乌鸦传媒

Expert Critique

FROM THE ASCO Reading Room

Aparna Parikh, MD, MS Attending Physician Massachusetts General Hospital

Gastric cancers (GC) and gastroesophageal (GEJ) cancers are heterogeneous with distinct molecular subtypes and an overall dismal prognosis with chemotherapy and targeted therapy offering only modest benefits. Given the benefit of immune checkpoint inhibitors (ICIs) in tumor types such a melanoma and NSCLC, a concerted effort has been made in GC and GEJ to explore the effects of ICIs. Within the various molecular subtypes, MMR-deficient tumors derive clear benefit from ICIs and pembrolizumab is approved in the setting.

More recently, other biomarkers of response to ICIs have been identified such as programmed death-ligand 1 (PD-L1), and pembrolizumab was approved for treatment of patients with PD-L1鈥損ositive recurrent or advanced GC or GEJ cancers in the third line based on one cohort of the Keynote-059 study. However, the majority of patients do not respond to ICIs. Pembrolizumab was also approved for patients with recurrent, locally advanced, or metastatic esophageal cancer but this was for SCC only and tumors whose tumors express PDL-1 (Combined Positive Score [CPS] 鈮�10), after disease progression after at least one therapy. Given that responses are still minimal, extensive efforts are underway to enhance the activity of ICIs in GC and GEJ.

As discussed in the article below, two relevant recent studies were presented at the 2019 ASCO annual meeting.

The KEYNOTE-059 trial of pembrolizumab in advanced GC and GEJ cancers confirmed activity of pembrolizumab in the first-line as a single agent and in combination in the PD-L1-positive CPS over 1, but further studies are needed to compare the combination versus monotherapy in a biomarker-enriched cohort. This study follows the Keynote-062 study in first-line PDL-1-positive patients demonstrating that pembrolizumab was not inferior to chemotherapy in terms of overall survival (OS) but the most OS benefit came for patients with a CPS score over 10. Disappointingly, pembrolizumab in combination with chemotherapy was not superior to chemotherapy alone in either the CPS1 or CPS10 group. The data suggests that, particularly in patients for whom chemotherapy may not be a first-line option and who have a CPS score of greater than 1, pembrolizumab is an option and may be preferred in the CPS over 10 patients.

Given the disappointing results with pembrolizumab in combination with chemotherapy, other combination studies are being explored. One such study was an investigator-initiated study by Dr. Janjigian, et al., which was a 37-patient phase II trial of first-line pembrolizumab in combination with trastuzumab, capecitabine, and oxaliplatin in patients with HER2-positive metastatic esophageal, and gastric adenocarcinoma. The primary endpoint was 6- month PFS. At a median follow-up of 10 months, the ORR was 88.6% and the 6-month PFS rate was 74% and there was no correlation between PD-L1 status and PFS or OS. The ORR is an impressive ORR, better than has been seen to date in this setting and has led to initiation of the Phase III Keynote-811 trial, a global, randomized, double-blind study of pembrolizumab, trastuzumab, and chemotherapy versus placebo, trastuzumab, and chemotherapy.

Immunotherapy has slowly been making inroads in gastric and GEJ cancers, and pembrolizumab is a viable option in several settings: (1) third-line or beyond PD-L1 positive CPS over 1 gastric or gastroesophageal cancer; (2) second-line PD-L1-positive CPS score over 10 in adenocarcinoma; (3) second-line all squamous cell esophageal cancer; and (4) first-line in PD-L1-positive CPS score over 1 gastric cancer.

Ongoing combination approaches are underway as well as attempts to identify biomarkers of response but more importantly, attempts to invoke a response in the patients with low CPS score. Ongoing trials need to consider biological differences such as Asian versus non-Asian populations given the known biological differences and will need to account for the high CPS patients who seem to benefit the most.

Full Critique

Novel immunotherapy drugs such as checkpoint inhibitors have been paradigm changing in melanoma, non-small cell lung cancer, and urothelial cancers. Now evidence points to improved efficacy in patients with gastroesophageal cancer -- in particular, those patients with biomarkers associated with response to treatment, such as programmed death-ligand 1 (PD-L1).

For patients with advanced gastric or gastroesophageal cancers, chemotherapy and targeted therapies such as trastuzumab and ramucirumab have yielded modest improvements in survival, but no long-term cure. Pembrolizumab, already approved for treatment of patients with PD-L1–positive recurrent or advanced gastric or gastroesophageal junction adenocarcinoma who have received multiple rounds of chemotherapy, is now also approved for patients with recurrent, locally advanced, or metastatic squamous cell carcinoma of the esophagus and tumors expressing PD-L1 that progressed after one or more lines of systemic therapy.

As reported at the 2019 ASCO annual meeting, the of pembrolizumab in advanced gastric and gastroesophageal cancer included patients in three different cohorts: cohort 1 had received two or more prior lines of chemotherapy with pembrolizumab alone; and cohorts 2 and 3 treated patients in the first-line setting.

