P1.04-05 PD-L1 Expression (SP263) in Lung Cancer and Paired Brain Metastases – A Single Center Study in 211 Patients

A. Kundig,C. Fung,A. Scherz,A. Ochsenbein,E. Herrmann,E. Ermis,R. Schmid,E. Hewer,S. Berezowska
DOI: https://doi.org/10.1016/j.jtho.2019.08.908
IF: 20.121
2019-01-01
Journal of Thoracic Oncology
Abstract:Lung cancer presents with brain metastases (BM) in a significant number of cases and resected BM are often the only tissue available for analysis of predictive biomarkers. Studies have shown some molecular divergence between BM and primary lung tumor, but data on paired samples is rare. Here we analyzed the expression of the predictive biomarker programmed cell death ligand 1 (PD-L1) in BM and paired primary tumors. In a single-center retrospective study, we assembled a cohort of consecutive BM resected and diagnosed at our University hospital between 2000 and 2015. We analyzed PD-L1 expression (Ventana SP263 kit) on full slides of lung cancer BM (N=211) and available paired primary tumors (N=87), using the increments <1%, 1-49% and ≥50%. Our BM cohort comprised 145 (68.7%) adenocarcinomas, 24 (11.4%) squamous cell carcinomas, 34 (16.1%) small cell carcinomas and large cell neuroendocrine carcinomas and 8 (3.8%) others. PD-L1 staining was evaluable in 193 BM and 87 paired primary tumors. There was no tumoral PD-L1 expression in the high-grade neuroendocrine carcinomas. Among the non-neuroendocrine cancer BM, 86/165 (52.1%) expressed <1% PD-L1, 36/165 (21.8%) 1-49%, and 43/165 (26.1%) ≥50%. PD-L1 scores significantly correlated between BM and primary tumors (p<0.001), with a complete congruence in 62/77 cases (80.5%). Clinically significant discrepancies (cut-off 50%) were present in 9/77 (11.7%) cases: 6 cases were primary tumor-positive/BM-negative, and 3 cases BM-positive/primary tumor-negative. Only 4/9 patients with discrepant PD-L1 scores received chemotherapy in between tissue acquisition of BM and primary tumor, failing to serve as an explanation. Tissue from at least 2 separate brain metastases was available in 12 patients, with discrepant PD-L1 scores regarding the 50% cut-off in only 1 patient (8.3%), and regarding the 1% cut-off in 2 patients (16.7%). There was no association between PD-L1 expression in the BM and overall survival (p=0.95). In conclusion, we show that PD-L1 scores are concordant in most of paired BM and primary tumors. The discordant cases could not be explained by longer time lapses or chemotherapy or radiotherapy in between tissue acquisition, nor by scarce tissue availability of the primary tumor, e.g. due to biopsies only, and may be due to tumor heterogeneity.
What problem does this paper attempt to address?