Real-World Data Indicate Favorable Efficacy With Nivolumab in Advanced Gastric Cancer

Yu Sunakawa, MD, PhD, further discusses the early real-world data from the DELIVER trial with nivolumab in patients with gastric/GEJ cancer and provided insight into what future analyses will examine.

Nivolumab (Opdivo) has demonstrated favorable efficacy with an acceptable safety profile according to early real-world data from the prospective, multicenter DELIVER trial (UMIN000030850), said lead study author Yu Sunakawa, MD, PhD, who added that the clinical outcomes reported thus far compare favorably with those observed in the phase 3 ATTRACTION-2 trial (NCT02267343).

“These real-world data demonstrated comparable clinical outcomes to [those reported in] the previous phase 3 ATTRACTION-2 trial with nivolumab in advanced gastric cancer,” said Sunakawa. “DELIVER revealed [similar] tumor behavior, with some patients experiencing rapid tumor growth [following] the nivolumab treatment within clinical practice. Biomarkers for hyper-progressive disease should [soon be] established.”

Results from the multicenter, observational/translational DELIVER trial showed a disease control rate (DCR) of 39.2% (95% CI, 34.9-43.7) in 487 evaluable patients. In 282 patients with measurable lesions, the response rate with nivolumab was 6.7% (95% CI, 4.1-10.3) with a slightly lower DCR of 36.5%. Results from a subgroup analysis done based on patient background showed DCR rates of 41%, 42%, and 24%, in patients with a performance status of 0, 1, and 2, respectively. Notably, the DCR was found to be lower in patients with ascites versus those without, at 28.6% versus 47.0%, respectively (P = .005).

Moreover, when the tumor growth rate (TGR) was calculated as a percentage increase in tumor volume during 1 month of treatment, the rate was found to decrease following the initiation of nivolumab in 56.6% of patients (n = 124/219). However, 20.5% of patients were reported to have hyper-progressive disease. Additional preliminary findings from the trial showed a median progression-free survival (PFS) of 1.8 months, as well as a median overall survival (OS) of 5.9 months in evaluable patients.

“The DCR [with nivolumab] was approximately 40% in the previous ATTRACTION-2 trial, [which was] comparable [to the DCR reported] in the DELIVER trial,” said Sunakawa. “In DELIVER, the preliminary survival data showed that the median PFS was 1.8 months and the median OS was 5.9 months; that survival time [is] comparable to what had been reported in the ATTRACTION-2 trial, although the data are still immature.”

In an interview with OncLive, Sunakawa, an associate professor in the Department of Clinical Oncology at the St. Marianna University School of Medicine, further discussed the early real-world data from the DELIVER trial with nivolumab in patients with gastric/GEJ cancer and provided insight into what future analyses will examine.

OncLive: Could you provide some background on the utility of nivolumab in gastric cancer? What was the rationale for the DELIVER trial?

Sunakawa: Nivolumab has demonstrated survival benefit and a manageable safety profile in previously treated [patients with] advanced gastric cancer or gastroesophageal junction (GEJ) cancer in the phase 3 ATTRACTION-2 trial. The objective response rate was approximately 11% and the disease control rate was 40%. Recently, the long-term survival benefit of nivolumab was most evident in patients with a complete response or a partial response. However, there [were few] real-world data with nivolumab in clinical practice; specifically, [there were] no prospective data for patients with severe peritoneal dissemination, or ascites. Also, it has been [observed] that some tumors grow rapidly after nivolumab treatment, but the proportion is uncertain. Therefore, we performed an observational, translational study to evaluate the safety and efficacy of nivolumab for advanced gastric cancer in the real-world setting.

Interestingly, it has been shown that gut microbiome is a potential predictive factor for efficacy of immunotherapy. Gut microbiota may impact the response to gastric cancer therapy, as well as toxicity. As such, we are trying to discover novel host-related immune biomarkers, including the gut microbiome in this study.

What was the design of this study and what patients were included in the analysis?

The DELIVER trial is a multicenter, observational, translational study that enrolled 500 patients with gastric/GEJ cancer who received nivolumab monotherapy in any lines of treatment. The primary end point of the trial was overall survival and the secondary end points included response rate, disease control rate, tumor growth rate, PFS, and safety.

For a biomarker study, blood samples have been prospectively collected at 2 timepoints before and after nivolumab treatment for measuring host-related factors, including gut microbiome and metabolome.

Could you discuss the findings presented during the ESMO World GI Congress?

We presented the clinical outcomes of nivolumab treatment in real-world setting at this year’s meeting, including tumor response, tumor growth rate, hyper-progressive disease rate, as well as survival time and safety; it’s preliminary data. The efficacy results showed that the response rate was 6.7%, the disease control rate was 39%. The subgroup analysis showed that the disease control rate was approximately 40% for those with a performance status of 0 and performance status of 1, but 24% for those with a performance status of 2. The disease control rate was lower in patients with ascites compared with patients without [the condition], at 28.6% and 47.0%, respectively. Interestingly, we observed hyper-progressive disease in approximately 20% of patients who received nivolumab treatment in this study.

Regarding safety, were those findings in line with what has historically been seen with nivolumab?

Safety data were collected within for first 200-patient cohort, meaning the data are still immature. However, we observed adverse events (AEs) such as diarrhea, fatigue, decreased appetite, and nausea, more frequently compared with what had been seen in the ATTRACTION-2 trial. However, no unexpected immune-related AEs were reported.

Is there anything else that you would like to add?

Common AEs [with nivolumab] were observed more frequently compared with [what had been reported in] the ATTRACTION-2 trial. This may be reflective that in clinical practice, [we’re seeing more] patients with poor performance or conditions. We should know that the DCR is lower in patients with ascites compared with patients without ascites in clinical practice. Tumor growth rate decreased after introduction of nivolumab treatment in 56% of patients in this study. However, approximately 20% of patients may experience hyper-progressive disease.

Are any next steps planned for this research?

Yes. Next time, we will present survival time, the primary end point [of the trial], along with [data from a] subgroup analysis [being done] by patient background. We will also analyze the association between hyper-progressive disease and survival time in this study. For a biomarker study, [we hope to learn more about] microbiome markers and present that information at a future meeting.


  1. Sunakawa Y, Sakamoto Y, Inoue E, et al. Updated analysis of DELIVER trial (JACCRO GC-08): a large observational/translational study of nivolumab treatment in advanced gastric cancer. Presented at: 2020 ESMO World Congress on Gastrointestinal Cancer Virtual Meeting; July 1-4, 2020; Virtual.