Combined Vaccine-Immune-Checkpoint Inhibition Constitutes a Promising Strategy for Treatment of dMMR Tumors

doi:10.21203/rs.3.rs-92168/v1

Download PDF

Research

Combined Vaccine-Immune-Checkpoint Inhibition Constitutes a Promising Strategy for Treatment of dMMR Tumors

https://doi.org/10.21203/rs.3.rs-92168/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 18 Apr, 2021

Read the published version in Cancer Immunology, Immunotherapy →

Version 1

posted

You are reading this latest preprint version

Background: Mlh1-knock-out-driven mismatch-repair-deficient (dMMR) tumors can be targeted immunologically. By applying therapeutic tumor vaccination, tumor growth is delayed but escape mechanisms evolve, characterized by the upregulation of immune-checkpoint molecules (LAG-3, PD-L1). To counteract immune escape, we investigated the therapeutic activity of a combined tumor vaccine-immune checkpoint inhibitor therapy using α-PD-L1 monoclonal antibody clone 6E11.

Design: In this trial, Mlh1-knock-out mice with established gastrointestinal tumors received single or thrice injections of α-PD-L1 monoclonal antibody clone 6E11 (2.5 mg/kg bw, q2w, i.v.) either alone or in combination with the vaccine. Longitudinal flow cytometry and PET/CT imaging studies were followed by ex vivo functional immunological and gene expression assays.

Results: 6E11 monotherapy slightly increased median overall survival (mOS: 6.0 weeks vs. control 4.0 weeks). Increasing the number of injections (n=3) improved therapy outcome (mOS: 9.2 weeks) and was significantly boosted by combining 6E11 with the vaccine (mOS: 19.4 weeks vs. 10.2 weeks vaccine monotherapy). Accompanying PET/CT imaging confirmed treatment-induced tumor growth control, with the strongest inhibition in the combination group. Three mice (30%) achieved a complete remission and showed long-term survival. Decreased levels of circulating splenic and intratumoral myeloid-derived suppressor cells (MDSC) and decreased numbers of immune-checkpoint-expressing splenic T cells (LAG-3, CTLA-4) accompanied therapeutic effects. Gene expression and protein analysis of residual tumors revealed downregulation of PI3K/Akt/Wnt and TGF-signaling, leading to T cell infiltration and reduced numbers of macrophages, neutrophils, and MDSC.

Conclusions: By successful uncoupling of the PD-1/PD-L1 axis, we provide further evidence for the safe and successful application of immunotherapies to combat dMMR-driven malignancies that warrants further investigation.

Oncology

Cancer Biology

α-PD-L1

MMR deficiency

in vivo imaging

tumor microenvironment

long-term survival

Immunotherapy with immune-checkpoint inhibitors (ICI) has become a mainstay of treatment for a range of solid cancers, including melanoma, bladder cancer, non-small cell lung cancer and Hodgkin’s lymphoma [1–5]. CTLA-4, PD-1, or PD-L1 are the so far most studied checkpoint molecules and ICI widely applied in the clinic to improve patients’ prognosis. This blockade reactivates exhausted T-cells, prevents T-cell inhibition, and promotes effector T-cell proliferation to stimulate T-cell-mediated tumor cell killing [6–8]. Atezolizumab, Avelumab, and Durvalumab are FDA-approved as PD-L1 blocking antibodies. Monotherapy results in antitumor immune responses yet has a limited long-term therapeutic efficacy in most cases.

Colorectal cancer (CRC) is the third most common cancer worldwide. There is a wide range in global patterns with rising incidences in low- and middle-income countries but stable or even declining numbers in high-income countries [9]. For the latter, this is because of implementing screening strategies and changing patterns in risk factors, such as smoking. However, the positive trend in the older population is repealed by rising incidence rates in the younger (< 50 years) [9–11]. In approximately 3% of patients, CRC occur as part of a defined hereditary tumor syndrome such as Lynch Syndrome [9, 12–14]. A hallmark of tumors arising in Lynch Syndrome is mismatch repair deficiency (dMMR) and an accordingly ultramutated tumor phenotype (= TMB high). Due to the resulting high abundance of frameshifted neo-epitopes on the tumor cells’ surface, dMMR defines the molecular subtype with tremendous potential for immunological targeting [15–17]. Indeed, in 2017, the FDA approved α-PD-1 ICI Pembrolizumab and Nivolumab for treatment of dMMR cancers agnostic of cancer site [18], which was extended lately for the first-line treatment of patients with unresectable or metastatic dMMR CRC. Pre-existing Th type1 immune responses and high numbers of tumor-infiltrating CD8⁺ T-cell clones (= IFNγ signature) constitute positive predictive biomarkers [19]. However, roughly 25% of patients show intrinsic resistance and in most cases initially responding patients gradually develop resistance, highlighting the necessity of improving treatment options [20–23]. As for PD-L1, limited preclinical data exist. PD-L1 expression on tumor-infiltrating lymphocytes is a potential predictor for patients’ response to α-PD-1 therapy [24]. A recent phase II study in patients with dMMR metastatic or unresectable CRC revealed antitumor activity of Avelumab monotherapy [25]. Additional clinical trials are ongoing with different combinations being employed. One of them is based on tumor lysates or specific neoantigen-derived peptides. The former act as “global” vaccines and induce objective responses in some patients. To refine combination approaches preclinically, we employed the Mlh1 knock-out mouse model for dMMR-related diseases. Preceding vaccination approaches yielded prolonged overall survival in the therapeutic and prophylactic setting [26, 27]. Residual tumor cells showed an upregulation of immune-checkpoint molecules as part of acquired resistance. To counteract vaccination-induced immune escape and improve overall survival, we here applied a murine α-PD-L1 antibody (clone: 6E11) in combination with repeated vaccination.

