Cancer in the head and neck, primarily squamous cell carcinoma, develops in the oral cavity, nasopharynx, oropharynx, hypopharynx, and larynx ^[1][37]. Head and neck squamous cell carcinoma (HNSCC) is one of the most common causes of cancer worldwide, displaying high levels of immune suppression and poor survival rates in patients diagnosed with late-stage tumors ^[2][3][38,39]. Limited progress has been made in the last few decades regarding survival outcomes, and this can be most attributed to the difficulty in detecting these cancers without imaging modalities. Since HNSCC originates deep within the head and neck region, they are not easily palpable or noticeable by patients ^[1][37]. Thus, the discovery of reliable and non-invasive biomarkers is necessary to detect these cancers earlier which can lead to drastically improved patient outcomes. EVs could serve as a potential biomarker of early-stage HNSCC tumors which could easily be collected non-invasively by blood collection. Given the potential, it is no surprise many studies have been published on the role of EVs in the metastasis, lymphocyte regulation, angiogenesis, microenvironment remodeling, and drug resistance in HNSCC.

Local lymph nodes and distant metastasis is a common feature of HNC which ultimately contributes to the disease’s poor outcomes. One study investigated the role of EVs in the metastasis of nasopharyngeal cancer (NPC) cells and found that MMP13 is highly expressed in NPC cells, EVs released by NPC cells, and in the plasma of patients with NPC cells ^[4][40]. Matrix metalloproteins (MMPs) are critical enzymes in the remodeling of the extracellular matrix and lead to reduced cell adhesion and promoted invasion and metastasis ^[5][41]. An additional study found that plasma EVs containing overexpressed MMP13 may have a role in the mediation of tumor migration and invasion by upregulating Vimentin and HIF-1α expression, and reducing E-cadherin expression ^[6][42]. Latent membrane protein 1 (LMP1) is an oncoprotein that is encoded by the Epstein–Barr virus (EBV) and is a known driver of NPC ^[1][37]. It has been reported that LMP1 can promote its own packaging into CD63+/HSP70+ EVs, which is likely an important mechanism for LMP1 to engage in intracellular signal exchange and promote tumor growth ^[7][8][43,44]. Furthermore, CD63+/CD81+/HSP70+ EVs containing LMP1 derived, in vitro, from NPC cells have been reported to cause radio-resistance to recipient NPC cells ^[9][45]. The relationship between LMP1 and EVs is quite complex, but it can be concluded from the findings that LMP1-rich EVs possess pro-tumorigenic abilities by inducing cell proliferation, invasion, and radio-resistance in NPC ^[1][37]. An additional study found that EVs isolated from the saliva of patients with oral cavity squamous cell carcinoma and tongue cancer highly expressed miR-24-3p. This study continued to delineate that these EVs enriched with miR-24-3p stimulated non-specific cell proliferation, likely contributing to disease progression ^[10][46]. One study analyzed the proteomic profile of EVs in the same cancer cell line (HSC-3) and found the overexpression of multiple oncogenic proteins, specifically EpCAM, EGFR, and HSP90 ^[11][12][47,48]. These EV-packaged oncogenes certainly contribute to the promotion of tumor growth and metastasis in the setting of oral cavity squamous cell carcinoma and tongue cancer.

