Neoadjuvant radioimmunotherapy in pancreatic cancer enhances effector T cell infiltration and shortens their distances to tumor cells

Junke Wang; Jessica Gai; Tengyi Zhang; Nan Niu; Hanfei Qi; Dwayne L Thomas, II; Keyu Li; Tao Xia; Christina Rodriguez; Rose Parkinson; Jennifer Durham; Thomas McPhaul; Amol K Narang; Robert A Anders; Arsen Osipov; Hao Wang; Jin He; Daniel A Laheru; Joseph M Herman; Valerie Lee; Elizabeth M Jaffee; Elizabeth D Thompson; Qingfeng Zhu; Lei Zheng

doi:10.1126/sciadv.adk1827

. 2024 Feb 7;10(6):eadk1827. doi: 10.1126/sciadv.adk1827

Neoadjuvant radioimmunotherapy in pancreatic cancer enhances effector T cell infiltration and shortens their distances to tumor cells

Junke Wang ^1,^2,^3,^4,^5,^†, Jessica Gai ^1,^2,^3,^4,^6,^†, Tengyi Zhang ^1,^2,^3,⁴, Nan Niu ^1,^2,^3,⁴, Hanfei Qi ^1,^2,^7,⁸, Dwayne L Thomas II ^1,^2,^3,^4,⁹, Keyu Li ^1,^2,^3,⁴, Tao Xia ^1,^2,^3,⁴, Christina Rodriguez ^1,^2,^4,⁶, Rose Parkinson ^1,^2,^4,^6,⁸, Jennifer Durham ^1,^2,^4,^6,⁸, Thomas McPhaul ^1,^2,^3,⁹, Amol K Narang ^1,^2,^3,^4,¹⁰, Robert A Anders ^2,^3,^6,¹¹, Arsen Osipov ^1,^2,^3,^4,^6,¹², Hao Wang ^1,^2,^3,^6,^7,⁸, Jin He ^1,^2,^3,^6,⁹, Daniel A Laheru ^1,^2,^3,^4,^6,⁸, Joseph M Herman ^1,^2,^6,^10,¹³, Valerie Lee ^1,², Elizabeth M Jaffee ^1,^2,^3,^4,^6,⁸, Elizabeth D Thompson ^1,^2,^3,^4,^11,^*, Qingfeng Zhu ^1,^2,^3,^4,^8,^*, Lei Zheng ^1,^2,^3,^4,^6,^8,^9,^*

^¹Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^²The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^³The Pancreatic Cancer Precision Medicine Center of Excellence Program, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^⁴The Bloomberg Kimmel Institute for Cancer Immunotherapy, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^⁵Division of Biliary Tract Surgery, Department of General Surgery, West China Hospital, Sichuan University, Chengdu, Sichuan 610041, China.

^⁶The Skip Viragh Pancreatic Cancer Center, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^⁷Quantitative Sciences Division, Department of Oncology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^⁸The Cancer Convergence Institute at Johns Hopkins, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^⁹Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^¹⁰Department of Radiation Oncology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^¹¹Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, MD 21287, USA.

^¹²Cedars Sinai Medical Center, Los Angeles, CA, 90048, USA.

^¹³Northwell Health System, New Hyde Park, NY, 11042, USA.

Corresponding author. Email: lzheng6@jhmi.edu (L.Z.); qzhu6@jhmi.edu (Q.Z.); ethomp36@jhmi.edu (E.D.T.)

^†

These authors contributed equally to this work.

Roles

Junke Wang: Data curation, Formal analysis, Investigation, Resources, Validation, Visualization, Writing - original draft, Writing - review & editing

Jessica Gai: Formal analysis, Investigation, Visualization, Writing - review & editing

Tengyi Zhang: Investigation

Nan Niu: Investigation, Methodology

Hanfei Qi: Formal analysis

Dwayne L Thomas II: Investigation, Resources, Writing - review & editing

Keyu Li: Methodology

Tao Xia: Investigation

Christina Rodriguez: Data curation, Investigation, Project administration, Writing - review & editing

Rose Parkinson: Data curation, Project administration, Validation, Writing - review & editing

Jennifer Durham: Investigation, Project administration, Validation

Thomas McPhaul: Formal analysis, Investigation, Software

Amol K Narang: Conceptualization, Methodology

Robert A Anders: Conceptualization, Investigation, Visualization, Writing - review & editing

Arsen Osipov: Data curation, Investigation, Methodology, Resources, Validation, Visualization, Writing - review & editing

Hao Wang: Data curation, Formal analysis, Project administration, Writing - review & editing

Jin He: Conceptualization, Investigation, Resources

Daniel A Laheru: Conceptualization, Methodology, Project administration, Supervision

Joseph M Herman: Conceptualization, Supervision, Writing - original draft, Writing - review & editing

Valerie Lee: Conceptualization, Investigation, Methodology

Elizabeth M Jaffee: Conceptualization, Funding acquisition, Methodology, Resources, Supervision, Validation, Writing - review & editing

Elizabeth D Thompson: Investigation, Methodology, Resources, Writing - review & editing

Qingfeng Zhu: Formal analysis, Methodology, Software, Visualization

Lei Zheng: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Visualization, Writing - original draft, Writing - review & editing

PMCID: PMC10849596 PMID: 38324679

Abstract

Radiotherapy is hypothesized to have an immune-modulating effect on the tumor microenvironment (TME) of pancreatic ductal adenocarcinoma (PDAC) to sensitize it to anti–PD-1 antibody (a–PD-1) treatment. We collected paired pre- and posttreatment specimens from a clinical trial evaluating combination treatment with GVAX vaccine, a–PD-1, and stereotactic body radiation (SBRT) following chemotherapy for locally advanced PDACs (LAPC). With resected PDACs following different neoadjuvant therapies as comparisons, effector cells in PDACs were found to skew toward a more exhausted status in LAPCs following chemotherapy. The combination of GVAX/a–PD-1/SBRT drives TME to favor antitumor immune response including increased densities of GZMB⁺CD8⁺ T cells, T_H1, and T_H17, which are associated with longer survival, however increases immunosuppressive M2-like tumor-associated macrophages (TAMs). Adding SBRT to GVAX/a–PD-1 shortens the distances from PD-1⁺CD8⁺ T cells to tumor cells and to PD-L1⁺ myeloid cells, which portends prolonged survival. These findings have guided the design of next radioimmunotherapy studies by targeting M2-like TAM in PDACs.

Radiotherapy brings immunotherapy-induced effector T cells closer to tumor cells, which is associated with prolonged survival.

INTRODUCTION

Pancreatic ductal adenocarcinoma (PDAC) has a dismal prognosis, with an overall 5-year survival rate of only 12% in the USA (1). Approximately 30% of newly diagnosed PDAC presents as locally advanced pancreatic cancer (LAPC) (2), which is considered surgically unresectable due to local involvement of adjacent vessels. A common treatment strategy is induction chemotherapy followed by chemoradiation or stereotactic body radiation (SBRT) (3). Historically, chemotherapy followed by SBRT resulted in downstaging and successful R0 or R1 resection in 20% of LAPCs (4). We reported that SBRT following induction chemotherapy resulted in higher rates of complete pathologic responses (5) and attenuated lymphocyte suppression (6). However, additional interventions are needed to increase this rate because surgical resection is often required for durable local tumor control.

Immunotherapy is a promising treatment alternative or adjunct to standard of care for some aggressive cancers. Unfortunately, PDACs respond poorly to immunotherapy (7). While tumor-infiltrating effector T cells (T_effs) portend a survival benefit in immunotherapy-responsive cancers, their presence is rare in the PDAC tumor microenvironment (TME) (8). Immunotherapy has focused on manipulating tumor antigen-specific T_effs either by potentiating them through vaccine therapy or T cell therapy or by unleashing their function through immune checkpoint inhibitor (ICI) therapies such as anti–programmed cell death protein 1 (PD-1)/programmed cell death ligand 1 (PD-L1) antibodies. However, the major immune “defect” in PDAC is its immunosuppressive TME, which impedes T cell infiltration with immunotherapy (9).

Radiotherapy (RT) has been reported as one conventional treatment modality for local or metastatic PDAC, although the clinical benefit has not been clearly established (10–12). Emerging evidence indicates that RT can prime the TME with T_effs by causing immunogenic cell death and then activating innate responses, which serves as an “in situ vaccination” (13, 14). RT has the potential to improve the efficacy of immunotherapies such as ICIs, which have achieved clinical success for many cancers, but has not yet been successful for PDACs. Preclinical and clinical studies suggest that the combination of RT and ICIs exhibits a synergistic effect and enhanced antitumor activity in varied tumors such as non–small cell lung cancer and melanoma (15). In the KPC and Panc02 murine PDAC models, RT combined with PD-L1 blockade significantly improved tumor response and prevented development of liver metastases (16). Although we found that this combination offered local tumor control, it failed to induce T_eff infiltration into the tumors (17). Moreover, in both subcutaneous and orthotopic mouse PDAC models, SBRT combined with dual cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) and PD-1 blockades led to a modest growth delay in tumors, which was associated with significant increases in CD8⁺ and CD4⁺ T cell infiltration (18). These findings provide a rationale for the testing of RT combined with ICIs in patients with PDAC. In a phase 1 clinical trial, SBRT combined with durvalumab or durvalumab plus tremelimumab demonstrated an acceptable safety profile and a modest treatment benefit in patients with metastatic PDAC (19). In a single-arm phase 2 clinical trial, RT combined with ipilimumab and nivolumab led to a short duration of disease control and clinical benefit in 25 patients with metastatic, microsatellite stable PDAC (20). In another randomized phase 2 study, SBRT combined with nivolumab plus ipilimumab, compared to SBRT combined with nivolumab, demonstrated a statistically significant improvement in progression-free survival and radiographic responses in both SBRT-targeted and non-targeted lesions in patients with chemotherapy-refractory, metastatic PDAC (21). Nevertheless, the optimal immune-based mechanistic role of RT combined with immunotherapy in treating PDAC remained unknown.

Our prior study showed that a granulocyte-macrophage colony-stimulating factor-secreting allogeneic PDAC cell vaccine (GVAX) as a neoadjuvant therapy can induce the formation of tertiary lymphoid aggregates (LAs) and the up-regulation of PD-L1 in the LAs, suggesting that GVAX may also prime PDAC to respond to ICIs (22). Immunohistochemical analysis showed these LAs, with organized T cell and B cell zones with evidence of lymphoid neogenesis, to be regulatory structures of adaptive immunity (22). More recently, our platform trial (23, 24) testing either GVAX or GVAX in combination with anti–PD-1 antibody (a–PD-1) as neoadjuvant therapy for resectable PDACs demonstrated that the combination of GVAX and a–PD-1 induced the infiltration of CD8⁺ T and CD4⁺ T cells into tumors but reduced the amount of PD-1⁺CD8⁺ and PD-1⁺CD4⁺ T cells. Therefore, we and others have shown that it is possible to turn the immunologic desert in PDACs into immune-responsive tumors; however, the response rates remain low in clinical trials investigating these regimens (25). To test the synergy between vaccine therapy and RT in sensitizing PDACs for a–PD-1 therapy, we have completed a single-center, single-arm, phase 2 clinical trial (NCT02648282) of combination GVAX, a–PD-1 pembrolizumab, and SBRT for patients with LAPC (26). Few studies have comprehensively investigated the immune-modulating roles of RT in combination with immunotherapy in human PDACs. Here, by performing multiplex immunohistochemistry (IHC) on the prospectively collected specimens from this phase 2 clinical trial, we conducted a comprehensive assessment of the changes of TME in PDAC following the combination treatment regimens of GVAX, a–PD-1, and SBRT.

