Compared with the cancer-stimulating risk of animal extracellular vesicles (Tan et al., 2021) and the toxicity of synthetic lipid carriers (Palazzolo et al., 2018), PDEVs are very safe. Because they are natural nanoparticles secreted by plants and already contained in foods currently consumed by humans, they are tolerated by the immune system and have good biocompatibility (Anusha and Priya, 2022). Both animal and cellular experiments have shown them to be non-toxic (Wang Q. et al., 2013; Raimondo et al., 2015; Zhuang et al., 2015; Umezu et al., 2021), and even PDEVS encapsulation can reduce the side effects of DOX to improve efficacy (Zhang et al., 2016c). They do not pass through the placental barrier (Wang Q. et al., 2013) and they show favourable safety profiles in organ pathology and haemolytic reaction as well as levels of cellular inflammatory factors (Dad et al., 2021; Sarvarian et al., 2022).

Plant-derived extracellular vesicles are mostly derived from common plants, with a common source of raw materials that can be produced economically on a large scale. Compared with animal cell-derived EVs, PDEVs not only reduce the cost of cell culture supply but also reduce the time and labour of large scale cell culture. PDEVs reduce the cost/yield ratio of animal cell-derived EVs by approximately 300-fold (Li et al., 2018). This facilitates increased productivity and can be used on a large scale in the clinic.

The lipid membranes of PDEVs protect bioactive substances from external substances, including changes in pH value and heat and light and have extremely high stability (Logozzi et al., 2021). PDEVs are kept in circulation in the body for a long time, which is conducive to the continuous accumulation of drug effects (Yu et al.,; Wang Q. et al., 2013). At the same time, PDEVs are tolerated and remain stable in the gastrointestinal environment. Grape-derived EVS were resistant to degradation of saliva, gastric acid, and proteolytic enzymes and were able to spread through the intestinal tract, migrate through intestinal mucus and eventually be absorbed by mouse intestinal stem cells (Ju et al., 2013). Ginger-derived EVS were highly stable in simulated gastrointestinal fluids (Zhang et al., 2016a). In a study by Wang et al. grapefruit-derived EVS remained stable in both acid-based and simulated gastrointestinal solutions, exhibiting excellent anti-digestive capacity (Wang et al., 2014).

PDEVs, as natural vectors, inherently carry the innate active compounds of the source plant. This makes PDEVs have good biological functions such as antioxidant, anti-inflammatory and anti-cancer. It has been shown that EVs derived from lemon carry micronutrients such as vitamin C and citrate, which exert antioxidant effects in human cells (Akuma et al., 2019). Broccoli-derived EVs, containing sulforaphane, an active component found in some vegetables, contributed in the prevention of colitis in mice (Deng et al., 2017). Ginger-derived nanoparticles were found to carry 6-gingerol and 6-shogaol, two anticancer, anti-inflammatory and antioxidant bioactive compounds, and good hepatoprotective effects (Zhuang et al., 2015; Zhang et al., 2016b). Thus, PDEVs as nano-delivery systems can synergize with loaded drugs to exert therapeutic potential to improve efficacy.

Therefore, based on the above advantages, PDEVs have significant potential in tumour treatment.

The main obstacle to the development of drugs in the central nervous system is the lack of access to the brain. A large number of drugs with the potential to treat brain tumours are not widely used because they cannot cross the blood–brain barrier depending on the therapeutic concentration (Zhuang et al., 2016). Wang et al. recombined lipids derived from grapefruit to form a grapefruit-derived nanovector (GNV). In vivo biological distribution of DiR-labelled GNVs assessed in mouse studies after intranasal administration found that most of these continued to be stably distributed in the brain; furthermore, in vitro experiments showed that GNVs can encapsulate functional siRNAs and efficiently deliver them to mouse glioma GL26 cells. This study also found that intranasal administration of GNVs that encapsulated with the Stat3 inhibitor JSI-124 significantly inhibited Stat3 activation in a mouse model of brain tumours planted with GL26 cells. Delays in tumour growth thereby increased the survival rate of brain tumour mice (Wang Q. et al., 2013).

