78P - PD-L1 expression assessment in Non-Small-Cell Lung Cancer shows stability on Ventana's XT Benchmark platform – “Harmonization study”

Date 10 October 2016
Event ESMO 2016 Congress
Session Poster display
Topics Translational Research
Basic Principles in the Management and Treatment (of cancer)
Presenter Tzahi Neuman
Citation Annals of Oncology (2016) 27 (6): 15-42. 10.1093/annonc/mdw363
Authors T. Neuman1, G. Vainer2
  • 1Pathology, Hadassah Ein Kerem, 91120 - Jerusalem/IL
  • 2Pathology, Tel Aviv Sourasky Medical Center-(Ichilov), Tel Aviv/IL



Pembrolizumab is a monoclonal antibody against programmed cell death 1 (CD279; PD-1), recently approved by the FDA as a 2nd line therapy in NSCLC with a companion diagnostic (PD-L1 22C3, Dako). Today, the only validated IHC platform for PD-L1 detection is the Link-48 platform (Dako), which lowers the availability of the test. Ventana's benchmark XT platform is a widespread IHC platform. However, data about its reliability and reproducibility using the 22C3 antibody is lacking.


A comprehensive calibration of the 22C3 PD-L1 staining on the Benchmark XT platform (Ventana) was performed by combining the FDA-approved, pre-diluted 22C3 anti PD-L1 primary antibody (Dako) with two of Ventana's DAB detection systems, UltraView and OptiView. After receiving a comparable IHC pattern, 41 NSCLC random cases were independently evaluated by 2 expert pathologists for PD-L1 protein expression, using both platforms, defining the tumor proportion score (TPS): the percentage of tumor cells showing complete or partial membrane staining. Each case was classified as PD-L1 negative, weakly positive, or strongly positive (


Using the Dako IHC platform 8, 7 and 26 cases were stratified as PD-L1 strongly positive, weakly positive and negative cases, respectively. Using the Ventana's UltraView protocol 36/41 cases (87.8%) had the same results, including the 8 strongly positive cases. Pearson's correlation score indicate a high concordance (0.91; p value


Pembrolizumab treatment for NSCLC patients is coupled to the PD-L1 TPS by Dako. The Ventana's benchmark XT platform can also be used safely to stratify patients with the same algorithm. We provide a defined PD-L1 IHC protocol which is easy to replicate as we used commercially available pre-made reagents only.

Clinical trial identification

Irrelevant. This is not a clinical trial.

Legal entity responsible for the study

Hadassah Medical Center, Jeruslaem, Isreal. Tel Aviv Medical Center, Tel Aviv, Israel.




G. Vainer: Received advisory or consultant fees from Roche, Pfizer, Astra Zeneca and MSD. All other authors have declared no conflicts of interest.