Abstract
Several drugs have shown in vitro and in vivo pharmacological activity against urinary bladder cancer. This review aims at compiling the different drugs evaluated in in vitro and in vivo models of urinary bladder cancer and to review the advantages and limitations of both types of models, as well as the different methodologies applied for evaluating antineoplastic drug activity.
Cancer is one of the most important public health issues and the most feared human disease (1). It is the second leading cause of death after coronary heart diseases and one in three persons suffers from cancer throughout their lives and one in four will die from this disease (2). Urinary bladder cancer is a common disease that ranks ninth in worldwide cancer incidence. It is the fourth most common cancer in men and the ninth in women, with a probability of developing in men three-times higher than in women, and with a ratio of 2:1 for caucasians and negros, respectively (3). The risk for developing this disease increases with age, with a peak between 60 and 70 years (4). Remarkable differences can be found in its incidence, it being predominately higher in developed countries such as in North America, Western and Southern Europe (5). In less industrialized countries, such as in Asia, Africa and the Middle East, the incidence of urinary bladder cancer is lower except for regions where Schistosoma haematobium is endemic. In these cases, urinary bladder squamous cell carcinoma is common (6). Involving exogenous and endogenous factors, the aetiology of urinary bladder cancer is multifactorial (Figure 1) (7, 8). Cigarette smoking, and environmental and occupational exposure to chemical agents remain the two major risk factors (7).
Urinary bladder cancer is classified into three main types: transitional cell carcinoma, squamous cell carcinoma and adenocarcinoma. At minor percentages are the small-cell tumours (1%) and sarcomatoid tumours (fewer than 1%) (9). Accounting for more than 90% of all cases, transitional cell carcinoma is the most common form of urinary bladder cancer (10). At diagnosis, nearly 70% of patients with urinary bladder cancer present with non-muscle-invasive lesions. Several clinical factors, such as tumour multiplicity, diameter, concomitant carcinoma in situ (CIS) and gender, have been identified as having prognostic significance for recurrence (11). CIS represents a major concern in the treatment of non-muscle-invasive lesions. CIS is a high-grade lesion that is characterized by disorderly proliferation of cells with marked cytological abnormalities (12). Although the European Association of Urology recommends transurethral resection and intra-vesical Bacillus Calmette-Guerin (BCG) immunotherapy for patients with CIS lesions, which achieves a complete response rate, 20% of patients will ultimately die of metastatic disease (12-14). The remaining 30% of patients at diagnosis have muscle-invasive lesions and 10% of these cases has a tendency to metastasize, with a poor prognosis (15).
Urinary Bladder Cancer Treatment
Treatment of non-muscle-invasive lesions. Complete transurethral resection is the standard treatment for non-muscle-invasive lesions (16). Despite good prospects of survival (success rate of 80%), these tumours recur in approximately 70% of patients (10). One of the major challenges in treating non-muscle-invasive tumours is to reduce the high frequency of early recurrences, detected in more than 45% of the patients, three months following transurethral resection (17). In order to reduce the recurrence risk and to delay or prevent progression to a muscle-invasive lesion, after transurethral resection, adjuvant intra-vesical instillations of chemotherapy or immunotherapy are widely applied with mitomycin C (MMC) and BCG, respectively (16). Use of BCG not only reduces the recurrence rate, but also reduces the risk of a non-muscle-invasive lesion progressing to a muscle-invasive lesion, improving the overall survival (7). The role of BCG in urinary bladder cancer was clinically studied for the first time in 1976, by Morales and collaborators, which found a complete response in seven out of nine patients treated (18). Although treatment with BCG provides better results than transurethral resection without immunotherapy, side-effects arising from its administration are a concern. The development of sepsis, cystitis, dysuria and mild haematuria are frequently reported (19). However, with increased experience in using BCG, the side-effects now appear to be less prominent (20). Intravesical chemotherapy with MMC, epirubicin and doxorubicin have all shown comparable beneficial effects (21).
