Serial passaging greatly induce the expansion of tumorigenic/cancer-stem cells. The more the passage, the more tumorigenic they are. Often (but not in all systems) tumorigenic cancer stem cells are drug resistant. So after serial passaging, you will have more proportion of CSCs with resistance property compared to the parental cell line. However, the cells may lose certain characteristics due to serial passaging, if the cells are heterogeneous. So I would recommend testing both parental cell line and serially passaged cell line for your studies.

             A good example is my paper (link below), where we used both parental and serially passaged cells and show the serially passaged cells can resurrect after induction of apoptosis and form spheres more efficiently than parental cells. You may also refer to the reference within (nature) about serial passaging. Good luck..!!

Here is my paper link

https://www.researchgate.net/publication/233742776_Blebbishields_the_Emergency_Program_for_Cancer_Stem_Cells_Sphere_Formation_and_Tumorigenesis_after_Apoptosis

Article Blebbishields, the Emergency Program for Cancer Stem Cells: ...

 The advantage of serial transfer of implanted tumor tissue is that it allows the cultured cells to become adapted to in vivo conditions.  According to some publications, cultured PANC-1 cells may require a long time to develop tumors after subq orthotopic implantation.   If you choose to inoculate with cultured cells for your study, the initiation rate and time for tumor growth may be lengthy for all experiments.  

 However, by first doing serial transfers in vivo, cells with more aggressive growth characteristics will be selected.  Subsequent implantation of these tumor pieces should result in rapid tumor growth.  As a result, it is much quicker and easier to obtain a lot of rapidly growing inoculum from in vivo tumors than from in vitro grown cells.  You can probably do this study either way, but it may be better in the long run to use serial passage.

There is also another thread about thIs on RG that you may want to read:

https://www.researchgate.net/post/What_kinds_of_pancreatic_cancer_are_easy_for_tumorigenic_Nude_PANC-1_BxPC-3_CFPANC-1_Capan-2_SW1990_OR_AsPC-1

Best luck!

http://www.ncbi.nlm.nih.gov/pubmed/7559796

Dear Mr Jinesh (@Goodwin G. Jinesh) and Mr Walter (@Robert J Walter),

Thank you very much for the useful information and the publications provided. 

Dear Quah,

I would not go on serial passaging (according to my own experience) with an established cell line as Panc-1.

I would go on very few serial passaging with primary cancer cell lines.

However, you must be aware that the tumor microenvironement is quite distinct for a given established cell line developing s.c. versus orthotopically.

Established pancreatic cancer cell lines display various biological behaviors even in vitro (see the here attached Mijatovic et al., 2007).

The BxPC-3 human pancreas cancer cells display a higher tumor take rate (in orthotopic site; about 100%) than the Panc-1 model.

I would definitively argue for a direct orthotopic transplantation of human established cell lines from pancreas origin in nude mice (see Van Quaquebeke et al., 2007).

If your Panc-1 cell line is "THE" cel line of interest, I would first try to go orthotopically, and if unsuccesful, s.c., but with direct s.c. injections of cells and without serial passaging.

Goodwin is still very young (I am unfortunately no more!). May be the "old" (very old?) articles here attached could also be of help.

Lastly, here attached are also three recent reviews on xenografts.

Best regards

Robert

Thank you, Mr Robert Kiss (@Robert Kiss), for all of the papers provided. They are really helpful for me.