Challenges in drug efficacy and toxicity screening with 3D bioprinted melanoma models

1 Introduction

Melanoma is the most aggressive and heterogeneous form of skin cancer developing from pigment-producing melanocytes. Melanomagenesis is a complex, multi-step process in which normal melanocytes undergo malignant transformation driven by the accumulation of sequential genetic and epigenetic mutations. This process is initiated predominantly by ultraviolet (UV) radiation-induced deoxyribose nucleic acid (DNA) damage, which, when repaired incorrectly, give rise to oncogenic mutations in key regulatory genes (Raaijmakers et al., 2016; Lotz et al., 2019; Li et al., 2022; Wang et al., 2025). Activating mutations in oncogenes such as B-Raf proto-oncogene (BRAF) and NRAS proto-oncogene (NRAS) lead to constitutive activation of the Mitogen-Activated Protein Kinase (MAPK) signaling pathway which regulates cell proliferation, differentiation, and survival. Concurrently, loss-of-function in genes encoding for the tumor suppressor genes including Cyclin-dependent kinase inhibitor 2A (CDKN2A), phosphatase and tensin1 homolog (PTEN), and tumor protein p53 (TP53) disrupt cell-cycle control, apoptosis, and genomic stability, thereby promoting melanoma progression (Karami Fath et al., 2022; Castellani et al., 2023; Ekinci et al., 2025). Melanoma development is further modulated by epigenetic reprogramming, hormone- and cytokine-mediated signaling within the skin microenvironment, particularly in genetically predisposed individuals (Li et al., 2022; Karami Fath et al., 2022). While early oncogenic events may result in benign melanocytic nevi through oncogene-induced senescence, progression to invasive melanoma requires additional molecular changes and interactions with the surrounding skin cells like keratinocytes and fibroblasts. The components of the tumor microenvironment (TME) such as stromal cells, tumor-infiltrating immune cells, and the tumor-associated blood vessels, collectively play important roles in tumor growth, immune evasion, and metastatic potential (Michielon et al., 2022; Tasdogan et al., 2025).

Surgical resection remains the cornerstone of melanoma treatment for early-stage melanoma. For advanced melanoma, systemic therapies including immune checkpoint inhibitors and targeted therapies have significantly improved patient outcomes (Popovic et al., 2022). The immune checkpoint inhibitors targeting the cytotoxic T-lymphocyte associated antigen 4 (CTLA-4), programmed death 1 (PD-1) receptor and its ligands PD-1 and PD-2 promote anti-tumor immunity. Mechanistically, the anti-CTLA-4 antibody binds to the inhibitory checkpoint CTLA-4 receptor that is located on the surface of regulatory T-cells, which blocks the interaction of T-cells with antigen presenting cells that expresses the CTLA-4 ligand, B7 co-stimulatory molecule. The binding of anti-CTLA-4 antibodies to CTLA-4 promotes the activation of T-cells and the inhibition of the immune checkpoint blockade. Activated T-cells can subsequently elicit an immunological response through cytokine release, clonal expansion and infiltration into the tumor leading to tumor regression (Sullivan et al., 2018; Brunton and Knollmann, 2023). T-cell activation may also be inhibited through an alternative receptor-ligand interaction, in which ligands PD-L1 and PD-L2 bind to the PD-1 receptor, resulting in a co-inhibitory effect. The PD-1 receptor is present on the surface of immune cells, such as T-cells. The anti-PD-L1 antibody inhibits the interaction of PD-L1 or PD-L2 with the PD-1 receptor, thereby enhancing T-cell activation (Carlino et al., 2021; Mahoney et al., 2015; Yi et al., 2021). Targeted therapy selectively inhibits oncogenic signaling pathways that drive melanoma growth and survival. In melanomas harboring BRAF mutations, small-molecule inhibitors targeting mutant BRAF (e.g., vemurafenib, dabrafenib) in combination with downstream MEK inhibitors (e.g., trametinib, cobimetinib) act by suppressing MAPK pathway activation, resulting in reduced proliferation, cell-cycle arrest, and induction of apoptosis in BRAF-mutant melanoma cells. In NRAS-mutant melanoma, where direct inhibition of NRAS remains challenging, downstream targeting of the MAPK pathway using MEK inhibitors provides limited pathway suppression. Additionally, traditional chemotherapeutic agents such as dacarbazine and temozolomide exert anti-tumor effects through DNA alkylation, leading to DNA damage and apoptotic cell death, although their clinical use in melanoma is reserved for palliative care (Brunton and Knollmann, 2023; Duarte de Oliveira et al., 2025). Despite advances in targeted therapies and immunotherapies, melanoma remains a public health concern due to its high metastatic potential, and the development of intrinsic and acquired resistance to both targeted and immune-based therapies (Bhatia et al., 2009; Saginala et al., 2021; Tasdogan et al., 2025; Winder and Virós, 2017).

