Article Text
PDF
Abstract
Background Microsatellite stable colorectal liver metastases (MSS CLM) maintain an immunosuppressive tumor microenvironment (TME). Historically, immune-based approaches have been ineffective. VB-111 (ofranergene obadenovec) is a genetically-modified adenoviral vector targeting the TME; its unique dual mechanism induces an immune response and disrupts neovascularization. Checkpoint inhibition may synergize the immune response induced by viral-mediated anti-angiogenic gene therapy. We aimed to examine the safety and antitumor activity of VB-111 and nivolumab in patients with refractory MSS CLM and to characterize immunological treatment-response.
Methods This was a phase II study of adult patients with histologically-confirmed MSS CLM who progressed on prior therapy. A priming dose of VB-111 1×1013 viral particles was given intravenously 2 weeks prior to starting biweekly nivolumab 240 mg and continued every 6 weeks. The combination continued until disease progression or unacceptable toxicity. The primary objectives were overall response rate and safety/tolerability. Secondary objectives included median overall survival and progression-free survival. Correlative studies were performed on paired tumor biopsies and blood.
Results Between August 2020 and December 2021, 14 patients were enrolled with median age 50.5 years (40–75), and 14% were women. Median follow-up was 5.5 months. Of the 10 evaluable patients, the combination of VB-111 and nivolumab failed to demonstrate radiographic responses; at best, 2 patients had stable disease. Median overall survival was 5.5 months (95% CI: 2.3 to 10.8), and median progression-free survival was 1.8 months (95% CI: 1.4 to 1.9). The most common grade 3–4 treatment-related adverse events were fever/chills, influenza-like symptoms, and lymphopenia. No treatment-related deaths were reported. Qualitative analysis of immunohistochemical staining of paired tumor biopsies did not demonstrate significant immune infiltration after treatment, except for one patient who had exceptional survival (26.0 months). Immune analysis of peripheral blood mononuclear cells showed an increase of PD-1highKi67highCD8+ T cells and HLA-DRhigh T cells after VB-111 priming dose. Plasma cytokines interleukin-10 and tumor necrosis factor-α increased after treatment with both drugs.
Conclusion In patients with MSS CLM, VB-111 and nivolumab did not improve overall response rate or survival but were tolerated with minimal toxicities. While challenging to distinguish between antiviral or antitumor, correlative studies demonstrated an immune response with activation and proliferation of CD8+ T cells systemically that was poorly sustained.
Trial registration number NCT04166383.
- Immunotherapy
- Tumor Microenvironment
- Gastrointestinal Neoplasms
- Liver Neoplasms
- Oncolytic Virotherapy
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
VB-111 (ofranergene obadenovec) is a unique viral-mediated anti-angiogenic gene therapy that can disrupt tumor-specific vascularization and induce a tumor-directed immune response. It has been successfully used in early-phase clinical trials with glioblastoma, thyroid, and ovarian cancers; however, the only phase III trial conducted thus far did not meet its primary endpoint in recurrent platinum-refractory ovarian cancer. Only one clinical trial has assessed tumor immunogenicity of VB-111 and was limited to biopsies of only two patients.
WHAT THIS STUDY ADDS
This study is the first clinical trial to evaluate the efficacy of VB-111 combined with an immune checkpoint inhibitor in treatment-refractory microsatellite stable colorectal liver metastases. This study provides unique insight into the immunologic mechanisms of this novel immunotherapy combination through immune profiling of paired peripheral blood mononuclear cells (n=11) and tumor biopsies (n=6), which showed activation and proliferation of CD8+ T cells after VB-111.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
With its novel insight into the immune mechanisms induced by this viral-mediated gene therapy, this study validates the importance of on-treatment tumor biopsies and paired peripheral blood samples. It also provides rationale for future trial design in colorectal liver metastases. A single priming dose may not be the optimal strategy for viral-mediated therapies; rather, multiple sequential doses of viral vector, administration of dual checkpoint inhibition, or combinatorial approaches with chemokines/cytokines may be required to elicit a stronger immune response to result in tumor killing.
Background
Immune-based approaches in colorectal cancer have been largely unsuccessful with the notable exception of immune checkpoint inhibition in microsatellite instability-high disease.1–4 The reasons for this are unclear, but no doubt related to the poor immunogenicity evident in advanced gastrointestinal malignancies such as lack of infiltrating lymphocytes in advancing T-stages and an immunosuppressive tumor microenvironment (TME).3 5 While advances in tumor immuno-oncology are promising, most patients will not benefit from these strategies. Mechanisms for tumor-mediated immune evasion and primary adaptive resistance to immunotherapeutics remain elusive.
