Osteosarcoma accounts for the majority of deaths from bone tumors, which are the third most-common cause of cancer-related death in children and young adults [19, 31]. Before the 1960s, the 5-year overall survival rate for osteosarcoma was approximately 20% despite amputation of the extremity with a primary tumor. Investigators at the time concluded that subclinical micrometastases were present in the majority of patients at diagnosis and were responsible for late recurrence and death [6, 14, 15]. Osteosarcoma has since been considered a systemic disease regardless of whether metastases are detectable at presentation, and adjuvant chemotherapy has become the standard of care. Multiple clinical trials in the 1970s to 1980s identified the combination of methotrexate, doxorubicin, and cisplatin (MAP) to be highly effective against osteosarcoma [19, 22]. Overall survival rates improved to 60% to 70% by the mid-1980s. However, the available chemotherapeutic options and thus survival have not changed substantially for 30 years . Novel therapeutics therefore are needed to target the progression of drug-resistant metastases in osteosarcoma.
Osteosarcoma presents a unique challenge to the existing drug discovery infrastructure because of prohibitive costs of developing new agents and the relative scarcity of patients available for clinical trials. A 2014 meeting of osteosarcoma experts called for candidate drugs to be rigorously evaluated in preclinical micrometastatic models to prioritize agents with activity against the progression of established metastases, the lethal disease process . Three-dimensional (3-D) multicellular tumor spheroids have been used for more than 40 years in cancer research as an intermediate between monolayer culture and in vivo studies. Three-dimensional spheroids are produced by various in vitro methods, including continuous agitation of a cell suspension or cellular aggregation by sedimentation through centrifugation, a hanging droplet, or microfluidics to restore the histomorphologic, functional, and microenvironmental features of in vivo human tumor tissue . Numerous studies have detailed the ability of spheroids to establish cell-cell interactions, cell-matrix interactions, diffusion gradients, growth kinetics, and genetic profiles that closely mimic corresponding microtumors in vivo [5, 8, 10, 11, 18, 21]. Not surprisingly, spheroids show chemoresistance and better correlate with the in vivo response to chemotherapy when compared with monolayer cultures [2, 10, 11, 18, 27]. These characteristics, combined with ease of production and high degree of uniformity through a centrifugation-based method, make spheroids a promising assay for preclinical drug screening in osteosarcoma.
This laboratory study evaluated application of osteosarcoma spheroids (sarcospheres) for drug screening to develop a high-yield assay to identify novel therapeutics. Specific purposes of this study were: (1) to characterize sarcosphere size in multiple highly metastatic human osteosarcoma cell lines; (2) to establish accurate measurement of sarcosphere growth; (3) to verify intra- and interexperimental uniformity; and (4) to apply this platform to conventional methotrexate, doxorubicin, and cisplatin (MAP) chemotherapy to identify variability in chemoresistance across multiple cell lines.
Materials and Methods
Sarcospheres were first characterized to establish accurate measurement of sarcosphere growth and uniform production before application to evaluate MAP chemotherapy (Fig. 1). To address the first purpose of the study, to characterize sarcosphere size, sarcospheres were seeded at various densities followed by quantitative measurement of sarcosphere diameter and volume. To address the second purpose of the study, to accurately measure sarcosphere growth, resazurin reduction was quantitatively assessed, with and without EDTA, with time and compared with sarcosphere volume to identify conditions that accurately reflect growth. The third purpose of the study, to confirm sarcosphere uniformity, was addressed by quantitatively measuring diameter and resazurin reduction at treatment Day 0 and the Z’ factor, a measure of assay suitability for high-throughput screening, at treatment Day 2.
