Strongly preserved modules between cancer tissue and cell line contribute to drug resistance analysis across multiple cancer types

The existence and emergence of drug resistance in tumor cells is the main burden of cancer treatment. Most cancer drug resistance analyses are based entirely on cell line data and ignore the discordance between human tumors and cell lines, leading to biased preclinical model transformation. Based on cancer tissue data in TCGA and cancer cell line data in CCLE, this study identified and excluded non-preserved module (NP module) between cancer tissue and cell lines. We used strongly preserved module (SP module) for clinically relevant drug resistance analysis and identified 2068 “cancer-drug-module” pairs of 7 cancer types and 212 drugs based on data in GDSC. Furthermore, we identified potentially ineffective combination therapy (PICT) from multiple perspectives. Finally, we found 1608 sets of predictors that can predict drug response. These results provide insights and clues for the clinical selection of effective chemotherapy drugs to overcome cancer resistance in a new perspective.     

A forward genetic approach for analyzing the mechanism of resistance to the anti-cancer drug, 5-fluorouracil, using Caenorhabditis elegans

Pyrimidine antagonists including 5-Fluorouracil (5-FU) have been used in chemotherapy for cancer patients for over 40 years. 5-FU, especially, is a mainstay treatment for colorectal cancer. It is a pro-drug that is converted to the active drug via the nucleic acid biosynthetic pathway. The metabolites of 5-FU inhibit normal RNA and DNA function, and induce apoptosis of cancer cells. One of the major obstacles to successful chemotherapy is the resistance of cancer cells to anti-cancer drugs. Therefore, it is important to elucidate resistance mechanisms to improve the efficacy of chemotherapy. We have used C. elegans as a model system to investigate the mechanism of resistance to 5-FU, which induces germ cell death and inhibits larval development in C. elegans. We screened 5-FU resistant mutants no longer arrested as larvae by 5-FU. We obtained 18 mutants out of 72,000 F1 individuals screened, and mapped them into three complementation groups. We propose that C. elegans could be a useful model system for studying mechanisms of resistance to anti-cancer drugs.     

Modeling the Transfer of Drug Resistance in Solid Tumors

ABC efflux transporters are a key factor leading to multidrug resistance in cancer. Overexpression of these transporters significantly decreases the efficacy of anti-cancer drugs. Along with selection and induction, drug resistance may be transferred between cells, which is the focus of this paper. Specifically, we consider the intercellular transfer of P-glycoprotein (P-gp), a well-known ABC transporter that was shown to confer resistance to many common chemotherapeutic drugs. In a recent paper, Durán et al. (Bull Math Biol 78(6):1218–1237, 2016) studied the dynamics of mixed cultures of resistant and sensitive NCI-H460 (human non-small lung cancer) cell lines. As expected, the experimental data showed a gradual increase in the percentage of resistance cells and a decrease in the percentage of sensitive cells. The experimental work was accompanied with a mathematical model that assumed P-gp transfer from resistant cells to sensitive cells, rendering them temporarily resistant. The mathematical model provided a reasonable fit to the experimental data. In this paper, we develop a new mathematical model for the transfer of drug resistance between cancer cells. Our model is based on incorporating a resistance phenotype into a model of cancer growth (Greene et al. in J Theor Biol 367:262–277, 2015). The resulting model for P-gp transfer, written as a system of integro-differential equations, follows the dynamics of proliferating, quiescent, and apoptotic cells, with a varying resistance phenotype. We show that this model provides a good match to the dynamics of the experimental data of Durán et al. (2016). The mathematical model shows a better fit when resistant cancer cells have a slower division rate than the sensitive cells.     

Targeted Casp8AP2 methylation increases drug resistance in mesenchymal stem cells and cancer cells

Casp8AP2 contains a FLASH functional domain and is critical for the formation of death complex and the relay of death signal into the cells. Genetic defects in Casp8AP2 are associated with several diseases. A CpG island within the Casp8AP2 promoter is differentially regulated during somatic stem cell differentiation, and aberrant DNA methylation within the Casp8AP2 promoter has been reported in cancers. We hypothesized that abnormal DNA methylation of Casp8AP2 promoter might contribute to prolonged cellular survival or drug resistance in cancer. The epigenetic state within the Casp8AP2 promoter was then determined in different cancer cell lines and patient samples by methylation-specific PCR. Targeted Casp8AP2 methylation within normal and tumor cells was performed to see whether methylation promoted drug resistance. We found differential Casp8AP2 methylation among the normal and tumoral samples. Global demethylation in a platinum drug-resistant human gastric cancer cell line reversed Casp8AP2 methylation and diminished drug resistance. Targeted methylation of the Casp8AP2 promoter in somatic stem cells and cancer cells increased their resistance to drugs including platinum drugs. These data demonstrate that methylation within the Casp8AP2 promoter correlates with the development of drug resistance and might serve as a biomarker and treatment target for drug resistance in cancer cells.     

