Hasumi et al., 911–916 Prostate cancer is an extremely common form of cancer in older men. The disease has a complex etiology and many open questions also remain on its development and progression. Although hormone therapy is highly effective in shrinking tumor size, most patients become resistant to this treatment after several years. The biological mechanism underlying hormonal treatment resistance, known as recurrent hormone refractory prostate cancer (HRPC), is not clear. In this study, Hasumi et al., hunted for novel genes that may be differentially expressed in prostate cancer patients. Using GeneChip technology with both malignant and normal tissue samples taken from untreated prostate cancer patients, they identified an almost 3-fold increased expression of the protease inhibitor neuroserpin. Real-time quantitative PCR analysis identified the presence of neuroserpin in all prostate cancer (n = 45) and paired normal tissue (n = 45) samples and in patients with HRPC. However, neuroserpin levels were significantly higher in malignant than in normal tissue and higher again in HRPC samples. Importantly, neuroserpin expression had a significant negative correlation with recurrence-free and overall survival. How this protease inhibitor plays a role in prostate cancer disease progression remains to be seen, but in the interim its expression could be a very useful parameter to monitor disease recurrence and survival. Giwercman et al., 994–997 Giwercman et al. also look at prostate cancer, but from an epidemiological perspective. Although it is believed that prostate carcinogenesis is dependent on the presence of testosterone and related androgens, an association between androgen exposure and prostate cancer risk is inconclusive. Androgen status is not an absolute measure of, but is associated with, fertility. In this study, the authors postulate that male fertility—indirectly calculated by whether an individual has fathered offspring or not—is associated with an increased risk of prostate cancer. Using data from the Swedish National Cancer Registry, the authors analyzed nearly 49,000 diagnosed cases of prostate cancer matched by year of birth to healthy controls. The association between the number offspring and the risk of developing prostate cancer was estimated. The risk of developing prostate cancer is slightly lower in men who are childless (OR = 0.83; 95% confidence interval (CI) = 0.81 - 0.86) or who have only fathered one child (OR = 0.93; 95% CI = 0.90 - 0.96). The Odds Ratio for the association between the number of children and the risk of prostate cancer. The authors plainly admit that having fathered children does not take female fertility or the desire to have children into account and, therefore, is clearly not a perfect measurement. However, as these data are based on a huge cohort, the hypothesis that fertility is associated with a slightly increased prostate cancer risk should be further considered. Yanamandra et al., 998–1005 Glioblastoma is a highly invasive cancer with a high morbidity and mortality as neighboring tissue is invaded and destroyed by the malignant cells. Tissue factor pathway inhibitor (TFPI-2) is a serine protease inhibitor that is underexpressed in gliomas. The authors had previously demonstrated that overexpression of TFPI-2 in the human malignant glioma cell line SNB19 reduced invasiveness of tumor growth both in vitro and in vivo. In this study, Yanamandra et al., investigated whether TFPI-2 could be reliably expressed and display its anti-cancer effects when produced by a recombinant adeno-associated virus (rAAV). rAAV is an attractive gene therapy tool as it gives stable expression and can infect nondividing cells. SNB19 human glioma cells were grown as multi-cellular tumor spheroids and then infected with rAAV-TFPI-2. Both the migration rate and the invasiveness of SNB19 sharply declined when cells were treated with rAAV-TFPI-2 compared to control constructs. Neovascularization is also crucial to tumor progression and survival. rAAV-TFPI-2 infected SNB19 cells repressed capillary structure formation by 85% in vitro and angiogenesis by up to 100% in vivo. To test whether this gene therapy model could inhibit tumor growth in vivo, nude mice were injected intracerebrally with SNB19 cells infected with rAAV-TFPI-2. After 4 weeks, the mice that received SNB19 with the control vector had developed tumors, whereas mice receiving rAAV-TFPI-2 infected SNB19 had more than 85% reduction in tumor formation. The mechanism of how TFPI-2 causes these anti-cancer effects is not clear. Nonetheless, this comprehensive study gives very promising results for a novel gene therapy approach to treat malignant brain tumors and thus warrants further investigation.