Dung T. Le, MD
The PD-1 inhibitor nivolumab (Opdivo) had an overall response rate (ORR) of 14% with an acceptable safety profile in patients with gastric or gastroesophageal junction (GEJ) cancer, according to results from the phase I/II CheckMate-032 trial.
“Nivolumab monotherapy was well tolerated in patients with metastatic gastric, esophageal, or gastroesophageal junction cancer,” lead author Dung T. Le, MD, an assistant professor of Oncology at Johns Hopkins Kimmel Cancer Center, said when presenting the results at the 2016 Gastrointestinal (GI) Cancers Symposium.
CheckMate-032 is a randomized, open-label, multitumor cohort study that accrued patients with lung, breast, bladder, pancreatic, and ovarian cancer, and treated them with nivolumab as a single-agent or in combination with ipilimumab.
Overall, 163 patients with gastric cancer were enrolled in the study. Fifty-nine patients with locoregionally advanced or metastatic gastric cancer or GEJ were assigned to single-agent nivolumab at 3 mg/kg IV every 2 weeks. The remaining 104 patients were assigned to 1 of 3 nivolumab/ipilimumab combination regimens. The data presented at the GI Symposium were the preliminary results for patients who received nivolumab monotherapy.
The median patient age was 60 years (range, 29-80), including 17 (29%) patients aged ≥ 65 years. Males comprised 76% (n = 45) of the cohort and 95% (n = 56) of patients were white. Fifty-three percent of patients (n = 31) had GEJ, 15% (n = 9) had esophageal cancer, and 31% (n = 18) had gastric cancer. The disease site was not known for 1 patient.
Overall, 83% of patients had received at least 2 prior therapies. Seventy-one percent (n = 42) of patients had received 2 to 3 prior regimens, 17% of patients (n = 10) had 1 prior regimen, and 12% had been treated with >3 prior regimens.
The primary outcome measure for the study was ORR. Secondary endpoints included progression-free survival, overall survival (OS), duration of response, and safety.
The average number of nivolumab doses received was 7.3 (median, 5). The maximum number of doses received was 31 and 58% (n = 34) of patients received at least 4 doses. The average cumulative dose was 21.8 mg/kg.
Four patients (7%) remain on treatment, with 93% (n = 55) having discontinued. The reasons for discontinuing nivolumab included progression (n = 46), treatment-related adverse events (AEs; n = 3), unrelated AEs (n = 3), withdrawn consent (n = 1), noncompliance (n = 1), and other (n = 1). The median follow-up was 4.6 months.
The median ORR was 14% (n = 8; 95% CI, 6%-25%), including 1 (2%) complete response (CR) and 7 (12%) partial responses (PR). The rates of stable disease (SD) and progressive disease were 19% (n = 11) and 58% (n = 34), respectively. Responses were not determined for 6 (10%) patients. The disease control rate (CR + PR + SD) was 32% (n = 19).
The median time to response was 1.6 months, with a median response duration of 7.1 months.
“There was one patient who progressed by RECIST criteria, but on subsequent imaging had a nice regression over time. And, in fact, this patient currently has no evidence of disease on imaging,” said Le.
The median OS was 5.0 months (95% CI, 3.4-12.4). The 3-, 6-, and 12-month OS rates were 70%, 49%, and 36%, respectively. Le noted that, “Forty-nine percent and 36% of chemorefractory patients were still alive at 6 and 12 months, respectively.”
Although PD-L1 expression was not an inclusion criteria for the study, PD-L1 levels were able to be tested on 40 patients using archived tumor samples or baseline biopsies, according to Le.
Patients with PD-L1 expression >1% had an ORR of 27% compared with 12% in patients with PD-L1 expression <1%. When using a 5% threshold, the rates were 33% versus 15% in patients with PD-L1 expression >5% versus <5%, respectively.
“PD-L1 expression appears to be numerically associated with a higher objective response rate,” said Le.
Le noted that treatment-related AEs were consistent to those observed in previous trials of single-agent nivolumab. Overall, 69% (n = 41) of patients experienced a treatment-related AE of any grade. The most common all-grade AEs were fatigue (32%), pruritus (17%), decreased appetite (15%), diarrhea (15%), nausea (14%), increased AST (12%), pyrexia (10%), vomiting (10%).
Grade 3/4 AEs occurred in 17% (n = 10) of patients. The most frequently occurring grade 3/4 AEs included increased AST at 5% of patients, and fatigue, vomiting, and diarrhea at 2% each.
Ten percent (n = 6) of patients experienced a treatment-related serious AE (SAE) of any grade. Grade 3/4 SAEs occurred in 5% of patients, including increased ALT, increased AST, pneumonitis, and vomiting, occurring in 1 patient each. There were no treatment-related deaths.
Although not approved by the FDA for gastric cancer, nivolumab has approved indications for the treatment of lung cancer, melanoma, and renal cell carcinoma.