In silico trials predict that combination strategies for enhancing vesicular stomatitis oncolytic virus are determined by tumor aggressivity

Abstract

Background: Immunotherapies, driven by immune-mediated antitumorigenicity, offer the potential for significant improvements to the treatment of multiple cancer types. Identifying therapeutic strategies that bolster antitumor immunity while limiting immune suppression is critical to selecting treatment combinations and schedules that offer durable therapeutic benefits. Combination oncolytic virus (OV) therapy, wherein complementary OVs are administered in succession, offer such promise, yet their translation from preclinical studies to clinical implementation is a major challenge. Overcoming this obstacle requires answering fundamental questions about how to effectively design and tailor schedules to provide the most benefit to patients.

Methods: We developed a computational biology model of combined oncolytic vaccinia (an enhancer virus) and vesicular stomatitis virus (VSV) calibrated to and validated against multiple data sources. We then optimized protocols in a cohort of heterogeneous virtual individuals by leveraging this model and our previously established in silico clinical trial platform.

Results: Enhancer multiplicity was shown to have little to no impact on the average response to therapy. However, the duration of the VSV injection lag was found to be determinant for survival outcomes. Importantly, through treatment individualization, we found that optimal combination schedules are closely linked to tumor aggressivity. We predicted that patients with aggressively growing tumors required a single enhancer followed by a VSV injection 1 day later, whereas a small subset of patients with the slowest growing tumors needed multiple enhancers followed by a longer VSV delay of 15 days, suggesting that intrinsic tumor growth rates could inform the segregation of patients into clinical trials and ultimately determine patient survival. These results were validated in entirely new cohorts of virtual individuals with aggressive or non-aggressive subtypes.

Conclusions: Based on our results, improved therapeutic schedules for combinations with enhancer OVs can be studied and implemented. Our results further underline the impact of interdisciplinary approaches to preclinical planning and the importance of computational approaches to drug discovery and development.

Trial registration: ClinicalTrials.gov NCT02285816 NCT02879760.

Keywords: clinical trials as topic; combination; computational biology; drug evaluation; drug therapy; oncolytic virotherapy; preclinical.

Conflict of interest statement

Competing interests: None declared.

© Author(s) (or their employer(s)) 2021. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

Figures

Figure 1

Schematic representation of the tumor growth model under combination OV-therapy. (A) Biological assumptions for the combination OV-therapy interactions between VV (enhancer) and VSV oncolytic viruses. Infection of cells by either VV or VSV results in cell lysis, whereby new virus particles are released along with a cocktail of antigens, antivirals and cytokines. For simplicity, we considered each virus to release associated cytokines concentrations that can independently instigate the recruitment of immune cells (such as phagocytes). However, in the presence of both VV and VSV infections, we assume the cytokine production decreases the recruitment of immune cells, allowing for a more targeted immune response and virus-induced cell lysis. Additionally, VSV releases antivirals that block the intracellular replication of the virus and the infection of neighboring cells. In comparison, lysis of VV infected cells produces antivirals that downregulate the antivirals produced by VSV, allowing for infection and replication to occur. Once activated, these immune cells induce cell apoptosis of uninfected cancer cells. (B) In the model, quiescent susceptible cells (light blue) activate and begin division by transitioning into the G1 phase of the cell cycle. Cells exit G1 to enter the active phase (mitosis) and complete division. Most susceptible cells in the active phase re-enter quiescence after mitosis, however, certain dividing cells may mutate into an immune-resistant lineage (red). Immune interactions are driven by immune cells who encounter quiescent,G1 and actively dividing susceptible tumor cells. Tumor-immune interactions increase proinflammatory cytokine concentrations to recruit additional immune cells to the tumor site. VSV and VV infect both normal and immune-resistant tumor cells, creating virus-specific infected cell pools. These infected cells undergo lysis releasing new virus progeny. The virus also influences the cytokine production which controls the immune cell production and activity. VSV: vesicular stomatitis virus; VV: vaccinia virus.

Figure 2

In silico trial strategy recapitulates experimentally observed variability. (A) (1) Model parameters are established by calibrating experimental results to the model’s predictions. A distribution of responses centered at the mean of the experimental data is then used to generate parameter sets representing virtual individuals. (2) To populate the trial, each virtual patient’s tumor growth is simulated to determine whether they are candidates for the trial. Patients whose tumor growth is acceptable (ie, clinically relevant) are placed into repeated identical cohorts. (3) Alternative treatment schedules are then tested on each cohort by simulating individual virtual patient responses with the mathematical model and summarizing cohort level outcomes (such as mean and SD of responses). (4) optimal actionable schedules are then inferred by comparing cohort level and individual outcomes. (B) Tumor growth (relative to tumor volume on day 6) over time in absence of treatment. Black line: model fit; red stars: experimental observations measured by Le Boeuf et al, gray shaded region: distribution of growth from full cohort of patients. (C) Virtual patients were ordered based on intrinsic tumor growth rates r (top and bottom 10% denoted by shaded regions). OV: oncolytic virus.