Even more complex, liver-dominant metastatic colorectal cancer (mCRC), as opposed to peritoneal-limited or lung-limited metastases, appear to be less responsive to immunotherapy compared with these other metastatic disease sites. For example, recent progress in a first-in-human trial (NCT0386272) with botensilimab (a novel innate/adaptive immune activator) and balstilimab (an Fc-enhanced next-generation anti-PD-1, or programmed cell death 1, antibody) led to a fast-track designation by the US Food and Drug Administration for this novel combination but limited to mCRC without liver involvement due to substantial survival differences based on the presence of active liver metastases: 12-month overall survival of 40% (95% CI: 17% to 62%) in active liver metastases versus 81% (95% CI: 66% to 90%) without active liver metastases.6 Therefore, currently, the standard of care for microsatellite stable colorectal liver metastases (MSS CLM) remains chemotherapy and may be combined with an anti-angiogenetic therapy, such as bevacizumab, but the overall prognosis for this subset of patients remains quite limited with 5-year overall survival of 15.6%.7
VB-111 (ofranergene obadenovec) is a viral-mediated anti-angiogenic gene therapy with a unique dual mechanism of action: vascular disruption and induction of a tumor-directed immune response (online supplemental figure 1). This viral gene therapy is an E1 deleted, non-replicating adenovirus-5 vector with a modified murine pre-proendothelin promoter (PPE-1–3 x) and a human Fas-chimera transgene (Fas-c) given intravenously.8 9 The murine PPE-1–3 x promoter requires endothelin-1 to express its pro-apoptotic transgene and specifically restricts expression of Fas-c to activated angiogenic endothelial cells, as compared with quiescent endothelial cells,10–12 resulting in its promising properties as an anti-angiogenic disrupter. The non-replicating viral vector is internalized by endothelial cells of blood vessels, and the Fas-chimera transgene is expressed, constructing a transmembrane chimeric death receptor with an extracellular portion of tumor necrosis factor (TNF) receptor 1 (p55) and an intracellular and transmembrane portion of Fas (CD95).13 Found abundantly within the TME, TNF-α activates the apoptotic pathway, resulting in tumor-specific angiogenic endothelial cell death. Subsequently, direct tumor cell death occurs as neovascularization is disrupted, generating neoantigens. A tumor-directed immune response then results as antigen presenting cells ingest neoantigens and prime tumor-specific effector T cells, inducing cellular and antibody-mediated immune responses.14 Like oncolytic viruses, the adenovirus vector may also trigger a localized inflammatory reaction in response to the virus itself, recruiting proinflammatory cytokines into the tumor milieu.
Supplemental material
Preclinically and clinically, VB-111 has shown highly specific anti-angiogenic activity and immunotherapeutic effects in a variety of tumors.9 13 15–21 In several rodent models, prolonged survival was observed in animals treated with one does of VB-111 (ie, nude rats with intracranial U87 glioma xenografts) or inhibition of tumor growth with central tumor necrosis (ie, nude mice with multiple thyroid cancer-derived cell lines and C57BL/6J mice with subcutaneous B16 melanoma or injected with Lewis lung carcinoma metastases).13 15 16 Immunological studies demonstrated an increase in tumor infiltrating CD8+ T cells after treatment with VB-111 in a Lewis lung carcinoma mouse model,17 and the anti-angiogenic properties of VB-111 were shown through significant reduction of CD31+ expression, an endothelial cell marker of angiogenesis, after treatment with VB-111 in mouse models with thyroid cancer.16 In early-phase clinical trials, promising results have been reported in glioblastoma, ovarian, and thyroid tumors.9 18–21 For example, in glioblastoma, primed VB-111 (given prior to) or unprimed VB-111 (given concurrently) with bevacizumab was well-tolerated, prolonged survival (median overall survival in the primed vs unprimed groups: 414 vs 141.5 days, HR 0.24, p=0.0056), and correlated to expansive areas of necrosis on imaging in the phase I/II trial18; however, the phase III GLOBE study was unable to validate these findings, showing worse survival outcomes with increased toxicity in the combination arm with VB-111 and bevacizumab in comparison to the control arm with bevacizumab monotherapy. Median overall survival was reported as 6.8 versus 7.9 months in the combination versus control arm (HR, 1.20; 95% CI: 0.91 to 1.59, p=0.19), and higher grade 3–5 adverse events were seen with the combination (67% vs 40%).19 The differences between the early phase and phase III trials were attributed to the change in treatment schedule where the priming dose of VB-111 was omitted in the larger randomized control trial, suggesting the importance of this priming dose in combination therapies. With the well-established role of anti-angiogenesis therapy and difficult-to-overcome immunosuppressive TME, MSS CLM is an interesting model for this novel immune inducing vascular disruptive agent. However, the antitumor immunity induced by VB-111 through release of neoantigens from hypoxic, dying tumors may not be enough to produce a clinically significant antitumor response in such an immunosuppressive TME. Therefore, the combination of VB-111 with an immune checkpoint inhibitor (ICI) may enhance responsiveness by reversing T-cell anergy and potentiating T-cell activation.
