Search

Somatic copy number variations (CNV, i.e., amplifications and deletions) have been implicated in the origin and development of multiple cancers and some of these aberrations are designated targets for therapies. Although FISH is still considered the gold standard for CNV detection, the increasing number of potentially druggable amplifications to be assessed makes a gene-by-gene approach time- and tissue-consuming. Here we investigated the potential of next generation sequencing (NGS) custom panels to simultaneously determine CNVs across FFPE solid tumor samples. DNA was purified from cell lines and FFPE samples and analyzed by NGS sequencing using a 20-gene custom panel in the GeneReader Platform®. CNVs were identified using an in-house algorithm based on the UMI read coverage. Retrospective validation of in-house algorithm to identify CNVs showed 97.1% concordance rate with the NGS custom panel. The prospective analysis was performed in a cohort of 243 FFPE samples from patients arriving at our hospital, which included 74 NSCLC tumors, 148 CRC tumors, and 21 other tumors. Of them, 33% presented CNVs by NGS and in 14 cases (5.9%) the CNV was the only alteration detected. We have identified CNV alterations in about one-third of our cohort, including FGFR1, CDK6, CDK4, EGFR, MET, ERBB2, BRAF, or KRAS. Our work highlights the need to include CNV testing as a part of routine NGS analysis in order to uncover clinically relevant gene amplifications that can guide the selection of therapies.

Biases in expendable bathythermograph (XBT) instruments have emerged as a leading uncertainty in reconstructions of historical ocean heat content change and therefore climate change. Corrections for these biases depend on the type of XBT used; however, this is unspecified for 52% of the historical XBT profiles in the World Ocean Database. Here, we use profiles of known XBT type to train a neural network that can classify probe type based on three covariates: profile date, maximum recorded depth, and country of origin. Whereas previous studies have shown an average classification skill of 77%, falling below 50% for some periods, our new algorithm maintains an average skill of 90%, with a minimum of 70%. Our study illustrates the potential for successfully applying machine learning approaches in a wide variety of instrument classification problems in order to promote more homogeneous climate data records.

We consider a non-linear filtering problem, whereby the signal obeys the stochastic Navier-Stokes equations and is observed through a linear mapping with additive noise. The setup is relevant to data assimilation for numerical weather prediction and climate modelling, where similar models are used for unknown ocean or wind velocities. We present a particle filtering methodology that uses likelihood informed importance proposals, adaptive tempering, and a small number of appropriate Markov Chain Monte Carlo steps. We provide a detailed design for each of these steps and show in our numerical examples that they are all crucial in terms of achieving good performance and efficiency.

Borderline ovarian tumors (BOTs) are a subset of epithelial ovarian tumors with low malignant potential but significant risk of relapse (10% to 30%). Unfortunately, surgical prognostic factors for BOT relapse have not been clearly identified, probably due to the use of heterogeneous surgical definitions and limited follow-up. The aim of this study was to assess potential relapse risk factors using standard surgical definitions and long follow-up.All patients diagnosed with BOT for a period of more than 10 years in a single institution were included in the analysis. Complete surgical staging was defined as the set of procedures that follow standard guidelines for staging surgery (except lymphadenectomy), performed either with one or two interventions. Fertility-sparing surgeries that preserved one ovary and the uterus but included all the remaining procedures were classified as complete staging. The relationship between potential risk factors and time to BOT relapse was assessed by log-rank tests corrected for multiple comparisons and Cox regression.Forty-six patients with a median follow-up of 5.4 years were included, of whom 91.3% had been diagnosed as FIGO stage I disease and 45.7% had received complete staging surgery. Five relapses were detected (10.9%), all of them in women who had been diagnosed with stage I disease and had received incomplete staging surgery. Log-rank tests confirmed the association between incomplete staging surgery and shorter time to BOT relapse.Complete staging surgery should be considered a cornerstone of BOT treatment in order to minimize the risk of relapse.

