Informatics in Medicine Unlocked 23 (2021) 100534

Contents lists available at ScienceDirect

Informatics in Medicine Unlocked

journal homepage: http://www.elsevier.com/locate/imu

Prospect for application of mathematical models in combination

cancer treatments
Joseph Malinzi a, b, Kevin Bosire Basita c, Sara Padidar d, Henry Ademola Adeola e, f, *
a
Department of Mathematics, Faculty of Science and Engineering, University of Eswatini, Private Bage 4, Kwaluseni, Eswatini
b
Institute of Systems Science, Durban University of Technology, Durban, 4000, South Africa
c
Department of Mathematics, Faculty of Applied Science and Technology, Technical University of Kenya, Haile Selassie Avenue, Nairobi, Kenya
d
Department of Biological Sciences, Faculty of Science and Engineering, University of Eswatini, Private Bage 4, Kwaluseni, Eswatini
e
Division of Dermatology, Department of Medicine, Faculty of Health Sciences and Groote Schuur Hospital, University of Cape Town, Cape Town, South Africa
f
Department of Oral and Maxillofacial Pathology, Faculty of Dentistry, University of the Western Cape and Tygerberg Hospital, Cape Town, South Africa

A R T I C L E I N F O A B S T R A C T

Keywords: The long-term efficacy of targeted therapeutics for cancer treatment can be significantly limited by the type of
Combination cancer therapy therapy and development of drug resistance, inter alia. Experimental studies indicate that the factors enhancing
Mathematical models of combination cancer acquisition of drug resistance in cancer cells include cell heterogeneity, drug target alteration, drug inactivation,
therapy
DNA damage repair, drug efflux, cell death inhibition, as well as microenvironmental adaptations to targeted
Chemoimmunotherapy
therapy, among others. Combination cancer therapies (CCTs) are employed to overcome these molecular and
Chemovirotherapy
Chemoradiotherapy pathophysiological bottlenecks and improve the overall survival of cancer patients. CCTs often utilize multiple
Radioimmunotherapy combinatorial modes of action and thus potentially constitute a promising approach to overcome drug resistance.
Immunovirotherapy Considering the colossal cost, human effort, time and ethical issues involved in clinical drug trials and basic
Radiovirotherapy medical research, mathematical modeling and analysis can potentially contribute immensely to the discovery of
better cancer treatment regimens. In this article, we review mathematical models on CCTs developed thus far for
cancer management. Open questions are highlighted, and plausible combinations are discussed based on the
level of toxicity, drug resistance, survival benefits, preclinical trials and other side effects.

1. Introduction Computational biology and mathematical modeling have enabled re­

Cancer is the second leading cause of death globally, with 70% of
cancer mortality from low- and middle-income countries [1]. To date,
cancer treatment has heavily relied on surgery, often followed by
chemotherapy and/or radiotherapy [2–4]. Recent advances with the use
of virological and immunological agents including nanotechnology de­
livery methods have increased our arsenal in treating this complex dis­
ease [5]. Given the high costs involved in treating cancer, and the
continued development of better and less toxic treatment approaches,
mathematical modeling provides a valuable and low-cost tool for re­
searchers by pre-dicting treatment outcomes and identifying potential
best combinations to therapeutic approaches with minimal adverse
effects.
The use of mathematical modeling in understanding a variety of
disease states, including cancer, has been well documented [6,7].
searchers to de-velop new treatments, from screening potential candi­
date compounds [8] and building and identifying chemical structures
that will best fit target sites [6,9] to predicting the outcome of novel
combinations of drug regi-mens [10–17]. This paper reviews a novel use
of mathematical modeling to enable researchers to predict the treatment
outcomes of combined therapeutic strategies in the treatment of cancers.
Benign cancers and locally confined malignant cells can often be
treated by a single treatment strategy such as surgery or radiation
therapy alone. In some cases, a combination approach is employed such
as the use of (neo) adjuvant radiotherapy or chemotherapy before or
after surgery to reduce the size of a tumour, thereby increasing chances
of complete surgical resection and/or destroying of any post-surgical
residual cancer cells [18,19]. Employing combined therapeutic strate­
gies is essential to many malignant tumours where a single approach (e.
g. surgery) is not always suitable, for instance in bladder cancer or

* Corresponding author. Division of Dermatology, Department of Medicine, Faculty of Health Sciences and Groote Schuur Hospital, University of Cape Town,
Observatory 7925, Cape Town, South Africa.
E-mail address: henry.adeola@uct.ac.za (H.A. Adeola).

https://doi.org/10.1016/j.imu.2021.100534
Received 4 September 2020; Received in revised form 29 January 2021; Accepted 5 February 2021
Available online 13 February 2021
2352-9148/© 2021 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
J. Malinzi et al. Informatics in Medicine Unlocked 23 (2021) 100534

non-small cell lung cancer that may not be operable.

Although mono-therapeutic strategies claim a large share in the
medical world for the treatment of several cancer types, they are
considered less effective in comparison with the combination therapy
approach [20]. Most standard mono-therapeutic strategies (e.g.
chemotherapy alone) destroy all the actively proliferating cells
non-selectively, ultimately causing the annihilation of both healthy and
cancerous cells [20]. In addition, chemother-apy is known to produce
toxic effects on the patients with a myriad of side effects, threat to life,
suppressing bone marrow cells and increasing vulnerability to infection
[20,21]. Hence, there is an overarching demand for tumour-targeted
therapies, that are minimally toxic especially in the era of precision
and personalized medicine. Combination cancer treatments hold great
appeals in boosting the efficacy of anticancer drugs, enhancing
apoptosis, suppressing tumour growth and decreasing cancer stem cell
population [22]. Although chemotherapy-based combination therapy
could be toxic, the toxicity can be greatly reduced when different
pathways are tar-geted [22,23]. Moreover, dosing for these agents works
in an additive manner, thereby lowering the cumulative amount of
therapeutic dosage needed in each individual drug [23].
Several advances have been made in tumour profiling, and deep
sequencing has revealed driver mutations as well as novel targets for the
development of new cancer drugs [24,25]. There have been remarkable
achieve-ments on combination therapy in recent years with much
attention focusing on most effective combinations that are likely to
produce significant effects on the tumour. However, this journey has not
been smooth with the process of designing plausible combinations fac­
ing numerous challenges. Current combination cancer therapy presumes
that better results are obtained by amalgamating therapeutic drugs at
their maximal tolerated doses [26]. Nevertheless, differences in the
pharmacokinetics of single drugs bring about inconsistency in the de­
livery of synergistic drug ratios to tumour cells. Compounded toxicity
observed in combination treatment protocols calls for application of
suboptimal dosages to counter the effects. Additionally, there are chal­
lenges regarding the identification of the best combination strategies
and the complications involved in the combination therapy [26].
Upon the selection of a reasonably designed combination, the pro­
cess of its early clinical development is complicated; thus, requiring
attention to detail [26]. Implementation of combination strategies is Fig. 1. A schematic diagram showing the evolution of cancer treatments. It
depicts better therapeutic results with the use of combination cancer strategies
influenced by several issues including toxicity, pharmacokinetic and
and that mathematical models play an integral role.
pharmacodynamic interactions, proper timing of resistance develop­
ment and the identification of robust biomarker in predicting the
response. Clearly, the questions of how to implement the combination, propose plausible combinations of treatment strategies to treat a variety
when to implement the combination and to whom it is implemented for of cancers.
are the major problems that face the clinical community during devel­
opment of treatment regimens [26]. 2. Mathematical models for combination cancer therapy
Production of new anticancer drugs takes a long time, is risky, and
requires costly medical experimentation; and thus, mathematical and 2.1. Models of chemoimmunotherapy
computational approaches are needed to supplement clinical trials in
making predictions. For this reason, efficient mathematical approaches Cancer chemotherapeutic drugs normally disrupt pathways neces­
are needed to try and understand different mechanisms of cancer ac­ sary for the growth and survival of tumours. However, they are often
tivities and effective combination therapy for cancer at a relatively limited due to the development of drug resis-tance and the damages
affordable cost. Many Mathematical models have been formulated to exerted to normal cells of the host tissue [27]. Recent years have focused
determine effective combination cancer therapies. Nonetheless, the cure on host-tumour interactions and it has been established that
for cancer is still far from reach, with complexities associated with host-tumour interactions play a pivotal role in determining clinical
therapeutic combinations remaining a major impediment. course and the outcomes of the treatments of human malignancies [27].
Here, we carry out a critical review of published mathematical Several mathematical models have been constructed in an attempt to
models that have been used to predict the outcome of combination understand tumour-immune interactions in depth and determine the
therapy strategies. The therapeutic strategies reviewed are chemo­ elements in the immune system that play a critical role in responding to
immunotherapy, chemovirothrapy, chemoradiotherapy, radio­ immunotherapy [28]. Several studies [29–35], have attempted to un­
immunotherapy, immunovirotherapy, radiovirotherapy and targeted derstand the effects of immune modulation by using mathematical
therapies (see Fig. 1: which shows the evolution of cancer treatments. approaches.
The figure further depicts that better therapeutic outcomes are achieved Modern cancer treatment methods rely on the ability of certain
when combination cancer strategies are used. It as well depicts that cancers to trigger the immune response. In- corporation of the immune
mathe-matical models shall play an integral role in deciding on strate­ component in the formulation of mathematical models has played a key
gies that are viable). Existing open problems are discussed, and we role in the clinically observed phenomena, for instance, tumour