This phase II study aimed to investigate the antitumor activity, safety, and tolerability of VB-111 in combination with nivolumab (an anti-PD-1 antibody) in previously treated patients with MSS CLM.
Methods
Study design
This was a single center, open label, single arm phase II clinical study of VB-111 in combination with nivolumab in patients with refractory, metastatic MSS CRC and was designed to evaluate antitumor activity, safety, and feasibility. This study was approved by the ethics committee at the National Cancer Institute of the National Institutes of Health. The trial was designed and conducted in accordance with the Declaration of Helsinki and the Ethical Guidelines for Clinical Research (NIH, Bethesda, Maryland, USA). All participants provided written informed consent. This study was registered at ClinicalTrials.gov, and the protocol is available in the supplemental material.
Supplemental material
Patient selection
Eligible patients included adults aged ≥18 years with histologically confirmed MSS metastatic colorectal adenocarcinoma and at least one measurable tumor within the liver. All patients must have progressed on at least two lines of systemic therapy, and patients were permitted to have previously received anti-angiogenic therapies such as bevacizumab or ICIs. Patients were excluded if they had received any prior cancer-targeting vaccinations. Adequate organ function and performance status were required. Full eligibility criteria are defined in the supplemental material (online supplemental figure 2).
Procedures
A prior phase I study established the recommended phase II dose of 1×1013 viral particles (VPs), which was used in this study design, and only adjusted for low weight (0.7×1013 VPs if weight <50 kg). Treatment was given in 2-week cycles; patients were given VB-111 1×1013 VPs intravenously every 6 weeks and received concomitant pre-dose acetaminophen to mitigate possible fever. Nivolumab was administered every 2 weeks at a flat dose of 240 mg intravenous infusion, starting 2 weeks after initial VB-111 priming dose. Treatment continued until disease progression or unacceptable toxicity; treatment beyond progression was permitted per Principal Investigator discretion. Study schema is provided in online supplemental figure 3.
Outcomes
The primary objectives were overall response rate, safety, and feasibility of VB-111 in combination with nivolumab in MSS CLM. The secondary objectives were 6-month progression-free survival, median progression-free survival, and median overall survival. The exploratory objectives were to characterize immunological responses with the combination as assessed by correlative studies of paired tumor biopsies and blood samples.
Radiographic response evaluation
Disease status was evaluated every 8 weeks with contrast-enhanced CT scans (or magnetic-resolution imaging if clinically indicated), and radiographic response and disease progression were assessed by investigators in accordance with the Response Evaluation Criteria in Solid Tumor criteria V.1.1. Patients were considered evaluable for response evaluation if they had received at least one dose of both study drugs.
Safety and tolerability
All patients who received at least one dose of either study drug were evaluated for safety. Physical examination, liver function test, carcinoembryonic antigen (CEA) level, and toxicities were assessed every week during the first 8 weeks of study and every 2 weeks thereafter. Adverse events related to study drugs were reported descriptively by type and graded according to the National Cancer Institute Common Terminology Criteria for adverse events (V.5.0).
Immune correlates
Blood samples were obtained at baseline and every 2 weeks while on treatment and used for high-dimensional flow cytometry and cytokine array analyses. Multicolor flow cytometry was performed on peripheral blood mononuclear cells (PBMCs) to study different immune cell subsets using the following antibodies: CD45(HI30), CD19 (HIB19), CD11c (B-ly6), CD27 (O323), PD-L1 (MIH1), HLA-DR (L243), CD16 (3G8), CD11b (ICRF44), CD38 (S17015A), TCR γ/δ (B1), CD3 (UCHT1), CD14 (M5E2), CD33 (WM53), CD56 (QA17A16), FoxP3 (259D), CD8 (SK1), TIM3 (F38-2E2), CD45RA (HI100), CD4 (RPA-T4), CCR7 (150503), ICOS (DX29), PD-1 (EH12.1), CD25 (M-A251), CTLA4 (L3D10), Ki67 (Ki67), CD127 (A019D). Data for all samples were collected on a Cytek Aurora spectral flow cytometer (Cytek) and analyzed using FlowJo software (FlowJo). Gating strategies were performed as previously described.22 The populations from either live, CD45+leukocytes (for immune profiling) or live, CD3+lymphocytes (for T-cell profiling) were then down sampled and exported as .csv files. After import of these .csv files, graph-based clustering and UMAPs (or Uniform Manifold Approximation and Projection) were generated using the Partek Flow software pipeline (Build V.10.0.22.0703, Partek, St. Louis, Missouri, USA). Plasma samples were used to analyze circulating cytokines. Eight targets were chosen, and plasma was analyzed using a Human Cytokine/Chemokine/Growth Factor ProcartaPlex Multiplex Immunoassay Panel A using Luminex xMAP technology.