HNCs are characterized as “hot tumors”, meaning they show immense infiltration of lymphocytes, macrophages, and additional active immune cells ^[13][49]. However, the immunosuppressive nature of HNCs prevents a coordinated immune response in these tumors. In vitro, it has been reported that HNC cell-derived EVs contribute to the reprogramming of immune cells. In the plasma of patients with active HNC, TSG101+ EVs induce significantly more apoptosis of CD8+ T cells, more inhibition of T cell proliferation, and increased upregulation of innate suppressive functions of CD4 + CD39+ T regulatory cells when compared with EVs of healthy patients ^[14][50]. Furthermore, an additional in vitro study found that EVs derived from oral cavity cells (Tu167) and additional HNC cells (SCC0209 and HN60) induced an immunosuppressive phenotype in CD8+ T cells ^[15][51]. Another in vitro study by Li et al. found that EVs released by hypoxic oral squamous cell carcinoma cells induce a pro-metastatic phenotype in surrounding cells via delivery of EV-miR-21 ^[16][52]. A subsequent report by the same research team found that delivery of miR-21 by CD61+/CD81+ EVs also downregulated PTEN, increased PD-L1 expression, and induced immunosuppressive activity of myeloid-derived stem cells ^[17][53]. The multiple functions of EV-miR-21 in inducing an immunosuppressive environment in HNC may make it a strong target of therapeutic interest. HPV16 infection is strongly associated with tumors of the oropharynx. Interestingly, studies have found that EVs play a differential role on immune cells depending on whether the tumor is HPV positive or HPV negative ^[18][54]. More specifically, they found that HSP70+/TSG101+ EVs from HPV-negative oral cancer (PCI-13 and PCI-30) cells reduced dendritic cell maturation while EVs from HPV-positive oral cancer (UMSCC2, UMSCC47, SCC90) cells upregulated maturation of dendritic cells, which contributes to improved outcomes of HPV-positive HNC patients ^[18][54]. In agreement, Ludwig et al. found that HPV-positive EVs were enriched with CD47 and CD247, two immune cell-related effectors, while HPV-negative EVs were enriched with MUC-1 and HLA-DA, two tumor-protective and growth-promoting antigens ^[14][50]. In the setting of immune checkpoint inhibitors, one recent study found that PD-L1+ plasma EVs reduced CD69 expression on activated T cells in HNSCC patients, offering another immunosuppressive mechanism ^[19][55]. This report further found PD-L1+ EVs in plasma of HNSCC patients was directly correlated with evidence of advanced disease, high tumor stage, and local lymph node involvement ^[19][55].

Multiple studies have reported the role of EV uptake in inducing upregulation of angiogenesis-related molecules resulting in the promotion of endothelial cell proliferation and migration ^[20][21][22][56,57,58]. One recent study uncovered that miR-142-3p selectively packaged into small EVs (sEVs) plays a role in enhanced angiogenesis and vascular density, both in vivo and in vitro ^[23][59]. In HNCs, multiple EV-miRNAs have been reported to play a role in regulating angiogenesis, specifically miR-23a, miR-17-5p, and miR-9 ^[24][25][26][60,61,62]. Selectively packaged proteins in EVs have also been shown to play a role in angiogenesis. One study found that HNC (OSC19, SCC61, and Detroit 562) cell-derived sEVs influence reverse signaling of Ephrin-B in endothelial cells, which promotes angiogenesis ^[27][63]. Additional studies have uncovered that CD61+/CD9+/Flotillin-1+ EVs from NPC (CNE1, CNE2, 6-18B, 5-8F) cells are enriched with 6-Phosphofructo 2-kinase/fructose 2, 6-bisphosphatase 3, a critical glycolysis-regulatory enzyme which is known to promote angiogenesis and various other pro-tumor functions ^[28][64]. Furthermore, another recent study reported that CD9+/TSG1-1+ EVs derived from HPV-positive oral cancer (UMSCC47) also promoted angiogenesis by a direct endothelial interaction and indirectly by upregulating adenosine and the A2B receptor pathway in macrophages ^[29][65]. These findings all suggest various mechanisms of EV-mediated angiogenesis in HNC.

