Previous studies on the composition of exosomes have revealed the presence of various cargo molecules, including membrane proteins, cytosolic and nuclear proteins, extracellular matrix proteins, metabolites, and nucleic acids, such as mRNA, noncoding RNA species, and DNA (34, 35). The composition can vary widely based on cellular origin, heterogeneity in size, cellular microenvironment, and inherent biology of cells (4).

The diverse composition of exosomes results in the complexity of their functions. Additionally, the effects of exosomes on receptor cells vary based on the expression of receptors on the cell surface. Generally, exosomes are involved in eliminating excess metabolites to maintain cellular homeostasis (36) and intercellular communication, which has been associated with numerous physiological and pathological functions (37). In cancer cells, exosomes are known to support tumor progression by promoting angiogenesis, regulating the immune system, and remodeling the surrounding parenchymal tissue (38). In breast milk, breast milk-derived exosomes contain miRNAs with immunomodulatory functions to regulate immune tolerance and promote postnatal health and growth (39).

Recent studies on the central nervous system have indicated that exosomes secreted by oligodendrocytes enhance neuronal viability as well as an increase in the neuron firing rate (40). In a syngeneic mouse model of melanoma, melanoma-derived exosomes induced lung vascular leakiness and recruited bone marrow-derived macrophages to metastatic sites. Moreover, chemotherapy and radiation therapy also impact the composition and functioning of exosomes with potential implications on therapy outcomes (41).

Current studies are involved in the exploration of exosomes as a new method for the cellular delivery of drug payload. There are several techniques being used to separate exosomes from EVs ( Table 1 ); however, there are still many challenges regarding the efficient and economic production, separation, and purification of exosomes in adequate quantities, which prevent their clinical use (9).

Therefore, it is imperative to promote a standardized method that is practical, efficient, and feasible to isolate and purify exosomes from the cell culture medium.

Exosomes can transport DNA, RNA, and proteins into biological fluids to mediate intercellular communication (4). Tumor cell-derived exosomes contain oncogenic information that facilitates cancer growth and metastasis (51). Thus, it was recently predicted that exosomes could act as tumor biomarkers (52, 53). A study showed that TAG72-rich and CA125-rich exosomes could act as biomarkers of colorectal cancer (54). A significant increase in the serum/plasma levels of miR-547-3p was observed in patients with osteosarcoma (55) or hepatocellular carcinoma (56) and considered as a diagnostic biomarker, and miR-547-3p levels were significantly reduced post-radiotherapy in patients with glioma (57). Exosomes can be detected in the early stages of cancer, suggesting that they might be vital candidates for laboratory examinations.

Tumor-derived exosomes possess tumor markers, and tumors can be identified based on the levels of tumor-derived exosomes in bodily fluids ( Figure 2A ) (60). However, during the early stages of development, tumors secrete few exosomes, which are difficult to detect, and thus, detachable microfluidic devices were invented to detect early-stage tumors using electrochemical aptamer sensors (DeMEA). This equipment can improve the sensitivity of the detection of tumor-secreted exosomes (61). It provides the possibility of early detection of cancer. When a tumor progresses to a terminal stage, it is critical to determine its level of metastasis to develop the treatment strategy. Thus, an ingenious microfluidic device was designed to detect exosomes based on the specific binding of lectins to unique glycans on the surface of cancer cell-derived exosomes, which could distinguish between metastatic and non-metastatic pancreatic cancer while improving the accuracy of pancreatic cancer diagnosis ( Figure 2B ) (58).

Compared with laboratory examinations, imaging examinations are more intuitive and visual (62). Exosomes are considered to be natural delivery vesicles that can slow down the endocytosis of macrophages and extend blood circulation time to avoid clearance via the kidney, liver, and the immune system (28), implying that their biocompatibility and transcellular permeability present them as probable vehicles for theranostic applications. Currently, for imaging analysis, the cargo molecules that can be loaded by the exosomes include inorganic molecules (63), magnetic materials ( Figure 2C ) (59), and fluorescent molecules (64). Besides, there is two main preparation method of exosome theranostics: in-vitro biosynthesis and in-situ biosynthesis.

