We have presented data showing the

We have presented data showing the discovery proteomic identification of three plasma proteins which are putative predictive marker of survival for SqCC NSCLC patients after radiotherapy. The level of two of these proteins (LRG1 and CRP) has been shown to be predictive with regard to survival, with elevation indicative of reduced survival time. A combined level of over 140mg/L in plasma was found in all of the patients with shorter survival. The validation was carried out using a non-MS orthogonal antibody-based method, improving confidence that the differences seen are real and therefore, potentially of use in a clinical setting. The agreement, observed in this study and others, between isobaric tagging experiments and ELISA supports the use of an isobaric tagging approaches to the identification of novel biomarkers in bodily fluid before validation by other methods like ELISA. One of the biomarkers we have highlighted, CRP, has previously been proposed as an agent allowing monitoring of chronic rho inhibitor (reviewed Volanakis, 2001) and also been studied as a possible biomarker of lung cancer risk (Chaturvedi et al., 2010; Van Hemelrijck et al., 2011). Elevated levels of circulating CRP were associated with increased risk of lung cancer; elevated levels were observed up to 5years pre-diagnosis. Since chronic inflammation has been proposed to generate an environment advantageous to cancer survival, as well as promoting tumourogenesis (NSCLC reviewed in O\'Callaghan et al., 2010) the identification of CRP as a risk factor may be due to chronic inflammation. Thus, because of the non-specific nature of circulating CRP levels Hemelrijick and colleagues measured multiple time points of CRP and observed that this increased confidence in the link between elevation of CRP and lung cancer risk (Van Hemelrijck et al., 2011). Our study also allowed multiple reading of CRP, thus allowing the increase in this protein during radiotherapy treatment to be observed; therefore multiple time points should be implemented into any potential future studies. The levels of CRP which indicated risk of lung cancer in the study by Hemelrijick and colleagues was over 10mg/mL. It is worth noting that the elevated levels we see during radiotherapy are higher (up to 80mg/L), with the pre radiotherapy levels averaging 34mg/mL. In NSCLC elevated levels of plasma CRP has been linked with poor prognosis for patients undergoing surgical resection and chemotherapy. But to our knowledge this is the only report of CRP elevation as a potential predictive marker for survival after radiotherapy. This indicates that CRP measurement may be useful for all lung cancer patients undergoing treatment. Prognosis for resection and chemotherapy were indicated by levels before treatment, where the link we have observed is a change during therapy. Therefore the mechanism for CRP release may differ between the different therapies and so more work needs to be carried out to investigate the mechanism of the CRP elevation. CRP as a predictive marker of survival after radiotherapy has been previously investigated by Dehing-Oberjie and colleagues as part of a nine protein panel; they did not see CRP levels as a significant indicator (Dehing-Oberije et al., 2011). However only a single time point prior to treatment was acquired. We see no significant difference in pre-treatment levels of CRP in this study which confirms their observation. Lipopolysaccharide binding protein is another acute phase response protein, involved in the immune response to gram negative bacteria. It has been monitored to predict outcome for sepsis and lung injury (Villar et al., 2009) and a study has shown it may be of use for patients with neutropenia associated with cancer to diagnose those with gram negative infections (Oude Nijhuis et al., 2003). Modulation of this protein has not previously been linked with lung cancer or radiotherapy but it is known that it can be expressed by lung epithelial cells upon stimulation by cytokines like IL-6 (Dentener et al., 2000). Therefore the LBP increase seen during treatment could be as a result IL-6 stimulation of the lung epithelial rather than an immune challenge. Previous reports have shown that IL-6 can be expressed in lung cancer cell lines and the lungs of patients after radiotherapy (Zhang et al., 1994). Measurement of IL-6 in the plasma of the patients during radiotherapy in our study showed some correlation with the level of CRP and LRG1, but not LBP. Previously it has been reported that elevated IL-6 plasma levels prior to treatment have been associated with poor prognosis for radiotherapy patients (Dehing-Oberije et al., 2011). With our findings, more investigation is required into whether CRP and LRG1 levels alone can be independent predictors of survival or whether a panel including IL-6 would have greater predictive ability. The identification of CRP and LBP in this study and IL-6 previously (Dehing-Oberije et al., 2011) as indicators of response to radiotherapy suggests that inflammation could be an important factor in radiotherapy response either through tumour response or radiotherapy toxicity. IL-6 levels have previously been shown to be associated with lung toxicity (Siva et al., 2014) and so it would be interesting to see if CRP and LRG1 modulation in plasma could indicate radiotherapy linked toxicity.