Amid Success, Immunotherapy Faces Challenges in GI Malignancies

Gregory L. Beatty, MD, PhD

Immunotherapy has shown promise in select patients with gastrointestinal (GI) malignancies, but challenges remain in immune recognition and optimal sequencing, says Gregory L. Beatty, MD, PhD.

Cancers that form in the GI tract often avoid immune recognition, but patients who have mismatch repair-deficient (dMMR) cancers have larger mutational loads, making it easier for the immune system to recognize those tumors. The PD-1 inhibitor pembrolizumab (Keytruda) is approved by the FDA for the treatment of patients with unresectable or metastatic, microsatellite instability-high (MSI-H) or dMMR solid tumors that have progressed after prior treatment, which includes patients with MSI-H or dMMR colorectal cancer (CRC) who have progressed on a fluoropyrimidine, oxaliplatin, and irinotecan.

More recently, the FDA granted a priority review to a supplemental biologics license application for the combination of nivolumab (Opdivo) and ipilimumab (Yervoy) for the treatment of adult patients with MSI-H or dMMR metastatic CRC following progression on a fluoropyrimidine, oxaliplatin, and irinotecan. This was based on findings from the phase II CheckMate-142 trial, where the investigator-assessed overall response rate in the cohort treated with the combination was 55% (95% CI, 45.2%-63.8%) at a median follow-up of 13.4 months, with 31% of patients having stable disease. The disease control rate for ≥12 weeks was 80%.

Nivolumab monotherapy was approved by the FDA in August 2017 for the treatment of adult and pediatric patients with MSI-H or dMMR mCRC that has progressed following treatment with a fluoropyrimidine, oxaliplatin, and irinotecan.

However, the majority of patients with GI malignancies still do not respond to immunotherapy, says Beatty, specifically those with pancreatic and gastroesophageal cancers.

OncLive: Can you provide an overview of your presentation?

Beatty, assistant professor of medicine, director of Translation Research, Pancreatic Cancer Research Center, University of Pennsylvania Perelman School of Medicine, discussed the promise for immunotherapy in GI cancers during a presentation at the 2018 OncLive^® State of the Science Summit™ on Gastrointestinal Cancers. In an interview during the meeting, he shared insight on the challenges with sequencing immunotherapies in light of recent advancements in the field.Beatty: There is a lot of promise for immunotherapy in GI malignancies, but there have been a lot of challenges. Part of that has emerged because the gut is where 70% of the immune system resides. One of the key challenges for the immune system is deciding whether to activate or to remain tolerant. There are a lot of pathogens that are in the gut, for which the immune system must [attack] on a daily basis. Many cancers in the GI tract have found ways in which to use that tolerance mechanisms to evade immune recognition. That creates a challenge for utilizing immune therapy. There are subsets of patients that appear to respond to immunotherapy, and those are ones who have dMMR [cancer]. Those tumors usually have larger numbers of mutations, so those tumors look more nonself, which alerts the immune system to attack. There, checkpoint therapies that target PD-1/PD-L1 have been quite successful.

Most patients with GI malignancies do not respond. The majority of [patients with] pancreatic cancer do not respond to immunotherapy. In CRC, [tumors] that do not have dMMR usually do not respond. For gastroesophageal cancer, the majority still do not respond.

Were there any updates at the 2018 AACR Annual Meeting of immunotherapy in GI malignancies?

Finding ways in which to condition tumors for improved responses to immune therapy is the next step. It is not just combinations of those therapies, but strategically sequencing therapies to change a tumor that is resistant to one that is sensitive. Whether it is sensitive to immune therapy or cytotoxic therapies, inflammation in the immune system is a key player in defining that balance.There are 2 new players in the field, so to speak: epigenetic therapies and rational ways to use targeted therapies. Here I presented some work using a MEK inhibitor, which seems to condition tumors to be more immunogenic, so they are more visible to the immune system because they upregulate MHC expression, which is important for T cells to recognize. However, the MEK inhibitors also seem to improve T-cell responses by improving the function of those T cells—not necessarily their ability to expand in the priming setting, but their ability to act in the effector setting.

Would these approaches be applicable in a broader range of patients?