In-vitro biosynthesis is a more controllable technique. In a study, chlorin e6 (Ce6) molecules were loaded on the surface of inorganic Au nanoparticles (Au@Ce6), followed by embedding the Au@Ce6 into exosomes, which were collected and purified from the urine of gastric cancer patients. The results showed that the nanovehicles could successfully improve the cellular uptake of cancer cells and reduce the endocytosis of macrophages, which further enhanced the accumulation of targeted tumor sites for superior real-time fluorescence imaging (63). Another study used superparamagnetic iron oxide (SPIO)-loaded exosomes for magnetic particle imaging (MPI), which enabled the acquisition of in vivo images with greater contrast, sensitivity, and quantitation than magnetic resonance imaging (MRI) (65). Apart from this, gold-iron oxide nanoparticles-loaded exosomes also demonstrated excellent T2 contrast in MRI ( Figure 2D ) (66).

In-situ biosynthesis is a more sensitive technique where the raw materials of the bioprobe are injected into the organism, followed by their self-assembly in the tumor microenvironment (TME). After the bioprobe assembly is complete, it gets embedded into the exosome as metabolic trash and excreted outside the cell. Thus, it can be found in the early stages of cancer. A study utilized TME for bioimaging with in-situ biosynthesized nanoscale gold and iron probes and subsequently disseminated Au-Fe nanoclusters from cargo exosomes within the circulation, which could be subsequently used as fluorescence, Computed Tomography (CT), and MRI imaging tools for cancer diagnosis (67).

In clinical studies, the application of biopsy to brain cancers is limited due to the location of the nidus. Usually, the diagnosis is confirmed by imaging analysis and laboratory examination, and biopsy is performed only when there are clinical symptoms; however, by then, the cancer has developed to middle or advanced stages. Therefore, it is necessary to improve the sensitivity of imaging techniques and laboratory examination and strive for early diagnosis. The use of exosomes in the diagnosis may solve this problem. In the laboratory diagnosis, once the TME is established, exosomes containing oncogenic information are released into the peripheral blood, which can be detected by laboratory tests (57). As for the image-based diagnosis, encapsulating particles with imaging capabilities in exosomes can be absorbed by the brain cancer cells due to their capability to cross the blood-brain barrier (BBB) (65, 68) or be transported from the nidus to the peripheral blood (67), which can effectively improve the imaging signals of CT or MRI and enhance the sensitivity of imaging diagnosis. Based on these conclusions, the combination of improved laboratory tests and imaging techniques by exosome can facilitate early diagnosis of brain cancers, reduce the rate of missed diagnosis, and achieve early treatment of brain cancers.

In summary, exosomes could improve the sensitivity of laboratory tests and imaging to detect tumors. In laboratory examinations, the presence of exosomes with tumor markers in bodily fluids could be used to detect tumors. Further research invented numerous devices to improve the sensitivity of detecting early tumors as well as for distinguishing whether tumors were metastatic by analyzing exosomes, with improved clinical efficacy. In imaging examinations, exosomes derived from tumor cells were found to enhance the signal at the tumor site by loading inorganic molecules, magnetic materials or fluorescent molecules. These involved two methods: (1) in vitro biosynthesis where fluorescent molecules were loaded into the exosomes in vitro and then transported to tumors through the homing ability of exosomes; (2) in-situ biosynthesis where the raw materials of the bioprobe were injected into the organisms and then self-assembled in the tumor microenvironment (TME).

