Virtual Library

Start Your Search

Shriram Velamuri



Author of

  • +

    P2.04 - Immuno-oncology (ID 167)

    • Event: WCLC 2019
    • Type: Poster Viewing in the Exhibit Hall
    • Track: Immuno-oncology
    • Presentations: 1
    • Moderators:
    • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall
    • +

      P2.04-05 - A Retrospective Evaluation of PD-L1 Expression Heterogeneity on Primary Non-Small Cell Lung Cancer and Metastatic Lymph Nodes (REPLICA) (ID 2686)

      10:15 - 18:15  |  Author(s): Shriram Velamuri

      • Abstract

      Background

      Immunotherapies have changed the treatment landscape in lung cancer. Programmed death ligand (PD-L1) protein which is defined as membranous PD-L1 expression on tumour cells regardless of staining intensity represents a reliable biomarker. The heterogeneity of PD-L1 expression in different sites of disease is unknown; this may have implications on biopsy, staging and potentially treatment choice. The primary objective of this study is to evaluate the relationship between PD-L1 expression in the primary site (lung) and metastatic lymph node LNs N1 and/or N2, in NSCLC.

      Method

      Paired samples (lung tumour and hilar/mediastinal LNs) from 300 patients who underwent lung resection and lymphadenectomy for NSCLC, between January 2013 and October 2018, have been collected and analysed for PD-L1 immunohistochemistry expression using the 22C3 pharmDx Agilent assay. PD-L1 stained slides have been reviewed by two pathologists independently, using Tumour Proportion Score in the categories of <1%, 1-49% and >=50%. The agreement between the pathologists was assessed using Kappa statistic, while cross tabulation was used to compare PD-L1 expression and assess association between the primary tumour and involved LNs.

      Result

      Of the 300 paired samples 285 have been assessed by pathologist 1 and 300 by pathologist 2. The median age of the patients is 69; 55% were males, 92% had a positive smoking history and 65% had an adenocarcinoma; 50% had hilar involvement only. There was moderate agreement between the two pathologists in PD-L1 assessment in both primary tumour [k=0.40 (95%CI 0.34-0.47)] and LNs [k=0.51(95%CI 0.47-0.59)]. Pathologist 1 and 2 reported 80% and 81% overall agreement between PD-L1 expression classification in the primary tumour and involved LNs respectively. When the primary tumour showed no PD-L1 expression, both pathologists reported no PD-L1 expression in 95% of the paired LNs. When the primary tumour expressed PD-L1 >=50%, pathologist 1 reported 30% of involved LNs showed <1% and 30% showed 1-49% staining, whilst pathologist 2 reported 4 % of involved LN showed PD-L1 <1% and 39% showed 1-49% PD-L1 staining.

      Conclusion

      The majority of cases show similar PD-L1 expression between the primary tumour and involved LNs while heterogeneity is present in 20% of cases. Where the PD-L1staining was >=50 in the primary tumour, the majority of involved LNs showed only 1-49% or <1% for PD-L1 expression. These findings may have implications on site of biopsy and management.