Treatment of muscle-invasive lesions. The treatment options that are currently available for the management of muscle-invasive urinary bladder cancer include radical cystectomy and chemotherapy-plus-radiation therapy, with the goal of bladder preservation. Combined chemotherapy based on methotrexate, vinblastine, adriamycin and cisplatin (MVAC) was initiated in the 1980s, leading to a disease-free survival rate of 3.7% at six years (22). However, this protocol was highly toxic, with severe side-effects, being associated with a mortality rate of about 4% (23). Thus, new approaches are being continuously investigated to provide superior efficacy with lower toxicity (24).
In Vitro and In Vivo Models for the Study of Urinary Bladder Cancer
Experimental models are used to better explain tumour behaviour, to evaluate the effect of chemopreventive agents, and to study the efficacy of antineoplastic drugs (25). Such experimental research can be achieved by means of in vitro and in vivo models.
In vitro models. To date, cultured urinary bladder cells represent the most frequently used in vitro bladder cell model. These models usually consist of isolated urinary bladder cancer cell lines and have been established as a valid in vitro model not only to study the mechanism involved in urinary bladder cancer development but also to evaluate anti-neoplastic drug efficacy (26). In 1970, Rigby and Franks established the first human urinary bladder cancer cell line, designated as RT4 (27). Since then, many other human urinary bladder cancer cell lines have been established and characterized according to their origin, grade and stage. A great proportion of these cell lines was established from invasive and metastic tumours, benefiting the investigation of late tumour progression and metastic lesions. On the other hand, few non-muscle-invasive human urinary bladder cancer cell lines are available, which is a disadvantage in the investigation of non-muscle-invasive urinary bladder cancer (26). Urinary bladder cancer cell lines may also be established from rodents exposed to urothelial chemical carcinogens. In 1971, Toyoshima and collaborators established the Nara urinary bladder cancer II (NBT-II) cell line, a rat cell line obtained from a urinary bladder tumour chemically induced by N-Butyl-N-(-4-hydroxybutyl) nitrosamine (BBN) (28). In the same year, three more urinary bladder cancer cell lines were established from tumours induced by the combined use of N-2-fluorenylacetamide and cyclophosphamide in Fischer 344 female rats, two of them epithelial (BC5 and BC6) and one fibroblastic (BC7) (29). Five years later, two mouse urinary bladder cancer cell lines were established using the carcinogen N-[4-(5-nitro-2-furyl)-2-thiazolyl]formamide (FANFT) and, more recently, seven chemically-induced mouse urinary bladder cancer cell lines (BC13, BC29, BC30, BC46, BC57, BC58 and BC59) were established from tumours developed in C57BL/6 mice exposed to BBN (30). Table I summarizes commercially available human and rodent urinary bladder cancer cell lines.
Under specific culture conditions, normal human urothelial cells may be also established from surgical specimens of urinary bladder (31). These normal cells can be used to evaluate mechanisms by which therapeutic agents interact with normal human urothelial cells and to verify if antineoplastic drugs can induce cell damage. Tumours are, however, three-dimensional complexes in which there is great interaction between tumourous and non-tumourous cells. Thus, cell culture represents an artificial system for investigating features such as vascularization and perfusion (32). Urothelium may be maintained in combination with stroma as a three-dimensional heterotypic or organotypic culture. Such organoids may either be established from intact tissues, or recombined from reconstructed urothelial and stromal compartments prior to culture and may be maintained up to 20 weeks in culture (33). Chang and collaborators were pioneers in using human urinary bladder tumour specimens in three-dimensional histoculture to assess the activity of a new platinum analog (34). The growth of tumour cells as three-dimensional multicellular spheroids in vitro has led to important insights in tumour biology, since properties of the in vivo tumour, such as proliferation, and nutrient gradients, can be studied under controlled conditions (35). Moreover, three-dimensional culture allows for an understanding of basic paracrine signalling mechanisms that regulate tissue homeostasis, development of new methods for urinary bladder reconstruction and tissue engineering, and generation of models of malignant and benign diseases (33). Furthermore, isolated organs allow an approach towards the assessment of organ physiology and morphology, providing with models that mimic conditions in humans more closely (26). Several years ago, Burgués and collaborators tested several drugs (epirubicin, thiotepa, adriamycin, MMC, verapamil and ciprofloxacin) on ex vivo spheroids of non-muscle-invasive urinary bladder cancer. Their study suggests that use of three-dimensional urinary bladder cultures could be a possible approach in clinical practice to select for the best antineoplastic drug for each patient and to investigate the effect of new antineoplastic drugs or drug combinations (36).