As mentioned above, the use of RNAs in oncology has been limited by their easy degradation by nucleases, which reduces their stability in the blood (95). Furthermore, the pore size of the glomerular basement membrane (GBM) is approximately 6-10 nm (96), while the length of naked RNA is only 7 nm; thus, it can be filtered through GBM and pass into the urine quickly (97). Currently, scientists assume that this defect in RNAs could be remedied by using the exosome. Several experiments have been designed to verify this hypothesis. For example, one study found a significant increase in the storage stability of RNAs after loading with exosomes compared with only PEIs-modified siRNAs (92). Plasma-derived exosomes also protected miRNAs from nuclease degradation and successfully promoted apoptosis of the HepG2 hepatocellular carcinoma cells (77). Again, siRNAs loaded with milk-derived exosomes are stable and resist degradation; the combination of exosomes and siRNAs eventually silence the expression of target genes by approximately 2-10 folds ( Figure 3A ) (69). Thus, exosomes, with no inherent toxicity and intrinsic biocompatibility, can protect RNAs from nuclease degradation and renal filtration and help maintain the stability of RNAs in blood.

However, only maintaining stability is not enough. Since naked RNAs have a poor ability of cell penetration, it is hard for RNAs to enter into the cell (98). Even after entering the cell, they run the risk of being hydrolyzed by lysosomes (89). Consequently, the success of therapeutic RNAs depends on efficient cell internalization and timely endosomal escape. It is well known that exosomes have the innate ability to cross biological barriers; thus, engineered exosomes could efficiently facilitate cellular uptake of miRNAs and significantly reduce cancer proliferation (80). Professor Lin has proved that milk-derived exosomes can transport across the gastrointestinal barrier (99), and small interfering RAN (bcl-2 siRNA) encapsulated by milk-derived exosomes can effectively pass through cell membranes and inhibit the growth of cancer cells (84). Additionally, a study demonstrated that sometimes, exosomes fused with the target cell membranes, releasing their cargos directly into the cytoplasm. Thus, exosomes could help avoid the endosome trapping, enabling the full functionality of the RNAs and effectively inhibiting tumor growth ( Figure 3B ) (100). If the exosome was internalized via the clathrin-mediated or clathrin-independent endocytosis, it could also help RNAs escape digestion by reverse fusion with the limiting membrane of the MVE or by being re-secretion via the early endocytic recycling pathway (30).

RNA therapy runs the risk of off-target effects, which can result in unanticipated phenotypes (89). Currently, active targeting might constitute the best way to solve this problem. Compared with other vehicles, exosomes have better targeting ability. First, exosomes have been proved to have a weak homing effect, i.e., cancer-derived exosomes are preferentially absorbed by the cancer cells (65, 101). Second, the exosomes originate from the cell membrane (4); thus, they can be modified via different targeted peptides or proteins and then can actively target cancer cells. Recently, exosomes were modified by a variety of target ligands, such as tLyp-1 (70), folic acid (69), iRGD peptide (102), and T7-peptide (79) in RNAs-based cancer therapy. Thus, this approach can limit the off-target effects and promote cellular uptake of RNAs by cancer cells, either passive targeting or active targeting.

The poor diagnostics, therapeutics, and prognosis of brain cancer are well known. It mainly includes glioblastoma, medulloblastoma, and oligodendroglioma. Currently, a standard treatment for brain cancer is surgical resection, followed by radiotherapy and chemotherapy. However, surgical resection is limited because of the diffusive nature of brain cancer and the intolerance in some patients. Also, chemotherapy and radiotherapy can have serious adverse effects; thus, brain cancer has a low cure rate and a high recurrence rate clinically. Therefore, a drug with low-toxicity, low-immunogenicity, high bioavailability, ability to cross the blood-brain barrier (BBB), and active targeting is needed. Most naked molecules do not meet these requirements. Fortunately, exosomes possess an invisibility cloak, increasing their functionality by wrapping them in a vesicle. In other words, they can carry the cargos and pass through biological barriers smoothly, including BBB; thus, improving the biological availability of cargos (103). Additionally, exosomes are non-toxic, which will not lead to acute immune rejection, thus giving the cargos high biocompatibility (7). Also, exosomes present a customizable surface and can be modified for different applications (104). These characteristics of exosomes are exclusive to them (105), making them the preferred carriers in brain treatment.

