (C) have demonstrated a modest activity in metastatic renal cell cancer (mRCC) patients (pts). The combination of Sunitinib (S) and C resulted in an acceptable safety profile in pts with advanced solid tumors. We hypothesized that the addition of S to G+C in advanced/mRCC would be synergistic and could increase the therapeutic efficacy. Methods: This is an open-label, phase I, dose-escalation study to assess the maximum-tolerated dose (MTD), safety and antitumor activity (in terms of response rate) of S in combination with C and G in naive pts with advanced mRCC. Advanced RCC pts with adequate organ function, performance status (ECOG 0-1), age >18 years, were eligible. Treatment consisted on iv G day 1 and 8, oral C bid days 1–14, and oral S continuous daily dosing (CDD) schedule, for six cycles (21 days per cycle), followed by sunitinib monotherapy, at the investigator’s discretion if clinical benefit was maintained. Dose level 1 G 1000 C 850 S 37.5; dose level 0 G 1000 C 650 S 37.5 (dose level -1 G 850 C 500 S 37.5). Results: 16 pts were enrolled in the study, median age: 67 yrs, histology: 9 clear cell, 4 papillar and 1 sarcomatoid have been treated at the first two dose levels. In DL1, two out of 4 pts had DLT, consisting on G3 diarrhea and mucositis, and G4 thrombocytopenia. Decision was made to reduce down to level 0; only 1/12 pts experienced DLT at this DL (mucositis G3 and thrombocytopenia G3). Other non-DLTs adverse events were asthenia, hand-foot syndrome, mucositis, diarrhea, infection, neutropenia, epistaxis, low gastrointestinal tract bleeding and hypertension. Dose reductions were frequent (58% of pts) and only 7 pts were able to receive the 3 drugs for > 3 cycles. One death was reported as possibly related to study drugs. A partial response was achieved in 3 pts, and stable disease in 6 patients. Conclusions: Although a low rate of DLT were observed at dose level 0, the safety profile of the combination does not seem to be manageable in most patients. In this trial there are no signs of a synergistic activity for the combination therapy of S plus G/C in therapy naïve pts with mRCC.

4568 Background: 3% to 5% of non-small cell lung cancers demonstrate translocation of the anaplastic lymphoma kinase (ALK) gene. This genetic event is associated with tumor sensitivity to ALK inhibitors. ALK gene translocations, copy number gains or amplification, mutations, and expression, have been not characterized in UC. Our aim was to determine the prevalence of these events in a cohort of clinically annotated and uniformly treated UC patients. Methods: Copy-number (CN) evaluation in all tumor samples were obtained on a single array platform, (Agilent 180K array). ALK status was evaluated by fluorescence in situ hybridization (FISH) with a break-apart probe and a centromeric probe of chromosome 2 (Abbott Molecular Inc) in selected samples with and without CN alteration (CNA) at the ALK locus (2p23). ALK expression was assessed by immunohistochemistry (IHC) (Cell Signaling antibody) and high throughput mutation analysis with the Oncomap 3 platform. Fusion transcripts were evaluated using HTG platform. Results: Tumors from ninety-four UC patients were evaluated. CNA near the ALK locus were found in 3 patients. Polysomy of chromosome 2 (3-6 copies) was observed in 2 patients, and both polysomy and amplification were observed in 1 patient. In patients without CNA, only 1 out of 4 patients presented polysomy (3 copies). ALK expression was undetectable in all 94 samples. mRNA transcript analysis showed 1 case with an atypical (non-EML4) translocation of ALK, and 1 patient with an unconfirmed mutation. Conclusions: ALK alterations are rare events in UC. This is the first description of ALK alterations in UC. The relevance and clinical implications of ALK amplification and translocation in UC merits further evaluation. Sequencing data of the observed mutation and translocation will be presented.