The tumor microenvironment is made up of a wide range of cells and proteins including tumor-associated macrophages (TAMs), immune cells, vascular cells, cancer-associated fibroblasts (CAFs), extracellular matrix, and more ^[30][66]. All of these components making up the tumor microenvironment contribute to tumorigenesis ^[30][66]. CAFs are abundantly found in the tumor microenvironment and are correlated with poor outcomes in HNC ^[31][67]. One study looking into CAFs found that NPC EVs enriched with LMP1 played a role in differentiating fibroblasts into CAFs ^[32][68]. This is a novel example of LMP1+ EVs inducing autophagy and metabolic switching of fibroblasts into CAFs, ultimately promoting proliferation and migration of NPC cells ^[32][68]. Furthermore, another study found that CD63+/CD81+/TSG101+ EVs generated by adenoid cystic carcinoma (SACC-81) cells were internalized by periodontal ligament fibroblasts and human umbilical vein endothelial cells (HUVECs), driving a malignant phenotype in both recipient cells ^[33][34][69,70]. High concentrations of TAMs in the tumor microenvironment, both pro-inflammatory M1 and anti-inflammatory M2, have been associated with poor outcomes in HNCs ^[35][71]. One study found that EVs from oral tongue cancers (CAL-27 and SCC9) promoted M2 macrophage polarization, influencing a pro-tumoral environment ^[36][72]. An additional study found that EV-miR-21 generated by transformed hypopharyngeal cells (FaDu) also induced M2 macrophage polarization via suppressing PDCD4 and IL12A ^[37][73]. An investigation into the role of EVs generated by HPV+ cancer cells found that EV-driven M1 macrophage polarization may play a role in the improved outcomes seen in HPV+ HNCs. Specifically, they found that CD9+/CD63+/TSG101+ EVs from HPV-positive cells (SCC47, SCC90, SCC104) stimulated M1 macrophage polarization while EVs from HPV-negative cells (SAS, CAL-27, CAL-33) stimulated M2 macrophage polarization ^[38][74]. These promising findings require further investigation and it is likely that multiple non-EV-related mechanisms also influence favorable outcomes in HPV+ HNCs. Finally, another recent report found that Alix+/TSG101+ EVs released by naive macrophages promoted migration of laryngeal cancer cells (BICR18) and induced PD-L1 expression, ultimately creating an immunosuppressive tumor microenvironment ^[39][75]. The tumor microenvironment is a complex environment that likely includes many EV- and non-EV-related regulatory mechanisms, all of which are important to further investigate.

A primary cause of treatment failure in patients with HNCs is drug resistance. Various mechanisms can influence drug resistance, including epigenetic modifications, DNA damage repair, drug efflux, cell death inhibition, drug inactivation, and more ^[40][76]. Two studies delineated a potential role of EVs in acquired HNC drug resistance, specifically finding that inhibition of EV secretion led to a significant increase in cisplatin concentration in cisplatin-resistant oral cancer cells (H314, HSC-3-R, SCC-9-R) ^[41][42][77,78]. One proposed mechanism was potential EV-miR-21 targeting of programmed cell death factor 4 (PDCD4) and phosphatase tensin homolog (PTEN) ^[42][78]. Furthermore, a more recent study found that EV-miR-30a may facilitate cisplatin resistance by the upregulation of Beclin1, a known autophagy-related gene ^[43][79]. A separate study uncovered an additional factor related to cisplatin resistance, specifically the transmission of EV-miR-155 ^[44][80]. The importance of further characterization of these EVs related to cisplatin resistance must be noted; however, early studies suggest they are a critical factor in resistance.

The roles of EVs on the metastasis and proliferation, lymphocyte regulation, angiogenesis, microenvironment remodeling, and drug resistance in HNCs are summarized in Table 1.

