O-007
ZEBRA: an ACCRU/IRCI multicenter phase 2 study of pembrolizumab in patients with advanced small bowel adenocarcinoma (SBA)
Introduction
SBA is a rare cancer that is frequently diagnosed in later stages. No standard of care exists for metastatic SBA beyond first-line FOLFOX/CAPOX. Later-line SBA therapy has been extrapolated from studies of colon and gastric cancers, but SBA has more frequent microsatellite instability (MSI), greater PD-L1 expression than CRC, and more T-lymphocyte tumor infiltration compared to those cancers. PD-1 inhibitors remain untested in unselected SBA patient; thus, we hypothesize that pembrolizumab will induce antitumor response.
SBA is a rare cancer that is frequently diagnosed in later stages. No standard of care exists for metastatic SBA beyond first-line FOLFOX/CAPOX. Later-line SBA therapy has been extrapolated from studies of colon and gastric cancers, but SBA has more frequent microsatellite instability (MSI), greater PD-L1 expression than CRC, and more T-lymphocyte tumor infiltration compared to those cancers. PD-1 inhibitors remain untested in unselected SBA patient; thus, we hypothesize that pembrolizumab will induce antitumor response.
Methods
Previously treated advanced SBA patients received pembrolizumab 200 mg IV q3 weeks until disease progression (PD) or toxicity. Primary endpoint was confirmed overall response rate (ORR), defined as meeting RECIST 1.1 criteria on two consecutive evaluations >4 weeks apart. Pembrolizumab was deemed ineffective if ORR is 30% (Ha), with 0.06 significance when ORR is 10%, and 94% power to detect ORR 30% in 35 patients. Seven or more responses may warrant further studies. Secondary endpoints were progression-free survival (PFS), overall survival (OS), and adverse events (AEs). MSI or MMR status was performed clinically for this analysis.
Previously treated advanced SBA patients received pembrolizumab 200 mg IV q3 weeks until disease progression (PD) or toxicity. Primary endpoint was confirmed overall response rate (ORR), defined as meeting RECIST 1.1 criteria on two consecutive evaluations >4 weeks apart. Pembrolizumab was deemed ineffective if ORR is 30% (Ha), with 0.06 significance when ORR is 10%, and 94% power to detect ORR 30% in 35 patients. Seven or more responses may warrant further studies. Secondary endpoints were progression-free survival (PFS), overall survival (OS), and adverse events (AEs). MSI or MMR status was performed clinically for this analysis.
Results
40 patients were enrolled and treated from 5/2017-9/2018; 60% were male with median age 63 yrs (29-85). Primary tumors were distributed in the duodenum (60%), jejunum (25%), and ileum (15%). MSI status was available for 22 patients (55%). 32 patients are off study due to: PD 25 (78%), death 5 (16%), and AE 2 (6%). Of these 32 patients, median treatment duration was 4 cycles (range: 1-23) with 3 confirmed partial responses (8%; 95% CI: 2-20%). This did not meet our pre-defined criteria for success. Median OS 6.9 mo (95% CI: 5.1, NR) and median PFS 2.8 mo (95% CI: 2.7-5.1) were similar among the three primary tumor sites. Of 18 MSS patients, there was 1 confirmed PR (6%) and disease control rate 50% (9 patients with partial response or stable disease). Of 2 MSI-L patients, the best response was 1 patient with stable disease. Both MSI-H pts achieved PR and remain alive without progression. 23 patients (58%) had grade >3 AEs, 9 pts (23%) had grade 4/5 AEs. Of five grade 5 AEs, all were deemed most likely secondary to disease progression, except one grade 5 sepsis judged possibly related to treatment.
40 patients were enrolled and treated from 5/2017-9/2018; 60% were male with median age 63 yrs (29-85). Primary tumors were distributed in the duodenum (60%), jejunum (25%), and ileum (15%). MSI status was available for 22 patients (55%). 32 patients are off study due to: PD 25 (78%), death 5 (16%), and AE 2 (6%). Of these 32 patients, median treatment duration was 4 cycles (range: 1-23) with 3 confirmed partial responses (8%; 95% CI: 2-20%). This did not meet our pre-defined criteria for success. Median OS 6.9 mo (95% CI: 5.1, NR) and median PFS 2.8 mo (95% CI: 2.7-5.1) were similar among the three primary tumor sites. Of 18 MSS patients, there was 1 confirmed PR (6%) and disease control rate 50% (9 patients with partial response or stable disease). Of 2 MSI-L patients, the best response was 1 patient with stable disease. Both MSI-H pts achieved PR and remain alive without progression. 23 patients (58%) had grade >3 AEs, 9 pts (23%) had grade 4/5 AEs. Of five grade 5 AEs, all were deemed most likely secondary to disease progression, except one grade 5 sepsis judged possibly related to treatment.
Conclusion
In the largest study of SBA to date, and the first using immunotherapy, Pembrolizumab did not achieve the goal ORR in pretreated advanced SBA pts; however, it may control disease in some MSS patients (1 PR, DCR 50%), though further study is needed to confirm this outcome. Efficacy in MSI-H is confirmed. Further predictive marker testing is underway.
In the largest study of SBA to date, and the first using immunotherapy, Pembrolizumab did not achieve the goal ORR in pretreated advanced SBA pts; however, it may control disease in some MSS patients (1 PR, DCR 50%), though further study is needed to confirm this outcome. Efficacy in MSI-H is confirmed. Further predictive marker testing is underway.
Publisher
Oxford University Press
Source Journal
Annals of Oncology
