Miniaturized pre-clinical cancer models as research and diagnostic tools

Abstract

Cancer is one of the most common causes of death worldwide. Consequently, important resources are directed towards bettering treatments and outcomes. Cancer is difficult to treat due to its heterogeneity, plasticity and frequent drug resistance. New treatment strategies should strive for personalized approaches. These should target neoplastic and/or activated microenvironmental heterogeneity and plasticity without triggering resistance and spare host cells. In this review, the putative use of increasingly physiologically relevant microfabricated cell-culturing systems intended for drug development is discussed. There are two main reasons for the use of miniaturized systems. First, scaling down model size allows for high control of microenvironmental cues enabling more predictive outcomes. Second, miniaturization reduces reagent consumption, thus facilitating combinatorial approaches with little effort and enables the application of scarce materials, such as patient-derived samples. This review aims to give an overview of the state-of-the-art of such systems while predicting their application in cancer drug development.

Keywords: Body-on-a-chip; Cancer models; Cell adhesion-mediated drug resistance; Cell culturing devices; Combinatorial screening platforms; Extracellular matrix; Microfabricated model systems; Microfluidics; Pre-clinical drug assessment; Tumor microenvironment.

Figure 1. Stages in the progression towards more relevant models in cell-based assays

The predictive value of drug development should theoretically improve when increasingly (in vivo) relevant models are used. This graph outlines some directions taken in cell culture development to improve these predictive values. Two parameters, cell type and culture system, are plotted in an x–y graph. The linear curve thereby explains the hypothetical increase in predictability of a cell culture system with the alteration of these parameters. Along the x-axis, some examples of cell-types that have a higher relevance than cell lines in monoculture are plotted. This ranges from primary cells [87] to co-culture [243] to patient-derived materials [41], which is envisioned to be the most relevant model system. Systems that enable the culture and analysis of patient cells can be used for tailored treatment, which holds a high promise for the improvement of cancer therapy. On the y-axis, the culture platforms of different relevance are plotted. Depending on the model, a system with controlled extrinsic parameters may increase the relevance compared to standard 2D culture platforms [28, 244]. For example, long 2D patterns makes it possible to control cell growth of endothelial cells as capillary structures in 2D [26]. Platforms that enable 3D culture and recapitulate the organization of cells in vivo are probably amongst the most relevant [13, 33, 57, 87].

Figure 2. The role of the microenvironment in drug response and new treatment strategies

Factors present in the tumor microenvironment induce environment-mediated drug resistance (EMDR) by two primary mechanisms: soluble factor-mediated drug resistance (SFM-DR) and cell adhesion-mediated drug resistance (CAM-DR) (A). When the cancer is treated for the first time, most tumor cells respond to the drugs. However, the interaction with microenvironmental factors can give enough protective signaling for some of the cells to survive therapy and eventually to repopulate the tumor with resistant cells. Therefore, therapeutic strategies that disrupt EMDR pathways would reduce the level of surviving cells and prevent the emergence of acquired resistance [6, 15]. Cell adhesion via integrins is fundamental for cell survival, behavior and cell migration. Therefore, these receptors are central in the development of new tumor-stromal interaction disrupting drugs (B). Inhibiting specific integrins can reduce the effect of CAM-DR and block additional stromal processes such as angiogenesis [–131] (left). On the other hand, the specific tumor microenvironment can also be used for targeted delivery of drugs by, for example, recognition of a cancer-specific fibronectin [245] (right).

Figure 3. Controlling microenvironmental parameters by microfabrication

Engineering on the single cell level makes it possible to control the cell microenvironment on many different levels. The controllable parameters include the ECM coating, cell spreading (cell shape), dimensionality and rigidity for an open microfabricated system (A). In a closed system (B), the liquid volume is controlled, and thereby the concentration of solutes and flow are additional parameters that can be controlled in such a system.

Figure 4. The application of miniaturized cell systems in cancer drug development

Different types of cell arrays can be useful to determine the effect and the interdependence of environmental parameters on drug response. (A) A PEG-hydrogel microwell array platform with protein only at the bottom of microwells is exploited to study the effect of cell-matrix interaction and cell-cell interaction in 3D cultured breast cancer cells. On this platform, the extrinsic parameters could be controlled to mimic the niche of a breast cancer. Further, confocal microscopy allowed insight into the role of cell positioning within the cluster and cell-density, comparing the images from different positions within the cluster (z1, z2 and z3). One result of this work was that not only cell-cell interactions are important in 3D drug response but also matrix-interaction, although its effect is spatially limited [62]. Multi-organ interaction can be studied on “body-on-a-chip” platforms (B), which represent an in vitro alternative to the animal or even patient customized human model, which can test the uptake, distribution and metabolic digestion of a drug in different organs. In one approach, a system was developed to integrate the effect of the intestinal absorption and the hepatic metabolism of drugs against breast cancer [221]. Miniaturized spotted adhesion arrays (C), with mixtures of different proteins, enables the combinatorial effect of different matrix proteins to be studied. By screening different combinations of fibronectin with other matrix proteins, it was found that the adhesion of one cell line was increased on fibronectin combined with collagen IV. The same study used the array to correlate ECM ligands to metastatic behavior [171]. Miniaturization enables the immediate capture and analysis of small cell samples from patients (D). The limited number of cells in patient samples has so far restricted their use in diagnosis and treatment prediction. However, the direct analysis of patient cells is an important aspect in enabling personalized treatment. In miniaturized systems, small cell numbers can be analyzed with a signal-to-noise ratio, which is similar to that in standard macro scale analysis of larger cell populations. In a versatile approach using self-assembly of magnetic beads decorated with antibodies, cells could not only be captured within a microsystem but also further cultured therein and analyzed [242].