Prognostic Values Of ROS1 In Lung Adenocarcinoma

Objective: FISH(Break-apart/Split Fluorescence in situ hybridization)is a common technology in clinical detection of rearranged genes.Based upon FISH,we aimed to reveal characteristics of rearranged patients and optimize their screening and diagnosis.By analysis of impact on survival and QOL(Quality of Life),we directed at evaluating prognostic values of ROS1 rearrangements and CNA(Copy Number Alterations)in lung adenocarcinoma.Methods: 1.Status of ROS1 FISH detection in primary NSCLC(Non-Small Cell Lung Cancer)patients from Mayo Clinic Lung Cancer Cohort(MCLCC)were reviewed and applicable specimen of undetected patients were matched.After the second confirmation by pathologist or cytologist,those specimen were sent to the Mayo Central Lab(MCL)for detection.Technicians independently read and recorded results under fluorescence microscopy.2.A retrospective study of clinical data were performed in 634 patients.Last follow up date was November 30,2016 or the day of death.The signal patterns obtained from FISH results were matched to patients’ samples.According to ratios of separated signals,results were dichotomized into positive(ratios of separated signals greater than cut-off points)and negative(ratios of separated signals smaller than cut-off points)rearrangement.Due to combinative patterns of separated signals,positive ones were divided into typical(separation),atypical(single 3’ signal)and rare pattern(single 5’ signal).According to the variation of copy numbers,CNA were divided into normal,amplification and deletion.3.All statistical analysis were performed by SAS,version 9.3(SAS Institute).Bland-Altmann analysis were used for evaluation of reliability for FISH results.Data were compared across groups using chi-square(χ2)test or Fisher exact test(as appropriate).A multivariate logistic regression analysis was performed using backward selection to identify predictive factors for ROS1 rearrangement.Survival curves were generated by Kaplan-Meier method and difference were assessed by Log-rank test.Cox proportional hazard modeling was used to control for confounding variables.A p-value less than 0.05 was considered statistically significant.Results: 1.A total of 634 patients were enrolled.24(3.8%)cases had ROS1 gene rearrangement,of which 17(2.7%)were typical or atypical patterns,and 7(1.1%)were rare patterns.2.ROS1 rearranged patients likely to be younger,had no or mild smoking history and advanced stage of lung adenocarcinoma.Multivariate analysis indicated that younger age(<60 years),female gender,non-smoker and advanced stage(IIIb-IV)were significant predictive factors for ROS1 rearrangement.3.Positive rearrangement was an independent prognostic factor(HR: 0.49,95% CI: 0.26-0.93)for OS(Overall Survival)in patients with adenocarcinoma of the lung,but have no effect on DFS(Disease Free Survival).Whereas CNA is not an independent prognostic factor for OS and DFS.4.Genetic heterogeneity(ROS1 gene rearrangements and CNA)did not predict short-term QOL.There were no significant differences in general physical condition,appetite,fatigue,daily activity,cough,dyspnea,hemoptysis,pain and other related symptoms.5.ROS1 rearranged patients had good response to ROS1 targeted drugs(Crizotinib and Ceritinib).RR(Response Rate)in first line treatment was 50.0%,DCR(Disease Control Rate)was 83.3%.RR in second line or above was 25.0%,DCR was 25.0%.Overall RR was37.5%,DCR was 75.0%.6.Differences in overall survival rate were observed among patients with different rearranged patterns,and patients with rare patterns had the worst prognosis.7.FISH is an effective strategy for detection of ROS1 genetic heterogeneity.Results for rearrangement had relatively good consistency in groups,while results for CNA had relatively bad consistency.Conclusions: 1.FISH for detection of ROS1 rearrangement is applicable.2.ROS1 rearranged patients are likely to be younger,have no or mild smoking history and advanced stage of lung adenocarcinoma.They had good response to targeted therapy and better prognosis on OS.3.CNA showed no exact implications,which might only act as an small aspect of genetic heterogeneity in tumor cells.