The refined platform then was applied to evaluate MAP chemotherapy, the fourth purpose of this study, to validate its future use in drug screening. Sarcospheres were treated with individual MAP agents (0 to 1000 μmol/L) to quantitatively determine concentrations at which 50% of growth from Days 0 to 2 was inhibited (GIC50). Qualitative analysis of sarcosphere response to MAP was assessed by fluorescent staining for live, dead, and apoptotic cells. Cell lines resistant to MAP in sarcospheres were treated in monolayer for comparison. Future studies will apply this platform to screen drugs in osteosarcoma before in vivo evaluation.
Cultures were maintained for three highly metastatic osteosarcoma cell lines. The 143B cell line , from the parental HOS-TE85 cell line, was obtained from the American Type Culture Collection (Manassas, VA, USA). The MG-63.3 cell line , from the parental MG-63 cell line, was obtained from the laboratory of C. Khanna DVM, PhD (National Cancer Institute, Bethesda, MD, USA). The LM7 cell line , from the parental SaOS-2 cell line, was obtained from the laboratory of E. S. Kleinerman MD (MD Anderson Cancer Center, Houston, TX, USA). Cells were maintained after receipt in liquid nitrogen. All experiments were performed on cells passed five to 10 times after resuscitation from liquid nitrogen and negative for Mycoplasma contamination within 6 months. Unless otherwise specified, all cell cultures contained minimum essential media supplemented with 10% fetal bovine serum, nonessential amino acids, sodium pyruvate, and penicillin-streptomycin and were maintained at 37° C in a humidified 5% CO2 environment.
Methotrexate, doxorubicin, and cisplatin were obtained from Tocris Biosciences (Bristol, UK) in solid form. Methotrexate was solubilized in Dulbecco’s phosphate-buffered saline (DPBS) after dropwise addition of sodium hydroxide with frequent warming and vortex; final pH was adjusted to 7.4 with hydrochloric acid as needed . Doxorubicin was solubilized in water . Cisplatin was solubilized in DPBS . All stock solutions were sterile-filtered and stored (methotrexate: 50 mmol/L at -20° C; doxorubicin: 5 mmol/L at -4° C; cisplatin: 5 mmol/L at -20° C). Experimental concentrations were obtained by serial dilution from stock solutions.
Sarcospheres were generated using a modified technique described for epithelial cancers . Each well of a 96-well, round-bottomed plate was coated with 50 μL of 0.5% poly 2-hydroxyethyl methacrylate (poly-HEMA) (Polysciences Inc, Warrington, PA, USA), dissolved in 100% ethanol, and incubated at 37° C for 3 days before use to obtain a nonadherent surface. Cells cultured in monolayer were lifted with Accutase® (EMD Millipore, Billerica, MA, USA) and resuspended in media followed by serial dilution. Wells were plated with the desired number of cells in 100 μL of media supplemented with 2.5% Matrigel® (Corning Inc, Corning, NY, USA). Peripheral wells were filled with 200 μL of DPBS to minimize edge effects—variability attributable to evaporative losses at the periphery of the plate. A minimum of three wells on each plate contained media and Matrigel alone to provide background values. Plates were centrifuged at 1000 g for 10 minutes at 4° C and then incubated at 37° C, 5% CO2. Sarcospheres were matured for 24 hours before chemotherapeutic treatments on Day 0, which were added directly to individual wells in 100 μL of media, and incubated for 48 hours before analysis on Day 2 (Fig. 2A). For all experiments, one plate was analyzed before treatment on Day 0 for comparison. Except in initial experiments characterizing size, where sarcospheres were seeded with a range of cell numbers, sarcospheres were plated at a density that provided a Day 0-diameter of approximately 400 μm 2500 cells (143B), 1000 cells (MG-63.3), and 8000 cells (LM7).
Monolayer cultures for chemotherapy experiments were generated after cells were plated on 96-well, flat-bottomed plates at near confluence in 100 μL of media. Peripheral wells were filled with 200 μL of DPBS to minimize edge effects, and a minimum of three wells contained media alone to provide background values. Plates then were incubated, treated with chemotherapeutics, and analyzed as described for sarcospheres.
