Background: With the approval of the PD1 Inhibitor pembrolizumab for
first line treatment of PD-L1+ NSCLC, PD-L1 testing by immunohistochemistry
(IHC) has become a necessity. However, the DAKO Autostainer
AS48 for the FDA approved DAKO 22C3 PharmDX assay (22C3PhDX) is
not available in Switzerland. The goal of this study was to cross validate the
concentrated 22C3 antibody on the Ventana BenchMark Ultra and Leica
Bond immunostainers.
Methods: We developed protocols for the concentrated anti-PD-L1 22C3
antibody for both platforms. 22C3PhDX on the AS48 was used as standard.
A tissue microarray was constructed from 23 NSCLC. Empty sections
and the antibody were distributed to 16 participants to be stained
on the Ventana (8) and/or Leica (12) platform. Central scoring and statistical
evaluation was performed in Basel. We also tested the performance
of the Ventana SP263 assay and Cell signaling E1L3N antibody in selected
centers using local protocols.
Results: There was strong variability of 22C3 PD-L1 tumor cell staining
as a continuous variable for both platforms. Categorical PD-L1 staining
(0–49% vs. 50–100%) did not significantly differ between the centers using
Ventana, whereas data from Leica remained highly variable (p<0.001).
Staining on Leica was generally weaker than on Ventana compared to 22C3PhDX making scoring more tiresome and requiring a higher magnification.
SP263 (4 centers) was highly concordant with 22C3 on Ventana
and with P22C3PhDX, whereas E1L3N (3 centers) was strongly variable.
Conclusions: PD-L1 IHC using a standardized protocol for 22C3 antibody
on Ventana is feasible and provides satisfactory results in most centers,
and also the approved SP263 assays is confirmed as a good alternative
to 22C3PhDx. 22C3 PD-L1 IHC on Bond is currently not recommended
due to major staining variability between centers. Local fine-tuning of
IHC protocols and participation in ring trials are critical when laboratory
developed 22C3 IHC is used.