RESULTS

Combination treatment with pembrolizumab, vaccine therapy, and SBRT induces CD4⁺ and CD8⁺ T lymphocyte infiltration into posttreatment PDAC tumors

To investigate changes in tumor-infiltrating immune cells in PDACs after neoadjuvant immunotherapy with a GVAX vaccine therapy, an a–PD-1 pembrolizumab, and SBRT, we used a sequential staining and stripping multiplex IHC technique (27). De-identified specimens were collected from 54 patients with evaluable LAPC who were enrolled in a clinical trial (NCT02648282) (26) to receive GVAX, pembrolizumab, and SBRT following induction chemotherapy with either the FOLFIRINOX or the gemcitabine/abraxane combination chemotherapy regimens from July 2016 to January 2021 (fig. S1A). A pre-GVAX/a–PD-1/SBRT treatment core biopsy specimen was obtained from every participant. We obtained archived surgically resected tumor samples from surgical candidates post-GVAX/a–PD-1/SBRT treatment. For patients with unresectable tumors, a posttreatment core biopsy was performed if feasible. Among 54 evaluable patients’ specimens (table S1), those from patients who have been followed on the clinical trial for at least 2 years between the start of the GVAX/a–PD-1/SBRT treatment and 1 June 2022, including those who had died within 2 years, were included in this study.

Slides were sectioned from formalin-fixed paraffin-embedded (FFPE) blocks and stained by multiplex IHC with a panel of myeloid and lymphoid cell markers (tables S2 and S3). The regions of interest (ROIs) were selected within the biopsy tumor slides to cover a minimum of three different tumor cores from the same biopsy (fig. S1B). A minimum of three ROIs in the tumor areas that each contained at least one vaccine-induced intratumoral tertiary LA were also selected from each posttreatment surgically resected tumor and divided into LAs and non-LA tumor areas (fig. S1C). For posttreatment core biopsy specimens, only tumor areas were selected. Pseudo-colors were assigned to each marker. ROIs were quantified as described previously (27). As the signals from each staining were captured separately, they would not interfere with each other. A combination of positive and negative markers was used to identify each immune cell subtype (Fig. 1A and table S4); thus, nonspecific staining of individual markers is unlikely to cause the misidentification of immune cell subtypes (27). The density of each immune cell subtype was calculated as the percentage of total cells identified by nuclear staining within each ROIs. In comparison to resected PDACs from patients treated by the standard of care chemotherapy and SBRT (designated SBRT cohort, table S5), those who had upfront surgery without neoadjuvant therapy (designated No-NAT cohort, table S6), and the published cohort from a neoadjuvant platform clinical trial (NCT02451982) where patients received one cycle of GVAX and a–PD-1 nivolumab 2 weeks before the surgical resection (designated GVAX/a–PD-1 cohort) (23), the immune cell composition in the resected PDACs from patients who received GVAX/a–PD-1/SBRT appears to have several key differences (Fig. 1B). Therefore, in the below studies, we used specimens from these cohorts as comparisons (dataset S1).

Fig. 1. — (A) Overlaid images of representative immune cell markers assigned with pseudo-colors in the resected PDACs following GVAX/a–PD-1/SBRT treatment. Scale bars, 50 μm. (B) Immune cell composition in the resected PDAC TME following different types of neoadjuvant therapies (n = 21 for “No-NAT,” n = 17 for “SBRT,” n = 10 for “GVAX/a–PD-1,” and n = 24 for “GVAX/a–PD-1/SBRT”). (C) Summary of the density of all immune cell subtypes examined as indicated in the pretreatment (n = 54) versus posttreatment (n = 45) tumor areas in PDACs with GVAX/a–PD-1/SBRT treatment. (D and E) Summary of the density of all immune cell subtypes analyzed as indicated in the pretreatment [(D); n = 24 for “resected” and n = 30 for “non-resected”] and posttreatment [(E); n = 24 for resected and n = 21 for non-resected] tumor areas from the resected versus non-resected PDACs with GVAX/a–PD-1/SBRT treatment. Data were shown as the means ± SEM or percentage and compared by unpaired t test; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; all others, not significant.

To assess the effect of the GVAX/a–PD-1/SBRT treatment on patient’s PDAC immune-microenvironment, we examined the immune cell population in patient’s pretreatment biopsy samples and the non-LA tumor areas in posttreatment tumor specimens from all available patients (Fig. 1C). We found that the combination neoadjuvant therapy induced a variety of immune cell infiltration, including a significant increase in the total CD4⁺ and CD8⁺ T cell infiltration (Fig. 1C). We also found a significant increase in M1-like tumor-associated macrophage (TAM) infiltration in posttreatment tumors (Fig. 1C). There was no significant difference in the levels of B cells, M2-like TAM, and tumor-associated neutrophil (TAN) infiltration between pre- and posttreatment tumor samples (Fig. 1C and fig. S2). These results suggested that the TME has changed toward favoring an antitumor immune response following GVAX/a–PD-1/SBRT treatment.

Next, we compared immune cell populations between patients who underwent PDAC resection and those with non-resectable PDACs following GVAX/a–PD-1/SBRT treatment (Fig. 1, D and E). It should be noted that LAs were not consistently observed in posttreatment biopsy specimens due to small tissue areas. Thus, tumor areas were compared between posttreatment resected PDAC tumors and non-resected PDAC biopsy specimens. With the exception of TANs, immune cell–subtype compositions between these two patient groups were comparable before receiving GVAX/a–PD-1/SBRT treatment (Fig. 1D). In posttreatment tumor areas, there were significantly increased infiltrations of immunosuppressive immune cell subtypes in non-resected PDACs in comparison to resected PDACs, including T helper 2 (T_H2), regulatory T cells (T_regs), and TANs, suggesting that maintaining a less-immunosuppressive TME as part of an immune response to combination GVAX/a–PD-1/SBRT treatment may contribute to surgical resectability following neoadjuvant treatment (Fig. 1E). Nevertheless, a higher density of TANs after induction chemotherapy and before GVAX/a–PD-1/SBRT may contribute to an inability to surgical resection following neoadjuvant therapy.

An increase in high-quality CD8⁺ T cell population occurs with combination GVAX + a–PD-1 + SBRT treatment and correlates with better survival in patients with resectable PDAC

To understand the response and resistance mechanisms elicited by our combination therapy in patients with PDAC, we examined potential correlations between each subtype of T cells, myeloid cells, and overall survival (OS). OS > 2 years was considered a “favorable” survival for patients with PDAC following surgical resection based on previously published correlative studies (22, 27). We did not observe a correlation between OS and the infiltration of CD8⁺ T cells and the granzyme B⁺ (GZMB⁺) CD8⁺ T cell subtype in tumor areas before neoadjuvant immunotherapy in patients with either resectable or non-resectable PDAC (Fig. 2A). A positive correlation between the infiltration of GZMB⁺CD8⁺ T cells, but not that of general CD8⁺ T cells, in tumor areas and OS was found in patients with resectable PDAC after treatment (Fig. 2B). However, this correlation was not observed in patients with non-resectable PDACs (Fig. 2B). Furthermore, we observed a significant increase in the infiltration of CD8⁺ T cells and GZMB⁺CD8⁺ T cells in tumor areas from pretreatment tumors to posttreatment tumors in resected PDACs associated with OS > 2 years, but not in unresected PDACs (Fig. 2C). A significant increase in CD8⁺ T cells was also observed in unresected PDACs associated with OS < 2 years (Fig. 2C), suggesting that an increased infiltration of general CD8⁺ T cells is not prognostic. Again, a significantly increased GZMB⁺CD8⁺ T cells, but not CD8⁺ T cells in general, in posttreatment LAs was associated with OS > 2 years in resected PDACs (Fig. 2D), suggesting that there is a specific association of T_eff cytotoxic function with survival. Overall, it appears that increased CD8⁺ T cells in the posttreatment PDACs may simply be a response to the GVAX/a–PD-1/SBRT treatment. However, increased GZMB⁺CD8⁺ T cells in the posttreatment PDACs may not only signify an immune response to this treatment but also represent improved clinical outcomes.

Fig. 2. — (A and B) Correlative analysis of OS with the densities of CD8⁺ and GZMB⁺CD8⁺ T cells, as indicated, in pretreatment [(A); n = 9 for “resected OS < 2y,” n = 15 for “resected OS > 2y,” n = 20 for “non-resected OS < 2y,” and n = 7 for “non-resected OS > 2y”] and posttreatment [(B); n = 9 for resected OS < 2y, n = 15 for resected OS > 2y, n = 11 for non-resected OS < 2y, and n = 7 for non-resected OS > 2y] tumor areas. (C) Correlative analysis of OS with the changes of the densities of CD8⁺ and GZMB⁺CD8⁺ T cells, as indicated, between matched pre- and posttreatment tumor areas (n = 9 for resected OS < 2y, n = 15 for resected OS > 2y, n = 11 for non-resected OS < 2y, and n = 7 for non-resected OS > 2y). (D) Correlative analysis of OS with the densities of CD8⁺ and GZMB⁺CD8⁺ T cells, as indicated, in the posttreatment LAs in the resected PDACs (n = 9 for “OS < 2y” and n = 15 for “OS > 2y”). 2y, 2 years. Data were shown as the means ± SD; comparisons by unpaired t test for (A), (B), and (D) and paired t test for (C); *P < 0.05 and **P < 0.01; NS, not significant.

An increase in T_H1 and T_H17 density correlates with improved OS in patients treated with the combination GVAX/a–PD-1/SBRT

We next evaluated the infiltration of CD4⁺ T cell subtypes in the pre- and posttreatment tissues and their correlation with OS (Fig. 3, A to F, and fig. S3, A to R). By comparing paired pre- and posttreatment tumor areas, we found that the GVAX/a–PD-1/SBRT treatment induced a significant increase in the general CD4⁺ T cells, the T_H1 and T_H17 subtypes, and the T_H1/T_H2 ratio in resected tumors with OS > 2 years, but not those with OS < 2 years (Fig. 3, A, B, D, and F). In contrast, GVAX/a–PD-1/SBRT treatment did not induce a significant increase in the general CD4⁺ T cells, T_H17, and the T_H1/T_H2 ratio in unresected tumors with OS > 2 years (Fig. 3, A, D, and F). However, the densities of essentially all CD4⁺ T cell subtypes in pretreatment tumors did not correlate with OS (fig. S3, A to F), suggesting that treatment-induced changes in CD4⁺ T cell subtypes may have changed the clinical outcome. In contrast, there were associations between higher T_H2 density and OS < 2 years (fig. S3I), between higher T_H17 density and OS > 2 years (fig. S3L) in posttreatment non-LA tumor areas in resected tumors, and between higher T_H1 density and OS > 2 years in posttreatment LAs in resected tumors (fig. S3N) and in posttreatment biopsies of unresected tumors (fig. S3H), further supporting the same notion that the GVAX/a–PD-1/SBRT treatment may have changed the clinical outcomes. These results are consistent with our previous findings showing that T_H17 signaling was enhanced in PDACs and associated with improved survival following GVAX/a–PD-1 treatment and that an enhanced T_H1 infiltration and T_H1/T_H2 ratio were also associated with improved OS (23). In this current study, we also found a significant increase in the infiltration of posttreatment T_regs, which are known to play an immunosuppressive role according to previously published studies (28, 29), in unresected PDACs associated with OS > 2 years (Fig. 3E).

The EOMES-mediated T cell exhaustion pathway predicts clinical outcome of non-resectable PDACs following GVAX/a–PD-1/SBRT treatment

We did not observe any significant difference in exhausted T cell infiltration PD-1⁺CD4⁺, PD-1⁺CD8⁺, and EOMES⁺CD8⁺ T cells between pre- and posttreatment tumor areas, in tumor areas or LAs between resected tumors and unresected tumors, or in tumor areas or LAs between tumors associated with OS > 2 years and OS < 2 years (fig. S4, A to L). The only exception is in the EOMES⁺CD8⁺ T cell infiltration that exhibited a statistically significant elevation in posttreatment tumor areas compared to that in pretreatment tumor areas in matched unresected PDACs associated with OS < 2 years (fig. S4K). We further examined different subgroups of exhausted CD8⁺ T cells and found no obvious difference in the composition of PD-1⁺EOMES⁺, PD-1⁻EOMES⁺, PD-1⁺EOMES⁻, and PD-1⁻EOMES⁻ CD8⁺ T cells in pretreatment biopsy tumor areas between resected and unresected PDACs (fig. S4M). We found a modest, but statistically significant, increase of the infiltration of PD-1⁺EOMES⁺ CD8⁺ T cells (2.20 versus 1.21% of total CD8⁺ T cells) and a modest but statistically significant decrease in infiltration of PD-1⁻EOMES⁺ CD8⁺ T cells (6.15 versus 9.11% of total CD8⁺ T cells) in unresected PDACs compared to resected PDACs (fig. S4N). These results suggested that the EOMES-mediated T cell exhaustion pathway likely affects the clinical outcome of non-resectable PDACs. Nevertheless, this pathway does not seem to be a major T cell exhaustion mechanism that accounts for the sensitivity and resistance to the GVAX/a–PD-1/SBRT treatment.

