Perinatal factors including high birth weight have been associated with childhood brain tumors in case-control studies. However, the specific contributions of gestational age and fetal growth remain unknown, and these issues have never been examined in large cohort studies with follow-up into adulthood. We conducted a national cohort study of 3,571,574 persons born in Sweden in 1973-2008, followed up for brain tumor incidence through 2010 (maximum age 38 years) to examine perinatal and familial risk factors. There were 2,809 brain tumors in 69.7 million person-years of follow-up. After adjusting for potential confounders, significant risk factors for brain tumors included high fetal growth (incidence rate ratio [IRR] per additional 1 standard deviation, 1.04; 95% CI, 1.01-1.08, P=0.02), first-degree family history of a brain tumor (IRR, 2.43; 95% CI, 1.86-3.18, P<0.001), parental country of birth (IRR for both parents born in Sweden vs. other countries, 1.21; 95% CI, 1.09-1.35, P<0.001), and high maternal education level (Ptrend=0.01). These risk factors did not vary by age at diagnosis. The association with high fetal growth appeared to involve pilocytic astrocytomas, but not other astrocytomas, medulloblastomas, or ependymomas. Gestational age at birth, birth order, multiple birth, and parental age were not associated with brain tumors. In this large cohort study, high fetal growth was associated with an increased risk of brain tumors (particularly pilocytic astrocytomas) independently of gestational age, not only in childhood but also into young adulthood, suggesting that growth factor pathways may play an important long-term role in the etiology of certain brain tumor subtypes.

Breast cancer brain metastasis is resistant to therapy and a particularly poor prognostic feature in patient survival. Altered metabolism is a common feature of cancer cells, but little is known as to what metabolic changes benefit breast cancer brain metastases. We found that brain metastatic breast cancer cells evolved the ability to survive and proliferate independent of glucose due to enhanced gluconeogenesis and oxidations of glutamine and branched chain amino acids, which together sustain the nonoxidative pentose pathway for purine synthesis. Silencing expression of fructose-1,6-bisphosphatases (FBP) in brain metastatic cells reduced their viability and improved the survival of metastasis-bearing immunocompetent hosts. Clinically, we showed that brain metastases from human breast cancer patients expressed higher levels of FBP and glycogen than the corresponding primary tumors. Together, our findings identify a critical metabolic condition required to sustain brain metastasis and suggest that targeting gluconeogenesis may help eradicate this deadly feature in advanced breast cancer patients. Cancer Res; 75(3); 1–12. ©2014 AACR.

Breast cancer brain metastasis is resistant to therapy and a particularly poor prognostic feature in patient survival. Altered metabolism is a common feature of cancer cells but little is known as to what metabolic changes benefit breast cancer brain metastases. We found that brain-metastatic breast cancer cells evolved the ability to survive and proliferate independent of glucose due to enhanced gluconeogenesis and oxidations of glutamine and branched chain amino acids, which together sustain the non-oxidative pentose pathway for purine synthesis. Silencing expression of fructose-1,6-bisphosphatases (FBPs) in brain metastatic cells reduced their viability and improved the survival of metastasis-bearing immunocompetent hosts. Clinically, we showed that brain metastases from human breast cancer patients expressed higher levels of FBP and glycogen than the corresponding primary tumors. Together, our findings identify a critical metabolic condition required to sustain brain metastasis, and suggest that targeting gluconeogenesis may help eradicate this deadly feature in advanced breast cancer patients.

Pan- or multidrug resistance is a central problem in clinical oncology. Here, we use a genetically engineered mouse model of BRCA2-associated hereditary breast cancer to study drug resistance to several types of chemotherapy and PARP inhibition. We found that multidrug resistance was strongly associated with an EMT-like sarcomatoid phenotype and high expression of the Abcb1b gene, which encodes the drug efflux transporter P-glycoprotein. Inhibition of P-glycoprotein could partly resensitize sarcomatoid tumors to the PARP inhibitor olaparib, docetaxel, and doxorubicin. We propose that multidrug resistance is a multifactorial process and that mouse models are useful to unravel this. Cancer Res; 75(4); 1–10. ©2014 AACR.

Pan- or multi-drug resistance is a central problem in clinical oncology. Here we use a genetically engineered mouse model of BRCA2-associated hereditary breast cancer to study drug resistance to several types of chemotherapy and PARP inhibition. We found that multi-drug resistance was strongly associated with an EMT-like sarcomatoid phenotype and high expression of the Abcb1b gene, which encodes the drug efflux transporter P-glycoprotein. Inhibition of P-glycoprotein could partly re-sensitize sarcomatoid tumors to the PARP inhibitor olaparib, docetaxel and doxorubicin. We propose that multi-drug resistance is a multi-factorial process and that mouse models are useful to unravel this.