Neoadjuvant PD-1 blockade plus chemotherapy induces a high pathological complete response rate and anti-tumor immune subsets in clinical stage III gastric cancer

Abstract

First-line PD-1 blockade plus chemotherapy significantly improves the survival benefits in late-stage gastric cancer (GC) patients. However, the pathological response rate and effects on the immune microenvironment of neoadjuvant PD-1 blockade plus chemotherapy in patients with cTNM-stage III GC remain to be elucidated. Patients with cTNM-stage III GC who underwent neoadjuvant PD-1 blockade plus chemotherapy and surgery were enrolled. Four in vivo models bearing GC were jointly established to investigate the specific roles of chemotherapy and PD-1 blockade for GC treatment. The tumor immune microenvironment was analyzed by hematoxylin and eosin (H&E) and IHC staining, multicolor flow cytometry and immunofluorescence. A total of 75 patients with cTNM-stage III (cT2-4N1-3M0) gastric cancer who received neoadjuvant PD-1 blockade plus chemotherapy (SOX/XELOX) were included in this study. After treatment, 21 (28.0%) and 57 (76.0%) patients achieved pathological complete response (pCR) and post-therapy pathological downstaging. Subgroup analyses revealed that patients with CPS >1 (32.6% vs 8.3%) and dMMR (35.7% vs 25.4%) subtype had better efficacy. Additionally, the resected specimens showed more anti-tumor immune infiltration indicating a response to neoadjuvant PD-1 blockade plus chemotherapy. Multicolor immunofluorescence and in vivo experiments on mouse models revealed that elevated M1/M2 ratio of macrophages, CD8 + T cells and plasma cells indicated effective response to treatment. Furthermore, neoadjuvant PD-1 blockade plus chemotherapy neither delayed surgery nor increased postoperative complication rate. The analyses indicate neoadjuvant PD-1 blockade plus chemotherapy is a promising therapeutic strategy in patients with cTNM-stage III GC with an encouraging pCR rate.

Keywords: cTNM-stage III gastric cancer; immune cell subpopulations; neoadjuvant PD-1 blockade plus chemotherapy; pathological complete response; tumor microenvironment.

Figure 1.

Pathological evaluation to response to neoadjuvant treatment in clinical TNM-stage III gastric cancer (GC) patients. (a) The information of tumor regression grades and several subgroups, including MMR, EBV, CPS and HER2. (b) The ratio of tumor necrosis patients following neoadjuvant PD-1 blockade plus chemotherapy. (c) The tumor regression grades in dMMR, pMMR and EBV-positive patients. (d) The tumor regression grades in CPS ≤ 1 and CPS > 1 patients. (e) The tumor regression grades in HER2-negative and HER2-positive patients.

Figure 2.

Histological features of cancer foci following different neoadjuvant PD-1 blockade plus chemotherapy. (a) Hematoxylin and eosin (H&E) results observed substantial tumor-infiltering immune cells in the tumor nests of responders to neoadjuvant PD-1 blockade plus chemotherapy, while rare in the non-responder group. (b) The histopathological features in resected specimens: presence of abundant infiltrating immune cells (tumor infiltrating lymphocytes, plasma cells, eosinophils and neutrophils), tertiary lymphoid structure (TLS), proliferative fibrosis and neovascularization in the regression bed. (c) Immunohistochemistry (IHC) results showed that responders to neoadjuvant treatment had higher levels of CD8⁺ T cells and CD68⁺ macrophages, compared to non-responders.

Figure 3.

Tumor microenvironment features of immune cells in responders and non-responders to PD-1 blockade. (a) Volcano plots of differentially expressed genes between non-responders and responders to PD-1 blockade. (b) Immune suppressive checkpoints between non-responders and responders to PD-1 blockade. (c) The top 20 enriched KEGG pathways of the differentially expressed genes. (d) Differentially infiltrated immune cells, identified by the CIBERSORT method, between non-responders and responders to PD-1 blockade. (e) Multicolor immunofluorescence in one representative responder and one non-responder treated with neoadjuvant PD-1 blockade plus chemotherapy. (f) The proportions of immune cells identified by multicolor immunofluorescence in 4 responders and 5 non-responders to neoadjuvant PD-1 blockade plus chemotherapy.

Figure 4.

PD-1 blockade plus chemotherapy inhibits tumor growth by inducing more anti-tumor immune subsets in mouse models. (a) Treatment protocol for PD-1 inhibitor and chemotherapy in 615-mice. At the end, tumors were collected and analyzed by multi-color flow cytometry. (b) The tumor sizes of 615 mice treated with chemotherapy (5-fluorouracil and oxaliplatin) and/or PD-1 inhibitor before flow cytometry. (c-d) The changes of tumor volumes of 615 mice treated with the four strategies in a longer period of observation. Staining of CD4⁺ T cells (e), CD8⁺ T cells (E), effector CD8⁺CD44⁺CD62⁻ T cells (f) and intracellular cytokine staining of TNF-α⁺CD8⁺ T cells (g) among CD45⁺ T cells.

Figure 5.

PD-1 blockade and chemotherapy regulated macrophages to inhibit tumor progression in mouse models. (a-b) Chemotherapy (5-fluorouracil and oxaliplatin) and PD-1 inhibitor decreased macrophages and exerted synergic effects in combinational group. (c-e) Combining PD-1 blockade and chemotherapy could increase M1-type macrophages and decreased M2-type macrophages to elevate the ratio of M1/M2. (f) The CT images at pre- and post-PD-1blockade plus chemotherapy of one case with clinically evaluated partial response. (g) The proportions of CD8 + T cells and macrophages identified by multicolor immunofluorescence in the above case. CBT: PD-1 blockade plus chemotherapy.