A major limitation in the development and evaluation of novel melanoma therapies is the continued use of two-dimensional (2D) cell cultures and animal models. While these models have contributed substantially to our understanding of melanoma biology, they fail to reproduce the cellular, biochemical, and architectural complexity of the human TME. As a result, findings obtained from these models show poor clinical translation, leading to inaccurate pharmacokinetic and toxicity profiles in clinical settings (Tiwari et al., 2021; Wu et al., 2023). In vivo tumor models are also commonly used, however, the differences in genetics and species produce data that does not accurately represent human tumorigenesis and progression. Additionally, ethical considerations on animal use, time constraints and the high cost of in vivo studies further reinforce the need for alternative, physiologically relevant in vitro models (Vanderburgh et al., 2016; Löwa et al., 2018).

Three-dimensional (3D) bioprinting technology has emerged as one of the promising strategies to address these challenges. Through pre-designed computer-aided models with layer-by-layer deposition of bioinks with cell-embedded hydrogels, 3D bioprinting enables the fabrication of melanoma constructs that more closely represent skin tissue architecture and better capture clinical heterogeneity of melanoma (Song et al., 2021; Dong et al., 2024). Hydrogels used as bioink matrices provide structural support, enable cell proliferation, and protect cells during the printing process while allowing the incorporation of multiple cell types to better mimic the TME (Tiwari et al., 2021). Importantly, 3D bioprinting aligns with the principles of replacement, reduction, and refinement (3R) in animal research, and offers opportunities for developing personalized melanoma models to support patient-specific therapeutic strategies (Clark, 2018; Han et al., 2020; Neufeld et al., 2022).

Despite these advantages, the transition from 2D cell culture models to complex 3D bioprinted cell constructs exhibits critical gaps in analytical methodologies (Liu and Jin, 2025). Biochemical and molecular techniques commonly used to assess cell viability, proliferation, as well as drug efficacy and toxicity include luminescence, colorimetric and fluorometric assays, microscopy, flow cytometry, quantitative polymerase chain reaction, and sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE). Most of these were previously optimized for cells in 2D monolayers and often yield inconsistent or misleading results in dense, multilayered 3D matrices due to matrix interference, diffusion limitations and uneven cell distribution, resulting in poor data interpretation, reproducibility and limited cross-study comparability (Liu and Jin, 2025). This perspective highlights some of these analytical challenges in the context of drug efficacy and drug-induced apoptosis in 3D bioprinted melanoma models.

2 The need for advanced cell culture models in melanoma

Melanoma accounts for approximately 1% of all skin cancer cases but has a high mortality rate due to its potential for metastasis and because it often displays intrinsic or acquired resistance to conventional chemotherapeutic agents, resulting in poor survival outcomes (Saginala et al., 2021; Winder and Virós, 2017). Improving therapeutic efficacy requires development of disease models that closely mimic the clinical behavior of melanoma, enabling drug screening and robust biomarker discovery.

Advanced cell culture models have transformed in vitro cancer research by producing 3D models that better recapitulate tumor heterogeneity, extracellular matrix composition, and complex cell–cell and cell–matrix interaction. Such models are important for accurate disease representation, developing personalized cancer treatments and understanding mechanisms of drug resistance (Jensen and Teng, 2020). Models are developed to be simplified representations of melanoma biology in patients. To enable a thorough understanding of drug efficacy and toxicity, various models must be used to ensure safe and effective treatments (Rebecca et al., 2020).