Vinflunine (VFL) has been approved in the European Union for second-line treatment of advanced transitional cell carcinoma of the urothelial tract (TCCU) in patients who progress after a platinum based regimen. However, very few patients achieve response by response evaluation criteria in solid tumours (RECIST). Therefore, another 'response' threshold may be more useful than RECIST 1.0 in this setting.One hundred and seventy nine patients with advanced TCCU treated with second-line VFL therapy had chest Computed Tomography (CT) and abdominal/pelvic CT or MRI performed at baseline and at first follow-up (6 weeks ± 3 days) after therapy initiation. Tumour measurements and response by RECIST 1.0 were correlated with overall survival (OS). Kaplan-Meier and receiver operating characteristic (ROC) analysis were then used to determine the optimal size threshold to define 'responders'. Impact of adverse prognostic factors including Eastern Cooperative Oncology Group Performance Status (ECOG PS)0, Hb10 g/dL, and liver metastases was analysed.Tumour response included 13 partial responses (PR) by RECIST 1.0 and 52 patients with ≥ 10% decrease in the sum of longest diameters. Responders by RECIST 1.0 did not have a statistically significant improvement in OS, while patients with sum long axis diameter (SLD) reduction of ≥ 10% had a longer OS than those with SLD reduction of10%: 11.3 versus 6.9 months (log rank p=0.0224). ROC analysis yielded ≥ 10% decrease in SLD as the optimal size change correlating with OS. These results persisted on multivariate analysis.In the study population, a ≥ 10% reduction in SLD at first follow-up imaging is a better early predictor of outcome than RECIST.

Uveal melanoma metastases develop in 6.5-35% of patients, most commonly to the liver. Metastatic uveal melanoma (MUM) survival is poor, with 5-7 months of median survival. We reviewed retrospectively all patients with MUM diagnosed between January 1990 and December 2008 at our institution. We analyzed a total of 58 patients with a median age of 61 years (31-84 years). Median time for metastases development was 25.63 months (0.17-102.43 months). Fifty-six patients had hepatic involvement, 63.8% bilobar and 51.7% had more than or equal to five hepatic metastatic lesions. Sixteen patients (27.6%) had two or more organs involved. Six patients (10.71%) were treated with surgery, 25 patients (44.67%) received systemic chemotherapy, and 23 (41.07%) had best supportive care (BSC). The median overall survival (OS) for all the patients was 10.83 months [95% confidence interval (CI): 6.92-14.74]. Patients who had undergone chemotherapy presented 10.83 months (95% CI: 5.35-16.308) of median OS whereas the patients who did not undergo this treatment had an OS of 8.033 months (95% CI: 2.46-13.61). There were more patients with poor survival characteristics such as worse Eastern Cooperative Oncology Group performance status in the BSC group. OS was poor in treated and BSC patients. Differences in survival are more likely to be related to patient characteristics rather than to a chemotherapy effect. Patients with MUM should be included in clinical trials evaluating other options with newer agents.

Background. The combination of gemcitabine plus capecitabine and sunitinib (GCS) shows activity in metastatic renal cell carcinoma (mRCC). We tested the multitargeted “chemo-switch” regimen as first-line treatment in patients with mRCC. Methods. We assessed the maximum tolerated dose and antitumor activity of GCS in treatment-naïve, advanced mRCC patients. Treatment consisted of intravenous gemcitabine on days 1 and 8, oral capecitabine twice daily on days 1–14, and oral sunitinib daily for six 21-day cycles, followed by sunitinib monotherapy at the investigator's discretion. Dose level 0 (DL0) was gemcitabine 1,000 mg/m2 per day plus capecitabine 650 mg/m2 per 12 hours plus sunitinib 37.5 mg/day; DL1 was gemcitabine 1,000 mg/m2 per day plus capecitabine 850 mg/m2 per 12 hours plus sunitinib 37.5 mg/day. Results. Sixteen patients were enrolled. At DL1, two of four patients had dose-limiting toxicity (DLT; grade 3 diarrhea and grade 4 thrombocytopenia). The dose was reduced to DL0 when only 1 of 12 patients experienced DLT (grade 3 diarrhea, grade 3 mucositis, and grade 3 thrombocytopenia). Dose reductions were frequent (58% of patients), and only seven patients were able to receive the three drugs for more than three cycles. One patient achieved a complete response, three had partial responses, and the best response for four was stable disease. Conclusion. The safety profile of the combination does not seem manageable in this patient population. No further development of the combination is recommended.

