Mathematical modeling of an immune checkpoint inhibitor and its synergy with an immunostimulant

Elpiniki Nikolopoulou; Steffen E. Eikenberry; Jana L. Gevertz; Yang Kuang

doi:10.3934/dcdsb.2020138

Article Contents

2021, Volume 26, Issue 4: 2133-2159. Doi: 10.3934/dcdsb.2020138

This issue Previous Article Dynamics of phytoplankton species competition for light and nutrient with recycling in a water column Next Article Carrying simplex in the Lotka-Volterra competition model with seasonal succession with applications

Mathematical modeling of an immune checkpoint inhibitor and its synergy with an immunostimulant

1.
School of Mathematical and Statistical Sciences, Arizona state University, Tempe, AZ 85287-1804, USA
2.
Department of Mathematics and Statistics, The College of New Jersey, Ewing, NJ, USA
3.
School of Mathematical and Statistical Sciences, Arizona state University, Tempe, AZ 85287-1804, USA

^* Corresponding author: Yang Kuang
^* Corresponding author: Yang Kuang

Received: July 2019

Revised: December 2019

Early access: April 26, 2020

Published online: April 26, 2020

Abstract / Introduction Full Text(HTML) Figure(7) / Table(3) Related Papers Cited by

Abstract

Immune checkpoint inhibitors (ICIs) are a novel cancer therapy that may induce tumor regression across multiple types of cancer. There has recently been interest in combining the ICIs with other forms of treatments, as not all patients benefit from monotherapy. We propose a mathematical model consisting of ordinary differential equations to investigate the combination treatments of the ICI avelumab and the immunostimulant NHS-muIL12. We validated the model using the average tumor volume curves provided in Xu et al. (2017). We initially analyzed a simple generic model without the use of any drug, which provided us with mathematical conditions for local stability for both the tumorous and tumor-free equilibrium. This enabled us to adapt these conditions for special cases of the model. Additionally, we conducted systematic mathematical analysis for the case that both drugs are applied continuously. Numerical simulations suggest that the two drugs act synergistically, such that, compared to monotherapy, only about one-third the dose of both drugs is required in combination for tumor control.

Keywords:

Mathematics Subject Classification: Primary: 58F15, 58F17; Secondary: 53C35.

Citation:

Full Text(HTML)

Figure 1. Schematic model representation for Tumor-Immune Interactions. Sharp arrows indicate proliferation/activation. Blocked arrows indicate killing/blocking. Dashed lines indicate proteins on the tumour (V) or T cells (T). IL-12 stimulates the proliferation of activated T-cells. NHS-muIL12's role is to administrate IL-12 to damaged intratumoral regions. PD-L1 is mainly expressed on tumor cells. Anti-PD-L1 binds to PD-L1 to prevent the formation of the PD-1-PD-L1 complex

Download: Full-size image PowerPoint slide

Figure 2. Tumor volume data and simulations using model (3)-(6) for each single-agent and combination therapy case. Case (a) administration of no drug. Case (b), (c) treatment with NHS-muIL12 $ 2\mu g $ and $ 10 \mu g $ respectively on day $ 0 $. Case (d) administration of avelumab $ 200 \mu g $ on days 0, 3 and 6. Case (e) treatment with both avelumab ($ 200 \mu g $) on days 0, 3, 6 and NHS-muIL12 ($ 2\mu g $) on day 0

Download: Full-size image PowerPoint slide

Figure 3. First graph depicts the model simulation for the combination treatment case of avelumab ($ 200 \mu g $) and NHS-muIL12 ($ 10\mu g $). The remaining two graphs correspond to each of the drugs administrated with respect to time

Download: Full-size image PowerPoint slide

Figure 4. Surface plot depicting the asymptotic behavior of systems (3)-(6) and (18)-(19) for combinations of $ \gamma_1 $ (avelumab) and $ \gamma_2 $ (NHS-muIL12) drugs, where both are applied continuously. Based on a threshold, three regions, namely, 'tumor control', 'tumor escape' and 'intermediate' are identified, which are separated by hyperbolic curves. The hyperbolic curve closer to the origin corresponds to the system (18)-(19) and the other one to (3)-(6). Six representative points indicated by different markers and colors, are chosen and shown on the plot. Their respective tumor volume behavior is plotted in Figure 5 employing the same marker and color notation. Note that the red point falls in the intermediate region, where the behavior is dependent on the system being used