In vivo models. Prediction of drug activity in patients with cancer based only on in vitro studies is not reliable, and animal models are widely recognized as being essential to the study of antineoplastic drug efficacy (32). Animal models were defined by Wessler in 1976 as “living organisms with an inherited naturally-acquired or induced pathological process that closely resembles the same phenomenon in man”. Their application in biomedical research, and pathophysiological and toxicological studies, allows us to determine aetiological factors, study cancer development and progression, and to develop new medical devices or therapies (37).
For an appropriate and valid animal model for the study of urinary bladder carcinogenesis to exist, several requirements are necessary: the tumour should be of urothelial origin with different stages of disease progression, it should grow intravesically so as to be directly exposed to antineoplastic drugs, it should mimic the pathogenesis of human urinary bladder cancer, and it should present stable molecular and genetic alterations similar to those found in human urinary bladder cancer (38). Several animal species such as dogs, rabbits, guinea pigs, and hamsters may be used. However, rats and mice are the animals most employed, due to their small size, innumerable anatomical, physiological and biochemical similarities to humans, clear genetic background and high reproductive rate. Furthermore, the occurrence of spontaneous tumours in laboratory rodents is uncommon, which is one reason why they are considered as ideal models for the study of chemically-induced urinary bladder cancer (38).
Recently, a study conducted by Palmeira and collaborators described the similarities found between human urinary bladder carcinogenesis and rodent chemically-induced urinary bladder carcinogenesis in regards to the histopathological features and biological alteration profile, namely: DNA aneuploidy, p53 overexpression and high Ki-67 proliferative index. They reported that the spectrum of lesions chemically-induced in rodents, such as hyperplasia, dysplasia, low- and high-grade tumours, CIS and invasive urothelial carcinoma are similar to those observed in humans (39).
Experimentally induced urinary bladder tumours. Three models are currently available for inducing urinary bladder tumours: chemically-induced, genetically-engineered, and transplantable (xenograft and syngeneic animals) (Figure 2) (40). A wide range of compounds are described as urinary bladder carcinogens, but of all, BBN is the most frequently used chemical carcinogen (41). However, it is possible to induce urinary bladder cancer with FANFT and N-Methyl-N-nitrosourea (MNU). The selection of each of these agents is made according to the animal facilities and aims of the experimental work (42). The xenograft model is the most routinely used transplantable model and BBN is the carcinogen most frequently applied to induce bladder tumours.
Advantages and Disadvantages of In Vitro and In Vivo Models
As yet, there is not an ideal experimental model for urinary bladder cancer study since both in vitro and in vivo models have limitations. However, with the information obtained from both models, a better understanding of urothelial bladder carcinogenesis is possible. In pre-clinical studies, the antitumour efficacy of a new drug is first evaluated in in vitro models and later in animal models. One of the greatest advantages of in vitro models application is that they offer the possibility to maintain cells in completely controlled environmental conditions, allowing the study of specific cellular and molecular pathways in shortened experimental timescales, being less expensive than animal models and less time-consuming. In contrast, the greatest limitation of this model is that cells growing in vitro are not the exact dissociated replicates of their in vivo counterparts. The use of monolayer cell cultures is usually restricted to a single or at most two cell types. Tumours are composed not only of neoplastic cells but also of stroma and inflammatory cells, which gives a tumour a three-dimensional structure, interacting and influencing its growth (33). The impossibility of tumour angiogenesis and metastasis studies can be considered as limitations of in vitro studies, since these are complex processes with many different mechanisms involved. It is, therefore, clearly difficult to perform in vitro assays which totally simulate these processes and only a combination of methods will be able to provide a clear picture (43). In vitro studies can also provide important information concerning the parameters of pharmacodynamics. To better-understand pharmacokinetics, it is necessary to use in vivo models, since these models offer the best approach for effectively combining and interpreting the major determinants of drug kinetics across species (44). Likewise, in vitro studies do not predict the adverse effects of drugs (45).