It is important that any developed model of melanoma reflects features of the physical, pathological, genomic, and immunological tumor tissue. As our understanding of melanoma biology advances, it becomes increasingly evident that traditional 2D cell culture models cannot replicate the complexity of the melanoma TME, which is needed to understand how melanoma cells behave and respond to drug treatments (Kharouf et al., 2023; Kuras, 2023). Consequently, the predictive power of 2D cell cultures for drug efficacy is limited, often failing to translate into successful clinical outcomes (Jensen and Teng, 2020; Abuwatfa et al., 2024). Advancements in 3D cell culture models present promising opportunities to overcome some of these challenges and to develop new treatment strategies (Patton et al., 2021; Karras and Kunz, 2024; Viegas and Sarmento, 2024).

Currently many technologies are used for creating complex 3D tumor-cell models such as spheroids, organoids (also referred to as tumor organoids or tumoroids in some instances), patient-derived models, reconstructed skin, tumor-on-chip and 3D bioprinted models as shown in Figure 1 top panel (Rebecca et al., 2020; Patton et al., 2021; Michielon et al., 2022; Nomdedeu-Sancho et al., 2023). All the models show great promise and are developed for different purposes, however, some limitations are attached to spherical models, such as their size heterogeneity, and inability to fully capture the cellular and microenvironment heterogeneity found in physiological skin tumor tissues, where spatiotemporal fluctuations in nutrition, cell proliferation, and oxygenation are common. 3D bioprinting of structural tissue such as layered scaffolds, which enables the manufacturing of complex skin tumor constructs comprised of melanoma cells, fibroblasts, keratinocytes, endothelial cells, and immune cells replicating natural tissue, is one biofabrication approach that has emerged as a valuable tool in cancer research (Augustine et al., 2021; Barghian Zarnaghi et al., 2025).

The most used 3D bioprinting technologies include laser-assisted bioprinting, inkjet-based bioprinting, stereolithography-based or light assisted bioprinting, and extrusion-based bioprinting indicated in Figure 1 bottom panel (Dong et al., 2024; Liu et al., 2024; Moon et al., 2024). Each bioprinting technology operates on a different principle and has its own advantages and disadvantages (Datta et al., 2018). Of these techniques, pneumatic extrusion-based bioprinting has become the most widely used approach for 3D cell culture modelling due to its accessibility, cost-effectiveness, and capacity to replicate tissue complexity (Jacob et al., 2022; Shin et al., 2022; Zanderigo et al., 2023). Digital light processing bioprinting has the added benefit of being able to fabricate complex structures with high resolution (Li et al., 2025). In 3D bioprinting, bioink comprised of the matrix with the cells is carefully deposited layer by layer, to form a layered skin-like architecture. The formulation of the appropriate bioink can significantly influence the success of the 3D bioprinting constructs (Geevarghese et al., 2022). A bioink is the combination of natural or synthetic biomaterials, living cells, additional biochemical factors, and, for digital light processing, photoinitiators (Jones, 2006; Chen et al., 2023; Li et al., 2025). Bioinks are designed to mimic the extracellular matrix (ECM), providing a suitable environment for cell growth, differentiation, and cell function in 3D (Raees et al., 2023; Shi et al., 2023). Hydrogels are highly favorable in 3D bioprinting due to their unique physicochemical properties, including high water content, viscosity and stiffness, and biocompatibility, which enable efficient cell encapsulation and support cell viability, growth, and proliferation while mimicking key features of the natural ECM (Chen et al., 2023; Benwood et al., 2021; Karvinen and Kellomäki, 2023).