Co-amplification of double minute chromosomes, multiple drug resistance, and cell surface P-glycoprotein in DNA-mediated transformants of mouse cells.

A genetic system comprised of mammalian cell mutants which demonstrate concomitant resistance to a number of unrelated drugs has been described previously. The resistance is due to reduced cell membrane permeability and is correlated with the presence of large amounts of a plasma membrane glycoprotein termed P-glycoprotein. This system could represent a model for multiple drug resistance which develops in cancer patients treated with chemotherapeutic drugs. We demonstrate here that the multiple drug resistance phenotype can be transferred to mouse cells with DNA from a drug-resistant mutant and then amplified quantitatively by culture in media containing increasing concentrations of drug. The amount of P-glycoprotein was correlated directly with the degree of drug resistance in the transformants and amplified transformants. In addition, the drug resistance and expression of P-glycoprotein of the transformants were unstable and associated quantitatively with the number of double minute chromosomes. We suggest that the gene for multiple drug resistance and P-glycoprotein is contained in these extrachromosomal particles and is amplified by increases in double minute chromosome number. The potential use of this system for manipulation of mammalian genes in general is discussed.     

Drug scheduling of cancer chemotherapy based on natural actor-critic approach

Recently, reinforcement learning methods have drawn significant interests in the area of artificial intelligence, and have been successfully applied to various decision-making problems. In this paper, we study the applicability of the NAC (natural actor-critic) approach, a state-of-the-art reinforcement learning method, to the drug scheduling of cancer chemotherapy for an ODE (ordinary differential equation)-based tumor growth model. ODE-based cancer dynamics modeling is an active research area, and many different mathematical models have been proposed. Among these, we use the model proposed by de Pillis and Radunskaya (2003), which considers the growth of tumor cells and their interaction with normal cells and immune cells. The NAC approach is applied to this ODE model with the goal of minimizing the tumor cell population and the drug amount while maintaining the adequate population levels of normal cells and immune cells. In the framework of the NAC approach, the drug dose is regarded as the control input, and the reward signal is defined as a function of the control input and the cell populations of tumor cells, normal cells, and immune cells. According to the control policy found by the NAC approach, effective drug scheduling in cancer chemotherapy for the considered scenarios has turned out to be close to the strategy of continuing drug injection from the beginning until an appropriate time. Also, simulation results showed that the NAC approach can yield better performance than conventional pulsed chemotherapy.     

Dictyostelium discoideum to human cells: pharmacogenetic studies demonstrate a role for sphingolipids in chemoresistance

Resistance to chemotherapy is a major obstacle for the treatment of cancer and a subject of extensive research. Numerous mechanisms of drug resistance have been proposed, and they differ for different drugs. Nevertheless, it is clear that our understanding of this important problem is still incomplete, and that new targets for modulating therapy still await discovery. The attractive biology and the availability of powerful molecular techniques have made the cellular slime mold Dictyostelium discoideum, a powerful non-mammalian model for drug target discovery, and the problem of drug resistance. To understand the molecular basis of chemoresistance to the widely used drug cisplatin, both genetic and pharmacological approaches have been applied to this versatile experimental system. These studies have resulted in the identification of novel molecular pathways which can be used to increase the efficacy of cisplatin, and brought attention to the role of sphingolipids in mediating the cellular response to chemotherapeutic drugs. In the following review, we will describe the history and utility of D. discoideum in pharmacogenetics, and discuss recent studies which focus attention on the role of sphingolipids in chemotherapy and chemoresistance.     

Mathematical modelling for the assessment of the effect of drug application delays in metronomic chemotherapy of cancer due to physiological constraints

Several anti-angiogenic drugs in metronomic chemotherapy (MCT) scheduling in cancer have the potentiality to kill cancer cells. Theoretical analysis and experimental models have substantially indicated this potentiality. Therefore MCT may provide an option to improve the quality of life (QoL) with a quite reduced level of toxicity in the patient concerned. However, the criticism regarding MCT is that this may have a different toxicity profiling. This may require frequent interventions in the long drawn-out treatment schedule. Hence, it is worthwhile to devise a reliable estimation technique. Mathematical modelling and computer simulation techniques can assist in this regard. This, in turn, may help in the drug application decision. It is particularly important in situations where a fluctuating clearance rate of the applied drug is present. So, from dynamical perspectives, the frequencies and timings of drug stoppages need to be determined appropriately for maintaining the stability of the system without overburdening the toxicity level. Present work addresses the issue of subsequent drug administration delays and arrives at bounds of system performance and explores the treatment response in presence of physiological constraints.     