2
J. Malinzi et al.
dormancy, unchecked growth of tumours and oscillations in tumour size
[36]. The first attempt to illustrate the immunotherapy related effects in
a suitable ordinary differential equations (ODEs) model was given by
Kirschner and Panetta [37]. The study considered inter-leukin (IL-2)
along with adaptive cellular immunotherapy (ACI) by developing
dynamical equations explaining external inflow of both IL-2 and im­
mune cells.
Chappell et al. [10] presented a high-level abstraction mathematical
model that explored the interaction of immune cells and tumour cells,
where they further explored the merits of combining immunothera­
peutic agents with chemotherapy and radiotherapy. The model was
tested, and numerical simulations were carried out using data similar to
that shown by Deng et al. [38]. Numerical results showed that tumour
mass reduced significantly when radiotherapy and immunotherapy are
used in combination whereas single therapy recorded no significant
decrease in tumour mass. Further, a combination of radiotherapy and
immunotherapy leads to an increase in the number of T-cells that are
activated compared to single therapies.
Another mathematical model, comprising of ODEs, focussed on the
interplay between NK and CD8+ T cells with different varieties of
tumour cell lines [39]. The results found that ligand transduced cells
trigger enough immune response to put tumour growth under control,
whereas control-transduced tumour cells evade immunity. In a more
recent study, de Pillis et al. [28] extended their earlier study [39] by
incorporating the latest research on baseline NK and activated CD8+
T-cells in both healthy donors and cancer patients. The model was
quali-tatively analyzed and supplemented with numerical simulations of
chemoimmunotherapy and vaccine therapy.
The results of the study intimated that, depending on the efficacy of
the CD8+ T-cells, chemoimmunotherapy is likely to successfully control
tumour cells in body tissue.
Rodrigues et al. [40] developed a simple ODE model to study the
dynamics of chemoimmunotherapy of chronic lymphocytic leukemia.
Their model analysis showed that the application of chemotherapy
paves way for immunotherapy in a way that it decreases the number of
cancer cells thus reducing the tumour density on which immune cells
acts upon. It was established that there is a minimum number of immune
cells needed to be transplanted in adoptive treatment so as to ensure a
complete cure or remission. Nonetheless, more studies are needed to
model adoptive cellular immunotherapy.
Pang et al. [41] discussed single chemotherapy, immunotherapy and
mixed treatments as well as conditions triggering tumour eradication.
They analyzed the effects of least effective concentration and the
half-life of the therapeutic drugs where they found that better results can
be achieved if the half-life of the drug is ex-tended. Further, the impact
of drug resistance on therapeutic results was considered and a mathe­
matical model explaining the cause of chemotherapeutic failure that
uses a single drug was proposed. In the end, numerical simulations
showed that chemoimmunotherapy is likely to achieve a better treat­
ment effect. However, de-spite the reported achievement, tumour cells
are continuously becoming resistant to numerous structurally and
mechanistically unrelated drugs, limiting the effectiveness of chemo­
immunotherapy. Thus, determining how to effectively amalgamate
those treatment modes and design optimal dosage combination regi­
mens deserves further research.
2.2. Models of chemovirotherapy
The use of viruses as targeted therapy in cancer originated in the
early 20th century with numerous viruses already tested in humans and
experimental animals [42]. Oncolytic viruses (OVs) are a heterogeneous
group of low or non-pathogenic viruses employed in the treatment of
cancers, taking advantage of their ability to selectively infect and lyse
cancer cells without damaging normal cells [43,44]. Many viruses, both
genetically engineered and naturally occurring, are currently being
researched as oncolytic cancer virotherapeutic agents [43]. Examples of
3
Informatics in Medicine Unlocked 23 (2021) 100534
such viruses include vaccinia virus, measles virus, reovirus, adenovirus
and vesicular stomatitis virus [45,46]. During carcinogenesis, tumour
cells are able to gain genetic and cellular changes that interferes with the
host immunosurveillance capabilities [23]. Such targeted,
tumour-specific properties make oncolytic virotherapy, either solo or in
combination (e.g. chemovirotherapy), an attractive therapy for cancer.
In recent years, chemovirotherapy has been shown to enhance
therapeutic effects that are far from reach when each therapy is used
independently [11,20,22,47–49]. Ottolio-Perry et al. [20] reviewed the
combination of oncolytic viruses with the existing treatment modalities
with a view of helping investigators make informed decisions regarding
the clinical development of these products. In another study, Nguyen
et al. [47] highlighted the mechanisms that guarantee the successful
combination of oncolytic viruses with other therapeutic drugs where,
however, he concluded that the success of these combinations is subject
to several factors including the type of oncolytic virus used, type of
cancer targeted, dosage and the timing.
There is a wealth of mathematical models that have been developed
in an attempt to comprehend as well as characterize viral dynamics
[50–56]. Not many mathematical models have, nonetheless-thus far,
been de-veloped and studied to investigate the combination of chemo­
therapeutic drugs and viruses. Malinzi et al. [11] constructed and
analyzed a model of parabolic non-linear partial differential equations
(PDEs) to examine the spatiotemporal dynamics of tumour cells under
the treatment of chemovirotherapy. The proposed model con-sisted
compartments of uninfected and infected tumour cells densities, a free
virus, and a chemotherapeutic drug. It was shown that the success of
virotherapy is highly determined by the virus infection rate and burst
size. Further, travelling wave solutions revealed that tumour diffusivity
and growth rate are critical parame-ters during chemovirotherapy
treatment and numerical simulations confirmed that combining
chemotherapeutic drugs with oncolytic viruses is more effective
compared to either monotherapies.
In an effort to investigate the enhancement of chemotherapy by
oncolytic viruses, Malinzi et al. [12] pro-posed an ODE mathematical
model and an optimal control problem for chemovirotherapy. The
model described the interplays between tumour cells, immune response
and a treatment combination by combining virotherapy and chemo­
therapy. Stability analysis showed that tumours could grow to their
maximum capacity given that there is no form of treatment. It was
shown that chemotherapy alone is likely to clear tumour cells on the
con-dition that the efficacy of the drug is more than the intrinsic growth
rate of the tumour. The combined effects of oncolytic virotherapy and
chemotherapy were evaluated using sensitivity analysis of the model
parameters. Optimal control simulations revealed that the half of the
maximum tolerated dozes (MTDs) for chemotherapy and virotherapy
optimize treatment outcomes. Sensitivity analysis and numerical simu­
lations affirmed that the success of chemovirotherapy depends on virus
burst size, the rate of viral infection and the dose. Moreover, the right
dose of chemotherapy required to produce effective results in unison
with virotherapy has been a concern for both clinicians and
mathematicians.
Malinzi [22] proposed a mathematical model for chemovirotherapy
where he considered three-drug infusion methods and compared their
efficacies, carried out mathematical analysis to forecast the outcomes of
the OVs in combination with chemotherapy and also compared the ef­
ficacy of each single treatment modality. The model was formulated
based on two aspects, that is, model without delay and with delay.
Numerical simulations for both models were carried out, where it was
shown that if the tumour burst is big, then regardless of the drug infusion
method, chemovirotherapy is more effective than any of the single
treatments. Simulations further indicated that the success of chemo­
virotherapy depends on virus burst size, the rate of viral infection and
the dose. Further, simulations showed that incorporating delays in the
system increased the time within which tumour clearance occurred in
body tissue. Despite this study, there still exists open questions which
J. Malinzi et al.
need further investigation, for example, the model does not address the
quantity of dose required to completely eradicate the tumour and at
what specific point in the tumour should the drug be infused to have a
greater effect. Interestingly, ascertaining the optimal dosing schedule is
still a challenge.
2.3. Models of chemoradiotherapy
The paradigm of the interplay between radiation and chemotherapy
was first conceptualized by Steel and Peckham in 1979 [57] and was
later summarized by Seiwert et al. [58]. Two of the main advantages for
the use of chemoradiotherapy, assuming there is no interaction between
the two, is to use cytotoxic drugs that will be used to tackle the disease,
not within the radiation field and to irradiate seclusion sites [57].
In a study by Goldie and Coldman [13], a stochastic model for
alternating radiation and chemotherapy was proposed. Their model was
based on the earliest approaches for combination therapy where they
used three compartments, that is, stem, differentiation and end cells in
modeling tumour growth. Their model incorporated chemo-resistant,
radio-resistant and extra parameters to measure cells with joint resis­
tance. They found that a regimen treatment with three doses of
chemotherapy and radiotherapy acting in an alternating manner is
pow-erful compared to sequential treatment under the same conditions
because it suppresses subpopulation resistant responsible for treatment
failure.
In another study by Beil and Wein [3], a mathematical model was
formulated with the aim of determin-ing the best way to sequence the
three standard therapies for cancer, that is, radiotherapy, chemotherapy
and surgery. Differential equations were used to model the growth of
tumours and its metastases with an underlying assumption that primary
and metastatic tumours behave identically. The model aimed to show
which combina-tion achieved higher curative probability: surgery fol­
lowed by chemotherapy and radiotherapy (SCR), Surgery followed by
radiotherapy and chemotherapy (SRC) and surgery followed by radio­
therapy then chemotherapy and radiotherapy (SRCR) given that the
primary tumour is big enough or that the metastatic cancer population
outnumbers the primary tumour. Their model included metastatic can­
cer originating from dormant cells and angiogenesis, which eventually
led to the interactions between primary tumour compartment and
metastatic compartment and thus the emergence of optimality
sequences.
Ergun et al. [59] used a similar approach to find optimal scheduling
and doses of angiogenic inhibitors and radiotherapy that maximize the
elimination of a primary tumour. The model was analyzed using two
compartments of tumour cells and vascular endothelial cells and the
damage by radiation was modelled in accordance with the
linear-quadratic which widely determines fractionation schedules in
radiobiology society. It was shown that optimal amalgamation treat­
ment regimen happens when antiangiogenetic therapy is constantly
increased in order to maintain an optimal tumour endothelial cells ratio,
and the fraction sizes of radiation keep changing with treatment so as to
maximize the probability of tumour control. However, the model was
solely concerned with the primary tumour and did not include the
antimetastatic effect of antiangiogenic treatment, which might inhibit
metastatic growth during radiotherapy if it is applied methodically.
Further, the model did not take into consideration the fact that radiation
may cause vascular endothelial growth factor (VEGF) expression, which
attenuates the destruction of endothelial cells by radiotherapy.
In a study by Salari et al. [60] it was shown that the addition of
chemotherapy drug and its mode of action makes optimal fractionation
regimen change for concurrent chemoradiotherapy. They contend that
the introduc-tion of chemotherapeutic drugs containing cytotoxic effects
only does not hamper optimal fractionation regimen when there is no
concurrent chemotherapy whereas a radiosensitization agent has the
capacity to tamper with the optimality choice of fraction size. They also
show that the emergence of optimal non-uniform fractionation may
4
Informatics in Medicine Unlocked 23 (2021) 100534
result if a drug exhibits both radiosensitization and independent cyto­
toxic properties [61].
Ghaffari et al. [62] proposed an ODE model that considered
chemotherapy and radiotherapy for metastatic cancer. They examined
the interplay between immune and cancer cells, with chemotherapy
regarded as a predator on both normal and cancer cells. The model
aimed at understanding the specific system dynamics and as well guide
the creation of combination therapies. It was shown that a decaying
dosage protocol is weaker compared to a constant dosage protocol in
eliminating small metastases at a given time.
Marcu et al. [63] simulated ciplatin-radiotherapy treatment with
emphasis on time sequencing and combined treatment scheduling of the
drug and radiation. They examined different time courses of cisplatin
and radio-therapy for advanced head and neck cancer. It was agreed that
to discover the true synergistic effect between cisplatin and radio­
therapy reported in the literature, they would have to incorporate a
proper radiosensitizing term. They further indicated that by adminis­
tering cisplatin on a weekly basis would result to enhanced ther-apeutic
effects compared to daily doses and that giving cisplatin in cross prox­
imity in time to the radiation fractions would definitely result to
maximal combined effect.
Barazzuol et al. [64] formulated a mathematical model of radio­
therapy in the treatment of glioblastoma where they used a linear
quadratic model and investigated the additional effect of temozolomide
in two simpli-fied scenarios. Their main aim was to address the question
of whether the benefit of temozolomide is drawn from its radio­
sensitizing properties in the concurrent phase or from the additional cell
kill of the adjuvant phase. They concluded that the model presuming
only radiosensitization fits the observed survival curves more closely
than assuming independent cytotoxicity.
Powathil et al. [65] presented a multiscale mathematical model
which relied on cellular automata to examine the spatial distribution
and microenvironment of a cell population during treatment. A log
cell-kill model was used to estimate chemotherapeutic effects whereas
the linear-quadratic model was used in the case of radiation effects. They
applied their model to show a clinical application to two known and one
hypothetical treatment regimen for esophageal cancer, where they
concluded that their proposed alternative regimen was more effective.
Several mathematical models allow broad disruptions regarding the
concepts of pure radio-sensitization and independent chemotherapy cell
kill [66–74].
2.4. Models of radioimmunotherapy
Successful implementation of radioimmunotherapy for cancer treat-
ment has proven to be significantly more difficult than was expected
initially [75–79]. This is, in part, as a result of highly diverse, complex
and interrelated biological and physical factors that are considered
when designing a successful protocol [75]. By utilizing mathematical
models incorporating plasma antibody pharmacokinetics, extravasation
and interstitial transport and the antibody-antigen interaction, Fujimori
et al. and van Osdol et al. [80–83] investigated the connection between a
variety of tumour and antibody specific parameters and the microscopic
distribution of antibody and absorbed dose within a tumour [75].
O’Donoghue et al. [14] proposed a Uniform tumour Dosimetry
mathematical model that compared single-dose and fractionated radi­
oimmunotherapy. The model compared large single administrations
(LSAs) with rapid fractionation (RF) administered in small quantities
within a short period of the time interval. It was indicated that for ho­
mogeneous absorbed dose distributions throughout tumours, LSAs are
predicted to give a greater possibility of tumour cure than rapidly
fractionated treatments of the same marrow toxicity. If dose distribu­
tions are heterogeneous, RF may have a therapeutic advantage,
depending on how tumour uptake varies from one fraction to another.
Kumar [84] proposed a mathematical model for radio­
immunotherapy where they formulated radiotherapy dose distributions
J. Malinzi et al.
with regard to optimizing tumour cure probability (TCP) where the rate
of dose distribution was assumed to be high enough to enable the de­
livery of dose distribution instantaneously. Numerical results showed
that the tumour cells density and dose distribution are not sensitive to
other various functional forms of tumour parameters. The results of this
study revealed that the immune response plays a critical role during
cancer treatment.
Flux et al. [85] described a dosimetry 3-D mathematical model that
quantified the absorption of dose dis-tribution as a result of adminis­
tration of an intralesional radiolabeled monoclonal antibody which
allowed the distribution of spatial and temporal heterogeneity of
radionuclide without a calibration scan. The model was tested by using a
set of registered patient data where dose profiles and histograms of the
dose volume were produced. It was established that 3-D dose distribu­
tion was significantly non-uniform. Initially, intralesional infusion re­
sults suggested that the model offered a means of determining how the
absorbed dose was distributed within a tumour.
Serre et al. [86] suggested a discrete-time pharmacodynamic math­
ematical model of the amalgamation of radiotherapy and immune
checkpoint inhibitors: PD1-PDL1 axis and the CTLA4. The model shows
how a growing tumour triggers and inhibits tumour-immune responses
and describes the effects of irradiation. The model’s ability to predict
pharmacodynamic endpoints was justified in retrospective by exam­
ining that it could explain clearly data obtained from experimental
studies, which took into consideration the combination of radiotherapy
and immune checkpoint inhibitors. They concluded that different de­
signs in silico could be com-pared by simulating Kaplan-Meier curves
and mathematical tools could be used to partially automate optimized
protocols.
More work is needed to focus on integrating the developed mathe­
matical methods into clinical practice that will heavily deal with man­
aging drug and radiation-induced toxicities during the process of
optimization. Fur-ther combining radioimmunotherapy with other
treatment regimens including targeted agents and metronomic chemo­
therapy is a subject to be explored in future research. More still, dose
effect relationships should be exam-ined and a pharmacokinetic model
integrated to clearly shade light on the relationship between the
experimental results and the actual dosing protocols.
2.5. Models of immunovirotherapy
Viruses have been engineered genetically to promote the expression
of pro-inflammatory cytokines and co-stimulatory molecules which,
upon release, instigates a targeted immune response on the tumour
[87–91]. Several treatment approaches combining viral oncolysis with
production of immunostimulatory molecules and dendritic cell (DC)
injections are currently under construction with the aim of bettering
treatment outcomes [91–93]. Several studies have been carried out to
investigate the outcome of immunovirotherapy on myeloma [94], to
establish how initial conditions hamper the efficacy of OV therapy on
melanoma [95] and determine factors inhibiting and enhancing the
transmission of oncolytic viruses via a tumour site [96]. Other recent
studies focusing on the interaction of immune cells with DC vaccines
include [97–99].
Lai and Friedman [16] considered a combination therapy where one
drug was used as a vaccine to activate dendritic cells so as to instigate
more T cells for the purpose of invading the tumour while the other drug
is a checkpoint inhibitor which suppresses cancer cells. They developed
a mathematical model in the form of PDEs to address the question of
whether combining treatment of two drugs administered at certain
levels is better than using a treatment of one drug with almost twice the
dosage level. The model describes the interaction of dendritic cells,
cancer cells, cytokines IL-I2 and IL-2, CD8+ cells (T8), GM–CFC–secret­
ing cancer cells (GVAX) and anti-PD-1. The concept of synergy between
drugs was introduced and a synergy map was developed suggesting the
proportion in which drugs should be administered so as to realize the
5
Informatics in Medicine Unlocked 23 (2021) 100534
maximum tumour volume to be reduced under the restriction of a
maximum tolerated dose. Results showed that combining GVAX and
anti-PD-1 in suitable quantities could consequentially reduce the growth
of a tumour. More studies need to be carried out to explore the effects of
treatment with GM–CFC–secreting vaccine (GVAX) and anti-PD- 1 drug.
To date, oncolytic virotherapy still possess an exercise in population
dynamics where the interplays between the viruses, tumour cells and the
immune system components play a major role in determining treatment
out-comes [17,50,94,95,100–105]. Several mathematical models have
been formulated to describe the outcome of such interactions [17,95,
100–105]. A study that addresses the combined effects of viral oncolysis
and T-cell- mediated oncolysis was carried out in Ref. [15] wherein a
mathematical model of virotherapy that induces release of cytokine
IL-12 and co-stimulatory molecule 4-1BB ligand was developed. It was
found that whereas viral oncolysis is important in reducing the tumour
burden, increased stimulation of cytotoxic T cells leads to a short-term
reduction in tumour size, but a faster relapse. In addition, it was
found that amalgamations of specialist viruses expressing either IL-12 or
4-1BBL might initially act more potently against tumours than a general
virus that simultaneously expresses both, but the advantage is likely not
large enough to replace treatment using the generalist virus.
A mathematical model incorporating cytokine and co-stimulatory
molecule expressing OVs in combination with DC injections was stud­
ied by Wares et al. [91]. The aim of this study was to investigate the
effect of varying doses of OV and DC injections during tumour treat­
ment. It was shown using simulations that treatment of a tumour with
immunostimulatory OVs first followed by a sequence of DC injections is
more effective than alternating OV and DC injections. It was concluded
that the efficacy of a dosing strategy depends on the ordering of onco­
lytic adenoviruses and dendritic cells, temporal tumour spacing and
dosages chosen. This model was later extended by Gevertz and Wares
[106] in which used different techniques such as information criteria
analyses, sensitivity analyses and a parameter sloppiness analysis and
discovered that it is possible to reduce their model to one variable and
three parameters and still be fitted to the data. They argued that
reduction of the model to minimal form allows for tractability of the
system in the face of parametric uncertainty.
Timalsina et al. [107] proposed a PDE model with the aim of
studying tumour virotherapy and mediated immunity. The model
incorporated adaptive and innate immunity responses with interactions
in tumour cells, OVs and immune system in a territory containing a
boundary in motion. They assumed that all tumour cells exhibit homo­
geneity and viruses follow a diffusion process. They used the results
obtained to examine tumour development and provide insight into
crucial aspects of virotherapy such as dependence of the efficacy on vital
parameters and the delay in the adaptive immunity.
Jenner et al. [108] suggested a mathematical model with a view of