Although 3D bioprinting has considerable potential for melanoma research, construct stability, bioink compatibility with cells, and reproducible fabrication of constructs across different 3D bioprinting technologies pose challenges, illustrated in Figure 1 bottom panel. Construct stability and printability are still strongly dependent on bioink formulation and the crosslinking method used. Melanoma-based studies indicate that printable hydrogels such as alginate/hyaluronic acid/gelatin and pectin–nanocellulose composites can improve rheology, printability, shape precision, and short-term A375 viability; however, these outcomes are formulation and bioprinter-specific, since variations in printing and crosslinking conditions may affect melanoma-cell viability and proliferation (Schmid et al., 2022; Schipka et al., 2024; Teixeira et al., 2024; de Villiers et al., 2024). Bioink compatibility with melanoma cells is not guaranteed, because both bioink composition and printing conditions can alter proliferation, spheroid morphology, migration, and cell–matrix interaction even when viability is maintained (Schipka et al., 2024). Additionally, the same bioink formulation may need re-optimization when applied across different printers, extrusion parameters, or crosslinking methods. Perhaps future studies may not only focus on bioink formulations, but also on validation of the same formulation in various bioprinters for reproducibility (Schmid et al., 2022; Germain et al., 2022; Schipka et al., 2024; Teixeira et al., 2024; de Villiers et al., 2024).

3 Summary of 3D melanoma models to date

Multiple advanced 3D melanoma models have been developed to improve the physiological relevance of preclinical cancer research, particularly within the fields of diagnostic oncology and pharmaceutical sciences. These include tumor spheroids, patient-derived organoids, tumor-on-chip models, and 3D bioprinted constructs, each offering distinct advantages for modelling melanoma biology, drug response, and mechanisms of resistance (Rebecca et al., 2020; Jensen and Teng, 2020; Patton et al., 2021; Nomdedeu-Sancho et al., 2023). For example, Müller and co-workers established an organotypic melanoma spheroid skin model using primary keratinocytes and fibroblasts isolated from juvenile foreskin, combined with 451-LU melanoma spheroids. The spheroids showed epidermal differentiation, basal lamina formation, and necrotic cores, thereby mimicking human melanoma characteristics (Müller and Kulms, 2018). Karimi et al. characterized WM266-4 and WM115 melanoma spheroids using hypoxia distribution, glucose uptake, proliferation, and positron annihilation lifetime spectroscopy, evaluating positronium lifetime as a potential biophysical biomarker of tumor aggressiveness (Karimi et al., 2023).

Spheroids of melanoma with mutations in the NRAS gene and multicomponent melanoma spheroids mimicking skin/dermal, lung, and liver TMEs were applied in high-throughput drug screening. The study identified daunorubicin hydrochloride and pyrvinium pamoate as potent inhibitors of NRAS-mutant melanoma spheroids (Angeli et al., 2025).

Patient-derived organoids from melanoma brain metastases, cutaneous melanoma, and mucosal melanoma have been developed to evaluate their response to targeted therapy and immunotherapy (Ou et al., 2023; Sun et al., 2023; Abedellatif et al., 2024). These organoids preserved key histological features and mutational profiles of BRAF- and NRAS-mutated tumors and responded to agents such as dabrafenib combined with trametinib (Abedellatif et al., 2024) and temozolomide (Ou et al., 2023). This provided clinically relevant models for targeted therapy and immunotherapy-response studies.

Microfluidic systems have enabled precise 2D/3D tissue construction for melanoma modelling and personalized drug testing (Jeffries et al., 2020). Pandya et al. developed a 3D melanoma-on-a-chip model that distinguished carboplatin-responsive from carboplatin-resistant tumors, supporting personalized therapy strategies (Pandya et al., 2017). Herreros and colleagues produced a melanoma-on-chip model integrating gemcitabine delivery, demonstrating the feasibility of chip-based local drug administration (Herreros et al., 2024).

3D bioprinted models have also been used to evaluate the effects of different bioinks on melanoma cell morphology and behavior, providing guidance for the selection of biomaterial/hydrogels in melanoma modelling (Schmidt et al., 2019). Other studies have shown the application of 3D bioprinting to produce constructs of advanced melanoma models using explants from patients with late-stage melanoma (Jeong et al., 2021). In addition, bioprinted constructs have been used to recreate vascular–lymphatic interactions in heterogeneous melanoma to study metastasis and tumor immunology (Passarelli et al., 2017; Cho et al., 2022). More recently, 3D bioprinted A375 melanoma constructs have been used for drug screening and were validated for apoptotic cell death analysis using flow cytometry, showing compatibility with quantitative, mechanistic assays commonly used in preclinical research (De Villiers et al., 2024; 2025). Collectively, these studies show the evolution of 3D bioprinted melanoma models towards physiologically relevant models, highlighting their growing potential for translational medicine, mechanistic studies, and personalized therapeutic evaluation. Table 1 summarizes the key characteristics and applications of these 3D melanoma models.

