As the first phase 3 trial to investigate an Fc-silent, anti-TIGIT agent in patients with solid tumors, the STAR-221 study (NCT05568095) may further clarify the therapeutic role of domvanalimab plus zimberelimab (Sepalizumab) and chemotherapy in patients with gastric cancers, potentially reshaping first-line treatment paradigms, particularly for PD-L1–positive disease, according to Kohei Shitara, MD.
STAR-221 is investigating the combination of the TIGIT monoclonal antibody domvanalimab, the PD-1 inhibitor zimberelimab, and FOLFOX (fluorouracil, leucovorin, and oxaliplatin) or CAPOX (capecitabine and oxaliplatin) vs nivolumab (Opdivo) plus FOLFOX or CAPOX in the first-line setting for patients with locally advanced unresectable or metastatic gastric, gastroesophageal junction (GEJ), and esophageal adenocarcinoma.1 Overall survival (OS) serves as the primary end point. Secondary end points include progression-free survival (PFS), overall response rate (ORR), duration of response (DOR), safety, and quality of life outcomes.
The initiation of STAR-221 is based on positive findings from the phase 2 EDGE-Gastric trial (NCT05329766).2 Updated findings from arm A1 of the study showed that in all evaluable patients with first-line, locally advanced unresectable or metastatic gastric/GEJ/esophageal adenocarcinoma (n = 41), at a median follow-up of 13.9 months, the confirmed ORR with domvanalimab plus zimberelimab and FOLFOX was 59% (95% CI, 42%-74%). Among patients with a PD-L1 tumor activity positivity (TAP) score of at least 5% (n = 16), this rate was 69% (95% CI, 41%-89%). Among those with a TAP score of less than 5%, the ORR was 50% (95% CI, 29%-71%).
“Maybe the benefit [will be limited to] patients in the PD-L1–high population, but we should see the actual data [from STAR-221],” Shitara said in an interview with OncLive®. “If we have positive findings, this may give us important insight about the structure of the antibody and whether Fc-silent or Fc-activated [agents] give us different stories [in the context of] past negative trials.”
In the interview, Shitara discussed the mechanism of action and clinical activity of the TIGIT-targeted agent domvanalimab; preclinical and phase 2 data that support the continued investigation of domvanalimab plus zimberelimab and chemotherapy in patients with gastric/GEJ/esophageal cancers; and the potential clinical implications of results from STAR-221, which may be influenced by PD-L1 status.
Shitara is the director of the Department of Gastrointestinal Oncology at the National Cancer Center Hospital East in Kashiwa, Japan.
OncLive: How do domvanalimab and zimberelimab work alongside chemotherapy to elicit antitumor responses?
Shitara: The important target molecule is TIGIT, especially for this combination. TIGIT is an inhibitory checkpoint receptor expressed on various types of immune cells, especially exhausted CD8 cells and regulatory T cells [Tregs], where it usually competes with activating molecules, such as CD20 and CD26, for binding its ligands. If there is TIGIT expression, this usually leads to inhibition of a specific immunophenotype like T cells, Tregs, and natural killer cells. Importantly, [TIGIT] is usually observed with other immune checkpoint molecules, such as PD-1 and LAG-3 [inhibitors]. It is expressed at a relatively later stage of exhaustion. That’s why blocking TIGIT has emerged as an attractive strategy in cancer immunotherapy. This is usually combined with anti–PD-1 and –PD-L1 agents.
An important aspect [of some] TIGIT antibodies is [the] Fc-maintained or Fc-activated [state, which] retains Fc receptor binding capabilities, leading to antibody-dependent cellular cytotoxicity or complement-dependent cytotoxicity to deplete TIGIT-expressing immune cells. For example, an Fc-maintained IgG1 antibody may deplete Tregs. This may be good for antitumor response, but it may also increase autoimmune toxicity because of the depletion of Tregs.
Another class of agents is the Fc-silent TIGIT antibody, which is engineered to lack Fc receptor engagement. It doesn’t deplete Tregs—instead, it [mainly] enhances the expansion of effector cells. On the other hand, if Fc is maintained, there is also a risk of depletion of activated effector cells. There are always pros and cons for this kind of structure.
In preclinical studies, there’s a good paper for both classification of Fc-silent and Fc-maintained antibodies. This was published in Cancer Research and supports that Fc-silent IgG1 antibodies for TIGIT may be better compared with Fc-maintained [agents]. However, another group published a different paper in Nature that supports that Fc-activated TIGIT antibodies are better, especially regarding depletion of immunosuppressive cells, like Tregs.
Based on these preclinical studies, TIGIT inhibition is attractive. However, there are 2 types of anti-TIGIT antibodies. Domvanalimab is mainly an Fc-silent IgG1 antibody. This may not deplete inhibitory cells, but [it may] mainly activate exhausted effector cells. Because of its lack of depletion of Tregs, the toxicity profile is expected to be feasible. This is usually combined with the anti–PD-1 [agent] zimberelimab. Zimberelimab is an IgG4 monoclonal antibody [with a mechanism of action] similar to that of nivolumab and pembrolizumab [Keytruda]. This combination has been tested in various clinical trials.
