Biomarkers Define Future Age of Immunotherapy in Cancer

Drew Pardoll, MD, PhD

The use of immunotherapy as a treatment for cancer is progressively increasing with a flood of recent approvals for immune checkpoint inhibitors directed against CTLA-4 and PD-1. These therapies have revolutionized cancer treatment, making immunotherapy the fourth pillar of cancer treatment.

“Immunotherapy is the perfect anti-cancer agent, more patients are being treated with immunotherapy and it continues to increase,” stated Drew Pardoll, MD, PhD, John Hopkins University, at the AASLD Liver Meeting. Discovering ways to look at patients’ endogenous immune systems can lead to treating patients with the right type of immunotherapy by targeting, detecting, and destroying cancer cells, which will be important as these approaches are explored in hepatocellular carcinoma (HCC).

Various immune checkpoints that appear to regulate T‐cell activity have been identified, including cytotoxic T-lymphocyte antigen 4 (CTLA‐4), programmed cell death protein 1 (PD‐1), and lymphocyte activation gene 3 (LAG‐3). Checkpoints are stimulatory and inhibitory signals that regulate the duration and level of the immune response. The inhibitory immune checkpoints act like brakes on the immune response. In this case, they block a signal that would prevent activated T‑cells from attacking the cancer.

The checkpoint inhibitor that is furthest along in development for HCC is nivolumab (Opdivo), a human IgG4 anti—PD‑1 monoclonal antibody. In an expansion cohort of the CheckMate-040 study trial, the agent was explored in 4 groups of patients with HCC, all of whom received nivolumab at 3 mg/kg every 2 weeks: patients without hepatitis who had not received or were intolerant to sorafenib (n = 54); participants who had progressed after receiving sorafenib (n = 58); patients who were infected with hepatitis C (n = 51); and individuals who were infected with hepatitis B (n = 51).

The response rate did not differ by PD-L1 expression status, as assessed by immunohistochemistry on ≥1% versus <1% of tumor cells. As such, there is a still a need to uncover a biomarker to determine which patients respond best to this approach. At this point, this is an important area of research that will help shape the future age of immunotherapy.

It is assumed that tumor specific T cells are the primary executor of anti-tumor immunity; however, there is still little known about the diversity and nature of tumor antigens recognized by a patient’s endogenous immune system. As a result, studies are attempting to better understand other antigens that can be used to help predict response.

New technologies are beginning to advance these research efforts, such as rapid exome sequencing. This approach is adding to the understanding of adaptive immunity having a genetic basis for diversity. Not only does the mutational load appear to be tied to immunotherapy response but also the number and types of antigens present.

Across a handful of small studies, DNA mismatch repair has proved effective at determining response to immunotherapy. Mismatch repair-deficient tumors harbor thousands of mutations that may produce neoantigens that can be recognized and targeted by T cells. This biomarker of response was tested in a study looking at the PD-1 inhibitor pembrolizumab (Keytruda) in patients with gastrointestinal cancers, which was conducted at Johns Hopkins. In studies exploring this approach, response rates have ranged between 45% and 60%.

In addition to this approach, T‑cells with individual clone types have been identified and studied. In one analysis, 100 clones were reviewed that had significant changes between pre-treatment and pre-operation timepoints. Ninety‑four of these were significantly increased following anti‑PD‑1 therapy. These combined markers could help personalize care in the future.

At John Hopkins School of Medicine, immunofluorscopy is being utilized for immunotherapy research in cancer. One study utilizing this approach focused on CTLA-4 and PD-1. The researchers used multiplex immunofluorescent panels to review melanoma for CD8 and Foxp3 regulatory T‑cells, and found that PD-L1 is expressed predominantly on melanocytes. This finding could allow researchers to better detect, target, and personalize immunotherapy, Pardoll said.

Further experiments continue to be conducted in mice models, which are the closest proxy to the human immune system. These mice models can be used to test hypotheses for treatment and to develop biomarkers. As biomarker research continues to advanced, future immunotherapies may also be considered targeted therapy.

<<<

View more from the 2016 AASLD Annual Meeting

The objective response rate (ORR) was 16% among 214 patients who participated in the expansion cohort of the study. Among the 35 patients who responded, 33 had partial responses and 2 experienced a complete response. Another 111 patients had stable disease. ORRs were similar across the groups. The ORR ranged from 12% for patients who had been infected with hepatitis B to 20% for uninfected patients who were sorafenib naïve or intolerant.