Brief Summary: The study consists in a co-clinical trial by using zebrafish embryos. Specifically, an observational prospective clinical trial on patients operated of epato-biliar-pancreatic cancers and gastro-intestinal cancers undergoing a chemotherapy treatment will be run concurrently to an animal trial on zebrafish embryos xenotransplanted with patient cancer cells in order to demonstrate that zebrafish model is able to predict the therapeutic regimen with the best efficacy for each patient.

Detailed Description: In the last years, a new concept of personalised medicine called 'Mouse Avatars' or 'co-clinical trials' has emerged. Development of mouse avatars implicates implantation of patient tumour samples in mice for use in drug efficacy studies. This model allows conducting preclinical trials in parallel with ongoing human phase I/II clinical trials. Murine and patient trials are conducted concurrently, and information obtained from the murine system is used for clinical management of the patient's tumour. The advantage of this approach is that each patient has his/her own tumour growing in an in vivo system, thereby allowing the identification of a personalised therapeutic approach. Such approach eliminates the cost and toxicity associated with non-targeted chemotherapy. Unfortunately, nowadays, it is clear that co-clinical trials using mouse avatars are not affordable because mice are expensive and time-consuming. The study challenge is to make Avatars available for every patient and the approach sustainable for the National Healthcare Systems. To implement this concept, the investigators propose to replace mouse Avatars with zebrafish Avatars. Indeed, the investigators propose to run co-clinical trials by using zebrafish embryos. Zebrafish embryos as model for human cancer cell xenografts offer several advantages: 1. Rejection-free model. In zebrafish thymus is not working until 9 days post-fertilization (dpf), allowing rejection-free xenografts during this period. Indeed, the use of expensive immune-permissive strains is not required, in sharp contrast to the mouse model. 2. Very low cost of husbanding. 3. Reduced amount of testing material. Zebrafish experiments require much less material to assess drug efficacy. 4. Short time for the analysis of xenotransplants. The time required for a drug efficacy study is 5 days, as opposite to several weeks or months in the mouse model. 5. Low experimental cost and simple procedures. The zebrafish produces large number of embryos with each fecundation (hundreds). This provides statistical power to the analysis and facilitates the collection of a large number of data. 6. Low ethical impact. Zebrafish is classified as insentient from fecundation to the time at which embryos become capable of independent feeding (120 hours post fertilization, hpf); therefore, during this time it does not require a license according to the Directive 2010/63/EU. 7. Data collected in zebrafish are relevant to humans. Zebrafish genome is closely related to that of humans. For instance, remarkable similarity in molecular signalling processes, cellular structure, anatomy, and physiology has been observed between zebrafish and other high-order vertebrates, including humans. This accounts for the exponential increase in the use of zebrafish in drug discovery during the last two decades. Collectively, these points outline the remarkable advantages of the zebrafish Avatar over the mouse Avatar, paving the way for a realistic and cost-sustainable implementation of the co-clinical trials. Specifically, the aim of the study is to perform an observational prospective clinical trial on patients operated of epato-biliar-pancreatic cancers and gastro-intestinal cancers undergoing a chemotherapy treatment in order to demonstrate that zebrafish Avatar is able to predict the therapeutic regimen with the best efficacy for each patient. To this effect, 120 patients meeting the trial inclusion criteria will be enrolled over a 18 months period. In this study, a fragment of tumor will be taken from the surgical specimen by the pathologist, fragmented and transplanted in the yolk of 48 hpf zebrafish embryos. The effect of different anticancer drugs and/or their combinations on the survival, proliferation and migration of the xenotransplanted cancer cells will be evaluated by exposing the xenotransplanted embryos to fish water modified with the drugs. The chemotherapy regimens to be tested in the xenografted embryos are selected in agreement to the common clinical practice, i.e.: * fluoropyrimidines, platinum compounds, irinotecan, taxans (docetaxel) in stomach \& esophageal cancers * fluoropyrimidines, platinum compounds, irinotecan, gemcitabine, nab paclitaxel in pancreatic/biliary cancers * fluoropyrimidines, platinum compounds, irinotecan in colorectal cancers. Embryos will be analysed 4, 24 and 48 hours post injection (hpi). Primary measures will include: * cell proliferation * cell migration * formation of secondary masses * induction of neo-angiogenesis The enrolled patients after the surgical operation will undergo to the adjuvant chemotherapy treatment. The comparative evaluation undertaken after closure of intake will be based on prospectively collected data on (i) clinical outcome and (ii) chemo-sensitivity data collected in zebrafish model.