reproducing the results for tumour growth synonymous to results pro­
duced experimentally during treatment with an adenovirus virus. The
model was fitted to data using parameters from the literature in order to
reduce the degrees of freedom of the model. Optimization of the model
was done using a least-squares non-linear fitting algorithm ‘lsqnonlin’ in
MATLAB. The model indicates that decreasing APC stimulation and
raising helper T cell stimulation is likely to improve treatment. None­
theless, it is still imperative to establish the range of these parameters
when the absolute elimination of tumour cells occur.
Mahasa et al. [109] presented a mathematical model that elucidated
the interplays between oncolytic viruses, tumour cells, normal cells and
the antitumoural and antiviral immune responses. The model consisted
of delay differential equations (DDEs) with one discrete delay indicating
the time required to trigger a tumour-specific immune response. The
model aimed to predict effects of antitumoural and antiviral immune
responses in the presence of oncolytic viruses, the response of the
tumour to the oncolytic viral infections and the oncolytic virus
tumour-specificity responsible for maximizing the reduction of the
tumour while concurrently ensuring that the toxicity on the normal
J. Malinzi et al.
tissue encircling tumour cells are minimized.
Cassidy and Humphries [110] formulated and analyzed a mathe­
matical model of tumour growth under on-colytic virotherapy and
immunotherapy. The model explicitly included heterogeneity in tumour
propagation speed by considering the cell cycle duration as a random
variable. The interrelationship between the estimated number of cells
surviving in the cell cycle and tumour eradication was established by the
help of linear stability analysis. The results of their analysis showed that
an increase in immune recruitment acts synergistically to lyse tumour
cells during oncolytic virotherapy. Nonetheless, they noted that their
model was limited by the fact that it oversimplified tumour-immune
interplays.
2.6. Models of radiovirotherapy
The interaction dynamics of oncolytic viruses with tumour cells and
im-mune responses are highly intricate [17,52,53,102,111]. Deter­
mining the outcome of radiovirotherapy is de-pendent on understanding
the complex interactions between the various components involved
during treatment with radiovirotherapy [17]. Thus, modeling the ki­
netics of virotherapy could act as an impetus for the devel-opment of
improved therapeutic protocols as well as gaining further insight on the
outcome of such therapies [17,52,53,102,111].
A study by Dingli et al. [17] focused on attenuated strains of measles
virus that highly infect tumour cells because of their high expression of
CD46, a property that gives viruses an advantage to attach and enter into
the target cells. They proposed and analyzed a mathematical model that
took into account population dynamics that encapsulated crucial com­
ponents of radiovirotherapy. Analysis of the model included deter­
mining and establishing existence of corresponding equilibria related to
complete cure, partial cure, and treatment failure. Through numerical
simulations, they analyzed the influence of factors, in the form of pa­
rameters, that determine the outcome of radiovirotherapy treatment.
Morestill, they evaluated relevant therapeutic scenarios for effective
radiovirotherapy. The study showed that radiotherapy is more effective
when used synergistically with oncolytic virotherapy.
Tao and Guo [112] developed a PDE model describing cancer radi­
ovirotherapy, which is a generalization of the existing ODE models given
in Refs. [17,51]. The model was developed as a moving boundary
problem. Global existence and uniqueness of solutions to the model was
proved, and a new explicit parameter condition corre-sponding to
treatment success was determined. By modeling combined action of
virotherapy and radiotherapy, they aimed to design the possible optimal
therapy strategies. Numerical experiments verified that
radiovirother-apy is more effective compared to monotherapy. To
explore possible optimal therapy strategies, a number of numerical
simulations were further carried out. The results suggested that there is
an optimal timing of radio-iodine administration and an optimal dose of
the radioactive iodide, which needs to be tested by experimental data.
2.7. Mathematical models of targeted therapies
Targeted therapies involve the use of novel cancer drugs that interact
with specific molecular targets that are pertinent to cancer progression
[113,114]. Examples of targeted therapies include; protein-kinase in­
hibitors: Imatinib (Glivec®)-used for the treatment of leukemia, Dasa­
tanib (Sprycel®)-used for treating prostate cancer, and Temsirolimus
(Torisel®)-used for treating several cancer types, monoclonal anti­
bodies: Bevacizumab (Avastin ®)-for treating colon cancer, and pro­
teasome in-hibitors: Bortezomib (Valcade®)-for treating myeloma and
leukemia [114]. Dozens of other targeted therapies for treating different
types of cancer do exist and have already been approved for use whereas
others are under-going clinical trials [115]. Single targeted therapy
drugs with short-lived treatment effects often fail [116]. The alternative
strategy previously proposed is to synergistically use a combination of
targeted therapy drugs which employ different pathways, or to use
6
Informatics in Medicine Unlocked 23 (2021) 100534
targeted therapies together with other treatment methods. For example,
with immunotherapy; a great number of preclinical and clinical trials
have revealed that there is a synergistic antitumor effect with the use of
targeted therapy with immunotherapy [116].
The development or improvement of these targeted therapies re­
quires a molecular understanding of cancer pathogenesis and a char­
acterization of the interplay between tumour cells and therapeutic
agents. Nonetheless, not many attempts have been made to mathemat­
ically investigate the dynamic relationships between cancerous cells and
targeted therapeutic agents; moreover only a few mathematical studies
exist, thus far, for the prediction of the outcome of combination treat­
ment with targeted therapies. A comprehensive review of mathematical
models for targeted therapy can be found in Abbott and Michor [114].
We complement these with a review of some mathematical studies on
targeted therapies and combination treatments involving targeted
therapies.
Green et al. [117] proposed a compartmental mathematical model
for antibody-targeted therapy of colorectal cancer, using clinical trial
data from bio-distribution of antibodies against carcinoembryonic an­
tigen. The study indicated the most significant parameters that deter­
mine antibody localization are the affinity for the antibody, the flow of
the antibody through the tumour and the rate of elimination of the
antibody from the tumour.
Shen et al. [118] developed a mathematical biphasic model that
accurately described the cell-drug response based on an analysis of
targeted inhibition of colorectal cancer cell lines. Their model used three
kinetic param-eters: ratio of target-specific inhibition, F1, potency of
target-specific inhibition, Kd1, and potency of off-target toxicity, Kd2 to
describe the drug response where the determination of these kinetic
parameters are valuable in predicting effective combination targeted
therapy for multi-driver cancer cells. The model and the mechanistic
insights give a new mechanistic perspective and mathematical tool for
predicting effective combination targeted blockades against
multi-driver cancer cells.
Stochastic differential equations (SDEs) have been employed to
model the stochastic evolution of resistance of therapeutic drugs. Sun
et al. [119] proposed a stochastic model by using a set of SDEs to
examine the dynamics of drug sensitive cells, drug-resistant cells and
new metastatic cells. They used clinical data to vali-date their model and
predicted distinct patterns of drug dose-dependent synergy for two
different sets of drug combinations in the treatment of melanoma. This
approach was expected to enhance the study of effective and robust
cancer drugs.
Kozlowska et al. [120] presented a comprehensive stochastic math­
ematical model and simulator approach that describes platinum resis­
tance and standard of care therapy in high-grade serous ovarian cancer
(HGSOC). Their study used post treatment clinical data, including
18F-FDG-PET/CT images, to accurately predict the model parameters
and simulate ‘virtual patients with HGSOC’. The model revealed that the
treatment responses exhibited by these virtual patients was in line with
those of real-life patients with HGSOC. They utilized their approach to
evaluate survival benefits of combination therapies containing up to six
drugs targeting platinum resistance mechanisms where they indicated
that combining standard of care with a drug specifically targeting the
most dominant resistance sub-population resulted in a significant sur­
vival benefit.
Jarrett et al. [121] constructed a mathematical model that evaluated
the combination of trastuzumab-paclitaxel therapies for drug interaction
relations on the basis of order, timing and quantity of the drug dosage.
Their model was based on time-resolved microscopy data that captured
changes in vitro cell confluence in response to the combination of
paclitaxel and trastazumub treatment. The model showed increased
synergy for treatment reg-imens where trastuzumab was administered
before paclitaxel and indicated that trastuzumab accelerates the cyto­
toxicity of paclitaxel.
Bozic et al. [115] proposed a mathematical model to predict the
J. Malinzi et al.
effect of combining targeted therapies. Their model was based on data
obtained from 20 melanoma patients. Their model simulation results
revealed that combination treatment with two targeted therapy drugs is
far more effective than using a single drug; and those drugs should
simultaneously be administered for optimal results.
Owen et al. [122] constructed a spatiotemporal mathematical model
to determine the outcome of combining macrophage-based hypo­
xia-targeted gene therapy with chemotherapy. Model simulation results
showed that combining conventional drugs and macrophage based tar­
geted therapies would work synergistically and result to better
anti-cancer effects than the individual effects from each of the therapies.
Arciero et al. [33] presented a mathematical model where they
considered a novel treatment strategy known as small interfering RNA
(siRNA) therapy in which the treatment suppresses TGF-β production by
targeting the mRNA codes for TGF-β, and thus reducing the presence and
effect of TGF-β in tumour cells. The model predicts conditions within
which siRNA treatment can be successful in transformation of TGF-β
producing tumours to either non-producing or producing a small value
of TGF-β tumours.
Tang et al. [123] introduced a novel model, called TIMMA (Target
Inhibition inference using Maximization and Minimization Averaging)
with the aim of demonstrating its feasibility in systematic investigation
of the model predictions using kinome-wide single and pairwise siRNA
knock-down experiments. The model used functional data on drugs for
its construction and target combination predictions. The validation of
TIMMA re-sults using systematic siRNA-mediated silencing of the
selected targets and their pairwise combinations showed an increased
ability to identify both druggable kinase targets and synergistic in­
teractions. The model was ap-plied to case studies in MCF-7 and
MDA-MB-231 breast cancer and BxPC-3 pancreatic cancer cells where it
was confirmed that TIMMA-predicted kinase targets are essential for
tumour survival, either individually or in combination. Further, it was
shown that the construction of model algorithm resulted in significantly
better predictive accuracy and computational efficiency in comparison
with an existing algorithmic solution.
In another study, Tang et al. [124] applied network pharmacology
model to predict synergistic drug combi-nations. Kinome-wide
drug-target profiles and gene expression data were used to pinpoint a
synergistic target interaction between Aurora B and ZAK kinase inhibi­
tion that resulted to an enhanced growth inhibition and cytotoxicity.
They employed stochastic simulation algorithm (SSA) to model
signaling pathways, implement and understand the mechanisms of ac­
tion of the identified target interactions. An MDA-MB-231 signaling
network was constructed using dynamic modeling approach to simulate
the effect of perturbing the genes of interests on the cell viability. The
observed similarity of the model predictions to the experimental ob­
servations indicates that the constructed signaling network consists of
key protein-protein interactions that may mediate the synergistic and
antagonistic relationship of Aurora B with ZAK and CSF1R, respectively.
Araujo et al. [125] presented a mathematical model to investigate
combination therapy in which multiple nodes in a signaling cascade
pathway are simultaneously targeted with specific inhibitors. The model
sug-gested that the attenuation of biochemical signals is significantly
enhanced when several upstream processes are inhibited, and that this
weakening is most pronounced in signals downstream of serially con­
nected targets.
Komarova and Dominik [126] developed a mathematical framework
to examine the principles underlying the emergence and prevention of
resistance in the treatment of cancers with targeted drugs. They
considered a stochastic dynamical system which took into account
turnover rate of tumour cells and the rate of generating resistant mu­
tants.The model was applied chronic myeloid leukemia where it was
suggested that combining three targeted drugs with different specific­
ities is likely to overcome the problem of resistance.
Charusanti et al. [127] presented a mathematical model describing
several signaling events in chronic myeloid leukemia (CML) cells. Their
7
Informatics in Medicine Unlocked 23 (2021) 100534
study examined dynamic effects of the drug STI-571 (Glivec) on the
autophosphorylation of the BCR-ABL oncoprotein and subsequent
signaling through the Crkl pathway, and they used simulations to pre­
dict a minimal concentration for drug effectiveness. The model was
fitted to mouse Bcr-Abl autophosphorylation data where it is revealed
that cellular drug clearance mechanisms such as drug efflux significantly
reduces the effectiveness of Glivec in blast crisis cells. Additionally, it is
stipulated that these resistance mechanisms might be present from the
onset of disease.
Other mathematical models of combination cancer therapy include
[75,104,128–160]. Bajzer et al. [104] modelled cancer virotherapy with
recombinant measles viruses where the interactions of the tumour and
virus were based on the already established biology. Hadjiandreou and
Mitsis [158] developed a mathematical model that used Gompertz
growth law and pharmacokinetic-pharmacodynamic approach to model
the effects of drugs on tumour progression and designed optimal ther­
apeutic patterns. Su et al. [159] formulated a mathematical model of
tumour therapy with oncolytic viruses and MEK inhibitor.
3. Discussion and conclusion
Combination cancer therapy strategies have fast become the estab­
lished method for oncologists to treat cancer and has the potential to
counter treatment side effects and resistance to drugs. There is much
concern over the toxicity and effectiveness of combining therapeutic
strategies. Ideally, there are several factors that necessitate the selection
of a certain cancer combination therapy strategy such as im-provement
in the quality of life of the patient, nonoverlapping safety profiles,
reduction of drug resistance and preclinical evidence of synergy be­
tween the interacting therapeutic agent. However, developing novel
combi-nations of therapeutic strategies is time-consuming and expen­
sive and depends on a myriad of factors unique to each cancer. More­
over, there are still several questions that need to be answered
pertaining to combination cancer therapy. These include determining
the optimal dosage combinations and investigating the side effects of
using the different combination treatments. Mathematics can help to
answer such questions by allowing researchers to model therapeutic
options (including new, repurposed and old drugs) and predict their
treatment and toxicity outcomes before embarking on lengthy and
expensive trials. Scientists use a range of methods, tools and software to
analyze mathematical models; the choice depends on the problem being
solved and the model type. Mathematical models in the form of differ­
ential equations are normally analyzed using qualita-tive methods; to
investigate the long-term characteristics of the model solutions
[161–163]. Model analysis is usually complemented with numerical and
computational simulations using several methods implemented by a
number of programing languages and software [164,165]. Some ex­
amples of tools and software that are used in analyzing mathematical
models reviewed in this paper and tools that can be used for users to
upload and analyze input data are summarized in Table 2. We as well
provide some existing databases for combinations of cancer therapy in
Table 3.
Complete drug development takes more than 10 years with huge
costs ranging from USD 12 million to USD 12 billion. Moreover, only a
small fraction of drugs that enter clinical trials are eventually approved
[166]. This makes it imperative to devise methods of reducing the cost of
drug development and one such way is through mathematical modeling.
Drug developers, researchers, clinicians and other bio-scientists can use
model simulations and results from mathematical analyses to assimilate
complex mechanisms of cancer therapy dynamics thus leading the
development of more effective treatments.
In Section 2, we reviewed mathematical models for a number of
combination cancer therapy strategies de-veloped thus far. In doing so,
we highlighted the important discoveries that have been made via
mathematical modeling and identified the gaps that still need to be fil­
led. All the mathematical studies, reviewed here, indi-cate that
J. Malinzi et al.
Table 1
A summary of merits and demerits associated with the use of certain forms of combination therapies.
Therapies Merits
Chemoimmunotherapy Application of low-dose chemotherapy with immune- stimulating
vaccination significantly lengthens sur- vival both in a prophy­
lactic and therapeutic setting.
Improved progression-free survival and the overall survival of
physically fit patients with previously un- treated symptomatic
chronic lymphocytic leukemia.
Chemovirotherapy Several studies have shown that combining chemother- apy with
oncolytic viruses re- sults to enhanced therapeutic effects.
Benefits include significant improvement of overall sur- vival,
enhanced efficacy and decreased tumour volume.
Chemoradiotherapy Cisplatin chemoradiotherapy (CRT) delays disease recur- rence
and boosts survival in contrast to radiotherapy (RT) alone in
women with stage IIIB squamous cell carci- noma of the cervix.
CCRT has a survival advan- tage compared with RT alone in
patients between 65 and 72 years.
Enhances local tumour con- trol in the treatment of cervi- cal
cancer.
Concurrent chemoradiother- apy has been shown to reduce the
risk of cancer recurring by 50% in patients having ad- vanced stage
disease regional spread.
Radioimmunotherapy Radioimmunotherapy treat- ment regimen has been used in the
treatment of haemato- logical malignancies with a fractionated
dosing regimen achieving remarkable regres- sion of bulky masses
with non-Hodgkin lymphoma.
Studies on other aggressive non-Hodgkin lymphoma have indi­
cated promising efficacy of anti-CD20 ra- dioimmunotherapy and
anti-CD22 radioimmunother- apy reduced hematologic toxicity,
improved survival benefits and improved efficacy.
Immunovirotherapy Selective virus replication in rapidly proliferating cells.
Directs tumour microenvi- ronment towards immune re- sponse.
Enhances survival.
OVs produces pathogens that are responsible for instigating as well
as enhanc- ing anti-tumour immunity. Combining OVs with im-
munostimulatory cytokines initiates an additional anti- tumour­
immune response with apoptosis of tumour cells occurring due to
suppressive properties of immune cells.
Radiovirotherapy In the treatment of prostate cancer, it is reported that MV- NIS
results to significant tu- mour regression as well as highly
significant prolonga- tion of survival in animals.
In addition, administration of iodine-131 enhances the anti-
tumour effect of MV-NIS vi- rotherapy.
Several HSV1 mutants act synergistically with radiation therapy
83–85, although in some experiments the impact of radiation is
essentially ad- ditive
Targeted therapies Foundation of precision medicine.
Drug combinations targeting different pathways can be used.
Attenuation of biochemical signals is highly enhanced. Imatinib
(Glivec®), the first targeted therapy to be ap- proved by the FDA,
has proven to be successful in the treatment of chronic myeloid
leukemia.
combining cancer treatments strategies leads to enhanced therapeutic
outcomes. Mathematical models of chemoimmunotherapy showed that
with the use of this strategy for the treatment of chronic lymphocytic
leukemia, the patient’s immune system would be boosted in addition to
8
Informatics in Medicine Unlocked 23 (2021) 100534
Demerits References
Although chemoim- munotherapy in animal models is well [10,38,61,
established, it is extremely hard to transfer this knowledge into 172–177]
clinical practice as there are sev- eral hurdles that must be
overcome.
Side effects include sec- ondary malignancies, myelo- and
immunosuppression which are likely to be the onset of myeloid
malig- nancies (acute myeloid leukemia and myelodysplas- tic
syndrome), hematological non-hematological toxicities.
Although this combination has proved promising, it is however [11,12,20,22,
limited by the side effects such as viremia and cardiac toxicity. 167,178–181]
Moderate to acute gastroin- testinal tract effects in the form of [3,13,20,57,
bleeding proctitis due to telangiectasia and ulcera- tion, which 182–185]
has been poorly reported in the literature.
Increased incidences of tu- mour recurrence, gastroin- testinal
and severe hemato- logic complications.
Has low overall survival rate and late toxicity
Limited by hematologic toxicity, secondary cancers and of [14,168,186,
myelodysplastic syndrome. 187]
There is considerable ev- idence indicating that the overall risk
for Radioim- munotherapy is no higher than chemotherapy.
Other demerits include the development of drug resis- tance as
well as being costly.
Convincing of the oncohe- matologist community to in- corporate
radioimmunother- apy would require designing of randomized
clinical tri- als and stratification of pa- tients for response to ra-
dioimmunotherapy. Clinical efficacy of radioimmunother- apy
applications in solid tu- mours remain limited.
Genetic mutations allow for the development of resis- tance of [87–91,106,
tumour cells to OVs. Moreover tumour cells can learn to avoid 108]
adaptive immu- nity via immunoediting.
Viruses can trigger anti-viral immune responses.
Side effects including gas- trointestinal toxicity and weight loss. [17,20,
179–181,
188–194]
The efficacy of radiovirother- apy can be limited by sev- eral
factors in thyroid tumour treatment.
The biodistribution of the virus, its propagation in the tumour
through numerous replication cycles and the strength of the
sodium/iodide symporter expression can be inadequate or
hetero- geneous and thus it may affect an effective therapeutic
response.
Side effects include gastroin- testinal toxicity and weight loss. [113,114,116,
125]
Cancer cells can eventually become resistant to targeted
therapies.
Drugs for some targets are hard to develop.
reducing the tumour volume. This ap-proach leads to an increase in the
number of T-cells that are activated when chemotherapy and immuno­
therapy are combined compared to a single therapy such as chemo­
therapy [10]. Further, ligand transduced cells can trigger enough
J. Malinzi et al. Informatics in Medicine Unlocked 23 (2021) 100534