Resazurin Reduction Assay
Sarcosphere and monolayer viability was assessed by resazurin reduction, a chemical reaction that occurs only in mitochondria of viable cells. Twenty microliters of alamarBlue® (Invitrogen, Carlsbad, CA, USA) were added to each well containing 200 μL of media. Except in initial experiments characterizing resazurin incubation time, where sarcospheres were incubated and measured at hourly intervals from 0 to 24 hours, incubation at 37° C was continued for 6 hours. Fluorescence readings were obtained at excitation and emission wavelengths of 535 nm and 590 nm, respectively (Tecan Genios Pro, Mannedorf, Switzerland). The average fluorescence of acellular background wells was subtracted from individual control and treatment wells.
Measurement of Sarcosphere Volume
Sarcosphere diameter and volume were assessed by direct imaging. Brightfield images of each sarcosphere were obtained using an inverted microscope with a x10 objective (Leica DMI 6000; Leica Microsystems, Wetzlar, Germany). The scale of images was determined using a calibration slide. Images were analyzed using the open-source software Image J Fiji (National Institutes of Health, Bethesda, MD, USA) and a macro to automate the process as previously described .
Resazurin penetration into sarcospheres was assessed as previously described in epithelial spheroids, where EDTA was required for permeabilization . EDTA was added to culture media on Day 0 and Day 2 to yield a final concentration of 5 mmol/L and incubated for 45 minutes before analysis through the resazurin reduction assay. Fluorescence readings were obtained at regular times with and without EDTA treatment for comparison.
Sarcospheres were observed on Day 2 after treatment by fluorescence microscopy with Ready ProbesTM (Molecular Probes, Eugene, OR, USA). Sixteen microliters of each reagent were added directly to individual wells and incubated for 2 hours before imaging. Live cells were stained with a Hoechst® 33342 (Molecular Probes) and observed with a 4',6-diamidino-2-phenylindole (DAPI) filter (excision/emission: 360 nm/460 nm). Dead cells were stained with propidium iodide (Molecular Probes) and observed with a red fluorescent protein filter (535 nm/617 nm). Apoptotic activity was assessed using CellEvent® Caspase-3/7 Green (Molecular Probes) and observed with a fluorescein isothiocyanate filter (502 nm/530 nm). Images were obtained using standard settings for each marker across treatment conditions with an inverted microscope and a x10 objective (Leica DMI 6000). Images shown are representative of two independent experiments, each with three sarcospheres per group. Positive controls for apoptotic activity were treated with staurosporine (1 μmol/L) in 0.1% dimethyl sulfoxide .
Unless otherwise mentioned, all quantitative data are presented as the mean of a minimum of three independent experiments. Each experiment included a minimum of three sarcospheres per group. Error bars represent SD and data are presented as mean ± SD. Spearman’s r values were calculated to determine the correlation between resazurin reduction and volume with significance set at a probability less than 0.05 using a two-tailed test. Z′ factors, a measure of assay suitability for high-throughput screening used to assess assay signal dynamic range and variability, were produced for a minimum of three independent experiments for each cell line as described . Sarcosphere growth after chemotherapy was calculated as described by Shoemaker . Mean chemotherapy concentration-response curves were fit using a three-parameter logistic model, from which the drug concentrations at which 50% of growth from Days 0 to 2 is inhibited (GIC50), were determined after interpolation and compared using Student’s t test. All curve-fitting, interpolation, and statistics were generated using Prism 7 (GraphPad, La Jolla, CA, USA).
