There seem to be a number of human clinical trials underway and completed that address these lifestyle variables: http://clinicaltrials.gov/ct2/show/NCT00739791?term=prostate+cancer+nutrition+physical+activity&rank=6

In general you only need to use animal models to approximate the human system when you need to do something that would be unethical in the human system (make mutants or test very expeimental, potentially dangerous therapies).

I would say that genetically engineered mouse models are far more accurate when it comes to monitor behavioral parameters. For prostate cancer, two models mimics to some extend the human disease, namely the PTEN and the Myc mice. Part of this text is taken from my thesis and describe a little bit those 2 models.

PTEN (Cancer Cell. 2003 Sep;4(3):209-21):

Specific loss of function of the tumor suppressor PTEN in the prostate epithelium (PE) PTENPE-/- is probably one of the most robust and relevant. PTEN loss is a central event in human PCa. PTEN haploinsufficiency in mice is sufficient to drive mouse prostatic intraepithelial neoplasia (mPIN) formation (Di Cristofano et al., 2001), while loss of two alleles specifically in the PE leads to PCa development in a swift and robust succession of events (Trotman et al., 2003; Wang et al., 2003). PTENPE-/- mice develop hyperplasia at 4 weeks, mPIN at 6 weeks, microinvasive lesions at 9 weeks and can display fully invasive adenocarcinoma from 12 weeks of age (Wang et al., 2003). Interestingly, progression to a metastatic disease might be dependent on the mouse genetic background since different groups have observed conflicting results (Trotman et al., 2003; Wang et al., 2003). Further analysis by the Pandolfi group demonstrated that epithelial cells of PTENPE-/- mice undergo a senescent state before progressing to invasive adenocarcinoma between 4 to 6 months of age. Interestingly, additional loss of Trp53PE-/- is sufficient to bypass senescence and lead to invasive adenocarcinoma in the majority of mice (Chen et al., 2005). Similarly, the PTEN loss of function mouse model was combined with numerous other mouse models and has been very informative in understanding PCa initiation and progression.

Myc (Cancer Cell. 2003 Sep;4(3):223-38):

The overexpression c-Myc that is widely implicated in many human cancers and is located on the most commonly amplified locus in human PCa (Taylor et al., 2010). Overexpression of c-Myc was driven by either the probasin promoter (Lo-Myc) or a composite ARR2/probasin promoter (Hi-Myc) that contains two additional androgen response elements leading to increased transgene expression. Interestingly, the kinetics of disease progression was faster in the Hi-Myc line since they developed mPIN at 2 weeks of age, a feature only observed at 4 weeks of age in the Lo-Myc line. As opposed to SV40-based transgenic mice, Lo-Myc and Hi-Myc mice do not develop neuroendocrine disease but instead progress to invasive adenocarcinoma by 6 or 12 months of age. Castration of 2 month old Hi-Myc mice causes the reversal of the mPIN phenotype, while castration of 8 month old mice leads to only partial regression of the adenocarcinoma (Ellwood-Yen et al., 2003). Myc overexpression is a hallmark of human PCa and although this model does not develop metastasis, it does provide valuable insight into disease development. Other mouse models have been based on the overexpression of oncogenes frequently deregulated in human PCa. These include overexpression of a mutated H-Ras (Scherl et al., 2004), ERG (Klezovitch et al., 2008; Tomlins et al., 2008) or an activated form of human AKT1 (Majumder et al., 2003). However, these models develop a relatively weak phenotype that is limited to mPIN lesions.

However, I must say that for dietary intervention, the Myc model is more attractive since it is highly sensitive to diet modulation (Kobayashi, Cancer Research 2008 or Blando, Cancer Prevention Research 2011) while the PTEN model seems a little less prone to react to diet changes (Kalaany, Nature 2009), probably due to the already hyperactive PI3K pathway.

So I would definitely select a GEM model, and probably go for the Hi-Myc mouse since it is also a simple transgenic mouse much easier to breed compared to the PTEN mouse that has to be crossed with PB-Cre mice and be maintained PTEN-floxed on both alleles.

I hope it does answer your question.

Maybe helpful to read the paper below

Cancer Res. 2013 Dec 19. [Epub ahead of print]

High fidelity patient-derived xenografts for accelerating prostate cancer discovery and drug development.

Lin D, Wyatt AW, Xue H, Wang Y, Dong X, Haegert A, Wu R, Brahmbhatt S, Mo F, Jong L, Bell RH, Anderson S, Hurtado-Cull A, Fazli L, Sharma M, Beltran H, Rubin MA, Cox ME, Gout PW, Morris J, Goldenberg L, Volik SV, Gleave ME, Collins CC, Wang Y.

The canine prostate shares with human the propencity to develop hyperplasia and cancer of the prostate. See references from P. Ofner et al. (1968-1980).

I agree with Robert. In spite of some differences, canine prostate is the closest match for the human prostate. It is also considered the best model for human prostate cancer, even though there is no implanted prostate cancer model in dogs, and most studies are done on the benign prostate. It is often used for evaluation of various prostate ablative techniques like PDT and thermal therapy. See the following articles in addition to Ofner at al:

J. M. Keller, G. R. Schade, K. Ives, X. Cheng, T. J. Rosol, M. Piert, J. Siddiqui, W. W. Roberts and E. T. Keller, "A novel canine model for prostate cancer," Prostate 73, 952-959 (2013).

C. L. Lai, R. van den Ham, G. van Leenders, J. van der Lugt and E. Teske, "Comparative characterization of the canine normal prostate in intact and castrated animals," Prostate 68, 498-507 (2008).

B. E. LeRoy and N. Northrup, "Prostate cancer in dogs: Comparative and clinical aspects," Vet. J. 180, 149-162 (2009).

S. Aggarwal, R. M. Ricklis, S. A. Williams and S. R. Denmeade, "Comparative study of PSMA expression in the prostate of mouse, dog, monkey, and human," Prostate 66, 903-910 (2006).

S. L. Chowning, R. C. Susil, A. Krieger, G. Fichtinger, L. L. Whitcomb and E. Atalar, "A preliminary analysis and model of prostate injection distributions," Prostate 66, 344-357 (2006).

C. M. Moore, D. Pendse and M. Emberton, "Photodynamic therapy for prostate cancer-a review of current status and future promise," Nat. Clin. Pract. Urol. 6, 18-30 (2009).

The nude mouse is an excellent model. There have been several studies. There's an article where part tiulado. Antitumor effect of acid on human meclofenamic androgen-independent prostate cancer: a preclinical evaluation. Posted in International Urology and Nephrology

It will depend on what do you want to study.

For the mechanism of PC from hormone-sensitive to hormone-resistant you may use LNCaP and Vcap xenograft models (inoculated into nude or scid mice.

For the mechnism of metastasis PC3, C4-2 xenograft can be used.

In terms of transgenic mice or k/o mice, they are used for study the role of a particular gene in PC pathogenesis and disease prograssion.

There is no perfact animal model for PC studies, all model has its limitation for use. Eventhough, based on your project, you need to choose one that is best fits your purpose.