Table 2 Table 2 (continued )

Examples of existing software for implementing mathematical models reviewed Software Packages Functionality References
in this paper.
to automatically quan-
Software Packages Functionality References tify phenomena such as
Matlab Contains several Popular system for data [195] synergism and
differential equa- tion visual- ization and inhibition.
solvers, for example, carrying out numeri- TIMMA-R R packages. Predicts the effects of [208]
ode 23 and pdepe and cal simulations of drug combi- nations
tool boxes, for ex- differential equa- tions. based on their binary
ample, jLab; used for drug- target
data analysis interactions and single-
Maple Differential equation Powerful software for [196] drug sensitivity profiles
packages like dsolve and symbolic computation in a given cancer
PDEtools. and qualitative analy- sample.
sis. Synergyfinder Web application. Allows for pre- [209]
Mathematica Contains functions such Powerful symbolic [197] processing, analyz- ing
as Dsolve and NDsolve computation; offers and visualising pairwise
numerical evaluation, drug combinations in
opti- mization and an interactive man- ner.
visualization of a very Combenefit Enables for the analysis, [210]
wide range of quantifica- tion and
numerical functions. visualization of drug
Python Contains packages like Open source software [198] com- bination effects in
Scilab; a numerical and a high level, terms of synergy and/or
computational package numerically oriented antagonism.
that can be used for pro- graming language. URDME Simulates stochastic [211]
handling ODEs and reaction- diffusion
PDEs. processes on arbitrary
SageMath Rich in several packages Open source software [199] meshes.
such as de- solve for covering nu- merous COMSOL Solves highly nonlinear [212]
solving differential areas including differential equations.
equa- tions. differential calculus,
algebra, numerical
analy- sis, calculus and immune response to put tumour growth under control, whereas
statistics. control-transduced tumour cells evade immunity [28]. However, none
Maxima desolve function for Used for symbolic [200] of these models can yet provide an optimal dosage for the combination
solving ordi- nary computation and
differential equations. solving differential
of chemotherapy and immunotherapy.
equations. Chemovirotherapy has experienced notable successes both clinically
GNU Octave ODE solver; LSODE. Powerful open source [201] and experimentally, however, much is still unknown about it [12]. The
software with built-in principal mechanisms underlying the anti-tumoural effects of many or­
2D/3D plotting and
dinary virotherapies in conjunction with chemotherapeutics still re­
visual- ization tools.
Mathcad Contains functions like Contains equation- [202] mains a mystery. Fundamental questions such as prospective
odesolve and rkfixed. solving pack- ages ramifications of chemotherapeutic agents on viral replication and the
which can be used to immunotherapeutic effects caused by virotherapy are yet to be fully
plot func- tions, solve answered. Another troubling problem is that of fully distinguishing
blocks of equations,
perform curve-fitting
chemotherapy-related toxicity from that of virotherapy induced toxicity
operations as well as [167]. Designing optimal scheduling of chemovirotherapy treatment
solving differential attracts questions in an attempt to decipher the right dose combination,
equa- tions using deciding the most effectual method of drug infusion and the critical
iterative methods.
treatment characteristics [12]. Furthermore, fig-uring out the potential
Julia Contains solvers like Open source software [203]
OrdinaryDif-fEq.jl and designed for high implications of chemotherapeutics on viral reproduction and/or the
LSODA.jl. performance immunotherapeutic effects possibly caused by virotherapeutics is never
computing; can be used easy [167]. Combinations related to chemotherapy produce com­
for determining pounded toxicity and bone-marrow suppression and therefore their
numerical simulations
stochastic ODEs and
potential remains to be seen. Despite a great deal of clinical research in
PDEs among others. virotherapy, not many mathematical models of chemovirotherapy yet
R Contains ODE solvers Open source software [204] exist. Mathematical modeling has shown enhancement of chemothera­
like deSolve. designed mainly for peutic drugs with oncolytic viruses leads to the depletion of all tumour
statistical computing
cells from a patient’s body tissue in less than a month [11,12,22]. This
and graphics; can be
used for carrying out time frame, nonetheless, leads to questions relating to toxicity. It is
numerical simulations therefore important for future mathematical studies to investigate the
of ODEs and PDEs. effect of oncolytic viruses on the tumour microenvironment. Mathe­
COPASI Contains solvers like An open source [205] matical models of immunovirotherapy showed that the influx of im­
Hybrid LSODA/SSA and software that can be
LSODAR. used for solving
mune cells around the tumour coupled with oncolytic viruses acts
mathematical mod-els synergistically to lyse cancer cells. Nonetheless, one important aspect of
of biological processes. oncolytic virotherapy that has not thoroughly been investigated, using
CompuSyn Analysis of [206] mathematical models, is the effect of the viruses on normal body cells.
combination data.
Moreover, viruses can trigger anti-viral immune responses, and this as
CalcuSyn Analyzes combined [207]
drug effects and is able well needs further investigation. Other models have shown radiotherapy
to be more effective when used synergistically with oncolytic virother­
apy [17,112], including estimating optimal dosage combinations [112].

9
J. Malinzi et al. Informatics in Medicine Unlocked 23 (2021) 100534

Table 3 are still open [169].

