ECAsT: a large dataset for conversational search and an evaluation of metric robustness.

The Text REtrieval Conference Conversational assistance track (CAsT) is an annual conversational passage retrieval challenge to create a large-scale open-domain conversational search benchmarking. However, as of yet, the datasets used are small, with just more than 1,000 turns and 100 conversation topics. In the first part of this research, we address the dataset limitation by building a much larger novel multi-turn conversation dataset for conversation search benchmarking called Expanded-CAsT (ECAsT). ECAsT is built using a multi-stage solution that uses a combination of conversational query reformulation and neural paraphrasing and also includes a new model to create multi-turn paraphrases. The meaning and diversity of paraphrases are evaluated with human and automatic evaluation. Using this methodology, we produce and release to the research community a conversational search dataset that is 665% more extensive in terms of size and language diversity than is available at the time of this study, with more than 9,200 turns. The augmented dataset not only provides more data but also more language diversity to improve conversational search neural model training and testing. In the second part of the research, we use ECAsT to assess the robustness of traditional metrics for conversational evaluation used in CAsT and identify its bias toward language diversity. Results show the benefits of adding language diversity for improving the collection of pooled passages and reducing evaluation bias. We found that introducing language diversity via paraphrases returned up to 24% new passages compared to only 2% using CAsT baseline.

Which Surgical Attitude to Choose in the Context of Non-Resectability of Ovarian Carcinomatosis: Beyond Gross Residual Disease Considerations.

BACKGROUND: In ovarian cancer, the increased rate of radical surgery comprising upper abdominal procedures has participated to improve overall survival (OS) in advanced stages by increasing the rate of complete cytoreductions. However, in the context of non-resectability, it is unclear whether radical surgery should be considered when it would lead to microscopic but visible disease (≤1 cm). We aimed to compare the survival outcomes among patients with incomplete cytoreduction according to the extent of surgery. METHODS: Overall, 148 patients presenting with advanced stage ovarian carcinomas were included in this retrospective study, regardless of treatment schedule. These patients were stratified according to the extent of surgery (standard or radical). Complete cytoreduction at the time of debulking surgery could not be carried out in all cases. RESULTS: Among our study population (n = 148), 96 patients underwent standard procedures (SPs) and 52 underwent radical surgeries (RP). Patients in the SP group had a lower Peritoneal Index Cancer (PCI) at baseline (12.6 vs. 14.9; p = 0.049). After PCI normalization, we observed similar OS in the SP and RP groups (39.7 vs. 43.1 months; p = 0.737), while patients in the SP group had a higher rate of residual disease >10 mm (p < 10(-3)). Patients in the RP group had an increased rate of relapse (p = 0.005) but no difference in disease-free survival compared with the SP group (22.2 for SP vs. 16.3 months; p = 0.333). Residual disease status did not impact survival outcomes. CONCLUSIONS: In the context of non-resectable, advanced stage ovarian cancer, standard surgery seems as beneficial as radical surgery regarding survival outcomes and should be considered to reduce surgery-associated morbidity.

Epithelial to Mesenchymal Transition in a Clinical Perspective.

Tumor growth and metastatic dissemination rely on cellular plasticity. Among the different phenotypes acquired by cancer cells, epithelial to mesenchymal transition (EMT) has been extensively illustrated. Indeed, this transition allows an epithelial polarized cell to acquire a more mesenchymal phenotype with increased mobility and invasiveness. The role of EMT is quite clear during developmental stage. In the neoplastic context in many tumors EMT has been associated with a more aggressive tumor phenotype including local invasion and distant metastasis. EMT allows the cell to invade surrounding tissues and survive in the general circulation and through a stem cell phenotype grown in the host organ. The molecular pathways underlying EMT have also been clearly defined and their description is beyond the scope of this review. Here we will summarize and analyze the attempts made to block EMT in the therapeutic context. Indeed, till today, most of the studies are made in animal models. Few clinical trials are ongoing with no obvious benefits of EMT inhibitors yet. We point out the limitations of EMT targeting such tumor heterogeneity or the dynamics of EMT during disease progression.