4559 Background: Sunitinib is a tyrosin kinase inhibitor with proven efficacy in renal clear cell carcinoma (RCCC). However properly validated molecular predictors of response or toxicity are still lacking. This study aims to identify such genetic markers focusing on the pharmacokinetics and pharmacodynamics pathways of the drug. Methods: An observational prospective study involving 15 centers of the Spanish Oncology Genitourinary Group (SOGUG) was designed to collect DNA from RCCC sunitinib-treated patients. Eligibility criteria included patients with locally, advanced or metastatic clear renal cell carcinoma treated with sunitinib in a daily practice setting and no prior systemic treatment (neither chemotherapy nor immunotherapy). A total of 15 key polymorphisms in 8 genes involved in the pharmacokinetics (CYP3A4, CYP3A5, ABCB1 and ABCG2) and pharmacodynamics (VEGF, VEGFR2, VEGFR3 and PDGFRa) of the drug were selected to perform a genotype-phenotype analysis. Results: From February 2008 trough December 2010, 100 patients have been included. Regarding clinical factors prognostic classification of the MSKC showed a significant correlation with TTP (p=0.014). Among the polymorphisms studied, two VEGFR3 SNPs presented statistically significant associations with sunitinib TTP (HR=4.3, 95%CI= 2.1-8.8, P=6x10^-5 and HR=3.6, 95%CI=1.9-6.8, P=10^-4, per allele in multivariate analysis), clinical benefit (HR=12.4, 95%CI= 2.0-77, P=0.007 and HR=5.9, 95%CI= 1.3-28, P=0.024, per allele in multivariate analysis) and one of them with OS (HR=7.5, 95%CI= 1.6-35, P=0.01) favouring wild type patients. Two ABCB1 SNPs also showed a significant association with TTP. In relation to toxicities, one CYP3A5SNP showed a trend towards an increased risk for grade 3/4 toxicity (P=0.052). Conclusions: To the best of our knowledge this is the first time a relevant role of VEGFR3 polymorphisms in response to first line treatment for renal clear cell carcinoma is found. If confirmed, inhibition of this target could have therapeutic implications.

274 Background: A four-weekly regimen of gemcitabine and cisplatin (GEMCIS) has similar activity and is less toxic than MVAC in advanced bladder cancer. However, full-dose cisplatin in unfit patients with impaired renal function is contraindicated and other carboplatin-based schedules have been developed. We have previously reported the feasibility of GEMCIS in a biweekly schedule in unfit patients with renal impairment. Here we report a multicenter retrospective study of this biweekly regimen in patients with impaired renal function. Methods: Between January 2004 and October 2009, 40 patients with locally advanced nonsurgically resectable or metastatic bladder cancer and impaired renal function were included. Treatment consisted of gemcitabine 2500 mg/m² on day 1 and cisplatin 35 mg/m² on day 1, every 14 days. Results: Median age of the patients was 73 years (range: 51-82 years). Median IK was 80% (range: 60-100%). Mean creatinine clearance was 49 ml/min (range:37-59 ml/min). Eight patients had previously received chemotherapy with gemcitabine and/or platinum based therapy. Metastatic localizations were: 17 lymph nodes, 10 pulmonary, 10 bone, 7 liver, 12 pelvic and 1 central nervous system. The median number of cycles/patient was 6 (1-13). Out of 36 patients evaluable for response, there was one complete response, 14 partial responses (ORR: 42%; 95% CI 27-58%), 11 stabilizations and 10 progressive diseases. Hematologic toxicities were grade 1 anaemia in 15 patients, grade 2 in 8; grade 3 in 2; grade 3 neutropenia in 5 patients and grade 4 in 1 patient; grade 3 plaquetopenia in 3 patients. Nonhematologic toxicities were grade 1-2 vomiting in 2 patients. Two patients showed a grade 2 hepatic toxicity. Worsening of the renal function was observed in two patients. Alopecia grade 1-2 was seen in three patients. There was one toxic death related to metabolic acidosis.The median progression-free survival is 15 weeks. The median OS from first cycle of GEMCIS is 35 weeks and 1-year OS is 43% (St error 9%). Conclusions: A two-weekly schedule of gemcitabine plus cisplatin is feasible, active, and generally well tolerated in an outpatient setting in unfit patients with poor renal reserve. No significant financial relationships to disclose.