Paired intrahepatic tumor biopsies were obtained within 28 days of treatment initiation and while on treatment, either after priming dose of VB-111 or after receiving both drugs (VB-111 priming dose and two infusions of nivolumab). H&E stains were performed on all paired research biopsies, and expression of CD3, CD8, and PD-L1 (or programmed cell death-ligand 1) was evaluated by immunohistochemical staining and analyzed qualitatively by the Department of Pathology of the National Cancer Institute.
Statistical considerations
The study was designed as a Simon optimal two-stage phase II trial23 to rule out an unacceptably low response (complete response and partial response, CR+PR) rate of 5%, in favor of an improved response rate of 25%. The null hypothesis was established based on historical control data of therapies in the third line setting for mCRC.24 25 With a type I error rate of 10% (α=0.10) and 90% power (type II error, or β=0.10), the trial was to enroll 9 patients in the first stage, and if 1 or more of the first 9 had a PR or CR as their best overall response, the trial would continue enrollment until 24 evaluable patients had been treated. At the conclusion of the trial, 3 or more patients with PR or CR out of 24 would be considered of interest for future evaluation. Under the null hypothesis, the probability of early termination after the first stage of nine evaluable patients was 63%.
Efficacy analyses were based on all patients who received at least one dose of both study drugs; safety analyses included all patients who received at least one dose of either study drug. The Kaplan-Meier method was used to describe time-to-event outcomes, including progression-free survival and overall survival, defined as time of enrollment until time of event (ie, progression-free survival measured as disease progression or death, which came earlier). Binomial proportions are reported with a 95% exact CI. Adverse events were summarized using descriptive statistics. Flow cytometry analyses were performed using FlowJo software, and non-parametric statistical tests were performed on biospecimen to detect treatment-related changes over time; all results are shown as means with SEM presented and p values of <0.05 were considered statistically significant. Statistical analyses were performed using GraphPad Prism V.9 for Windows (GraphPad, San Diego, California, USA) and SAS (V.9.4, Cary, North Carolina, USA).
Results
Between August 2020 and December 2021, 14 patients with CLM were enrolled on study. The median follow-up was 5.5 months (data cut-off of March 2023). Patient characteristics are summarized in table 1. The median age of the population was 50.5 years (range 40–75), 14% of patients were women, and 64% were non-Hispanic white. All patients had an ECOG (or Eastern Cooperative Oncology Group) performance status of 0, except for one patient who had an ECOG of 1. The most common primary site of disease was left colon (64%), and the most frequent sites of metastasis were liver and lung with the majority presenting as synchronous metastatic disease at diagnosis (71%). Baseline CEA level was 219 ng/mL (range 2.3–5,664). Most patients had received at least three lines of systemic chemotherapy in the
Patient characteristics
metastatic setting. All patients were confirmed to be MSS, and half of the patients had either a KRAS or BRAF mutation.
As shown in the study consort diagram (online supplemental figure 4), the safety analyses included 13 patients who received at least one study drug, and the efficacy analyses included 12 patients who received at least one dose of each study drug. One patient did not receive any investigational agent due to disease progression and was removed from study prior to treatment, and another patient was taken off treatment after receiving only the VB-111 priming dose due to clinical disease progression. The median doses of VB-111 given was 2 (range 0–2). The most common reason for treatment discontinuation was disease progression. All patients died from progressive disease.
Antitumor activity and efficacy
Ten patients were considered evaluable for radiographic response. Although the study design called for a pause after nine evaluable patients in the first stage, one patient had been initially excluded from evaluability due to clinical disease progression. In retrospect, early restaging scans were evaluated for tumor measurements, and thus, this patient was included in the analysis of overall response rate. Ultimately, the combination of VB-111 and nivolumab did not result in any radiographic responses. At best, two patients had stable disease, and eight patients had progressive disease.
As shown in figure 1, the median progression-free survival was 1.8 months (95% CI: 1.4 to 1.9 months), and all patients experienced progressive disease by 6 months. The median overall survival was 5.5 months (95% CI: 2.3 to 10.8 months) with three patients living longer than 10 months (26.0, 25.7, and 10.8 months). Tumor progression was the primary cause of death for all patients.