For the treatment of brain tumors, RNA therapy is not only hindered by the BBB but also by the off-target effects. The complex and mysterious structure of the brain makes the adverse effects of off-target unpredictable. One study analyzed the delivery of miRNA-21 using T7-peptide-decorated exosomes via intravenous injection. After systemic administration, this compound showed a brilliant brain-targeting ability and reduced tumor size in orthotopic glioblastoma models without adverse effects. Similarly, Lamp2b and rabies virus glycoprotein (RVG) are also known to target glioblastoma ( Figure 3C ) (79). If unmodified exosomes are used to deliver RNAs, then intra-tumor injection may be the most efficient and safe method (106).

In summary, exosome-loaded RNA for tumor treatment could improve the disadvantages of RNA, such as rapid hydrolysis, poor bioavailability, and inability to cross the biological barriers. Experiments have shown that exosomes can store RNA, avoiding RNA degradation by enzymes and maintaining its stability in the blood. In addition, exosomes can carry RNA across biological barriers, such as the blood-brain barrier and the gastrointestinal barrier. Finally, RNA therapy has the risk of off-target effects. Thus, exosomes can help RNA overcome off-target effects and improve targeting ability, in turn improving anti-tumor efficacy.

Several natural products and extracts, such as paclitaxel, curcumin, celastrol, anthocyanin, etc. suffer from low bioavailability in vivo despite having excellent anti-cancer properties. Numerous experiments have shown that exosomes can improve the bioavailability of these drugs and improve their efficacy in cancer treatment. For example, aminoethylanisamide-polyethylene glycol-vectorized exosomes loaded with paclitaxel (AA-PEG-exoPTX) exhibit a high loading capacity, excellent accumulation in cancer cells upon systemic administration, and improved therapeutic efficiency (111). In this previous study, AA-PEG-vectorized exosomes increased the aqueous solubility of PTX and promoted the accumulation of PTX in tumor tissues. In another experiment, the poor aqueous solubility and instability in the blood of celastrol were improved via exosome wrapping. Thus, exosomes improved the bioavailability of celastrol and enhanced its anti-cancer efficacy (119).

Compared with systemic administration, oral administration is non-invasive and easy for the patient; thus, it constitutes an excellent treatment method for chemotherapy. However, the gastrointestinal barrier hinders the absorption of drugs, especially gastric acid, which reduces the stability of drugs. Fortunately, milk-derived exosomes can stably exist in gastric acid; thus, avoiding the degradation of the cargo molecules (99). Curcumin is well known for its excellent anti-cancer properties but poor aqueous solubility. Its hydrophobic nature results in not only poor aqueous solubility but also rapid intestinal hepatic metabolism. A study showed that exosome-encapsulated curcumin could stably exist in gastric acid and smoothly cross the gastrointestinal barrier to improve the bioavailability of curcumin. Additionally, compared with the pure curcumin group, exo-curcumin showed significantly improved accumulation in tumor cells as well as enhanced anti-tumor efficacy (108).

Most chemical drugs, such as adriamycin, cisplatin, oxaliplatin, etc. suffer from low biocompatibility. The adverse effects of these drugs limit their clinical application. Theoretically, carriers with active targeting ability, low immunogenicity, and self-degradability should solve this problem. Autologous-derived exosomes elicit a negligible immune response or an inflammatory reaction, as well as possess the ability of self-degradation and active/passive targeting via surface or gene modification. These characteristics make exosomes more suitable than other carriers to improve the biocompatibility of chemical drugs (120). A study developed A33 antibody, which is uniformly expressed in colorectal cancer, positive exosomes loading doxorubicin (Dox) to improve the targeting ability, and this system showed high efficacy for the treatment of colorectal cancer without causing cardiotoxicity. Additionally, it demonstrated that exosomes fused with target cells more efficiently under acidic conditions, implying that exosomes facilitated drug release in acidic cancer cells ( Figure 4A ) (112). Furthermore, autologous exosomes can also reduce the hepatotoxicity and nephrotoxicity of gemcitabine by mediating the delivery of drugs to the cancer site due to its special homing effect (117). Additionally, the use of exosomes increased cancer cell uptake, minimized immunogenicity, and enhanced the biocompatibility of aspirin ( Figure 4B ) (115).

