Additional Immunotherapy-Predictive Biomarker Strategies on the Horizon

As cancer immunobiology becomes better understood, new predictive biomarkers are likely to become translated into clinical tools. In some cases, anti–PD-L1 antibodies exhibit antitumor activity even when tumors do not express detectable PD-L1 protein at all, for example.

A recent study by researchers from the United States and China might explain why this type of activity can occur.5 In preclinical experiments, they showed that T cells are not the only immune cell populations affected by PD-L1; natural killer (NK) cells also express PD-L1 in some people with acute myeloid leukemia (AML). PD-L1–positive NK cells exhibit antitumor activity, they noted. When bound by an anti–PD-L1 antibody, the antitumor activity of NK cells was enhanced whether or not the tumor cells themselves expressed PD-L1.5

Related Articles

“Natural killer cells are the body’s first line of defense against cancer and viral infections,” said senior coauthor Michael A. Caligiuri, MD, president of City of Hope National Medical Center in Duarte, California, in a press release. “Tumors have developed mechanisms to circumvent NK cells and T cells. We believe PD-L1 expression on NK cells identifies tumors that could be susceptible to destruction by NK cells, thereby providing a new immunotherapeutic avenue to explore.”

The most promising strategy is to use multiple biomarkers that reveal both tumor immune phenotype (like PD-L1 expression) and measures of tumor mutation burden, according to Dr Maleki.

“In immunotherapy, the drugs do not target the tumor itself,” he noted. “You’re targeting the immune system. The nature of the drugs is different, so obviously the biomarkers need to be different too. What activates the immune system — or the T cells that are targeted with these drugs — are foreign antigens. Because tumors undergo mutation, they create neoantigens that look foreign to the immune system even though they’re derived from our own cells.”

Tumor mutation burden is a proxy for the number of likely neoantigens present, and should therefore predict a tumor’s visibility to the immune system and theoretically, the likely efficacy of immunotherapy.

“If there’s a higher tumor mutation burden, there is a higher likelihood of presence of neoantigens that can be recognized by the immune system,” Dr Maleki said.