Kaplan-Meier survival analyses of progression-free survival (PFS) and overall survival (OS) in patients with microsatellite stable metastatic colorectal cancer treated with VB-111 and nivolumab. All patients who received at least one dose of both study drugs were included (n=12); one patient was excluded due to disease progression prior to nivolumab initiation, and one patient was taken off protocol due to malignant bowel perforation with abdominal sepsis prior to treatment. Time-to-event started from date of enrollment until disease progression or death, respectively. After median a follow-up of 5.5 months, median PFS was 1.8 months (95% CI: 1.4 to 1.9), and median OS was 5.5 months (95% CI: 2.3 to 10.8).
Safety and tolerability
Nearly all patients experienced at least one treatment-related adverse event (trAE). The combination of VB-111 and nivolumab did not cause any unexpected trAE and was generally well-tolerated. As summarized in table 2, the most common trAEs of all grades was nausea/vomiting (n=10), fatigue (n=9), lymphocyte count decreased (n=7), fevers/chills (n=6), and influenza-like symptoms (n=6). The most common
Summary of treatment-related adverse events
grade 3 and 4 trAE was lymphopenia, which may have been related to pretreatment disease status or nivolumab. No unexpected toxicities were observed. Treatment was discontinued for one patient due to suspected immunotherapy-related pancreatitis. No treatment-related deaths were observed.
Immune correlatives
Immune correlative analyses of serial peripheral blood samples and paired tumor biopsies prior to treatment and on-treatment were performed to characterize immune cells and treatment-related changes: flow cytometry of PBMCs, multiplex cytokine immunoassay of plasma, and immunohistochemical staining of tumor tissue.
As shown in figure 2 and online supplemental figure 5, high-plex multicolor spectral cytometry was performed to study different immune cell subsets and activation status. Unsupervised analysis of immunoprofiling revealed 11 distinct immune cell clusters (containing T cells, B cells, dendritic cells, granulocytes, natural killer or NK cells, and monocytes). Next, we performed a pairwise analysis of changes in the cell composition before and on treatment in all samples (baseline: C2D1 to assess VB-111 effect; on treatment: C4D1 to assess effect of VB-111 and nivolumab), and immune cells were quantified. The immune cell cluster with HLA-DRhigh T cells represented activated T cells expanded after treatment but decreased following treatment with nivolumab, although remained significantly higher than baseline levels.
Immune and T-cell profiling of peripheral blood mononuclear cells (PBMC) after treatment with VB-111 and nivolumab. PBMCs were collected at baseline, after receiving a priming dose of VB-111 alone (C2D1), and after receiving both the priming dose of VB-111 and 2 infusions of nivolumab (C4D1). A high-plex multicolor full-spectrum flow cytometry panel for immunophenotyping was performed unsupervised to study different immune cell subsets and activation status. The stacked bar graphs show the frequency of distinct immune (top left) and T-cell (bottom left) clusters and changes with treatment, stratified by patient. The best radiographic response is shown below each patient’s stacked bar graph, depicted in color. Of the immune cell clusters, the cluster of HLA-DRhigh T cells (top right) represented activated circulating T cells, which increased significantly after VB-111 infusion and trended down after the following 4 weeks; although decreased, the activated T cells remained significantly higher than baseline following treatment with nivolumab and VB-111. Of the T-cell clusters, the cluster of PD-1highKi67highCD8+ T cells (bottom right) represented proliferating T cells and increased significantly after VB-111 priming dose; however, these proliferating cells decreased by the next time point, 4 weeks later after addition of nivolumab. The T-cell cluster labeled “CD4” was an unidentified population of CD4+ T cells. All results are shown as mean±SEM. *p<0.05, **p<0.01, and ***p<0.001 were significantly different when compared with baseline blood samples using Friedman test. cDC, classical dendritic cells; EM, effector memory; EMRA, effector memory T cells re-expressing CD45RA; NK, natural killer; PD-1, programmed cell death 1; reg, regulatory; Treg, regulatory T cells.
better characterize potential effector cells of the therapy, a separate analysis for T cells was done. In-depth analysis of T cells revealed 10 distinct T-cell clusters. The T-cell cluster with PD-1highKi67highCD8+ T cells represented proliferating T cells expanded after treatment but not sustained after 6 weeks of treatment with nivolumab.
To validate findings from unsupervised analysis, a conventional supervised analysis was done. In line with findings from the unsupervised analysis, we observed an increase in frequency of PD-1highKi67highCD8+ T cells and HLA-DRhigh T cells after the first dose of VB-111, which decreased later. Effector memory CD8+ T cells increased after treatment with combination of VB-111 and nivolumab, but myeloid-derived suppressor cells (MDSC), regulatory T cells (Treg), and NK cells did not change significantly with treatment over time (online supplemental figure 6).