TABLE 1. Overview of selected 3D melanoma models from spheroids and organoids to microfluidic and 3D bioprinted systems.

4 Analysis of anti-cancer drug efficacy and toxicity in melanoma

Assessing drug efficacy and mechanisms of cell death in melanoma depends on analytical methods that quantify cell viability, proliferation, and the molecular pathways targeted by anti-cancer agents. In cancer treatment, regulated cell death, particularly apoptosis, is generally preferred because it limits inflammatory responses that may promote tumor progression and damage surrounding tissues (Kroemer et al., 2009; Galluzzi et al., 2012; 2018; Ye et al., 2024). Apoptosis is activated by two pathways, the intrinsic and extrinsic pathways, which converge at effector caspase-3 and ultimately result in DNA and protein breakdown. The intrinsic pathway, also known as the mitochondrial pathway, is controlled by pro-apoptotic and anti-apoptotic molecules in the Bcl-2 family, which co-ordinate to increase mitochondrial outer membrane permeability and subsequent release of apoptogenic factors like cytochrome c, second mitochondria-derived activator of caspases (smac/DIABLO), high temperature requirement A serine protease 2 (Omi/HtrA2), apoptosis inducing factor, and endonuclease G. Cytochrome c activates caspase-9, while apoptosis inducing factor (AIF) contributes to caspase-independent cell death. Smac/DIABLO and Omi/HtrA2 promotes caspase activity by blocking inhibitor of apoptosis proteins (Galluzzi et al., 2018; Mustafa et al., 2024). On the other hand, the extrinsic pathway is death receptor mediated by Fas ligand and tumor necrosis factor-α (TNF-α) binding to their respective receptors, Fas, and TNF. The binding of these receptors activates an intracellular death domain that activates proteins such as the TNF receptor-associated death domain (TRADD), the Fas-associated death domain (FADD), and the active caspase-8, which subsequently activates effector caspases-3/7 (Tait and Green, 2010; Broussard et al., 2018; Galluzzi et al., 2018; Green, 2022; Wróblewska et al., 2023; Mustafa et al., 2024).

Evasion of apoptosis is a hallmark of cancer, including melanoma (Dai et al., 2016). Melanoma cells bypass apoptosis through dysregulation of Bcl-2 family proteins, overexpression of apoptosis proteins (IAPs) inhibitors, and rewiring of pro-survival signaling pathways like heat shock proteins 40, 70, and 90 (Hartman, 2020; Cyran and Zhitkovich, 2022). Accurate analysis of apoptosis is therefore essential, as the success of many anti-cancer drugs in melanoma is measured by their ability to restore or enhance apoptotic signaling and support the development of therapeutic strategies that overcome tumor growth, invasion, and therapy resistance. For precise and comparable results in drug efficacy and toxicity testing, it is often necessary to have similar numbers of cells or similar-sized spheroids, or similar-sized 3D bioprinted constructs (Couto et al., 2019).

In melanoma, growing evidence shows that apoptosis plays a complex role in tumor biology and the TME, influencing both cancer initiation and disease progression (Ye et al., 2024). Importantly, apoptotic response in melanoma is highly context-dependent and regulated. Apoptotic cells contribute to adaptive immune responses by serving as a source of tumor antigens. Following apoptosis, dendritic cells in different layers of the skin rapidly engulf dying cells and transport antigens to lymphoid tissues, where they activate T lymphocytes and B lymphocytes (T and B cells), promoting B-cell clonal expansion and antibody production (Albert, 2004; Ye et al., 2024). Concurrently, apoptotic cells can also promote tumor progression by influencing macrophage polarization towards activated (M2) macrophage-like phenotype by suppressing activated macrophage (M1)-mediated anti-tumor immune activity (Morana et al., 2022).

