By: Natalie Paskoski, NSH Communications Specialist
PD-L1 or programmed death-ligand 1 is expressed as a transmembrane protein, which on regular cells, binds with the PD-1 receptor on T-cells to stop them from attacking the healthy cell. One of the ways that tumors evade eradication by the body’s natural immune system, is by taking advantage of this mechanism. Tumor cells expressing PD-L1 bind with the PD-1 receptor like a healthy cell, thus inactivating the cytotoxic T-cell, stopping it from killing the tumor cell.
If we know that a tumor is expressing PD-L1 and binding with T-cells to inactivate them, can’t we block them from connecting? That’s where immune checkpoint inhibitors come in as a cancer treatment. PD-1 and PD-L1 inhibitors are drugs that target either PD-1 or PD-L1 to block the PD-L1 to PD-1 connection so that the T-cells can stay activated and fight the cancer cells.
In this case, PD-L1 can be considered a prognostic predictive biomarker, in other words, the decision for whether to pursue a treatment plan that includes an inhibitor drug to fight a patient’s tumor is based off of the degree to which the tumor is expressing PD-L1. Obviously, you don’t want to use a PD-L1 checkpoint inhibitor if the tumor isn’t expressing it, as it won’t have any effect.
This is where histology plays a crucial role in determining treatment of a patient. There are 4 FDA approved assays for PD-L1, PD-L1 (SP142) from Ventana, PD-L1 (SP263) from Ventana, PD-L1 IHC 22C3 pharmDx from Agilent, and PD-L1 IHC 28-8 pharmDx from Agilent. They all stain a little bit differently though and cannot be used interchangeably as they all have their own scoring algorithms.
The 4 scoring algorithms for PD-L1 are Tumor Proportion Score (TPS), Combined Positive Score (CPS), Tumor Cells (TC), and Immune Cells (IC). TPS looks at tumor cells only, dividing the number of tumor cells showing partial or complete membrane staining divided by total tumor cell count in the area used to evaluate. CPS also looks at immune cells, so it is stained tumor cells plus immune cells, divided by the total number of tumor cells in the area. These two are Agilent methods that require cell counting. The Ventana methods, TC and IC are just based on visual percentage of stained cells, tumor cells in TC and immune cells in IC.
When running a test for PD-L1 in your lab, it is important to understand that this isn’t just a typical stain, where you might change something in the process to get a “good” stain, that is pleasing to the pathologist. Particularly if you’re running PD-L1 testing as a companion diagnostic, it should be considered not as a stain, but as an assay. The results can only be reproducible if the manufacturer’s guidelines are strictly adhered to when validating an antibody for an approved companion diagnostic test. It isn’t just a matter of is the protein present. The quantitative scoring element is essential for determining how much of the protein is present, as this will determine whether or not a checkpoint inhibitor treatment will be likely to produce results.