Plasma circulating cytokine analysis demonstrated an increase in interleukin (IL)-10 and TNF-α after treatment with both study drugs (figure 3). No difference with treatment was observed in the following cytokines: interferon (IFN)-γ, IL-1b, IL-6, IL-8, IL-12p40, or VEGFa (or vascular endothelial growth factor A).
Plasma circulating cytokines after treatment with VB-111 and nivolumab. Peripheral blood mononuclear cells were collected at baseline, after receiving a priming dose of VB-111 alone (C2D1), and after receiving both the priming dose of VB-111 and two infusions of nivolumab (C4D1). Plasma samples were used to analyze circulating cytokines. Eight targets were chosen, and plasma was analyzed using a Human Cytokine/Chemokine/Growth Factor ProcartaPlex Multiplex Immunoassay Panel A with Luminex xMAP technology. Treatment-related changes in cytokine levels are described above. IL-10 and TNF-α were significantly increased after treatment with VB-111 and nivolumab. All results are shown as mean±SEM. *p<0.05, **p<0.01, and ***p<0.001 were significantly different when compared with baseline blood samples using Friedman test. IFN, interferon; IL, interleukin; PD-L1, programmed cell death-ligand 1; TNF, tumor necrosis factor; VEGFa, vascular endothelial growth factor A.
Figure 4 summarizes the qualitative analysis of immunohistochemical staining of paired tumor biopsies, correlated to clinical outcomes and sorted by survival. Many of the baseline tumor biopsies did not demonstrate much infiltration of lymphocytes (TILs) or PD-L1 expression by tumor cells; CD3+T cells were mostly confined to the stroma (pts 5, 13, 7, 23, 11 and 6), whereas a few patients (3, 12, and 14) had sparse to few T cells infiltrating tumor prior to treatment with study drugs. Only one patient (pt 11) had any PD-L1 expression within their baseline tumor biopsy. Paired biopsies were available in sufficient quantity for analysis in six patients (only one after both study drugs, pt 14). Although sparse in general, TILs were more commonly observed after treatment and varied in composition of CD8+T cells.
Immune compositional changes of paired tumor biopsies as correlated to clinical outcomes. Ten patients underwent a baseline biopsy of an intrahepatic tumor within 28 days of treatment initiation and were later randomized to an on-treatment biopsy either1 after receiving only priming dose of VB-111 (ie, prior to cycle 2, day 1) or2 after receiving both priming dose of VB-111 and two infusions of nivolumab (ie, prior to cycle 4, day 1). Immunohistochemistry staining of tissue samples were analyzed qualitatively for CD3+ T cells, CD8+ T cells, and PD-L1 expression by the Department of Pathology at the National Cancer Institute. This table demonstrates treatment-mediated immunophenotypic changes observed within tumor as correlated to survival. ICI, immune checkpoint inhibitor (ie, nivolumab); OS, overall survival; PD-L1, programmed cell death-ligand 1; PFS, progression-free survival; pt, patient.
As shown in figure 5, the most “inflamed” tumor after treatment correlated to exceptional survival (26.0 months): patient 9 had left-sided colon cancer (MSS, KRAS and BRAF wild type) with metastases to lung and liver and had previously been treated with neoadjuvant FOLFOX prior to a lower anterior resection and followed by adjuvant FOLFOX, microwave ablation to liver metastasis, oral capecitabine, irinotecan/cetuximab/TARE-Y90, and right hemi-hepatectomy. This patient remained on treatment for 2.1 months, slightly longer than median duration of treatment (1.8 months) among the 12 patients treated with both study drugs. After treatment with VB-111, many TILs (mostly comprised of CD8+T cell) were observed, and focal weak expression of PD-L1 was seen within the tumor. Treatment-related changes were difficult to fully ascertain as the baseline biopsy contained non-viable, necrotic tumor tissue without any TILs.
Immunohistochemical analyses of paired tumor biopsies in a patient with exceptional survival. Tumor biopsies, obtained at baseline and after treatment with VB-111, were stained with H&E to characterize viable tumor and tumor-infiltrating lymphocytes. Antibodies were used to assess expression of T-cell markers (CD3, CD8) and PD-L1 expression of tumor cells at baseline (100× magnification) and while on-treatment (200× magnification): (A) H&E staining; (B) CD3+ T cells; (C) CD8+ T cells; (D) PD-L1+ expression. PD-L1, programmed cell death-ligand 1.
