The compounds employed in photopharmacology are mainly classified into five categories according to the mechanism by which light induces structural modification. A distinction is then made among photocleavage control, photodynamic therapy, photothermal control, photodegradation control and, photoconformational control.

Photocleavage control. In this approach molecules remain pharmacologically inactive as long as a suitable protective group remains bound to the drug’s structure (Figure 1a). Exposure to light causes its irreversible removal, releasing the pharmacologically active compound.

The two most studied photocleavable groups in this area are ortho-nitrobenzyl and coumarin derivatives [5,6]. The former absorbs in the far ultraviolet (~365 nm) and exhibits limited tissue penetration due to its relatively short wavelength. Coumarin derivatives, on the other hand, have been extensively studied because they have enabled the development of molecules activatable at higher wavelengths, even beyond 500 nm, thus improving compatibility with activation at tissue depths [7]. The limitation of these derivatives lies in their tendency to undergo competing radical pathways during photolysis, which can generate nitroso-aldehydes and other reactive byproducts instead of clean drug release. This side reactivity depends strongly on both the nature of the leaving group and the architecture of the linker connecting the cage to the pharmacophore; accordingly, new o-nitrobenzyl and coumarin-based spacers are being developed to favour heterolytic cleavage, suppress radical channels, and minimize cytotoxic photoproducts [8,9,10,11]. An illustrative example of the biological potential of photoreleasable compounds is provided by Zhao and co-workers, who demonstrated that vanilloid derivatives can be photoreleased in situ to activate TRPV1 receptors on nociceptive neurons, with one-photon quantum efficiency as shown in Figure 1b [12]. Most of the initial examples were cleaved with UV light, which suffers from low tissue permeability and cell damage. Recently, efforts have been made to obtain visible light excitable scaffolds and far-red/near-infrared (FR/NIR) systems operating in the therapeutic window (650–900 nm) [13,14]. These FR/NIR photocages provide more efficient and versatile drug release—especially in hypoxic cancer environments—and show enhanced performance when combined with targeted therapies or functional conjugates.

Photodynamic therapy (PDT). It is currently the only clinically approved form of photopharmacology [15]. The treatment involves the administration of a photosensitizing dye, followed by localized irradiation with visible light, which generates an in situ excited state leading to the formation of cytotoxic singlet oxygen or other reactive oxygen species (ROS) that induce localized cell death (Figure 2a).

PDT has been successfully employed in the treatment of cancers, dermatological conditions, and ophthalmic diseases, and has pioneered the therapeutic use of light in combination with pharmacological agents. Clinically approved photosensitizers include Foscan^® (depicted in Figure 2b), a chlorin used for the PDT treatment of advanced head and neck, prostate and pancreatic cancers, Photofrin^®, and other [16,17,18].

Despite its clinical success, PDT still faces several challenges that limit its widespread clinical application:

1. PDT requires effective light delivery to the target site, restricting its therapeutic use to superficial tumours or lesions located in tissues adjacent to accessible organs [19].
2. PDT often relies on high-energy, short-wavelength excitation (UV, blue, or green light), which significantly limits tissue penetration [20].
3. The abnormal vasculature and rapid proliferation of cancer cells in solid tumours lead to hypoxic microenvironments with reduced oxygen availability, which severely compromises the generation of reactive oxygen species and consequently diminishes PDT efficacy.

To address these limitations, chemo-photodynamic therapy (chemo-PDT) has emerged as a complementary strategy, combining conventional chemotherapy with PDT to achieve synergistic anticancer effects [21]. By integrating light-activated cytotoxicity with chemotherapeutic mechanisms that are less dependent on oxygen availability or light penetration, chemo-PDT can partially overcome tumour hypoxia and depth-related constraints while enhancing overall therapeutic efficacy. Recent work has shown that nanoplatform-based chemo-PTD strategies can modulate the hypoxic tumour microenvironment, limiting hypoxia-driven tumour progression [22]. Beyond chemo-PDT, recent evidence indicates that integrating PDT with ferroptosis-inducing mechanisms enhances ROS-driven cytotoxicity and mitigates hypoxia-associated resistance, offering a powerful extension of conventional chemo-photodynamic strategies [23].

