New Standard of Care in PD-L1+ Esophageal Cancer?

SAN FRANCISCO — Immunotherapy with pembrolizumab (Keytruda, Merck) was superior to standard chemotherapy in improving overall survival for patients with advanced esophageal cancer who had high levels of programmed cell death ligand 1 (PD-L1).

In patients whose PD-L1 combined positive score (CPS) was ≥10, regardless of histology, treatment with pembrolizumab yielded a 12-month overall survival of 43%, vs 20% for those treated with either paclitaxel, docetaxel, or irinotecan.

"These data suggest that pembrolizumab should be considered a new standard of care in patients with a PD-L1 CPS of 10 or greater in the second-line setting," said lead study author Takashi Kojima, MD, professor in the Department of Gastroenterology and Gastrointestinal Oncology at the National Cancer Center Hospital East in Kashiwa, Japan.

"The overall response rate was also higher with pembrolizumab than with chemotherapy," Kojima added, "and the safety profile was better with pembrolizumab vs chemotherapy."

The findings were presented here at the Gastrointestinal Cancers Symposium (GICS) 2019.

"Patients with advanced esophageal cancer after first-line chemotherapy have a poor prognosis and limited treatment options," explained Kojima. "Taxanes and irinotecan have been used after first-line therapy, but no overall survival benefit has been seen in a phase 3 study with chemotherapy."

Pembrolizumab showed activity in metastatic esophageal cancer in an earlier phase 2 KEYNOTE 180 study. "Durable responses were observed in patients with squamous cell carcinoma (SCC), adenocarcinoma, and with the PD-L1 combined positive score of 10 or greater," said Kojima.

Survival Benefit in PD-L1+ Patients

The current phase 3 KEYNOTE-181 study compared pembrolizumab with the investigator's choice of chemotherapy as second-line therapy for patients with advanced/metastatic SCC and adenocarcinoma of the esophagus or Siewert type I adenocarcinoma of the esophagogastric junction that had progressed following standard frontline therapy.

The cohort included 628 patients who were randomly assigned to receive either pembrolizumab 200 mg every 3 weeks or chemotherapy. Patients were stratified by histology (SCC vs adenocarcinoma) and region (Asia vs rest of world).

The primary endpoint was overall survival, which was evaluated in the entire intent-to-treat (ITT) population, in the subgroup of patients with PD-L1 CPS ≥10 and in patients with SCC.

At a median follow-up of 7.1 months for pembrolizumab and 6.9 months for chemotherapy, the overall survival rate for patients with a PD-L1 CPS ≥10 (n = 222) who received pembrolizumab was 9.3 months, vs 6.7 months in the chemotherapy arm (hazard ratio [HR], 0.69; P = .0074).

Although it was directionally favorable, the difference in overall survival was not statistically significant in the ITT group (7.1 months vs 7.1 months; HR, 0.89; P = .0560).

Kojima noted that in the SCC subset, there was "clinically meaningful" improvement in overall survival for patients who received pembrolizumab vs chemotherapy, but it did not reach statistical significance because of prespecified boundaries (n = 401; 8.2 months vs 7.1 months; HR, 0.78; P = .0095).

Progression-free survival (PFS) was also improved by pembrolizumab in those patients with PD-L1 CPS ≥10, at 21%, vs 7% with chemotherapy. In the SCC group, PFS was 15% vs 9%; and in the ITT group, PFS was 12% vs 10%.

Overall response was improved with pembrolizumab in all three groups: 21.5% vs 6.5%, 16.7 vs 7.4%, and 13.1% vs 7.1%.

At the data cutoff, nine patients in the pembrolizumab arm were still receiving therapy, and five had completed 2 years of treatment. None of the patients in the chemotherapy arm were still receiving treatment.

Pembrolizumab was associated with a lower rate of adverse events of any grade (64% vs 86%) or of grade 3-5 (18% vs 41%), compared with chemotherapy. "No new safety signals were observed with pembrolizumab," said Kojima.

Reasonable to Consider Pembro

In a discussion of the paper, Harry Yoon, MD, co-chair of the Esophageal/Gastric Cancer Disease Group at the Mayo Clinic Cancer Center, Rochester, Minnesota, noted that studies such as this one "help us move incrementally towards enhanced molecular selection for combination therapy in this disease."

He pointed out that among the SCC patients, pembrolizumab missed the prespecified P value by .0020, and the overall survival curves separate. "One can imagine that if there had not been three primary endpoints but only two, this may have been a positive result," speculated Yoon. "So how can we implement these data into our clinical practice, if at all?"

Some may advocate use of pembrolizumab off protocol, especially for patients who cannot tolerate chemotherapy, because it is better tolerated. "I think this can be a discussion point for guideline committees," he said. He added that it is "important to note that KEYNOTE 181 was not designed as a noninferiority study, and not all groups seem to benefit equally."

One limitation was that the patients were not stratified by PD-L1 status, only by histology and region. "The lack of stratification by PD-L1 status potentially increases the chance of confounding if favorable patient characteristics were disproportionately assigned to the pembro arm," he said.

A multivariate analysis may help clarify whether positive results in patients with PD-L1 CPS ≥10 are explained by a higher frequency of favorable prognostic variables in the patients in the pembro arm. "The strength of these results could influence guideline recommendations and implementation in clinical practice," Yoon explained.

Overall, he summarized, the results are "internally valid with a caveat."

In translating these findings to the clinic, Yoon explained that in his practice, he currently orders a PD-L1 immunohistochemistry assay (IHC) and an HER2 IHC at the first metastatic diagnosis of gastroesophageal adenocarcinomas. He also thinks that it is "reasonable to consider a practice change.

"This would mean ordering a PD-L1 IHC in patients with squamous cell carcinoma at first metastatic diagnosis," he said. "It would also mean that pathology labs would need to report a more detailed PD-L1 CPS score, if they don't already."

As for treatment, for patients with a PD-L1 CPS ≥10, it is "reasonable to consider pembro," Yoon said. "This should be discussed in guideline committees, and these results will be submitted to regulatory authorities worldwide."

The study was funded by Merck Sharp & Dohme Corp. Dr Kojima has received honoraria from Oncolys BioPharma and institutional payments for patents, royalties, or other intellectual property from Astellas Amgen BioPharama, MSD, Oncolys BioPharma, Ono Pharmaceutical, and Shionogi. Several coauthors have disclosed relationships with industry. Dr Yoon has relationships with Merck Sharp & Dohme, Astellas, LSK Biopharma, Lilly/ImClone, Roche/Genentech (inst), Boston Biomedical (inst), and Merck (inst).

Gastrointestinal Cancers Symposium (GICS) 2019: Abstract 2. Presented January 17, 2019.

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.