Download: Full-size image PowerPoint slide

Figure 5. A subset of tumor volume behaviors of both the full system (3)-(6) and the limiting system (18)-(19) for continuous dosage with $ \gamma_1 $ (avelumab) and $ \gamma_2 $ (NHS-muIL12) respectively. The threshold chosen is shown on the plot with a dashed line. We observe that the qualitative tumor volume behaviors between the systems are comparable with the exception of the red point, which lies in the intermediate region as shown in Figure 4

Download: Full-size image PowerPoint slide

Figure 6. Surface plot depicting the asymptotic behavior of the full system (3)-(6) for combinations of $ \gamma_1 $ (avelumab) and $ \gamma_2 $ (NHS-muIL12) drugs, where both are applied continuously. Based on a threshold, two regions, namely, 'Tumor Control' and 'Tumor Escape' are identified, which are separated by a hyperbolic curve. Six representative points indicated by different markers and colors, are chosen and shown on the plot

Download: Full-size image PowerPoint slide

Figure 7. Surface plot depicting the asymptotic behavior of the limit case system (18)-(19) for combinations of $ \gamma_1 $ (avelumab) and $ \gamma_2 $ (NHS-muIL12) drugs, where both are applied continuously. Based on a threshold, two regions, namely, 'Tumor Control' and 'Tumor Escape' are identified, which are separated by a hyperbolic curve. Six representative points indicated by different markers and colors, are chosen and shown on the plot

Download: Full-size image PowerPoint slide

Table 1. State Variables of the model system (1)-(2) and (3)-(6).

Variable	Meaning	Unit
$V$	tumor cell volume	mm$^3$
$T$	volume of activated T cells	mm$^3$
$L$	free PD-L1 volume	mm$^3$
$P$	free PD-1 volume	mm$^3$
$A_{1}$	anti-PD-L1 concentration	g
$A_{2}$	NHS-muIL12 concentration	g
$Q$	PD-1-PD-L1 volume	mm$^3$

| Show Table

DownLoad: CSV

Table 2. Parameters and variables (Var.) of the model system (3)-(6)

Var.	Meaning	Value	Reference
$r$	Tumor cell growth rate	$0.213\text{ day}^{-1}$	fitted
$\eta$	Kill rate of tumor cells by T cells	$1$ mm$^{-3}$ $\cdot$ day$^{-1}$	fitted
$\delta$	Source of T cell activation	0.02 mm$^{3}$ /day	estimated
$\lambda_{TI_{12}}$	Activation rate of T cells by IL-12	8.81 day$^{-1}$	[14]
$K_{A_{2}}$	Dissociation constant of $A_{2}$	$7 \cdot 10^{-14}$ moles/liter	estimated
$K_{TQ}$	Inhibition of function of T cells by PD-1-PD-L1	$10^{-13}$ mm$^{6}$	estimated
$d_{T}$	Death rate of T cells	$0-0.5\text{ day}^{-1}$	[14]
$d_{A_{1}}$	Degradation rate of Anti-PD-L1	$ 0.1136\text{ day}^{-1}$	[17]
$d_{A_{2}}$	Degradation rate of NHS-muIL12	0.69 day$^{-1}$	[12]
$\rho_{p}$	Expression level of PD-1	$3.19\cdot10^{-7}$ - $8.49\cdot 10^{-7}$	[11]
$\rho_{L}$	Expression level of PD-L1	$3.56\cdot10^{-7}$ - $1.967\cdot 10^{-6}$	[11]
$K_{A_{1}}$	Dissociation constant of free PD-L1 with anti-PD-L1	10$^{-13}$ mol/liter	estimated
$\epsilon_{v}$	Expression of PD-L1 in tumor cells vs. T cells	1-100	[14]
$\sigma$	fraction of complex association and dissociation	0.01mm$^{-3}$	estimated
$\gamma_1$	continuous infusion rate of avelumab	$10^{-7}-9\cdot 10^{-5}$ g/day	estimated
$\gamma_2$	continuous infusion rate of NHS-muIL12	$10^{-9}-2\cdot 10^{-6}$ g/day	estimated
$c_1$	conversion constant for $A_1$ drug	$55^{-1}10^{-7}-55^{-1}10^{-6}$	estimated
$c_2$	conversion constant for $A_2$ drug	$75^{-1}10^{-7}- 75^{-1}10^{-6}$	estimated

| Show Table

DownLoad: CSV

Table 3. Parameter values used in simulations for equations (3)-(6)

