Biomarker Developments Continue to Shape Immunotherapy Treatment Paradigm for GI Cancers

Sarbajit Mukherjee, MD, MS

The utilization of monoclonal antibodies such as nivolumab [Opdivo] and Ipilimumab [Yervoy] have become vital in shifting the treatment paradigm in gastroesophageal and gastrointestinal (GI) cancers, according to Sarbajit Mukherjee, MD, MS.

Furthermore, Mukherjee stressed the importance of biomarkers when treating patients with GI cancers, since these have major impact when making immunotherapy treatment decisions.

“[Biomarkers] tell you how the patient was at baseline when you started treatment. We know that chemotherapy can modify the tumor microenvironment, and then you are not sure about how those changes during treatment are going to affect a patient in the future,” Mukherjee said. “If we can have something that is more dynamic and more reflective of what's going on within the patient as they get treatment, those biomarkers are going to be important.”

In an interview OncLive^®, Mukherjee, assistant professor, Oncology, Department of Medicine, member, Tumor Immunology and Immunotherapy Program, Roswell Park Comprehensive Cancer Center and clinical assistant professor, Jacobs School of Medicine and Biomedical Sciences, University at Buffalo, discussed questions that clinicians and patients should take into consideration when using immunotherapy to treat gastroesophageal and GI cancers.

OncLive^®: What are some considerations to keep in mind when utilizing immunotherapy in gastroesophageal cancers?

Mukherjee: The last couple years have been exciting for those treating GI cancers because there have been several new approvals for immunotherapy drugs. What naturally comes to mind is, “When is the right time to use these drugs?” If you look at the recent approvals, nivolumab is approved in the adjuvant setting for esophageal and GE-junction cancers based on the exciting data from the phase 3 Checkmate 577 Study [NCT02743494] for those patients who undergo tri-modality treatment with chemoradiation and do not achieve a complete pathologic CR [complete response].

Nivolumab was found to improve the disease-free survival [DFS] significantly. This approval is not based on tumor histology or biomarker status. So you could use it for both [adenocarcinoma] and squamous cell, regardless of what the PD-L1 status is. This is something we have been commonly using in clinic as well. The tolerability for this treatment is good.

The others are a bit more controversial. If you look at [National Cancer Care Network] NCCN guidelines vs the FDA approvals in the frontline or metastatic setting, the NCCN puts category 1 recommendations for certain PD-L1 cutoffs. However, FDA approval is based on all comers. It’s not based on biomarker status. That’s when it becomes a little bit tricky [in terms of] when we should use it. From a reimbursement perspective, sometimes insurance companies go by what's mentioned on NCCN guidelines.

As for my clinical practice, I don't judge my decision based on biomarker expression alone. Of course, if the patient meets the criteria that is mentioned on NCCN guidelines, then it's a no brainer. But when [biomarker expression] does not fit those criteria, then it becomes a discussion between the physician and the patient. What are the potential risks of using this treatment in conjunction with chemo in the frontline metastatic setting vs the potential risks?

Sometimes PD-L1 expression could be heterogeneous. There's some interobserver variability in how you measure the PD-L1 [and] combined positive score [CPS]. One thing I always mention, if there is a clinical trial available, then that is possibly the best option for the patient. The other thing to keep in mind is you want to check the PD-L1/CPS at an institution that does it regularly, so it becomes more standardized.

Are there any additional immunotherapy regimens that are expected to expand treatment options?

Ipilimumab and nivolumab showed some great promise in squamous cell cancer, but not so much in [adenocarcinoma]. In the future, we might be able to use a chemo-free regimen for frontline metastatic patients for esophageal squamous cell carcinoma, but patient selection is the key. We must carefully select patients and see who would benefit from a chemo-free approach. But the fact that we are even thinking about it, that it might be available for our patients in the future, that alone is so exciting. We couldn’t have thought about it a few years back.