Clinically, multidrug resistance is a fatal blow to cancer chemotherapy. Additionally, the TME is well known for its potential for drug resistance. In fact, exosomes play a crucial role in this process since they transmit the information related to drug resistance between cancer cells. On the contrary, for cancer treatment, exosomes can also be used to transmit information about reversing drug resistance between cancer cells. Thus, caudally injected exo-anti-miR-214 was shown to reverse the resistance to cisplatin in gastric cancer and repress cancer growth in vivo by downregulating the expression of miR-214, which promoted gastric cancer chemoresistance (121). Additionally, in pancreatic ductal adenocarcinoma, chemoresistance to gemcitabine (GEM) significantly reduced the sensitivity of chemotherapeutic agents. The downregulation of key proteins, such as human equilibrium nucleotide transporter 1 (hENT1) and deoxynivalenol kinase (dCK), reduced the transport of GEM to the cytoplasm and insufficient conversion of GEM phosphorylation products, resulting in GEM-related chemoresistance. The internalization of exosomes into the cell was shown to be energy-dependent and was mediated by clathrin-independent endocytosis and macropinocytosis. Thus, GEM could enter the cytoplasm through a different internalization via exosomes, avoiding the primitive transport involved in human equilibrium nucleotide transporter protein 1 (hENT1) and thus, resist resistance (116). Drug resistance has been associated with an efflux transporter known as P-glycoprotein (P-gp) that is located in the cellular membrane (122). A study showed that the PTX-loaded exosome could bypass P-gp-mediated drug efflux in the resistant cancer cells to overcome the chemoresistance ( Figure 4C ) (123).

Due to chemoresistance and the limitations related to BBB, most chemotherapy drugs show poor therapeutic efficacy in brain tumors. The exosome is an extraordinary carrier with improved drug bioavailability, biocompatibility, and the ability to cross the BBB, which is helpful in improving the efficacy of drugs. For example, doxorubicin- and paclitaxel-loaded exosomes can cross BBB smoothly, with good bioavailability and anticancer efficacy ( Figure 4D ) (124). In another study, neuropilin-1-targeted peptide (RGE) was conjugated with the exosome membrane to improve its biocompatibility (110). As previously mentioned, brain cancer is also affected by multidrug resistance. It has been proved that miR-9 plays an important role in the development of drug resistance in glioblastoma cells against temozolomide by inducing an increase in P-gp. After entering the cells by exosomes, anti-miR-9 could reverse the expression of the multidrug transporter and sensitize glioblastoma cells to temozolomide (125). Thus, exosomes can be used to increase the therapeutic efficacy in brain cancer, especially in crossing BBB and overcoming chemoresistance.

In summary, the anti-tumor effect of chemotherapy was mainly inhibited by the low bioavailability and poor biocompatibility of the drug and the drug resistance in cancer cells. Exosomes could improve their bioavailability, stability, trans-biological barrier capacity, and overcome tumor drug resistance, improving the antitumor effect of chemotherapy.

Tumor-derived exosomes are known to carry immunosuppressive signals from tumor cells and transmit them between cells (146, 147). However, they are rich in tumor antigens as they carry tumor cell-related information (134). Therefore, isolating TEXs and injecting them into the organism could help with antigen presentation, which could generate an immune response, and produce enough cytotoxic T cells to kill tumor cells.