Variable	Meaning	Value
$r$	Tumor cell growth rate	$0.213\text{ day}^{-1}$
$\eta$	Kill rate of tumor cells by T cells	$1$ mm$^{-3}$ $\cdot$ day$^{-1}$
$\delta$	Source of activation	0.02 mm$^{3}$ /day
$\lambda_{TI_{12}}$	Activation rate of T cells by IL-12	8.81 day$^{-1}$
$K_{A_{2}}$	Dissociation constant of $A_{2}$	$7 \cdot 10^{-14}$ moles/liter
$K_{TQ}$	Inhibition of function of T cells by PD-1-PD-L1	$10^{-13}$ mm$^{6}$
$d_{T}$	Death rate of T cells	$0.05\text{ day}^{-1}$
$d_{A_{1}}$	Degradation rate of Anti-PD-L1	$ 0.1136\text{ day}^{-1}$
$d_{A_{2}}$	Degradation rate of NHS-muIL12	0.69 day$^{-1}$
$\rho_{p}$	Expression level of PD-1	$5.84\cdot 10^{-7}$
$\rho_{L}$	Expression level of PD-L1	$2.7635\cdot 10^{-7}$
$K_{A_{1}}$	Dissociation constant of PD-L1 with anti-PD-L1	10$^{-13}$ mol/liter
$\epsilon_{v}$	Expression of PD-L1 in tumor cells vs. T cells	50
$\sigma$	fraction of complex association and dissociation	0.001 mm$^{-3}$
$\gamma_1$	prescribed infusion rate of avelumab	$10^{-7}-9\cdot 10^{-5}$ g/day
$\gamma_2$	prescribed infusion rate of NHS-muIL12	$10^{-9}-2\cdot 10^{-6}$ g/day
$c_1$	conversion constant for $A_1$ drug	$55^{-1}10^{-7}$
$c_2$	conversion constant for $A_2$ drug	$75^{-1}10^{-7}$