Sarcosphere Size Is Dependent on Cell Line and Number of Cells Seeded
Therapeutically relevant sarcosphere diameters of 300 to 500 μm were achieved on Day 0 for 143B, MG-63.3, and LM7 cell lines at 2500 (diameter: 398 ± 15 μm; 95% CI, 398-434 μm), 1000 (392 ± 11 μm; 95% CI, 364-420 μm), and 8000 (410 ± 10 μm; 95% CI, 384-436 μm) cells seeded, respectively (Fig. 2B). Sarcosphere growth by volume was quantitatively cell line-dependent and greater for small sarcospheres (Fig. 2C). 143B cells showed the greatest change in volume from Days 0 to 2 (610% ± 186% [95% CI, 149%-1073%] to 137% ± 7% [95% CI, 121%-153%] for 500 and 10,000 cells seeded, respectively). In contrast, LM7 sarcosphere volume changed minimally (12% ± 7% [95% CI, -5% to 30%] to -26% ± 6% [95% CI, -41% to -11%] for 1000 and 20,000 cells seeded, respectively).
Accurate Measurement of Sarcosphere Growth
Measurement of resazurin reduction with time showed that 6 hours after addition provided the highest signal-to-noise ratio in the linear range of the assay at Days 0 and 2 for 143B, MG-63.3, and LM7 cell lines (143B [Fig. 3]; MG-63.3 and LM7 not shown). The addition of EDTA did not affect resazurin reduction at times up to 6 hours and was cytotoxic at longer times (143B [Fig. 3]; MG-63.3 and LM7 not shown). Resazurin reduction and volume were related to the number of cells seeded and correlated with each other for 143B (Fig. 4; Spearman’s r: 0.98; p < 0.001), MG-63.3 (0.99; p < 0.001), and LM7 (0.98; p < 0.001).
Sarcosphere Uniformity Is Compatible With Drug Screening
Untreated sarcospheres showed uniform diameter and resazurin reduction on Day 0 across multiple experiments (Fig. 5). Mean Z′ factors on Day 2 for resazurin reduction and volume were greater than 0.5 for all cell lines (Table 1).
Response to MAP Therapy Is Dependent on Cell Line and Culture Model
Concentration-response curves depicting sarcosphere growth by resazurin reduction and volume were generated after MAP therapy (Fig. 6). MG-63.3 and LM7 sarcospheres showed greater than 2000-fold resistance to methotrexate (GIC50 by resazurin reduction: 88 ± 36 μmol/L [95% CI, 24-279 μmol/L] and 174 ± 16 μmol/L [95% CI, 137-221 μmol/L], respectively) compared with the 143B sarcospheres (0.04 ± 0.01 μmol/L; 95% CI, 0.03-0.05 μmol/L; p < 0.001 for MG-63.3 and LM7). LM7 sarcospheres were most sensitive to doxorubicin and cisplatin (0.08 ± 0.04 μmol/L [95% CI, 0.02-0.23 μmol/L] and 0.30 ± 0.11 μmol/L [95% CI, 0.12-0.70 μmol/L], respectively) compared with 143B and MG-63.3 sarcospheres. MG-63.3 sarcospheres were most resistant to doxorubicin and cisplatin (0.61 ± 0.12 μmol/L [95% CI, 0.37-0.98 μmol/L] and 1.7 ± 0.70 μmol/L [95% CI, 0.51-5.0 μmol/L], respectively) compared with 143B and LM7 sarcospheres. Volume and resazurin reduction produced similar results at low drug concentrations (up to 1 μmol/L). However, at high drug concentrations (10 μmol/L), volume as a marker of cell death was limited by the persistence of nonviable cells and did not decrease below the Day 0 threshold.
Fluorescent imaging of live, dead, and apoptotic cell populations showed by qualitative analysis a cell line-dependent response to chemotherapeutic treatment at the GIC50. Staurosporine, a potent inducer of apoptosis, served as a positive control and showed that LM7 and 143B sarcospheres undergo apoptosis effectively. Qualitatively, cisplatin treatment induced the greatest apoptotic and dead cell staining in LM7 sarcospheres (Fig. 7), whereas doxorubicin treatment showed the most apoptotic and dead cell staining in 143B sarcospheres (Fig. 8). Methotrexate induced the least apoptotic and dead cell staining in both cell lines.
