Identification of biomarkers of clinical response to PD-1 blockade in lung cancer patients

Approximately 4,000 new cases of lung cancer are diagnosed each year in Sweden, with similar incidences in men and women. There are four clinical stages: I to IV depending on where the cancer is located, if or where it has spread. Current treatment for stage III-IV lung cancer patients is chemo and radiation therapy followed by immune therapy by blockade of the Programmed Cell Death Protein-1 (PD-1).

PD-1 is expressed by antigen-experienced CD8 T cells residing in the tumor tissues and can bind to its ligands PDL-1 and PDL-2 expressed by myeloid derived suppressor cells and tumor cells of multiple tumor types. The binding of PD-1 to PDL-1 leads to a block in T cell receptor signaling, activation and differentiation of CD8+ T cells to cytotoxic T cells in the tumor tissue. Thus, when the tumor infiltrating CD8 cytotoxic T cells engages with the tumor cells with the intention to kill it, the PD-1-PDL-1 interaction instead shuts down the activation and tumor killing function of the CD8 T cells – a mechanism whereby cancer cells effectively protect themselves from immune-cell-mediated killing.

Antibodies binding to PD-1 or PDL-1 can block the PD-1-PDL-1 interaction (PD-1 blockade), resulting in reactivation of the tumor specific CD8 cytotoxic T cells. Data from clinical trials report durable clinical response in 45 percent stage III-IV lung cancer patients after PD-1 blockade. Currently the only predictor of clinical response to PD-1 blockade is PD-L1 expression on tumor cells; current standard at the Sahlgrenska University hospital for newly diagnosed lung cancer patients. However, it is has been shown in several clinical trials, that it a less than perfect predictor of clinical response. Not all patients with high PD-L1 expression have a durable clinical benefit, and some patients with low or no expression of PDL-1 can respond to PD-1 blockade therapy.

Thus, clinicians treating patients with PD-1 antibodies are urgently looking for additional biomarkers than PDL-1 expression alone that can correlate with clinical response after immune therapy. Identifying biomarkers will:

1. avoid exposing patients to the side effects of immune therapy by identifying patients not likely to respond to therapy and
2. save on the very high costs of immune therapy by treating patients likely to respond.

In lung cancer patients tumor associated mutations and expansion of CD8 T cells subsets in the blood, correlate with response to PD-1 blockade. However, the effects of PD-1 blockade on the tumor biology and the immune system in the same patient is lacking and can enhance the specificity of biomarkers predicting clinical response or resistance to PD-1 blockade, which I will explore in the project.

During my sabbatical year at Merck Research Labs, Palo Alto CA, USA, I have developed a tumor model in mice that lack PD-1 expression to define mechanisms of resistance to immune therapy. In addition, I have established a clinical workflow involving genetic analysis of lung tumor tissue and the immune response in the same patient.

The team consists of an oncologist (Andreas Hallqvist), a clinical geneticist (Anna Rohlin), and me as an immunologist (Sukanya Raghavan); each with a strong commitment to improve immune therapy in lung cancer patients at the Sahlgrenska University Hospital. Using specialized strains of mice, patient samples and high throughput screening methods, the current project is poised to successfully identify biomarkers of clinical response after PD-1 blockade.