Cell culture & vaccine preparation

Cells were cultured in DMEM medium, supplemented with 10% FCS (fetal calf serum), 6 mM Glutamine, and antibiotics (all from Biochrom, Berlin, Germany). The tumor lysate was prepared from a A7450 tumor allograft as described [28].

Mlh1^-/- mouse model and in vivo treatment protocol

Ethical statement

The German local authority approved all animal experiments: Landesamt für Landwirtschaft, Lebensmittelsicherheit und Fischerei Mecklenburg‐Vorpommern (7221.3‐1‐026/17; -026/17-3), under the German animal protection law and the EU Guideline 2010/63/EU. Mice were bred in the animal facility of the University Medical Center in Rostock under specific pathogen‐free conditions. Mlh1 genotyping was done according to [21]. During their whole life-time, all animals got enrichment in the form of mouse-igloos (ANT Tierhaltungsbedarf, Buxtehude, Germany), nesting material (shredded tissue paper, Verbandmittel GmbH, Frankenberg, Deutschland), paper roles (75 × 38 mm, H 0528–151, ssniff‐Spezialdiäten GmbH), and wooden sticks (40 × 16 × 10 mm, Abedd, Vienna, Austria). During the experiment, mice were kept in type III cages (Zoonlab GmbH, Castrop‐Rauxel, Germany) at 12‐h dark:light cycle, the temperature of 21 ± 2 °C, and relative humidity of 60 ± 20% with food (pellets, 10 mm, ssniff‐Spezialdiäten GmbH, Soest, Germany) and tap water ad libitum.

Experimental protocol

Mice with PET/CT proven gastrointestinal tumors (GIT) were conducted to therapy using four weekly tumor lysate boosts. Vaccination was sustained (10 mg/kg bw, biweekly, n = 10 mice) until tumors progressed, but for a maximum of 12 times. Treatment with α-PD-L1 (clone 6E11, kindly provided by Genentech, a subsidiary of Roche, South San Francisco, USA, dissolved in PBS) given at 2.5 mg/kg bw intravenously was done once (n = 4 mice) or thrice (n = 10 mice) every second week (q2wx3). Mice receiving the combination of α-PD-L1 were given vaccine first, followed by α-PD-L1 injection. Here again, combinations included single or triple α-PD-L1 applications (n = 10 mice/group; q2wx1 and q2wx3). Control mice were left untreated (n = 10 mice). Reduction of suffering was guaranteed by providing daily prepared soaked pellets, twice-daily monitoring of the health status using a score sheet and by applying humane endpoints (weight loss > 15%, pain/distress, changes in social behavior). All mice were sacrificed before they became moribund to prevent pain and distress. At this time, blood samples, spleens, lymph nodes, and GIT were removed for further analyses.

PET/CT imaging

PET/CT imaging scans were performed on a small animal PET/CT scanner (Inveon PET/CT, Siemens Medical Solutions, Knoxville, TN, USA) according to a standard protocol as described before [29, 30].

Immune phenotyping

Blood samples were taken routinely from the retrobulbar venous plexus. Single cell suspensions of spleens and GIT were obtained upon passing them through a cell strainer (100 µm). Samples (2 × 10⁵/Well) were stained with a panel of conjugated monoclonal antibodies (mAb, 1 µg each) followed by lysis of erythrocytes (155 mM NH₄Cl (MERCK Millipore, Darmstadt, Germany), 10 mM KHCO₃ (MERCK Millipore), and 0.1 mM EDTA (Applichem, Darmstadt, Germany). Negative controls consisted of lymphocytes stained with the appropriate isotypes (Biolegend, San Diego, USA). Cells were washed, resuspended in PBS and analyzed by flow cytometry on a Flow Cytometer (BD FACSVerse™, BD Pharmingen). Data analysis was performed using BD FACSuite software (BD Pharmingen).

Procartaplex Cytokine Assay

Cytokine levels in plasma samples were determined according to the manufacturer’s instructions of the Procartaplex™ multiplex immunoassay (Thermo Fisher Scientific, Schwerte, Germany). Measurement as well as cytokine quantification was performed on a Bioplex 2000 (Bio-Rad Laboratories GmbH, Munich, Germany) in combination with the Bio-Plex Manager Software. Absolute plasma cytokine and chemokine level are presented [ng/ml].

Fragment length analysis of cMS target genes

A panel of non-coding and coding MS marker was analyzed as described before [31]. MSI is defined by mono- and/or bialellic band shifts usually characterized by deletions (indicated with minus symbol + number).

Nanostring targeted gene expression profiling

The T cell–inflamed tumor microenvironment was analyzed by targeted gene expression profiling of tumor RNA from fresh frozen or Tissue-Tek® embedded treatment and control samples (n = 3 samples/group). Total RNA was isolated using the RNeasy Mini Kit according to the manufacturers’ instruction (Qiagen, Hilden, Germany). Total RNA concentrations were measured using the NanoDrop ND1000 (Thermo Fisher Scientific). Gene expression analysis was conducted on the NanoString nCounter gene expression platform (NanoString Technologies, Seattle, WA) applying the PanCancer IO 360™ Panel. This panel enables digital profiling of 770 genes that shape the tumor-immune interface and allows for characterization of pathways relevant in immune response and escape. Quality control, normalization and data analysis was done by applying the nSolver™ Analysis Software 4.0 including nCounter Advanced Analysis (version 2.0.115). Data are presented as Heatmap and log10 (p-value) as well as log2 fold change.

Immunofluorescence

Cryostat sections of 4 µm were air-dried and fixed in cold pure methanol for 8 min. Unspecific binding sites were blocked in 2% BSA (Roth) for 2 h followed by incubation with 1 µg of the following FITC- and PE-labeled mAbs: CD4, CD8α, CD11b, Gr1 (Immunotools, Friesoythe, Germany), CD11c, CD104, LAG-3, PD-1, F4/80, and PD-L1 (Biolegend). Sections were washed, embedded in Roti Mount Flour Care DAPI (Roth, Karlsruhe)and target proteins visualized on a confocal laser scanning microscope (LSM780, Zeiss, Jena, Germany) using 20x objectives.

