Advertisement
MA07 LIQUID BIOPSY APPLICATIONS FOR IMMUNO-ONCOLOGY, TARGETED THERAPIES, AND EARLY STAGE NSCLC THURSDAY, SEPTEMBER 09, 2021 - 10:45-11:45| Volume 16, ISSUE 10, SUPPLEMENT , S906-S907, October 01, 2021

MA07.05 A Novel Blood-Based microRNA Diagnostic Test With Very High Accuracy for Early Lung Cancer Screening and Monitoring

      Introduction

      Lung cancer is the leading cause of cancer death worldwide. Stage I lung cancers have a very good 5-year overall survival rate of about 80%, but these patients are typically asymptomatic and thus unlikely to be diagnosed. In fact, a staggering 75% of lung cancers are diagnosed at stages III and IV with a 5-year overall survival rate of only 30% and 6%, respectively. Lung cancer is in need of a good early diagnosis method. The current standard screening technique, low dose computed tomography (LDCT) has only 60% specificity, resulting in unnecessary invasive lung biopsies. We report here, the development of a serum-derived, microRNA (miRNA)-based genomic signature for potential early lung cancer diagnosis, by making use of a large public microarray dataset (Communications Biol 2020;3:134).

      Methods

      The miRNA microarray dataset (1566 lung cancer patients and 2178 non-cancer controls across 2565 miRNAs) was downloaded from GEO. The data were divided into the same training and test sets as in the original publication. Linear Models for Microarray Data (limma) analysis was used to identify and rank differentially expressed miRNAs in the training set for building diagnostic models by calculating a diagnostic index using top rank miRNAs. Cross-validation was performed to determine the optimal number of miRNAs for the final diagnostic model, which was then evaluated in the test set for the diagnostic performance.

      Results

      TableSensitivity of the new diagnostic model vs. the original diagnostic model in relevant clinical subgroups
      Clinical SubgroupsNOriginal ModelNew Model
      Clinical stage IA6860.9610.996
      Clinical stage IB2850.9370.996
      Clinical stage IIA1460.9730.979
      Clinical stage IIB610.9670.984
      Clinical stage IIIA1640.9020.994
      Clinical stage IIIB60.8331.000
      Clinical stage IV81.0001.000
      T stage T1a4660.9610.996
      T stage T1b2970.9560.993
      T stage T2a4350.9360.991
      T stage T2b520.9231.000
      T stage T3890.9440.989
      T stage T4170.9411.000
      N stage N010470.9550.995
      N stage N11660.9580.982
      N stage N21420.9010.993
      Adenocarcinoma10380.9510.992
      Squamous cell carcinoma2050.9420.995
      Small cell lung cancer220.9091.000

      Conclusion

      A serum-based diagnostic test with nearly perfect accuracy has been developed, which showed exceedingly better performance than LDCT. If validated in independent studies, this test has the potential for effective clinical screening of early stage lung cancers.

      Keywords

      lung cancer, serum microRNA, early diagnosis