MG-63.3 monolayers showed nearly 10,000-fold greater sensitivity to methotrexate (Fig. 9A; GIC50 by resazurin reduction: 0.01 ± 0.01 μmol/L; 95% CI, 0.002-0.02 μmol/L) compared with relatively resistant MG-63.3 sarcospheres (88 ± 36 μmol/L; 95% CI, 24-279 μmol/L; p < 0.001). LM7 monolayers and sarcospheres were highly resistant to methotrexate (Fig. 9B, GIC50 not reached for LM7 monolayers).
Osteosarcoma survivorship has plateaued during the past 30 years, demanding continued pursuit of novel therapeutics to improve patient outcomes . A major barrier to inquiry in this context is the inadequacy of current in vitro models to capture the complexity of the micrometastatic environment in osteosarcoma; and as a result, many promising in vitro drugs fail to show in vivo efficacy . We evaluated sarcospheres to develop a micrometastatic platform to identify novel therapeutics in osteosarcoma. The platform was characterized in multiple cell lines to meet conventional drug screening benchmarks. It then was applied to evaluate conventional MAP therapy, which highlighted heterogeneity across multiple cell lines.
A limitation of this study is that the sarcosphere platform is slightly lower throughput than traditional monolayer screening studies, as described in osteosarcoma and other cancers [30, 33]. However, benefits of this approach are that it more accurately reflects the in vivo microenvironment, compared with monolayer cultures, and that it requires minimal specialized equipment and training [10, 11, 18]. We also did not directly examine the ability of sarcospheres to mimic the behavior of corresponding microtumors in vivo, although it has been shown in multiple nonsarcoma [5, 10, 11, 18, 21] and sarcoma spheroid models [2, 8, 27]. Another limitation is that this platform uses osteosarcoma cell lines without the inclusion of primary-derived cancer or stromal cells, which some advocate more accurately represent the in vivo microenvironment. Although likely true, these additions come at the cost of decreased throughput with an unclear marginal benefit in efficacy for the purpose of drug screening. This platform therefore represents an intermediate between monolayer and in vivo models. An additional limitation is that cell viability was assessed indirectly by resazurin reduction and sarcosphere volume. Direct measurement is possible after sarcosphere disassociation, live cell staining, and counting by flow cytometry, but it is time-intensive and previously shown to be directly related to volume in other cancers .
Sarcosphere diameter and growth were dependent on the number of cells seeded and cell type. This finding supports similar results in epithelial spheroids to suggest that diffusion gradients limit growth as spheroids increase in size . Although this method can produce sarcospheres with diameters of 200 to 800 μm, the therapeutically relevant size for analysis has been shown to be 300 to 500 μm . This ensures that pathophysiologic gradients of oxygen and nutrients are present, along with a core of hypoxic quiescent cells, thought to be responsible for the increased chemo- and radioresistance of spheroids and solid tumors . Growth was highly related to cell line, reflecting known heterogeneity in growth rates and genetics among osteosarcoma cell lines, which parallels the heterogeneity seen in patients with osteosarcoma and may have important implications for response to chemotherapy.
Accurate measurement of sarcosphere viability occurred after resazurin incubation for 6 hours, without EDTA, and was directly related to sarcosphere volume and the number of cells seeded. A previous study  showed that epithelial-based spheroids required EDTA-mediated permeabilization to accurately measure resazurin reduction. However, EDTA did not show an effect in our mesenchymal model, which may be explained by the absence of intercellular epithelial tight junctions. Ivanov et al.  explored the relationship between viable cells in neurospheres, measured by dissociation and cell counts, and neurosphere volume to determine that volume was an excellent predictor for the number of viable cells in healthy spheroids. They recommended that volume be used as a reference method for cytotoxicity assays in spheroid models. Our current study confirmed that volume is directly related to resazurin reduction in mitochondria of healthy sarcospheres. However, after drug treatment, volume measurement is limited by the persistence of nonviable cellular debris. This finding suggests that resazurin reduction provides a more accurate representation of cell viability in the setting of cytotoxic drug therapy.