What phase 2 data contributed to the rationale for STAR-221?
This combination’s development is further supported by results from the ongoing EDGE-Gastric trial. This is a first-line gastric cancer trial [planning] to enroll approximately 40 patients to test domvanalimab and zimberelimab in combination with a cytotoxic FOLFOX regimen. Initial data were presented at the 2024 ASCO Annual Meeting and showed an ORR of 59% in the entire population, with a median PFS of 12.9 months. Although the small sample size is a relatively major limitation of the study, [the findings], in direct comparison with [those from] previous pivotal trials like the phase 3 CheckMate 649 [NCT02872116] and KEYNOTE-859 [NCT03675737] studies, suggested an increased ORR and longer median PFS.
The subgroup analysis [data] also support [population] enrichment by PD-L1 status. This TIGIT-directed domvanalimab/zimberelimab/chemotherapy combination seemed to work better in patients with high PD-L1 expression, such as a combined positive score [CPS] or TAP score of 5% or higher. This is in line with our expectations, because TIGIT is usually co-expressed with other checkpoints like PD-1 [T] effector cells. Additionally, the safety profile looks manageable; because of a lack of Treg depletion, autoimmune reactions seem to not be increased.
The phase 3 STAR-221 trial is ongoing to test domvanalimab/zimberelimab in combination with first-line FOLFOX or CAPOX vs the current standard of chemotherapy plus nivolumab in the first-line setting. We needed to test PD-L1 status before patient enrollment. Patients could be enrolled regardless of PD-L1 status, but stratification by PD-L1 status was important, considering the mode of action [of the combination demonstrated in] the subgroup analysis of the phase 2 study. That’s why PD-L1 TAP score was included as one of the stratification factors for the primary end point for OS in this study. Subgroups will also be tested according to TAP score or CPS status. Enrollment [to STAR-221] is completed, and we are waiting for survival follow-up and analysis.
Concerning the current other trials for TIGIT-targeted therapy, unfortunately, no phase 3 trial has shown the benefit of TIGIT inhibitors in any solid tumors. For example, there were negative trials in melanoma, gynecological cancer, and lung cancer. The lung cancer trial showed a borderline benefit of TIGIT inhibition, especially in the PD-L1–positive population.
Although all trials [with TIGIT inhibition] have been negative, there [are opportunities to use TIGIT inhibitors with] different modes of action, especially [regarding] the Fc portion. Previously reported phase 3 trials have almost always applied Fc-maintained or Fc-activated TIGIT inhibitors. STAR-221 is the first study to test an Fc-silent anti-TIGIT antibody. [Therefore], we may observe a difference [in outcomes from this trial] compared with previous negative trials.
How might positive results from STAR-221 affect the development of domvanalimab and the placement of the investigational combination among current chemoimmunotherapy regimens for gastric cancers?
It should depend on the result. If [domvanalimab/zimberelimab/chemotherapy] shows an OS benefit compared with chemotherapy plus nivolumab, it should be at least the standard in some populations. However, this depends on the subgroup analysis [where patients are stratified] by PD-L1 status. If it only shows a benefit in the PD-L1–high population, it should be the treatment used for that population [only].
However, if [the combination] shows an [OS] benefit regardless of PD-L1 status, this is also attractive, because usually patients with low PD-L1 status achieve limited benefit with chemotherapy plus a PD-[L]1 inhibitor. If this trial turns out to be positive, why not test this agent in other types of tumors [beyond those] where previous anti-TIGIT therapy failed to show a survival benefit? It should be interesting.
Chemotherapy/nivolumab is standard, especially for patients with a PD-L1 CPS of 1 or higher. [In patients with a] CPS less than 1, [this combination] didn’t show benefit, but there’s no detrimental effect. The [STAR-221] trial was started before the FDA’s Oncologic Drugs Advisory Committee meeting about the restriction of [frontline PD-1 inhibitors to patients with HER2-negative, microsatellite-stable gastric/GEJ adenocarcinoma with a PD-L1 CPS greater than 1]. I don’t expect any large barriers or any differences in this [trial] population compared with the general gastric cancer population regarding CPS frequency.
Additionally, the main objective in this study is to see the treatment effect, especially in patients with a CPS or a TAP score of 5% or higher. [CPS] is being monitored during the trial. It does not seem different from the usual frequency [seen in this population], but the exact results should be reported after the analysis.
References
- A clinical trial of a new combination treatment, domvanalimab and zimberelimab, plus chemotherapy, for people with an upper gastrointestinal tract cancer that cannot be removed with surgery that has spread to other parts of the body (STAR-221). ClinicalTrials.gov. Updated June 26, 2025. Accessed June 30, 2025. https://www.clinicaltrials.gov/study/NCT05568095
- Janjigian Y, Oh D, Pelster M, et al. Updates on abstract 433248: EDGE-Gastric arm a1: phase 2 study of domvanalimab, zimberelimab, and FOLFOX in first-line (1l) advanced gastroesophageal cancer. Presented at: 2024 ASCO Annual Meeting. May 31-June 4, 2024. Chicago, IL.