In further studies, Zhuang et al. (Zhuang et al., 2016) demonstrated that GNVs can carry microRNA (miR)-17 to treat brain tumours in mice. This study shows that the targeting of GNV coated with folic acid (FA) to cells is enhanced and that FA-GNVs can bind to folic acid receptors GL-26 brain tumour cells. In addition, FA-GNV-coated polyethyleneimines (FA-pGNVs) not only enhance the ability to carry RNA but also the toxicity of polyethyleneimines is eliminated by using GNVs. Intranasal administration of brain tumour mice with FA-pGNV/miR-17 can quickly deliver miR-17 to mouse brains and inhibit expression of major histocompatibility complex I (MHCI) on GL-26 cells to trigger activation of natural killer (NK) cells thereby killing tumour cells and playing a role in treating brain tumours in mice. The above studies reported that PDEVs can be used as nanocarriers to help therapeutic drugs enter the brain through the blood–brain barrier, and in combination with other modifiers, can effectively and stably fulfil the role of drugs in treating brain tumours.

Colon cancer is the third most common cancer globally, and recently, the incidence of colon cancer has grown year by year, increasing what is already a threat to human life and health (Sung et al., 2005; Twelves et al., 2005; Yeo et al., 2015; Arnold et al., 2017). Chemotherapy remains the most common treatment option for colon cancer patients (de Gramont et al., 2012). However, toxic sie effects of chemotherapeutic drugs constrain their efficacy. Therefore, a nano-delivery system that can both maintain (or improve) the effect of drug treatment and reduce associated toxicity has broad prospects. Zhang et al. (Zhang et al., 2016c) reported an advance in treating colon cancer using a ginger-derived nano-GDLV modified with FA modifications of the therapeutic agent doxorubicin (DOX) to successfully inhibit tumour growth in colon-26 xenograft tumour models. This study shows that FA-GDLV has an increased ability to target colon-26 tumours compared to unmodified GDLV, possibly through active FA-FUR interactions. However, the accumulation of FA-GDLV in the spleen and liver was found to be significantly less than that with GDLV, suggesting that using FA-GDLV as a carrier can reduce the systemic toxicity of the drug to normal tissues. In vivo studies reported that FA-GDLV is detectable after 48 h of circulation after intravenous injection, giving them a greater chance of penetrating the tumour. Therefore, FA-GDLV can be an ideal drug delivery platform that can exert anti-tumour effects while avoiding adverse reactions of free-circulating drugs.

Li et al. (Li et al., 2022) performed in vitro anticancer studies using broccoli-derived extracellular vesicle-coated astaxanthin nanoparticles that had a more significant inhibitory effect on the proliferation of human colon cancer HT-29 cells than that of astaxanthin alone. Other studies reported that GNVs, after modification, can effectively deliver chemotherapy drugs to colon cancer cells and significantly inhibit tumour growth in colon cancer model mice (Wang Q. et al., 2013; Wang et al., 2015).

PDEVs have been introduced into clinical trials. Researchers from the James Graham Brown Cancer Centre used plant-derived nano-vesicle-coated curcumin to treat intestines in cancer patients. Curcumin has a powerful inhibitory effect on colon cancer cell lines; however, because of the low solubility, poor stability and simple metabolism, oral curcumin exhibits only limited bioavailability even at very high doses. By using PDEV delivery, most of the major obstacles to curcumin application can be addressed, including increased solubility, bioavailability and stability. This remains ongoing work, so further clarification of the conclusions is needed (NCT01294072).

Breast cancer as a malignancy seriously threatens the health of women globally (Yu et al., 2021). In particular, once metastasis occurs in the later stages of breast cancer, a poor prognosis can mean limited life expectancy for the patient (Chaffer and Weinberg, 2011). Currently, there are still several defects in commonly used chemotherapy and photothermal therapy, such as tumour cell resistance, insufficient accumulation of drug properties and non-specific distribution and toxicity to normal organs, which seriously limit their application (Minko et al., 2013; Liu et al., 2019; Chen et al., 2021).

Zeng et al. (Zeng et al., 2022) argue that to overcome the drawbacks of traditional treatment methods, a functional nanocarrier loaded with mixed anticancer drugs has favourable prospects in breast cancer treatment. They found that aloe vera gel-derived nanovesicles (gADNV) is an example of an excellent nanocarrier in this regard. gADNV is modified with an active integrin-targeted peptide (Arg–Gly–Asp, RGD) to effectively deliver the photothermal drug indocyanine green (ICG) and DOX for breast cancer treatment. The ‘competitive’ relationship between ICG and DOX can be transformed into a ‘cooperative’ relationship by π–π stacking interactions within gADNVs to improve their loading efficiency. This dual-drug co-delivery nanosystem called DIARs has good stability, leak resistance and exhibits high breast tumour targeting capabilities both in vitro and in vivo. Simultaneously, this nanosystem significantly inhibits cell growth and migration and induces apoptosis through a combination of phototherapy and chemotherapy. Intravenous administration of DIARs showed high therapeutic efficacy in 4T1 tumour-bearing mouse models and no significant damage to other organs. Wang et al. (Wang et al., 2015) confirmed that the DOX and curcumin can be successfully delivered to the desired site of inflammation through GNVs and achieved therapeutic effects by modifying GNVs with individually activated leukocyte membranes (IGNV). In vivo experiments also proved that intravenous injection of IGNV-DOX or IGNV-Cur can significantly inhibit the growth of breast tumours in mice. These studies showed that PDEVs can be successfully combined with various modifications to improve the targeting and loading efficiency of breast cancer treatment drugs and reduce their side effects to give full play to their therapeutic roles, providing an effective approach in treating breast tumours.