Sarcosphere uniformity met conventional drug screening benchmarks. The previously described Z′ factor , a statistical measure reflective of the assay signal dynamic range and variability, provides a measure of overall quality for high-throughput screening assays. The Z’ factor threshold for an “excellent” assay was met in the current study for volume and resazurin reduction across all cell lines. These findings assert that this sarcosphere platform is appropriate for drug screening applications, as shown for spheroid-based models in other cancers [9, 10]. Previous osteosarcoma drug screening studies, although limited, have focused on monolayer cultures and in vivo murine xenografts [28, 33]. Rimann et al.  explored 3-D osteosarcoma microtissues produced by a hanging drop method in comparison to the centrifugation-based method presented here. In their study, they showed relative chemoresistance of 3-D tissues compared with monolayer but did not provide detailed growth or uniformity analysis required before drug screening applications.
Finally, response to MAP therapy was cell line and culture model-dependent. MG-63.3 and LM7 sarcospheres showed striking resistance to methotrexate (greater than 2000-fold) compared with 143B sarcospheres. Prior studies have shown the ability of MG-63 and SaOS-2 subclones, parental cell lines for MG-63.3 and LM7, respectively, to upregulate dihydrofolate reductase and explain in part the observed chemoresistance [3, 29]. Another explanation may be that the slower-growing cell lines, MG-63.3 and LM7, are not as sensitive to the antimetabolite methotrexate, which disrupts the folate synthesis pathway, because of relatively low cellular turnover. In contrast, doxorubicin and cisplatin invoke widespread DNA damage regardless of metabolic demand. Interestingly, the methotrexate resistance of MG-63.3 sarcospheres was abolished by growth in the monolayer, suggesting that 3-D culture leads to increased chemoresistance in these cells. Previous studies, as mentioned, have identified chemoresistance in spheroid models, which better correlate with the in vivo response to chemotherapy [10, 11, 18, 27]. It recently was suggested that this phenomenon may be attributable in part to a shift in metabolic phenotype when cells are cultured in 3-D models, which leads to elevated multidrug resistance transport capacity . Overall these findings reflect the known patient-to-patient heterogeneity in osteosarcoma and underscore the importance of evaluating multiple tumor models in osteosarcoma research. They also serve as a proof of concept for drug screening using the sarcosphere platform. Future application of this technique to patient-derived tumors offers an intriguing pathway to identify personalized and effective agents to combat chemoresistance in osteosarcoma.
In this laboratory study, we developed and validated a three-dimensional in vitro drug screening platform for osteosarcoma requiring only standard cell culture equipment. This study also highlights heterogeneity in highly metastatic osteosarcoma cell lines by showing differences in growth rates and susceptibility to MAP therapy. Future efforts will apply this platform to screen existing drug libraries to identify and prioritize promising drugs for in vivo analysis. Expansion of this platform to other osteosarcoma cell lines, particularly those showing drug resistance, and patient biopsy tissue is desirable and would require similar evaluation of sarcosphere growth, seeding density, resazurin response, and sample uniformity to ensure accurate analysis and screening efficacy. The described approach is a promising starting point for drug screening in osteosarcoma because it is tailored to evaluate the progression of micrometastatic disease.
We thank C. Khanna DVM, PhD (National Cancer Institute, Bethesda, MD, USA) and E.S. Kleinerman MD (MD Anderson Cancer Center, Houston, TX, USA) for providing the MG-63.3 and LM7 cell lines, respectively. We also thank E.S. Din BA (Case Western Reserve University, Cleveland, OH, USA) for contributing to preliminary experimental data.
1. Allison DC, Carney SC, Ahlmann ER, Hendifar A, Chawla S, Fedenko A, Angeles C, Menendez LR. A meta-analysis of osteosarcoma outcomes in the modern medical era. Sarcoma. 2012;2012:704872.