| Show Table

DownLoad: CSV

Related Papers

Cited by

References

[1]	H. O. Alsaab, S. Sau, R. Alzhrani, K. Tatiparti, K. Bhise, S. K. Kashaw and A. K. Iyer, PD-1 and PD-L1 checkpoint signaling inhibition for cancer immunotherapy: Mechanism, combinations, and clinical outcome, Frontiers in Pharmacology, 8 (2017), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5572324/. doi: 10.3389/fphar.2017.00561.
[2]	R. H. Blair, D. L. Trichler and D. P. Gaille, Mathematical and statistical modeling in cancer systems biology, Frontiers in Physiology, 3 (2012), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3385354/. doi: 10.3389/fphys.2012.00227.
[3]	R. H. Byrd, J. C. Gilbert and J. Nocedal, A trust region method based on interior point techniques for nonlinear programming, Mathematical Programming, 89 (2000), 149-185. doi: 10.1007/PL00011391.
[4]	K. Chin, V. K. Chand and D. S. A. Nuyten, Avelumab: Clinical trial innovation and collaboration to advance anti-PD-L1 immunotherapy, Annals of Oncology: Official Journal of the European Society for Medical Oncology, 28 (2017), 1658–1666, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5834034/. doi: 10.1093/annonc/mdx170.
[5]	S. Halle, K. A. Keyser, F. R. Stahl, A. Busche, A. Marquardt, X. Zheng, M. Galla, V. Heissmeyer, K. Heller, J. Boelter, K. Wagner, Y. Bischoff, R. Martens, A. Braun, K. Werth, A. Uvarovskii, H. Kempf, M. Meyer-Hermann, R. Arens, M. Kremer, G. Sutter, M. Messerle and Reinhold Förster, In vivo killing capacity of cytotoxic T Cells is limited and involves dynamic interactions and T Cell cooperativity, \emphImmunity, 44 (2016), 233-245, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4846978/. doi: 10.1016/j.immuni.2016.01.010.
[6]	L. F. Han, S. Eikenberry, C. H. He, L. Johnson, M. C. Preul, E. J. Kostelich and Y. Kuang, Patient-specific parameter estimates of glioblastoma multiforme growth dynamics from a model with explicit birth and death rates, Mathematical Biosciences and Engineering, 16 (2019), 5307–5323, https://www.aimspress.com/fileOther/PDF/MBE/mbe-16-05-265.pdf. doi: 10.3934/mbe.2019265.
[7]	V. R. Juneja, K. A. McGuire, R. T. Manguso, M. W. LaFleur, N. Collins, W. N. Haining, Gordon J. Freeman and Arlene H. Sharpe, PD-L1 on tumor cells is sufficient for immune evasion in immunogenic tumors and inhibits CD8 T cell cytotoxicity, The Journal of Experimental Medicine, 214 (2017), 895–904, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5379970/. doi: 10.1084/jem.20160801.
[8]	J. Kang, S. Demaria and S. Formenti, Current clinical trials testing the combination of immunotherapy with radiotherapy, Journal for Immunotherapy of Cancer, 4 (2016), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5028964/. doi: 10.1186/s40425-016-0156-7.
[9]	Y. Kuang, J. D. Nagy and S. E. Eikenberry, Introduction to Mathematical Oncology, CRC Press, Boca Raton, FL, 2016.
[10]	V. A. Kuznetsov, I. A. Makalkin, M. A. Taylor and A. S. Perelson, Nonlinear dynamics of immunogenic tumors: Parameter estimation and global bifurcation analysis, Bulletin of Mathematical Biology, 56 (1994), 295–321, https://www.sciencedirect.com/science/article/pii/S0092824005802605.
[11]	X. L. Lai and A. Friedman, Combination therapy of cancer with cancer vaccine and immune checkpoint inhibitors: A mathematical model, PLoS ONE, 12 (2017), 1-24. doi: 10.1371/journal.pone.0178479.
[12]	T. List and D. Neri, Immunocytokines: A review of molecules in clinical development for cancer therapy, Clinical Pharmacology, 5 (2013), 29–45, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3753206/. doi: 10.2147/CPAA.S49231.
[13]	X.-Y. Meng, N.-N. Qin and H.-F. Huo, Dynamics analysis of a predator-prey system with harvesting prey and disease in prey species, Journal of Biological Dynamics, 12 (2018), 342-374. doi: 10.1080/17513758.2018.1454515.
[14]	E. Nikolopoulou, L. R. Johnson, D. Harris, J. D. Nagy, E. C. Stites and Y. Kuang, Tumour-immune dynamics with an immune checkpoint inhibitor, Letters in Biomathematics, 5 (2018), S137–S159. doi: 10.1080/23737867.2018.1440978.
[15]	Z. P. Parra-Guillen, A. Janda, P. Alzuguren, P. Berraondo, R. Hernandez-Alcoceba and I. F. Troconiz, Target-mediated disposition model describing the dynamics of IL12 and IFN$\gamma$ after administration of a mifepristone-inducible adenoviral vector for IL-12 expression in mice, AAPS Journal, 15 (2013), 183–194, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3535095/.