Changes in M2-like TAMs following GVAX/a–PD-1/SBRT treatment correlates with patient outcomes

We also examined changes in myeloid cell populations following neoadjuvant therapy (Fig. 3, G to J, and fig. S5, A to L), because they are known suppressive regulators of T cell responses in the PDAC TME (30). The densities of myeloid cell subpopulations including TAMs and TANs in pretreatment tumor areas were mostly not associated with OS in either resectable or non-resectable patients following GVAX/a–PD-1/SBRT treatment (fig. S5, A to D). An increase in M1-like TAMs in posttreatment tumor areas compared to that in pretreatment tumor areas was noted in both resected PDACs associated with OS < 2 years and those with OS > 2 years (Fig. 3G), suggesting that this increase is a response to the treatment that does not affect clinical outcomes. However, an increase in the M1/M2-like TAM ratio, likely due to an increase of M1-like TAMs, was significantly associated with OS > 2 years in resected PDACs, whereas a decrease in the M1/M2-like TAM ratio, likely due to an increase of M2-like TAMs, was significantly associated with OS < 2 years in unresected PDACs (Fig. 3I). In addition, an increase in the M2-like TAMs was significantly associated with OS < 2 years in unresected PDACs (Fig. 3H). Consistently, M2-like TAMs in posttreatment tumor areas in unresectable PDACs is the only myeloid cell subpopulation that negatively correlated with OS in the posttreatment non-LA tumor areas or LAs (fig. S5F). These results suggest that a treatment-induced increase in the M1/M2-like TAM ratio or M2-like TAM density is not only an immune response to this treatment but also potentially changes clinical outcomes. In contrast, we did not observe any significant association between TANs and OS in posttreatment tumor areas or LAs (fig. S5, H and L).

Similarly, the densities of PD-L1⁺ myeloid cell subpopulations including PD-L1⁺ TAMs and PD-L1⁺ TANs in pretreatment tumors or posttreatment tumors did not correlate with OS in either resectable or non-resectable patients (fig. S5, M to U). When we compared PD-L1⁺ TAMs and PD-L1⁺ TANs between pre- and posttreatment tumor areas from the same patients (fig. S5, V to X), we found that a decrease in PD-L1⁺ M2-like TAMs was statistically significant in resected PDACs associated with OS > 2 years (fig. S5W), suggesting that the PD-L1⁺ subtype may account for the impact of the treatment-induced increase in M2-like TAMs on clinical outcomes. Furthermore, there was a statistically significant decrease in PD-L1⁺ TANs in unresected PDACs associated with OS < 2 years (fig. S5X), suggesting that the role of TANs following GVAX/a–PD-1/SBRT treatment remains elusive and that TAN may be inferior to the PD-L1⁺ M2-like TAM as the additional target in PDAC following induction chemotherapy and the GVAX/a–PD-1/SBRT treatment in contrast to its role in resectable PDAC described previously (23).

Untreated PDACs and chemotherapy/RT-treated PDACs are characterized by distinct TMEs

Our previous studies have demonstrated that GVAX/a–PD-1 neoadjuvant therapy induces CD8⁺ T cell infiltration into the PDAC TME, but not the high-quality T_eff subpopulation such as GZMB⁺CD8⁺ T cells (23). To assess the specific immune-modulating effect of chemotherapy and SBRT when given with the GVAX/a–PD-1 regimen, we compared the immune cell population between PDACs resected following chemotherapy and the GVAX/a–PD-1/SBRT treatment from the current study and those from the prior study following GVAX/a–PD-1 treatment (23). We first compared the pretreatment biopsy specimens before the GVAX/a–PD-1/SBRT to those before the GVAX/a–PD-1 treatment (Fig. 4A). Of note, the pretreatment biopsy specimens in the clinical trial of LAPCs were collected after induction chemotherapy, thus demonstrating the effect of chemotherapy on the TME, whereas the pretreatment biopsy specimens in the prior clinical trial of resectable PDAC were collected before starting any treatment. We observed a significantly higher density of infiltrated lymphoid cells in the pre-radioimmunotherapy biopsies from patients following induction chemotherapy, including CD8⁺ T cells and its exhausted subtype EOMES⁺CD8⁺ T cells, CD4⁺ T cells and the T_H1 subtype, PD-1⁺ T cells, and DC-SIGN⁺ immature dendritic cells (DCs), as well as a significantly lower density of natural killer (NK) cells and TANs, comparing to those who had not started any treatment (Fig. 4A). There were no significant differences between the levels of GZMB⁺CD8⁺ T cells, T_H2 cells, T_H17 cells, T_regs, B cells, mast cells, TAMs, and DC-LAMP⁺ mature DCs (Fig. 4A). These results suggest that chemotherapy or a more advanced stage of PDAC skews the effector cells in the TME toward a more exhausted status.

Fig. 4. — (A) Summary of the density of all immune cell subtypes analyzed as indicated in the pretreatment tumor areas in PDACs with GVAX/a–PD-1 treatment (n = 10) versus GVAX/a–PD-1/SBRT treatment (n = 24). (B and C) Summary of the density of all immune cell subtypes analyzed as indicated in the posttreatment non-LA tumor areas (B) and LAs (C) in PDACs with GVAX/a–PD-1 treatment (n = 10) versus GVAX/a–PD-1/SBRT treatment (n = 24). Data were shown as the means ± SEM and compared by unpaired t test; *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001; all others, not significant.

SBRT enhances the GVAX/a–PD-1–mediated increase of GZMB⁺CD8⁺ T_effs

In contrast to the pre-immunotherapy specimens, the post-immunotherapy resected tumor specimens exhibited nearly no significant differences in immune cell subtypes between the non-LA tumor areas following either the GVAX/a–PD-1/SBRT treatment or the GVAX/a–PD-1 treatment, including CD8⁺ T cells and its exhausted subtype EOMES⁺CD8⁺ T cells, PD-1⁺ T cells, or NK cells (Fig. 4B). The only exception was that tumors collected after GVAX/a–PD-1/SBRT treatment had a significantly lower T_H2 infiltration than the tumors collected after GVAX/a–PD-1 treatment (Fig. 4B).

We also compared the immune cells compositions within the GVAX-induced LAs between these two cohorts (Fig. 4C). We observed a significantly higher density of GZMB⁺CD8⁺ T cells and a significantly lower density of T_regs in the posttreatment LAs in tumors collected after the GVAX/a–PD-1/SBRT treatment in comparison to those collected after GVAX/a–PD-1 treatment (Fig. 4C). This result suggests that either chemotherapy, more likely SBRT, or both enhance the ability of the GVAX/a–PD-1 treatment to increase high-quality effector CD8⁺ T cells and reduce T_regs in the setting of anti–PD-1 treatment that is known to increase T_regs. It is likely based on these results that SBRT mediates the immune modulating activity of the GVAX/a–PD-1 therapy specifically in LAs as previously shown (22, 23).

Although we did not observe an increase in GZMB⁺CD8⁺ T cells in tumor areas following GVAX/a–PD-1/SBRT treatment compared to GVAX/a–PD-1 treatment, we suspected that there may be a more significant treatment-induced change in GZMB⁺CD8⁺ T cells in the tumors following GVAX/a–PD-1/SBRT treatment. We therefore calculated the fold changes in CD8⁺ T cells and GZMB⁺CD8⁺ T cells from pretreatment biopsy tumors to posttreatment resected tumors in non-LA tumor areas and compared them between the two cohorts (fig. S6). Tumors following GVAX/a–PD-1/SBRT treatment had a significantly lower fold change in CD8⁺ T cells in the tumor areas compared to tumors following GVAX/a–PD-1 treatment (fig. S6A). In contrast, tumors following GVAX/a–PD-1/SBRT treatment did not show a significantly higher fold change in GZMB⁺CD8⁺T cells in the tumor areas compared to PDACs following GVAX/a–PD-1 treatment (fig. S6A). However, tumors following GVAX/a–PD-1/SBRT treatment had a significantly lower fold change in CD8⁺ T cells in the tumor areas in patients with OS < 2 years but not in those with OS > 2 years in comparison with PDACs following GVAX/a–PD-1 treatment (fig. S6B). Nevertheless, compared to PDACs following GVAX/a–PD-1 treatment, tumors following GVAX/a–PD-1/SBRT treatment had a significantly higher fold change in GZMB⁺CD8⁺ T cells in the tumor areas in patients with OS > 2 years but not in those with OS < 2 years (fig. S6C). These results suggested that, although adding SBRT to GVAX/a–PD-1 treatment leads to a decrease in the general population of CD8⁺ T cells in the tumors, it leads to an increase in the high-quality subpopulation of CD8⁺ T cells, such as the GZMB⁺CD8⁺ T cells, and subsequently results in an improved survival.

Next, we assessed the role of GVAX/a–PD-1 treatment in the GVAX/a–PD-1/SBRT combination treatment. We obtained a cohort of PDACs resected after neoadjuvant gemcitabine or fluorouracil-based chemotherapy followed by SBRT (designated SBRT cohort) for comparison (Fig. 5 and table S5). Within the tumors from this SBRT cohort, we did not anticipate the observation of LAs often induced by vaccine therapy. Thus, we compared only the tumor areas from tumors in the SBRT cohort, and the non-LA tumor areas following GVAX/a–PD-1/SBRT treatment showed a significantly lower density of CD8⁺ T cells, a significantly higher density of T_H1, and a significantly lower density of T_H2 and T_regs. Tumors following GVAX/a–PD-1/SBRT treatment also showed a significantly lower density of PD-1⁺CD8⁺ T cells in the tumor areas. Nevertheless, tumors following GVAX/a–PD-1/SBRT treatment exhibited an increased trend in GZMB⁺CD8⁺ T cells in the tumor areas, although not statistically significant, likely due to the small sample size in the SBRT cohort (P = 0.0595). Together, these results suggest that it is not SBRT alone that modulates T_eff infiltration but rather the addition of SBRT to the GVAX/a–PD-1 combo.

Fig. 5. — Data were shown as the means ± SEM and compared by unpaired t test; *P < 0.05, **P < 0.01, and ****P < 0.0001; all others, not significant.

SBRT enhances GVAX/a–PD-1 treatment–mediated recruitment of immunosuppressive M2-like TAMs

It has been described that the infiltration of immunosuppressive cells such as T_regs and M2-like TAM into the TME are associated with poor prognosis and aggressiveness of PDAC (31–33). Although we observed a decreased density of T_regs in LAs from PDAC tumors following GVAX/a–PD-1/SBRT treatment compared to those following GVAX/a–PD-1 treatment, we found an increased M2-like TAM infiltration in LAs following GVAX/a–PD-1/SBRT treatment (Fig. 4C). Therefore, our results suggest that adding SBRT to the GVAX/a–PD-1 treatment has led to increased M2-like TAM infiltration.

We did not observe differences in myeloid cell populations including TANs, M1-like TAMs, DC-LAMP⁺ mature DCs, DC-SIGN⁺ immature DCs, and mast cells in tumors between the SBRT cohort and the cohort that received GVAX/a–PD-1/SBRT treatment (Fig. 5). However, we observed a significantly lower density of M2-like TAMs in the cohort that received GVAX/a–PD-1/SBRT treatment compared to the SBRT cohort (Fig. 5), further suggesting that the addition of SBRT in the GVAX/a–PD-1/SBRT treatment combination caused an elevation in M2-like TAM infiltration.