IFN-γ ELISpot

2.5 × 10³ targets/well (2 GIT cell lines: Mlh1^−/− A7450, Mlh1^−/− 328, 1 lymphoma cell line: Mlh1^−/− 1351, and YAC-1 cells) were seeded in IFNγ–specific mAb (Mabtech, 3321-3)–coated, 96-well microtiter plates. Peripheral blood leukocytes (5 × 10⁴/Well) or splenocytes (1 × 10⁴/well) from vaccinated and control mice were added in triplicates and co-cultured overnight. Bound antibody (Mabtech, 3321-6) was visualized by BCIP/NBT (KPL, Gaithersburg, Maryland, USA); spots were counted using an ELISpot reader. Presented are the numbers of IFNγ–secreting cells per 10,000 effector cells corrected for background levels counted in the absence of target cells, which was always ≤ 5 spots/well. Target cells without effector cells showed no background level.

Statistics

All values are expressed as mean ± SD. After proving the assumption of normality (Kolmogorov-Smirnov test), differences between vaccinated and control mice were determined using the unpaired Student’s t-test or one-way ANOVA (Bonferroni or Dunnett’s multiple comparison). Kaplan-Meier survival analysis was done by applying the log rank (Mantel Cox) test. Statistical analyses were performed using GraphPad Prism 5 (San Diego, CA). The criterion for significance was set to p < 0.05.

Combination of α-PD-L1 and vaccines significantly improves outcome of Mlh1^−/− mice

In a first cohort, the ICI α-PD-L1 was administered once because of its long half-value period. Effects on survival were only marginal (Fig. 1A) and may indicate that MLH1^−/−- associated tumors are refractory to ICI monotherapy. By combining single α-PD-L1 with the vaccine (combination 1), overall survival was not significantly improved. In a next step, α-PD-L1 was given three times, to see whether tumor are indeed ICI-refractory or single application was simply not sufficient to induce immune responses in this model. Indeed, thrice α-PD-L1 injection extended the life span of mice to a degree comparable to the vaccine monotherapy (Fig. 1B). In combination with the vaccine (= combination 2), α-PD-L1 antibody treatment could even quintuple the life of mice from four weeks (control) to ~ 20 weeks (p < 0.001 vs. control; p < 0.01 vs. 1x α-PD-L1; p < 0.05 vs. 3x α-PD-L1; p < 0.05 vs. vaccine). Hence, this combination partially abrogated intrinsic ICI resistance and we consequently continued to move on with the triple α-PD-L1 treatment (combination 2, hereafter referred as combination) for subsequent functional analyses.

Combinational therapy leads to tumor reduction

Longitudinal PET/CT measurement revealed significant tumor size reduction by either therapy (vaccine, α-PD-L1 and combination) compared to controls (Fig. 2A). Still, analysis of the tumor size within the treatment groups identified significant reduction over time only in the combination (Exemplary pictures are given in Fig. 2B Fig. 2B) finally resulting in partial or even complete remission. For the latter, this promising result was seen in three mice. Two of them remained alive until the experimental endpoint and one mouse had to be euthanized because of a progressive cutaneous benign lesion (week 23).

Peripheral immune activation by vaccine-based immunotherapy

To investigate the immunological changes during therapy, blood was taken from mice every four weeks and analyzed via flow cytometry (Fig. 3A). The vaccine treatment induced a temporary increase of CD3⁺/CD4⁺ T-helper cells at day 84 which was not seen in the other groups. The level of CD3⁺/CD8⁺ cytotoxic T-lymphocytes (CTL) remained constant over time, while the amount of NK cells increased continually in all three treatment groups. The CD11b⁺/Gr1⁺ MDSC were doubled during the therapy with every treatment. The effects on CD19⁺ B-lymphocytes were oppositional. B-lymphocytes increased in the combination and decreased during vaccine or α-PD-L1 treatment. CD83⁺ dendritic cells (DC) were mainly found in mice treated with the vaccine only or the combination, likely because of stimulating the humoral arm of the immune system.

To investigate changes in the cytokine levels that act as growth factors, we analyzed plasma levels from different time-points and at the end using a multiplex cytokine assay. TNF-α showed only marginal changes with the vaccine, a remarkable peak at day 56 in the α-PD-L1 treatment and a constant slight increase over time in the combination (Fig. 3B). This latter increase was also seen for the chemoattractant MIP1β. The IL10 level fluctuated in all three treatments. In contrast, the vaccine induced IL13, while it remained unaffected upon α-PD-L1 treatment and decreased in the combination group, indicative for minor relevance of Th2-cytokines in treatment response. The levels of RANTES and Eotaxin decreased with vaccination, but for α-PD-L1 and the combination it initially increased.

Changes in important sites for immune reactions: spleens and residual tumors

Additionally to the blood immune-monitoring, spleens and residual tumors were resected at the experimental endpoint and infiltrating cells analyzed by flow cytometry. In spleens, levels of CD3⁺/CD4⁺ T-helper cells and CD3⁺/CD8⁺ CTL did not change with the therapies. By contrast, levels of CD11b⁺/Gr1⁺ MDSC decreased and the CD83⁺ DC significantly increased in all treatment groups with a trend towards stronger effects in the combination (p < 0.05; and p < 0.01 vs. control; Fig. 4). The amount of PD-1⁺ cells increased slightly, PD-L1⁺ cells reduced. LAG-3⁺ cells decreased significantly, especially in the α-PD-L1 and the combination treatment (p < 0.05; and p < 0.01 vs. control). The same effect was seen for CTLA4⁺ cells (p < 0.05; p < 0.01; p < 0.001 vs. control and vaccine) and TIM-3⁺ cells.