Several experiments have shown that the immunostimulatory ability of dendritic cell-derived exosomes (DEXs) is the same as that of the donor cells (148), and they directly or indirectly stimulate cytotoxic T cells via the major histocompatibility complex class I (MHC I) and major histocompatibility complex class II (MHC II) (149). Additionally, DCs are highly immunogenic, and different mature DC-derived exosomes have different functions. Mature, activated DC-derived exosomes often express MHC-I and MHC-II molecules as well as co-stimulatory molecules (CD40, CD80, CD86, etc.) and are therefore able to induce effective antigen-specific T and B cell responses and activate cytotoxic T cells and natural killer (NK) cells. Studies have shown that DC-derived exosomes induce antigen-specific responses against tumor cells by activating innate and adaptive immune cells and overcoming tumor-induced immunosuppression ( Figure 5A ) (150). Based on these characteristics, DEXs have been extensively used in free-cell tumor vaccines (133). Along with dendritic cells, the exosomes derived from macrophages, natural killer cells, B cells, and T cells also possess the immunoregulatory ability and can be used in tumor immunotherapy (151, 152).

The successful delivery of tumor antigens or adjuvants to antigen-presenting cells (APCs) is the most effective method to induce an anti-tumor immune response. Therefore, a novel carrier that can deliver tumor antigens and adjuvants to APCs is required for the development of cancer immunotherapy. Exosomes are natural vesicles produced by cells, which carry cargo molecules stably and safely, can smoothly cross biological barriers, and can target cells accurately. Furthermore, exosomes are also suitable as carriers in immunotherapy because (i) they can simulate the transmission of pathogens to antigen-presenting cells (153), (ii) they can carry a wide range of cargo molecules, such as tumor-specific antigens, peptides, and toll-like receptor (TLR) ligands (154), (iii) the loading method is optional and convenient, the cargos can be loaded directly into the exosomes isolated from the donor cells or indirectly by co-incubation with donor cells ( Figure 5B ) (155).

Based on their carrier and immunomodulatory ability, exosomes play a vital role in immunotherapy against the brain tumor. For example, co-delivery of tumor-derived exosomes with α-galactosylceramide (α-GalCer) on a DC-based vaccine induced strong activation and proliferation of tumor-specific cytotoxic T lymphocytes and destroyed tumor immunosuppressive environment, showing a powerful effect in glioblastoma (132). Additionally, DEXs loaded with chaperone-rich cell lysates (CRCLs) derived from GL261 glioma cells promoted proliferation and activity of CD4⁺ and CD8⁺ T cells, enhanced T cell infiltration in intracranial glioma tissues, and induced the generation of anti-tumor cytokines, including IL-2 and IFN-γ. Therefore, they significantly prolonged the survival of mice with tumors and inhibited tumor growth in vivo (172). A study used natural killer cell-derived exosomes (NK-Exos) to treat glioblastoma and found that NK-Exos exerted an innate therapeutic effect in glioblastoma. It demonstrated that NK-Exos possessed strong antitumor activity toward glioblastoma cells and could facilitate the transfer of other anti-cancer agents through the BBB by wrapping them (138).

In summary, collecting tumor-derived exosomes and reinjecting them into the organism could aid antigen presentation and elicit an immune response. While the dendritic cell-derived exosomes have the same immunostimulatory capacity as donor cells, the tumor-derived exosomes are able to activate innate and adaptive immune cells and overcome tumor-induced immunosuppression, triggering an antigen-presentation response against the tumor cells. Furthermore, in addition to their own biological activity, exosomes can transport tumor antigens or adjuvants to antigen-presenting cells and induce antitumor immune responses.

