Spheroid Cultures for Cancer Research and Treatment

Ron Allison, M.D.

Introduction

Spheroid models are sphere-shaped cell colonies that permit growth and function studies of diverse normal and malignant tissues. Spheroid growth mimics the growth of naturally occurring human tumors. In both, cells in contact with nutrients grow quickly while the kinetics and activity of cells further inward depend on diffusion.

Sophisticated microtechniques permit analysis of whole spheroids, spheroid subpopulations within the colony, or even individual cells. Manipulations of the spheroid environment have provided insights into tumor physiology, and research with therapeutic agents, such as radiation, offers hope for improved cancer treatment.

History

Since the late 1950s, multicellular aggregates of malignant cells have been analyzed for proliferation, differentiation, invasion potential, and structural similarity to human tumors on a large scale.¹ However, it was not until the early 1970s that Sutherland and coworkers systematically investigated the response of tumor cell aggregates to antineoplastic therapy. Because the cell lines formed nearly perfect sphere-shaped aggregates, they were called "spheroids."²

Spheroid Cultures

A wide range of normal and malignant cell types can be grown as spheroids. Embryonic, fetal, postnatal, and adult cells from virtually all organ systems, including the central nervous system, may be maintained in culture for several weeks. Recently, cells that would not form spheroids have been induced to do so by co-culturing with spheroid-forming non-clonogenic feeder cells.³

Histologic examination of these spheroids reveal striking similarities between their structure and that of naturally occurring human tumors. Furthermore, the response of culture spheroids to cancer therapy is so similar to the response obtained in humans that many investigators consider spheroids useful for studying cancer cells' basic biological properties.⁴

Spheroids may be used for studies of viability, clonogenicity, LD5O, and metastatic potential under a variety of experimental conditions. In certain systems, spheroids can be mechanically or chemically "peeled like an onion" so that specific cell subpopulations can be isolated and studied.⁵ Research on spheroids grown from a biopsy of a patient's tumor may one day allow individualized treatment that would increase the chance of a cure.

Common Characteristics of Spheroids and Tumors

Analysis of human tumors and their spheroid counterparts has revealed numerous structural, functional, and biological similarities.⁶ Structurally, spheroids are virtually indistinguishable from samples taken from the original tumor. Functional analysis of spheroids reveals that their secretory and metabolic activity is similar to that of the tumors or organs from which the spheroids derive. For example, human colon carcinoma spheroids secrete measurable quantities of carcinoembryonic antigen, thyroid cell spheroids produce thyroid hormones, and hepatocyte spheroids can exhibit the same metabolic functions as liver tissues. These characteristics are not necessarily seen when the same cells are cultured as monolayers.⁴

Biologically, both spheroids and natural tumors are composed of heterogeneous cells. In naturally occurring tumors, cells close to blood vessels are well-supplied with oxygen and nutrients, and they actively proliferate and secrete various substances. Cells farther from the nutrient vessels are progressively more hypoxic and acidic, grow more slowly, and appear to be relatively resistant to antineoplastic therapy.

Spheroids share these characteristics. Cells close to the sphere surface, in contact with oxygen and with environmental nutrients, actively proliferate and secrete various substances. Further inward, cell viability depends on diffusion of nutrients in and waste out. Interestingly, the areas of hypoxia in spheroid cultures occur 200 microns from the sphere surface, and analysis of human tumor biopsies reveals areas of hypoxia 200 microns from nutrient vessels.

Because the spheroid environment can be controlled and manipulated, milieu effects on tumor cell viability can be carefully examined. Using microtechniques and automated cell sorting, each cell can be evaluated and the effects of varied growth factors, such as nutrients, hormones, and therapeutic interventions, can be systematically studied and analyzed. Using these techniques, certain areas of spheroids have been identified that contain

chronically hypoxic cells resistant to all forms of therapy. These studies, as well as research with human tumors in vivo, support the belief that the chronically hypoxic cells are often responsible for cancer treatment failure.^5,7 Much research is being conducted in the area of therapeutic techniques to promote cancer cell oxygenation. For example, hypoxic cells produce specific proteins,⁸ and manipulation of these proteins may influence cell viability and growth.

Cell-to-cell communication within tumors can also be studied using spheroids. In spheroids, as in naturally occurring tumors, numerous communication channels, including gap junctions, desmosomes, and electrical coupling, develop. Studies of these processes may provide insight into molecular mechanisms regulating cell proliferation and differentiation.

An exciting area of research involves co-culturing normal cells with malignant cells within a spheroid. Tumor invasion of the normal cell often occurs. This technique, called "confrontation," appears to be a valid method of evaluating invasion.⁹ Modifications, including the use of human amnion as tissue with spheroid tumor aggregates as metastatic lesions, have proved valuable in the study of metastasis.¹⁰

Spheroids are also used to elucidate immune factors in tumor dynamics. Sutherland developed an in vitro system analyzing the response of tumor spheroids to diverse human lymphocytes.¹¹ He found that diminished clonogenicity follows certain tumor-lymphocyte reactions. Another area of active research involves evaluation and modification of synthetic antibodies for therapeutic effect.

