Osteosarcoma Spheroid Models Provide Key Tool for Preclinical Drug Development

With survival rates in osteosarcoma stagnating over the last few decades with long-standing standard chemotherapies, osteosarcoma spheroid models may serve as important tool for obtaining a deeper understanding of the tumor microenvironment in preclinical studies, as well as a platform for discovering novel therapeutic targets and identifying more effective therapies for osteosarcoma, according to R. Lor Randall, MD, FACS.

Randall and colleagues at the University of California (UC) Davis, developed a novel spheroid model of osteosarcoma to better assess the combined effects of oxygen deficiency, represented by spheroid diameter and extracellular matrix (ECM) deposition, both of which are known to reduce the efficacy of various cell-based therapies. The in vitro study utilized 2 murine osteosarcoma cell lines with differing metastatic potentials to create spheroids of varying sizes. Cell responses were analyzed under standard (21% oxygen) and physiological (5% oxygen) oxygen conditions, and chemotherapeutic responses to doxorubicin treatment were also evaluated.

The study demonstrated that ECM production and oxygen tension significantly influenced spheroid size by affecting cell organization, which itself was dependent on the distribution of nutrients and oxygen. Notably, highly metastatic osteosarcoma exhibited greater susceptibility to chemotherapies vs less metastatic disease when matrix production increased.

“Since we've been able to show that there is variable response between the different types of cell lines and efficacy that recapitulates some of the clinical situations, [we’ve shown that] this would be a good model to introduce other agents that are in the pipeline for clinical delivery,” said Randall, the David Linn Endowed Chair for Orthopedic Surgery, chair of the Department of Orthopedic Surgery, and a professor at UC Davis Comprehensive Cancer Center in Sacramento.

In an interview with OncLive®, Randall detailed the construction and subsequent use of osteosarcoma spheroid models, highlighted their advantages and complementary role alongside engineered bone marrow models for preclinical drug development, and emphasized the importance of collaborating with engineering teams at academic centers to develop innovative tumor models.

OncLive: What advantages do engineered bone marrow models offer for preclinical research in sarcoma, and how might they provide insights into drug response?

Randall: Our engineered bone marrow model is in pursuit of a more ideal model for preclinical studies for patients with sarcomas. We all realize sarcomas are relatively rare, and preclinical work is critical because we have a limited ability to conduct clinical trials. It's that much more important in the sarcoma and rare cancer world. We have talked about these engineered bone marrow models in the past. These models are nice because, unlike the two-dimensional petri dish types of models, this mimics the bone marrow and the nascent environment for sarcomagenesis.

One of the advantages is that you can manipulate the microenvironment with oxygen tension, macrophage polarization, ECM induction, [and] a variety of other factors, in a three- dimensional environment that mimics the bone marrow, where we know these osteosarcomas arise. Compare that with some of the preclinical models and mice models that have come about. The problem is that it's harder to control some of the very precise variables that may lead to the formation of sarcoma and then tease out the mechanism. Because we have a strong engineering background here at UC Davis Health, I've been collaborating with J. Kent Leach, PhD, [a professor of Orthopaedic Surgery and Biomedical Engineering, Lawrence Ellison Endowed Professor of Musculoskeletal Research, and Vice Chair of Research at UC Davis] on that engineered bone marrow.

How do three-dimensional osteosarcoma spheroid models complement more complex engineered bone marrow models in the development of new therapies?

More recently, [we’ve been] looking at a simpler model so that we can perform even more high throughput [research] using spheroid [models of] osteosarcoma based upon a variety of different cell lines. In a recently published paper, we looked at a highly metastatic cell line and a less aggressive cell line, and we were able to induce them into these spheroids, which, just like the name implies, are spheres of sarcoma cells [into which] we can introduce an agent to evaluate at therapeutic response.

The beauty of it is that, because it's three-dimensional and there are some mechanical properties to it as well, we believe it is a better model than the two-dimensional monolayer work that [has been] done. It's not as elegant as the engineered bone marrow [model], but we want to reserve the engineered bone marrow [model] for those agents that we think have high likelihood for success because it's so much more complex. The goal of the spheroids is to potentially introduce agents that we're considering for clinical development, into the spheroids. Then, if we [observe] an effect, we can take [this agent] to the engineered bone marrow [model] and then take it into a clinical trial. That's the big picture.

