No abstract

We have previously established that adipose tissue adjacent to breast tumors becomes inflamed by tumor-derived cytokines. This stimulates autotaxin (ATX) secretion from adipocytes, whereas breast cancer cells produce insignificant ATX. Lysophosphatidate produced by ATX promotes inflammatory cytokine secretion in a vicious inflammatory cycle, which increases tumor growth and metastasis and decreases response to chemotherapy. We hypothesized that damage to adipose tissue during radiotherapy for breast cancer should promote lysophosphatidic acid (LPA) signaling and further inflammatory signaling, which could potentially protect cancer cells from subsequent fractions of radiation therapy. To test this hypothesis, we exposed rat and human adipose tissue to radiation doses (0.25-5 Gy) that were expected during radiotherapy. This exposure increased mRNA levels for ATX, cyclooxygenase-2, IL-1β, IL-6, IL-10, TNF-α, and LPA and LPA receptors by 1.8- to 5.1-fold after 4 to 48 h. There were also 1.5- to 2.5-fold increases in the secretion of ATX and 14 inflammatory mediators after irradiating at 1 Gy. Inhibition of the radiation-induced activation of NF-κB, cyclooxygenase-2, poly (ADP-ribose) polymerase-1, or ataxia telangiectasia and Rad3-related protein blocked inflammatory responses to γ-radiation. Consequently, collateral damage to adipose tissue during radiotherapy could establish a comprehensive wound-healing response that involves increased signaling by LPA, cyclooxygenase-2, and other inflammatory mediators that could decrease the efficacy of further radiotherapy or chemotherapy.-Meng, G., Tang, X., Yang, Z., Benesch, M. G. K., Marshall, A., Murray, D., Hemmings, D. G., Wuest, F., McMullen, T. P. W., Brindley, D. N. Implications for breast cancer treatment from increased autotaxin production in adipose tissue after radiotherapy.

Lysophosphatidate (LPA) signaling through 6 receptors is regulated by the balance of LPA production by autotaxin (ATX) vs. LPA degradation by lipid phosphate phosphatases (LPPs). LPA promotes an inflammatory cycle by increasing the synthesis of cyclooxygenase-2 and multiple inflammatory cytokines that stimulate further ATX production. We aimed to determine whether the anti-inflammatory glucocorticoid (GC) dexamethasone (Dex) functions partly by decreasing the ATX-LPA inflammatory cycle in adipose tissue, a major site of ATX secretion. Treatment of human adipose tissue with 10-1000 nM Dex decreased ATX secretion, increased LPP1 expression, and decreased mRNA expressions of IL-6, TNF-α, peroxisome proliferator-activated receptor (PPAR)-γ, and adiponectin. Cotreatment with rosiglitazone (an insulin sensitizer), insulin, or both abolished Dex-induced decreases in ATX and adiponectin secretion, but did not reverse Dex-induced decreases in secretions of 20 inflammatory cytokines and chemokines. Dex-treated mice exhibited lower ATX activity in plasma, brain, and adipose tissue; decreased mRNA levels for LPA and sphingosine 1-phosphate (S1P) receptors in brain; and decreased plasma concentrations of LPA and S1P. Our results establish a novel mechanism for the anti-inflammatory effects of Dex through decreased signaling by the ATX-LPA-inflammatory axis. The GC action in adipose tissue has implications for the pathogenesis of insulin resistance and obesity in metabolic syndrome and breast cancer treatment.-Meng, G., Tang, X., Yang, Z., Zhao, Y., Curtis, J. M., McMullen, T. P. W., Brindley, D. N. Dexamethasone decreases the autotaxin-lysophosphatidate-inflammatory axis in adipose tissue: implications for the metabolic syndrome and breast cancer.