Adding SBRT to GVAX/a–PD-1 treatment shortens the distances between PD-1⁺CD8⁺ T cells and tumor cells and distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells

Previously, we found that CD8⁺ T_effs mainly reside in LAs but rarely in the vicinity of tumor cells following GVAX/a–PD-1 treatment. However, the higher density of GZMB⁺CD8⁺ T cells in LAs correlated with shorter distance between CD8⁺ T cells and tumor cells (23). Although our results showed that tumors following GVAX/a–PD-1/SBRT treatment had a significantly higher fold change in GZMB⁺CD8⁺ T cells in the tumor areas in patients with OS > 2 years compared to PDACs following GVAX/a–PD-1 treatment (fig. S6C), we wondered whether CD8⁺ T_effs were getting closer to tumor cells following GVAX/a–PD-1/SBRT treatment. Thus, we measured the spatial distances between several key tumor immune cells and tumors cells in the different resected PDAC cohorts following different types of neoadjuvant therapies. The ROIs selected from the above-described images were reprocessed in Halo software (Fig. 6A), and the spatial distances between epithelial tumor cells marked by epithelial cellular adhesion molecule (EpCAM), CD8⁺ T cells marked by CD8, and myeloid cells marked by CSF-1R were measured, respectively (Fig. 6B and dataset S2). Distances from CD8⁺ T cell subtypes (PD-1⁺ and PD-1⁻) and those from CSF-1R⁺ myeloid cell subtypes (PD-L1⁺ and PD-L1⁻; and CD68⁺CD163⁻ M1-like and CD68⁺CD163⁺ M2-like macrophages) were also measured.

Fig. 6. — (A) A representative region from six-marker multiplex IHC images integrated by Halo software. Scale bar, 100 μm. (B) Distance measurement schema of a representative region. Positive signals (exemplified by the EpCAM staining signals) and the nearest neighbor cells (exemplified by CD8⁺ cells) are connected by gray lines whose lengths are measured as the distances. (C to E) Comparison of percentages of PDACs with shorter (<100 μm) and longer (>100 μm) average distances between PD-1⁺ or PD-1⁻ CD8⁺ T cells and tumor cells [(C) and (D)] and between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells (E) in PDACs following different types of neoadjuvant therapies (n = 19 for No-NAT, n = 15 for SBRT, n = 10 for GVAX/a–PD-1, and n = 17 for GVAX/a–PD-1/SBRT). Data were shown as the percentage and compared by chi-square test; *P < 0.05, **P < 0.01, and ****P < 0.0001.

As previously described (23, 34), we used 100 μm as the cutoff to divide PDACs into those with the shorter average distances (<100 μm) and those with the longer average distances (>100 μm) between two types of cells. Here, we did the analysis by using different cutoffs to divide PDACs into two subgroups and found that different cutoffs yielded consistent results and that the 100-μm cutoff remains the most optimal one (fig. S7). A significantly higher percentage of tumors with shorter distances between PD-1⁺CD8⁺ T cells and tumor cells was observed in PDACs following GVAX/a–PD-1/SBRT treatment (13/17, 76.47%) than in those without neoadjuvant therapy (2/19, 10.53%), with neoadjuvant SBRT therapy (4/15, 26.67%) or with GVAX/a–PD-1 treatment (3/10, 30.00%), respectively (Fig. 6C). In contrast, a significantly higher percentage of tumors with shorter distances between PD-1⁻CD8⁺ T cells and tumor cells was observed in PDACs following GVAX/a–PD-1 treatment (10/10, 100.00%) and GVAX/a–PD-1/SBRT treatment (16/17, 94.12%) than in those in the SBRT cohort (9/15, 60.00%) (Fig. 6D).

The percentage of tumors with shorter distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells in tumors following GVAX/a–PD-1/SBRT treatment (16/17, 94.12%) was higher than that in tumors following GVAX/a–PD-1 treatment (7/10, 70.00%); however, the difference did not reach statistical significance (Fig. 6E). Nevertheless, a significantly higher percentage of tumors with shorter distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells was observed in the cohort following GVAX/a–PD-1/SBRT treatment when compared to other cohorts (Fig. 6E). Furthermore, there were no statistically significant differences in the percentages of tumors with shorter distances between CSF-1R⁺ myeloid cells regardless of PD-L1 status and tumor cells among different cohorts (fig. S8A). However, a significantly higher percentage of tumors with shorter distances between PD-L1⁻CSF-1R⁺ myeloid cells and tumor cells was noted in the cohort following GVAX/a–PD-1/SBRT treatment (17/17, 100.00%) than in the SBRT cohort (11/15, 73.33%) (fig. S8A).

We also compared the distances as a continuous variable between PD-1⁺CD8⁺ or PD-1⁻CD8⁺ T cells and tumor cells and distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells in different cohorts (fig. S8B). The distances between PD-1⁺CD8⁺ T cells and tumor cells and the distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells were significantly shorter in the GVAX/a–PD-1/SBRT treatment cohort compared to any other cohort (fig. S8B). The distances between PD-1⁻CD8⁺ T cells and tumor cells were also significantly shorter in the GVAX/a–PD-1/SBRT treatment cohort compared to the SBRT or GVAX/a–PD-1 treatment cohort (fig. S8B). Comparing the GVAX/a–PD-1/SBRT treatment cohort and the No-NAT cohort, the distances between PD-1⁻CD8⁺ T cells or general CD8⁺ T cells and tumor cells had no significant difference (fig. S8B). It is possible that the distance between PD-1⁻CD8⁺ T cells and tumor cells was shorter due to the abundance of tumor cells in untreated tumors. Together, these results showed that adding SBRT to GVAX/a–PD-1 treatment shortened the distances between CD8⁺ T cells, particularly PD-1⁺CD8⁺ T cells and tumor cells, and also the distance between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells in PDACs. Thus, these results suggest that adding SBRT to GVAX/a–PD-1 treatment brings more T_effs into the tumor vicinity and may also potentiate the effect of anti–PD-1/PD-L1 blockade antibody by bringing PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells closer.

We also did the correlative analysis between the density of CD8⁺ T cells, PD-1⁺CD8⁺ T cells, PD-1⁻CD8⁺ T cells, or GZMB⁺CD8⁺ T cells and their distance to other cell types of interest (fig. S9, A to H). The densities of CD8⁺ T cells and its subtypes including PD-1⁺CD8⁺ T cells do not appear to have an impact on the distance between PD-1⁺CD8⁺ T cells and tumor cells or between other cell types of interest and tumor cells (fig. S9, A to D). The higher density of PD-1⁺CD8⁺ T cells is in a strong trend associated with both the longer, not shorter, distance between PD-1⁺CD8⁺ T cells and M1-like TAM and the longer distance between PD-1⁺CD8⁺ T cells and M2-like TAM (fig. S9F). Although it is unknown how higher density of PD-1⁺CD8⁺ T cells could make their distance to TAMs longer, these results further suggests that an increased abundance of CD8⁺ T cells or a CD8⁺ T cell subtype is likely independent from their distance to other cell types.

Bringing the PD-1⁺CD8⁺ T cells closer to tumor cells and closer to PD-L1⁺ myeloid cells by adding SBRT to the GVAX/a–PD-1 treatment correlated with better OS

Next, we correlated the distances between CD8⁺ T cells, myeloid cells, and tumor cells with OS (Fig. 7). The shorter distances between PD-1⁺CD8⁺ T cells and tumor cells correlated with longer OS in PDACs following GVAX/a–PD-1/SBRT treatment; this correlation is nearly statistically significant (P = 0.0889) (Fig. 7A). Such a correlation was not present in other cohorts.

The distances between PD-1⁻CD8⁺ T cells and tumor cells did not correlate with OS in any of these cohorts (Fig. 7B). The distances between general CD8⁺ T cells and tumors similarly did not correlate with OS in any of the cohorts (Fig. 7C). The shorter distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells correlated with longer OS only in the GVAX/a–PD-1/SBRT treatment cohort (Fig. 7D). These results suggest that adding SBRT to the GVAX/a–PD-1 treatment enhances activated T cell and myeloid cell proximity that may also lead to prolonged survival in PDACs.

Adding SBRT to GVAX/a–PD-1 treatment elongated the spatial distances between M2-like TAMs and tumor cells

Since discovering that the shorter distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells correlated with longer OS only in the GVAX/a–PD-1/SBRT treatment cohort, we hypothesized that PD-L1⁺CSF-1R⁺ myeloid cells consist of TAMs. We found that the distances between PD-1⁺CD8⁺ T cells and M1-like TAMs or M2-like TAMs, respectively, were near significantly or significantly elongated in the GVAX/a–PD-1/SBRT treatment cohort compared to the GVAX/a–PD-1 cohort (fig. S10A). Similarly, the distances between tumor cells and M1-like TAMs or M2-like TAMs, respectively, were significantly prolonged in the GVAX/a–PD-1/SBRT treatment cohort compared to the GVAX/a–PD-1 cohort (fig. S10A). Therefore, these TAMs must not fully overlap with PD-L1⁺CSF-1R⁺ myeloid cells. It is possible that the distances between PD-L1⁻ TAMs and CD8⁺ T cells were modulated differently by the GVAX/a–PD-1/SBRT treatment. Supporting this notion, the distances between PD-1⁺CD8⁺ T cells and PD-L1⁻CSF-1R⁺ myeloid cells were not significantly different between the GVAX/a–PD-1/SBRT treatment cohort and the GVAX/a–PD-1 cohort (fig. S10A). Because of the technical limitations in the number of markers to be included in the distance analysis, we were not able to assess the distances between PD-1⁺CD8⁺ T cells and PD-L1⁺ or PD-L1⁻ TAMs at present, which should be measured in the future when there is appropriate technology.

Nevertheless, we found that the distances between tumor cells and M2-like TAMs are potentially prognostic as they were notably elongated in the patients with OS > 2 years compared to those with OS < 2 years in the GVAX/a–PD-1/SBRT treatment cohort (P = 0.1049) but not in the GVAX/a–PD-1 cohort (fig. S10B). This result suggests that adding SBRT to GVAX/a–PD-1 treatment may offer survival benefit by elongating the distances between tumor cells and pro-cancerous macrophages.

DISCUSSION

To our knowledge, this is the most comprehensive studies to analyze the major immune subtypes within the TME of LAPCs following neoadjuvant immunotherapy and to compare the effect of different types of neoadjuvant therapies on the TME of PDACs. This is the largest study of PDACs with pre- and post-immunotherapy tumor specimens and also with spatial analysis of immune cells by multiplex IHC. The results from this study may potentially guide the design of combination treatment strategies and development of previously unidentified immunotherapy strategies for patients with PDAC, particularly those at the locally advanced stage, as well as inform the development of immunotherapeutics in combination with RT.

Both receiving chemotherapy and being diagnosed with a locally advanced stage skew the effector cells in the TME of PDACs toward a more exhausted state. However, SBRT specifically enhances T_eff population induced by a–PD-1 combination immunotherapy that is not observed when given with standard of care chemotherapy. This response favored improved OS, suggesting an immunomodulatory mechanism for RT in sensitizing PDAC tumors to anti–PD-1 therapy. Our results also suggested that such favorable antitumor immune responses are not merely a result of SBRT but a combination of GVAX/a–PD-1/SBRT. These findings may explain the improved survival observed in patients who received GVAX/a–PD-1/SBRT as the neoadjuvant therapy (26) and GVAX/a–PD-1 as adjuvant therapy (24) compared with that in patients who received only standard of care SBRT before surgical resection (4).

The resectability rate in this study was 44% following the neoadjuvant therapy of GVAX/a–PD-1/SBRT (26) compared to 20% following SBRT in the historical control group (4), after both received similar standard of care chemotherapies. A randomized controlled study is needed to prove the significance of this approach for improving the treatment of patients with PDAC. However, this study suggests a potential underlying mechanism for improved resectability, the reduction in suppressive T cell and TAN populations, and the enhanced proximity of T_effs to tumor cells in the PDAC TME. More specific targeting of these immunosuppressive immune cell subtypes may further improve resectability.

