Novel Immunotherapy Approaches Emerging, With Potential for Difficult-to-Treat Tumors

Novel Immunotherapy Approaches Emerging, With Potential for Difficult-to-Treat Tumors

Dr. Michael A. Postow speaks during General Session 7.
With numerous approvals for a variety of different types of agents and therapies, including immune checkpoint inhibitors, CAR T-cell therapy, and others, immunotherapy is now well established within oncology. As the field matures, new targets, combinations, and concepts have begun to emerge, with particular interest in expanding the uses of immunotherapy into malignancies that have so far proven largely resistant to the approach. A group of experts discussed some of those new targets and concepts, including the use of IDO inhibition and optimal ways to target CD40, during a session held on January 27.

“What do we mean by ‘new’?” Michael A. Postow, MD, of Memorial Sloan Kettering Cancer Center, asked. He answered the question with some possibilities: we can “teach an old dog new tricks,” or we can “get a brand new pet.” In other words, we can take our existing immunotherapies and alter them to increase their function, or use the antibodies to create prodrugs that only bind in the desired microenvironment. Or further still, we can explore new T-cell regulators either through enhanced costimulation or by blocking other checkpoints. Dr. Postow discussed some of the newer T-cell checkpoint inhibitors and how they might be used both as monotherapy and in combination with other immune checkpoint inhibitors.

With regard to the “old dogs,” Dr. Postow discussed how there are some different views of the CTLA-4 checkpoint. “There is some thought that CTLA-4 induces intratumoral Treg (regulatory T cell) depletion,” he said. One of the theories of how ipilimumab works is that by binding to CTLA-4 expressed on Tregs, macrophages are recruited, and eventually this results in antibody-dependent cytotoxicity and depletion of the intratumor Tregs.

“Are there ways that we could enhance this antibody-dependent cytotoxicity?” Dr. Postow asked. For example, it may work to modify the Fc component (non-fucosylated ipilimumab), which in theory could deplete more Treg cells and enhance the cytotoxicity. Dr. Postow said this concept remains in early development, with no hard data yet on its efficacy.

Another approach would be to adjust existing agents such that their effects only occur where needed, meaning in the tumor microenvironment. Dr. Postow described using an agent known as a probody, where the antibody has, essentially, a mask over it so that it cannot recognize its target. But it also has a linker, and when the probody arrives in the tumor microenvironment, dysregulated proteases cleave off the mask and allow the agent to bind to its target.

“It doesn’t bind PD-1 everywhere, but binds only where the proteases are activated and remove that mask,” Dr. Postow said. “The idea is to relieve the toxicity of some of these checkpoint antibody approaches.”

He also discussed the “new pet” approaches, meaning either “turning up” the activating T-cell receptors such as OX40, CD137, and others, or blocking the inhibitory receptors—of which there are many aside from the commonly targeted CTLA-4 and PD-1.

On the costimulatory side, some early development is ongoing with agents targeting the CD27 receptor, for example. CD27 is expressed constitutively on unstimulated T cells, and it binds to CD70 on antigen-presenting cells. Varlilumab is currently in early trials using this approach.

OX40 is another costimulatory molecule; engagement of this receptor increases proliferation, effector function, and survival of T cells. A phase I trial of an agent known as 9B12 has been published, and Dr. Postow said it was promising in particular because of its tolerability; no maximum tolerated dose was reached, and two patients with melanoma had mixed responses to the agent. Other trials are ongoing both as monotherapy and in combination with PD-1/PD-L1 agents.

Dr. Kunle Odunsi speaks during General Session 7.
“There are some considerations with costimulation that may be different from other checkpoint inhibitors, such as timing and dose,” Dr. Postow said. There is preclinical evidence that with too high a dose, there may be inferior antitumor immunity because of a hyperactivation of T cells or enhanced apoptosis. “We should be guided by preclinical literature on how best to move forward,” he said.

On the inhibitory side, an agent targeting LAG-3 known as relatlimab has been tested in combination with nivolumab in patients with melanoma who progressed on prior immunotherapy. The study was promising, with some objective responses seen, although it remains unclear if LAG-3 expression might be used as a marker of response to the therapy.

“What’s the best checkpoint?” Dr. Postow asked, noting that this is an increasingly important question. “We’re discovering more checkpoints all the time. How do we know what really to pursue, and what level of evidence is enough to move forward?”

He also pointed out that with new approaches and targets, a second step often seems to be to combine the new agent with existing PD-1/PD-L1 agents, and this may not always be the best approach.