Cytomegalovirus (CMV) infects 40–70% of women, but infection has been reported in >95% of breast cancer patients. We investigated the consequences of these observations by infecting mice with mCMV or a negative control medium for 4 days, 11 days or 10 weeks to establish active, intermediate or latent infections, respectively. Syngeneic 4T1 or E0771 breast cancer cells were then injected into a mammary fat pad of BALB/c or C57BL/6 mice, respectively. Infection did not affect tumor growth in these conditions, but latently infected BALB/c mice developed more lung metastases. The latent mCMV infection of MMTV-PyVT mice, which develop spontaneous breast tumors, also did not affect the number or sizes of breast tumors. However, there were more tumors that were multilobed with greater blood content, which had enhanced vasculature and decreased collagen content. Most significantly, mCMV infection also increased the number and size of lung metastases, which showed a higher cell proliferation. Viral DNA was detected in breast tumors and lung nodules although viral mRNA was not. These novel results have important clinical implications since an increased metastasis is prognostic of decreased survival. This work provides evidence that treating or preventing HCMV infections may increase the life expectancy of breast cancer patients by decreasing metastasis.

Human cytomegalovirus (HCMV) infects 40–70% of adults in developed countries. HCMV proteins and DNA are detected in tumors and metastases, suggesting an association with increased invasion. We investigated HCMV infection in human breast cancer cell lines compared to fibroblasts, a component of tumors, and the role of platelet-derived growth factor receptor-α (PDGFRα). HCMV productively infected HEL299 fibroblasts and, to a lesser extent, Hs578T breast cancer cells. Infection of another triple-negative cell line, MDA-MB-231, and also MCF-7 cells, was extremely low. These disparate infection rates correlated with expression of PDGFRA, which facilitates HCMV uptake. Increasing PDGFRA expression in T-47D breast cancer and BCPAP thyroid cancer cells markedly increased HCMV infection. Conversely, HCMV infection decreased PDGFRA expression, potentially attenuating signaling through this receptor. HCMV infection of fibroblasts promoted the secretion of proinflammatory factors, whereas an overall decreased secretion of inflammatory factors was observed in infected Hs578T cells. We conclude that HCMV infection in tumors will preferentially target tumor-associated fibroblasts and breast cancer cells expressing PDGFRα. HCMV infection in the tumor microenvironment, rather than cancer cells, will increase the inflammatory milieu that could enhance metastasis involving lysophosphatidate.

Human cytomegalovirus (HCMV) infects 40–70% of adults in developed countries. Detection of HCMV DNA and/or proteins in breast tumors varies considerably, ranging from 0–100%. In this study, nested PCR to detect HCMV glycoprotein B (gB) DNA in breast tumors was shown to be sensitive and specific in contrast to the detection of DNA for immediate early genes. HCMV gB DNA was detected in 18.4% of 136 breast tumors while 62.8% of 94 breast cancer patients were seropositive for HCMV. mRNA for the HCMV immediate early gene was not detected in any sample, suggesting viral latency in breast tumors. HCMV seropositivity was positively correlated with age, body mass index and menopause. Patients who were HCMV seropositive or had HCMV DNA in their tumors were 5.61 (CI 1.77–15.67, p = 0.003) or 5.27 (CI 1.09–28.75, p = 0.039) times more likely to develop Stage IV metastatic tumors, respectively. Patients with HCMV DNA in tumors experienced reduced relapse-free survival (p = 0.042). Being both seropositive with HCMV DNA-positive tumors was associated with vascular involvement and metastasis. We conclude that determining the seropositivity for HCMV and detection of HCMV gB DNA in the breast tumors could identify breast cancer patients more likely to develop metastatic cancer and warrant special treatment.

Previous studies have shown that there is a strong correlation between radiologists' diagnoses and their gaze when reading medical images. The extent to which gaze is attracted by content in a visual scene can be characterised as visual saliency. There is a potential for the use of visual saliency in computer-aided diagnosis in radiology. However, little is known about what methods are effective for diagnostic images, and how these methods could be adapted to address specific applications in diagnostic imaging. In this study, we investigate 20 state-of-the-art saliency models including 10 traditional models and 10 deep learning-based models in predicting radiologists' visual attention while reading 196 mammograms. We found that deep learning-based models represent the most effective type of methods for predicting radiologists' gaze in mammogram reading; and that the performance of these saliency models can be significantly improved by transfer learning. In particular, an enhanced model can be achieved by pre-training the model on a large-scale natural image saliency dataset and then finetuning it on the target medical image dataset. In addition, based on a systematic selection of backbone networks and network architectures, we proposed a parallel multi-stream encoded model which outperforms the state-of-the-art approaches for predicting saliency of mammograms.

No abstract