Discussion
In this small phase II study of patients with MSS CLM treated with VB-111 and nivolumab, this therapeutic combination did not improve overall response rates, nor meaningfully prolong survival but was generally tolerable with mild toxicities. Nevertheless, two important contributions resulted1: three heavily pretreated patients had unexpectedly impressive prolonged survival (26.0, 25.7, and 10.8 months), suggesting that this novel combination of immunotherapies may have impacted the immunosuppressive environment of CLM in certain patients, and2 the availability of paired tumor biopsies and peripheral blood allowed for immune profiling before and on-treatment, enriching our understanding of viral-mediated oncology treatments.
For patients with mCRC treated in the third-line setting, the median overall survival approaches 7–8 months.24 25 Therefore, the three patients with survival times greater than 10 months was quite impressive given their treatment-refractory advanced disease; the clinical features and treatment history of these three individual patients with exceptional survival are outlined in the supplemental materials (online supplemental figure 7). In particular, patient 9 had marked TILs after treatment with only a VB-111 priming dose, mostly comprised of CD8 T cells, and PD-L1 expression was seen focally within the tumor (figure 5). We were unable to fully assess baseline infiltration due to necrotic tissue comprising most of the sample, limiting our interpretation on contribution of VB-111 in inducing this observed response. Nonetheless, the on-treatment tissue biopsy of this patient exhibited the most significant number of lymphocytes of the six available tumor biopsies after treatment, and this patient went on to survive for 26.0 months after enrollment, only receiving 2.2 months of treatment with VB-111 and nivolumab before dying from progressive disease. Interestingly, while this patient shared many molecular features and baseline clinical characteristics as others in study, patient 9 was the only participant to undergo liver-directed therapies (microwave ablation and Y-90 embolization) prior to enrolling in study. It is possible that these liver-directed treatments modulated the TME in a manner that overcame immune regulation and sensitized nearby liver tumors to subsequent immune-based therapies, such as VB-111 or nivolumab.26–28
In addition to these observed exceptional survivals, this study provides unique insight into the immune mechanisms induced by this novel immunotherapy combination, adding to the sparse literature on the immunologic mechanisms of viral-mediated therapies. Specifically, there is only one other clinical trial that assessed tumor immunogenicity of VB-111 and was limited to biopsies of only two patients.20 Our study provides greater insight due to larger tumor sample size and suggests a trend towards prolonged survival with increased tumor-infiltration of lymphocytes in some patients. We were unable to correlate treatment-related changes with radiographic response like Arend et al, which may provide more direct evidence of VB-111 induced immunological changes. However, we were able to show VB-111 induced expansion of activated and proliferating CD8+ T cells in the flow cytometry analysis of PBMCs, validating the immunogenicity of this gene therapy. Cytokine analysis suggested a trend up in several proinflammatory cytokines (IFN-ɣ, IL-1b, and IL-12) over time; however, our ability to observe peak changes in cytokines that would be expected from an antiviral immune response within the 5–7 days after treatment with an adenoviral vector was limited by the small sample size and timing of phlebotomy draws every 2 weeks. While we did observe an increase in IL-10 and TNF-α, this likely reflected ongoing, longitudinal disease progression, rather than treatment effect. Notably, the observed immunogenicity in PBMCs was transient and not sustained after two infusions of nivolumab, suggesting that a single priming dose or an ICI alone may not be the optimal strategy for this viral-mediated gene therapy; rather, multiple sequential doses of viral vector clustered closer together may be required to elicit a stronger, prolonged immune response to result in clinically meaningful tumor killing.
It is well-demonstrated that the strength of the immune response, as reflected by IFN-ɣ production, is critical for tumor killing, not simply the presence of an immune response29 30; therefore, we decreased the frequency between VB-111 infusions that had been previously tested in the clinical trials with VB-111 in other solid tumors. The original dosing frequency of every 8 weeks was based on the pharmacokinetics observed in the phase I trial of VB-111 with advanced solid tumors that established a decline in adenoviral vector levels by at least 2-log fold or undetectable after 56 days post-infusion,9 and the studies reported in table 3 demonstrated safety and tolerability of this dosing schedule. A shorter interval of 28 days was even evaluated in a small cohort of the phase I/II study of VB-111 with bevacizumab in recurrent glioblastoma, showing toleration but lacked clinical significance likely due to the small sample size, negative impact of precluding a priming dose, and counteraction of bevacizumab to the mechanism of VB-111.18 Intending to strengthen the immune response, we tested an interval of 6 weeks between VB-111 infusions, which induced a transient immune response and was well-tolerated, but future considerations should reduce dosing intervals further, potentially to every 4 weeks and using IFN-ɣ production as a measure of activity. The limited clinical efficacy of VB-111 in colorectal cancer, as compared with other early phase clinical trials in glioblastoma, thyroid, and ovarian cancers (table 3), may be related to different immunosuppressive pressures within the liver. CLM (in contrast to lung-dominant or peritoneal-limited colorectal cancer) tend to poorly respond to ICIs and may be related to the TME composition of immune cells and chemokines/cytokines that mediate tumor migration, drug resistance, and immune evasion.31 32 An increase in tumor-associated macrophages, MDSC, and Treg have been reported in CLM; these specialized cells can maintain the immunosuppressive environment and inhibit adaptive antitumor immune responses, disrupting the immunologic balance by also
Published clinical trials of adult patients with solid tumors treated with VB-111 (ofranergene obadenovec) monotherapy or in combination with other agents
depleting protective NKs cells.33 34 Combinatorial approaches with adding or targeting cytokines/chemokines (ie, recombinant IFN-ɣ, anti-IL-10 therapies, IL-12, IL-2 or related cytokines) may overcome this challenge and restore equilibrium.