Existing database for combinations of cancer therapy. Several mathematical models reviewed in this article, for simplicity,
Database Potential applications Website References ignore pertinent factors in cancer dy-namics for example, the pharma­
cogenomics of cancer treatment and tumour-immune interactions yet
DrugComb A data portal for https://drugcomb. [213]
integrative cancer fimm.fi/ these would ultimately influence the experimental outcomes and
drug combinations. mathematical results. It is therefore essential for future studies to
NCI Provides data showing [214] employ more complex mathematical models, considering detailed in­
ALMANAC how well pairs of FDA- formation for example the tumour microenvironment. Furthermore,
approved cancer drugs
kill tumour cells from
future mathematical models should focus on using different tools other
the NCI-60 Human than differential equations to consider scenarios that cannot be captured
Tumour Cell Lines. by ODEs and PDEs. Generally, there is also a dire need for more research
DrugCombDB Comprehensive http://drugcombdb. [215] to focus on consolidating the discovered mathematical insights into
database dedicated to denglab.org/main
clinical practice.
collecting drug
amalgamations from In contrast to the combinations summarized in Table 1, chemo­
various data sources. virotherapy represents a rationally designed combination due to the
CMTTdb Focuses on the key https://www.biosino.or [216] improved overall survival benefits, nonoverlapping safety profiles,
information for g/CMTTdb/ reduction of drug resistance, the impact on the tumour and of note is the
randomized clinical
trials (RCTs), such as
preclinical evidence of synergy between the components. In support of
patient selection this are the mathematical models on chemovirotherapy which have been
criteria, tumour presented in the previous sections. For example, a study by Malinzi [22]
conditions, therapy on mathematical analysis of chemovirotherapy indicated that the
modes (single or
tumour is likely to be cleared from the body in less than a month if both
combined therapy).
CANCROX Provides important http://cancrox.gmb.bio. [217] chemotherapy and virotherapy are used. Further, the study contended
information about br/view/index.php that in the treatment of cancer, oncolytic viruses enhance chemothera­
cancer treatment, drug peutic drugs which is also in agreement with [12]. Nonetheless,
combinations, genes designing an effective combo of viruses and chemothera-peutic drugs
and types of cancer.
will need to have a perfect balance between immune anti-viral and
OncoRx Detecting drug https://onco-informa [218]
interactions with tics.com/ anti-tumour responses.
chemotherapy Whilst some forms of cancer are incurable and treatment is only
regimens. administered to improve the quality of life and/or prolong life, a com­
SynToxProfiler Promotes the https://syntoxprofiler.fi [219]
bination therapy strategy that proves clinically to be of benefit to the
prioritization of drug mm.fi/stp/eerr
combinations based on nKuW9DeMSZ3r patient taking into account the cost incurred will prove highly valuable.
integrated analysis of clWtuLyrBpHbUvN% Today there are many treatment options for patients diagnosed with
synergy, efficacy and 2BWDUAXDxZLUg% cancer than there was a decade or two ago. Some of these treatments can
toxicity profiles. 3D/ produce significant responses that could completely eradicate metasta­
ReDO-DB Focuses exclusively on http://www.redo-pr [220]
static tumours [170]. Nonetheless, these treatments all experience the
the potential use of oject.org/db/
licensed non-cancer same problem, that is, drug resistance. This resistance of cancer cells to
medica- tions as drugs could ultimately lead to cancer relapse or recurrence. Molecular
sources of new cancer alterations may play a pivotal role in either acquired or intrinsic drug
ther- apeutics.
resistance and some of these alterations contributing to these resistance
CancerDR Identification of https://webs.iiitd.edu. [221]
mutations in can-cer in/ragh
include mutation of the drug’s molecular target, changes in the way the
genes (targets) as well ava/cancerdr/#thumb drug interacts with the tumour, broad cellular changes, and changes in
as natural variations in the tumour microenvironment among others [170]. Efforts are being
targets. made in order to understand how cells develop this resistance and more
importantly, how it can be overcome. While everyone knows that
Nonetheless, just like with other combina-tions, it is important that finding a cure to cancer is a hard task, it is evident today that combining
future studies focus on investigating the effect of using both treatments different strategies and minimizing the chances that it recurs, is defi­
on body tissue. Mathematical models of chemoradiotherapy and radio­ nitely the correct path to seek.
immunotherapy have shown that such combinations may lead to better We anticipate our understanding of the mechanisms of cancer to
treatment outcomes compared to a single approach. Whilst some studies ameliorate its burden in the near future, but the success will depend on a
[13,14] have attempted to identify optimal dosage combinations, no multidisciplinary approach across multiple sections of science, including
mathematical study has investigated the toxicity of the combina-tion mathematics, ecology, epidemiology, genetics and immunology. The
despite the adverse effects of radiotherapy alone being well docu­ early euphoria following breakthroughs exploiting combination thera­
mented. Although radioimmunotherapy is a promising treatment strat­ peutic strategies has been tainted by the existence of numerous chal­
egy, questions regarding its long-term safety issues still exist. There still lenges. To this end, we highlight the outstanding open problems that still
exist difficulties in estimating responses based on dosimetry and radio­ exist despite the reported success and numerous efforts in the research
biological models [168]. An understanding of how these treatments can community to combat cancer using combination anti-cancer therapy.
be best administered, how frequent the treatments can be administered Despite the growing expectations, in practice, combination cancer
and how to best integrate them with other agents to increase the overall therapy strategies have demonstrated in-consistent results that can be
response is yet to be developed [168]. In spite of the fact that it is still attributed to the problems incurred when conjoining these molecularly
possible to administer higher doses of radioactivity, it is still not known targeted agents [171]. It is however hard to establish whether the het­
if the ratio of the tumour to that of non-tumour is higher. In addition, erogeneity of response can be ascribed to the complications of tumour
practical questions such as ascertaining the efficacy of the amalgamation biology, poor drug penetration into the tumour or sub­
specifically the abscopal effect, how to sequence the combination to give optimal/uncoordinated exposure to the conjoined agents [171].
the best result and determining the optimal timing of the amalgamation Furthermore, the problem of determining the ratio at which therapeutic
agents should be mixed to produce maximal synergistic and minimal