Our previous study showed that a higher density of TANs in pretreatment biopsies is associated with poorer prognosis following neoadjuvant and adjuvant therapy of GVAX/a–PD-1 for resectable PDACs (23). Here, we did not observe a survival correlation with TANs density either in pretreatment or in posttreatment tumors following GVAX/a–PD-1/SBRT treatment within either the resected PDACs or unresected PDACs. Nevertheless, the role of TANs in RT remains interesting to be explored, as this study showed that the density of TANs was significantly increased in the LAs in the resected tumors following GVAX/a–PD-1/SBRT treatment compared to GVAX/a–PD-1 treatment. It is possible that TANs would only influence the resectability but not the long-term outcome. In contrast, further targeting M2-like TAM is strongly supported by the finding showing that RT-induced M2-like TAMs is associated with poorer prognosis following GVAX/a–PD-1/SBRT treatment. We previously conducted a preclinical study targeting TAMs with a CCR2/CCR5 dual antagonist in combination with radiation (35) and have recently completed enrollment into a clinical trial of CCR2/CCR5 dual antagonist in combination with nivolumab with or without GVAX following induction chemotherapy and SBRT for LAPCs (NCT03767582). The hypotheses raised by this study will be tested once the outcome results of the second radioimmunotherapy clinical trial become available.

There are several limitations in this study. First, some conclusions drawn from this study are based on the comparisons between tumors from different cohorts of patients obtained in different clinical trials. Although such comparisons would be considered a strength of this study, confounding factors such as different stages and neoadjuvant chemotherapies likely exist and vary among these patient cohorts (table S7). Therefore, some conclusions must be further validated by correlative studies using specimens from a randomized clinical trial. Nevertheless, comparisons of specimens between different cohorts of patients have provided precious opportunities for understanding the immunomodulating role of RT in combination with immunotherapy in PDACs. Second, the analysis of the TME is limited to multiplex IHC. Because of the treatment effect, tumor-infiltrating immune cells were only harvested in a small percentage of resected tumors; thus, a higher resolution study by single-cell RNA sequencing was not feasible. Bulk RNA sequencing was performed but could not distinguish gene expression patterns between tumor and tumor-infiltrating immune cells. Third, although this study was conducted using the most sophisticated cell-to-cell distance measurements to date, this technique is still limited by the number of overlaid markers and minimal densities of cells in the ROIs. Because of these limitations, we were not able to assess the distances related to PD-L1⁺ TAMs and GZMB⁺CD8⁺ T cells. This technique would also require further improvement to feasibly apply this distance measurement method to biopsy specimens.

Although our study supports an antitumor immunomodulating role of RT, it also suggested that adding SBRT to the GVAX/a–PD-1 treatment leads to an increase in M2-like TAM infiltration, suggesting that targeting M2-like TAMs may further enhance antitumor immune response. Supporting this notion, this study finds that adding SBRT to GVAX/a–PD-1 treatment elongates the distance between M2-like TAMs and tumor cells in PDACs with longer survival. We may learn in the near future whether targeting TAMs would further enhance the antitumor immune response of a radio-immunotherapy in PDAC by analyzing the specimens from the above-described clinical trial of CCR2/CCR5 dual antagonist in combination with nivolumab following chemotherapy and SBRT.

MATERIALS AND METHODS

Patients and specimens

From July 2016 to January 2021, 58 patients with LAPC at the Johns Hopkins Hospital were enrolled in a clinical trial under a Johns Hopkins Medical Institution Institutional Review Board approved protocol (26). Written informed consent was obtained from all patients. Of this group, 54 patients completed two cycles of cyclophosphamide (Cy)/GVAX/a–PD-1/SBRT therapy and were evaluable for immune response (table S1). In previous studies, low-dose Cy before GVAX has proved to deplete immunosuppressive T_regs (22, 36). Twenty-four patients had tumors resected after combination therapy, and the remaining 30 patients were unresected. For the 54 enrolled patients, FFPE tissue blocks were obtained, including 54 pretreatment tumor specimens by endoscopic ultrasound-guided fine needle core biopsies and 45 posttreatment tumor specimens composed of 24 surgically resected and 21 biopsied specimens. OS was calculated from the first day of the first cycle of immunotherapy, which would be 3 weeks before the first day of a 5-day course of SBRT at 6.6 gray (Gy) per day, to the date of death or to the last follow-up date on 1 June 2022. To conduct the survival correlation analysis, patients whose OS is greater than 2 years or those who died within 2 years were included. Patients from whom we obtained both pre- and posttreatment specimens were included in a matched pre- and posttreatment comparison.

Seventeen surgically resected PDACs following neoadjuvant therapy of gemcitabine or fluorouracil-based chemotherapy and SBRT at 6.6 Gy per day for 5 days from March 2011 to October 2015 (designated SBRT cohort, table S5) and 21 surgically resected PDACs without neoadjuvant therapy from February 1998 to May 2003 (designated No-NAT cohort, table S6) were also included in the study for comparison as historical control cohorts. For the SBRT cohort, OS was calculated from the first day of SBRT treatment to the date of death or to the last follow-up date on 21 May 2020. For the No-NAT cohort, OS was calculated from the date of diagnosis to the date of death or to the last follow-up date on 16 October 2006. Such an old cohort was chosen because neoadjuvant therapy was given to the majority of localized PDACs more recently. FFPE tissue sections were also obtained for these comparison groups. Furthermore, we obtained our previously published immunological data without modification from the 10 surgically resected PDACs following neoadjuvant therapy of Cy/GVAX/a–PD-1 (designated GVAX/a–PD-1 cohort) (23), which were enrolled in another clinical trial (NCT02451982).

Sequential IHC and image acquisition

Previous studies have described the sequential staining-stripping multiplex IHC protocol (27). In brief, deparaffinized 5-μm FFPE tissue sections were stained by hematoxylin (Dako, S3301), followed by whole-slide bright-field scanning using NanoZoomer (Hamamatsu). After subsequent endogenous peroxidase blocking and microwave heat-induced antigen retrieval with citrate buffer (pH 6.0; BioGenex, HK080-9K), sequential IHC iterations composed of staining, digital image acquisition, and chromogen stripping were performed as previously described (27). Information on the optimized concentrations and incubation times of primary antibodies, the incubation time of horseradish peroxidase-polymer antibody (Nichirei Biosciences Inc.), and the aminoethyl carbazole (Vector Laboratories, SK-4200) reaction times for chromogenic detection were summarized in tables S2 and S3. Negative control images were chosen from those after the last cycle of staining and chromogen stripping without primary antibody staining. In this study, two separate staining panels were designed, and each panel included 15 markers primarily defining either myeloid or lymphoid cells (tables S2 and S3). Every immune marker examined represented a specific aspect of the immune response of interest, which was determined on the basis of the previously published study (27); thus, multiplicity was not controlled in the analysis.

Image processing and analysis

The digitized image processing and analyzing workflow included sequential steps of image co-registration, visualization, and quantitative image analysis. First, digitized images scanned by NanoZoomer were co-registered via the specific CellProfiler v.2.2.1 pipeline designed as previously described (27), shown in fig. S11. Tumor areas for subsequent analyses were circled by pathologists. A minimum of three rectangular ROIs (~3000 × 3000 pixels per ROI) in the vicinity of tumor epithelia identified by the EpCAM staining, which are known to be sufficiently representative of the whole-tumor area as previously described (27), were chosen (fig. S12). Each ROI from the posttreatment surgically resected tumor contains at least one vaccine-induced intratumoral tertiary LA and would be divided into LAs and non-LA tumor areas for a secondary analysis (fig. S1C). According to our prior studies (23), quantifying any three ROIs would yield similar results, suggesting that three ROIs are representative for the whole-tumor area. For pretreatment, posttreatment biopsy specimens or resected tumors from the No-NAT and SBRT cohorts, ROIs were selected in the vicinity of tumor epithelia and also contain CD45⁺ or CD3⁺ cells as previously described (27). According to our prior studies (23), quantifying any three ROIs that cover a minimum of three biopsy cores yielded consistent immune cell density results. Tumor epithelia, large intratumoral blood vessels, and areas with tissue detachment were maximally excluded. As shown in fig. S13, visualization was performed by converting co-registered images into individually pseudo-colored single-marker staining images via ImageJ software v1.48 (National Institutes of Health) (37) and Aperio ImageScope software v.12.3.2.8013 (Leica Biosystems). In the quantitative image analysis step, single-cell segmentation was obtained via the specific CellProfiler v.2.2.1 pipeline as previously described (27), followed by image cytometry analysis and quantification using the FCS Express 7 Image Cytometry software (De Novo Software) (fig. S14). Lymphoid and myeloid cell subtypes were defined by multiple markers as listed in table S4.

Spatial relationship assessment

All image analysis steps of the spatial relationship assessment were performed via Halo Image Analysis Platform software v3.4.2986 (Indica Labs). The steps were as follows: (i) six different sequentially stained images were deconvolved, registered, and fused into a pseudomultiplex immunofluorescent image; (ii) a pathologist annotated the entire tumor region and excluded the necrotic region for the next analysis; (iii) positive staining signals were identified by the threshold adjustment of the signal intensity of each marker via the HighPlex FL module; and (iv) distance and proximity measurements were performed on the basis of the positive signal detected with coordinates and by the spatial module included in the software.

Quantification and statistical analysis

The density of each immune cell subtype was calculated by its percentage among all cells in ROIs. Student’s t test or chi-square test was used for comparison. Paired t test was used for the comparison between pre- and posttreatment biospecimens from same patients, while unpaired t test was used for the comparison between two independent groups. Multiple testing correction was not used because the analysis was either hypothesis testing or explorative in nature. chi-square test was used for the comparison between percentages. All statistical analyses and figures were completed using GraphPad Prism software v9.3.1 (GraphPad Software). A two-sided P value of <0.05 was considered statistically significant.

Acknowledgments

Funding: This study was supported by NIH grants R01 CA169702 (L.Z.), R01 CA197296 (L.Z. and E.M.J.), and P50 CA062924 (E.M.J. and L.Z.); a MISP grant from Merck Sharp & Dohme LLC, a subsidiary of Merck & Co. Inc., Rahway, NJ, USA (L.Z.); a Gateway Foundation grant (L.Z.); and the NCI Cancer Center Support Grant P30 CA006973.

Author contributions: Conceptualization: A.K.N., R.A.A., J.H., D.A.L., J.M.H., V.L., E.M.J., and L.Z. Methodology: N.N., K.L., A.K.N., A.O., D.A.L., V.L., E.M.J., E.D.T., Q.Z., and L.Z. Investigation: J.W., J.G., T.Z., N.N., D.L.T., T.X., C.R., J.D., T.M., R.A.A., A.O., J.H., V.L., and E.D.T. Visualization: J.W., J.G., R.A.A., A.O., Q.Z., and L.Z. Supervision: D.A.L., J.M.H., E.M.J., and L.Z. Validation: J.W., R.P., J.D., A.O., and E.M.J. Formal analysis: J.W., J.G., H.Q., T.M., H.W., and Q.Z. Data curation: J.W., C.R., R.P., A.O., and H.W. Software: T.M. and Q.Z. Writing—original draft: J.W., J.M.H., and L.Z. Writing—review and editing: J.W., J.G., D.L.T., C.R., R.P., R.A.A., A.O., H.W., J.M.H., E.M.J., E.D.T., and L.Z. Funding acquisition: E.M.J. and L.Z. Project administration: C.R., R.P., J.D., H.W., D.A.L., and L.Z. Resources: J.W., D.L.T., A.O., J.H., E.M.J., E.D.T., and L.Z.