Residual tumors harbored higher numbers of infiltrating CD3⁺/CD4⁺ T-helper cells and CD3⁺/CD8⁺ CTL, particularly for the combination group. CD11b⁺/Gr1⁺ MDSC increased with α-PD-L1 treatment and dropped in the combination (p < 0.05 vs. α-PD-L1). Levels of CD83⁺ DC were constant and similar to the control, while CD200R⁺ cells reduced in the combination (p < 0.05 vs. α-PD-L1). Looking at the frequency of immune-checkpoint molecules, there were additional differences. The abundance of PD-1⁺ cells remained unchanged in all groups, PD-L1⁺ cells increased with α-PD-L1 treatment and decreased in the combination (p < 0.01 vs. α-PD-L1). Infiltrating LAG-3⁺ cell numbers were high in the monotherapies, whereas CTLA4⁺ infiltration was mainly confined to groups of the vaccine (vaccine monotherapy and combination). Still, TIM-3⁺ cells decreased significantly upon combination (p < 0.01 vs. control).

Gene expression analysis identifies downregulation of PI3K/Akt/Wnt and TGF-signaling

To have a closer look on the complex interplay between the tumor, microenvironment, and immune response, the PanCancer IO 360 Gene Expression Panel was applied (Fig. 5).

The cluster, left to the heat map (Fig. 5A), schematically illustrates the relations of the three individuals of the four treatment groups dependent on their tumor infiltrating lymphocyte (TIL) levels. The amounts of different immune cells differs the most in the combination group compared to the other groups. Here, overall immune cell expression levels increased in two of three individuals. Treatment with α-PD-L1 also changed immune cell expression patterns compared to the control and vaccine treatment.

Total TIL levels (Fig. 5B, upper left) were elevated in the α-PD-L1 and the combination therapy, because of increasing amounts of cytotoxic T and B cells (upper and middle). Conspicuously, levels of exhausted CD8⁺ T cells and neutrophils exclusively decreased in the combination (middle and lower). Macrophages only decreased in the vaccine and combination group.

As can be taken from Fig. 5C, effects on common signaling pathways in the combination group correlate more with the vaccine therapy than with α-PD-L1 treatment. In detail, the myeloid compartment, TGF-beta and Wnt signaling pathways were downregulated in all treatment groups in comparison to the control. Additionally in the combination group, genes related to angiogenesis and PI3K/Akt pathway were downregulated. For the latter, LAMA1 and Comp were downregulated, whereas the phosphatase PTEN, a well-established tumor suppressor was upregulated (Fig. 5D). Genes for costimulatory signaling and lymphoid compartment were higher in tumors of mice receiving α-PD-L1 and the combination, whereas no changes were seen in the vaccine group. Furthermore, the JAK/STAT signaling was activated by the combination (Fig. 5D).

We summarize the detailed analysis of differential expression at the gene set level (supplementary Fig. 1). In the combination, genes belonging to the interferon signaling (such as H2-Q1/H2-Q2, Ifi203, and Vcam1) and cytotoxicity (Gzma and Tnfsf10) were upregulated, whereas all genes of the myeloid compartment genes were downregulated (including Ly6C1, Olr1, and Ccl20).

Combination therapy alters the tumor microenvironment

While above findings already showed changes between individual treatment groups, we additionally studied the tumor microenvironment by immunofluorescence (Fig. 6).

MDSC and F4/80⁺ tumor-associated macrophages (TAM) were detectable in control tumors, indicative a suppressive microenvironment. While MDSC were effectively eliminated upon therapy, irrespective of the applied treatment regimen, only the combination was able to impact on numbers of infiltrating TAM (Fig. 6A, B). Besides, CD11c⁺ cells increased in the combination. Numbers of CD4⁺ T-helper cells remained the same, while the CD8⁺ CTL increased. This resulted in a significant difference between the vaccine and the α-PD-L1 treatment (Fig. 6C, D). Vice versa, the amount of LAG-3⁺ T-cells significantly increased upon vaccine treatment but significantly decreased upon α-PD-L1 and the combination. These results were similar to the levels of PD-L1⁺ cells, which significantly decreased in these two groups (Fig. 6C, D). Though PD1 was highly abundant on tumor cells, we observed intratumoral differences, ranging from high to low PD1 expression within individual specimen (supplementary Fig. 2). This was, however, treatment-independent.

Treatment-induced molecular changes in cMS

Residual tumors of the different treatment groups were scanned for typical gene mutations at cMS (Fig. 7A, B). Depending on the treatment, tumors harbored different mutation frequencies in cMS. NKtr1 and Kcnma1 (left of the dotted line) had the lowest mutation rates for the control, whereas the treatments resulted in high mutation rates. Noticeably, GIT from all three treatment groups showed no mutation in Spen, Apc and Casc3 (highlighted with the grey box in the middle), while a mutation rate of 20–30% was evidently in control tumors. Residual tumors from the combination harbored the lowest mutation frequencies in Akt3, Clock, Il1F9, and Rfc3 (highlighted with the right grey box), especially compared with α-PD-L1 treatment (= 100% mutation frequency).

ELISpot analysis reveals increased immune activation upon combination treatment

To asses immune activation, IFN-γ secretion by T-lymphocytes was detected by ELISpot-assays after coincubation of splenocytes from treated and control mice with different cancer cell lines (A7450 T1 M1, 328, 1351, and Yac-1) (Fig. 7C). Splenocytes of mice from the combination group responded with significantly higher IFN-γ secretion than those treated with α-PD-L1. NK cell reactivity was excluded by lacking IFN-γ secretion against Yac-1 cells. Notably, IFN-γ secretion levels against 1351 MLH1^−/− lymphoma cells were the lowest irrespective of the treatment.