We created these 2 different types of osteosarcoma cell line spheroids, and we've been able to manipulate the oxygen tension within those spheroids to physiologic oxygen tension, which means [taking it from standard atmospheric oxygen tension of] 21% or higher [to] the relatively hypoxic oxygen tension in the bone marrow, which is approximately 5%.Then we've been able to manipulate the ECM in these spheroids using L-ascorbic acid 2-phosphate, which stimulates ECM to form. In osteosarcoma, we know there is a large ECM. We can create a sphere of high or low metastatic potential in osteosarcoma and then introduce agents.

What did these osteosarcoma spheroid models reveal about the influence of extracellular matrix and oxygen tension on the efficacy of doxorubicin, according to results from your preclinical study?

In this case, we introduced doxorubicin, and we found that the doxorubicin was more likely to kill the highly metastatic lesions, which was a bit counterintuitive. We also realized that that [result] was dependent upon the ECM being present in those samples. When we took out the ECM, doxorubicin was less effective. When these osteosarcoma cells metastasize to the lung, they tend to have less ECM that may be [part of the explanation for] why they're resistant to doxorubicin.

What are some of the limitations and benefits of osteosarcoma spheroid models?

It's very important [to note that this model] is still in vitro. It, like the nude mouse models, does not have a competent immune system. There are other variables at play, so take it for what it's worth. The idea here is that it's relatively cheap to produce and it can give us a signal of efficacy for a drug that we then put into a more sophisticated [model], whether it be an engineered bone marrow or murine model.

The other nice thing about it is we can control the variables. We can control oxygen tension and the degree of ECM [production]. We could introduce macrophages in the immune system to try to polarize them. Then [we could] inject the drug into those different spheroids and look for where we kill off the tumor cells most effectively. Is it with high oxygen tension or low oxygen tension? Is it with the macrophages being polarized in a certain way? Or is it a function of the ECM? We can manipulate singular variables, which we can't do in a mouse model.

What are the next steps for leveraging osteosarcoma spheroid models to evaluate novel agents and address challenges in preclinical drug development for rare tumor types and beyond?

This is a model [for use in further] investigations, and the fidelity seems pretty good. For this study we just evaluated doxorubicin, which is standard therapy for osteosarcoma. It was a proof of principle paper. Since we've been able to show that there is variable response between the different types of cell lines and efficacy that recapitulates some of the clinical situations, [we’ve shown that] this would be a good model to introduce other agents that are in the pipeline for clinical delivery. We would propose evaluating exciting new agents and putting them into these spheroids and looking at responses manipulating the oxygen tension and the ECM. When we find something that affects the proliferation of these cells in a negative way, which is what we want, we would then put them into the bone marrow for more elaborate testing.

I believe [this model has utility outside of rare tumor types] and I'm sure that there are individuals exploring it. The intrigue for us is, again, because there's so few patients with this [disease] relatively speaking. Relative to the other tumor types, however, we just don't have the ability to perform some clinical trials, so we really need these preclinical models.

How can oncologists leverage partnerships with engineering teams to develop innovative models such as this one across tumor types?

What I’ve found most intriguing in my time here at UC Davis is the strength of the engineering group. As oncologists, we don't tend to rub elbows with engineers. We tend to rub elbows with molecular biologists, epidemiologists, and everything in between, in terms of cell biology and science, as opposed to the applied sciences of engineering. I've gained a mass amount of appreciation for the applied sciences, and an engineer is trained in a slightly different way than a pure scientist. Our oncology community, if they have access to an engineering school in their academic medical center, might look around at who's in their biomedical engineering group. [They could] potentially partner up to build models or execute models are already established [in some of these tumor types] and answer some profound questions in terms of therapy.

Reference

Sagheb IS, Coonan TP, Randall RL, et al. Extracellular matrix production and oxygen diffusion regulate chemotherapeutic response in osteosarcoma spheroids. Cancer Med. 2024;13(18):e70239. doi:10.1002/cam4.70239