[16]	A. Radunskaya, R. Kim and T. Woods II, Mathematical modeling of tumor immune interactions: A closer look at the role of a PD-L1 inhibitor in cancer immunotherapy, Spora: A Journal of Biomathematics, 4 (2018), 25–41, https://ir.library.illinoisstate.edu/cgi/viewcontent.cgi?article=1022&context=spora. doi: 10.30707/SPORA4.1Radunskaya.
[17]	A. Rao and M. R. Patel, A review of avelumab in locally advanced and metastatic bladder cancer, Therapeutic Advances in Urology, 11 (2019), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6354303/. doi: 10.1177/1756287218823485.
[18]	E. A. Saparata and L. G. Pillis, A Comparison and catalog of intrinsic tumor growth models, Bulletin of Mathematical Biology, 76 (2014), 2010–2024, https://www.ncbi.nlm.nih.gov/pubmed/25081547. doi: 10.1007/s11538-014-9986-y.
[19]	R. Serre, S. Benzekry, L. Padovani, C. Meille, N. André, J. Ciccolini, F. Barlesi, X. Muracciole and D. Barbolosi, Mathematical modeling of cancer immunotherapy and its synergy with radiotherapy, Cancer Research, 76 (2016), 4931–4940, http://cancerres.aacrjournals.org/content/76/17/4931. doi: 10.1158/0008-5472.CAN-15-3567.
[20]	L. Shi, S. Chen, L. Yang and Y. Li, The role of PD-1 and PD-L1 in T-cell immune suppression in patients with hematological malignancies., Journal of Hematology and Oncology, 6 (2013), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3851976/. doi: 10.1186/1756-8722-6-74.
[21]	S. Shi, J. Huang and Y. Kuang, Global dynamics in a tumor-immune model with an immune checkpoint inhibitor, DCDS-B, in review.
[22]	S. Simon and N. Labarriere, PD-1 expression on tumor-specific T cells: Friend or foe for immunotherapy?, \emphOncoimmunology, 7 (2017), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5739549/.
[23]	A. Stéphanou, P. Ballet and G. Powathil, Hybrid modelling in oncology: Success, challenges and hopes, (2019), https://arXiv.org/pdf/1901.05652.pdf.
[24]	A. Talkington and R. Durett, Estimating tumor growth rates in vivo, Bulletin of Mathematical Biology, 77 (2015), 1934–1954, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4764475/. doi: 10.1007/s11538-015-0110-8.
[25]	S. G. Tan, K. F. Liu, Y. Chai, C. W.-H. Zhang, S. Gao, G. F. Gao and J. X. Qi, Distinct PD-L1 binding characteristics of therapeutic monoclonal antibody durvalumab, Protein and Cell, 9 (2018), 135-139. doi: 10.1007/s13238-017-0412-8.
[26]	J. Tang, J. X. Yu, V. M. Hubbard-Lucey, S. T. Neftelinov, J. P. Hodge and Y. Q. Lin, The clinical trial landscape for PD1/PDL1 immune checkpoint inhibitors, Nature Reviews Drug Discovery, 17 (2018), 854-855. doi: 10.1038/nrd.2018.210.
[27]	J. R. Wares, J. J. Crivelli, C. Yun, I. Choi, J. L. Gevertz and P. S. Kim, Treatment strategies for combining immunostimulatory oncolytic virus therapeutics with dendritic cell injections, Mathematical Biosciences and Engineering, 12 (2015), 1237-125. doi: 10.3934/mbe.2015.12.1237.
[28]	A. B. Warner and M. A. Postow, Combination controversies: Checkpoint inhibition alone or in combination for the treatment of melanoma?, \emphOncology, 32 (2018), 228–34, https://www.ncbi.nlm.nih.gov/pubmed/29847853.
[29]	C. X. Xu, Y. P. Zhang, P. A. Rolfe, V. M. Hernández, W. Guzman, G. Kradjian, B. Marelli, G. Qin, J. Qi, H. Wang, H. Yu, R. Tighe, K. Lo, J. M. English, L. Radvanyi and Y. Lan, Combination therapy with NHS-muIL12 and avelumab (anti-PD-L1) enhances antitumor efficacy in preclinical cancer models, Clinical Cancer Research, 23 (2017), 5869–5880, http://clincancerres.aacrjournals.org/content/23/19/5869. doi: 10.1158/1078-0432.CCR-17-0483.
[30]	Y. Y. Yan, A. B. Kumar, H. Finnes, S. N. Markovic, S. Park, R. S. Dronca and H. Dong, Combining immune checkpoint inhibitors with conventional cancer therapy, Frontiers in Immunology, 9 (2018), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6072836/. doi: 10.3389/fimmu.2018.01739.
[31]	D. Zamarin and M. A. Postow, Immune checkpoint modulation: Rational design of combination strategies, Pharmacology and Therapeutics, 150 (2015), 23–32, https://www.sciencedirect.com/science/article/pii/S0163725815000042. doi: 10.1016/j.pharmthera.2015.01.003.