10
J. Malinzi et al.
anti-cancer effects is unanswered. The problems of how to find the best
drug combinations, as well as how to effectively combine therapy ex
vivo in human cells to avoid doing thousands of clinical trials is still
disturbing.
It will be interesting to know how to effectively combine therapeutic
agents to produce effective results with appropriate sequencing of these
agents. It is likely that combining two or more therapies will probably
increase toxicity if not ideally designed and thus the challenge of opti­
mally constructing these agents to circumvent the adverse side effects to
a tolerable level is of great concern. This is likely to lead to suboptimal
administration of drugs and thus encouraging drug resistance in the
course of the treatment. However, in clinical trials, it has always been
essential to de-escalate the dose until the drugs become least toxic, a
process which is time-consuming and laborious.
In conclusion, while there has been some remarkable achievement
and promise in the utilization of mathe-matical models in the combi­
nation treatment strategies of cancer, there is still a long way left to go.
Since cancer is a complex group of diseases, their mechanisms of
development are poorly understood and thus mathematical models
developed, thus far, are relatively simple. Thus, an ideal and novel
combination therapy selected for cancer treatment should significantly
lyse only cancer cells, increase the overall survival of a patient, reduce
drug resistance, increase the efficacy and show preclinical evidence of
synergy.
Mathematical modeling is making a significant contribution to the
treatment of cancer. Thus given the pressure to find a value-based
healthcare cancer therapy, there is absolutely no qualm that a mathe­
matical approach will be an integral tool in determining the best com­
bination strategy at a reduced cost while improving cancer treatment
outcomes in the years to come. However, despite the promising results
reported in various studies towards cancer treatment, most of them have
not been integrated for routine clinical use.
It is widely acknowledged that defeating cancer is unlikely to happen
from a single therapeutic agent but rather a reasonable amalgamation of
harmonious compatible strategies chosen from diverse therapeutics
includ-ing traditional and modern cancer therapies. Ideally, there is a
dire need to optimize dosing, duration as well as relative scheduling of
the diverse components by considering the potential synergies and the
probable antag-onisms, as well as toxicities. With the numerous possi­
bilities of combinatorial therapies and the outstanding challenge of
figuring out the interplays between complex and intricated pharmaco­
logic processes, we trust that optimization of these combinations in
silico before testing them in patients could be made possible by devel-
oping and analyzing integrated mathematical models.
Finally, we would like to note that this article does not provide an
intricate review of the mathematical equations which constitute the
models that have been discussed. Rather, a comprehensive review of
mathematical models for combination cancer therapy has been carried
out. We envisage that these models shall potentially play an essential
role in deciding on strategies that are viable both medically and
economically with a high likelihood of yielding clinical benefits.
Ethical approval and consent to participate
This study is exempt from ethical approval in all the authors’
affiliations.
Consent to publish
The authors guarantee that this work has not been previously pub­
lished elsewhere.
Data availability
The authors declare that all data supporting the findings of this study
are available within the article. Any other supplementary information
11
Informatics in Medicine Unlocked 23 (2021) 100534
that is not included in the article is available on request from the cor­
responding author.
Funding
No author received funding to specifically carry out this study.
Authors’ contributions
J.M conceptualized the idea and designed the study. K.B.B and J.M
wrote the initial draft of the article. H.A.A. and S.P reviewed and revised
the manuscript and made critical intellectual inputs. H.A.A designed the
figures and finalized the manuscript. All authors proofread and revised
the article before submission.
Declaration of competing interest
The authors declare that they have no known competing financial
interests or personal relationships that could have appeared to influence
the work reported in this paper.
Acknowledgement
H.A.A thanks the South African Medical Research Council (SAMRC)
for a mid-career scientist and Self-initiated research grant; and the South
African National Research Foundation (NRF) for Research Development
Grant (RDG) for rated researchers.
References
[1] World Health Organisation. https://www.who.int/news-room/fact-sheets/deta
il/cancer. Accessed: April 2020.
[2] Hu Q, Sun W, Wang C, Gu Z. Recent advances of cocktail chemotherapy by
combination drug delivery systems. Adv Drug Deliv Rev 2016;98:19–34.
[3] Beil DR, Wein LM. Sequencing surgery, radiotherapy and chemotherapy: insights
from a mathematical analysis. Breast Canc Res Treat 2002;74(3):279–86.
[4] Benzekry S. Contributions in mathematical oncology: when theory meets reality.
PhD thesis. Université de Bordeaux; 2017.
[5] Heath JR, Davis ME. Nanotechnology and cancer. Annu Rev Med 2008;59:
251–65.
[6] Byrne HM. Dissecting cancer through mathematics: from the cell to the animal
model. Nat Rev Canc 2010;10(3):221–30.
[7] Araujo RP, McElwain DLS. A history of the study of solid tumour growth: the
contribution of mathematical modelling. Bull Math Biol 2004;66(5):1039–91.
[8] Barbolosi D, Ciccolini J, Lacarelle F, B, Barlési, André N. Computational
oncology-mathematical modelling of drug regimens for precision medicine. Nat
Rev Clin Oncol 2016;13(4):242.
[9] Kuang Y, Nagy JD, Eikenberry SE. Introduction to mathematical oncology, vol.
59. CRC Press; 2016.
[10] Chappell M, Chelliah V, Cherkaoui M, Derks G, Dumortier T, Evans N,
Ferrarini M, Fornari C, Ghazal P, Guerriero ML, et al. Mathematical modelling for
combinations of immuno-oncology and anti-cancer therapies. In: Quantitative
systems pharmacology UK meeting. Macclesfeld; 2015.
[11] Malinzi J, Eladdadi A, Sibanda P. Modelling the spatiotemporal dynamics of
chemovirotherapy cancer treatment. J Biol Dynam 2017;11(1):244–74.
[12] Malinzi J, Ouifki R, Eladdadi A, Torres DFM, White KA. Enhancement of
chemotherapy using oncolytic virotherapy: mathematical and optimal control
analysis. Math Biosci Eng 2018;15(6):1435.
[13] Goldie JH, Coldman AJ, Ng V, Hopkins HA, Looney WB. A mathematical and
computer-based model of alternating chemotherapy and radiation therapy in
experimental neoplasms 1. In: Treatment modalities in lung cancer, vol. 41.
Karger Publishers; 1988. p. 11–20.
[14] O’Donoghue JA, Sgouros G, Divgi CR, Humm JL. Single-dose versus fractionated
radioimmunotherapy: model compar- isons for uniform tumor dosimetry. J Nucl
Med 2000;41(3):538–47.
[15] Kim PS, Crivelli JJ, Choi I, Yun C, Wares JR. Quantitative impact of
immunomodulation versus oncolysis with cytokine- expressing virus
therapeutics. Math Biosci Eng 2015;12(4):841–58.
[16] Friedman A, Lai X. Combination therapy for cancer with oncolytic virus and
checkpoint inhibitor: a mathematical model. PloS One 2018;13(2):e0192449.
[17] Dingli D, Cascino MD, Josić K, Russell SJ, Bajzer Ž. Mathematical modeling of
cancer radiovirotherapy. Mathematical Biosciences; 2006.
[18] Hsieh MC, Yu HH, Chang WW, Lu CY, Chang CL, Chow JM, Chen SU, Cheng Y,
Wu SY. Adjuvant radiotherapy and chemotherapy improve survival in patients
with pancreatic adenocarcinoma receiving surgery: adjuvant chemotherapy alone
is insufficient in the era of intensity modulation radiation therapy. Cancer
Medicine 2018;7(6):2328–38.
J. Malinzi et al.
[19] Trimble EL, Ungerleider RS, Abrams JA, Kaplan RS, Feigal EG, Smith MA,
Carter CL, Friedman MA. Neoadjuvant therapy in cancer treatment. Cancer 1993;
72(S11):3515–24.
[20] Ottolino-Perry K, Diallo J, Lichty BD, Bell JC, McCart JA. Intelligent design:
combination therapy with oncolytic viruses. Mol Ther 2010;18(2):251–63.
[21] Chan W, Lam K, Siu W, Yuen K. Chemotherapy at end-of-life: an integration of
oncology and palliative team. Support Care Canc 2016;24(3):1421–7.
[22] Malinzi J. Mathematical analysis of a mathematical model of chemovirotherapy:
effect of drug infusion method. Computational and Mathematical Methods in
Medicine 2019;2019.
[23] B Mokhtari R, Homayouni TS, Baluch N, Morgatskaya E, Kumar S, Das B, Yeger H.
Combination therapy in combating cancer. Oncotarget 2017;8(23):38022.
[24] Dry JR, Yang M, Saez-Rodriguez J. Looking beyond the cancer cell for effective
drug combinations. Genome Med 2016;8(1):125.
[25] Doroshow JH, Simon RM. On the design of combination cancer therapy. Cell
2017;171(7):1476–8.
[26] Lopez JS, Banerji U. Combine and conquer: challenges for targeted therapy
combinations in early phase trials. Nat Rev Clin Oncol 2017;14(1):57.
[27] Emens LA. Chemoimmunotherapy. Cancer Journal (Sudbury, Mass.) 2010;16(4):
295.
[28] de Pillis L, Fister KR, Gu W, Collins C, Daub M, Gross D, Moore J, Preskill B.
Mathematical model creation for cancer chemo-immunotherapy. Computational
and Mathematical Methods in Medicine 2009;10(3):165–84.
[29] De Lillo S, Salvatori MC, Bellomo N. Mathematical tools of the kinetic theory of
active particles with some reasoning on the modelling progression and
heterogeneity. Math Comput Model 2007;45(5–6):564–78.
[30] Adam JA. General aspects of modeling tumor growth and immune response. In:
A survey of models for tumor-immune system dynamics. Springer; 1997.
p. 15–87.
[31] Chaplain M, Matzavinos A. Mathematical modelling of spatio-temporal
phenomena in tumour immunology. In: Tutorials in mathematical biosciences III.
Springer; 2006. p. 131–83.
[32] De Boer RJ, Hogeweg P, Dullens HF, De Weger RA, Den Otter W. Macrophage T
lymphocyte interactions in the anti-tumor immune response: a mathematical
model. J Immunol 1985;134(4):2748–58.
[33] Arciero JC, Jackson TL, Kirschner DE. A mathematical model of tumor-immune
evasion and sirna treatment. Discrete Continuous Dyn Syst - Ser B (DCDS-B) 2004;
4(1):39–58.
[34] de Pillis LG, Gu W, Radunskaya AE. Mixed immunotherapy and chemotherapy of
tumors: modeling, applications and biological interpretations. J Theor Biol 2006;
238(4):841–62.
[35] de Pillis LG, Radunskaya AE, Wiseman CL. A validated mathematical model of
cell-mediated immune response to tumor growth. Canc Res 2005;65(17):7950–8.
[36] Isaeva OG, Osipov VA. Different strategies for cancer treatment: mathematical
modelling. Computational and Mathematical Methods in Medicine 2009;10(4):
253–72.
[37] Kirschner D, Panetta JC. Modeling immunotherapy of the tumor–immune
interaction. J Math Biol 1998;37(3):235–52.
[38] Deng L, Liang H, Burnette B, Beckett M, Darga T, Weichselbaum RR, Fu Y.
Irradiation and anti–pd-l1 treatment synergis- tically promote antitumor
immunity in mice. J Clin Invest 2014;124(2):687–95.
[39] de Pillis LG, Radunskaya A. A mathematical model of immune response to tumor
invasion. In: Computational fluid and solid mechanics 2003. Elsevier; 2003.
p. 1661–8.
[40] Rodrigues DS, Mancera PFA, Carvalho T, Gonçalves LF. A mathematical model for
chemoimmunotherapy of chronic lymphocytic leukemia. Appl Math Comput
2019;349:118–33.
[41] Pang L, Shen L, Zhao Z. Mathematical modelling and analysis of the tumor
treatment regimens with pulsed immunotherapy and chemotherapy.
Computational and Mathematical Methods in Medicine 2016;2016.
[42] Vähä-Koskela JV, Heikkilä JE, Hinkkanen AE. Oncolytic viruses in cancer
therapy. Canc Lett 2007;254(2):178–216.
[43] Guo ZS, Thorne SH, Bartlett DL. Oncolytic virotherapy: molecular targets in
tumor-selective replication and carrier cell- mediated delivery of oncolytic
viruses. Biochim Biophys Acta Rev Canc 2008;1785(2):217–31.
[44] Fukuhara H, Ino Y, Todo T. Oncolytic virus therapy: a new era of cancer
treatment at dawn. Canc Sci 2016;107(10):1373–9.
[45] Li S, Tong J, Rahman MM, Shepherd TG, McFadden G. Oncolytic virotherapy for
ovarian cancer. Oncolytic Virotherapy 2012;1(1).
[46] Yu Z, Li S, Brader P, Chen N, Yong AY, Zhang Q, Szalay AA, Fong Y, Wong RJ.
Oncolytic vaccinia therapy of squamous cell carcinoma. Mol Canc 2009;8(1):45.
[47] Nguyen A, Ho L, Wan Y. Chemotherapy and oncolytic virotherapy: advanced
tactics in the war against cancer. Frontiers in Oncology 2014;4:145.
[48] Ungerechts G, Frenzke ME, Yaiw K, Miest T, Johnston PB, Cattaneo R. Mantle cell
lymphoma salvage regimen: synergy between a reprogrammed oncolytic virus
and two chemotherapeutics. Gene Ther 2010;17(12):1506.
[49] Ulasov IV, Sonabend AM, Nandi S, Khramtsov A, Han Y, Lesniak MS. Combination
of adenoviral virotherapy and temo- zolomide chemotherapy eradicates
malignant glioma through autophagic and apoptotic cell death in vivo. Br J Canc
2009;100(7):1154.
[50] Friedman A, Tian JP, Fulci G, Chiocca EA, Wang J. Glioma virotherapy: effects of
innate immune suppression and increased viral replication capacity. Canc Res
2006;66(4):2314–9.
[51] Friedman A, Tao Y. Analysis of a model of a virus that replicates selectively in
tumor cells. J Math Biol 2003;47(5):391–423.
12
Informatics in Medicine Unlocked 23 (2021) 100534
[52] Wein LM, Wu JT, Kirn DH. Validation and analysis of a mathematical model of a
replication-competent oncolytic virus for cancer treatment: implications for virus
design and delivery. Canc Res 2003;63(6):1317–24.
[53] Wu JT, Byrne HM, Kirn DH, Wein LM. Modeling and analysis of a virus that
replicates selectively in tumor cells. Bull Math Biol 2001;63(4):731.
[54] Tao Y, Guo Q. The competitive dynamics between tumor cells, a replication-
competent virus and an immune response. J Math Biol 2005;51(1):37–74.
[55] Eftimie R, Eftimie G. Tumour-associated macrophages and oncolytic
virotherapies: a mathematical investigation into a complex dynamics. Letters in
Biomathematics 2018;5(sup1). S6–S35.
[56] Malinzi J, Sibanda P, Mambili-Mamoboundou H. Analysis of virotherapy in solid
tumor invasion. Journal of Mathematical Biosciences 2015;263:102–10.
[57] Steel GG, Peckham Michael J. Exploitable mechanisms in combined radiotherapy-
chemotherapy: the concept of additivity. Int J Radiat Oncol Biol Phys 1979;5(1):
85–91.
[58] Grassberger C, Paganetti H. Methodologies in the modeling of combined chemo-
radiation treatments. Phys Med Biol 2016;61(21):R344.
[59] Ergun A, Camphausen K, Wein LM. Optimal scheduling of radiotherapy and
angiogenic inhibitors. Bull Math Biol 2003;65(3):407–24.
[60] Salari E, Unkelbach J, Bortfeld T. A mathematical programming approach to the
fractionation problem in chemoradiotherapy. IIE Trans Healthc Syst Eng 2015;5
(2):55–73.
[61] Medina-Echeverz J, Berraondo P. How can chemoimmunotherapy best be used
for the treatment of colon cancer? Immunotherapy 2012;4(12):1787–90.
[62] Ghaffari A, Bahmaie B, Nazari M. A mixed radiotherapy and chemotherapy model
