With rapid advances in molecular oncology, the availability of preclinical in vitro cell models and in vivo animal models with specific genomic aberrations is critical for improved prediction of clinical outcomes in cancer patients. One of the most widely used preclinical models is conventional cell lines, such as the NCI-60 panel of cell lines;these cell lines are widely used in preclinical testing for novel targeted drugs, partially owing to the low expense and reduced labor associated with cell culture compared with other preclinical models, such as animal xenografts. However, recent studies have shown that accumulation of genetic aberrations in cancer cell lines occurs with increasing passage number. These models also lack the heterogeneity of tumors and do not exhibit a proper microenvironment, highlighting the limitations of cell-based models. Consistent with this, Johnson et al. demonstrated that in vivo activities of the cell lines within the NCI-60 panel did not closely correlate with corresponding human cancers.

Therefore, to better preserve the genomic integrity and tumor heterogeneity observed in patients, patient-derived xenograft (PDX) models are being used more frequently. PDX is generated by directly transplanting freshly resected patient tumors into immunocompromised murine hosts with or without an intermediate in vitro culture step. This PDX model is an improvement over cell lines because it can provide both an appropriate tumor microenvironment and heterogeneity of tumor cells. However, the engraftment success rates and growth rates of implanted tumors are highly variable depending on the tumor type, possibly due to insufficient numbers of hematopoietic cells and/or ineffective microenvironmental cues in the mouse stroma. The extent to which tumor cells from freshly resected tumors are able to withstand mechanical stresses and xenotransplantation barriers is also unclear. Furthermore, the use of PDX models for application in clinical oncology is limited owing to the time required for PDX establishment (> 4 months) since most patients with refractory cancer live less than 1 year. Recently, PDC line models have been suggested as an alternative preclinical model to be used as a prediction tool for preclinical drug sensitivity.

Therefore, in this study, the investigators aimed to overcome these potential barriers of pre-existing models by examining the capacity of PDC line models to recapitulate the histological and genomic features of primary patient tumors. In selected cases, the investigators screened drug sensitivity in vitro using PDC lines and compared the results with real-life clinical treatment outcomes.

Inclusion Criteria:

• Patients older than 18 years.
• Patients with histologically confirmed cancer
• Written informed consent form
• Presence of metastatic lesion(s) not amenable to surgical treatment and having malignant effusion in the body cavity which needed to be drained by percutaneous methods for therapeutic purpose.

Exclusion Criteria:

• Concurrent disease or condition that would make the subject inappropriate for study participation or any serious medical condition that would interfere with the subject's safety.

inclusion criteria are as follows: age ≥ 18 years; pathologically confirmed solid cancer; presence of metastatic lesion(s) not amenable to surgical treatment and having malignant effusion in the body cavity which needed to be drained by percutaneous methods for therapeutic purpose.