Another possibility for the differences observed in our clinical trial as compared with others with VB-111 is that liver metastases, unlike other sites of disease, may be impervious to anti-angiogenic therapies by using vessel co-option, or incorporating pre-exiting vessels from surrounding normal tissue, for tumor vascularization, avoiding immune surveillance.35 36 VB-111 requires activated angiogenic proliferating endothelial cells for the transgene to be expressed and disrupt neovascularization; hence, if CLM co-opt quiescent endothelial cells for their blood supply in this non-angiogenic mechanism, avoiding neovascularization, VB-111 and other anti-angiogenic therapies may be ineffective. We were limited in our ability to assess the role of anti-angiogenesis in the mechanism of action of VB-111, or even demonstrate viral expression of VB-111 within on-treatment tumor biopsies, since the transgene is specifically expressed in angiogenic endothelial cells, and the inclusion of blood vessels within biopsy samples is difficult to reliably reproduce with image-guided core needle biopsies. Therefore, our tumor-directed biopsies were insufficient to address these questions fully.
The success of viral-mediated oncology therapies depends on the balance between antiviral immune response and antitumor immunity. With viral vectors that use endemic viruses, such as adenovirus or herpes simplex virus, there is a valid concern that patients may have pre-existing neutralizing antibodies that could result in rapid viral clearance. This may be even more detrimental in formulations given intravenously where viruses can be cleared systemically by circulating neutralizing factors (antibodies and complement factors) prior to reaching the tumor. Neutralization of the adenoviral vector in VB-111 may have played a role in the observed transient immune response and lack of clinical response; however, in the phase I trial of VB-111 in advanced solid tumors, there was no correlation between baseline neutralizing antibody levels and post-therapy disease state or peak anti-Ad-5 IgG titers.9 Furthermore, pre-existing neutralizing antibodies (or development of these antibodies during treatment) have failed to correlate with generation of tumor-specific immune responses in other viral-based therapies.37 38
An important limitation to this study was the small sample size (n=14). However, despite the low number of biospecimen, this study provided insightful contributions into the molecular mechanisms of this novel combination. While an immune response was observed, it was challenging to distinguish between antiviral or antitumor responses, which will require further characterization in future studies with VB-111.
Conclusion
In summary, although negative, the novel anti-angiogenic gene therapy with immune-inducing properties, VB-111, in combination with nivolumab demonstrated activation and proliferation of CD8+ T-cells after treatment in patients with MSS CLM. Future considerations may involve shorter intervals between doses of VB-111, administration of dual ICIs that may synergistically enhance immune responses, and combinatorial approaches with chemokines/cytokines to overcome the immunosuppressive pressures within the liver.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
Ethics statements
Patient consent for publication
Ethics approval
This study involved human participants and was approved by NCI Institutional Review Board (20C0022). This study was approved by the ethics committee at NIH/NCI. The trial was designed and conducted in accordance with the Declaration of Helsinki and the Ethical Guidelines for Clinical Research (NIH, Bethesda, Maryland, USA). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The author(s) disclose receipt of the following financial support for the research, authorship, and/or publication of this article: TFG is supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute (ZIA BC 011343). The study was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research and a Cooperative Research and Development Agreement between NCI and VBL Therapeutics (CRADA 03295). The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the US Government.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @coffman_kelley, @Cecilimonge4
Contributors Study concept and design: TFG. Analysis and interpretation of data: KC-D’A, CM, CX, YM, DMH, TFG. Statistical analysis: SMS. Administrative, technical, or material support: DK, BJW, EBL, YM. Patient enrollment and clinical care: CM, CX, DMH, TFG, BR. Writing: KC-D’A, TFG. Guarantor author: TFG. All authors reviewed and approved the manuscript. There are no relevant financial or non-financial competing interests to report.
Competing interests No, there are no competing interests.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.