Pivoting Away From the Biomarker Hunt in Gastric Cancer

David H. Ilson, MD, PhD

On the heels of the KEYNOTE-062 trial, exploration to identify actionable biomarkers may not yield the greatest benefit for patients with gastric cancer, said David H. Ilson, MD, PhD. 

"[We need to] move the field forward," said Ilson. "We are focused on PD-1/PD-L1 agents, but we need to look at other targets and newer drugs to try and overcome immune suppression. The biomarker hunt alone is not going to be the answer because at the end of the day, the vast majority of patients do not benefit from these drugs and we already know the patients who will benefit. It is important to identify biomarkers, but I'm not sure it will move the field that much forward." 

Recent trials have evaluated the utility of PD-L1 and microsatellite instability (MSI) status. Results of the randomized, phase III KEYNOTE-062 trial showed that the combination of pembrolizumab (Keytruda) with chemotherapy did not improve progression-free or overall survival (OS) compared with chemotherapy alone as a frontline treatment in patients with advanced gastric or gastroesophageal junction adenocarcinoma.¹

However, in an arm evaluating pembrolizumab monotherapy, the PD-1 inhibitor showed noninferior OS versus chemotherapy in the entire intent-to-treat population of patients whose tumors express PD-L1 (CPS ≥1).

In an updated analysis of patient subsets in the trial, pembrolizumab showed a clinically meaningful improvement for patients with MSI—high (MSI-H) tumors. Though OS was not reached in either arm, the median progression-free survival (PFS) for these patients was 11.2 months with pembrolizumab alone compared with 6.6 months with chemotherapy (HR, 0.72; 95% CI, 0.31-1.68).² In patients with MSI-H tumors who had pembrolizumab/chemotherapy, the median PFS was not reached and was 6.6 months in those who received chemotherapy alone.

"It is clear that it is a small subset of MSI-H patients who benefit from checkpoint inhibitors," Ilson said. "Those are the patients we would use these drugs earlier on with an expectation of a high-degree of effectiveness, but they are not 100% effective."

In an interview with  OncLive  during the International Society of Gastrointestinal Oncology 2019 Gastrointestinal Oncology Conference, Ilson, a medical oncologist at Memorial Sloan Kettering Cancer Center, discussed current treatment options for patients with gastric cancer, the challenges with the KEYNOTE-062 trial, and other immunotherapy agents coming down the pipeline.  

OncLive:  What are the current treatment options for patients with gastric cancer?

Ilson:  We know that adjuvant chemotherapy needs to be added in gastric cancer. The standard for resectable gastric cancer in the United States is pre- and postoperative chemotherapy with the FLOT regimen, which comprises 5-fluorouracil, oxaliplatin, docetaxel, and leucovorin. Adding radiation to the management of gastric cancer does not add benefit, so it's perioperative chemotherapy. In Asia, it is more upfront surgery followed by postoperative chemotherapy. 

For advanced disease, we have the advent of immunotherapy agents. A number of results have come from randomized phase III trials looking at first-, second-, and later-line use of checkpoint inhibitors with mixed results. The most important recent study was KEYNOTE-062, which was a first-line trial of chemotherapy alone, chemotherapy plus pembrolizumab, or pembrolizumab alone in previously untreated patients with gastric cancer. Also, patients had to be ≥1% PD-L1 positive. 

Overall, it was a negative trial. It did not show that adding pembrolizumab to chemotherapy improved survival compared with chemotherapy alone. However, pembrolizumab alone as initial treatment may benefit some subsets of patients, particularly MSI-H patients. In that subgroup, pembrolizumab had a high response rate. These are certainly, data to indicate consideration to use pembrolizumab earlier for MSI-H patients. 

Even though PD-L1 status was somewhat of a predictive biomarker, in microsatellite-stable patients with a combined positive score (CPS) of ≥10%, there was a higher response rate. However, even in those patients, adding pembrolizumab to chemotherapy showed a slightly higher response rate but no improvement in survival compared with chemotherapy alone.