IDO Inhibition in Ovarian Cancer

Kunle Odunsi, MD, PhD, of the Roswell Park Cancer Institute, spoke about the possibilities of IDO inhibition in ovarian cancer. He said that tryptophan catabolism is a pivotal regulatory of innate and adaptive immunity, a pathway that involves IDO1, IDO2, and TDO, and research has suggested that the IDO pathway may be a critical mechanism of immune suppression in this malignancy.

Dr. Odunsi’s group identified a panel of genes and a gene signature that they believe may be significant for understanding ovarian cancer responses to immunotherapy. They arrived at an IDO and tumor-infiltrating lymphocyte (TIL) signature that can identify four categories of patients: IDO-low/TIL-low, IDO-high/TIL-low, IDO-low/TIL-high, and IDO-high/TIL-high.

Dr. Katelyn T. Byrne speaks during General Session 7.
In a cohort of 265 patients with ovarian cancer, these categories had prognostic significance. The IDO-low/TIL-high patients had a progression-free survival of 30 months, compared with only 19 months for other groups (p = 0.0000822); the same was true for overall survival, which was 59 months for the IDO-low/TIL-high group and 44 months in other groups (p = 0.001).

Immunohistochemistry revealed that tumoral IDO1 expression inversely impacts TIL infiltration, Dr. Odunsi said. In a group of 198 patients with ovarian cancer, the correlation coefficient for IDO1 expression and TIL infiltration was 0.41 (p = 1.9 x 10-11).

“Perhaps reducing IDO enzyme activity might be associated with clinical benefit in ovarian cancer,” Dr. Odunsi said. They examined several inhibitors of IDO, focusing on epacadostat, an agent with promising preclinical and animal model data. They found that the drug was efficient at suppressing IDO activity in ovarian cancer cells, and fared better than 1-methyltryptophan.

“The IDO blockade results in significant changes in gene expression,” Dr. Odunsi said, noting that the most important pathways affected were metabolism and stress response. There are clinical trials now underway, he said, that will further evaluate the IDO inhibition approach, including attempting to answer the question of whether IDO inhibition can enhance vaccine efficacy.

“It will be important to think of combinations, as well as the sequence” of therapy in relation to immune checkpoint blockade, Dr. Odunsi said.

CD40 and Pancreatic Cancer

Katelyn T. Byrne, PhD, of the University of Pennsylvania, described some of the newest research on the possibilities of using CD40 as a target to attack pancreatic ductal adenocarcinoma (PDA).

“This cancer is very resistant to every type of therapy we’ve been trying, and it is especially resistant to immunotherapy,” Dr. Byrne said. “Pancreatic cancer is also increasing in incidence and mortality, so we’re hoping we can develop better therapies.”

A panel discussion takes place during General Session 7.
She said that in many malignancies, there is a “pool” of exhausted T cells that can be rescued by giving anti–CTLA-4 or anti–PD-1/PD-L1 therapy. “We think this pool doesn’t exist yet in pancreatic cancer,” Dr. Byrne said. “We want to prime the T-cell response.”

Agonistic CD40 antibodies might be used to activate the T-cell responses in patients with PDA. PDA is an immunologically “cold” tumor, with very few T cells making it into the tumor microenvironment. In a mouse model, agonistic CD40 antibody combined with gemcitabine and nab-paclitaxel can essentially flip the tumor from “cold” to “hot.” Dr. Byrne said that many mice can be cured of their tumors using this approach, and a challenge with a second tumor showed they had a protective memory response against those new challenges.

Two clinical trials are ongoing in this field now. In one phase I study, the anti-CD40 agent RO7009789 either alone or in combination with gemcitabine and nab-paclitaxel will be given as neoadjuvant therapy in patients with resectable pancreatic tumors. Another, a phase Ib/II study in patients with metastatic PDA, chemotherapy will be combined with an anti-CD40 agent along with an anti–PD-1 agent; Dr. Byrne said this is currently enrolling at the University of Pennsylvania and hopefully will open in other sites soon.

Dr. Byrne’s group also wanted to understand why the approach was failing in some mice. They were able to inject tumor clones that had either high or low T-cell infiltration and found that mice injected with a T-cell “low” clone treated with those same combination therapies fared poorly, all eventually succumbing to the tumor burden. Those injected with a T-cell “high” clone, meanwhile, were receptive to the therapy and again developed a protective memory response. “T-cell infiltration is really dictating response to therapy,” Dr. Byrne said. “If you can target those negative regulators […] you can flip the tumor from cold to hot, and therefore render it sensitive to therapy.”

–Dave Levitan