Competing interests: L.Z. receives grant support from Bristol-Meyer Squibb, Merck, AstraZeneca, iTeos, Amgen, NovaRock, Inxmed, Halozyme, and Abmeta. L.Z. is a paid consultant/Advisory Board member at Biosion, Alphamab, NovaRock, Ambrx, Akrevia/Xilio, QED, Natera, Novagenesis, Snow Lake Capitals, BioArdis, Tempus, Amberstone, Pfizer, Tavotek, and Mingruizhiyao. L.Z. holds shares at Alphamab, Amberstone, and Mingruizhiyao. E.M.J. receives grant support from Lustgarten Foundation, Bristol-Meyer Squibb, Genentech, and AstroZeneca; is a paid consultant for NextCure, Genocea, DragonFly, Stimit, CSTONE, Achilles, and Candel; is on the advisory board of Parker Institute and Break Through Cancer; is a founder of Abmeta Biotech; and is the Chief Medical Advisor for the Lustgarten Foundation. J.M.H. consults and holds equity from Histosonics, consults for BTG, and receives research support to the Northwell Health System from Canopy Cancer Collective/1440 Foundation. R.A.A. serves on the advisory board of RAPT Therapeutics, Astra Zeneca, Merck SD, and Bristol-Myers Squib and receives research funding from RAPT Therapeutic, Bristol-Myers Squib, and the National Institute of Health. There is no relevant conflict of interest disclosed by all other authors.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

This PDF file includes:

Figs. S1 to S14

Tables S1 to S7

Legends for datasets S1 and S2

Click here for additional data file.^{(6.4MB, pdf)}

Other Supplementary Material for this manuscript includes the following:

Datasets S1 and S2

Click here for additional data file.^{(57.7KB, zip)}

REFERENCES AND NOTES

1.Siegel R. L., Miller K. D., Wagle N. S., Jemal A., Cancer statistics, 2023. CA Cancer J. Clin. 73, 17–48 (2023). [DOI] [PubMed] [Google Scholar]
2.Hidalgo M., Pancreatic cancer. N. Engl. J. Med. 362, 1605–1617 (2010). [DOI] [PubMed] [Google Scholar]
3.Zhong J., Patel K., Switchenko J., Cassidy R. J., Hall W. A., Gillespie T., Patel P. R., Kooby D., Landry J., Outcomes for patients with locally advanced pancreatic adenocarcinoma treated with stereotactic body radiation therapy versus conventionally fractionated radiation. Cancer 123, 3486–3493 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Gemenetzis G., Groot V. P., Blair A. B., Laheru D. A., Zheng L., Narang A. K., Fishman E. K., Hruban R. H., Yu J., Burkhart R. A., Cameron J. L., Weiss M. J., Wolfgang C. L., He J., Survival in locally advanced pancreatic cancer after neoadjuvant therapy and surgical resection. Ann. Surg. 270, 340–347 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.He J., Blair A. B., Groot V. P., Javed A. A., Burkhart R. A., Gemenetzis G., Hruban R. H., Waters K. M., Poling J., Zheng L., Laheru D., Herman J. M., Makary M. A., Weiss M. J., Cameron J. L., Wolfgang C. L., Is a pathological complete response following neoadjuvant chemoradiation associated with prolonged survival in patients with pancreatic cancer? Ann. Surg. 268, 1–8 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Wild A. T., Herman J. M., Dholakia A. S., Moningi S., Lu Y., Rosati L. M., Hacker-Prietz A., Assadi R. K., Saeed A. M., Pawlik T. M., Jaffee E. M., Laheru D. A., Tran P. T., Weiss M. J., Wolfgang C. L., Ford E., Grossman S. A., Ye X., Ellsworth S. G., Lymphocyte-sparing effect of stereotactic body radiation therapy in patients with unresectable pancreatic cancer. Int. J. Radiat. Oncol. Biol. Phys. 94, 571–579 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Foley K., Kim V., Jaffee E., Zheng L., Current progress in immunotherapy for pancreatic cancer. Cancer Lett. 381, 244–251 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Kleponis J., Skelton R., Zheng L., Fueling the engine and releasing the break: Combinational therapy of cancer vaccines and immune checkpoint inhibitors. Cancer Biol. Med. 12, 201–208 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Wu A. A., Drake V., Huang H.-S., Chiu S., Zheng L., Reprogramming the tumor microenvironment: Tumor-induced immunosuppressive factors paralyze T cells. Onco. Targets. Ther. 4, e1016700 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Sultana A., Tudur Smith C., Cunningham D., Starling N., Tait D., Neoptolemos J. P., Ghaneh P., Systematic review, including meta-analyses, on the management of locally advanced pancreatic cancer using radiation/combined modality therapy. Br. J. Cancer 96, 1183–1190 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Goodman K. A., Hajj C., Role of radiation therapy in the management of pancreatic cancer. J. Surg. Oncol. 107, 86–96 (2013). [DOI] [PubMed] [Google Scholar]
12.Zhu X., Li F., Ju X., Shen Y., Cao Y., Cao F., Fang F., Qing S., Jia Z., Zhang H., Prediction of overall survival after re-irradiation with stereotactic body radiation therapy for pancreatic cancer with a novel prognostic model (the SCAD score). Radiother. Oncol. 129, 313–318 (2018). [DOI] [PubMed] [Google Scholar]
13.Formenti S. C., Demaria S., Radiation therapy to convert the tumor into an in situ vaccine. Int. J. Radiat. Oncol. Biol. Phys. 84, 879–880 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Yasmin-Karim S., Bruck P. T., Moreau M., Kunjachan S., Chen G. Z., Kumar R., Grabow S., Dougan S. K., Ngwa W., Radiation and local anti-CD40 Generate an effective in situ vaccine in preclinical models of pancreatic cancer. Front. Immunol. 9, 2030 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Aliru M. L., Schoenhals J. E., Venkatesulu B. P., Anderson C. C., Barsoumian H. B., Younes A. I., K Mahadevan L. S., Soeung M., Aziz K. E., Welsh J. W., Krishnan S., Radiation therapy and immunotherapy: What is the optimal timing or sequencing? Immunotherapy 10, 299–316 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
16.Azad A., Yin Lim S., D'Costa Z., Jones K., Diana A., Sansom O. J., Kruger P., Liu S., McKenna W. G., Dushek O., Muschel R. J., Fokas E., PD-L1 blockade enhances response of pancreatic ductal adenocarcinoma to radiotherapy. EMBO Mol. Med. 9, 167–180 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Fujiwara K., Saung M. T., Jing H., Herbst B., Zarecki M. K., Muth S., Wu A., Bigelow E., Chen L., Li K., Jurcak N., Blair A. B., Ding D., Wichroski M., Blum J., Cheadle N., Koenitzer J., Zheng L., Interrogating the immune-modulating roles of radiation therapy for a rational combination with immune-checkpoint inhibitors in treating pancreatic cancer. J. Immunother. Cancer 8, e000351 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Stump C. T., Roehle K., Manjarrez Orduno N., Dougan S. K., Radiation combines with immune checkpoint blockade to enhance T cell priming in a murine model of poorly immunogenic pancreatic cancer. Open Biol. 11, 210245 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Xie C., Duffy A. G., Brar G., Fioravanti S., Mabry-Hrones D., Walker M., Bonilla C. M., Wood B. J., Citrin D. E., Gil Ramirez E. M., Escorcia F. E., Redd B., Hernandez J. M., Davis J. L., Gasmi B., Kleiner D., Steinberg S. M., Jones J. C., Greten T. F., Immune checkpoint blockade in combination with stereotactic body radiotherapy in patients with metastatic pancreatic ductal adenocarcinoma. Clin. Cancer Res. 26, 2318–2326 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Parikh A. R., Szabolcs A., Allen J. N., Clark J. W., Wo J. Y., Raabe M., Thel H., Hoyos D., Mehta A., Arshad S., Lieb D. J., Drapek L. C., Blaszkowsky L. S., Giantonio B. J., Weekes C. D., Zhu A. X., Goyal L., Nipp R. D., Dubois J. S., van Seventer E. E., Foreman B. E., Matlack L. E., Ly L., Meurer J. A., Hacohen N., Ryan D. P., Yeap B. Y., Corcoran R. B., Greenbaum B. D., Ting D. T., Hong T. S., Radiation therapy enhances immunotherapy response in microsatellite stable colorectal and pancreatic adenocarcinoma in a phase II trial. Nat. Cancer 2, 1124–1135 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Chen I. M., Johansen J. S., Theile S., Hjaltelin J. X., Novitski S. I., Brunak S., Hasselby J. P., Willemoe G. L., Lorentzen T., Madsen K., Jensen B. V., Wilken E. E., Geertsen P., Behrens C., Nolsoe C., Hermann K. L., Svane I. M., Nielsen D., Randomized phase II study of nivolumab with or without ipilimumab combined with stereotactic body radiotherapy for refractory metastatic pancreatic cancer (CheckPAC). J. Clin. Oncol. 40, 3180–3189 (2022). [DOI] [PubMed] [Google Scholar]
22.Lutz E. R., Wu A. A., Bigelow E., Sharma R., Mo G., Soares K., Solt S., Dorman A., Wamwea A., Yager A., Laheru D., Wolfgang C. L., Wang J., Hruban R. H., Anders R. A., Jaffee E. M., Zheng L., Immunotherapy converts nonimmunogenic pancreatic tumors into immunogenic foci of immune regulation. Cancer Immunol. Res. 2, 616–631 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Li K., Tandurella J. A., Gai J., Zhu Q., Lim S. J., Thomas D. L. II, Xia T., Mo G., Mitchell J. T., Montagne J., Lyman M., Danilova L. V., Zimmerman J. W., Kinny-Köster B., Zhang T., Chen L., Blair A. B., Heumann T., Parkinson R., Durham J. N., Narang A. K., Anders R. A., Wolfgang C. L., Laheru D. A., He J., Osipov A., Thompson E. D., Wang H., Fertig E. J., Jaffee E. M., Zheng L., Multi-omic analyses of changes in the tumor microenvironment of pancreatic adenocarcinoma following neoadjuvant treatment with anti-PD-1 therapy. Cancer Cell 40, 1374–1391.e7 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Heumann T., Judkins C., Li K., Lim S. J., Hoare J., Parkinson R., Cao H., Zhang T., Gai J., Celiker B., Zhu Q., McPhaul T., Durham J., Purtell K., Klein R., Laheru D., de Jesus-Acosta A., le D. T., Narang A., Anders R., Burkhart R., Burns W., Soares K., Wolfgang C., Thompson E., Jaffee E., Wang H., He J., Zheng L., A platform trial of neoadjuvant and adjuvant antitumor vaccination alone or in combination with PD-1 antagonist and CD137 agonist antibodies in patients with resectable pancreatic adenocarcinoma. Nat. Commun. 14, 3650 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Tsujikawa T., Crocenzi T., Durham J. N., Sugar E. A., Wu A. A., Onners B., Nauroth J. M., Anders R. A., Fertig E. J., Laheru D. A., Reiss K., Vonderheide R. H., Ko A. H., Tempero M. A., Fisher G. A., Considine M., Danilova L., Brockstedt D. G., Coussens L. M., Jaffee E. M., le D. T., Evaluation of cyclophosphamide/GVAX pancreas followed by Listeria-Mesothelin (CRS-207) with or without nivolumab in patients with pancreatic cancer. Clin. Cancer Res. 26, 3578–3588 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Lee V., Ding D., Rodriguez C., Onners B., Narang A., Meyer J., Herman J. M., Hacker-Prietz A., Burkhart R. A., Burns W., He J., de Jesus-Acosta A., Klein R. B., Laheru D., le D. T., Ryan A., Sugar E., Zheng L., A phase 2 study of cyclophosphamide (CY), GVAX, pembrolizumab (Pembro), and stereotactic body radiation (SBRT) in patients (pts) with locally advanced pancreas cancer (LAPC). J. Clin. Oncol. 39, 4134–4134 (2021). [Google Scholar]
27.Tsujikawa T., Kumar S., Borkar R. N., Azimi V., Thibault G., Chang Y. H., Balter A., Kawashima R., Choe G., Sauer D., el Rassi E., Clayburgh D. R., Kulesz-Martin M. F., Lutz E. R., Zheng L., Jaffee E. M., Leyshock P., Margolin A. A., Mori M., Gray J. W., Flint P. W., Coussens L. M., Quantitative multiplex immunohistochemistry reveals myeloid-inflamed tumor-immune complexity associated with poor prognosis. Cell Rep. 19, 203–217 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Wang X., Lang M., Zhao T., Feng X., Zheng C., Huang C., Hao J., Dong J., Luo L., Li X., Lan C., Yu W., Yu M., Yang S., Ren H., Cancer-FOXP3 directly activated CCL5 to recruit FOXP3⁺Treg cells in pancreatic ductal adenocarcinoma. Oncogene 36, 3048–3058 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Mougiakakos D., Choudhury A., Lladser A., Kiessling R., Johansson C. C., Regulatory T cells in cancer. Adv. Cancer Res. 107, (2010). [DOI] [PubMed] [Google Scholar]
30.Pushalkar S., Hundeyin M., Daley D., Zambirinis C. P., Kurz E., Mishra A., Mohan N., Aykut B., Usyk M., Torres L. E., Werba G., Zhang K., Guo Y., Li Q., Akkad N., Lall S., Wadowski B., Gutierrez J., Kochen Rossi J. A., Herzog J. W., Diskin B., Torres-Hernandez A., Leinwand J., Wang W., Taunk P. S., Savadkar S., Janal M., Saxena A., Li X., Cohen D., Sartor R. B., Saxena D., Miller G., The pancreatic cancer microbiome promotes oncogenesis by induction of innate and adaptive immune suppression. Cancer Discov. 8, 403–416 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Tang Y., Xu X., Guo S., Zhang C., Tang Y., Tian Y., Ni B., Lu B., Wang H., An increased abundance of tumor-infiltrating regulatory T cells is correlated with the progression and prognosis of pancreatic ductal adenocarcinoma. PLOS One 9, e91551 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Hiraoka N., Onozato K., Kosuge T., Hirohashi S., Prevalence of FOXP3⁺ regulatory T cells increases during the progression of pancreatic ductal adenocarcinoma and its premalignant lesions. Clin. Cancer Res. 12, 5423–5434 (2006). [DOI] [PubMed] [Google Scholar]
33.Yang S., Liu Q., Liao Q., Tumor-associated macrophages in pancreatic ductal adenocarcinoma: Origin, polarization, function, and reprogramming. Front. Cell Dev. Biol. 8, 607209 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Yan C., Huang H., Zheng Z., Ma X., Zhao G., Zhang T., Chen X., Cao F., Wei H., Dong J., Tang P., Jiang H., Wang M., Wang P., Pang Q., Zhang W., Spatial distribution of tumor-infiltrating T cells indicated immune response status under chemoradiotherapy plus PD-1 blockade in esophageal cancer. Front. Immunol. 14, 1138054 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Wang J., Saung M. T., Li K., Fu J., Fujiwara K., Niu N., Muth S., Wang J., Xu Y., Rozich N., Zlomke H., Chen S., Espinoza B., Henderson M. K., Funes V., Herbst B., Ding D., Twyman-Saint Victor C., Zhao Q., Narang A., He J., Zheng L., CCR2/CCR5 inhibitor permits the radiation-induced effector T cell infiltration in pancreatic adenocarcinoma. J. Exp. Med. 219, (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Laheru D., Lutz E., Burke J., Biedrzycki B., Solt S., Onners B., Tartakovsky I., Nemunaitis J., le D., Sugar E., Hege K., Jaffee E., Allogeneic granulocyte macrophage colony-stimulating factor-secreting tumor immunotherapy alone or in sequence with cyclophosphamide for metastatic pancreatic cancer: A pilot study of safety, feasibility, and immune activation. Clin. Cancer Res. 14, 1455–1463 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Schneider C. A., Rasband W. S., Eliceiri K. W., NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figs. S1 to S14