2. Arai K, Sakamoto R, Kubota D, Kondo T. Proteomic approach toward molecular backgrounds of drug resistance of osteosarcoma cells in spheroid culture system. Proteomics. 2013;13:2351–2360.
3. Diddens H, Niethammer D, Jackson RC. Patterns of cross-resistance to the antifolate drugs trimetrexate, metoprine, homofolate, and CB3717 in human lymphoma and osteosarcoma cells resistant to methotrexate. Cancer Res. 1983;43:5286–5292.
4. Duan X, Jia SF, Zhou Z, Langley RR, Bolontrade MF, Kleinerman ES. Association of alphaavbeta3 integrin expression with the metastatic potential and migratory and chemotactic ability of human osteosarcoma cells. Clin Exper Metastasis. 2005;21:747–753.
5. Fennema E, Rivron N, Rouwkema J, van Blitterswijk C, de Boer J. Spheroid culture as a tool for creating 3D complex tissues. Trends Biotechnol. 2013;31:108–115.
6. Friedman MA, Carter SK. The therapy of osteogenic sarcoma: current status and thoughts for the future. J Surg Oncol. 1972;4:482–510.
7. Friedrich J, Seidel C, Ebner R, Kunz-Schughart LA. Spheroid-based drug screen: considerations and practical approach. Nature Protoc. 2009;4:309–324.
8. Gaebler M, Silvestri A, Haybaeck J, Reichardt P, Lowery CD, Stancato LF, Zybarth G, Regenbrecht CRA. Three-dimensional patient-derived in vitro sarcoma models: promising tools for improving clinical tumor management. Front Oncol. 2017;7:203.
9. Gupta PB, Onder TT, Jiang G, Tao K, Kuperwasser C, Weinberg RA, Lander ES. Identification of selective inhibitors of cancer stem cells by high-throughput screening. Cell. 2009;138:645–659.
10. Hickman JA, Graeser R, de Hoogt R, Vidic S, Brito C, Gutekunst M, van der Kuip H; IMI PREDECT Consortium. Three-dimensional models of cancer for pharmacology and cancer cell biology: capturing tumor complexity in vitro/ex vivo. Biotechnol J. 2014;9:1115–1128.
11. Hirschhaeuser F, Menne H, Dittfeld C, West J, Mueller-Klieser W, Kunz-Schughart LA. Multicellular tumor spheroids: an underestimated tool is catching up again. J Biotechnol. 2010;148:3–15.
12. Ivanov DP, Parker TL, Walker DA, Alexander C, Ashford MB, Gellert PR, Garnett MC. Multiplexing spheroid volume, resazurin and acid phosphatase viability assays for high-throughput screening of tumour spheroids and stem cell neurospheres. PloS One. 2014;9:e103817.
13. Ivascu A, Kubbies M. Rapid generation of single-tumor spheroids for high-throughput cell function and toxicity analysis. J Biomol Screen. 2006;11:922–932.
14. Jaffe N. Historical perspective on the introduction and use of chemotherapy for the treatment of osteosarcoma. Adv Exp Med Biol. 2014;804:1–30.
15. Jeffree GM, Price CH, Sissons HA. The metastatic patterns of osteosarcoma. Br J Cancer. 1975;32:87–107.
16. Khanna C, Fan TM, Gorlick R, Helman LJ, Kleinerman ES, Adamson PC, Houghton PJ, Tap WD, Welch DR, Steeg PS, Merlino G, Sorensen PH, Meltzer P, Kirsch DG, Janeway KA, Weigel B, Randall L, Withrow SJ, Paoloni M, Kaplan R, Teicher BA, Seibel NL, Smith M, Uren A, Patel SR, Trent J, Savage SA, Mirabello L, Reinke D, Barkaukas DA, Krailo M, Bernstein M. Toward a drug development path that targets metastatic progression in osteosarcoma. Clin Cancer Res. 2014;20:4200–4209.