Figure 3

Influence of enhancer injection multiplicity on tumor burden. (A) The effects of enhancer multiplicity (NE) were investigated by simulating 1–7 VV enhancers, with VSV administered 7 days after the final enhancing dose. Tumor growth was assessed 15 days after the administration of VSV. (B) Distribution in number of tumor cells 15 days after VSV administration with respect to the multiplicity of enhancers. Central mark (red) indicates median, bottom edge denotes the bottom quartile, top edge denotes the top quartile. Significance indicators report the non-significant results of a Kolmogorov-Smirnov test for significance of difference between distributions (p§amp;lt;0.05). (C) Tumor growth dynamics from last enhancer to 15 days after VSV administration for protocols with 1 (blue) and 7 (red) enhancers. Mean is denoted by a solid line, SD by shaded regions of same color and individual virtual patient values are plotted as circles. (D) Kaplan-Meier survival curves for protocols with 1 (dark blue), 2 (light blue) and 7 (red) enhancers. No significant difference between protocols from 2 to 7 enhancers was found (measured by log-rank test for significance, p§amp;lt;0.05). VSV: vesicular stomatitis virus; VV: vaccinia virus.

Figure 4

Individual responses to multiple enhancer injections protocols are stratified by intrinsic tumor growth rates. (A) Waterfall plot of the change in tumor size 15 days after VSV administration. Bar color depicts the optimal number of enhancers. Corresponding tumor growth rate (r) for each patient are plotted in gray. (B) Ordering of protocols from best (bottom row) to worst (top row) for each patient based on the tumor size 15 days after VSV administration. Corresponding tumor growth rates are plotted above (patient ordering identical based on intrinsic tumor growth rate as in A). VSV: vesicular stomatitis virus; VV: vaccinia virus.

Figure 5

Longer VSV lags have a detrimental effect on therapeutic success. (A) Inspired by the results for enhancer multiplicity, the effects of the length of VSV lags were investigated by simulating either a single enhancer protocol (left) or a seven enhancer protocol (right). VSV lags were varied from 1 to 15 days after the final enhancer. Tumor growth was assessed 15 days after the administration of the last VSV. (B) Distribution in number of tumor cells 15 days after VSV administration with respect to the duration of VSV lag after 1 (orange) or 7 (blue) enhancers. Central mark (red) indicates median, bottom edge denotes the bottom quartile, top edge denotes the top quartile and individual virtual patient values are plotted as circles. (C) Tumor growth as a function of time. Markers indicate significant differences in final tumor size (t-test, p§amp;gt;0.05). (D) Cytokine concentrations as a function of time. (E) Immune cells as a function of time. In C, D, E: 1 enhancer, 1-day VSV lag (solid blue), 7 enhancers, 1-day VSV lag (dashed blue), 1 enhancer, 15-day VSV lag (solid red), 7 enhancers, 15-day VSV lag (dashed red). VSV: vesicular stomatitis virus; VV: vaccinia virus.

Figure 6

Optimal VSV lag is stratified by intrinsic tumor growth rates. (A) Waterfall plot of the change in tumor size 15 days after the last VSV administration for the one enhancer protocol. Bar color depicts the optimal VSV lag. Corresponding tumor growth rate (r) for each patient are plotted in gray. (B) Waterfall plot of the change in tumor size 15 days after the last VSV administration for the seven enhancer protocol. Bar color depicts the optimal VSV lag. Corresponding tumor growth rate (r) for each patient are plotted in gray. (C) Ordering of protocols from best (bottom row) to worst (top row) for each patient based on the tumor size 15 days after the last VSV administration for seven enhancer protocol. Corresponding tumor growth rates are plotted above (patient ordering identical based on intrinsic tumor growth rate as in B. The ordering of the optimal VSV lag for the one enhancer protocol is provided in online supplemental figure S6 in online supplemental information). VSV: vesicular stomatitis virus.

Figure 7

Individualized schedules are determined by tumor aggressivity and risk stratification according to tumor aggressivity is necessary for optimal outcomes. (A) Optimal number of enhancers (yellow), optimal VSV lag (fuchsia) and relative tumor size 15 days after last VSV versus untreated control (purple) as a function of intrinsic tumor growth rate. For all but a subset of the least aggressive tumors, individualized protocols called for a VSV lag of 1 day, with fewer than seven enhancers. (B) Two new cohorts of patients were generated with either aggressive tumor growth (0.0629<r<0.0657, purple) or slow tumor growth (0.0196<r<0.0260, green). Original Le Boeuf et al data (red stars) and model fit to the original data (black curve) as in figure 2C. (C) Each cohort was simulated according to the previously determined optimal aggressive protocol (one enhancer followed by a VSV 1 day later) and optimal slow protocol (seven enhancers followed by a VSV 15 days later). To assess the effect specificity of each protocol, a cross-over trial wherein virtual patients with fast growth were treated with the slow protocol and vice versa was performed. (D) Kaplan-Meier survival curves for the two cohorts under the two different protocols. (E) To confirm the robustness of the aggressive and slow protocols, the optimal number of enhancers and VSV lag was then determined for patients in the new cohorts. The results of the newly generated cohort were then compared with the original cohort in A. Tumor size 15 days after the VSV was assessed. Original cohort individualized therapy compared with the new slow growth (top left) and new aggressive growth (top right) individualized schedules. Overlays of the corresponding optimal number of enhancers and VSV lag for each patient from old protocol versus the new slow growth (bottom left) and aggressive growth (bottom right). VSV: vesicular stomatitis virus; VV: vaccinia virus.