Tables S1 to S7

Legends for datasets S1 and S2

Click here for additional data file.^{(6.4MB, pdf)}

Datasets S1 and S2

Click here for additional data file.^{(57.7KB, zip)}

[R1] 1.Siegel R. L., Miller K. D., Wagle N. S., Jemal A., Cancer statistics, 2023. CA Cancer J. Clin. 73, 17–48 (2023). [DOI] [PubMed] [Google Scholar]

[R2] 2.Hidalgo M., Pancreatic cancer. N. Engl. J. Med. 362, 1605–1617 (2010). [DOI] [PubMed] [Google Scholar]

[R3] 3.Zhong J., Patel K., Switchenko J., Cassidy R. J., Hall W. A., Gillespie T., Patel P. R., Kooby D., Landry J., Outcomes for patients with locally advanced pancreatic adenocarcinoma treated with stereotactic body radiation therapy versus conventionally fractionated radiation. Cancer 123, 3486–3493 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Gemenetzis G., Groot V. P., Blair A. B., Laheru D. A., Zheng L., Narang A. K., Fishman E. K., Hruban R. H., Yu J., Burkhart R. A., Cameron J. L., Weiss M. J., Wolfgang C. L., He J., Survival in locally advanced pancreatic cancer after neoadjuvant therapy and surgical resection. Ann. Surg. 270, 340–347 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R5] 5.He J., Blair A. B., Groot V. P., Javed A. A., Burkhart R. A., Gemenetzis G., Hruban R. H., Waters K. M., Poling J., Zheng L., Laheru D., Herman J. M., Makary M. A., Weiss M. J., Cameron J. L., Wolfgang C. L., Is a pathological complete response following neoadjuvant chemoradiation associated with prolonged survival in patients with pancreatic cancer? Ann. Surg. 268, 1–8 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Wild A. T., Herman J. M., Dholakia A. S., Moningi S., Lu Y., Rosati L. M., Hacker-Prietz A., Assadi R. K., Saeed A. M., Pawlik T. M., Jaffee E. M., Laheru D. A., Tran P. T., Weiss M. J., Wolfgang C. L., Ford E., Grossman S. A., Ye X., Ellsworth S. G., Lymphocyte-sparing effect of stereotactic body radiation therapy in patients with unresectable pancreatic cancer. Int. J. Radiat. Oncol. Biol. Phys. 94, 571–579 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Foley K., Kim V., Jaffee E., Zheng L., Current progress in immunotherapy for pancreatic cancer. Cancer Lett. 381, 244–251 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R8] 8.Kleponis J., Skelton R., Zheng L., Fueling the engine and releasing the break: Combinational therapy of cancer vaccines and immune checkpoint inhibitors. Cancer Biol. Med. 12, 201–208 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R9] 9.Wu A. A., Drake V., Huang H.-S., Chiu S., Zheng L., Reprogramming the tumor microenvironment: Tumor-induced immunosuppressive factors paralyze T cells. Onco. Targets. Ther. 4, e1016700 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R10] 10.Sultana A., Tudur Smith C., Cunningham D., Starling N., Tait D., Neoptolemos J. P., Ghaneh P., Systematic review, including meta-analyses, on the management of locally advanced pancreatic cancer using radiation/combined modality therapy. Br. J. Cancer 96, 1183–1190 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R11] 11.Goodman K. A., Hajj C., Role of radiation therapy in the management of pancreatic cancer. J. Surg. Oncol. 107, 86–96 (2013). [DOI] [PubMed] [Google Scholar]

[R12] 12.Zhu X., Li F., Ju X., Shen Y., Cao Y., Cao F., Fang F., Qing S., Jia Z., Zhang H., Prediction of overall survival after re-irradiation with stereotactic body radiation therapy for pancreatic cancer with a novel prognostic model (the SCAD score). Radiother. Oncol. 129, 313–318 (2018). [DOI] [PubMed] [Google Scholar]

[R13] 13.Formenti S. C., Demaria S., Radiation therapy to convert the tumor into an in situ vaccine. Int. J. Radiat. Oncol. Biol. Phys. 84, 879–880 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Yasmin-Karim S., Bruck P. T., Moreau M., Kunjachan S., Chen G. Z., Kumar R., Grabow S., Dougan S. K., Ngwa W., Radiation and local anti-CD40 Generate an effective in situ vaccine in preclinical models of pancreatic cancer. Front. Immunol. 9, 2030 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Aliru M. L., Schoenhals J. E., Venkatesulu B. P., Anderson C. C., Barsoumian H. B., Younes A. I., K Mahadevan L. S., Soeung M., Aziz K. E., Welsh J. W., Krishnan S., Radiation therapy and immunotherapy: What is the optimal timing or sequencing? Immunotherapy 10, 299–316 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R16] 16.Azad A., Yin Lim S., D'Costa Z., Jones K., Diana A., Sansom O. J., Kruger P., Liu S., McKenna W. G., Dushek O., Muschel R. J., Fokas E., PD-L1 blockade enhances response of pancreatic ductal adenocarcinoma to radiotherapy. EMBO Mol. Med. 9, 167–180 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Fujiwara K., Saung M. T., Jing H., Herbst B., Zarecki M. K., Muth S., Wu A., Bigelow E., Chen L., Li K., Jurcak N., Blair A. B., Ding D., Wichroski M., Blum J., Cheadle N., Koenitzer J., Zheng L., Interrogating the immune-modulating roles of radiation therapy for a rational combination with immune-checkpoint inhibitors in treating pancreatic cancer. J. Immunother. Cancer 8, e000351 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Stump C. T., Roehle K., Manjarrez Orduno N., Dougan S. K., Radiation combines with immune checkpoint blockade to enhance T cell priming in a murine model of poorly immunogenic pancreatic cancer. Open Biol. 11, 210245 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R19] 19.Xie C., Duffy A. G., Brar G., Fioravanti S., Mabry-Hrones D., Walker M., Bonilla C. M., Wood B. J., Citrin D. E., Gil Ramirez E. M., Escorcia F. E., Redd B., Hernandez J. M., Davis J. L., Gasmi B., Kleiner D., Steinberg S. M., Jones J. C., Greten T. F., Immune checkpoint blockade in combination with stereotactic body radiotherapy in patients with metastatic pancreatic ductal adenocarcinoma. Clin. Cancer Res. 26, 2318–2326 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R20] 20.Parikh A. R., Szabolcs A., Allen J. N., Clark J. W., Wo J. Y., Raabe M., Thel H., Hoyos D., Mehta A., Arshad S., Lieb D. J., Drapek L. C., Blaszkowsky L. S., Giantonio B. J., Weekes C. D., Zhu A. X., Goyal L., Nipp R. D., Dubois J. S., van Seventer E. E., Foreman B. E., Matlack L. E., Ly L., Meurer J. A., Hacohen N., Ryan D. P., Yeap B. Y., Corcoran R. B., Greenbaum B. D., Ting D. T., Hong T. S., Radiation therapy enhances immunotherapy response in microsatellite stable colorectal and pancreatic adenocarcinoma in a phase II trial. Nat. Cancer 2, 1124–1135 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R21] 21.Chen I. M., Johansen J. S., Theile S., Hjaltelin J. X., Novitski S. I., Brunak S., Hasselby J. P., Willemoe G. L., Lorentzen T., Madsen K., Jensen B. V., Wilken E. E., Geertsen P., Behrens C., Nolsoe C., Hermann K. L., Svane I. M., Nielsen D., Randomized phase II study of nivolumab with or without ipilimumab combined with stereotactic body radiotherapy for refractory metastatic pancreatic cancer (CheckPAC). J. Clin. Oncol. 40, 3180–3189 (2022). [DOI] [PubMed] [Google Scholar]

