Author of

• 17398

  +

  P2.01 - Poster Session with Presenters Present (ID 461)

  • Event: WCLC 2016
  • Type: Poster Presenters Present
  • Track: Biology/Pathology
  • Presentations: 1
  • Moderators:
  • Coordinates: 12/06/2016, 14:30 - 15:45, Hall B (Poster Area)

  • 17465

    +

    P2.01-067 - The Relevance of CEA and CYFRA21-1 as Predictive Factors in Nivolumab Treated Advanced Non-Small Cell Lung Cancer (NSCLC) Patients (ID 6121)

    14:30 - 14:30  |  Author(s): C. Genova
    • Abstract

    Background:
    CEA, CYFRA21-1 and NSE are tumor markers acknowledged as useful predictors of response to chemotherapy for advanced adenocarcinoma, squamous and small-cell lung cancer, respectively. However, their role in cancer immunotherapy needs to be investigated.
    
    Methods:
    We analyzed 56 patients with advanced NSCLC treated with nivolumab (3 mg/kg) every 14 days within a single-institutional translational research study. Blood samples were collected at baseline and at each cycle up to 5 cycles, and then every two cycles. All patients underwent a CT-scan every 4 cycles and responses were classified according to RECIST and Immune-Related Response Criteria (irRC). The serum level of CEA was measured with a Chemiluminescent Microparticle Immunoassay while CYFRA21-1 and NSE with an Immuno Radiometric Assay. The markers levels at baseline and after 4 cycles were used to analyze the relationship between their median variation and the objective response rate (ORR). The performance of tumor markers in predicting ORR was analyzed by ROC analysis and a reduction of 20% was used as cut-off level.
    
    Results:
    Forty-eight patients were evaluated: median age: 71 years (44-85); male/female: 73%/27%; current or former smokers: 87.5%; non-squamous/squamous histology: 79%/21%. Baseline median levels were 4.8 ng/ml for CEA, 3.47 ng/ml for CYFRA21-1 and 7.51 ng/ml for NSE. At baseline, values over the upper normal limit of CEA, CYFRA21-1 and NSE were detected in 23 (48%), 26 (54%), and 7 (14%) patients respectively. Significant differences were observed between responders and non-responders and CEA variation (-9% vs.+41%, p=0.003 for RECIST; -10% vs.+31%, p=0.015 for irRC), CYFRA21-1variation (-39% vs.+92%, p<0.001 for RECIST; -35% vs.+72%, p=0.003 for irRC) and NSE variation (-30% vs.+23%, p=0.005 for RECIST; -23% vs.+36%, p=0.004 for irRC). Significant correlations were observed between CEA and CYFRA21-1 decrease with RECIST or irRC: with RECIST, a decrease of 20% of CEA was achieved in 43% of responders and in 8% of non-responders (p=0.013), while a decrease of 20% of CYFRA21-1 occurred in 67% of responders and in 8% of non-responders (p<0.007). With irRC, a decrease of 20% of CEA was achieved in 42% of responders and in 9% of non-responders (p=0.018), while a decrease of 20% of CYFRA21-1 occurred in 58% of responders and in 14% of non-responders (p=0.002). Multivariate analysis confirmed the positive association between CYFRA 21-1 (≤20%) and ORR (RECIST: p=0.004; irRC: p=0.016).
    
    Conclusion:
    The reduction in serum level of CEA and CYFRA21-1 might be a reliable biomarker to predict immunotherapy efficacy in NSCLC patients.