for treatment of cancer with metastasis. Math Methods Appl Sci 2016;39(15):
4603–17.
[63] Marcu L, Bezak E, Olver I. Scheduling cisplatin and radiotherapy in the treatment
of squamous cell carcinomas of the head and neck: a modelling approach. Phys
Med Biol 2006;51(15):3625.
[64] Barazzuol L, Burnet NG, Jena R, Jones B, Jefferies SJ, Kirkby NF. A mathematical
model of brain tumour response to radiotherapy and chemotherapy considering
radiobiological aspects. J Theor Biol 2010;262(3):553–65.
[65] Powathil GG, Gordon KE, Hill LA, Chaplain MAJ. Modelling the effects of cell-
cycle heterogeneity on the response of a solid tumour to chemotherapy: biological
insights from a hybrid multiscale cellular automaton model. J Theor Biol 2012;
308(1–19).
[66] Jones B, Dale RG. Inclusion of molecular biotherapies with radical radiotherapy:
modeling of combined modality treatment schedules. Int J Radiat Oncol Biol Phys
1999;45(4):1025–34.
[67] Jones B, Dale RG. The potential for mathematical modelling in the assessment of
the radiation dose equivalent of cytotoxic chemotherapy given concomitantly
with radiotherapy. Br J Radiol 2005;78(934):939–44.
[68] Plataniotis GA, Dale RG. Use of concept of chemotherapy-equivalent biologically
effective dose to provide quantitative eval- uation of contribution of
chemotherapy to local tumor control in chemoradiotherapy cervical cancer trials.
Int J Radiat Oncol Biol Phys 2008;72(5):1538–43.
[69] Plataniotis GA, Dale RG. Assessment of the radiation-equivalent of chemotherapy
contributions in 1-phase radio-chemotherapy treatment of muscle-invasive
bladder cancer. Int J Radiat Oncol Biol Phys 2014;88(4):927–32.
[70] Moraru IC, Tai A, Erickson B, Li XA. Radiation dose responses for chemoradiation
therapy of pancreatic cancer: an analysis of compiled clinical data using
biophysical models. Practical Radiation Oncology 2014;4(1):13–9.
[71] Durante M, Tommasino F, Yamada S. Modeling combined chemotherapy and
particle therapy for locally advanced pancreatic cancer. Frontiers in Oncology
2015;5:145.
[72] Hartley A, Sanghera P, Glaholm J, Mehanna H, McConkey C, Fowler J.
Radiobiological modelling of the therapeutic ratio for the addition of
synchronous chemotherapy to radiotherapy in locally advanced squamous cell
carcinoma of the head and neck. Clin Oncol 2010;22(2):125–30.
[73] Meade S, McConkey C, Sanghera P, Mehanna H, Hartley A. Revised
radiobiological modelling of the contribution of syn- chronous chemotherapy to
the rate of grades 3–4 mucositis in head and neck cancer. Journal of Medical
Imaging and Radiation Oncology 2013;57(6):733–8.
[74] Meade S, Sanghera P, McConkey C, Fowler J, Fountzilas G, Glaholm J, Hartley A.
Revising the radiobiological model of synchronous chemotherapy in head-and-
neck cancer: a new analysis examining reduced weighting of accelerated
repopulation. Int J Radiat Oncol Biol Phys 2013;86(1):157–63.
[75] Sgouros G. Plasmapheresis in radioimmunotherapy of micrometastases: a
mathematical modeling and dosimetrical analysis. J Nucl Med: official
publication, Society of Nuclear Medicine 1992;33(12):2167–79.
[76] Epenetos AA, Snook D, Durbin H, Johnson PM, Taylor-Papadimitriou J.
Limitations of radiolabeled monoclonal antibodies for localization of human
neoplasms. Canc Res 1986;46(6):3183–91.
[77] Order SE, Sleeper AM, Stillwagon GB, Klein JL, Leichner PK. Radiolabeled
antibodies: results and potential in cancer therapy. Canc Res 1990;50(3
Supplement):1011s–3s.
[78] Begent RHJ, Pedley RB. Antibody targeted therapy in cancer: comparison of
murine and clinical studies. Canc Treat Rev 1990;17(2–3):373–8.
[79] Sharkey RM, Blumenthal RD, Hansen HJ, Goldenberg DM. Biological
considerations for radioimmunotherapy. Canc Res 1990;50(3 Supplement):
964s–9s.
[80] Fujimori K, Covell DG, Fletcher JE, Weinstein JN. Modeling analysis of the global
and microscopic distribution of im- munoglobulin G, F (ab’)2, and fab in tumors.
Canc Res 1989;49(20):5656–63.
J. Malinzi et al.
[81] Fujimori K, Covell DG, Fletcher JE, Weinstein JN. A modeling analysis of
monoclonal antibody percolation through tumors: a binding-site barrier. J Nucl
Med: Official Publication, Society of Nuclear Medicine 1990;31(7):1191–8.
[82] Fujimori K, Fisher DR, Weinstein JN. Integrated microscopic-macroscopic
pharmacology of monoclonal antibody radiocon- jugates: the radiation dose
distribution. Canc Res 1991;51(18):4821–7.
[83] van Osdol W, Fujimori K, Weinstein JN. An analysis of monoclonal antibody
distribution in microscopic tumor nodules: consequences of a “binding site
barrier. Canc Res 1991;51(18):4776–84.
[84] Kumar D, Kumar S. A mathematical model of radioimmunotherapy for tumor
treatment. Afr J Math Comput Sci Res 2010;3(6):101–6.
[85] Flux GD, Webb S, Ott RJ, Chittenden SJ, Thomas R. Three-dimensional dosimetry
for intralesional radionuclide therapy using mathematical modeling and
multimodality imaging. J Nucl Med 1997;38(7):1059–66.
[86] Serre R, Benzekry S, Padovani L, Meille C, André N, Ciccolini J, Barlesi F,
Muracciole X, Barbolosi D. Mathematical modeling of cancer immunotherapy and
its synergy with radiotherapy. Canc Res 2016;76(17):4931–40.
[87] Melcher A, Parato K, Rooney CM, Bell JC. Thunder and lightning: immunotherapy
and oncolytic viruses collide. Mol Ther 2011;19(6):1008–16.
[88] Cheema TA, Fecci PE, Ning J, Rabkin SD. Immunovirotherapy for the treatment of
glioblastoma. OncoImmunology 2014;3(1):e27218.
[89] Hardcastle J, Mills L, Malo CS, Jin F, Kurokawa C, Geekiyanage H, Schroeder M,
Sarkaria J, Johnson AJ, Galanis E. Immunovirotherapy with measles virus strains
in combination with anti–pd-1 antibody blockade enhances antitumor activity in
glioblastoma treatment. Neuro Oncol 2017;19(4):493–502.
[90] Cerullo V, Vähä-Koskela M, Hemminki A. Oncolytic adenoviruses: a potent form
of tumor immunovirotherapy. OncoImmunology 2012;1(6):979–81.
[91] Wares JR, Crivelli JJ, Yun C, Choi I, Gevertz JL, Kim PS. Treatment strategies for
combining immunostimulatory oncolytic virus therapeutics with dendritic cell
injections. Math Biosci Eng 2015;12(6):1237–56.
[92] Bagheri N, Shiina M, Lauffenburger DA, Korn WM. A dynamical systems model
for combinatorial cancer therapy enhances oncolytic adenovirus efficacy by mek-
inhibition. PLoS Comput Biol 2011;7(2):e1001085.
[93] Huang J, Zhang S, Choi K, Choi I, Kim J, Lee M, Kim H, Yun C. Therapeutic and
tumor-specific immunity induced by combination of dendritic cells and oncolytic
adenovirus expressing il-12 and 4-1bbl. Mol Ther 2010;18(2):264–74.
[94] Dingli D, Offord C, Myers R, Peng K, Carr TW, Josic K, Russell SJ, Bajzer Z.
Dynamics of multiple myeloma tumor therapy with a recombinant measles virus.
Canc Gene Ther 2009;16(12):873.
[95] Rommelfanger DM, Offord CP, Dev J, Bajzer Z, Vile RG, Dingli D. Dynamics of
melanoma tumor therapy with vesicular stomatitis virus: explaining the
variability in outcomes using mathematical modeling. Gene Ther 2012;19(5):543.
[96] Mok W, Stylianopoulos T, Boucher Y, Jain RK. Mathematical modeling of herpes
simplex virus distribution in solid tumors: implications for cancer gene therapy.
Clin Canc Res 2009;15(7):2352–60.
[97] DePillis L, Gallegos A, Radunskaya A. A model of dendritic cell therapy for
melanoma. Frontiers in Oncology 2013;3.
[98] Pappalardo F, Pennisi M, Ricupito A, Topputo F, Bellone M. Induction of t-cell
memory by a dendritic cell vaccine: a computational model. Bioinformatics 2014;
30(13):1884–91.
[99] Dritschel H, Waters SL, Roller A, Byrne HM. A mathematical model of cytotoxic
and helper T cell interactions in a tumour microenvironment. Letters in
Biomathematics 2018;5(sup1):S36–68.
[100] Wu JT, Kirn DH, Wein LM. Analysis of a three-way race between tumor growth, a
replication-competent virus and an immune response. Bull Math Biol 2004;66(4):
605.
[101] Berg DR, Offord CP, Kemler I, Ennis MK, Chang L, Paulik G, Bajzer Z,
Neuhauser C, Dingli D. In vitro and in silico multidimensional modeling of
oncolytic tumor virotherapy dynamics. PLoS Comput Biol 2019;15(3):e1006773.
[102] Wodarz D. Gene therapy for killing p53-negative cancer cells: use of replicating
versus nonreplicating agents. Hum Gene Ther 2003;14(2):153–9.
[103] Wodarz D, Hofacre A, Lau JW, Sun Z, Fan H, Komarova NL. Complex spatial
dynamics of oncolytic viruses in vitro: mathematical and experimental
approaches. PLoS Comput Biol 2012;8(6):e1002547.
[104] Bajzer Ž, Carr T, Josić K, Russell SJ, Dingli D. Modeling of cancer virotherapy
with recombinant measles viruses. J Theor Biol 2008;252(1):109–22.
[105] Biesecker M, J K, Lu H, Dingli D, Bajzer Ž. Optimization of virotherapy for cancer.
Bull Math Biol 2010;72(2):469–89.
[106] Gevertz JL, Wares JR. Developing a minimally structured mathematical model of
cancer treatment with oncolytic viruses and dendritic cell injections.
Computational and Mathematical Methods in Medicine 2018;2018.
[107] Timalsina A, Tian JP, Wang J. Mathematical and computational modeling for
tumor virotherapy with mediated immunity. Bull Math Biol 2017;79(8):1736–58.
[108] Jenner AL, Yun C, Yoon A, Coster ACF, Kim PS. Modelling combined virotherapy
and immunotherapy: strengthening the antitumour immune response mediated
by il-12 and gm-csf expression. Letters in Biomathematics 2018;5(sup1):S99–116.
[109] Mahasa KJ, Eladdadi A, De Pillis L, Ouifki R. Oncolytic potency and reduced virus
tumor-specificity in oncolytic virotherapy. a mathematical modelling approach.
PloS One 2017;12(9):e0184347.
[110] Cassidy T, Humphries AR. A mathematical model of viral oncology as an immuno-
oncology instigator. Math Med Biol 2020;37(1):117–51. 04.
[111] Wodarz D. Viruses as antitumor weapons: defining conditions for tumor
remission. Canc Res 2001;61(8):3501–7.
[112] Tao Y, Guo Q. A free boundary problem modelling cancer radiovirotherapy. Math
Model Methods Appl Sci 2007;17:1241–59. 08.
13
Informatics in Medicine Unlocked 23 (2021) 100534
[113] National Cancer Institute. https://www.cancer.gov/about-cancer/treatment/t
ypes/targeted-therapies/targeted-therapies-fact-sheet. Accessed: October 2020.
[114] Abbott LH, Michor F. Mathematical models of targeted cancer therapy. Br J Canc
2006;95(9):1136–41.
[115] Bozic I, Reiter JG, Allen B, Antal T, Chatterjee K, Shah P, Moon Y, Yaqubie A,
Kelly N, Le DT, et al. Evolutionary dynamics of cancer in response to targeted
combination therapy. eLife 2013;2:e00747.
[116] Yu C, Liu X, Yang J, Zhang M, Jin H, Ma X, Shi H. Combination of immunotherapy
with targeted therapy: theory and practice in metastatic melanoma. Front
Immunol 2019;10:990.
[117] Green AJ, Johnson CJ, Adamson KL, Begent RHJ. Mathematical model of
antibody targeting: important parameters defined using clinical data. Phys Med
Biol 2001;46(6):1679.
[118] Shen J, Li L, Yang T, Cohen PS, Sun G. Biphasic mathematical model of cell–drug
interaction that separates target-specific and off-target inhibition and suggests
potent targeted drug combinations for multi-driver colorectal cancer cells.
Cancers 2020;12(2):436.
[119] Sun X, guang Bao J, Shao Y. Mathematical modeling of therapy-induced cancer
drug resistance: connecting cancer mecha- nisms to population survival rates. Sci
Rep 2016;6:22498.
[120] Kozłowska E, Färkkilä A, Vallius T, Carpén O, Kemppainen J, Grénman S,
Lehtonen R, Hynninen J, Hietanen S, Hau- taniemi S. Mathematical modeling
predicts response to chemotherapy and drug combinations in ovarian cancer.
Canc Res 2018;78(14):4036–44.
[121] Jarrett AM, Shah A, Bloom MJ, T McKenna M, Hormuth DA, Yankeelov TE,
Sorace Anna G. Experimentally-driven mathematical modeling to improve
combination targeted and cytotoxic therapy for HER2+ breast cancer. Sci Rep
2019;9(1):1–12.
[122] Owen MR, Stamper IJ, Muthana M, Richardson GW, Dobson J, Lewis CE,
Byrne HM. Mathematical modeling predicts synergistic antitumor effects of
combining a macrophage-based, hypoxia-targeted gene therapy with
chemotherapy. Canc Res 2011;71(8):2826–37.
[123] Tang Jing, Karhinen Leena, Xu Tao, Szwajda Agnieszka, Yadav Bhagwan,
Wennerberg Krister, Aittokallio Tero. Target in- hibition networks: predicting
selective combinations of druggable targets to block cancer survival pathways.
PLoS Comput Biol 2013;9(9):e1003226.
[124] Tang Jing, Gautam Prson, Gupta Abhishekh, He Liye, Timonen Sanna,
Akimov Yevhen, Wang Wenyu, Szwajda Agnieszka, Jaiswal Alok, Turei Denes,
et al. Network pharmacology modeling identifies synergistic aurora b and zak
interaction in triple-negative breast cancer. NPJ systems biology and applications
2019;5(1):1–11.
[125] Araujo RP, Petricoin EF, Liotta LA. A mathematical model of combination therapy
using the egfr signaling network. Biosystems 2005;80(1):57–69.
[126] Komarova Natalia L, Wodarz Dominik. Drug resistance in cancer: principles of
emergence and prevention. Proc Natl Acad Sci Unit States Am 2005;102(27):
9714–9.
[127] Charusanti Pep, Hu Xiao, Chen Luonan, Neuhauser Daniel, DiStefano III Joseph J.
A mathematical model of bcr-abl au- tophosphorylation, signaling through the
crkl pathway, and gleevec dynamics in chronic myeloid leukemia. Discrete &
Continuous Dynamical Systems-B 2004;4(1):99.
[128] Shen S, Duan J, Meredith RF, Buchsbaum DJ, Brezovich IA, Pareek PN,
Bonner JA. Model prediction of treatment plan- ning for dose-fractionated
radioimmunotherapy. Cancer: Interdisciplinary International Journal of the
American Cancer Society 2002;94(S4):1264–9.
[129] Callahan MK. Two drugs are better than one-modeling drug combinations in
cancer therapy. Sci Transl Med 2013;5(194). 194ec116–194ec116.
[130] Chakwizira A, Ahlstedt J, Nittby Redebrandt H, Ceberg C. Mathematical
modelling of the synergistic combination of radio- therapy and indoleamine-2, 3-
dioxygenase (Ido) inhibitory immunotherapy against glioblastoma. Br J Radiol
2018;91(1087):20170857.
[131] Monjazeb AM, Grossenbacher SK, Sckisel GD, Canter R, Sparger EE, Culp W,
Kent MS, Murphy WJ. Combined radiotherapy and immunotherapy using CPG
oligodeoxynucleotides and indolamine 2, 3 dioxygenase (Ido) blockade. Journal
for Immunotherapy of Cancer 2013;1(1):P256.
[132] Radunskaya A, Kim R, Woods II, et al. 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 2018;4(1):25–41.
[133] Nazari F, Pearson AT, Nör JE, Jackson TL. A mathematical model for il-6-
mediated, stem cell driven tumor growth and targeted treatment. PLoS Comput
Biol 2018;14(1):e1005920.
[134] Wang S, Schättler H. Optimal control of a mathematical model for cancer
chemotherapy under tumor heterogeneity. Math. Biosci. and Engr.-MBE 2016;13:
1223–40.
[135] Malinzi J. Mathematical modeling of cancer treatments and the role of the
immune system response to tumor invasion. PhD thesis. 2015.
[136] Spring BQ, Lang RT, Kercher EM, Rizvi I, Wenham RM, Conejo-Garcia JR,
Hasan T, Gatenby RA, Enderling H. Illu- minating the numbers: integrating
mathematical models to optimize photomedicine dosimetry and combination
therapies. Frontiers in Physics 2019;7:46.
[137] Rihan FA, Abdelrahman DH, Al-Maskari F, Ibrahim F, Abdeen MA. Delay
differential model for tumour-immune response with chemoimmunotherapy and
