**1. Introduction**

Scaffolds are solid structures usually made of a polymeric material that is used for a wide range of applications. They provide a necessary support for three-dimensional (3D) cell growth, thanks to their biocompatibility and biodegradability [1], and are extremely useful in in vitro 3D cell cultures. Traditional cell culture is applied to two-dimensional (2D) models on flat surfaces, but this methodology is not representative of the cells' physiological environment and usually confers them with less malignancy. The literature has reported that 3D cell culture with scaffolds can increase the cancer stem cell (CSC) population [2–4]. CSCs correspond to a small population within the tumor

which is resistant to chemotherapy and capable of dividing to form the tumor again after treatment (this is known as recurrence or metastasis) [5,6]. Since this malignant subpopulation represents a small percentage within the tumor, the population expansion and enrichment described would help in their study and promote further development of therapeutic strategies.

Additive manufacturing (AM) technologies have arisen as a novel set of tools with which to fabricate scaffolds [5,7]. In particular, 3D printers based on fused filament fabrication (FFF) technology are one of the most accessible and simplest options [8]. They are open-source, low-cost machines which usually use thermoplastic materials [9,10] and can easily be modified to improve the quality of the printed 3D products [11]. A variety of biocompatible polymers can be used for scaffold production with FFF. Poly-L-lactic acid (PLA) is a biodegradable thermoplastic aliphatic polyester that has great potential in clinics thanks to its biocompatibility and restorability. Consequently, it is widely used in tissue engineering [12]. Poly(ε-caprolactone) (PCL; Figure 1) is also a biodegradable polyester proven to be biocompatible and toxic-dye-free, but it has a slower degradation rate and different mechanical and physical features. For instance, PCL has a lower melting point (60 ◦C), reflecting its lower hydrogen bonding and polarity which determine its chemical and molecular behavior. Moreover, PCL does not have any isomers so there are no variances in the melting temperature and biological degradation. Due to these characteristics, its use in tissue engineering, drug delivery, and cell cultures is increasing [2,3,6,13,14]. PCL can be also used as copolymers, such as PCL-collagen and PCL-gelatin, and in combination with other polymers, for example PLA or PEG [13,15].

**Figure 1.** Poly(ε-caprolactone) chemical structure.

As scaffold production with 3D printers is a new area, greater effort should be made to determine the optimal parameters for the process [1,6,9]. The processing parameters in question are closely related to the properties of the polymer chosen and the subsequent application intended for the scaffold(s). First, the design parameters determine the architecture of the scaffold and can comprise the filament diameter, the distance between filaments, and the deposition angle [16]. They can also be modified depending on the desired design and application of the scaffold. Second, fabrication parameters control the printing process. These parameters include the extruder and bed temperature, deposition velocity, and layer height, and are closely linked to the material of the polymer and the environment [9,17].

When scaffolds are produced for tissue engineering or regenerative medicine, controlling features, such as pore size, pore shape, or mechanical strength, is mandatory [9,18]. Although there are some studies into the 3D printing of scaffolds based on fused deposition modeling (FDM) [19,20] very few analyze the effects the architecture of the scaffold may have on cell proliferation, and none develop schematic procedures or methods aimed at retaining any knowledge gained. Grémare et al., [21] studied the physicochemical and biological properties of PLA scaffolds produced by 3D printing (FFF). The authors studied four different square pore sizes (0, 150, 200, and 250 um). Results showed that scaffold pore size had negligible effects on their mechanical properties. After three and seven days of human bone marrow stromal cell (HBMSC) culture being applied, the scaffolds exhibited excellent viability and homogeneous distribution regardless of the pore size. Hutmancher et al. [22] studied the mechanical and cell culture response of PCL scaffolds using 61 ± 1% porosity and two matrix architectures. Results showed that five-angle scaffolds had significantly lower stiffness under compression loading than those with a three-angle pattern. Data also revealed that in terms of cell proliferation, while a scaffold with a 0/60/120◦ lay-down pattern had a higher proliferation rate in the first 2 weeks, the scaffolds with a 0/72/144/36/108◦ lay-down overtook the three-angle matrix

architecture in Weeks 3 and 4. Recently, Rabionet et al. [23] analyzed the effects of tubular scaffold architecture on cell proliferation for vascular applications. Results showed the strong influence the 3D process parameters have on the scaffold architecture and, subsequently, cell proliferation. Narrow pores produced lower cell proliferation due to the lower oxygen and nutrient exchange.

As the literature has reported, cell proliferation onto a scaffold depends on the material, the architecture, and cell kinetics. Whenever physicians need to work with cells, they require the best scaffolding features to obtain ideal cell culture results. In fact, the main problem was that scaffolds did not provide the same results for different lines of cells when the cells are cultured. When working with cells, physicians have different purposes and goals. For instance, they may want to enrich or treat the cells or to determine the impact a drug is having/has had on the cells. While identical scaffold features do not provide the same results, the cell line does. In fact, each cell line works better with different scaffold features. For this reason, this work aims to optimize the design features and the selection of the manufacturing process parameters when the open-source 3D extruder machine RepRap is utilized. This methodology focuses on manufacturing PCL scaffolds suitable for 3D cancer cell cultures and CSCs expansion as a first step before expanding to other cell lines. Both design and fabrication parameters have been optimized by following a specific flowchart step by step, and checking a measurable variable. In addition, preliminary in vitro experiments were performed to study the impact the scaffold design and fabrication have on the efficiency physicians require from the 3D cell culture and the scaffolds produced. Therefore, a sample application for the mass production of PCL scaffolds using a low-cost machine could be used to improve cancer stem cell research. The flowchart developed here provides a novel methodology to adjust process parameters to print micrometric scaffolds suitable for three-dimensional cell culture because, as is demonstrated, each cell line required different scaffold features. Hence, an optimization diagram could represent a common procedure which could be used by non-engineering professionals when a 3D cell culture protocol has to be established de novo. Physicians working with 3D cell cultures usually need some kind of rules or guidelines to follow to set up the cell culture. This paper's contribution is the methodology required to set up the 3D printing technology for a new line of cell culture by first defining the design characteristics and then the parameter selection for the manufacturing process. This paper does not contribute to the knowledge about PCLs or the 3D printing machine itself, but instead provides a methodology for physicians. The contribution is the method and steps to follow when scaffolds need to be manufactured for a new cell line.