Drug Resistance in Glioblastoma: A Mini Review

Glioblastoma multiforme (GBM) is recognized as the most common and lethal form of central nervous system cancer. Currently used surgical techniques, chemotherapeutic agents, and radiotherapy strategies have done very little in extending the life expectancies of patients diagnosed with GBM. The difficulty in treating this malignant disease lies both in its inherent complexity and numerous mechanisms of drug resistance. In this review, we summarize several of the primary mechanisms of drug resistance. We reviewed available published literature in the English language regarding drug resistance in glioblastoma. The reasons for drug resistance in glioblastoma include drug efflux, hypoxic areas of tumor cells, cancer stem cells, DNA damage repair, and miRNAs. Many potential therapies target these mechanisms, including a series of investigated alternative and plant-derived agents. Future research and clinical trials in glioblastoma patients should pursue combination of therapies to help combat drug resistance. The emerging new data on the potential of plant-derived therapeutics should also be closely considered and further investigated.     

Vesicles mimicking normal and cancer cell membranes exhibit differential responses to the cell-penetrating peptide Pep-1

The cell-penetrating peptide (CPP) Pep-1 presents a great potential in drug delivery due to its intrinsic property to cross plasma membrane. However, its mechanism of entry into the cell remains unresolved. In this study, we compare the selectivity of Pep-1 towards vesicles mimicking normal and cancer cell membranes. The interaction was performed in a wide range of peptide-to-lipid molar ratios using infrared (IR), fluorescence, scanning electron microscopy (SEM), thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC) techniques. At low peptide concentration, fluorescence experiments show that lipid-phosphatidylserine (PS) seems to enable Pep-1 translocation into cancer cell membrane as evidenced by the blue shift of its maximal emission wavelength. DSC data show that Pep-1 induces segregation of lipids. At high peptide concentration, IR data indicate that the interaction of Pep-1 is relatively stronger with normal cell membrane than with cancer cell membrane through the phosphate groups, while the interaction is weaker with normal cell membrane than with cancer cell membrane through the carbonyl groups. TGA and DSC data reveal that vesicles of normal cell membrane are thermally more stable than vesicles of cancer cell membrane. This suggests that the additional lipid PS included in cancer cell membrane has a destabilizing effect on the membrane structure. SEM images reveal that Pep-1 form superstructures including spherical particles and fibrils in the presence of both model membranes. PS seems to enhance peptide transport across cellular membranes. The biophysical techniques in this study provide valuable insights into the properties of CPPs in drug delivery systems.     

The 1991 Merck Frosst Award. Multidrug resistance in small cell lung cancer.

The two-year survival rate of patients with small cell lung cancer is less than 10%. The major reason for this poor outcome is the development of drug resistance. Panels of small cell lung cancer cell lines have been established, providing models for the study of drug resistance in this tumour. One such model is the doxorubicin-selected H69AR cell line. H69AR displays the typical multidrug resistance phenotype in that it is cross-resistant to anthracyclines, Vinca alkaloids (e.g., vinblastine) and epipodophyllotoxins (e.g., VP-16). However, H69AR cells do not overexpress P-glycoprotein, the membrane drug efflux pump frequently found on multidrug resistant cells. Some alterations in glutathione levels and associated enzyme activities were found but the data do not support the notion that enhanced drug detoxication is involved in H69AR cell resistance. Fewer drug-induced DNA strand breaks, reduced levels of topoisomerase II, and reduced formation of drug-stabilized DNA/topoisomerase II complexes were observed in H69AR cells. These data implicate topoisomerase II in the resistance phenotype of H69AR cells, but cannot explain H69AR cell resistance to the Vinca alkaloids, which do not have topoisomerase II as a target. Monoclonal antibodies against antigens overexpressed on H69AR cells have been derived and four have been characterized. Immunoscreening of an H69AR cDNA expression library has allowed the identification of one of these antigens as p36 (annexin II), a Ca2+/phospholipid binding protein. Chemosensitizers and novel xenobiotics have been examined for their ability to circumvent the drug resistance of H69AR cells. The limited success of these investigations suggests that innovative approaches may be required. In conclusion, the data obtained with H69AR and other models of small cell lung cancer indicate that multiple mechanisms contribute to drug resistance in this disease.     