Introduction/Background Patients with metastatic renal cell carcinoma (mRCC) in whom first-line therapies have failed might derive clinical benefit with sequential targeted agents. Limited data are available on the efficacy and toxicity of subsequent therapies after disease progression during pazopanib therapy. Patients and Methods Patients with mRCC who received subsequent systemic treatment after pazopanib treatment failure were identified across 7 institutions. Pazopanib was given as first-line therapy in 28 patients and after cytokines therapy in 7 patients. Clinical outcome and toxicity analyses of 2 sequential treatment options (anti-vascular endothelial growth factor [VEGF] or mammalian target of rapamycin inhibitor [mTORi]) is presented. Results Subsequent therapy was anti-VEGF in 22 patients and mTORi in 13. One patient who received bevacizumab and temsirolimus combination was excluded. VEGF-targeted therapies included sorafenib (n = 10), sunitinib (n = 3), bevacizumab (n = 2), cediranib (n = 4) and cabozantinib (n = 3). Patients treated with mTORi received everolimus. Median progression-free survival was 5.6 months from the start of subsequent therapy with anti-VEGF and 2.4 months with mTORi (P = .009). Overall survival (OS) was not significantly different (P = .68). Clinical benefit (including partial response and stable disease) on subsequent therapy was observed in 15 patients (64%) and 4 patients (31%) of anti–VEGF- and everolimus-treated patients, respectively (P = .021). Conclusion In this retrospective study, targeting VEGF was an effective strategy after disease progression during pazopanib treatment, although OS was not different among patients treated with VEGF or mTORi.

Purpose Previous studies have highlighted the importance of an appropriate human epidermal growth factor receptor 2 (HER2) evaluation for the proper identification of patients eligible for treatment with anti-HER2 targeted therapies. Today, the relationship remains unclear between the level of HER2 amplification and the outcome of HER2-positive gastric cancer treated with first-line chemotherapy with trastuzumab. The aim of this study was to determine whether the level of HER2 gene amplification determined by the HER2/CEP17 ratio and HER2 gene copy number could significantly predict some benefit in overall survival and response to therapy in advanced gastric cancer treated with trastuzumab-based chemotherapy. Patients and Methods Ninety patients with metastatic gastric cancer treated with first-line trastuzumab-based chemotherapy were studied. The optimal cutoff values for HER2/CEP17 ratio and HER2 gene copy number (GCN) for discriminating positive results in terms of response and prolonged survival were determined using receiver operating characteristic curves analyses. Results In this study, a median HER2/CEP17 ratio of 6.11 (95% CI, 2.27 to 21.90) and a median HER2 gene copy number of 11.90 (95% CI, 3.30 to 43.80) were found. A mean HER2/CEP17 ratio of 4.7 was identified as the optimal cutoff value discriminating sensitive and refractory patients (P = .005). Similarly, the optimal cutoff for predicting survival longer than 12 months was 4.45 (P = .005), and for survival longer than 16 months was 5.15 (P = .004). For HER2 GCN, the optimal cutoff values were 9.4, 10.0, and 9.5, respectively (P = .02). Conclusion The level of HER2 gene amplification significantly predicts sensitivity to therapy and overall survival in advanced gastric cancer treated with trastuzumab-based chemotherapy.

Neoadjuvant chemotherapy followed by cystectomy improves survival compared with surgery alone. To prevent overtreatment is of outmost importance to define molecular predictors of response for patient selection. We present the currently available data outlining a variety of potential markers to aid for a personalized decision-making process.Apart from p53, other markers of cell cycle regulation and apoptosis such as p21WAF1/CIP1 (p21) gene, Bcl-2, mouse double minute-2 and pRB have also been related to survival. The clinical relevance of epidermal growth factor receptor and HER2 expression has also been investigated with no success. Regarding Ki67, overexpressing tumors may potentially benefit from neoadjuvant therapy and conversely overexpression of vascular endothelial growth factor and bFGF have been linked to resistance to cisplatin-induced apoptosis. The role of multidrug resistance gene 1 and excision repair cross-complementing rodent repair deficiency complementation group 1 supports that enhanced DNA repair in the tumor decreases the benefit of platinum-based treatment. A 20-gene expression model has shown to predict lymph node involvement, helping on decision-making. A gene expression profiling has been proposed as predictive for response to neoadjuvant chemotherapy.Predictive markers will eventually aid in the selection of patients that most likely benefit from preoperative treatment. In the coming years, a panel of markers will become available to achieve the predicted goal.

