Olatz Etxaniz

Neoadjuvant cisplatin plus temozolomide versus standard treatment in patients with unresectable glioblastoma or anaplastic astrocytoma: a differential effect of MGMT methylation

AbstractPatients with unresectable glioblastoma or anaplastic astrocytoma have a dismal prognosis. The role of neoadjuvant chemotherapy prior to irradiation in these patients has been studied primarily in non-randomized studies. We have compared the effect of neoadjuvant chemotherapy plus radiotherapy versus concomitant radiotherapy plus temozolomide in a retrospective analysis of two consecutive series of patients in whom surgery consisted of biopsy only. From 2003 to 2005, 23 patients received two cycles of temozolomide plus cisplatin followed by radiotherapy (Cohort 1), and from 2006 to 2010, 23 additional patients received concomitant radiotherapy and temozolomide followed by adjuvant temozolomide (Cohort 2). In Cohort 1, 91.3xa0% of patients received all planned chemotherapy cycles. Progression-free and overall survival were 3.3 and 8.5xa0months, respectively. In Cohort 2, progression-free and overall survival were 5.1 and 11.2xa0months, respectively. No differences between the two groups were observed in rate of completion of radiotherapy, progression-free or overall survival. MGMT methylation was assessed in 91.3xa0% of patients. In Cohort 1, patients without MGMT methylation showed a trend towards shorter progression-free survival (Pxa0=xa00.09), while in Cohort 2, patients without MGMT methylation had longer progression-free survival (Pxa0=xa00.04). In the overall patient population, neoadjuvant temozolomide plus cisplatin had neither a positive nor negative influence on outcome. However, our findings indicate that patients with methylated MGMT may derive greater benefit from neoadjuvant temozolomide than those with unmethylated MGMT.n

Phase II clinical trial of PM00104 (Zalypsis®) in urothelial carcinoma patients progressing after first-line platinum-based regimen

AbstractPurposeThis exploratory phase II clinical trial evaluated the antitumor activity, safety profile and pharmacokinetics of PM00104 (Zalypsis®) 3xa0mg/m2 1xa0h every 3-week intravenous infusion in patients with advanced and/or metastatic urothelial carcinoma progressing after first-line platinum-based chemotherapy.nMethodsThe primary efficacy end point was the disease control rate (DCR), defined as the percentage of patients with confirmed objective response or progression-free at 3xa0months, according to the response evaluation criteria in solid tumors.nResultsIn a first stage (nxa0=xa019 patients evaluable for efficacy), only one patient achieved DCR (stable disease as best response and progression-free survival of 3.1xa0months). According to the 2-stage design used, the primary efficacy objective was unmet, and therefore, the trial was finalized without opening the second stage. The most common adverse events related to PM00104 were fatigue, anorexia, nausea, troponin I increase and neutropenia, which were transient and manageable with dose modifications or administration delays. Mean PK results (Cmaxxa0=xa048.57xa0μg/l and area under the curve (AUC)xa0=xa0154.97xa0hxa0μg/l) were similar to those observed in a previous phase I trial evaluating the same dose and schedule. Few troponin I concentrations were higher than 0.10xa0ng/ml, and none of them were related to parameters of PM00104 exposure such as AUC or Cmax.ConclusionsNo recommendation is given for further evaluation of PM00104 as single-agent treatment of patients with pretreated advanced and/or metastatic urothelial carcinoma. No new safety signals were observed.

The Wide Experience of the Sequential Therapy for Patients with Metastatic Renal Cell Carcinoma

Sequential targeted therapies are the standard of care for patients with metastatic renal cell carcinoma (mRCC). Several drugs are available for patients whose disease progresses while they receive initial tyrosine kinase inhibitor (TKI) therapy; these include nivolumab (an inhibitor of PD-1 receptor), everolimus (an inhibitor of the mechanistic target of rapamycin) or additional TKIs. Until now, there has been no clinical evidence to support the use of one strategy versus another, so investigators and physicians rely on experience, judgement and findings from molecular analyses to select the appropriate treatment. However, with the arrival of nivolumab and cabozantinib that provide an overall survival higher than other alternative treatments, therapeutic strategies may have changed. Here, we discuss findings from preclinical and clinical studies that might help clinicians to choose the optimal treatment approach for patients with mRCC who progress to initial therapy.