In this study, we describe a strategy to combine active tumor vaccination with an ICI in a clinically relevant dMMR mouse model [32]. DMMR is associated with high tumor mutational burden [33–35] and thus harbors a tentatively high likelihood of being susceptible to immunotherapy.

Using a murine α-PD-L1 antibody, monotherapy itself marginally improved outcome after single application. By increasing the number of injections, overall survival of Mlh1^−/− mice extended to a degree comparable to the vaccine monotherapy. The latter was prepared from a whole tumor lysate with proven antitumor activity from previous studies [26, 36]. Hence, both treatments prolonged mice’ survival suffering from highly aggressive Mlh1^−/−-driven GIT. Given the fact that Mlh1^−/− tumors, despite their high TMB, do not have a high IFNγ signature and are thus not targetable by ICI per se, the improved outcome after α-PD-L1 monotherapy is intriguing. It is therefore unlikely that mice’ outcome after targeting the PD-L1 axis is better if α-PD-L1 antibodies are applied more often or over a longer time. Rather targeting both MHC-I and II restricted tumor epitopes – with whole tumor lysates – in combination with PD-L1 blockade seems necessary to affect growth of poorly immunogenic and thus ICI refractory tumors, as recently shown for triple-negative breast cancer [37].

Dosing schedules and accurate timing of each combination partner remain undefined for combined vaccine and ICI strategies. Here, we performed alternating treatment starting with vaccine first. The rationale is based on our previous observations in which repetitive vaccine monotherapy provoked upregulation of immune-checkpoint-molecules on residual tumors [26]. To counteract therapy-induced upregulation, we here applied α-PD-L1 therapy during vaccination. This combined treatment yielded complete remission in 30% of mice and finally resulting in significantly improved overall survival. Although complete remission was not achieved in all mice, we would like to stress the point that tumor burden massively reduced in the combination likely because of inducing a T cell–inflamed tumor microenvironment. Other studies reported superior effects when checkpoint-inhibition was given after cessation of the vaccine [38]. Still, the significantly prolonged overall survival of MLH1^−/− mice achieved in this study argues in favor of concomitant application.

Most previous trials focused on α-PD-1 antibodies to increase antitumoral effects of vaccine-induced immunity [39–41]. Rare preclinical data exist on vaccine-α-PD-L1 combinations. A recent study described prolonged survival and increased tumor cell apoptosis in a hepatocellular carcinoma model treated with a combined DC vaccine and α-PD-L1 inhibitor [42]. Likewise, Sun et al. found enhanced tumor-antigen-specific immunity upon combined vaccine-PD-L1-blockade [43]. By reversing the immunosuppressive status of the micromilieu, PD-L1 is indeed a promising target. Here, we also identified a shaped tumor microenvironment accompanied by peripheral immune activation. By performing a detailed and longitudinal analysis, we found decreased numbers of circulating MDSC and T cell exhaustion markers after combined treatment. Accompanying in-depth gene expression analysis of residual tumors identified increased numbers of total TIL, mainly being cytotoxic T and B cells. Vice versa, levels of exhausted CD8⁺ T cells, tumor-associated macrophages, and neutrophils reduced in the combination group. Neutrophils are a group of tumor-associated cells which, in conjunction with MDSC, play a major role during cancer development and progression. Their specific location within the tumor (i.e. intra-, peritumoral or stromal) has prognostic relevance [44]. Abundance of tumor-associated neutrophils may even correlate with local TGFβ expression; in fact, TGFβ blocking improves outcome in preclinical cancer models [44]. In support of this, TGF-signaling was downregulated here upon combination. This vaccine-α-PD-L1 combination also altered additional common pathways with prognostic relevance, such as PI3K/Akt and Wnt-signaling as well as genes responsible for angiogenesis, matrix remodeling and metastasis. By contrast, genes belonging to the JAK/STAT signaling were upregulated. These cumulative data nicely explain the improved overall survival in mice treated with the combined vaccine-α-PD-L1 approach.

While most pronounced effects were in fact seen in the combination therapy and thus interpretable as synergistic, the monotherapy itself modulated the tumor microenvironment. Anti-PD-L1 treatment induced genes relevant for autophagy and downregulated NF-κB-signaling, which is in line with data from a recent trial on triple negative breast cancer cells, treated with Atezolizumab [45]. Upon vaccination, matrix remodeling/metastasis-related genes and genes of the Wnt-and TGF-signaling were downregulated.

Inter-individual differences throughout the treatment groups reflect the different overall survival times of mice. Here, short-term survivors had low TIL scores and vice versa. Teasing out what are the (patient-) individual baseline differences is the challenge for the next wave of pre- and clinical trials with immunotherapy to refine treatment on the long run.

Another interesting finding was the altered molecular profile in typical cMS marker upon treatment. One may speculate that treatment successfully eliminated single mutated clones, whereas other emerged under the immune-selective pressure. We identified somatic cMS mutations in NKtr1 and Kcnma1 in all treatment groups that were infrequent in control tumors. By contrast, somatic mutations in Spen, Apc, and Casc3 were no longer detectable. Notably, residual tumors from the combination therapy harbored the lowest mutation frequencies in Akt3, Clock, Il1F9, and Rfc3, especially compared with α-PD-L1 treatment (= 100% mutation frequency).

Amongst others, question remains why some tumors regressed while others finally progressed. Sustained tumor IFN signaling induces PD-L1 expression on tumor and immune cells and is considered a acquired resistance mechanism [46]. However, this only partially explains the different in vivo response. Reports from human dMMR CRC describe contradictory PD-L1 abundance on tumor-infiltrating lymphocytes or tumor cells [47, 48]. Its role to mediate immune escape is undebatable and results from a phase II study already confirmed antitumor activity of Avelumab with manageable toxicity in most, but clearly not all patients with previously treated dMMR mCRC and recurrent/persistent endometrial cancer [25, 49]. Heterogeneity among tumors, such as the varying TMB, different genomic variations (in cMS), and the activated Wnt/β-catenin signaling may provide an explanation for the heterogeneity in response. Understanding how the different cells react to our treatments holds promise for novel immune-modulating strategies and will hopefully help to guide the way for clinical vaccine-based immune-checkpoint regimens.