Together, these processes highlight the differential and context-dependent role of apoptosis in shaping the tumor microenvironment of complex cancers like melanoma. Apoptotic signaling in melanoma is strongly regulated by microenvironmental factors, including cell–cell interactions, cell–ECM adhesion, biomechanical cues, and gradients of oxygen, nutrients, and therapeutic agents progression (Ye et al., 2024). These factors modulate cell survival signaling, mitochondrial outer membrane permeabilization, and drug penetration, yet are poorly represented in 2D cultures or most spheroid models (Popovic et al., 2022). 3D bioprinted melanoma models may overcome these limitations by enabling reproducible organization of tumor and stromal cells within ECM-mimetic hydrogels, accurate control of construct architecture, and incorporation of microenvironmental complexity (Kharouf et al., 2023; Kuras, 2023). Although 3D bioprinted melanoma models offer a highly promising and physiologically relevant platform for investigating drug-induced apoptotic signaling, their implementation introduces mechanical and material-related stresses that affect cell viability and tissue organization, while variability in construct geometry, cellular distribution, and bioink properties complicate analytical methodologies. Together, these challenges limit reproducibility, assay sensitivity, and comparability across studies. Additionally, the effect of different bioprinting technologies on apoptotic signaling in melanoma is underexplored. Current melanoma-specific evidence is mainly in extrusion-based bioprinting, especially pneumatic extrusion systems. de Villiers et al. developed a BIOX pneumatic extrusion-based A375 melanoma model and used Annexin V/propidium iodide flow cytometry to quantify drug-induced apoptosis, showing that 3D bioprinted cultures had different apoptotic responses and greater drug resistance than 2D cultures (de Villiers et al., 2024; 2025). By contrast, other melanoma models have focused more on microenvironmental complexity, vascular interaction, and therapy response than on direct measurement of apoptosis in melanoma. Cho et al. generated heterogeneous melanoma spheroids with blood and lymphatic vessels, but the study emphasised stromal protection, endothelial dysfunction, invasion, and drug response rather than direct comparison of apoptosis biomarkers (Cho et al., 2022). Taken together, these studies suggest that present knowledge about apoptosis in bioprinted melanoma is limited to nozzle-based extrusion bioprinting, while direct comparison with other bioprinting technologies such as inkjet, stereolithography, or laser-assisted bioprinting has been poorly explored. These methods exposed the cells to various mechanical stresses like photochemical and formation of reactive oxygen species during the printing phase, apoptotic response may differ based on the method used, but this still needs to be evaluated in melanoma models (Liu X. et al., 2026).

5 Challenges in the analysis of 3D bioprinted constructs

Conventional assays were originally optimized for 2D monolayers and present significant challenges when applied to thick 3D bioprinted constructs. Bioprinted constructs often exceed 500 µm in thickness, and diffusion limitations restrict the penetration of reagents to deep layers of the construct. Incomplete reagent penetration can result in an underestimation of viable cells in the inner regions of the constructs, leading to incorrect drug potency prediction and misinterpretation of pharmacodynamic behavior in 3D models. Variability in construct sizes, shapes, and cell numbers further complicates analysis as comparisons may become unreliable and skew potency estimation (Riss and Trask, 2021). In addition, the bioprinting process itself influences cell viability. Pneumatic extrusion bioprinting damages cells through mechanical forces, whereas digital light processing induces damage by light radiative stress and chemically via the photoinitiator (Li et al., 2025). It is therefore imperative to include appropriate controls and validation processes in drug efficacy and toxicity studies using 3D bioprinted constructs.