IDH mutation status trumps the Pignatti risk score as a prognostic marker in low-grade gliomas

Management of low-grade gliomas (LGG) is based on clinical and radiologic features, including the Pignatti prognostic scoring system, which classifies patients as low- or high-risk. To determine whether molecular data can offer advantages over these features, we have examined the prognostic impact of several molecular alterations in LGG. In a cohort of 58 patients with LGG, we have retrospectively analyzed clinical and molecular characteristics, including the Pignatti criteria, IDH mutations, TP53 mutations, the 1p/19q deletion, and MGMT methylation, and correlated our findings with progression-free survival (PFS) and overall survival (OS). Mean age of patients was 45 years; 71% were classified as low-risk by the Pignatti system. IDH mutations were detected in 62%, p53 mutations in 17%, the 1p/19q codeletion in 46%, and MGMT methylation in 40% of patients. Survival analyses were performed in the 49 patients without contrast enhancement. In the univariate analysis, IDH mutations, the 1p/19q codeletion, and the combination of IDH mutations with the 1p/19q codeletion were associated with both longer PFS (Pu2009=u20090.006, Pu2009=u20090.037, and Pu2009=u20090.003, respectively) and longer OS (Pu2009<u20090.001, Pu2009=u20090.02, and Pu2009<u20090.001, respectively). The multivariate analysis identified absence of IDH mutations as a factor for greater risk of progression [hazard ratio (HR)u2009=u20093.1; Pu2009=u20090.007]and death (HRu2009=u20096.4; Pu2009<u20090.001). We suggest that IDH mutations may be more effective than the Pignatti score in discriminating low- and high-risk patients with LGG.

849PREAL LIFE EFFICACY AND SAFETY OF AXITINIB (AXI) IN PATIENTS WITH RENAL CELL CARCINOMA (RCC): RESULTS FROM THE SPANISH COMPASSIONATE USE PROGRAM

ABSTRACT Aim: AXI is a vascular endothelial growth factor receptor tyrosin kinase inhibitor recently approved for the treatment of patients (pts) with RCC after failure of sunitinib or a cytokine. In a phase III study, AXI showed an improved progression free survival (PFS) compared to sorafenib after failure to one prior treatment. Methods: A retrospective data collection of pts treated under the Spanish compassionate use program was carried out. The objective of the study was to analyze the efficacy and safety of AXI in the real life setting. The study was approved by the regulatory authorities. Results: From September 2011 to March 2014, a total of 225 pts from 45 centers were included, 73.3% were male and the median age was 55 years (range 29-86). 80% had clear cell RCC and 88.4% had prior nephrectomy. 47.6% received AXI in 2nd line, 29.8% in 3rd line and 24.7% in 4th or subsequent line. The starting dose was 5u2003mg/12h in 97.3% of pts, 13.1% had at least one dose titration and 16.4% a dose reduction, mainly due to toxicity. At present, 30.8% of the pts continue on treatment with AXI. The response rate (RR) was 24% with 41.5% of disease stabilization (SD). Median PFS for all pts was 4.86 months (95%CI 3.92-5.81). Median PFS for pts that received AXI in 2nd, 3rd, or 4th or further line was 3.45, 5.03 and 5.49 months respectively (p= 0.383). Patients that achieved a response had a longer PFS than those who achieved only SD with a PFS of 12.32 and 7.00 months respectively (p = 0.009). PFS was also analyzed for several patient subgroups: Patient subgroup PFS (months) p value Clear cell vs. non-clear cell 4.93 vs. 3.88 0.098 Prior nephrectomy vs. no nephrectomy 5.03 vs. 2.69 0.106 Age 5.03 vs. 3.45 0.322 PS 0-1 vs. PS ≥2 5.68 vs. 2.56 No or mild renal impairment vs. moderate or severe renal impairment 4.60 vs. 4.90 0.484 Median overall survival (OS) for all pts was 9.63 months. 76.9% of pts developed at least one adverse event (AE). Most common AE were, all grades (grade 3-4): asthenia 52% (9.3%), diarrhea 30.6%(4.8%) and hypertension 30.1% (2.2%). Conclusions: The overall PFS is similar to the one reported in the phase III trial in patients previously treated with sunitinib. These data confirm the efficacy and safety of AXI in an unselected population. Disclosure: All authors have declared no conflicts of interest.

Abstract 2777: TMPRSS2-ERG predictive value for taxanes resistance according to prior second-line hormonal manipulations in metastatic castration resistant prostate cancer