The primary endpoint of cohort 1 was safety and overall response rate (ORR) in all patients and those with a PD-L1-positive combined positive score (CPS) over 1. After a median follow-up of 6 months, the ORR was 11.6%, which was higher for PD-L1-positive patients (15.5%) than for PD-L1-negative patients (6.4%), reported Zev A. Wainberg, MD, of the David Geffen School of Medicine at UCLA, who presented the results.

"The long-term analysis data presented showed durable responses in PD-L1-positive groups," commented the discussant for the study, Osama E. Rahma, MD, of Dana-Farber Cancer Institute in Boston. "This study showed higher and durable response rates in PD-L1-positive CPS over 1 in gastric and gastroesophageal cancer. The safety profile was consistent with prior reports. I think pembrolizumab remains the standard of care in third-line and beyond in PD-L1-positive CPS score over 1 in gastric and gastroesophageal cancer."

In cohort 2, 25 patients received pembrolizumab plus cisplatin and fluorouracil or capecitabine in combination. After a median follow-up of 14 months, the ORR was 60%, which again was higher in PD-L1-positive patients (73.3%) compared with PD-L1-negative patients (37.5%). In cohort 3, 31 patients who had PD-L1 positive tumors received pembrolizumab alone. After 21 months of follow-up, the ORR was 25.8%.

In all three cohorts, the rates of grade 3-5 treatment-related adverse events were 18%, 80%, and 26%, respectively. This led to discontinuation in 2% and 12% of patients in cohorts 1 and 2, respectively; two patients died in cohort 1, and there were no discontinuations or deaths in cohort 3.

"There is promising activity of pembrolizumab as a single agent in PD-L1-positive CPS over 1 in first-line gastroesophageal cancer, and in combination with chemotherapy," said Rahma, who noted that cohorts 2 and 3 had small sample sizes.

He also discussed the results of the in PD-L1-positive patients with CPS over 1. "This abstract showed that pembrolizumab was not inferior to chemotherapy in terms of overall survival (OS) in the first-line setting and had higher magnitude of OS in CPS over 10. However, pembrolizumab in combination with chemotherapy was not superior to chemotherapy alone in terms of OS in CPS 1 or in CPS 10," said Rahma. "I would consider in my clinic pembrolizumab as first-line for PD-L1-positive gastric cancer, especially if the CPS score is over 10, and in patients who are unfit for chemotherapy."

Pembrolizumab in Combination

Also presented at the poster discussion session was a phase II trial of with trastuzumab, capecitabine, and oxaliplatin in 37 patients with HER2-positive metastatic esophageal and gastric adenocarcinoma. The primary endpoint was 6 months progression-free survival (PFS).

After a median follow-up of 10 months, the ORR was 88.6% and the 6-month PFS rate was 74%. There was no correlation between PD-L1 status and PFS or OS, reported Yelena Y. Janjigian, MD, of Memorial Sloan Kettering Cancer Center in New York City.

In his discussant remarks, Rahma again noted the small sample size, but said the results show "a promising signal."

The preliminary safety and efficacy results of this study led to initiation of the definitive , a global, randomized, double-blind study of pembrolizumab, trastuzumab, and chemotherapy versus placebo, trastuzumab, and chemotherapy. This study is currently open for enrollment.

Rahma said that based on the data presented at the ASCO meeting, pembrolizumab should remain the standard of care in third-line or beyond PD-L1 positive CPS over 1 gastric or gastroesophageal cancer. "I think pembrolizumab could now be considered the new standard of care in second-line PD-L1-positive CPS score over 10 in adenocarcinoma and all squamous cell esophageal cancer. Pembrolizumab could also be considered as a first-line in PD-L1-positive CPS score over 1 gastric cancer based on the KEYNOTE-062."

The ongoing KEYNOTE-811 trial raises the following open questions, he added:

• Based on the data presented at this meeting, how should we move forward with future design in esophageal cancer with immunotherapy?

• Do we have to take into account race as a condition?

• Do we have to look at Asian versus non-Asian populations?

• Should trials focus more on CPS score over 10 based on the separation of the curve that we saw in KEYNOTE-062?

• Should we look for another biomarker?

• Should we look for tumor mutational burden or TILs as another biomarker to help us determine which patients would respond?

Rahma also suggested that researchers should consider if they have picked the right backbone chemotherapy, and should think about different chemotherapies that may have different immune modulation effects. In addition, sequencing therapy instead of concurrent administration should be investigated, he said.

"Do we know that the immunogenic cells occur at the same time that it should happen when we target with anti-PD-1? Or should we sequence these therapies starting with chemotherapy followed by pembrolizumab or vice versa?"

Clinicians need combination strategies of chemotherapy with immune checkpoint inhibitors, or possibly other novel agents, to treat their gastric cancer patients in the clinic, he concluded.