In general, acute otitis media (OM) is a dysfunction of the eustachian tube resulting from an upper respiratory infection ^[45][81]. Diagnosis is made based on clinical examination findings, specifically physical evidence of middle ear inflammation, presence of middle ear effusion, and symptoms which include pain, irritability, and fever ^[45][81]. Chronic otitis media (COM) is a result of multiple recurrent infections of the middle ear and is often characterized by consistent middle ear effusion (MEE), which is most commonly mucoid ^[46][47][82,83]. This MEE is mostly made up of innate immune mediators from neutrophils, specifically neutrophil extracellular traps (NETs) and secretory mucin glycoproteins ^[48][49][84,85]. In vitro, OM models have shown that disease-related bacterial products drive differential expression of miRNAs within human middle ear epithelial cells (HMEECs) ^[50][86]. One study in 2017 analyzed the exosomal miRNA expression profile in middle ear effusions collected from patients with COM ^[51][87]. The research team found that exosomes collected from COM patients had a high abundance of miRNAs, most significantly being miR-223-3p, miR-451a, miR-16a-2p, miR-320e, and miR-25-3p ^[51][87]. Pathway analysis predicted these miRNAs to regulate 442 target genes; most notable were genes that upregulated many IL-8-mediated cellular functions, CXCR1/2-mediated signaling which is associated with antimicrobial defense, chemoattraction of monocytes/lymphocytes, and NF-kB pro-inflammatory upregulation resulting in angiogenesis and inflammation ^[51][87]. This was the first report of exosomal miRNAs in MEEs.

Chronic rhinosinusitis (CRS) is a persistent inflammatory disease of the paranasal sinus mucosa with multifactorial etiology, including genetic, environmental, bacterial, and immunological contributions ^[52][88]. Traditionally, CRS was classified phenotypically as occurring with (CRSwNP) or without nasal polyps, which failed to account for the range of mechanisms which can cause the disease ^[53][89]. The European Position Paper on Rhinosinusitis and Nasal Polyps 2020 advises against phenotypic classification and instead focuses on the pathophysiology of CRS to classify ^[54][90]. Given the simplicity in obtaining a clinical sample from CRS patients via nasal fluids, EVs pose a potentially useful diagnostic tool in categorizing CRS and quickly determining the most effective treatment option. Few studies have tested the cargo profile of EVs isolated from nasal fluids of CRS patients. Proteomic analysis of these samples was found in one study to show 123 differentially expressed proteins, which played a role in over 40 dysregulated signaling pathways ^[55][91]. Furthermore, significant differences in EV-cargo protein profiles were found between CRSwNP and control individuals, specifically molecular markers of CRSwNP including cystatin, glycoprotein VI, and peroxiredoxin-5 ^[55][91]. An additional report found elevated levels of cystatin-1 and -2, both epithelial protease inhibitors, in EVs isolated from nasal fluids of CRS patients ^[56][92]. These findings suggest both cystatin-1 and -2 may serve as markers of CRS with the ability to potentially predict disease phenotype as well. Another study found EVs from CRS patients to potentially lead to the formation of polyps via their role in upregulating pappalysin and serpins ^[57][93]. These findings provide another potential biomarker for diagnosis and future target for treatment of CRSwNP.

Acquired cholesteatoma is a chronic inflammatory disease that involves an overgrowth of hyperkeratinized squamous epithelium and erosion of the bone in the middle ear ^[58][94]. Nearly all cases of acquired cholesteatoma are a result of chronic infections, with 85% of cases being bacterial infections, most often from Pseudomonas aeruginosa ^[59][95]. Acquired cholesteatoma ultimately causes issues due to the erosion of body structures in the middle ear, which results in breakdown of the ossicular chain and otic capsule and subsequent hearing loss, facial paralysis, vestibular dysfunction, and intracranial complications ^[58][94]. Surgical intervention by tympanomastoid surgery to remove the lesion is the only effective treatment; however, bone loss and recurrence are unavoidable, and 70% of patients require follow-up surgery within 10 years ^[60][96]. Unfortunately, repeat surgeries for acquired cholesteatoma oftentimes results in further hearing loss ^[61][97]. Inadequate treatment options highlight the importance of further understanding acquired cholesteatoma and discovery of novel targets to improve patient outcomes. One study focusing on the role of exosomes in this disease found that exosomal miR-17 of keratinocyte origin led to the upregulation of fibroblast protein expression ^[62][98]. This upregulation promotes the differentiation of osteoclasts leading to bone destruction, a hallmark of acquired cholesteatoma ^[62][98]. Furthermore, studies into the pathogenesis of this disease are required; however, this report offers a novel therapeutic target and further studies should seek to identify additional exosomal miRNAs involved in disease processes.