Photothermal control. Photothermal approaches exploit light energy, typically in the near-infrared (NIR) region, to generate localized heating that can induce thermal ablation or activate controlled drug-release platforms such as gold or gold/silver hybrid nanoparticles (Figure 3a).

These structures can be functionalized with temperature-sensitive polymers or proteins, which release the active compound following conformational changes or phase transitions induced by heat [4,24]. An example of NIR-triggered photothermal-chemotherapy is illustrated by the schematic preparation of Mo154 complex micelles (Figure 3b) [25]. From a therapeutic perspective, photothermal strategies offer several advantages, including deep tissue penetration using NIR light and compatibility with minimally invasive light-delivery techniques. As a result, photothermal control has been widely explored for tumour ablation, combination therapies, and stimulus-responsive drug release in precision medicine [26,27]. Photothermal approaches have already found clinical application primarily as device-based therapies, such as laser interstitial thermal therapy (LITT), which are approved for the treatment of brain tumours and drug-resistant epilepsy [28]. In addition, clinically approved dyes such as indocyanine green have enabled drug-assisted photothermal effects in selected pharmacological settings [29].

Photodegradation control. In this case, polymeric supports or vectors that irreversibly degrade under the action of light, causing controlled local drug release, are used (Figure 4a). These systems exploit photocleavable groups inserted into the polymer or vector structure, which upon irradiation break chemical bonds that maintain material cohesion, causing fragmentation or increased porosity. This mechanism allows targeted and rapid release of entrapped drug molecules with high spatial precision. An additional advantage is the ability to modulate release kinetics by adjusting the intensity and duration of irradiation. However, an important challenge remains the assessment of the biocompatibility and clearance of residual polymer degradation products, which may have systemic effects not yet fully understood [30,31].

As demonstrated by Kubota and Ouchi [32], the incorporation of a photosensitive ortho-nitrobenzyl group into vinyl copolymers enables selective backbone degradation via photo-deprotection, without the need for acidic catalysts (Figure 4b).

Photoconformational control. These molecules are able to undergo reversible isomerization between two conformational states (e.g., cis and trans), modifying their pharmacological activity in response to specific wavelengths (Figure 5a). This type of approach has gained traction in recent years, with the advent of molecular switches such as azobenzene, stilbenes, and diarylethenes [33].

The introduction of azobenzenes can be accomplished by using azo-extension or azo-logization strategies. Azo-extension refers to the strategy of inserting the azoarene photoswitch directly at a strategic position on the parent molecule, extending the conjugation system without significantly altering the original pharmacophore. Azologization instead involves replacing a central structural unit of the parent molecule with an azo moiety, fundamentally changing the molecular scaffold to incorporate the photoswitch (Figure 6) [34].

Following the conformational change induced by light (photoswitch), spontaneous return to the initial conformational state is often observed, generally through a thermally induced process (T type-thermally reversible- photochromic molecules).

This mechanism offers a potential advantage over photocleavage, as molecules activated in the target tissue, once diffused into surrounding areas, can revert to the inactive form, thus reducing the risk of systemic toxicity. Azobenzene derivatives have been extensively studied for their light-triggered isomerization; while regular azobenzene absorbs in the UV, many engineered derivatives show visible-light responsiveness [35]. Although some azobenzene derivatives can undergo metabolic reduction in vivo, thus resulting in rapid elimination or formation of potentially toxic metabolites, their metabolic stability is highly structure-dependent. In fact, many azobenzenes show resistance toward glutathione-mediated reduction and are not substrates for all azoreductases [36]. An example of photoconformational control is provided by azobenzene-modified imidacloprid derivatives, which exhibit light-dependent activity modulation (Figure 5b) [33]. In contrast, systems based on isomerization of carbon–carbon double bond absorb in the ultraviolet region and require more energetic light radiation, which is less effective at penetrating tissues. On the other hand, they exhibit greater chemical stability, being less susceptible to metabolic degradation and, in many cases, safer toxicology [37,38,39,40,41]. In line with stilbene derivatives, diarylethenes (DAEs) are P-type (thermally irreversible, but photochemically reversible) photochromic molecules, endowed with high fatigue resistance and efficient photoconversion between their open and closed forms (Figure 7) [42,43].

Although the structural rearrangement associated with switching is relatively small, DAEs offer exceptional photochemical robustness and can be operated with visible light, making them well suited for applications requiring prolonged retention of a selected photoisomer without continuous irradiation [44].