optimal control. Computational and Mathematical Methods in Medicine 2014;
2014.
[138] Oke S, Matadi M, Xulu S. Optimal control analysis of a mathematical model for
breast cancer. Math Comput Appl 2018;23(2):21.
J. Malinzi et al.
[139] Mamat M, Subiyanto KA, Kartono A. Mathematical model of cancer treatments
using immunotherapy, chemotherapy and biochemotherapy. Appl Math Sci 2013;
7(5):247–61.
[140] Lai X, Stiff A, Duggan M, Wesolowski R, Carson WE, Friedman A. Modeling
combination therapy for breast cancer with bet and immune checkpoint
inhibitors. Proc Natl Acad Sci Unit States Am 2018;115(21):5534–9.
[141] Joshi B, Wang X, Banerjee S, Tian H, Matzavinos A, Chaplain MAJ. On
immunotherapies and cancer vaccination protocols: a mathematical modelling
approach. J Theor Biol 2009;259(4):820–7.
[142] Kosinsky Y, Dovedi SJ, Peskov K, Voronova V, Chu L, Tomkinson H, Al-Huniti N,
Stanski DR, Helmlinger G. Radi- ation and pd-(l) 1 treatment combinations:
immune response and dose optimization via a predictive systems model. Journal
for Immunotherapy of Cancer 2018;6(1):17.
[143] Rihan FA, Lakshmanan S, Maurer H. Optimal control of tumour-immune model
with time-delay and immuno-chemotherapy. Appl Math Comput 2019;353:
147–65.
[144] Akman Yıldız T, Arshad S, Baleanu D. Optimal chemotherapy and
immunotherapy schedules for a cancer-obesity model with caputo time fractional
derivative. Math Methods Appl Sci 2018;41(18):9390–407.
[145] Ratajczyk E, Ledzewicz U, Schättler H. Optimal control for a mathematical model
of glioma treatment with oncolytic therapy and TNF-α inhibitors. J Optim Theor
Appl 2018;176(2):456–77.
[146] Sharma S, Samanta GP. Analysis of the dynamics of a tumor–immune system with
chemotherapy and immunotherapy and quadratic optimal control. Differential
Equations and Dynamical Systems 2016;24(2):149–71.
[147] Ledzewicz U, Schättler H. An optimal control approach to cancer chemotherapy
with tumor–immune system interactions. In: Mathematical models of tumor-
immune system dynamics. Springer; 2014. p. 157–96.
[148] Nazari M, Ghaffari A, Arab F. Finite duration treatment of cancer by using vaccine
therapy and optimal chemotherapy: state- dependent riccati equation control and
extended kalman filter. J Biol Syst 2015;23:1–29. 01.
[149] Parra-Guillen ZP, Berraondo P, Grenier E, Ribba B, Troconiz IF. Mathematical
model approach to describe tumour re- sponse in mice after vaccine
administration and its applicability to immune-stimulatory cytokine-based
strategies. AAPS J 2013;15(3):797–807.
[150] Rodrigues DS, de Arruda Mancera PF, et al. Mathematical analysis and
simulations involving chemotherapy and surgery on large human tumours under
a suitable cell-kill functional response. Math Biosci Eng 2013;10(1):221–34.
[151] Villasana M, Ochoa G, Aguilar S. Modeling and optimization of combined
cytostatic and cytotoxic cancer chemotherapy. Artif Intell Med 2010;50(3):
163–73.
[152] Chareyron S, Alamir M. Mixed immunotherapy and chemotherapy of tumors:
feedback design and model updating schemes. J Theor Biol 2009;258(3):444–54.
[153] Ledzewicz U, Schattler H, d’Onofrio A. Optimal control for combination therapy
in cancer. In: 2008 47th IEEE conference on decision and control. IEEE; 2008.
p. 1537–42.
[154] d’Onofrio A, Ledzewicz U, Maurer H, Schättler H. On optimal delivery of
combination therapy for tumors. Math Biosci 2009;222(1):13–26.
[155] Powathil G, Kohandel M, Sivaloganathan S, Oza A, Milosevic M. Mathematical
modeling of brain tumors: effects of radio- therapy and chemotherapy. Phys Med
Biol 2007;52(11):3291.
[156] Imbs D, El Cheikh R, Boyer A, Ciccolini J, Mascaux C, Lacarelle B, Barlesi F,
Barbolosi D, Benzekry S. Revisiting bevacizumab+ cytotoxics scheduling using
mathematical modeling: Proof of concept study in experimental non-small cell
lung carcinoma. CPT Pharmacometrics Syst Pharmacol 2018;7(1):42–50.
[157] Tao Y, Guo Q. A mathematical model of combined therapies against cancer using
viruses and inhibitors. Sci China, Ser A: Mathematics 2008;51(12):2315–29.
[158] Hadjiandreou MM, Mitsis GD. Mathematical modeling of tumor growth, drug-
resistance, toxicity, and optimal therapy design. IEEE (Inst Electr Electron Eng)
Trans Biomed Eng 2014;61(2):415–25.
[159] Su Y, Jia C, Chen Y. Optimal control model of tumor treatment with oncolytic
virus and mek inhibitor. BioMed Res Int 2016;2016.
[160] Kim PS, Lee PP. Modeling protective anti-tumor immunity via preventative cancer
vaccines using a hybrid agent-based and delay differential equation approach.
PLoS Comput Biol 2012;8(10):e1002742.
[161] Preziosi L. Cancer modelling and simulation. CRC Press; 2003.
[162] Kelley WG, Peterson AC. The theory of differential equations: classical and
qualitative. Springer Science & Business Media; 2010.
[163] Brauer F, Castillo-Chavez C. Mathematical models in population biology and
epidemiology, vol. 2. Springer; 2012.
[164] Eqworld: The world of mathematical equations. http://eqworld.ipmnet.
ru/en/software.htm. Accessed: October 2020.
[165] Mathematical software. https://www.mat.univie.ac.at/~neum/software.html.
Accessed: October 2020.
[166] DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of
drug development costs. J Health Econ 2003;22(2):151–85.
[167] Binz E, Lauer UM. Chemovirotherapy: combining chemotherapeutic treatment
with oncolytic virotherapy. Oncolytic Virotherapy 2015;4:39.
[168] Sharkey RM, Goldenberg DM. Cancer radioimmunotherapy. Immunotherapy
2011;3(3):349–70.
[169] Shun L, Meng X, Feng R, Yang L, Xing L, Yu J. The role of radiation oncology in
immuno-oncology. Oncol 2019;24(Suppl 1):S42–52.
[170] Identifying novel drug combinations to overcome treatment resistance. National
Cancer Institute, https://www.cancer.gov/about-cancer/treatment/research/dru
g-combo-resistance, Accessed May 2019.
14
Informatics in Medicine Unlocked 23 (2021) 100534
[171] Tolcher AW, Mayer LD. Improving combination cancer therapy: the combiplex®
development platform. Future Oncol 2018;14(13):1317–32.
[172] Shrivastava S, Mahantshetty U, Engineer R, Chopra S, Hawaldar R, Hande V,
Kerkar RA, Maheshwari A, Shylasree TS, Ghosh J, et al. Cisplatin
chemoradiotherapy vs radiotherapy in figo stage iiib squamous cell carcinoma of
the uterine cervix: a randomized clinical trial. JAMA Oncology 2018;4(4):506–13.
[173] 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;7(1):8.
[174] Brown JR, Hallek MJ, Pagel JM. Chemoimmunotherapy versus targeted treatment
in chronic lymphocytic leukemia: when, how long, how much, and in which
combination? American Society of Clinical Oncology Educational Book 2016;36:
e387–98.
[175] Hallek M, Fischer K, Fingerle-Rowson G, Fink AM, Busch R, Mayer J, Hensel M,
Hopfinger G, Hess G, Von Grünhagen U, et al. Addition of rituximab to
fludarabine and cyclophosphamide in patients with chronic lymphocytic
leukaemia: a randomised, open-label, phase 3 trial. Lancet 2010;376(9747):
1164–74.
[176] Galluzzi L, Vacchelli E, Bravo-San Pedro J, Buqué A, Senovilla L, Baracco EE,
Bloy N, Castoldi F, Abastado J, Agostinis P, et al. Classification of current
anticancer immunotherapies. Oncotarget 2014;5(24):12472.
[177] Sasse AD, Sasse EC, Go Clark L, Ulloa L, Clark OAC. Chemoimmunotherapy versus
chemotherapy for metastatic malig- nant melanoma. Cochrane Database Syst Rev
2007;(1).
[178] Zhang P, Xi M, Zhao L, Shen J, Li Q, He L, Liu S, Liu M. Is there a benefit in
receiving concurrent chemoradiotherapy for elderly patients with inoperable
thoracic esophageal squamous cell carcinoma? PloS One 2014;9(8):e105270.
[179] Advani SJ, Sibley GS, Song PY, Hallahan DE, Kataoka Y, Roizman B, R
Weichselbaum R. Enhancement of replication of genetically engineered herpes
simplex viruses by ionizing radiation: a new paradigm for destruction of
therapeutically intractable tumors. Gene Ther 1998;5(2):160.
[180] Chung SM, Advani SJ, Bradley JD, Kataoka Y, Vashistha K, Yan SY, Markert JM,
Gillespie GY, Whitley RJ, Roizman B, et al. The use of a genetically engineered
herpes simplex virus (r7020) with ionizing radiation for experimental hepatoma.
Gene Ther 2002;9(1):75.
[181] Blank SV, Rubin SC, Coukos G, Amin KM, Albelda SM, Molnar-Kimber KL.
Replication-selective herpes simplex virus type 1 mutant therapy of cervical
cancer is enhanced by low-dose radiation. Hum Gene Ther 2002;13(5):627–39.
[182] Eifel PJ. Chemoradiotherapy in the treatment of cervical cancer. In: Seminars in
radiation oncology, vol. 16. Elsevier; 2006. p. 177–85.
[183] Neuner G, Patel A, Suntharalingam M. Chemoradiotherapy for esophageal cancer.
Gastrointestinal Cancer Research: GCR 2009;3(2):57.
[184] Shitara K, Muro K. Chemoradiotherapy for treatment of esophageal cancer in
Japan: current status and perspectives. Gastroin- testinal Cancer Research: GCR
2009;3(2):66.
[185] Baxi SS, O’Neill C, Sherman EJ, L Atoria C, Lee NY, Pfister Dd G, Elkin EB. Trends
in chemoradiation use in elderly patients with head and neck cancer: changing
treatment patterns with cetuximab. Head Neck 2016;38(S1):E165–71.
[186] Kraeber-Bodéré F, Barbet J, Chatal J. Radioimmunotherapy: from current clinical
success to future industrial breakthrough? J Nucl Med 2016;57(3):329–31.
[187] Gopal AK, Gooley TA, Maloney DG, Petersdorf SH, Eary JF, Rajendran JG,
Bush SA, Durack LD, Golden J, Martin PJ, et al. High-dose radioimmunotherapy
versus conventional high-dose therapy and autologous hematopoietic stem cell
transplantation for relapsed follicular non-hodgkin lymphoma: a multivariable
cohort analysis. Blood 2003;102(7):2351–7.
[188] Msaouel P, Iankov ID, Allen C, Aderca I, Federspiel MJ, Tindall DJ, Morris JC,
Koutsilieris M, Russell SJ, Gala- nis E. Noninvasive imaging and radiovirotherapy
of prostate cancer using an oncolytic measles virus expressing the sodium iodide
symporter. Mol Ther 2009;17(12):2041–8.
[189] Li H, Peng K, Russell SJ. Oncolytic measles virus encoding thyroidal sodium
iodide symporter for squamous cell cancer of the head and neck radiovirotherapy.
Hum Gene Ther 2011;23(3):295–301.
[190] Dingli D, Peng K, Harvey ME, Greipp PR, O’Connor MK, Cattaneo R, Morris JC,
Russell SJ. Image-guided ra- diovirotherapy for multiple myeloma using a
recombinant measles virus expressing the thyroidal sodium iodide symporter.
Blood 2004;103(5):1641–6.
[191] Touchefeu Y, Franken P, Harrington KJ. Radiovirotherapy: principles and
prospects in oncology. Curr Pharmaceut Des 2012;18(22):3313–20.
[192] Chiocca EA. Oncolytic viruses. Nat Rev Canc 2002;2(12):938.
[193] Spear MA, Sun F, Eling DJ, Gilpin E, Kipps TJ, Chiocca EA, Bouvet M.
Cytotoxicity, apoptosis, and viral replication in tumor cells treated with oncolytic
ribonucleotide reductase-defective herpes simplex type 1 virus (hrr3) combined
with ionizing radiation. Canc Gene Ther 2000;7(7):1051.
[194] Jorgensen TJ, Katz S, Wittmack EK, Varghese S, Todo T, Rabkin SD, Martuza RL.
Ionizing radiation does not alter the antitumor activity of herpes simplex virus
vector g207 in subcutaneous tumor models of human and murine prostate cancer.
Neoplasia 2001;3(5):451–6.
[195] Mathworks. Accessed: October 2020, http://www.mathworks.com/help/matlab.
[196] Differential equations in maple. Accessed: October 2020, https://www.maplesoft.
com/applications/view.aspx?sid=103786&view=html.
[197] Wolfram Language & System Documentation Centre. https://reference.wolfram.
com/language/guide/EquationSolving.html. Accessed: October 2020.
[198] Python. https://www.python.org/. Accessed: October 2020.
[199] Sage. https://doc.sagemath.org/html/en/reference/calculus/sage/calculus/desol
vers.html. Ac- cessed: October 2020.
J. Malinzi et al.
[200] Maxima, a computer algebra system. http://maxima.sourceforge.net/. Accessed:
October 2020.
[201] GNU Octave. https://www.gnu.org/software/octave/index. Accessed: October
2020.
[202] Mathcad. https://www.wikiwand.com/en/Mathcad. Accessed: October 2020.
[203] Julia. https://julialang.org. Accessed: October 2020.
[204] The R project for statistical computing. https://www.r-project.org. Accessed:
October 2020.
[205] Copasi. http://copasi.org/. Accessed: November 2020.
[206] Compusyn. http://www.combosyn.com/. Accessed: November 2020.
[207] Calusyn. http://www.biosoft.com/w/calcusyn.htm. Accessed: November 2020.
[208] He Liye, Wennerberg Krister, Aittokallio Tero, Tang Jing. Timma-r: an r package
for predicting synergistic multi-targeted drug combinations in cancer cell lines or
patient-derived samples. Bioinformatics 2015;31(11):1866–8.
[209] Ianevski Aleksandr, Giri Anil K, Aittokallio Tero. Synergyfinder 2.0: visual
analytics of multi-drug combination synergies. Nucleic Acids Res 2020.
[210] Veroli Giovanni Y Di, Fornari Chiara, Wang Dennis, Mollard Severine,
Bramhall Jo L, Richards Frances M, Jodrell Duncan I. Combenefit: an interactive
platform for the analysis and visualization of drug combinations. Bioinformatics
2016;32(18):2866–8.
[211] URDME. http://urdme.github.io/urdme/. Accessed January 2021.
[212] Understand, predict and optimize physics-based designs and processes with
COMSOL Multiphysics. https://www.comsol.com/comsol-multiphysics.
Accessed: November 2020.
[213] Zagidullin B, Aldahdooh J, Zheng S, Wang W, Wang Y, Saad J, Malyutina A,
Jafari M, Tanoli Z, Pessia A, et al. Drugcomb: an integrative cancer drug
combination data portal. Nucleic Acids Res 2019;47(W1):W43–51.
15
Informatics in Medicine Unlocked 23 (2021) 100534
[214] Nci Almanac. A new tool for research on cancer drug combination. https://www.
cancer.gov/news-events/cancer-currents-blog/2017/nci-almanac-drug-combina
tions", Accessed January 2021.
[215] Liu Hui, Zhang Wenhao, Zou Bo, Wang Jinxian, Deng Yuanyuan, Deng Lei.
Drugcombdb: a comprehensive database of drug combinations toward the
discovery of combinatorial therapy. Nucleic Acids Res 2020;48(D1):D871–81.
[216] Xue Bai, Yang Xiaobo, Wu Liangcai, Zuo Bangyou, Lin Jianzhen, Wang Shanshan,
Jin Bian, Sang Xinting, He Yungang, Yang Zhen, et al. Cmttdb: the cancer
molecular targeted therapy database. Ann Transl Med 2019;7(22).
[217] Muniz de Ávila Paulo, Cesar de Melo Bernardo Paulo, Teodoro Marques da
Silva Ramon Gustavo. Ana Lúcia Fachin, Mozart Marins, Edilson Carlos Caritá,
et al. Cancrox: a cross-species cancer therapy database. Database 2019:2019.
[218] Yi-Lwern Yap Kevin, Chan Alexandre, Keung Chui Wai, Chen Yu Zong. Cancer
informatics for the clinician: an interaction database for chemotherapy regimens
and antiepileptic drugs. Seizure 2010;19(1):59–67.
[219] Ianevski Aleksandr, Timonen Sanna, Alexander Kononov, Aittokallio Tero,
Giri Anil K. Syntoxprofiler: an interactive analysis of drug combination synergy,
toxicity and efficacy. PLoS Comput Biol 2020;16(2):e1007604.
[220] Pan Pantziarka, Verbaanderd Ciska, Sukhatme Vidula, Capistrano I Rica,
Crispino Sergio, Gyawali Bishal, Rooman Ilse, Van Nuffel An MT, Meheus Lydie,
Sukhatme Vikas P, et al. Redo_db: the repurposing drugs in oncology database.
ecancermedicalscience 2018;12.
[221] Kumar Rahul, Chaudhary Kumardeep, Gupta Sudheer, Singh Harinder,
Kumar Shailesh, Gautam Ankur, Kapoor Pallavi, Raghava Gajen- dra PS.
Cancerdr: cancer drug resistance database. Sci Rep 2013;3:1445.