Hochster Shares Best Practices on Molecular Markers and Sequencing in Gastric Cancer

Howard S. Hochster, MD, discusses molecular markers for patients with gastric cancer, using gene expression profiles as biomarkers, and choosing between anti–PD-1 and anti–CTLA-4 agents.

For patients with gastric cancer, utilizing molecular markers and genomic profiling is becoming more crucial in guiding treatment decisions, according to Howard S. Hochster, MD.

“For patients with gastric cancer, immunotherapy has become part of the armamentarium in either the first-line or in maintenance setting,” Hochster said. “It depends on [a patient’s] anti–PD-1 score on the tissue staining, but it's clear that people who have high scores benefit a lot from immunotherapy.”

HER2 positivity is another pivotal marker in the gastric cancer population that can be used when choosing a treatment approach. Fam-trastuzumab deruxtecan-nxki (Enhertu) is available for the treatment of adult patients with locally advanced or metastatic HER2-positive gastric or gastroesophageal junction adenocarcinoma who have received a previous trastuzumab (Herceptin)-based regimen.

In an interview with OncLive, Hochster, who is an associate director of clinical research and director of gastrointestinal oncology at Rutgers Cancer Institute of New Jersey and director of oncology research for RWJBarnabas Health, discussed molecular markers for patients with gastric cancer, using gene expression profiles as biomarkers, and choosing between anti–PD-1 and anti­–CTLA-4 agents.

OncLive®: At what point in a patient’s treatment course are you evaluating molecular markers and in this patient population, and with what assays?

Hochster: The use of molecular markers, specifically next-generation sequencing [NGS], is becoming more and more common, and its utility depends on what kind of cancer we are treating. For example, in pancreatic cancer, we are interested in patients who may have BRCA mutations, but [those patients are] uncommon. About 95% of patients have RAS mutations, [which] we [currently] cannot do anything about. Therefore, [that molecular marker] is not that helpful.

However, in colorectal cancer [CRC] and gastric cancer, there are important mutations, and lung cancer therapy is built around certain translocation mutations. Those are areas where you cannot even get started [with treatment] without knowing [a patient’s mutational status]. In CRC, if we know the [patient’s] RAS status, if they have a KRAS or nRAS mutation, we know what antibody to pair with chemotherapy. These have become standard practice.

What is going to be even more helpful is circulating tumor DNA [ctDNA], which uses NGS to look at how much DNA is coming from the tumors, and what mutations we can maybe treat as the tumor evolves. Sometimes under the pressure of treatment, we see that the mutational landscape evolves, and we can use treatments that we did not [previously] think we could use.

How have some recent approvals affected the frontline landscape of gastric cancer as well as treatment sequencing for patients with and without HER2-positive disease?

Today in the treatment of patients with gastric cancer, we have a lot of options. First, we need to know if patients have HER2 overexpression as [there is] in [patients with] breast cancer. It is not nearly as common in gastric cancer, [but it occurs in approximately] 25% to 30% of patients. However, they have a [slightly] different pathway. We are using anti-HER2 drugs, either [as] single [therapy], doublet [therapy]—maybe [in combination with] immunotherapy as well.

How strongly are you following the National Comprehensive Cancer Network (NCCN) guidelines when it comes to cutoffs for treatment selection with immunotherapy?

The NCCN guidelines are evidence-based, and are good guidelines to help us. Each patient is individual, and we need to decide what is best for them. My personal approach is to look at their PD-L1 expression score and try to decide if they can tolerate chemotherapy, or if they can tolerate immunotherapy, and then make a recommendation based on those criteria.

What are your thoughts on using gene expression profiles as biomarkers, perhaps in combination with more established markers, such as PD-L1, to better guide treatment selection?

When we are choosing therapy today, we have a lot of options to help guide us besides [a standard] chemotherapy regimen. We are able to look at the genomic profiling of the tumor, and we are always looking for actionable targeted mutations, which could provide a maintenance approach or an easier therapy besides chemotherapy.

Many times, we have a challenge of getting [agents] approved because these mutational changes are not confined to any 1 specific tumor. If [the tumor is] rare—for example, if we have a patient with HER2-overexpression in cholangiocarcinoma—it is difficult to get these drugs approved, even though they are approved in breast cancer. The first [challenge is] trying to find out if there is a targetable mutation, or an actionable mutation, and the second challenge is getting that drug [approved] for the patient today.

After seeing the results of the third arm from the phase 3 CheckMate 649 trial (NCT02872116), do you believe dual immunotherapy has a role to play?

The question of using an anti–PD-1 and anti–CTLA-4 drug is always a good one. It is a tradeoff of toxicity vs benefit. In general, using more than 1 checkpoint inhibitor tends to be better for response, but it also comes at a cost. However, this is just the beginning. There are a lot more checkpoints out there, and more checkpoint inhibitory drugs besides PD-1 and CTLA-4. We are looking forward to having other checkpoint inhibitors, [and] maybe we will have different doublets that will provide better responses with less toxicity.

Do you tend to reach for trastuzumab deruxtecan in the second-line setting for patients with HER2 positivity?

Initially, we tend to use an anti-HER2 antibody such as trastuzumab. However, the antibody-drug conjugate [ADC] trastuzumab deruxtecan is quite active. We have tried to use that as a second-line therapy if we do not have a clinical trial available.

Given the data from the phase 2 DESTINY-Gastric02 trial (NCT04014075) from the 2021 ESMO Congress, what is your expectation for the phase 3 DESTINY-Gastric04 trial (NCT04704934)?

The DESTINY-Gastric02 looked at trastuzumab deruxtecan in patients with previously treated HER2 overexpressing gastric cancer. In the DESTINY-Gastric04 study, patients be randomized to trastuzumab deruxtecan against paclitaxel and ramucirumab (Cyramza). I am sure that the ADC will be less toxic, easier for patients, and hopefully it will be more effective, as well.

Is there anything else about the current gastric cancer paradigm that you would like to share or emphasize?

There is still a lot to learn about gastric cancer. We know that occasionally patients have mismatch repair enzyme deficiency, HER2 overexpression, or viral infection that may make [a patient] more susceptible to immunotherapy. However, there is a lot more involved to understanding the biology. We look forward to seeing some of these basket trials looking at different ways of stratifying patients that will hopefully translate into better clinical outcomes.

I want to emphasize that this is a great opportunity for our service line to come together. The academic medical system between Rutgers and RWJBarnabas Health is going to become a major force for state-of-the-art cancer care and health care delivery in general, plus amazing amounts of research and utilizing some of Rutgers’ opportunities. It is an amazing, large, and broad university with many opportunities to help make our patients lives better.