17. Khanna C, Prehn J, Yeung C, Caylor J, Tsokos M, Helman L. An orthotopic model of murine osteosarcoma with clonally related variants differing in pulmonary metastatic potential. Clin Exp Metastasis. 2000;18:261–271.
18. Kimlin LC, Casagrande G, Virador VM. In vitro three-dimensional (3D) models in cancer research: an update. Mol Carcinog. 2013;52:167–182.
19. Kleinerman ES, ed. Current Advances in Osteosarcoma. Heidelberg, Germany: Springer International Publishing Switzerland; 2014.
20. Koshkin V, Ailles LE, Liu G, Krylov SN. Metabolic suppression of a drug-resistant subpopulation in cancer spheroid cells. J Cell Biochem. 2016;117:59–65.
21. Kunz-Schughart LA, Freyer JP, Hofstaedter F, Ebner R. The use of 3-D cultures for high-throughput screening: the multicellular spheroid model. J Biomol Screen. 2004;9:273–285.
22. Link MP, Goorin AM, Miser AW, Green AA, Pratt CB, Belasco JB, Pritchard J, Malpas JS, Baker AR, Kirkpatrick JA, Ayala AG, Shuster JJ, Abelson HT, Simone JV, Vietti TJ. The effect of adjuvant chemotherapy on relapse-free survival in patients with osteosarcoma of the extremity. N Engl J Med. 1986;314:1600–1606.
23. Luu HH, Kang Q, Park JK, Si W, Luo Q, Jiang W, Yin H, Montag AG, Simon MA, Peabody TD, Haydon RC, Rinker-Schaeffer CW, He TC. An orthotopic model of human osteosarcoma growth and spontaneous pulmonary metastasis. Clin Exp Metastasis. 2005;22:319–329.
27. Rimann M, Laternser S, Gvozdenovic A, Muff R, Fuchs B, Kelm JM, Graf-Hausner U. An in vitro osteosarcoma 3D microtissue model for drug development. J Biotechnol. 2014;189:129–135.
28. Sampson VB, Gorlick R, Kamara D, Anders Kolb E. A review of targeted therapies evaluated by the pediatric preclinical testing program for osteosarcoma. Front Oncol. 2013;3:132.
29. Serra M, Reverter-Branchat G, Maurici D, Benini S, Shen JN, Chano T, Hattinger CM, Manara MC, Pasello M, Scotlandi K, Picci P. Analysis of dihydrofolate reductase and reduced folate carrier gene status in relation to methotrexate resistance in osteosarcoma cells. Ann Oncol. 2004;15:151–160.
30. Shoemaker RH. The NCI60 human tumour cell line anticancer drug screen. Nat Rev Cancer. 2006;6:813–823.
31. Smith MA, Altekruse SF, Adamson PC, Reaman GH, Seibel NL. Declining childhood and adolescent cancer mortality. Cancer. 2014;120:2497–2506.
32. Walzl A, Unger C, Kramer N, Unterleuthner D, Scherzer M, Hengstschlager M, Schwanzer-Pfeiffer D, Dolznig H. The resazurin reduction assay can distinguish cytotoxic from cytostatic compounds in spheroid screening assays. J Biomol Screen. 2014;19:1047–1059.
33. Yu D, Kahen E, Cubitt CL, McGuire J, Kreahling J, Lee J, Altiok S, Lynch CC, Sullivan DM, Reed DR. Identification of synergistic, clinically achievable, combination therapies for osteosarcoma. Sci Rep. 2015;5:16991.
© 2018 Lippincott Williams & Wilkins LWW
34. Zhang JH, Chung TDY, Oldenburg KR. A simple statistical parameter for use in evaluation and validation of high throughput screening assays. J Biomol Screen. 1999;4:67–73.