Tumor-lysate vaccination in combination with α-PD-L1 prolongs the lifetime of Mlh1 knock-out mice significantly and shows strong tumor growth inhibition via downregulation of PI3K/Akt/Wnt and TGF-signaling. This combination regimen results in decreased levels of myeloid-derived suppressor cells (MDSC), splenic and intratumoral checkpoint-expressing T cells (PD-L1, LAG-3 and CTLA-4) and therefore positively modulates the tumor microenvironment. Combined vaccine-immune checkpoint inhibition provides a safe approach especially for patients having a likelihood of being non-responsive towards immune checkpoint monotherapy.

cMS – coding microsatellite

CTL – cytotoxic T-lymphocytes

DC – dendritic cells

dMMR – mismatch repair deficiency

GIT – gastrointestinal tumor

ICI – immune-checkpoint inhibitors

MDSC – myeloid-derived suppressor cells

TMB – tumor mutational burden

Ethics approval and consent to participate

Consent for publication

Genentech, a subsidiary of Roche, South San Francisco, USA gave their consent for publication.

Availability of data and material

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Funding

This work was supported by a grant from the German research foundation to CM [DFG grant number MA5799/2-1 and MA5799/2-2].

Authors’ contributions

CM – conducted the study, performed the in vivo experiments, analyzed data and wrote the manuscript; IS and SK – performed ex vivo experiments, analyzed data and participated in writing; AK and RF – performed nanostring analysis; LH and CJ – critically revised the manuscript

Acknowledgements

We gratefully thank Genentech, a subsidiary of Roche, South San Francisco, USA for providing the clone 6E11 for in vivo experiments. We additionally thank Mrs. Ilona Klamfuss for breeding mice, Brigitte Vollmar and Bernd Krause for their continuous support in their efforts of chairing the Core Facility of Multimodal Small Animal Imaging. We also gratefully acknowledge the excellent technical assistance of Mrs. Joanna Förster. Furthermore, we thank Carina Bergner and Anja Gummesson, radiopharmacy team of the Department of Nuclear Medicine of the University Medical Centre Rostock, for providing ¹⁸F-FDG for the small animal PET/CT experiments.

