Johan Vansteenkiste, MD, PhD, Respiratory Oncology Unit/Pulmonology, University Hospital KU Leuven, Belgium private × Which markers are available that define patients who are suitable for immunotherapy? When we look at immunotherapy for NSCLC, we should realize that approximately 20 % of treated patients show a response. To direct treatment to patients with a higher likelihood of response, biomarkers might be of value. One biomarker that is established in clinical practice is PD-L1 expression on the tumour, according to immunohistochemistry. For the time being, there are different methodologies to assess PD-L1 expression, but they appear to coalesce slowly, so probably we will eventually have quite a reliable read-out based on immunohistochemistry. PD-L1 is not an absolute marker like the molecular aberrations that we are using for the prescription of targeted agents, but it is what is called an enrichment marker. This means that the higher the PD-L1 expression is, the higher the likelihood that the patient will respond to immunotherapy. Patients can be PD-L1–negative and still have a response, but these are only very few cases. Hence, PD-L1 expression helps to select patients for that type of expensive therapy, and it helps to select patients for immunotherapy when there are perhaps other treatment options available that might be preferable (Figure). × Figure: Likelihood of response to immunotherapy according to PD-L1 expression, and preferred therapeutic approaches In which settings should immunotherapeutic approaches be avoided? Immunotherapy certainly represents great progress in the treatment of NSCLC, but there are patients who are less suitable, or even who have contraindications to this type of therapy. General exclusion criteria include prior allogenic bone marrow transplantation or solid organ transplantation, as suppression of the immune system is of vital importance for those patients. Another contraindication is autoimmune disease or a history of autoimmune disease, because frequently these patients already receive immunosuppressive treatment. In these cases, stimulation of the immune system is of course not an option, and certainly not stimulation of the lymphocytes, which are often at the centre of the pathogenesis of autoimmune diseases. Furthermore, there are some other vulnerabilities that are less absolute, such as interstitial lung disease, active hepatitis, or conditions outside the autoimmune context that require systemic treatment with corticosteroids at daily doses of more than 10 mg prednisone equivalent. Cancer patients who require high doses of corticosteroids for the treatment of their brain metastases are a typical example. How would you rate the global situation regarding practical restrictions, such as reimbursements and availability? We have seen progress with immunotherapy in NSCLC, but most of these agents are expensive. Therefore, access to this therapy for patients is very variable across different regions of the world. In the more developed countries, we see that reimbursement for immunotherapy is gradually being implemented, but even developed nations can look very critically at the incremental value according to the incremental cost. For instance, in the United Kingdom, the national body, NICE, rejected immunotherapy for NSCLC because the cost is not in relation to the real extra value to the patient. Obviously, there is still a long way to go. We can only hope that once more agents are approved, there will be competition, which might decrease the cost of immunotherapy. This would create the possibility for access to this important therapy for an increasing number of patients and in an increasing number of countries.