**4. Discussion**

The theranostics substrates that we produced are akin to the cell culture glass coverslips used to plate cells for research and analysis—with the notable exception that they enable the targeted delivery of therapeutics to the cells and cell sensing, simultaneously. Moreover, thanks to the fabrication process capability to attain maximum control over surface morphology at the nanoscale, the substrates can be designed to guide cell adhesion, proliferation, and organization. For these characteristics, this bio-device—and its more sophisticated evolutions that will be developed over time—can be integrated into conventional cell culture dishes or multi-wells to test the adhesion and growth of cells against different external factors, including substrate geometry and a controlled delivery of drugs. Researchers can plate cells over several different replica of the device, each of them with its characteristic topography and drug release profile. Then, the researchers will find the combination of surface topography and device payload that guarantees maximum/minimum cell adhesion and proliferation, depending on whether the aim of the research is optimize a structure for tissue engineering or the effects of a drug for personalized medicine. The search for the optimal values of surface topography and kinetics of release should be possibly conducted within the bounds identified by this and other similar works: where the roughness of the surface and the pore size is varied in the 0–30 nm interval, while drugs are released with a maximum initial rate of ≈0.4 μM/h. The output of the experiment—i.e., cell colonies—can be verified at different time steps from seeding using either confocal microscopy or Raman spectroscopy that is, notably, made possible by the distinctive design of the device. While the first technique provides information about cell adhesion, growth, and clustering, Raman analysis of samples describes the conditions of a cell at the level of its adhesion molecules. Thus, the combination of techniques gives a picture of the evolution of a cell over different scales, bridging the divide between the behavior of cells being observed in isolation (individual behavior of cells) or in-group (collective behavior of cells). Consistency between results may indicate that the substrate operates efficiently towards either improving or impairing cell adhesion and organization. Thus, the device can potentially be the basis for a test campaign aimed to optimize the characteristics of biomaterials for tissue engineering, regenerative medicine, or in-vitro-model applications. After identification of the optimal surface characteristics and drug dosage that assure the wanted effect, these values should be copied to the scaffold intended to support cell-growth, or to the implantable device that will release drugs to a disease, for real-life applications. Nonetheless, this implies a process of engineering of the device, aimed to overtake those complications that can possibly emerge when similar devices are used outside of a research context. A list of possible caveats is identified as follows:


system, and an inactive (off) state, in which release is temporarily paused. This can be possibly accomplished by varying, in a controlled fashion, the levels of pH and temperature of the system, having previously conjugated the drug with a pH-sensitive cleavable linker as described, as for an example, for injectable nanoparticle generators in [56].