Generally, fluorescence, luminescence or colorimetric techniques, such as MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide), MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium), CellTiter-Glo®, and resazurin-based assays are favored for ongoing monitoring of cell viability, proliferation, and metabolic activity without damaging the structural integrity of the 3D bioprinted models. These methods use measurements such as absorbance or fluorescence directly or indirectly in cell culture media to analyze cells based on specific indicators to provide insights into a wide range of cell functions and behaviors (Mosmann, 1983; Berridge et al., 2005; László, 2011; Gong et al., 2020; Dominijanni et al., 2021; Engel et al., 2022; Malakpour-Permlid and Oredsson, 2021; Ryoo et al., 2024). When using colorimetric assays for 3D bioprinted cell culture models, it is important to consider limitations such as the diffusion and bioactivity of compounds within the 3D bioprinted biomaterial matrix. This can cause the dilution of dyes or drugs and affect the accuracy of cell viability and behavior evaluations (Liu X. et al., 2026). Dominijanni et al. (2021) found that, based on different hydrogel and 3D construct properties, enzymatic dissociation of the hydrogel may be necessary for more accurate results. Non-destructive colorimetric methods tend to only provide an average response for a cell population, which could result in the loss of crucial biological information. Therefore, 3D bioprinted constructs are often cut into small sections, slices, or pieces, or dissociated into a single cell suspension before using traditional assays (Avnet et al., 2024; Mazzoldi et al., 2024; Chastagnier et al., 2025; Fang et al., 2025).

In contrast, destructive methods such as flow cytometry require the dissolution of the 3D bioprinted constructs, enabling the formation of a single-cell suspension to release encapsulated cells for analysis (de Villiers and du Plessis, 2023; Gretzinger et al., 2018). Flow cytometry has been widely used in biological analyses due to its ability to rapidly analyze quantitative and multi-parametric data on individual cells within a heterogeneous population, (de Villiers and du Plessis, 2023; Liu X. et al., 2026). Additionally, this technique can be highly effective due to its ability to manage large sample volumes and detect subtle variations in cell responses to drug treatments. This advantage is crucial to understand the mechanisms underlying drug efficacy and resistance, ultimately improving the identification of more effective personalized therapeutic strategies (Hernández et al., 2017). Although flow cytometry is a destructive analysis technique, its compatibility with high-throughput screening and its ability to measure diverse cell properties make it a promising tool for evaluating 3D bioprinted cell cultures for drug screening applications (de Villiers and du Plessis, 2023; Gretzinger et al., 2018).

Microscopy is another complementary method for analysis of 3D bioprinted melanoma constructs because it enables cell visualization within the constructs and cell response to treatment through monitoring of changes in morphology. Reviews of melanoma bioprinted constructs show that skin-like multicellular 3D models capture tumor architecture better than 2D cultures and are therefore more useful for drug screening (Boutin et al., 2018; Schmidt et al., 2019; Browning et al., 2020; Jeong et al., 2021; Cho et al., 2022; Caramenti et al., 2024; Schmid et al., 2024; Vázquez-aristizabal et al., 2024). Brightfield/phase-contrast microscopy is useful for observing/monitoring cell proliferation, health, gross morphology, and assessing contamination. Fluorescence microscopy allows specific cellular targets to be visualized within the cells using dyes, fluorescent proteins, or antibody labels, which makes it valuable for viability staining and pathway-specific assays (e.g., apoptosis markers) (Thorn, 2016; Atale et al., 2014; Klontzas and Protonotarios, 2021). Confocal microscopy is a form of fluorescence microscopy that removes out-of-focus signal, producing thin optical “slices” called Z-stacks that can be reconstructed into a 3D image. This is especially useful for assessing drug penetration effects at different depths within the construct (Elliott, 2021).

A key limitation of fluorescence/confocal imaging in intact 3D constructs is poor imaging depth, mainly due to light scattering in hydrogels. Hydrogels are heterogeneous, composed of polymers embedded with the cells/ECM. As such, refractive index differences often scatter light and reduce clarity in deeper sections of the construct (Richardson and Lichtman, 2015; Boutin et al., 2018; Richardson et al., 2022; Schmid et al., 2024). Practically, this is often addressed by physical sectioning, improved mounting strategies (holders/adhesives), and clearing (glycerol-based solutions), or refractive-index matching to improve transparency (Teodori et al., 2017; Boutin et al., 2018; Klontzas and Protonotarios, 2021; Spiller and Campos, 2025). Similarly, immunolabelling can be unreliable in thick matrices because antibodies and dyes may not diffuse well, which can produce weak or false-negative staining in deeper layers. This has driven approaches such as improving construct permeability and using refractive-index-matched hydrogels that support antibody movement (Richardson et al., 2022; Ni et al., 2017).