For this reason, in vivo models remain important: they preserve the three-dimensional tumour structure with cell-cell interactions and allow for pharmacokinetic and toxicity evaluation of the compounds. Significant limitations of in vivo studies include the necessity for animal facilities, they are also more time consuming, and involve high involved (40). The advantages and disadvantages of in vitro and in vivo models, as well as their possible applications are summarized in Figure 3.
In Vitro Studies to Assess the Efficacy of Antineoplastic and Other Drugs
In the past thirty years, more than 40 studies were conducted in order to evaluate the activity of antineoplastic drugs, making use of several human urinary bladder cancer cell lines, as well as a wide range of methodologies (Table II). It is clearly perceptible that out of the 40 different urinary bladder cancer cell lines used, the T24 muscle-invasive cell line is the one most employed, followed by the HT1376, RT112 and RT4 cell lines. Currently, there is a broad range of methodologies available to assess the in vitro efficacy of drugs, but assays such as 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), clonogenic, flow cytometry and western blot are regularly used. In vitro models have already provided the basis for study of the activity of many antineoplastic drugs. However, alkylating drugs, namely cisplatin and MMC, as well as anti-metabolite drugs, such as gemcitabine, are among the ones most investigated, whether in isolation or combination therapy. Other antineoplastic drugs with different mechanisms of action have also been analyzed, examples are inhibitors of the mammalian target of rapamycin (mTOR) (rapamycin, everolimus), topoisomerase II (epirubicin, doxorubicin, etoposide) and of mitotic spindle formation (paclitaxel, vincristine, vinorelbine). Many of these drugs already tested in vitro (approximately 26) have progressed to clinical studies of urinary bladder cancer. Nevertheless, none of them demonstrated superior efficacy when compared to the drugs currently used in urinary bladder cancer treatment, which is why none of them have been approved as a new therapy.
Through analysing Table III it is clear that non-antineoplastic drugs have also been studied using human urinary bladder cancer cell lines. Similarly to what happens with antineoplastic drugs, T24 and MTT remain the most used urinary bladder cancer cell line and the most frequent methodology applied, respectively. More than 20 compounds classified as non-antineoplastic drugs were tested, being in the majority non-steroidal anti-inflammatory drugs (NSAIDs). However, very few of these drugs have advanced to clinical studies.
Rodent urinary bladder cancer cell lines are used to assess the efficacy of antineoplastic and other drugs. The mouse MBT-2 and the rat AY-27 urinary bladder cancer cell lines are widely employed (Table IV).
In Vivo Studies to Assess the Efficacy of Antineoplastic and Other Drugs
As previously stated, in vivo models can be used to evaluate drug efficacy. As shown in Table V, in the last twenty years since chemical carcinogens were discovered, more than 25 studies were carried out with rats and mice to analyze antineoplastic drug efficacy in urinary bladder tumours, chemically-induced or implanted. The chemical induction of urinary bladder tumours is commonly implemented using the carcinogen BBN, and for tumour implantation, rodent (MB49, MBT-2, AY-27) and human (HT1376, KU-19-19, 5637, RT4, T24, UM-UC-3) urinary bladder cancer cell lines may be used. Immunotherapy with BCG, in isolation or combined with other drugs, has been one of the therapeutic approaches most extensively investigated in rodent models of urinary bladder cancer. Similarly to in vitro studies, in in vivo models, the use of NSAIDs and other drugs are common, using both chemically-induced and implanted urinary bladder tumours. We verify that the association of two or more drugs is frequent and histopathology and immunohistochemistry are commonly carried out, the expression of Ki-67, proliferating cell nuclear antigen (PCNA), tumour protein 53 (p53), cyclin, cyclo-oxygenase-2 (COX-2), vascular endothelial growth factor (VEGF) and platelet endothelial cell adhesion molecule (CD31) being widely evaluated on experimental models of urinary bladder cancer. In order to complement the information obtained from the investigation, many studies evaluate the activity of the same drug in both in vivo and in vitro models (Table V and Table VI). None of the drugs tested in vivo, not tested in vitro has advanced to clinical trials, these drugs did not eradicate tumours but only restricted their growth.