Background Borderline ovarian tumors (BOTs) are a subset of epithelial ovarian tumors with low malignant potential but significant risk of relapse (10% to 30%). Unfortunately, surgical prognostic factors for BOT relapse have not been clearly identified, probably due to the use of heterogeneous surgical definitions and limited follow-up. The aim of this study was to assess potential relapse risk factors using standard surgical definitions and long follow-up. Methods All patients diagnosed with BOT for a period of more than 10 years in a single institution were included in the analysis. Complete surgical staging was defined as the set of procedures that follow standard guidelines for staging surgery (except lymphadenectomy), performed either with one or two interventions. Fertility-sparing surgeries that preserved one ovary and the uterus but included all the remaining procedures were classified as complete staging. The relationship between potential risk factors and time to BOT relapse was assessed by log-rank tests corrected for multiple comparisons and Cox regression. Results Forty-six patients with a median follow-up of 5.4 years were included, of whom 91.3% had been diagnosed as FIGO stage I disease and 45.7% had received complete staging surgery. Five relapses were detected (10.9%), all of them in women who had been diagnosed with stage I disease and had received incomplete staging surgery. Log-rank tests confirmed the association between incomplete staging surgery and shorter time to BOT relapse. Conclusions Complete staging surgery should be considered a cornerstone of BOT treatment in order to minimize the risk of relapse.

Variant forms of bladder cancer are non-urothelial neoplasms or urothelial carcinomas mixed with other histologies. Compared to pure urothelial carcinoma, they all present with a high stage and grade. Prognosis is variable and there is a lack of evidence regarding the ideal treatment approach, because of their relative infrequency and the noninclusion in bladder cancer randomized trials. Despite this, basic recommendations can be extracted from case series. In the present report, existent literature about variant forms of bladder cancer is reviewed with focus on the most frequent: squamous cell, adenocarcinoma, small cell, micropapillary, sarcomatoid, and lymphoepithelioma-like.

e15007 Background: Standard first-line therapy of advanced TCCU is based on cisplatin-based combinations. In patients deemed unfit to receive CISPLATIN, carboplatin-based or non-platinum combinations are considered. A phase III trial comparing vinflunine (VFL) and best supportive care (BSC) with BSC alone for second-line treatment of advanced TCCU patients demonstrated a survival advantage for VFL+BSC. We studied the impact of the first-line platinum therapy on overall survival in second line setting. Methods: Eligible 357 patients of the phase III study were split in the two following subsets: CISPLATIN (Patients with prior CISPLATIN administration) and NO CISPLATIN (patients without prior CISPLATIN administration). Survival was measured from the date of random assignment. Overall Survival (OS) was calculated using Kaplan-Meier method, with log-rank comparisons. Multivariate Analysis of OS was analyzed with the Cox proportional hazards model, including prognostic factors for second-line setting previously identified (Bellmunt, 2010). Updated survival data in 11/2008 cut-off date was used. Results: CISPLATIN group represented 70.3% (n=251) and NO CISPLATIN 29,7% (n= 106). CISPLATIN group had less Liver involvement (25% vs 43%, p=0.0007) and better WHO-PS (>1: 66% vs 76%; p=0.0478). OS was higher in CISPLATIN group for all eligible patients (HR: 0.77; CI 95% 0.61-0.97; p=0.0294), for VFL+BSC arm (HR: 0.78; CI 95% 0.59-1.02; p=0.0693) and for BSC arm (HR: 0.68; CI 95% 0.42-1.08; p=0.0978). Multivariate analysis including prognostic factors (liver involvement, hemoglobin, PS) and prior platinum administration, did not show effect of CDDP on OS. VFL reduced the risk of death by 24% in CDDP-group (HR: 0.76; CI 95% 0.58-0.99; p=0.043) and by 35% in NO CDDP –group (HR: 0.65; CI 95% 0.41-1.04; p=0.0724). Conclusions: Differences in prognostic factors between CISPLATIN and NO CISPLATIN groups may explain the differences in OS in patients who undergo 2nd line therapy. The choice of Cisplatin or no Cisplatin chemotherapy in the first line did not impact subsequent benefit of vinflunine over best supportive care.

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 i...

10585 Background: Neuroendocrine tumors (NETs) comprise a heterogeneous group of neoplasms containing biologically active peptides and amines in their cytoplasm. Anaplastic lymphoma kinase (ALK) tr...