[R22] 22.Lutz E. R., Wu A. A., Bigelow E., Sharma R., Mo G., Soares K., Solt S., Dorman A., Wamwea A., Yager A., Laheru D., Wolfgang C. L., Wang J., Hruban R. H., Anders R. A., Jaffee E. M., Zheng L., Immunotherapy converts nonimmunogenic pancreatic tumors into immunogenic foci of immune regulation. Cancer Immunol. Res. 2, 616–631 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R23] 23.Li K., Tandurella J. A., Gai J., Zhu Q., Lim S. J., Thomas D. L. II, Xia T., Mo G., Mitchell J. T., Montagne J., Lyman M., Danilova L. V., Zimmerman J. W., Kinny-Köster B., Zhang T., Chen L., Blair A. B., Heumann T., Parkinson R., Durham J. N., Narang A. K., Anders R. A., Wolfgang C. L., Laheru D. A., He J., Osipov A., Thompson E. D., Wang H., Fertig E. J., Jaffee E. M., Zheng L., Multi-omic analyses of changes in the tumor microenvironment of pancreatic adenocarcinoma following neoadjuvant treatment with anti-PD-1 therapy. Cancer Cell 40, 1374–1391.e7 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.Heumann T., Judkins C., Li K., Lim S. J., Hoare J., Parkinson R., Cao H., Zhang T., Gai J., Celiker B., Zhu Q., McPhaul T., Durham J., Purtell K., Klein R., Laheru D., de Jesus-Acosta A., le D. T., Narang A., Anders R., Burkhart R., Burns W., Soares K., Wolfgang C., Thompson E., Jaffee E., Wang H., He J., Zheng L., A platform trial of neoadjuvant and adjuvant antitumor vaccination alone or in combination with PD-1 antagonist and CD137 agonist antibodies in patients with resectable pancreatic adenocarcinoma. Nat. Commun. 14, 3650 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R25] 25.Tsujikawa T., Crocenzi T., Durham J. N., Sugar E. A., Wu A. A., Onners B., Nauroth J. M., Anders R. A., Fertig E. J., Laheru D. A., Reiss K., Vonderheide R. H., Ko A. H., Tempero M. A., Fisher G. A., Considine M., Danilova L., Brockstedt D. G., Coussens L. M., Jaffee E. M., le D. T., Evaluation of cyclophosphamide/GVAX pancreas followed by Listeria-Mesothelin (CRS-207) with or without nivolumab in patients with pancreatic cancer. Clin. Cancer Res. 26, 3578–3588 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R26] 26.Lee V., Ding D., Rodriguez C., Onners B., Narang A., Meyer J., Herman J. M., Hacker-Prietz A., Burkhart R. A., Burns W., He J., de Jesus-Acosta A., Klein R. B., Laheru D., le D. T., Ryan A., Sugar E., Zheng L., A phase 2 study of cyclophosphamide (CY), GVAX, pembrolizumab (Pembro), and stereotactic body radiation (SBRT) in patients (pts) with locally advanced pancreas cancer (LAPC). J. Clin. Oncol. 39, 4134–4134 (2021). [Google Scholar]

[R27] 27.Tsujikawa T., Kumar S., Borkar R. N., Azimi V., Thibault G., Chang Y. H., Balter A., Kawashima R., Choe G., Sauer D., el Rassi E., Clayburgh D. R., Kulesz-Martin M. F., Lutz E. R., Zheng L., Jaffee E. M., Leyshock P., Margolin A. A., Mori M., Gray J. W., Flint P. W., Coussens L. M., Quantitative multiplex immunohistochemistry reveals myeloid-inflamed tumor-immune complexity associated with poor prognosis. Cell Rep. 19, 203–217 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R28] 28.Wang X., Lang M., Zhao T., Feng X., Zheng C., Huang C., Hao J., Dong J., Luo L., Li X., Lan C., Yu W., Yu M., Yang S., Ren H., Cancer-FOXP3 directly activated CCL5 to recruit FOXP3⁺Treg cells in pancreatic ductal adenocarcinoma. Oncogene 36, 3048–3058 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R29] 29.Mougiakakos D., Choudhury A., Lladser A., Kiessling R., Johansson C. C., Regulatory T cells in cancer. Adv. Cancer Res. 107, (2010). [DOI] [PubMed] [Google Scholar]

[R30] 30.Pushalkar S., Hundeyin M., Daley D., Zambirinis C. P., Kurz E., Mishra A., Mohan N., Aykut B., Usyk M., Torres L. E., Werba G., Zhang K., Guo Y., Li Q., Akkad N., Lall S., Wadowski B., Gutierrez J., Kochen Rossi J. A., Herzog J. W., Diskin B., Torres-Hernandez A., Leinwand J., Wang W., Taunk P. S., Savadkar S., Janal M., Saxena A., Li X., Cohen D., Sartor R. B., Saxena D., Miller G., The pancreatic cancer microbiome promotes oncogenesis by induction of innate and adaptive immune suppression. Cancer Discov. 8, 403–416 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Tang Y., Xu X., Guo S., Zhang C., Tang Y., Tian Y., Ni B., Lu B., Wang H., An increased abundance of tumor-infiltrating regulatory T cells is correlated with the progression and prognosis of pancreatic ductal adenocarcinoma. PLOS One 9, e91551 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Hiraoka N., Onozato K., Kosuge T., Hirohashi S., Prevalence of FOXP3⁺ regulatory T cells increases during the progression of pancreatic ductal adenocarcinoma and its premalignant lesions. Clin. Cancer Res. 12, 5423–5434 (2006). [DOI] [PubMed] [Google Scholar]

[R33] 33.Yang S., Liu Q., Liao Q., Tumor-associated macrophages in pancreatic ductal adenocarcinoma: Origin, polarization, function, and reprogramming. Front. Cell Dev. Biol. 8, 607209 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Yan C., Huang H., Zheng Z., Ma X., Zhao G., Zhang T., Chen X., Cao F., Wei H., Dong J., Tang P., Jiang H., Wang M., Wang P., Pang Q., Zhang W., Spatial distribution of tumor-infiltrating T cells indicated immune response status under chemoradiotherapy plus PD-1 blockade in esophageal cancer. Front. Immunol. 14, 1138054 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Wang J., Saung M. T., Li K., Fu J., Fujiwara K., Niu N., Muth S., Wang J., Xu Y., Rozich N., Zlomke H., Chen S., Espinoza B., Henderson M. K., Funes V., Herbst B., Ding D., Twyman-Saint Victor C., Zhao Q., Narang A., He J., Zheng L., CCR2/CCR5 inhibitor permits the radiation-induced effector T cell infiltration in pancreatic adenocarcinoma. J. Exp. Med. 219, (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R36] 36.Laheru D., Lutz E., Burke J., Biedrzycki B., Solt S., Onners B., Tartakovsky I., Nemunaitis J., le D., Sugar E., Hege K., Jaffee E., Allogeneic granulocyte macrophage colony-stimulating factor-secreting tumor immunotherapy alone or in sequence with cyclophosphamide for metastatic pancreatic cancer: A pilot study of safety, feasibility, and immune activation. Clin. Cancer Res. 14, 1455–1463 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] 37.Schneider C. A., Rasband W. S., Eliceiri K. W., NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 9, 671–675 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Neoadjuvant radioimmunotherapy in pancreatic cancer enhances effector T cell infiltration and shortens their distances to tumor cells

Junke Wang

Jessica Gai

Tengyi Zhang

Nan Niu

Hanfei Qi

Dwayne L Thomas II

Keyu Li

Tao Xia

Christina Rodriguez

Rose Parkinson

Jennifer Durham

Thomas McPhaul

Amol K Narang

Robert A Anders

Arsen Osipov

Hao Wang

Jin He

Daniel A Laheru

Joseph M Herman

Valerie Lee

Elizabeth M Jaffee

Elizabeth D Thompson

Qingfeng Zhu

Lei Zheng

Roles

Abstract

INTRODUCTION

RESULTS

Combination treatment with pembrolizumab, vaccine therapy, and SBRT induces CD4+ and CD8+ T lymphocyte infiltration into posttreatment PDAC tumors

Fig. 1. Multiplex IHC analysis of PDACs with different types of neoadjuvant therapies.

An increase in high-quality CD8+ T cell population occurs with combination GVAX + a–PD-1 + SBRT treatment and correlates with better survival in patients with resectable PDAC

Fig. 2. Multiplex IHC analysis of CD8+ and GZMB+CD8+ T cells and their correlation with OS in PDACs with GVAX/a–PD-1/SBRT treatment.

An increase in TH1 and TH17 density correlates with improved OS in patients treated with the combination GVAX/a–PD-1/SBRT

Fig. 3. Correlative analysis of OS with the changes of the densities of CD4+ T cell subtypes and myeloid cell subtypes between matched pre- and posttreatment tumor areas in PDACs with GVAX/a–PD-1/SBRT treatment.

The EOMES-mediated T cell exhaustion pathway predicts clinical outcome of non-resectable PDACs following GVAX/a–PD-1/SBRT treatment

Changes in M2-like TAMs following GVAX/a–PD-1/SBRT treatment correlates with patient outcomes

Untreated PDACs and chemotherapy/RT-treated PDACs are characterized by distinct TMEs

Fig. 4. Summary of the density of all immune cell subtypes analyzed in resected PDACs with GVAX/a–PD-1 treatment versus GVAX/a–PD-1/SBRT treatment.

SBRT enhances the GVAX/a–PD-1–mediated increase of GZMB+CD8+ Teffs

Fig. 5. Summary of the density of all immune cell subtypes analyzed as indicated in the posttreatment tumor areas in resected PDACs with SBRT treatment (n = 17) versus GVAX/a–PD-1/SBRT treatment (n = 24).

SBRT enhances GVAX/a–PD-1 treatment–mediated recruitment of immunosuppressive M2-like TAMs

Adding SBRT to GVAX/a–PD-1 treatment shortens the distances between PD-1+CD8+ T cells and tumor cells and distances between PD-1+CD8+ T cells and PD-L1+CSF-1R+ myeloid cells

Fig. 6. Spatial distance measurement between CD8+ T cells, CSF-1R+ myeloid cells, and tumor cells in PDACs following different types of neoadjuvant therapies.

Bringing the PD-1+CD8+ T cells closer to tumor cells and closer to PD-L1+ myeloid cells by adding SBRT to the GVAX/a–PD-1 treatment correlated with better OS

Fig. 7. Correlative analysis of OS with the average distances between CD8+ T cells and tumor cells and between PD-1+CD8+ T cells and PD-L1+CSF-1R+ myeloid cells in PDACs following different types of neoadjuvant therapies.

Adding SBRT to GVAX/a–PD-1 treatment elongated the spatial distances between M2-like TAMs and tumor cells

DISCUSSION

MATERIALS AND METHODS

Patients and specimens

Sequential IHC and image acquisition

Image processing and analysis

Spatial relationship assessment

Quantification and statistical analysis

Acknowledgments

Supplementary Materials

This PDF file includes:

Other Supplementary Material for this manuscript includes the following:

REFERENCES AND NOTES

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

Combination treatment with pembrolizumab, vaccine therapy, and SBRT induces CD4⁺ and CD8⁺ T lymphocyte infiltration into posttreatment PDAC tumors

An increase in high-quality CD8⁺ T cell population occurs with combination GVAX + a–PD-1 + SBRT treatment and correlates with better survival in patients with resectable PDAC

Fig. 2. Multiplex IHC analysis of CD8⁺ and GZMB⁺CD8⁺ T cells and their correlation with OS in PDACs with GVAX/a–PD-1/SBRT treatment.

An increase in T_H1 and T_H17 density correlates with improved OS in patients treated with the combination GVAX/a–PD-1/SBRT

Fig. 3. Correlative analysis of OS with the changes of the densities of CD4⁺ T cell subtypes and myeloid cell subtypes between matched pre- and posttreatment tumor areas in PDACs with GVAX/a–PD-1/SBRT treatment.

SBRT enhances the GVAX/a–PD-1–mediated increase of GZMB⁺CD8⁺ T_effs

Adding SBRT to GVAX/a–PD-1 treatment shortens the distances between PD-1⁺CD8⁺ T cells and tumor cells and distances between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells

Fig. 6. Spatial distance measurement between CD8⁺ T cells, CSF-1R⁺ myeloid cells, and tumor cells in PDACs following different types of neoadjuvant therapies.

Bringing the PD-1⁺CD8⁺ T cells closer to tumor cells and closer to PD-L1⁺ myeloid cells by adding SBRT to the GVAX/a–PD-1 treatment correlated with better OS

Fig. 7. Correlative analysis of OS with the average distances between CD8⁺ T cells and tumor cells and between PD-1⁺CD8⁺ T cells and PD-L1⁺CSF-1R⁺ myeloid cells in PDACs following different types of neoadjuvant therapies.