On the other hand, spiropyrans (SPs) function through a light-induced ring-opening mechanism that interconverts a neutral, non-planar spiropyran with a highly polar, conjugated merocyanine form (Figure 8) [45,46].

This transition entails profound changes in polarity, geometry, and electronic distribution, often resulting in marked differences in biological activity between the two isomers.

Together, these photoswitch families exemplify how diverse photochemical scaffolds can expand the scope of photopharmacology, providing complementary properties to azobenzenes and enabling tailored light-responsive behaviour for specific therapeutic or mechanistic applications.

In photopharmacology, reversible photoswitches stand out for their numerous advantages over other light-control strategies of pharmacological activity. Unlike photocleavage and photodegradation, which involve irreversible modifications and the production of potentially toxic chemical residues, photoswitches allow for dynamic and reversible modulation of the therapeutic effect, ensuring precise temporal and spatial control. Compared to photothermal control, photoswitches offer greater chemical compatibility, absence of toxicity from external materials, and high temporal resolution, ideal for modulating rapid or recurring biological events. Furthermore, the possibility of integrating photoswitches into various bioactive molecules without compromising their functionality greatly expands the therapeutic potential of this technology, making it one of the most promising and safest strategies in the field of photopharmacology [47,48].

When discussing reversible photopharmacological tools, several key physicochemical parameters must be considered, as they ultimately determine whether a compound can be effectively applied in biological systems. The most critical among these are activation wavelength, thermal half-life, chemical stability and water-solubility.

Activation wavelength refers to the specific value of light radiation required to induce the transition of the drug from an inactive to an active state. While longer wavelengths (e.g., NIR) offer deeper tissue penetration, shorter wavelengths such as UV are more energetic but have minimal penetration and can induce tissue damage. Since many common photoswitches are activated in the UV range, these limitations must be carefully considered when evaluating their applicability in vivo [49].

Thermal half-life represents the time required for the photoinduced and thermodynamically less stable form of the drug to spontaneously revert to the more stable form. This property is governed by the energy barrier between the two isomers and determines the duration of the drug’s effect in the absence of further light stimulation. An adequate half-life allows effective temporal control of pharmacological activity [50].

Chemical stability of the photoactivatable compound refers to the drug’s resistance to chemical and metabolic degradation, both during storage and within the organism. Good stability ensures that the active principle maintains its efficacy and does not lead to the formation of undesired or potentially harmful products [38].

Water solubility represents a major challenge in photopharmacology, as most photoswitch scaffolds are intrinsically hydrophobic owing to their extended aromatic π-systems. Limited aqueous solubility constitutes not only a pharmacokinetic constraint but also a photophysical and photopharmacological bottleneck, as it can alter the photochemical behaviour of photoswitches by promoting aggregation and affecting photostationary states and switching efficiency under physiological conditions [51]. To overcome these limitations, several design strategies have been explored, including the introduction of polar or ionizable substituents, the use of heteroaromatic or charged photoswitch scaffolds, prodrug or photocaging approaches, and conjugation to hydrophilic carriers. Notably, emerging classes of photoswitches, such as hemiphosphoindigos incorporating phosphinate groups, exemplify innovative solutions for achieving water compatibility while preserving robust photophysical properties [52].

In summary, the efficacy and safety of a light-activatable drug depend on the balanced optimization of these interrelated parameters. A rational and integrated design of photochemical, thermodynamic, and physicochemical properties is therefore essential for the development of precise and reliable light-based therapies [53,54].

Despite significant progress, photopharmacology still presents some critical issues limiting its practical application. Among these, the limited penetration of light in biological tissues, especially for ultraviolet and blue wavelengths, represents a significant obstacle for clinical use at depth. Currently, the clinical applicability of photopharmacology remains largely at the preclinical stage and requires further studies on safety and efficacy [55,56].

Looking ahead, the field of photopharmacology appears extremely promising and continuously evolving. Among the main development lines is the design of photoswitches activatable with light in the red or near-infrared range, which would favour greater tissue penetration and broader clinical applicability [57,58]. On the other hand, the development of new implantable devices for in situ irradiation could represent a turning point for the use of numerous compounds with photomodulable activity [59,60,61].