Currently, the transport efficiency of photosensitizers in PDT is low. Recent studies on the excellent carrier performance of exosomes have attracted the attention of researchers. Studies have shown that exosomes can smoothly carry the photosensitizer to the tumor site. The engineered exosome can be endowed with the ability to target tumor cells; thus, enhancing the therapeutic efficiency of PDT. For example, exosomes collected and purified from the urine of gastric cancer patients have the ability to passively target tumor sites due to natural cell membranes and antigens. Meanwhile, the exosomes are known to promote the internalization of Exo-PMA/Au-BSA@Ce6. Au NPs are transferred into aqueous solution by amphiphilic polymer (PMA) coating, and after surface coupling of the BSA layer, Ce6 is trapped into the BSA network and finally embedded in the exosome nanocarriers toward the cancer cells, avoiding endocytosis by macrophages and prolonging their blood circulation time. The results have shown that the nanoparticles could be released inside cancer cells under 633 nm laser irradiation, producing a large number of singlet oxygen species, killing the tumor cells ( Figure 6A ) (63). However, the short lifespan and limited diffusion length of cytotoxic reactive oxygen species (ROS) generated by the photosensitizer reduces the anti-tumor effects. Therefore, a nucleus targeted exosome was designed to enhance the PDT effect, engineered using a chimeric peptide (ChiP-Exo). In this study, a dual-stage light strategy was used for precise PDT by sequentially destroying the PM and nucleus of the cancer cells. Based on these exosomes, it could activate ROS in situ to disrupt the nuclei, which greatly inhibited tumor growth with minimized systemic toxicity ( Figures 6B, C ) (175). As mention above, tumor-derived exosomes can be used as immunostimulants. Based on this feature, the combination of photodynamic therapy and immunotherapy can be used to treat tumors through exosomes. For instance, recombinant exosomes derived from tumors (R-Exo) are used as drug delivery vehicles and immunostimulants at the same time. During exosomal re-assemble, a chlorin e6 photosensitizer was loaded into tumor-derived exosomes. After this modification, R-Exo retains its original average size and has the same membrane protein, so that it can target tumor cells. The R-Exo loaded with chlorine e6 (Ce6-R-Exo) can be observed by photoacoustic imaging, and it can effectively produce reactive oxygen species inside tumor cells under laser irradiation. In addition, Ce6-R-Exo enhances the ability of immune cells to release cytokines, which indicates that these modified exosomes can also be used as immunotherapeutic ( Figure 6D ) (176)

Compared with traditional therapy, PTT is controllable, noninvasive, and has minimal adverse effects. With the help of exosome delivery, photothermal transducers are carried to tumor cells and produce a thermal effect after NIR irradiation. For instance, an exosome with gold nanoparticles has been designed that assembles into a popcorn-like nanostructure. The formulated nanopopcorn, consisting of self-grown gold nanoparticles and exosomes loaded with DOX, was shown to retain the photothermal conversion of gold nanoparticles and cytotoxicity of DOX. The use of exosomes improved cellular internalization, controlled drug release, enhanced anti-tumor efficacy, and minimized adverse effects ( Figure 7A ) (177). However, it was found that PTT was blocked by heat shock proteins and other stress proteins, resulting in thermal tolerance. Thus, a higher temperature was required to achieve anti-tumor effects. However, the high temperature could damage the surrounding tissue. Therefore, the low-temperature nucleus-targeted PTT in the NIR-II region is conceptualized. For example, vanadium carbide quantum dots (V2C QDs) with good photothermal effect in the NIR-II region were modified with TAT peptides and packaged into exosomes with RGD modification. This carrier exhibited good biocompatibility, long circulation time, and endosomal escape ability. More importantly, it could target the cell and enter into the nucleus to achieve low-temperature PTT with advanced tumor destruction efficiency ( Figure 7B ) (178).

In summary, both PDT and PTT require delivery of the photosensitizer to the treatment site. Exosomes, as carriers of the photosensitizer, can increase the circulation time of the photosensitizer in the blood, and help the photosensitizer cross the blood-brain barrier. Additionally, the modified exosomes could carry photosensitizers to target the nucleus and improve the effectiveness of PDT or PTT.