e16554 Background: VETC has been described as an epithelial-to-mesenchymal-transition independent process of metastasis in hepatocellular carcinoma (HCC): endothelium covers small clusters of neoplastic cells allowing tumor dissemination. It has also been noted that VETC-positive HCCs benefit more from tyrosine kinase inhibitors (TKI) therapy. Interestingly, this type of angiogenesis was also found in RCC - and other cancers - and associated with poor prognosis. The objectives of the present study are: 1) to investigate the prognostic role of VETC in a cohort of primary consecutive RCC. 2) to model the predictive role VETC to TKI response in a cohort of metastatic RCC patients treated with sunitinib or pazopanib. Methods: The study was observational and retrospective. To evaluate the VETC effect on overall survival (OS) for Cox regression, with power .8, Hazard Ratio 2.35, proportion of subject VETC+ .4, proportion of events .6, correlation among covariates .4 and a type I error rate .5, the estimated sample size was n=89. We included consecutive patients undergoing surgery at our institution for primary RCC from 2005 to 2007, (Surgical Series; n=92 cases). Moreover, to investigate the possible role of VETC in predicting TKIs benefit, we considered all RCC patients treated with first line TKIs at our center (TKI Series; sunitinib, n = 39; pazopanib n = 17), and recorded the progression free survival (PFS). VETC was assessed with CD34 immunohistochemistry and defined as a continuous endothelial lining around tumor clusters. We used Bayesian probabilistic modeling to detect small effects and multilevel hierarchical modeling to reduce overfitting. Models were fit using Stan and R. The study was approved by institutional review board (n. 865/20) and registered on ClinicalTrials.gov. Results: VETC+ RCC had a worse prognosis in the Surgical Series, with a posterior probability density (PPD) of median OS of 88 months (mo) (standard deviation, SD: 16 mo) for VETC positive Vs 136 mo (SD: 26 mo) for VETC-; the expected loss of median OS was 48 mo (SD: 31 mo) for patients having a VETC + RCC. Conversely in the TKI Series, VETC+ RCC showed longer PFS compared to VETC- ones: with sunitinib a PPD of median PFS of 35 mo (SD:11 mo) for VETC+ Vs 19 mo (SD: 5 mo) for VETC-. Under Pazopanib a PPD of median PFS of 20 mo (SD: 8 mo) for VETC+ Vs 11 mo (SD: 7 mo) for VETC-. The expected gain of median PFS for of VETC+ cases, was 17 and 9 mo (SD: 12 and 11 mo), respectively in sunitinib and pazopanib treated cases. Conclusions: Our results confirmed the general adverse prognostic role of VETC in RCC, however this phenotype gave a substantial PFS gain for patients treated with TKI, similarly to what have been observed in HCC. VETC could be a new predictive bio-marker that allows the delivery of a personalized treatment: patients affected by RCC might directly benefit from a better selection of already approved drugs.
Molecular alterations in clinical stage III cutaneous melanoma: Correlation with clinicopathological features and patient outcome