Hearing loss, which is most often the result of cell death of the sensory hair cells of the inner ear, is a major problem across the world, affecting nearly 6.1% of the global population ^[63][99]. Sensory hair cell death can result from a variety of stressors, including noise trauma, aging, and treatment with platinum-based cancer therapeutics or aminoglycoside antibiotics, referred to as ototoxic drugs ^[63][99]. Permanent hearing loss due to ototoxic drugs affects nearly 500,000 people every year in the United States alone ^[64][100]. Uncovering the mechanisms underlying hearing loss induced by ototoxic drugs is critical to developing therapies that can avoid this therapeutic consequence. The upregulation of heat shock proteins (HSPs), specifically Hsp70, has been shown to protect sensory hair cells from aminoglycoside-induced cell death ^[65][101]. One report found that in response to heat stress, cells of the inner ear release exosomes carrying Hsp70 which then interacts with TLR4 on the hair cells ^[66][102]. These isolated exosomes provided a protective effect, thereby improving survival of hair cells treated with aminoglycoside antibiotics ^[66][102]. Further proving this mechanism, the report also found that the protective benefit was removed when these exosomes carrying Hsp70 were depleted ^[66][102]. This mechanism clearly highlights a potential therapeutic use of EVs in the setting of aminoglycoside antibiotic-induced deafness. Another report analyzed the proteomic profile of EVs isolated from mice treated with cisplatin, a platinum-based therapeutic ^[63][99]. Compared with the control, inner ear EVs from the cisplatin-treated group were found to be reduced in number and have significantly lower cargo protein concentration ^[63][99]. However, proteomic analysis showed a significant increase in protein expression of Tmem 33, Pgm1, Eif3f, Rps24, Cct8, Hsd17b4, Aldh3a1, Ddost, Aldh3a1, Eif3c, Luc7l2, and Acadvl—proteins with known roles in hearing loss ^[63][99]. Using an ischemia and reperfusion model in C57BL/6 mice, a study by Hao et al. discovered that delivery of NPC-EVs transfected with miR-21 to these mice reduced caspase-3/parvalbumin expression, increased IL-10 expression, and prevented an increase in TNF-α and IL-1β expression of cochlear hair cells ^[67][103]. This suggests that delivery of EV-miR-21 could serve to improve outcomes of cochlear damage related to ischemia ^[67][103]. Another in vivo report found that, compared with normoxic bone marrow mesenchymal stem cells (BMSCs), hypoxic BMSCs reduced cisplatin-induced ototoxicity by upregulating HIF-1α, superoxide dismutase 1 (SOD1), and SOD2 expression ^[68][104]. These findings suggest hypoxic preconditioning may offer protective effects to cisplatin-induced ototoxicity ^[68][104]. An additional in vivo report studied the effect of delivery of exosomes produced by BMSCs pretreated with heat shock (HS-BMSC-Exos) which were delivered to cisplatin-injected C57BL/6 mice ^[69][105]. This report found that trans-tympanic delivery of these HS-BMSC-Exos to cisplatin-dosed mice, reversed the upregulation of IL-1β, IL-6, TNF-α, NLRP3, ASC, cleaved caspase-1, and pro-caspase-1, thereby resulting in improved auditory sensitivity and reduced inner ear hair cell death ^[69][105]. Lastly, an in vitro investigation of gentamicin-induced ototoxicity found that delivery of Exo-miR-182-5p from mouse inner ear cells (IECs) to gentamicin-treated HEI-OC1 cells resulted in increased Bcl-2 expression and decreased FOXO3 and Bax expression ^[70][106]. These findings offer a protective mechanism of IEC-Exos in gentamicin-induced HEI-OC1 cell death ^[70][106].

The roles of EVs in Otitis Media, Chronic Rhinosinusitis, Acquired Cholesteatoma, and Ototoxicity are summarized in Table 2.