In recent years, there has been a substantial upsurge in interest in the development of systems for monitoring cell cultures. This could be attributed to a series of factors. On the one hand, the exploration of microfluidics and lab-on-a-chip devices has allowed the generation of intricate systems for cell culturing devices. The ability to create distinct and separated compartments for cell culture and biosensing, each tailored with their specific chemical and biochemical functionalizations, along with the capacity to construct intricate fluidic networks that interconnect these sections represents just a subset of the advantages offered by these systems [51,52,53,54]. On the other hand, advances on materials and biomaterials research has enabled both the development of suitable substrates for the generation of complex microenvironments and the fabrication of novel biosensors. The arising of smart, functional, and biocompatible materials in the recent years has allowed to speed up the development of multifunctional platforms for cell culture [55,56,57]. While most of what has been achieved so far offers a promising look on what can be accomplished in the near future, these new platforms are still on very early stages of development and can only be defined as proof-of-concept.

Ideally, a complete integrated platform should combine the culture of cells under a controlled microenvironment and the monitoring of cell secretion within the same device, independently of the applied biosensing method. However, the absence of a standardized definition for what constitutes a platform that truly integrates cell culture and secretion, as well as the specific components that should be included within such a platform, can be primarily attributed to the early developmental stage of this research area. Consequently, what has been described as “integrated” up to this point exhibits differing levels of implementation concerning cell culture and secretion monitoring within the same system.

Taking all into account, we have focused our literature survey on recent cell monitoring platforms that combine the capture and maintenance of cells, whether through regular 2D cultures, microfluidic-based cell trapping, or the incorporation of 3D cultures, with the integration of bioreceptors for capturing signaling molecules within a single device. This includes both devices that require further extraction of the bioreceptors for analysis, usually employing barcode systems and microbeads [58,59,60,61,62] and those that enable direct analysis of secretion within the same device. Table 1 lists the publications considered within these parameters, detailing the biological and sensing characteristics.

Cell monitoring platforms that integrate sensing for secreted signaling molecules can be categorized into two main groups based on the placement of the bioreceptors: those where bioreceptors are placed in a biosensing compartment independent from the cells, and those where bioreceptors are placed in direct proximity to the cells (see Table 1, column 6 “Bioreceptors”).

The first approach employs microfluidics to create complex systems with fluidic networks that transport secretions from each cell culture compartment to a dedicated biosensing compartment [58, 63,64,65,66,67, 67,68,69,70,71,72,73] (Fig. 3A). These compartments can be either fully integrated into a single device or exist as modular components that can be interconnected.

Strategies followed for the monitoring of the secretion of signaling molecules of a cell culture. A Transport of the secreted biomolecules from the culture chamber to the biosensing chamber within a microfluidic device. (i) A schematic drawing of the devices. (ii) and (iii) Examples of real devices, adapted from Rodriguez-Moncayo et al. [67], with the permission of ACS publications, and from van Neel et al. [82] with the permission of ACS publications. B Controlled placement of the biosensors in the proximity to the cell culture. (i) A schematic drawing of the devices. (ii) and (iii) Examples of real devices, adapted from Armbrecht et al. [70], with the permission of Wiley, and from Ramadan et al. [58] with the permission of AIP publishing

The second strategy relies on directly detecting secreted biomolecules from each cell event by placing bioreceptors in close proximity to the cells, often utilizing particles or features like barcodes. Integrating both cell culture and biosensing within a single compartment enables immediate capture of cell secretion and its precise correlation with the secretory cell. The bioreceptors or biosensors can be positioned around the cells [70, 74,75,76], underneath the cells [59, 77,78,79,80,81], on the cell surface [82, 83], in the medium [60, 64, 85,86,87,88], and on top of the cells [61, 62, 89, 90] (Fig. 3B).

Both strategies offer their own set of advantages and limitations. Placing bioreceptors in close proximity to the cells enables monitoring cell secretion with a spatial resolution unattainable by conventional methods, enhancing the ability to correlate cell secretion with individual cell events. However, this advantage comes with increased complexity in modulating substrates to accommodate both cellular and sensing elements, and it is more restrictive when combining cell secretion analysis with the other optical methods for cell monitoring. This complexity is particularly evident when unconventional materials are required as part of the system, such as polydimethylsiloxane (PDMS) for most microfluidic devices and gold layers for SPR-based sensing. On the other hand, generating individual but interconnected compartments facilitates fabrication, enabling adaptation of each compartment to its specific requirements. This not only allows for proper cell culture without the burdens that arise from the sensing requirements, both from a chemical and material point of view, but also enables easy integration of conventional assays such as immunocytochemistry. However, this approach sacrifices the spatial resolution achieved by the first strategy. In summary, one strategy does not surpass the other; each approach should be considered based on the type of analysis desired.

Most platforms developed to date have focused on exploring well-known secretion models to test and validate the technologies. This approach has led to a limited catalog of signaling biomolecules applied in cell sensing and a clear trend in the biological models utilized. Notably, immune cell-related models, often coupled with cytokine sensing, are the most extensively explored (Table 1, column 2 “Cell type measured”, column 8 “Biomolecules”). Interleukins (IL)-2, 4, 6, and 8; tumor necrosis factor (TNF); and interferon-γ (IFN) are some of the secreted cytokines whose detection and quantification have been incorporated in the most recent cell culture platforms [58,59,60,61,62,63,64, 70, 72, 73, 75,76,77,78,79,80,81,82, 85, 86, 89,90,91,92,93,94,95]. There are several reasons for researchers to use this particular model. Firstly, cytokine secretion models are well known and have been thoughtfully studied over the past years, presenting a broader window of standardized detection methodologies that can be adapted to new platforms, which facilitates the development and their validation of novel biosensing techniques. Secondly, cytokines are widely used as biomarkers and reporters to address pathological conditions, giving to the resulting product an excellent potential to be applied in clinical practice [96, 97]. Finally, the type of cell cultures used to generate these models have been mostly based on cells derived from the immune system [58, 60, 63, 64, 66, 67, 72, 73, 75,76,77, 85, 86, 89, 93,94,95], which are notably easier to integrate into a platform when compared to other cell types. This ease of integration is largely attributed to their non-adherent or transient nature, which in turn requires a less intricate microenvironment for their proper cultivation, especially in contrast to cell types derived from complex physiological microenvironments [98]. Therefore, it is easier to implement the replication of their microenvironment inside a platform.

A second model also widely used is the monitoring of growth factor secretion, such as the vascular endothelial growth factor (VEGF) and the hepatocyte growth factor (HGF) [60, 68, 70, 82, 88, 92]. These well-established models also present a huge clinical potential due to their regulatory actions and relation with pathological conditions. Another example of a secreted molecule that has been reported is a hormone like insulin [6