New IHC Biomarker Test Predicts Which Gastric Cancer Patients Benefit from Immunotherapy

Researchers from Peking University discovered a tumor-specific biomarker that can predict which gastric cancer patients will benefit from immunotherapy before surgery. In the study, tsMHC-II-positive tumors showed higher pathological complete response rates, 36.84% versus 11.11%, and major pathological response rates, 63.16% versus 25.93%. This simple technique is available in most pathology laboratories and could improve patient selection for neoadjuvant immunotherapy.

For patients with locally advanced gastric cancer, the promise of immunotherapy has always come with a frustrating caveat: some patients benefit substantially, while others do not. The challenge has been finding a biomarker that is not only biologically meaningful, but also practical enough to use in routine care. A new study from researchers at Zhejiang Cancer Hospital and Peking University suggests that tumor-specific MHC-II, or tsMHC-II, may do both.

Published in Science Bulletin, the work combines single-cell RNA sequencing with prospective clinical validation and points to tsMHC-II expression as a potentially useful way to identify which patients are most likely to respond to neoadjuvant immunotherapy before surgery.

"This biomarker could transform how we select patients for neoadjuvant immunotherapy," Professor. Xiangdong Cheng said, one of the study's corresponding authors. “By identifying patients most likely to benefit, we can improve outcomes while sparing non-responders from unnecessary treatments and side effects.”

Immune checkpoint inhibitors targeting PD-1 and PD-L1 have already changed the treatment landscape in advanced gastric cancer, and recent studies have raised hopes that they may also improve outcomes in the neoadjuvant setting. But the field has lacked a robust and clinically workable predictive marker. PD-L1 testing is widely used, yet its interpretation can be technically difficult and inconsistent.

In this phase 2 prospective study, investigators enrolled 46 patients with locally advanced gastric cancer who received neoadjuvant therapy followed by D2 gastrectomy. Ten patients, or 21.7%, achieved pathological complete response, and 19, or 41.3%, achieved major pathological response.

To understand why some tumors were sensitive and others were not, the team performed single-cell RNA sequencing on pretreatment tumor samples from 44 patients. Their analysis highlighted one signal in particular: a co-expression module enriched for MHC-II genes in tumor cells. This tumor-specific MHC-II expression was strongly associated with a better response to treatment.

The difference was clinically meaningful. Among patients with tsMHC-II-positive tumors, pathological complete response occurred in 36.84%, compared with 11.11% in tsMHC-II-negative disease. Major pathological response was also substantially higher, 63.16% versus 25.93%.

The researchers then moved beyond retrospective association and tested the marker prospectively. Of 66 patients screened, 32 were found to have tsMHC-II-positive tumors. After exclusions, 30 patients entered the validation cohort. Their outcomes were striking: 11 patients, or 36.67%, achieved pathological complete response, and 20, or 66.67%, achieved major pathological response after preoperative treatment and surgery. 

Notably, these response rates remained high regardless of PD-L1 combined positive score. In patients with CPS 5 or higher, pathological complete response and major pathological response rates were 40% and 70%, respectively. In those with CPS below 5, the corresponding rates were 35% and 65%.

That finding is important because it suggests tsMHC-II may capture clinically relevant biology that is not fully reflected by PD-L1 scoring alone. Mechanistically, the study found that interferon-gamma signaling appears to drive MHC-II upregulation in tumor cells, potentially enhancing antigen presentation and immune recognition.

The practical appeal is equally significant. Unlike more complex molecular assays, tsMHC-II can be assessed by standard immunohistochemistry using commercially available antibodies in routine pathology laboratories. That makes it far easier to implement than many emerging biomarkers. The authors also note a key advantage over PD-L1 testing, whose reproducibility in real-world practice has been questioned because of low agreement among pathologists.

The study also offers a broader view of the tumor microenvironment. The researchers found that higher levels of SPP1-positive macrophages may be associated with resistance to therapy, hinting that both tumor-intrinsic antigen presentation and the surrounding immune milieu shape response.

The findings are encouraging, but the validation cohort was small, and larger multicenter studies are needed before tsMHC-II can enter routine clinical use. Even so, the study combines mechanistic insight, prospective validation, and practical feasibility, positioning tsMHC-II as a potentially useful tool for selecting gastric cancer patients most likely to benefit from immunotherapy.