Modeling the Dynamics of Heterogeneity of Solid Tumors in Response to Chemotherapy

In this paper, we extend the model of the dynamics of drug resistance in a solid tumor that was introduced by Lorz et al. (Bull Math Biol 77:1–22, 2015). Similarly to the original, radially symmetric model, the quantities we follow depend on a phenotype variable that corresponds to the level of drug resistance. The original model is modified in three ways: (i) We consider a more general growth term that takes into account the sensitivity of resistance level to high drug dosage. (ii) We add a diffusion term in space for the cancer cells and adjust all diffusion terms (for the nutrients and for the drugs) so that the permeability of the resource and drug is limited by the cell concentration. (iii) We add a mutation term with a mutation kernel that corresponds to mutations that occur regularly or rarely. We study the dynamics of the emerging resistance of the cancer cells under continuous infusion and on–off infusion of cytotoxic and cytostatic drugs. While the original Lorz model has an asymptotic profile in which the cancer cells are either fully resistant or fully sensitive, our model allows the emergence of partial resistance levels. We show that increased drug concentrations are correlated with delayed relapse. However, when the cancer relapses, more resistant traits are selected. We further show that an on–off drug infusion also selects for more resistant traits when compared with a continuous drug infusion of identical total drug concentrations. Under certain conditions, our model predicts the emergence of a heterogeneous tumor in which cancer cells of different resistance levels coexist in different areas in space.     

ncRNAs associated with drug resistance and the therapy of digestive system neoplasms

Digestive system cancer remains a common cancer and the main cause of cancer-related death worldwide. Drug resistance is a major challenge in the therapy of digestive system cancer, and represents a primary obstacle in the treatment of cancer by restricting the efficiency of both traditional chemotherapy and biological therapies. Existing studies indicate that noncoding RNAs play an important role in the evolution and progression of drug resistance in digestive system cancer, mainly by modulating drug transporter-related proteins, DNA damage repair, cell-cycle-related proteins, cell apoptosis-related proteins, drug target-related proteins, and the tumor microenvironment. In this review, we address the potential mechanisms of ncRNAs underlying drug resistance in digestive system tumors and discuss the possible application of ncRNAs against drug resistance in digestive system tumors.     

基于多组学数据的肿瘤药物敏感性预测北大核心CSCD

肿瘤药物敏感性预测在指导患者临床用药方面具有重要意义。本文基于癌症药物敏感性基因组学数据库(genomics of drug sensitivity in cancer, GDSC) 198种药物的细胞系敏感性IC50数据,通过Stacking集成学习构建了包含基因表达、基因突变、拷贝数变异数据的多组学癌症药物敏感性预测模型。采用多种特征选择方法对基因特征进行降维,使用Stacking方法集成6种初级学习器和1种次级学习器进行建模,采用5折交叉进行模型验证。预测结果中AUC大于0.9的占比为36.4%,在0.8–0.9之间的占比为49.0%,最低AUC为0.682。基于Stacking构建的多组学预测模型较已有单组学和多组学模型的准确性和稳定性具有优势。多组学整合预测药物敏感性优于单一组学。特征基因功能注释和富集分析解析了肿瘤对sorafenib潜在的耐药机制,从生物学角度提供了模型可解释性及其应用于临床用药指导的价值。     

Dynamics of resistance development to imatinib under increasing selection pressure: a combination of mathematical models and in vitro data

In the last decade, cancer research has been a highly active and rapidly evolving scientific area. The ultimate goal of all efforts is a better understanding of the mechanisms that discriminate malignant from normal cell biology in order to allow the design of molecular targeted treatment strategies. In individual cases of malignant model diseases addicted to a specific, ideally single oncogene, e.g. Chronic myeloid leukemia (CML), specific tyrosine kinase inhibitors (TKI) have indeed been able to convert the disease from a ultimately life-threatening into a chronic disease with individual patients staying in remission even without treatment suggestive of operational cure. These developments have been raising hopes to transfer this concept to other cancer types. Unfortunately, cancer cells tend to develop both primary and secondary resistance to targeted drugs in a substantially higher frequency often leading to a failure of treatment clinically. Therefore, a detailed understanding of how cells can bypass targeted inhibition of signaling cascades crucial for malignant growths is necessary. Here, we have performed an in vitro experiment that investigates kinetics and mechanisms underlying resistance development in former drug sensitive cancer cells over time in vitro. We show that the dynamics observed in these experiments can be described by a simple mathematical model. By comparing these experimental data with the mathematical model, important parameters such as mutation rates, cellular fitness and the impact of individual drugs on these processes can be assessed. Excitingly, the experiment and the model suggest two fundamentally different ways of resistance evolution, i.e. acquisition of mutations and phenotype switching, each subject to different parameters. Most importantly, this complementary approach allows to assess the risk of resistance development in the different phases of treatment and thus helps to identify the critical periods where resistance development is most likely to occur.     

