Over the past 100 years, medicine has taken a huge step forward, which has led to a meaningful increase in the average human life expectancy. However, the cumulative risk of developing cancer or dying from cancer also increased with age (Oeppen and Vaupel, 2002; Sung et al., 2020). This makes the search for safer and more effective cancer therapies with fewer side effects one of the important challenges in biomedicine (Pucci et al., 2019). New perspectives for cancer therapies were opened by the development of bioceramics, suitable to be used as carriers for drug delivery (Sethu et al., 2017; Rödel et al., 2018; Vallet-Regí, 2019; Khalifehzadeh and Arami, 2020). The incorporation of traditional drugs within drug delivery systems aids in enhancing their pharmacokinetics and biodistribution.

The most popular bioceramic materials are based on calcium phosphates (Bal et al., 2020). They demonstrate good biocompatibility, osteoconductivity, and osteoinductivity. Currently, it is possible to obtain bioceramic matrices with a high degree of porosity, which allows achieving a high drug load capability and effective kinetic release. However, porosity significantly decreases the mechanical properties of implants (Indra et al., 2021). Calcium phosphate-based bioceramics initially have low mechanical strength and toughness, so porous calcium phosphate-based drug carriers are constrained to fill only small bone defects.

Recently, diopside (CaMgSi2O6) has been proposed as a new promising drug carrier for osteosarcoma treatment (Suleman et al., 2021) and potentially can be a candidate for glioblastoma treatment in the form of cranial implant-drug depot (Sheleg et al., 2002). The current standard of care for osteosarcoma and glioblastoma is surgery, followed by adjuvant chemotherapy with cytostatic agents. Systemic administration of chemotherapeutic agents has some serious adverse effects, including cardiotoxicity, hematotoxicity, and tissue necrosis. Local use of bone regeneration-promoting scaffolds loaded with cytostatic drugs to eliminate tumor cells remaining after resection will improve treatment efficacy and reduce the risk of local recurrence.

Unlike calcium phosphate-based bioceramic materials, diopside has a high mechanical strength which can lead to an increase in the mechanical properties of implantable materials (Ramezani et al., 2017). Additionally, it demonstrates good bioactivity, degradability and controllable drug release (Saravanan and Sasikumar, 2012; Pang et al., 2022). Therefore, studies aimed at the use of diopside in biomedicine, in particular, in anticancer therapy, are of high relevance.

An important step in the development of in vitro assays for assessing the effect of released drugs on cells is the choice of an adequate cell model. It should be simple, inexpensive, and well reproducible, but, at the same time, it should closely mimic in vivo conditions. 2D cell culture fails this task because of many limitations, such as the lack of most cell-cell and cell-extracellular matrix interactions; changes in cell morphology and polarity; and unlimited access to oxygen and nutrients from the culture medium (Fang and Eglen, 2017; Kapałczyńska et al., 2018). In this regard, 3D cell models have become increasingly popular because they better mimic in vivo environments and responses to drug treatment (Bissell, 2017; Langhans, 2018; Badr-Eldin et al., 2022). The most convenient 3D models that meet all of the above criteria are spheroids (Zanoni et al., 2019; Han et al., 2021). They are spherical, dense cell aggregates that can be reproducibly obtained by different methods (Białkowska et al., 2020; Koudan et al., 2020). The possibility of forming size-controlled spheroids of uniform size, cellular composition, and degree of maturity allows their application to assess the effect of various factors on cells in a physiological three-dimensional environment (Sapogova et al., 2020; Borodina et al., 2021).

In the present work, we have proposed an in vitro approach for the evaluation of the effect of doxorubicin (DOX) released from diopside particles on MG-63 cells and primary human fibroblasts (HF) in 3D cell culture. MG-63 spheroids modeled tumor tissue while HF spheroids modeled healthy tissue. We demonstrated the effective adsorption of doxorubicin by diopside particles, its gradual release and impact on spheroids in a cell-dependent manner. The activity of released doxorubicin on MG-63 spheroids was significantly higher than on HF spheroids. The proposed in vitro approach could be used to model the treatment of tissue and bone defects, infections, and anticancer therapy using local drug delivery systems.