The PD-1 inhibitor pembrolizumab (Keytruda) elicited encouraging activity with mild adverse events as a treatment for patients with advanced PD-L1â€“positive esophageal carcinoma.
Toshihiko Doi, MD, PhD
The PD-1 inhibitor pembrolizumab (Keytruda) elicited encouraging activity with mild adverse events (AEs) as a treatment for patients with advanced PD-L1—positive esophageal carcinoma, according to findings from the phase Ib KEYNOTE-028 study presented at the 2015 Gastrointestinal Cancers Symposium.
After a median follow-up duration of 7.1 months, the objective response rate (ORR) with pembrolizumab was 30%, which consisted entirely of partial responses. Nine percent of patients had stable disease and more than half of patients treated with pembrolizumab demonstrated some degree of tumor shrinkage.
“KEYNOTE-028 shows promising antitumor activity in a heavily pretreated population. It showed a manageable toxicity profile,” lead investigator Toshihiko Doi, MD, PhD, National Cancer Center Hospital East, Chiba, Japan, said during a presentation of the results. “Further evaluation of pembrolizumab in esophageal cancer is ongoing.”
In the KEYNOTE-028 study, 23 patients with PD-L1—positive esophagus or gastroesophageal junction carcinoma received intravenous pembrolizumab at 10 mg/kg every 2 weeks. The median age of patients, who were primary Asians (52%), was 65 years and most were male (83%). Patients primarily had squamous cell histology (74%) and an ECOG PS of 1 (65%).
Prior to entering the trial, patients had received adjuvant or neoadjuvant therapy (26%) and various lines of therapy for advanced disease. Most patients had received 2 or ≥3 prior therapies (39% and 48%, respectively). The most common prior therapies were platinum-based chemotherapy (100%) and fluoropyrimidine (91%). One patient had received prior treatment with trastuzumab.
PD-L1—positivity was defined as ≥1% expression on tumor or inflammatory cells by immunohistochemistry using the 22C3 antibody. Of those screened, 44.6% were deemed PD-L1–positive. The primary endpoint of the study was ORR, with key secondary outcome measures focused on progression-free survival (PFS), overall survival (OS), duration of response, and safety.
At a data cutoff of November 4, 2015, a statistically significant difference was not observed based upon histology. The response rate for those with squamous cell carcinoma (n = 17) was 29%. In the group with adenocarcinoma (n = 5), the ORR was 40%. Across the study, 56% of patients developed progressive disease. Data for OS and PFS were not yet mature.
A decrease in target lesion burden, defined as any amount of tumor shrinkage from baseline, was experienced by 52.2% of patients treated with pembrolizumab. In most cases, tumor shrinkage persisted on subsequent scans, with up to 70 weeks of follow-up for some patients. The median duration of response was not yet reached (range, 5.5-11.8+ months) and the time to response was 3.7 months (range, 1.8-8.3 months).
“Some patients showed durable responses but in contrast some of the patients had rapid progression within the first cycle, so it is very important to define a biomarker for immune checkpoint treatment,” said Doi.
To help uncover a potential biomarker of response, investigators selected a 6-gene signature that was associated with interferon gamma-related adaptive immune response within the tumor microenvironment. To help inform the findings, a signature score was created based on the normalization of expression values for the 6 genes (IDO1, CXCL10, CXCL9, HLA-DRA, STAT1, IFN-gamma).
Analysis of the signature was conducted following gene expression profiling on formalin-fixed paraffin-embedded tissues samples that were collected at baseline. Overall, those with higher signature scores experienced a more robust response to pembrolizumab and substantial delays in progression. In the non-inflamed, low score group, the ORR was 11% compared with an 43% for those with a higher signature score.
In addition to esophageal cancer, the 6-gene signature demonstrated similar results in samples from patients with head and neck cancer and gastric cancer; however, the test still needs validation and further study. “We have to further examine the responders and the non-responders,” explained Doi.
Treatment-related AEs of any grade occurred in 39% of patients, with 17% experiencing a grade 3 event. There were not any pembrolizumab-related deaths or treatment discontinuations. The most commonly observed grade 1/2 AEs were decreased appetite (9%), hypothyroidism (9%), rash (9%), and adrenal insufficiency (4%). The most frequently seen grade 3 AEs were lymphocytopenia (9%), pruritic rash (4%), liver disorder (4%), and decreased appetite (4%).
“Immune-related adverse events—hypothyroidism, adrenal insufficiency, and rash—these adverse events were recovered from after treatment with proper management, such as corticosteroids and interruption of dosing,” said Doi.
Pembrolizumab is currently FDA approved for patients with melanoma and PD-L1—positive advanced non–small cell lung cancer. In November 2015, the FDA granted a breakthrough therapy designation to pembrolizumab as a potential therapy for patients with microsatellite instability-high metastatic colorectal cancer. In addition to these indications, the humanized antibody continues to be explored across a number of settings.
The phase III KEYNOTE-181 study is currently recruiting patients with advanced esophageal/esophageal junction carcinoma to compare pembrolizumab with investigator's choice of standard therapy. The primary endpoints of the study are PFS and OS. The study, which started in December 2015, is estimated to enroll 600 patients (NCT02564263).
Doi T, Piha-Paul SA, Jalal SI, et al. Updated results for the advanced esophageal carcinoma cohort of the phase Ib KEYNOTE-028 study of pembrolizumab (MK-3475). J Clin Oncol. 2016;34 (suppl 4S; abstr 7).