New Approaches of PARP-1 Inhibitors in Human Lung Cancer Cells and Cancer Stem-Like Cells by Some Selected Anthraquinone-Derived Small Molecules

Poly (ADP-ribose) polymerase-1 (PARP-1) and telomerase, as well as DNA damage response pathways are targets for anticancer drug development, and specific inhibitors are currently under clinical investigation. The purpose of this work is to evaluate anticancer activities of anthraquinone-derived tricyclic and tetracyclic small molecules and their structure-activity relationships with PARP-1 inhibition in non-small cell lung cancer (NSCLC) and NSCLC-overexpressing Oct4 and Nanog clone, which show high-expression of PARP-1 and more resistance to anticancer drug. We applied our library selected compounds to NCI''s 60 human cancer cell-lines (NCI-60) in order to generate systematic profiling data. Based on our analysis, it is hypothesized that these drugs might be, directly and indirectly, target components to induce mitochondrial permeability transition and the release of pro-apoptotic factors as potential anti-NSCLC or PARP inhibitor candidates. Altogether, the most active NSC747854 showed its cytotoxicity and dose-dependent PARP inhibitory manner, thus it emerges as a promising structure for anti-cancer therapy with no significant negative influence on normal cells. Our studies present evidence that telomere maintenance should be taken into consideration in efforts not only to overcome drug resistance, but also to optimize the use of telomere-based therapeutics. These findings will be of great value to facilitate structure-based design of selective PARP inhibitors, in general, and telomerase inhibitors, in particular. Together, the data presented here expand our insight into the PARP inhibitors and support the resource-demanding lead optimization of structurally related small molecules for human cancer therapy.     

An ecological resilience perspective on cancer: Insights from a toy model

In this paper we propose an ecological resilience point of view on cancer. This view is based on the analysis of a simple ODE model for the interactions between cancer and normal cells. The model presents two regimes for tumor growth. In the first, cancer arises due to three reasons: a partial corruption of the functions that avoid the growth of mutated cells, an aggressive phenotype of tumor cells and exposure to external carcinogenic factors. In this case, treatments may be effective if they drive the system to the basin of attraction of the cancer cure state. In the second regime, cancer arises because the repair system is intrinsically corrupted. In this case, the complete cure is not possible since the cancer cure state is no more stable, but tumor recurrence may be delayed if treatment is prolongued. We review three indicators of the resilience of a stable equilibrium, related with size and shape of its basin of attraction: latitude, precariousness and resistance. A novel method to calculate these indicators is proposed. This method is simpler and more efficient than those currently used, and may be easily applied to other population dynamics models. We apply this method to the model and investigate how these indicators behave with parameters changes. Finally, we present some simulations to illustrate how the resilience analysis can be applied to validated models in order to obtain indicators for personalized cancer treatments.     

Standing genetic variation and the evolution of drug resistance in HIV

Drug resistance remains a major problem for the treatment of HIV. Resistance can occur due to mutations that were present before treatment starts or due to mutations that occur during treatment. The relative importance of these two sources is unknown. Resistance can also be transmitted between patients, but this process is not considered in the current study. We study three different situations in which HIV drug resistance may evolve: starting triple-drug therapy, treatment with a single dose of nevirapine and interruption of treatment. For each of these three cases good data are available from literature, which allows us to estimate the probability that resistance evolves from standing genetic variation. Depending on the treatment we find probabilities of the evolution of drug resistance due to standing genetic variation between 0 and 39%. For patients who start triple-drug combination therapy, we find that drug resistance evolves from standing genetic variation in approximately 6% of the patients. We use a population-dynamic and population-genetic model to understand the observations and to estimate important evolutionary parameters under the assumption that treatment failure is caused by the fixation of a single drug resistance mutation. We find that both the effective population size of the virus before treatment, and the fitness of the resistant mutant during treatment, are key-arameters which determine the probability that resistance evolves from standing genetic variation. Importantly, clinical data indicate that both of these parameters can be manipulated by the kind of treatment that is used.