Cousin S, Seneschal J, Italiano A. Toxicity profiles of immunotherapy. Pharmacol. Ther. Elsevier Inc.; 2018. p. 91–100.
Pistillo MP, Carosio R, Grillo F, Fontana V, Mastracci L, Morabito A, et al. Phenotypic characterization of tumor CTLA-4 expression in melanoma tissues and its possible role in clinical response to Ipilimumab. Clin Immunol. Academic Press Inc.; 2020;215.
Suzman DL, Agrawal S, Ning Y, Maher VE, Fernandes LL, Karuri S, et al. FDA Approval Summary: Atezolizumab or Pembrolizumab for the Treatment of Patients with Advanced Urothelial Carcinoma Ineligible for Cisplatin‐Containing Chemotherapy. Oncologist. Alphamed Press; 2019;24:563–9.
Crist M, Balar A. Atezolizumab in invasive and metastatic urothelial carcinoma. Expert Rev Clin Pharmacol. Taylor and Francis Ltd; 2017;10:1295–301.
Plimack ER, Bellmunt J, Gupta S, Berger R, Chow LQM, Juco J, et al. Safety and activity of pembrolizumab in patients with locally advanced or metastatic urothelial cancer (KEYNOTE-012): a non-randomised, open-label, phase 1b study. Lancet Oncol. Lancet Publishing Group; 2017;18:212–20.
Leng C, Li Y, Qin J, Ma J, Liu X, Cui Y, et al. Relationship between expression of PD-L1 and PD-L2 on esophageal squamous cell carcinoma and the antitumor effects of CD8+ T cells. Oncol Rep. Spandidos Publications; 2016;35:699–708.
Kalim M, Iqbal Khan MS, Zhan J. Programmed cell death ligand-1: A dynamic immune checkpoint in cancer therapy. Chem. Biol. Drug Des. Blackwell Publishing Ltd; 2020.
Liu J, Zhang S, Hu Y, Yang Z, Li J, Liu X, et al. Targeting PD-1 and Tim-3 pathways to reverse CD8 T-cell exhaustion and enhance ex vivo T-cell responses to autologous dendritic/tumor vaccines. J Immunother. Lippincott Williams and Wilkins; 2016;39:171–80.
Siegel RL, Torre LA, Soerjomataram I, Hayes RB, Bray F, Weber TK, et al. Global patterns and trends in colorectal cancer incidence in young adults. Gut. England; 2019;
Connell LC, Mota JM, Braghiroli MI, Hoff PM. The Rising Incidence of Younger Patients With Colorectal Cancer: Questions About Screening, Biology, and Treatment. Curr Treat Options Oncol. United States; 2017;18:23.
Glover M, Mansoor E, Panhwar M, Parasa S, Cooper GS. Epidemiology of Colorectal Cancer in Average Risk Adults 20--39 Years of Age: A Population-Based National Study. Dig Dis Sci [Internet]. 2019; Available from: https://doi.org/10.1007/s10620-019-05690-8
Evrard, Tachon, Randrian, Karayan-Tapon, Tougeron. Microsatellite Instability: Diagnosis, Heterogeneity, Discordance, and Clinical Impact in Colorectal Cancer. Cancers (Basel). 2019;11:1567.
Ait Ouakrim D, Dashti SG, Chau R, Buchanan DD, Clendenning M, Rosty C, et al. Aspirin, Ibuprofen, and the Risk of Colorectal Cancer in Lynch Syndrome. J Natl Cancer Inst. 2015;107:djv170.
Seth S, Ager A, Arends MJ, Frayling IM. Lynch syndrome – cancer pathways, heterogeneity and immune escape. J Pathol. 2018;246:129–33.
Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. United States; 2017;357:409–13.
Campbell BB, Light N, Fabrizio D, Zatzman M, Fuligni F, de Borja R, et al. Comprehensive Analysis of Hypermutation in Human Cancer. Cell. 2017;171:1042-1056.e10.
Luchini C, Bibeau F, Ligtenberg MJL, Singh N, Nottegar A, Bosse T, et al. ESMO recommendations on microsatellite instability testing for immunotherapy in cancer, and its relationship with PD-1/PD-L1 expression and tumour mutational burden: a systematic review-based approach. Ann Oncol Off J Eur Soc Med Oncol. England; 2019;30:1232–43.
Lemery S, Keegan P, Pazdur R. First FDA approval agnostic of cancer site — When a biomarker defines the indication. N. Engl. J. Med. Massachussetts Medical Society; 2017. p. 1409–12.
Loupakis F, Depetris I, Biason P, Intini R, Prete AA, Leone F, et al. Prediction of Benefit from Checkpoint Inhibitors in Mismatch Repair Deficient Metastatic Colorectal Cancer: Role of Tumor Infiltrating Lymphocytes. Oncologist. Alphamed Press; 2020;
Ariyan CE, Brady MS, Siegelbaum RH, Hu J, Bello DM, Rand J, et al. Robust antitumor responses result from local chemotherapy and CTLA-4 Blockade. Cancer Immunol Res. American Association for Cancer Research Inc.; 2018;6:189–200.
Bozorgmehr F, Hommertgen A, Krisam J, Lasitschka F, Kuon J, Maenz M, et al. Fostering efficacy of anti-PD-1-treatment: Nivolumab plus radiotherapy in advanced non-small cell lung cancer - study protocol of the FORCE trial. BMC Cancer. BioMed Central Ltd.; 2019;19.
Sullivan RJ, Hamid O, Gonzalez R, Infante JR, Patel MR, Hodi FS, et al. Atezolizumab plus cobimetinib and vemurafenib in BRAF-mutated melanoma patients. Nat. Med. Nature Publishing Group; 2019. p. 929–35.
Sahin IH, Akce M, Alese O, Shaib W, Lesinski GB, El-Rayes B, et al. Immune checkpoint inhibitors for the treatment of MSI-H/MMR-D colorectal cancer and a perspective on resistance mechanisms. Br. J. Cancer. Nature Publishing Group; 2019. p. 809–18.
Kong P, Wang J, Song Z, Liu S, He W, Jiang C, et al. Circulating lymphocytes, PD-L1 expression on tumor-infiltrating lymphocytes, and survival of colorectal cancer patients with different mismatch repair gene status. J Cancer. Ivyspring International Publisher; 2019;10:1745–54.
Kim JH, Kim SY, Baek JY, Cha YJ, Ahn JB, Kim HS, et al. A Phase II Study of Avelumab Monotherapy in Patients with Mismatch Repair-Deficient/Microsatellite Instability-High or POLE-Mutated Metastatic or Unresectable Colorectal Cancer. Cancer Res Treat. Korean Cancer Association; 2020;
Maletzki C, Gladbach YS, Hamed M, Fuellen G, Semmler M-L, Stenzel J, et al. Cellular vaccination of MLH1 ^−/− mice – an immunotherapeutic proof of concept study. Oncoimmunology. Taylor & Francis; 2018;7:e1408748.
Maletzki C, Wiegele L, Nassar I, Stenzel J, Junghanss C. Chemo-immunotherapy improves long-term survival in a preclinical model of MMR-D-related cancer. J Immunother Cancer. Journal for ImmunoTherapy of Cancer; 2019;7:1–14.
Maletzki C, Gladbach YS, Hamed M, Fuellen G, Semmler M-L, Stenzel J, et al. Cellular vaccination of MLH1^−/−mice–an immunotherapeutic proof of concept study. Oncoimmunology. 2018;7.
Maletzki C, Beyrich F, Hühns M, Klar E, Linnebacher M. The mutational profile and infiltration pattern of murine MLH1^-/- tumors: concurrences, disparities and cell line establishment for functional analysis. Oncotarget. 2016;7.
Rohde S, Lindner T, Polei S, Stenzel J, Borufka L, Achilles S, et al. Application of in vivo imaging techniques to monitor therapeutic efficiency of PLX4720 in an experimental model of microsatellite instable colorectal cancer. Oncotarget. 2017;8.
Maletzki C, Huehns M, Bauer I, Ripperger T, Mork MM, Vilar E, et al. Frameshift mutational target gene analysis identifies similarities and differences in constitutional mismatch repair-deficiency and Lynch syndrome. Mol Carcinog. 2017;56:1753–64.
Edelmann W, Yang K, Kuraguchi M, Heyer J, Lia M, Kneitz B, et al. Tumorigenesis in Mlh1 and Mlh1 / Apc1638N Mutant Mice. Cancer Res. 1999;59:1301–7.
Chae YK, Anker JF, Bais P, Namburi S. Mutations in DNA repair genes are associated with increased neo-antigen load and activated T cell infiltration in lung adenocarcinoma. 2018;9:7949–60.
Chang L, Chang M, Chang HM, Chang F. Microsatellite Instability: A Predictive Biomarker for Cancer Immunotherapy. Appl Immunohistochem Mol Morphol [Internet]. 2018;26. Available from: https://journals.lww.com/appliedimmunohist/Fulltext/2018/02000/Microsatellite_Instability___A_Predictive.11.aspx
Salem ME, Puccini A, Grothey A, Raghavan D, Goldberg RM, Xiu J, et al. Landscape of Tumor Mutation Load, Mismatch Repair Deficiency, and PD-L1 Expression in a Large Patient Cohort of Gastrointestinal Cancers. Mol Cancer Res. United States; 2018;16:805–12.
Maletzki C, Wiegele L, Nassar I, Stenzel J, Junghanss C. Chemo-immunotherapy improves long- term survival in a preclinical model of MMR-D-related cancer. Journal for ImmunoTherapy of Cancer; 2019;1–14.
Feola S, Capasso C, Fusciello M, Martins B, Tähtinen S, Medeot M, et al. Oncolytic vaccines increase the response to PD-L1 blockade in immunogenic and poorly immunogenic tumors. Oncoimmunology. Taylor and Francis Inc.; 2018;7.
Kodumudi KN, Ramamoorthi G, Snyder C, Basu A, Jia Y, Awshah S, et al. Sequential Anti-PD1 Therapy Following Dendritic Cell Vaccination Improves Survival in a HER2 Mammary Carcinoma Model and Identifies a Critical Role for CD4 T Cells in Mediating the Response. Front Immunol [Internet]. Frontiers Media S.A.; 2019 [cited 2020 Jun 29];10:1939. Available from: https://www.frontiersin.org/article/10.3389/fimmu.2019.01939/full
Yoo SY, Badrinath N, Jeong S-N, Woo HY, Heo J. Overcoming Tumor Resistance to Oncolyticvaccinia Virus with Anti-PD-1-Based Combination Therapy by Inducing Antitumor Immunity in the Tumor Microenvironment. Vaccines [Internet]. Multidisciplinary Digital Publishing Institute; 2020 [cited 2020 Jun 29];8:321. Available from: https://www.mdpi.com/2076-393X/8/2/321
Kadam P, Sharma S. Pd-1 immune checkpoint blockade promotes therapeutic cancer vaccine to eradicate lung cancer. Vaccines. MDPI AG; 2020;8:1–13.
Xu G, Feng D, Yao Y, Li P, Sun H, Yang H, et al. Listeria-based hepatocellular carcinoma vaccine facilitates anti-PD-1 therapy by regulating macrophage polarization. Oncogene. Springer Nature; 2020;39:1429–44.
Teng CF, Wang T, Wu TH, Lin JH, Shih FY, Shyu WC, et al. Combination therapy with dendritic cell vaccine and programmed death ligand 1 immune checkpoint inhibitor for hepatocellular carcinoma in an orthotopic mouse model. Ther Adv Med Oncol. SAGE Publications Inc.; 2020;12.
Sun NY, Chen YL, Wu WY, Lin HW, Chiang YC, Chang CF, et al. Blockade of PD-L1 enhances cancer immunotherapy by regulating dendritic cell maturation and macrophage polarization. Cancers (Basel). MDPI AG; 2019;11.
Shaul ME, Fridlender ZG. Tumour-associated neutrophils in patients with cancer. Nat. Rev. Clin. Oncol. Nature Publishing Group; 2019. p. 601–20.
Saleh R, Taha RZ, Nair VS, Alajez NM, Elkord E. PD-L1 blockade by atezolizumab downregulates signaling pathways associated with tumor growth, metastasis, and hypoxia in human triple negative breast cancer. Cancers (Basel). MDPI AG; 2019;11.
Benci JL, Xu B, Qiu Y, Wu TJ, Dada H, Twyman-Saint Victor C, et al. Tumor Interferon Signaling Regulates a Multigenic Resistance Program to Immune Checkpoint Blockade. Cell. Elsevier; 2016;167:1540-1554.e12.
Ho HL, Chou TY, Yang SH, Jiang JK, Chen WS, Chao Y, et al. PD-L1 is a double-edged sword in colorectal cancer: the prognostic value of PD-L1 depends on the cell type expressing PD-L1. J Cancer Res Clin Oncol. Springer Verlag; 2019;145:1785–94.
Korehisa S, Oki E, Iimori M, Nakaji Y, Shimokawa M, Saeki H, et al. Clinical significance of programmed cell death-ligand 1 expression and the immune microenvironment at the invasive front of colorectal cancers with high microsatellite instability. Int J Cancer. Wiley-Liss Inc.; 2018;142:822–32.
Konstantinopoulos PA, Luo W, Liu JF, Gulhan DC, Krasner C, Ishizuka JJ, et al. Phase II study of avelumab in patients with mismatch repair deficient and mismatch repair proficient recurrent/persistent endometrial cancer. J Clin Oncol. American Society of Clinical Oncology; 2019;37:2786–94.