Histology is another gold standard method for validation of skin tissue model structural characteristics and melanoma diagnostic markers in thick constructs (Schmid et al., 2024). Sectioning (commonly 3–5 µm) allows high-quality imaging of internal regions that are difficult to visualize in whole-mount samples. Histological stains such as haematoxylin and eosin, as well as immunostaining (immunohistochemistry/immunofluorescence) can be used to evaluate visualization and localization of diagnostic or apoptotic markers. Hematoxylin is a basic dye that stains proteins a blue color, while eosin stains proteins a pink color. These two stains are commonly used together to define intracellular organelles and proteins (Tamizhazhagan and Pugazhendy, 2017; Latonen et al., 2024; Xu et al., 2025; Magaki et al., 2019). Additionally, histology techniques are useful because they reduce dependence on lysate-based assays, which can be technically challenging in hydrogels. For example, Western blotting and RT-qPCR (reverse transcription quantitative polymerase chain reaction) require complete lysis of all cells to obtain proteins and DNA but may be affected by inefficient lysis and protein extraction from dense or highly crosslinked matrices. This can bias measured protein levels or cleavage products (Da Silva et al., 2024). Standardizing extraction conditions and lysis buffers is therefore important when lysate-based assays are used to complement microscopy.

In addition to assay-specific limitations, the physical handling and imaging of thick 3D bioprinted constructs present significant practical challenges. Unlike 2D models, 3D bioprinted constructs cannot be readily mounted on standard glass coverslips without deformation or disruption of their normal architecture, particularly when composed of soft hydrogels (Klontzas and Protonotarios, 2021). Their thickness and mechanical fragility complicate sample orientation for inverted microscopy, often requiring constructs to be flipped or repositioned to visualize cells located deeper within the matrix, which may introduce artefacts or alter matrix organization (Teodori et al., 2017; Xu et al., 2022). To reduce this, constructs may be cultured in an “upside-down” insert configuration that allows transfer to a glass-bottom dish for live imaging without flipping of the constructs (Zaderer et al., 2019). Furthermore, sophisticated data collection and interpretations are needed due to the natural heterogeneity of cells in 3D bioprinted cell culture models. For instance, variability in spheroid size, cell distribution, and diverse cell populations within the same 3D bioprinted construct can lead to inconsistencies in assay results, particularly when high-throughput screening methods are employed.

This variability is less pronounced in 2D cell cultures, where cells are typically more uniform and easier to analyze. However, in 3D cell cultures, the complexity of the environment and the interactions between cells and the surrounding biomaterial matrix can introduce some variability (Belfiore et al., 2021; Joshi and Lee, 2015). As analysis apparatus, techniques and methods improve, especially within the context of high-throughput drug screening, so does the number of parameters increase that can be analyzed simultaneously. This necessitates more sophisticated data storage and automated analysis. Miniaturization is also at the forefront of technological advancements, with the need to have immediate detection. Ever expanding into all fields and equally important is artificial intelligence and machine learning to improve manufacturing and high-throughput drug screening. However, all these advancements are in some way limited by the lack of development of regulatory and global standards (Liu X et al., 2026; Liu S et al., 2026). Collectively, these challenges highlight that current methodologies cannot readily be adopted from 2D culture to 3D bioprinted models without re-optimization, standardization, and validation.

6 Future perspectives and conclusion

3D bioprinted melanoma models represent physiologically relevant in vitro platforms that can better mimic key features of the in vivo melanoma TME, including architectural complexity, cellular heterogeneity, and clinically relevant drug responses. However, the analytical assays used to evaluate drug efficacy and toxicity require optimization and reagent reformulation for 3D bioprinted constructs to allow for penetration and suitability for hydrogels. This will be essential for improving reproducibility and comparability. It will be important for drug efficacy assays to standardize methods to allow for improved drug development. The development of bioprinted models and tailored analytical methods will encourage the application of 3D melanoma models in drug discovery and toxicity testing, positively impacting development in these fields.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Souren Paul, TERI School of Advanced Studies (TERI SAS), India

Reviewed by: Betül Çelebi Saltik, Hacettepe University, Türkiye

Wan’an Xiao, Shengjing Hospital of China Medical University, China