E14018 Background: The validation of KRAS mutations as a negative marker of response to the anti-epidermal growth factor receptor (EGFR) antibodies cetuximab and panitumumab has meant a seminal advance towards treatment individualization of colorectal cancer (CRC) patients. However, since KRAS wild-type status does not guarantee response to anti-EGFR antibodies, identification of other biomarkers of response is urgently needed. We hypothesized that MKP-1, a phosphatase that inactivates phospho-MAPKs including the EGFR downstream protein ERK, could be a mediator of resistance to anti-EGFR antibodies. Methods: Tumor specimens from 48 metastatic CRC patients treated with cetuximab-based chemotherapy were evaluated for KRAS and BRAF mutational status as well as MKP-1 expression as assessed by immunohistochemistry. Results: As expected, clinical benefit was confined to wild-type KRAS and BRAF patients. MKP-1 was overexpressed in 16 patients (33%) and was not associated with patient baseline clinical characteristics or KRAS mutational status. All patients with BRAF mutations (n = 3) had MKP-1 overexpression. Among KRAS wild-type patients, MKP-1 overexpressors had a 7% response rate while patients non- overexpressing MKP-1 had a 44% response rate (p = 0.03). Moreover median time to progression was significantly longer in MKP-1 non-overexpressing patients (32 vs. 13 weeks, p = 0.009). Conclusions: These results support the concept of MKP-1 as a promising marker of resistance to cetuximab-based treatment in colorectal cancer patients with KRAS wild-type status. No significant financial relationships to disclose.

4540 Background: There is a need for improving therapy in "unfit" patients (pts) who are not able to receive standard cisplatinum-based chemotherapy and to look for response predictors. VEGF is expressed in transitional cell carcinoma and its overexpression is associated with an aggressive behavior; targeting angiogenesis maybe a rational approach. Methods: This is a multicenter, open-label, phase II clinical trial assessing oral sunitinib at 50 mg daily in a 4/2-week schedule as first-line in "unfit" urothelial cancer pts. Eligibility criteria to define "unfit" included creatinine clearance under 60 ml/min but over 30 ml/min and performance status ≥ 1 (ECOG). Pretreatment sequential serum samples and measurements of Hounsfield units were collected in pts undergoing treatment. 14 potential serum markers (cytoquines, soluble receptors, growth factors and methaloproteases) were studied (xMAP Technology of Luminex Corp). Results: 37 pts with median age of 75 years were enrolled between July 2006 and September 2009. Bladder was the most frequent primary tumor site. Visceral disease included lung (33.3 %), liver (12%) and bone metastases (12%). Median number of cycles was 2.4 (1-6). Twenty-seven pts are evaluable for efficacy; 3 PRs (8.1%) were obtained, one not confirmed due to brain mets progression; 14 pts (53.8%) had SD lasting > 3m; clinical benefit rate of 62% and median PFS being 5.9 m. Three pts are still on therapy. Most frequent toxicities were fatigue, neutropenia and thrombopenia, hypertension, hand-foot syndrome, stomatitis, haematuria, and epistaxis. IL8 baseline levels were significantly increased in nonresponders and significantly associated with PFS. Final correlative analysis with Hounsfield units is ongoing. Conclusions: Sunitinib first line in “unfit” bladder pts shows significant benefit in 62 % of pts with a median PFS of 5.9 m. leading to a delay on time to start chemotherapy. Baseline levels of IL8 were able to predict benefit and warrant further study as predictive marker of sunitinib activity. Supported by FISS PI061239 and Cellex Foundation. No significant financial relationships to disclose.