Rodent models are particularly valuable for defining the molecular pathways participating in urothelial cell transformation and disease progression, for identifying modifier genes that affect penetrance of the manipulated genes, and for testing various therapeutic and preventive approaches.
Methodologies for Assessing Drug Activity
The use of reliable methodologies to determine and quantify the efficacy of several drugs facilitates the selection of promising candidate drugs for clinical trials (32). As shown in Figure 4, over the years, several methodologies have been developed and used to evaluate the efficacy of different drugs in in vitro and in vivo models. Tests for cytotoxicity and cellular growth inhibition are the oldest and most commonly used assays (43). For in vitro experiments, it should be borne in mind that if a substance leads to a reduced number of viable cells in comparison to the untreated cells after the incubation period, this can be due to the death of some cells, but without affecting growth of another sub-population, or to a general deceleration of growth but survival of all cells. Thus, several assays have been established to measure not only the viable cell number, but also to differentiate between the arrest of growth and cytotoxicity (43). The earliest test methods introduced the use of MTT, and 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide (XTT), and are based on cellular metabolism. The reduction of these colourless tetrazolium salts into coloured products is correlated with the number of viable cells. However, these assays do not differentiate between the growth arrest and cell death. Thus, it is necessary to additionally scrutinize the number of dead cells, commonly analyzed by microscopic analysis (trypan blue) or flow cytometry (propidium iodide). Other methods commonly applied for measuring cell growth are based on the incorporation of thymidine analogues, such as the 3[H]thymidine and the 5-bromo-2’-deoxyuridine (BrdU) incorporation assays (43). However, even if a drug is effective, this evidence does not necessarily translate into efficacy, we can only see an effect for which our assay was specifically designed. If a substance gives good results when tested, it does not necessarily mean that this substance will be effective (43, 46). Taking this into consideration, it is absolutely necessary to establish guidelines for performing in vitro studies in order to create a basic framework with good predictive proprieties (43, 45).
Conclusion
In the present article, we have reviewed experimental data related to the investigation of antineoplastic drugs for urinary bladder cancer, highlighting the use of in vitro and in vivo models. Used for many decades, well-established in vitro and in vivo models have been available for experimental evaluation of new antineoplastic drugs, providing invaluable pharmacological and toxicological data that may predict for the clinical efficacy of new compounds.
Standard cell culture studies are widely used to delineate the biological, chemical and molecular cues of living cells. Urinary bladder cancer cell lines established from human tumours of different stages and grades allow us to understand the heterogeneous response observed for the same drug between different patients. However, three-dimensional models that recapitulate the tumour microenvironment remain essential. In vivo models overcome this in vitro drawback. The development and optimization of reliable, sensitive and reproducible methodologies are indispensable tools for assessing the in vitro and in vivo efficacy of novel antineoplastic drugs.
From this review, we conclude that to date, more than 50 drugs have been tested on urinary bladder cancer, with BCG, cisplatin, gemcitabine and MMC being the most common, in isolation or applied simultaneously with other drugs, in in vitro or in vivo models. Nowadays, combination therapy is particularly important since the actions of target-based drugs may be supplemented or potentiated by other drugs. This therapeutic approach is increasingly used in experimental models.
Novel targeted-therapies are needed to further improve the chemotherapy efficacy of urinary bladder cancer treatment. Some of the drugs evaluated and described in this article have been shown to be promising in the treatment of this disease. Currently, drugs targeting angiogenesis are promising. Therapeutic investigations should be continued, with the development of new drugs, as well as of targeted therapies to improve treatment results for bladder cancer.
Acknowledgements
This study was supported by a grant from the Fundação para a Ciência e Tecnologia, Ministério da Educação, Portugal, grant number SFRH/BD/47612/2008 and by FCT Pest-OE/AGR/UI0772/2011 unity.
- Received January 21, 2013.
- Revision received March 7, 2013.
- Accepted March 7, 2013.
- Copyright© 2013 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved