Hypoxia is a feature of most solid tumours and is associated with a poor prognosis. The hypoxic environment can reduce the efficacy of radiotherapy and some chemotherapeutics, and has been investigated extensively as a therapeutic target. The clinical use of hypoxia-targeting treatment will benefit from the development of a biomarker to assess tumour hypoxia. There are several possible techniques that measure either the level of oxygen or the tumour molecular response to hypoxia. The latter includes gene expression profiling, which measures the transcriptional response of a tumour to its hypoxic microenvironment. A systematic review identified 32 published hypoxia gene expression signatures. The methods used for their derivation varied, but are broadly classified as: (i) identifying genes with significantly higher or lower expression in cancer cells cultured under hypoxic versus normoxic conditions; (ii) using either previously characterised hypoxia-regulated genes/biomarkers to define hypoxic tumours and then identifying other genes that are over- or under-expressed in the hypoxic tumours. Both generated gene signatures useful in furthering our understanding of hypoxia biology. However, signatures derived using the second method seem to be superior in terms of providing prognostic information. Here we summarise all 32 published hypoxia signatures, discuss their commonalities and differences, and highlight their strengths and limitations. This review also highlights the importance of reproducibility and gene annotation, which must be accounted for to transfer signatures robustly for clinical application as biomarkers.

Background:The lipogenic transcription factor carbohydrate response element-binding protein (ChREBP) may play a key role in malignant progression of breast cancer by allowing metabolic adaptations to take place in response to changes in oxygenation.Methods:Immunohistochemical analysis of ChREBP was carried out in human breast tumour tissue microarrays representative of malignant progression from normal breast through to metastatic cancer. The ChREBP protein and mRNA expressions were then analysed in a series of breast cancers for correlative analysis with common and breast-specific hypoxia signatures, and survival.Results:In invasive ductal carcinoma, ChREBP correlated significantly with mean ‘downregulated' hypoxia scores (r=0.3, P<0.015, n=67) and in two distinct breast progression arrays, ChREBP protein also increased with malignant progression (P<0.001). However, bioinformatic analysis of a large data set (2136 cases) revealed an apparent reversal in the relationship between ChREBP mRNA level and clinical outcome – not only being significantly correlated with increased survival (log rank P<0.001), but also downregulated in malignant tissue compared with adjacent normal tissue.Conclusion:The ChREBP expression may be reflective of an aerobic metabolic phenotype that may conflict with hypoxia-induced signalling but provide a mechanism for growth at the oxygenated edge of the tumours.

Neuroinflammation after surgery and its contribution to peri-operative neurocognitive disorders (PND) is not well understood. Studying the association between central and peripheral cytokines and neuroinflammation is a prelude to the development of treatments for PND. Here, we investigate the hypotheses that there is a greater cytokine response in cerebrospinal fluid (CSF) than plasma after orthopaedic surgery, and that plasma cytokine levels are directly related to CSF cytokine levels, indicating that plasma cytokine levels may have potential as biomarkers of neuroinflammation. Patients admitted with a fractured neck of femur were invited to participate in this study. Participants had a spinal catheter inserted just prior to induction of anaesthesia. Samples of blood and CSF were taken before, immediately after, and on the first day following emergency surgery. The catheter was then removed. Samples were analysed for the presence of ten cytokines by immunoassay. A spinal catheter was successfully inserted in 11 participants during the 18-month study period. Five plasma cytokines (IL-4, IL-6, IL-10, IL-12p70 and IL-13) rose significantly following surgery, whereas all ten CSF cytokines rose significantly (IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, IFN-γ and TNF-α) (adjusted-p < 0.05). Central (CSF) cytokine levels were consistently higher than their peripheral (plasma) counterparts after surgery, with some patients having a particularly marked neuroinflammatory response. The greatest increases occurred in IL-8 in CSF and IL-6 in plasma. There were significant, strong positive correlations between several of the measured cytokines in the CSF after surgery, but far fewer in plasma. There was no significant correlation between cytokine levels in the plasma and CSF at each of the three time points. To our knowledge, this is the first study to analyse paired samples of plasma and CSF for cytokine levels before and after emergency orthopaedic surgery. This study demonstrates that following surgery for a fractured neck of femur, there is a far greater rise in cytokines in the CSF compared to plasma. The lack of correlation between peripheral and central cytokines suggests measurement of peripheral cytokines are not necessarily related to which patients may have a large neuroinflammatory response.

No abstract