Therapy

Therapeutic interventions using spheroids have received wide attention. Modalities include radiation, chemotherapy, hyperthermia, immunotherapy and combinations of these. For example, researchers have found that irradiated spheroids are more resistant than monolayers to cell death because hypoxic areas within spheroids are radioresistant. Since hypoxic areas in naturally occurring tumor may limit curability, modification or elimination of these cells is necessary. One promising approach utilizes agents that label the hypoxic fraction.¹² Also, radiation sensitizing agents have been developed using spheroids for screening and testing,⁵ but clinical efficacy remains to be determined. Interestingly, spheroid analysis has shown that multiple-dose radiation regimens, similar to those used clinically, are able to minimize hypoxic cell populations and increase cancer cell kill.^5,7

Analysis of the effects of chemotherapeutic agents has shown that certain drugs will not diffuse far enough into spheroids to achieve adequate cell kill. Testing modified drug structure on spheroids can lead to agents of improved effectiveness.¹³

Certain cells within spheroids may develop genetic resistance. Perhaps new and more effective agents can be developed once these cells are isolated and evaluated. Additionally, by tagging the chemotherapeutic agents, microanalysis can reveal which cell sub-populations are most affected by particular drugs. In this way, combinations of chemotherapeutic agents may be evaluated to optimize tumor kill. Finally, new antiproliferative agents may be screened and analyzed using spheroids, which offer a promising approach to drug discovery.¹³

Spheroids have already modified immunotherapy. Initially, anti-tumor antibodies were too large to penetrate effectively into spheroid tumors. This was also true of antibodies used against human tumors. The use of spheroids to study immune therapy has led to a new generation of cytotoxic antibodies with increased tumor penetration.¹⁴ These are in the process of clinical trials.

Eventually, spheroids grown from a patient's tumor may be used to test the efficacy of a wide range of therapeutic options. This approach is not readily accomplished using monolayer cell cultures because monolayers lack the extensive cell-to-cell contact, the hypoxic cell populations, and the variable cell cycle times seen in both spheroids and naturally occurring human tumors.¹⁵ Most importantly, monolayer cultures generally employ established, immortal cell lines rather than biopsy specimens from actual tumors of patients. Because monolayers select for immortal cells, they may not be representative of most cells within a tumor. Nevertheless, monolayer cell cultures retain some value in screening new therapeutics because they are fully automated and reliable and many different human cell lines are available for testing.¹⁶

Use of animal models to screen therapeutic regimens and individualize therapy is not practical. The clinician cannot withhold cancer therapy for the months required to grow tumor cells in animals and to test treatment modalities.⁵ Additionally, it is virtually impossible to obtain from the patient the amount of tumor mass needed for transplantation into the many animals required to test the many individual and combination therapies available. Also, the costs of animals, manpower, and laboratory space would be prohibitive. Most importantly, these systems are unreliable. While the tumor cells are of human origin, the animal's immune response, metabolism, and toxic response to therapy all influence outcome. Even inbred animal strains can have variable tumor responses.^5,17

Future Applications

Spheroids are an established model for biological and therapeutic studies of malignancy. They are reliable, reproducible, and economical.⁵ As technology advances, the applications of spheroids will increase. High resolution MRI, MRS, and PET are already being used to evaluate tumor physiology within intact spheroids. Perhaps the greatest contribution of spheroids will come once the biopsies of cancer patients are grown as spheroids. These cultures may then serve to individualize and optimize the best therapies available. This approach is currently under evaluation by the Radiation Therapy Oncology Group, a consortium of leading clinical and experimental radiation oncologists.

References

1. Moscona A: The development in vitro of chimeric aggregates of dissociated embryonic chick and mouse cells. Proc Natl Acad Sci 1957;43:184-194.

2. Sutherland RM, McCredie JA, Inch WR: Growth of multicellular spheroids in tissue culture as a model of nodular carcinomas. J Nat Canc Inst 1971;46:113-120.

3. Djordjevic B, Lange C: Clonogenicity of mammalian cells in hybrid spheroids: A new assay method. Radiat Environ Biophys 1990;29:31-46.

4. Mueller-Klieser W: Multicellular spheroids. J Canc Res Clin Onc 1987;113:101-22.

5. Hall E.J: Radiobiology for the radiologist. Philadelphia, Lippincott, 1988.

6. Nederman T, Norling B, Glemelius B, Carlsson J, Bruntz U: Demonstration of an extracellular matrix in multicellular spheroids. Cancer Res 1984;44:3090-3097.

7. Sutherland RM: Cell-environment interaction in tumor microregions: The multicell spheroid model. Science 1988;240;177-184.

8. Freyer JP, Sutherland RM: Regulation of growth saturation in the development of necrosis in spheroids. Cancer Res 1986;46:3504.

9. Schleich A: The confrontation of normal and malignant cells in vitro, in Garattino S (ed): Chemotherapy of Cancer Dissemination with Metastasis. New York, Raven Press, 1973, pp 5 1-58.

10. Marcel MM: The use of culture to study invasion in vitro, in Liotta LA, Hart IR (eds): Tumor Invasion and Metastases. The Hague, Martinus Nijhoff, 1982, pp 207-230.

11. Sutherland RM, MacDonald HR, Howell RL: Multicellular spheroids: A new model target for in vitro studies of immunity. J Nat Canc Inst 1977;58:1849-1853.

12. Franko AJ: Hypoxic fraction with binding of misonidazole in multicellular tumor spheroids. Radiation Tes 1985;103:89-97.

13. Mederman T: Growth of tumor cells as spheroids and antitumor drug evaluations, in Dendy PP, Hall BT (eds): Human Tumor Drug Testing In Vitro. London, Academic Press, 1983, pp 147-161.

14. Sutherland RM, Bucheggcr F, Shreyer M, Vacca A, Mach J-P:Penetration and Binding of radiolabeled anti-carcinoembryonic antigen monoclonal antibodies and their antigen binding fragments in human colon multicellular tumor spheroids. Cancer Res 1987;47:1627-1633.

15. Sutherland R, Durand R: Radiation response in multicell spheroids. Current Topics Radiation Res 1976;11:87-139.

16. Boyd M: Status of the NCI preclinical antitumor drug discovery screen, PPO Updates of Cancer 1989;Oct(10).

17. Kallman R: Rodent Tumors in Experimental Cancer Therapy. New York: Pergamon, 1987.

Contents