Oral squamous cell carcinoma (OSCC) is the most common oral malignancy characterised by rapid local invasion and a high recurrence rate (Chi et al., 2015). Patients with recurrent OSCC have been reported to have failed treatment due to resistance to chemotherapy drugs, resulting in low survival rates (Wang B. et al., 2013; Wang et al., 2016). 5-Fluorouracil (5-FU) is one of the most used chemotherapy drugs for treating OSCC, although the anti-tumour efficacy is affected by side effects and tumour resistance (Longley et al., 2003; Atashi et al., 2021). Meng et al. (Yang et al., 2021) found that bitter melon-derived extracellular vesicles (BMEVs) are an intrinsically biologically active form of cancer treatment and potential nano-delivery carriers. BMEVs combined with 5-FU can enhance OSCC apoptosis by increasing the production of reactive oxygen species while BMEVs can enhance the cytotoxic effect of 5-FU during OSCC treatment and reduce resistance by downregulating the expression of NLRP3.

Another study reports that ginger-derived exosome-like nanovesicles (GDENs) can be engineered to inhibit tumours via intravenous delivery of siRNA by exhibiting ligands. The researchers mixed GDENs with arrow tail RNA nanoparticles and modified these with FA and a ligand of arrow tail RNA to prepare a GDEN carrying siRNA that targets the BIRC5 gene. In vitro experiments demonstrated that GDEN modified to carry siRNA can effectively bind to human oral epithelial cancer kB cells and effectively knockdown target gene expression. The results of in vivo experiments showed that mice with oral epithelial cell carcinoma xenotransplanted from humans could be injected intravenously with GDEN loaded with siRNA, which can significantly inhibit tumour growth in mice (Li et al., 2018). This, PDEVs have the potential to treat nano transmission systems for oral cancer.

Zeng et al. identified a nanocarrier made of aloe vera that is stable and leak-proof. Nanovesicles were isolated from the gel and outer skin of aloe vera (gADNVs and rADNVs) with high quality and yield. gADNVs exhibits good structural and storage stability, oxidation resistance and impermeability. They can be efficiently absorbed by melanoma cells and are not toxic in vitro or in vivo. Indocyanine green (ICG) loaded in gADNVs (ICG/gADNVs) showed high stability in both a heated system and in serum, and the retention rate exceeded 90% after 30°days stored in gADNVs. This high degree of retention for ICG/gADNVs can enable them effectively destroy melanoma cells and inhibit the growth of melanoma and is superior to treatment using free ICG and ICG liposomes. Interestingly, ADNVs exhibited significant penetration of mouse skin, which may facilitate non-invasive transdermal administration (Zeng et al., 2021).

Liver metastases are the cause of cancer death in most patients (de Ridder et al., 2016; Milette et al., 2017). Effective treatment of metastatic liver tumours is not currently available (Alberts and Wagman, 2008; Sabanathan et al., 2016). Teng et al. (Teng et al., 2016) found that miR-18a wrapped in GNVs mediates inhibition of liver metastasis, depending on the induction of M1 (F4/80 + IFNγ + IL-12+) macrophages. The depletion of macrophages eliminates the anti-metastatic effect. In addition, IFNγ induction of macrophages mediated by targeting IRF2-mediated miR-18a is required for the subsequent induction of IL-12. IL-12 then activates NK and NKT cells to inhibit liver metastasis in colon cancer. Knockout of IFNγ expression eliminates miR-18a-mediated IL-12 induction, and miR-18a therapy has an anti-metastatic effect in T cell-deficient mice, but no anti-metastatic effect on NK and NKT-deficient mice. Co-delivery of miR-18a and siRNA IL-12 to macrophages does not lead to activation of co-cultured NK and NKT cells.