Background: TMPRSS2-ERG is a genetic alteration specific of prostate cancer, present in primary tumors and maintained under castration resistant prostate cancer (CRPC) progression. It results in androgen-driven overexpression of ERG, which is involved in resistance to taxanes in preclinical models. In prior work, we found that TMPRSS2-ERG expression in blood correlated with docetaxel resistance in metastatic CPRC. Here, we investigated if TMPRSS2-ERG expression in primary tumors predicts taxanes resistance in CPRC and the potential impact of prior second-line hormonal manipulations with abiraterone (A) or enzalutamide (E). Methods: Patients with metastatic CRPC treated with taxanes were included. Formalin-fixed paraffin-embedded (FFPE) tumors and peripheral blood mononuclear cells (PBMCs) fraction were tested for TMPRSS2-ERG by RT-qPCR. FFPE from hormone-sensitive disease (primary diagnosis) were retrospectively obtained. PBMCs were prospectively collected prior taxane initiation. TMPRSS2-ERG expression was tested by RT-qPCR. TMPRSS2-ERG detection was correlated with taxane response and clinical outcome. Results: A total of 84 tumor samples from 74 patients were included: 65 (87.3%) treated with docetaxel, 19 (25.7%) with cabazitaxel and 10 (13.5%) with both. Forty-six tumor samples (54.7%) were TMPRSS2-ERG+ and 38 (45.2%) TMPRSS2-ERG-. Overall, no correlation between tumor TMPRSS2-ERG expression and taxanes response or clinical outcome was observed. In 42 (50%) samples matched tumor and PBMC samples were available at the time of this analysis: 23 (54.7%) had detectable TMPRSS2-ERG on tissue and 11 (26.2%) on PBMCs fraction. In 27 patients, taxanes were administered as a first-line therapy and in 15 after A or E progression. TMPRSS2-ERG was detected in PBMC from 8 (29.6%) and 3 (20%) patients without or with prior A or E. In patients without prior A or E, TMPRSS2-ERG expression in primary tumors predicted a lower median PSA-PFS (5.5 vs 10.1 months for TMPRSS2-ERG+ vs -, respectively; p Conclusions: The role of TMPRSS2-ERG in taxane resistance may be different according to prior exposure to second-line hormone-therapy in CRPC. Prior androgen receptor inhibition may result in TMPRSS2-ERG downregulation and/or activation of alternative mechanisms of resistance. Further data according to this hypothesis will be presented. Citation Format: Mercedes Marin-Aguilera, Oscar Reig, Natalia Jimenez, Ivan Victoria, Lydia Gaba, Sandra Lopez, Javier Prato, Maria Veronica Pereira, Teresa Vilella, Montserrat Domenech, Josep Badal, Albert Font, Juan Jose Garcia-Mosquera, Olatz Etxaniz, Cristina Carrato, Cristina Suarez, Joan Carles, Fabriccio Racc, Pedro Luis Fernandez, Aleix Prat, Begona Mellado. TMPRSS2-ERG predictive value for taxanes resistance according to prior second-line hormonal manipulations in metastatic castration resistant prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2777. doi:10.1158/1538-7445.AM2017-2777

842PCLINICAL OUTCOMES IN PATIENTS (P) WITH METASTATIC RENAL CELL CARCINOMA (MRCC) RECEIVING SEVERAL LINES OF TARGET THERAPY: RETROSPECTIVE ANALYSES OF A COHORT OF THREE MULTIDISCIPLINARY CENTRES FROM SPAIN

ABSTRACT Aim: New targeted therapeutics approved for mRCC offer multiple options in each line (L) of treatment (tto). Approximately 10-15% of mRCC p are treated with a 3rd L target agent. However, there are few data about which is the most appropriate strategy in the sequential tto in these p. Our aim is to analyze the efficacy of these strategies in terms of overall survival (OS). Methods: P diagnosed with mRCC from January 2005 to December 2013 in 3 multidisciplinary centres of Spain were retrospectively studied. Therapies were classified as: vascular endothelial growth factor receptor inhibitors (VEGFi) or mamaliam target of rapamycin inhibitors (mTORi). P were classified into subgroups depending on which strategy of tto they received: VEGFi-VEGFi-VEGFi, VEGFi-mTOR-VEGFi or VEGFi-VEGFi-mTOR. Analyses OS curve and medians were estimated using Kaplan Meier Method. Results: 279p were diagnosed with mRCC. P characteristics were: mean age 61.07y; 69.5% males, 29% females; Histology: 79.2% Clear Cell (CC), 7.2% Papilar, 3.2% Chromophob, 7.2% others; Prognosis: 13.6% poor, 49.1% Intermediate, 34.1% Good; 76.3% Prior nephrectomy. 97.5% (272p) of p received at least 1L of tto, 40% (109p) 2nd L and 16.5% (45p) 3rd L. 78.37% of p from the group of VEGFi-VEGFi tto recived a 3rd L (27% VEGFi and 51% mTORi) in front of 56% from VEGFi-mTOR (all VEGFi as 3rd L). OS of all p was 25 months (m) 95%CI(17.8-32.23). OS from p who only recived 2nd L were: VEGFi-mTOR 20m 95%CI(4.63-35.38); VEGFi-VEGFi 32m 95%CI(8.67-55.33). OS for 3rd L of tto were: VEGFi-mTOR-VEGFi 35m 95%CI(15.05-54.95); VEGFi-VEGFi-mTOR 30m 95%CI(19.37-40.68); VEGFi-VEGFi-VEGFi not reached. Conclusions: Results suggest that sequence with VEGFi has a trend in better outcomes: median OS has not been reached in VEGFi-VEGFi-VEGFi subgroup at the time of analyses and OS from 3rd L subgroups treated with an mTORi was similar than p receiving only 2nd L with two VEGFi. However, the study is limited by unbalanced factors and small number of p. More details of p characteristics subgroups will be provided. Disclosure: All authors have declared no conflicts of interest.