PD-L1 Status Shapes Choice of Using Nivolumab in Upper GI Cancer


During a Case-Based Roundtable® event, David Zhen, MD, discussed the issue of treating patients with upper gastrointestinal cancer with a PD-L1 composite positive score less than 5 with nivolumab vs chemotherapy alone, in the first article of a 2-part series.

David Zhen headshot

David Zhen, MD

Associate Professor, Clinical Research Division, Fred Hutch

Associate Professor

Division of Hematology and Oncology

University of Washington School of Medicine

Seattle, WA

Targeted Oncology: When treating a patient with gastrointestinal (GI) cancer, how important is mismatch repair testing (MMR), and does this apply to localized disease?

DAVID ZHEN, MD: [With regard to] the importance of doing MMR testing for localized disease, now in the National Comprehensive Cancer Network guidelines [recommend it] based on 2 studies, NEONIPIGA [NCT04006262] and INFINITY [NCT04817826]. Those studies have demonstrated response rate benefit [of immune checkpoint inhibitors], but we don’t have long-term benefit data such as event-free survival and overall survival [OS] yet, but it’s enough response that it is in the guidelines that we should be checking for MMR on all locally advanced gastric cancer.

What data support the addition of nivolumab (Opdivo) to chemotherapy in patients with advanced upper GI cancer?

This was the landmark phase 3 CheckMate 649 study [NCT02872116]. It…[enrolled patients with] newly diagnosed unresectable, advanced, or metastatic upper GI cancers, adenocarcinoma, with no HER2 positive status and good [ECOG] performance status of 0 or 1. These patients were randomly assigned to either chemotherapy alone—and the choices of chemotherapy could have been CapeOx [capecitabine plus oxaliplatin] or FOLFOX [leucovorin calcium (folinic acid), fluorouracil, and oxaliplatin]—and the experimental arm was essentially combining [one of] those 2 chemotherapy regimens with nivolumab. The nivolumab was dosed in such an approach that it would be timing with the chemotherapy as well [every 2 weeks with FOLFOX or every 3 weeks with CapeOx]. Patients were treated until disease progression or unacceptable toxicity [up to 2 years]. The primary end point in this study was dual, looking at both OS and progression-free survival [PFS] in patients with PD-L1 CPS [composite positive score] of 5 or above, and then the secondary end points were looking at the whole randomized patient population as well as in those with CPS of 1 or above.

What were the outcomes in terms of the primary and secondary end points?

When we look at the primary end point, focusing on patients who were enriched with PD-L1 of CPS 5 or above, there was an OS benefit of a [difference[ of 3 months...[median OS of] 14.4 months with nivolumab and chemotherapy vs 11.1 months with chemotherapy alone.1 The HR was 0.70 [95% CI, 0.61-0.81]. The PFS was [a median of 8.3 months with nivolumab vs 6.1 with chemotherapy alone (HR, 0.70; 95% CI, 0.60-0.81)]. There is also a benefit in the all-randomized patient population. The issue is when you look at the all-randomized patient population, it includes everybody with CPS less than 1 and CPS of 1 to 5, so you need that breakdown as well. Even though there was a benefit in the all-randomized patient population, if you do get into the nitty-gritty, the patients who have a PD-L1 CPS of less than 1 tend to get not a lot of benefit. The patients with CPS of 1 to 5 do get a little benefit but probably not as dramatic as the ones who have CPS of 5 or above.

What are your thoughts on using nivolumab in patients with PD-L1 CPS below 5?

This is not based on evidence; this is my personal opinion. I have some issues in terms of PD-L1 expression itself. One is, we’re all basing our decisions on the initial biopsy, and we all know about the heterogeneity in gastroesophageal cancer. Depending on where the biopsy was, that [location] might have been a high PD-L1, but this [location] might be low PD-L1, so we don’t know what the whole tumor is doing. If I treat somebody and re-biopsy some point, I might get a different [CPS]. There are the issues of the assay, the technical aspects of the PD-L1, and how reliable it is. Then based on the data, when we’re looking at the all-randomized population, they might get less benefit.

In my personal practice, I’ve had patients with high PD-L1 [CPS] who don’t get benefit, [and those with] even low PD-L1 who get a benefit, and I don’t know why. So I tend to give everybody chemoimmunotherapy with the exception of if I’m concerned about immunotherapy toxicity, and I need to justify based on PD-L1 to help stratify that. I think it’s just a test of the varying practicing cost and toxicity…. Even if you come to our GI group in the Fred Hutchinson Cancer Center, we’ll debate this all day long.

There is now a finalized paper for CheckMate 649 that has all the breakdowns for these patient [subgroups].2

What is important to know in terms of safety and tolerability of this regimen?

[Based on the] 3-year safety data, when we look at the toxicity rates, they are overall fairly similar, although with the exception that there was some more grade 3/4 toxicities in the nivolumab/chemotherapy arm [60% with nivolumab vs 45% with chemotherapy alone].1 Some of that is probably related to immune related toxicities. Overall rates of discontinuation were a fair number [42% with nivolumab vs 26% with chemotherapy alone], but I think a lot of that might be also due to chemotherapy. When we look at patient-reported outcomes, focusing on the patients with CPS of 5 or greater for nivolumab and chemotherapy together, looking at the change in terms of the quality of life, there were very minor changes.3 Overall, I think the point [the investigators] were trying to make is that there were not significant [differences] in quality of life for patients.


1. Janjigian YY, Shitara K, Moehler MH, et al. Nivolumab (NIVO) plus chemotherapy (chemo) vs chemo as first-line (1L) treatment for advanced gastric cancer/gastroesophageal junction cancer/esophageal adenocarcinoma (GC/GEJC/EAC): 3-year follow-up from CheckMate 649. J Clin Oncol. 2023;41(suppl_16):4025. doi:10.1200/JCO.2023.41.16_suppl.402

2. Janjigian YY, Ajani JA, Moehler M, et al. First-line nivolumab plus chemotherapy for advanced gastric, gastroesophageal junction, and esophageal adenocarcinoma: 3-year follow-up of the phase III CheckMate 649 trial. J Clin Oncol. 2024: JCO2301601. doi:10.1200/JCO.23.01601

3. Moehler M, Xiao H, Blum SI, et al. Health-related quality of life with nivolumab plus chemotherapy versus chemotherapy in patients with advanced gastric/gastroesophageal junction cancer or esophageal adenocarcinoma from CheckMate 649. J Clin Oncol. 2023;41(35):5388-5399. doi:10.1200/JCO.23.00170

Related Videos
Rohit Gosain, MD; Rahul Gosain, MD; and Pamela L. Kunz, MD, presenting slides
Related Content