There are several other PD-1/PD-L1 checkpoint inhibitors that have shown efficacy. Especially [at the 2021 ESMO Conference], we saw data regarding frontline use of these agents in several clinical trials, and the results are all consistent. We’re seeing that the chemo and immunotherapy approach is beneficial for most of our patients. Having said that, biomarkers still play a huge role, as we have consistently seen across different studies not every patient benefit from adding immunotherapy.

No matter how little the risk is, there is still some risk for immune-related adverse events. We need to carefully select our patients and we need to do better on biomarkers. Of course, PD-L1, CPS, [tumor mutational burden], and [microsatellite instability] MSI status, they are commonly being used clinically. But biomarker development is going to be a huge area in this field, and over the next few years, we're going to see some activity in that area as well. That will help us select patients better and minimize the risk of toxicity.

Are there any clinical trials currently looking into biomarker development?

All these clinical trials that have been presented, [most] have stratified patients based on PD-L1 expression. Some initial trials looked at tumor cell PD-L1 expression, but the newer trials are looking more at the CPS when you look at the expression on both the tumor cells, as well as immune cells. There are some exciting new biomarkers that could play a role in the future, including the inflammation in the tumor. This is something that some studies have looked at, including some earlier studies like the phase 2 KEYNOTE-059 trial [NCT02335411], and they found that the inflammatory gene signature correlated with response.

We often forget about some very basic clinical biomarkers that could also be useful. These biomarkers need to be rigorously examined prospectively in clinical trials. Some of my own research has looked at obesity as a biomarker. We have shown that obese patients tend to benefit more from the use of immune checkpoint blockade, and this is something that I've been personally working on. Gender, race, all these things could potentially come into play. All the clinical trials, especially the large phase 3 clinical trials, are using these as stratifying markers, based on gender, geographic location, and regular areas like PD-L1 or CPS.

The other important thing that's going to play a big role in the future is if we can have some dynamic biomarkers. The biomarkers that spoken about so far are mostly static biomarkers. You check for PD-L1 expression in the tumor tissue, and then the race, the gender. All of these things are not static.

Something that is going to be very important [to remember, biomarkers] tell you how the patient was at baseline when you started treatment. We know that chemotherapy can modify the tumor microenvironment, and then you are not sure about how those changes during treatment are going to affect a patient in the future. If we can have something that is more dynamic and more reflective of what's going on within the patient as they get treatment, those biomarkers are going to be important.

Sometimes it’s difficult to do multiple biopsies on patients, so if you develop biomarkers that are less invasive, [if you] can draw blood and predict the outcome [of treatment], that's something that's going to be important in the future.

In that regard, circulating tumor DNA is something that is being looked at. It has shown some early promise. And some other researchers are looking at the immune-cell population in the peripheral blood. Sometimes it becomes difficult to look at what's going on in the tumor, but if the clinical trials incorporate serial biopsies, so you know how that’s going to predict the treatment response in the future, that's going to be interesting.

In summary, we need to look beyond PD-L1, MSI, and CPS. We need to look at clinical biomarkers like gender, race, or obesity, and then circulating biomarkers like circulating tumor DNA.

What is your overall take home message?

In gastroesophageal cancers, adjuvant nivolumab is something that completely changed the landscape of treatment. We didn't have any adjuvant treatment after trimodality therapy in esophageal or GE-junction cancer. This is something I’m expecting our clinician colleagues to use more commonly in clinical practice.

Chemoimmunotherapy is going to be the standard of care for most patients. If there is an option for a patient to be enrolled in a clinical trial, that is something that should be discussed. I would always encourage my clinician colleagues to discuss clinical trial options with their patients. For patients, their families, and their caregivers, I always encourage them to talk to their oncologists about these options, or even potentially consider going to a center where there are lots of clinical trial options. All these new approvals these days, they have been possible because patients graciously participated in clinical trials. That’s how we advance science, that's how we progress.