Folie1.TIF
Genes are tested for differential expression in response to each selected covariate. For each gene, a single linear regression is fit using all selected covariates to predict expression. Volcano plot displaying each gene's log10 (p-value) and log2 fold change with the selected covariate. Highly statistically significant genes fall at the top of the plot above the horizontal lines, and highly differentially expressed genes fall to either side. Horizontal lines indicate various False Discovery Rate (FDR) thresholds or p-value thresholds if there is no adjustment to the p-values. The 40 most statistically significant genes are labeled in the plot.
Folie2.TIF
Immunofluorescence of residual tumor specimen. Representative images taken from a tumor treated with the combination. Note the intratumoral heterogeneity in terms of PD1 expression. Regulatory myeloid cells, defined as CD11b/PD-L1 double positive, were only occasionally detected.

Download PDF

Journal Publication

published 18 Apr, 2021

Read the published version in Cancer Immunology, Immunotherapy →

Version 1

posted

You are reading this latest preprint version

Combined Vaccine-Immune-Checkpoint Inhibition Constitutes a Promising Strategy for Treatment of dMMR Tumors

Status:

Journal Publication

Version 1

Abstract

Figures

Background

Methods

Cell culture & vaccine preparation

Results

Discussion

Conclusion

Abbreviations

Declarations

References

Supplementary Files

Status:

Journal Publication

Version 1