[PD1/PD-L1 immunohistochemistry in thoracic oncology: Where are we?].

The assays for the assessment of the PD-L1 status by immunohistochemistry are available in clinical studies in thoracic oncology to predict response to immunotherapies targeting the PD-1/PD-L1 pathway. With the arrival of this new class of molecules in second line and very soon in first line of treatment for patients with advanced or metastatic non-small cell lung cancer, these tests will certainly be required in routine once these new drugs will be granted marketing authorization. The rapid introduction of these "companion" or "complementary" tests seems essential to select patients to benefit from these effective but also expensive and sometimes toxic therapies. Although challenged by some oncologists (as some patients not expressing PD-L1 may sometimes respond to PD-1/PD-L1 blockade), the anti-PD-L1 immunohistochemically approach seems inevitable in 2017. This new activity developed in the pathology laboratories raises several questions: which anti-PD-L1 clone should be used? On which device? What threshold of positivity should be considered? Should PD-L1 expression be assessed on tumor cells as well as on the immune cells? What controls should be used? Comparative studies are underway or have been already implemented in order to answer some of these questions. This review addresses the different evaluation criteria for immunohistochemistry using the main anti-PD-L1 antibodies used to date as well the recently published studies using these antibodies in thoracic oncology.