Epigenetic modulators are also garnering a bit of interest, and that is because of the chromatin remodeling that takes place with the epigenetic modulators, which allows for new genes to appear and be expressed. That seems to lead to an interferon response, which can do very similar things to subtarget therapies. This means those tumors become more visible to the immune system, and harnessing that change in the tumor is now, at least at the preclinical level, pretty exciting for making immunotherapy work better. The next challenge is seeing whether a similar biology is seen in patients treated with these immune epigenetic modulators like HDAC inhibitors.One of the big goals is to take T-cell “cold” tumors and try to turn them into T-cell “hot” tumors. One of the challenges there is the fact that if you look at GI malignancies, the immune contexture of the tumors can be quite diverse. Therefore, there can be regions that are hot and regions that are cold.

What are your thoughts on the priority review designation of the combination of nivolumab and ipilimumab in CRC?

In reality, you can’t call them cold or hot. It becomes the question of what is the rate-limiting step. Part of that may actually be at the priming phase as well. We have found that if you have a strong enough priming agonist, such as delivering 2 myeloid agonists, that you can invoke a T-cell response that is effective independent of any immune checkpoint modulators. That suggests [that we should toggle] that equation to get a strong enough T-cell response going. That may be enough to get over that threshold—to get a domino effect of an immune response thereafter. There are different places where you can target that other priming.That is a logical next step for the field. PD-1 shows responses there, and we know that these 2 agents can act synergistically. CTLA-4 can help at the priming phase, and PD-1 can help at the effector phase of the immune response. Combining inhibitors that target both of those makes logical sense.

One of the challenges with that combination has been the adverse events (AEs). There will be ways in which to tackle that. Certainly, in CRC, lessons from melanoma and lung cancer will be able to be applied. There are data around sequencing, leading to fewer AEs than giving the combination together.

What other ongoing trials are you excited about?

One of the other challenges that we still have using nivolumab and ipilimumab or other combinations, is when to use them together. The use of combination therapy is not necessary for all patients. Some of that will emerge out of [data from] patients who respond or do not respond to initial PD-1 therapy. If they do respond, continue them on single-agent therapy. If they do not respond, then you try to harness a response by adding in the second therapy. The challenge is whether the sequence of the therapies is correct. That still needs to be worked out, I believe. What is the algorithm that one uses to move forward? The AEs can be debilitating for some patients.There is an approach that is attractive for incorporating immunotherapy into a standard-of-care regimen. It is a trial that is being run out of the University of Pennsylvania by Dr Kim A. Reiss Binder. The idea is to take patients with metastatic pancreatic cancer who are responding to a platinum-based therapy; they have stable disease or a partial response for 4 months or longer. Rather than just continuing that therapy until they progress, patients come off therapy and go onto a PARP inhibitor in combination with a checkpoint inhibitor.

Is there anything else that you would like to highlight?

Those patients are responding. How do we maintain that response? How can we loop in the immune system to make a better or longer response for the patients? Rather than utilizing the same approach that we have done, which is to treat until progression, this study is saying to treat until that inflection point has been hit and then give a different therapy. You are conditioning tumors to hopefully harness an immune response and maintain that immune response with a new therapy. Those types of strategies of clinical trials designs will be very informative moving forward.Sequencing is the biggest aspect. The other piece is the fact that a lot of us are using tissue-based biomarkers to inform treatment responses to immunotherapy. One of the challenges of that in GI malignancies is that the immune contexture of these tumors is not homogenous. Utilizing those biomarkers can be challenging and, ultimately, what we need to have are liquid-based approaches that are noninvasive. For some malignancies, that is more important than for others.

A perfect example of that is glioblastoma. Patients have to undergo a biopsy to understand what type of malignancy they have, but those biopsies come with risks. The risk of a bleed can be quite detrimental, and patients can decline quite rapidly because of it. Having ways in which we can look at liquid biopsies and other approaches, such as noninvasive imaging strategies, are going to be absolutely essential for immune-based therapy to be implemented across a wide range of malignancies.

Overman MJ, Lonardi S, Wong KYM, et al. Durable clinical benefit with nivolumab plus ipilimumab in DNA mismatch repair-deficient/microsatellite instability-high metastatic colorectal cancer. J Clin Oncol. 2018;36(8):773-779. doi: 10.1200/JCO.2017.76.9901.