Sarah Hill, MD, PhD.
Immune therapies declare open season on cancer, rousing immune system cells to take up an attack on tumors. But which immune cells join the hunt, which sit it out, and what happens within immune cells that causes them to go on the offensive?
Such questions are especially relevant when immunotherapies show only limited effectiveness against certain forms of cancer. To understand why such treatments sometimes fall short, and how they can be made more potent, scientists need to identify the immune cells that spring into action in response to immunotherapy and the changes those cells undergo.
Dana-Farber scientists accomplished just that in a recent study involving high-grade serous ovarian cancer (HGSOC), the most common form of ovarian cancer and one that has largely resisted immunotherapy drugs known as immune checkpoint inhibitors. Their findings, published in the journal Cancer Research, provide the first detailed look at the mechanism by which these drugs act on immune system cells in HGSOC and point to other drugs that may be more effective against the disease.
“In clinical trials, checkpoint inhibitors haven’t had as much success against HGSOC as other types of cancer,” says Dana-Farber’s Sarah Hill, MD, PhD, the senior author of the study. “To learn why that is, we need to better understand these drugs’ molecular mechanism of action in HGSOC and pinpoint the immune cells they target.”
Novel experimental tools
The study focused on checkpoint inhibitors known as PD-1 and PD-L1 inhibitors. These drugs — the product of discoveries by Dana-Farber’s Gordon Freeman, PhD, and others — work by sparking an immune system attack on cancer cells. Cancer cells hang PD-L1 proteins on their surface as a “Do Not Disturb” sign to the immune system. The PD-1 protein on immune system T cells heeds the sign and slams the brake on an attack. Inhibitors of PD-1 or -L1 are made from antibodies that release that brake.
To probe HGSOC’s ability to withstand these drugs, Hill and her colleagues used some novel experimental tools. These included collections of HGSOC and immune system cells derived directly from patient tumors — called organoid co-cultures — that more closely resemble the content of actual tumor tissue than conventional samples of tumor cells alone. Another tool was a “bispecific antibody,” the biological equivalent of a double-ended wrench.
“All antibodies have two arms, but most target only one protein,” Hill explains. “In the checkpoint inhibitor pembrolizumab, for example, both arms block PD-1. Our bispecific antibody was pretty cool, because one arm targeted PD-1 and the other targeted PD-L1. We hypothesized that it would be more effective against HGSOC than either a PD-1 or PD-L1 inhibitor alone.”
There was an additional reason for optimism. “Not only do bispecific antibodies have this dual mechanistic function but they can also bring immune cells right up to tumor cells,” Hill remarks. “That spatial proximity could increase the strength of the immune attack.”
In a series of experiments with the HGSOC organoid/immune cell co-cultures, the bispecific antibodies proved far better at killing tumor cells than PD-1 or PD-L1 antibodies alone. Encouraging as those results were, they weren’t enough to answer researchers’ main question: what is the mechanism that prompts the immune system to respond to checkpoint inhibitors, and why is it not effective in HGSOC?
For this, researchers turned to another cutting-edge technology, single-cell RNA sequencing, which can determine which genes within a cell are switched on. The investigators treated one set of organoid co-cultures with an inert substance, another with single-target antibodies, and another with bispecific antibodies. Each organoid co-culture contained multiple types of immune cells – CD4+T cells, CD8+ T cells, natural killer (NK) cells, B cells, myeloid cells, and others. After antibody treatment, researchers extracted cells of each type and, using single-cell RNA sequencing, ascertained which types were active against tumor cells and why.
They found that antibodies that target only PD-1 or -L1 failed to unleash two key members of the immune system arsenal – NK cells, the first wave of defenders against disease, and a subset of CD8+ T cells. The bispecific antibody, by contrast, activated both of these.
“The group of CD8+ T cells that we identified were in the early stages of ‘exhaustion’ — meaning they’re showing signs of being unable to engage in an attack on HGSOC,” Hill comments. “We found that the bispecific antibody reverses this exhaustion so they can take up the fight.”
Single-cell RNA sequencing enabled researchers to trace the molecular events that resulted in the activation of NK cells. They found that treatment with the bispecific antibody drove down the cells’ production of a protein called BRD1. That led the researchers to ask whether drug molecules that specifically target BRD1 could have the same NK-stimulating effect. When they tested one such molecule in the organoid co-cultures, it mobilized NK cells and reversed exhaustion in CD8+ T cells. When they tested it in immune competent mouse models of HGSOC, it had the same effect and also reduced tumor growth, suggesting that small-molecule drugs could be as or more effective in prompting an immune attack on cancer as antibody immunotherapies have been.
“Our findings demonstrate that a mechanistic understanding is critical to improving treatments for cancers like HGSOC that don’t respond well to current immune therapies,” Hill observes. “They also show that small-molecule, targeted drugs may offer a viable alternative to antibody-based immunotherapy in some cases.”
Learn more about research happening at Dana-Farber Cancer Institute.