In the absence of heterogeneity, studies were pooled using a fixed-effect model. the pooled specificity of 0.59 (95% CI: 0.55-0.62). In wild-type KRAS mCRC patients, the sensitivity and the specificity were 0.80 (95% CI: 0.70-0.87) and 0.60 (95% CI: 0.53-0.66), respectively. Increased EGFR GCN was associated with increased PFS (HR=0.557; 95% CI: 0.382-0.732) and OS (HR=0.579; 95% CI: 0.422-0.737). Conclusions This meta-analysis suggests that EGFR GCN represents a predictive biomarker for tumor response in mCRC patients treated with MoAbs regardless of KRAS mutation. mCRC patients with increased EGFR GCN are more likely to have a better response, PFS, and OS when treated with cetuximab or panitumumab. hybridization (FISH) had an increased EGFR copy number. By contrast, one of 21 nonresponders had an increased EGFR copy number (P 0.0001 for responders hybridization (CISH) in all or part of the patients in the studies; and (III) reported or allowed the calculation of odd ratio (OR) with Polygalaxanthone III corresponding 95% confidence intervals (95% CIs) comparing objective response rate (ORR) stratified by EGFR GCN, reported or allowed the calculation of hazard ratios (HRs) with 95% CIs comparing progression-free survival (PFS) and overall survival (OS) stratified by EGFR GCN. When the same Rabbit polyclonal to ABCC10 patient population was used in several papers, only the most recent studies were included in the meta-analysis. We excluded case reports and case series. Assessment of study quality and data extraction Because there is no validated instrument to measure study quality Polygalaxanthone III for predictive marker studies in an observational setting, we adapted the Newcastle-Ottawa Scale and the frame work suggested by Wells (15). The Newcastle-Ottawa Scale (NOS) contains eight items, categorized into three dimensions including Selection [4], Comparability [1], and Exposure [3]. A high-quality study can be awarded a maximum of one star for each numbered item within the Selection and Exposure categories. A maximum of two stars can be given for Comparability. The NOS ranges between zero up to nine stars. The following data were abstracted onto standardized forms: (I) basic information from papers such as first author, publication year, country; (II) characteristics of patients such as age and gender; (III) information of treatment such as type of MoAbs (cetuximab or panitumumab); (IV) information of the outcome impact factors such as detection method, response criteria, GCN cutoff, and KRAS; and (V) information of outcome such as ORR, PFS and OS. Study quality assessment and data extraction were carried out independently by two reviewers. Disagreements were resolved Polygalaxanthone III by discussion between the two reviewers. Statistical analysis For the meta-analysis, ORR was defined as the primary outcome and PFS and OS as secondary outcomes. For the primary outcome, the association between ORR and EGFR GCN was expressed as pooled OR. Overall effects were determined using the Z test. Predictive Polygalaxanthone III value was accessed by pooled sensitivity, pooled specificity and summary receiver operator characteristic (SROC). The area under the curve (AUC) and an index Q* are useful summaries of the curve (16). For the secondary outcome, the association between PFS and OS and EGFR GCN was expressed as pooled hazard ratio (HR). The methods to combine time-to-event outcomes were summarized by the log HR and its variance (17,18). If the individual trials didnt provide sufficient data, we extracted the data from the Kaplan-Meier survival curves by previously reported method (19) and the HR calculations spreadsheet (Additional file 1 of the paper, The survival curves were read by Engauge Digitizer version 4.1 (free software downloaded from For the primary outcome (ORR), we also did subgroup analyses. It were performed to evaluate the effect by ethnicity (Asian or Europe), MoAbs (cetuximab or panitumumab), EGFR GCN detection method (FISH or CISH), and response criteria [Response Evaluation Criteria in Solid Tumors (RECIST) or WHO]. Statistical heterogeneity was explored by 2 and inconsistency (I2) statistics; an I2 value of 50 percent or more represented substantial heterogeneity (20). In the absence of heterogeneity, studies were pooled using a.