The problem is that the vast majority of those patients progress on pembrolizumab and crossover to chemotherapy early on, arguing that it is not a fair comparison. Virtually all patients receiving pembrolizumab receive chemotherapy within a few months.

There was a suggestion of a survival benefit later in patients who got upfront pembrolizumab, but there was no crossover. Patients who received chemotherapy alone did not have access to immunotherapy later. Based on those data, we can't move pembrolizumab to first-line treatment yet, with the exception for MSI-H patients. Now the debate is whether MSI-H patients should receive checkpoint inhibitor therapy initially or in the second-line setting. 

What other trials are looking at pembrolizumab for gastric cancer?

The [KEYNOTE-061] trial comparing paclitaxel versus pembrolizumab did not show superiority for pembrolizumab over paclitaxel in the second-line setting. Again, the MSI-H patients had a high response rate of 50% to 60% with prolonged PFS. Therefore, in the MSI-H patients, there is a clear indication that this drug is a high priority in second-line use. 

Paclitaxel [alone] is no longer the standard second-line treatment. It is paclitaxel plus ramucirumab (Cyramza). 

The approved indication of pembrolizumab is in later-line settings. Right now, checkpoint inhibitors are approved agnostic of tumor type. If a patient has an MSI-H cancer and is resistant to chemotherapy, we can use checkpoint inhibitors in second- or third-line treatment. 

In cancer, we have third-line indication approval for pembrolizumab in PD-L1—positive patients who are refractory to chemotherapy. Recent data do not indicate we should move pembrolizumab earlier in non–MSI-H patients. It should probably [stay in the] later-line setting. Most patients progress on these drugs early. There is a subset of 10% to 15% of patients who do benefit with potentially durable control and responses. 

What are the challenges in treating these patients now?

Trying to identify better biomarkers to select patients [is a challenge]. Some patients do not respond to checkpoint inhibitors even though they are MSI-H. Then, there is the possibility of hyperprogression, when exposed to these drugs, for a very small percentage of patients.

There is a lot of interest in making an immunologically “cold” tumor environment “hot” to be more receptive to immune agents. The challenge will be identifying new drugs and targets. Biomarker development remains a priority, but because so few patients benefit from these drugs, trying to parse out who those few are won’t move the field forward. 

Are any other immunotherapy drugs being looked at?

In gastroesophageal cancer, there are two targets in key phase III trials. One is tumor-specific Claudin-18.1, a gap junction protein that is highly overexpressed in gastric cancer. Claudin-18.1 is targeted by an antibody called zolbetuximab, which may be involved in stimulating antibody-dependent cellular cytotoxicity. There are encouraging phase II data with zolbetuximab combined with chemotherapy in Claudin-18.1—positive patients. There are two ongoing phase III trials combining zolbetuximab with chemotherapy in high Claudin-18.1-expressors. 

The FIGHT trial is looking at bemarituzumab, a FGFR-targeted, receptor-blocking antibody, combined with FOLFOX compared with FOLFOX alone in biomarker-selected patients. It is a small subset of patients, but the results may be intriguing.

We are moving away toward rushing to large phase III trials to ensure we identify a viable target and have strong, randomized phase II data before taking a drug further. There is a greater interest in global collaboration because this is a rare disease in the United States. Important studies are being done in Japan, Korea, and China where gastric cancer is very common.

References

1. Merck Provides Update on Phase 3 KEYNOTE-062 Trial Evaluating KEYTRUDA  (pembrolizumab) as Monotherapy and in Combination with Chemotherapy for First-Line Treatment of Patients with Advanced Gastric or Gastroesophageal Junction Adenocarcinoma. Merck. Published April 25, 2019. https://bit.ly/2IXZtwG. Accessed April 25, 2019.
2. Shitara K, Van Cutsem E, Bang Y-J, et al. Pembrolizumab with or without chemotherapy vs chemotherapy in patients with advanced G/GEJ cancer (GC) including outcomes according to microsatellite instability-high (MSI-H) status in KEYNOTE-062. Ann Oncol. 2019;30(suppl_5). doi: 10.1093/annonc/mdz394.035.