Gene therapy. Many groups are investigating the utility of gene therapy in the treatment of RIX. These studies focus on increasing expression of water channels in salivary cells to facilitate fluid transport and hopefully increase saliva production. The water channel gene AQP1 (aquaporin 1) encodes a constitutively active water channel that facilitates fluid secretion along an osmotic gradient. AQP1 is expressed in the myoepithelial and endothelial cells of humans and mouse salivary glands and is limited to endothelial cells in rat submandibular gland (SMG) (120). This section addresses both in vitro and in vivo studies that aim to increase expression of AQP1.

An in vitro study explored artificial induction of AQP1 expression in established salivary gland cell lines and primary human salivary progenitor cells via two different methods: artificial transcriptional complexes and epigenetic alterations. Wang et al. introduced an artificial transcriptional complex at the AQP1 gene and saw increased expression of AQP1 in cell lines after delivery of guide RNAs targeting the promoter region (121). This group also explored the impact of epigenetic modification on AQP1 expression, in which they performed chemical demethylation of A253 cells (derived from human SMG tumor) and saw increased expression of AQP1 through demethylation alone (121). These data support other literature that suggested methylation was the primary method for AQP1 gene silencing (122). This in vitro study suggests that epigenetic editing can hold potential for inducing AQP1 expression in salivary cells.

The adenoviral delivery of human AQP1 (AdhAQP1) utilizes an adenovirus-derived vector to deliver the AQP1 gene into infected salivary cells. Many groups use this method to investigate AQP1 gene delivery in in vitro and in vivo models. One study observed a two- to three-fold increase in salivary fluid secretion compared with control animals after AdAQP1 delivery to rat SMG 3–4 months after radiation (17.5 or 21 Gy in a single fraction) (123). Another group delivered AdhAQP1 to the minipig parotid gland and showed improved saliva secretory volume, but not changes in salivary composition, within 8 weeks after administration (124). The minipig is a highly translational model animal, useful for evaluating novel therapies to prevent or reverse RIX in humans. AdhAQP1 has shown clinical promise in a phase I clinical trial in patients with RIX (ClinicalTrials.gov NCT00372320). Researchers evaluating late responses to the therapy found that AdhAQP1 resulted in both short- and long-term improvement of parotid salivary flow and sustained symptomatic relief for 2–3 years (125). This approach is being further developed in a recently completed phase I study (ClinicalTrials.gov NCT02446249) and an ongoing randomized double-blind placebo-controlled study (ClinicalTrials.gov NCT05926765).

Stem cell therapy. There is substantial interest in modulating salivary gland stem cells to improve salivary function. Studies in mice and humans have demonstrated that there are regions with high stem cell density within the parotid gland, and damage to these regions is a good predictor of salivary dysfunction after radiation (126). A double-blind randomized controlled trial was performed to determine the impact of dose reduction in these regions on salivary flow from the parotid gland. Salivary flow was reduced 16.8% and 8.5%, and patient-reported xerostomia was 50.0% and 45.9% in the standard IMRT versus high stem cell density–sparing IMRT groups, respectively. Unfortunately, in this trial, salivary flow and perceived xerostomia were not significantly improved (127).

Salivary stem cell populations are only beginning to be defined in recent years, as prior to this there was not a consensus definition of what constitutes the salivary stem cell population. Researchers identified a population of SOX2+ adult human salivary gland progenitor cells in the three major salivary glands that could potentially differentiate into acinar cells (128). Emmerson et al. demonstrated that SOX2 was essential for salivary gland regeneration following a single dose of 10 Gy to the murine sublingual gland (128). Using an ex vivo model, SOX2+ cells could repopulate the irradiated murine sublingual gland. It is possible that the presence of senescent cells, as a consequence of radiation, can enhance the self-renewal potential of the remaining, nonsenescent salivary gland stem cells, like these SOX2+ progenitor cells (126). Several groups have also been utilizing induced pluripotent stem cells to establish salivary tissue for in vitro and in vivo modeling (129–131). These models appear to recapitulate the stem cell populations seen in the developing gland and may serve as important models for future translational research.

There are two populations of stem cells identified in the ductal regions of the gland that have demonstrated the ability to regenerate ductal tissue after radiation exposure; these populations are marked by cytokeratin 14 (KRT14) and Kit, respectively (132, 133). Both KRT14+ and Kit+ cells demonstrated the ability to regenerate salivary tissue through distinct mechanisms. KRT14+ cells are fast-cycling cells that maintain a K14+ cell population in granulated ducts under homeostasis and also expand in response to radiation or severe injury and divide to produce cells in the larger granulated ducts, Aqp5+ acinar cells, and Kit+ intercalated duct cells (132–135). KRT14+ cells are now considered to be a bona fide salivary stem cell marker, as researchers have demonstrated the capacity of KRT14+ cells to replenish injured glands on many occasions; however, most studies show that KRT14+ cells replenish cells in the ductal compartment and do not contribute to acinar regeneration (132, 134–136).

Early work in identifying these salivary stem cell types used 3D salispheres generated from murine SMGs to study the expansion of potential salivary gland stem cells. Lombaert et al. generated salispheres from murine SMG tissue, and within these populations they identified cells expressing the stem cell markers Sca-1, Kit, and Musashi-1 (137). They demonstrated histologically that the isolated spheres initially expressed mostly ductal marker and were eventually capable of producing mucins and amylase. These data indicate that the spheres they isolated were of ductal origin, and with time, differentiated into cells with an acinar phenotype. After characterization, the spheroids were transplanted into irradiated murine salivary glands and the group saw improvement in glandular structure and increased proliferation compared with controls. They used fluorescence-activatedcell sorting (FACS) to enrich for Kit+ stem cells to further characterize the salivary stem cells present in the spheroid cultures. Kit+ cells were then transplanted into the salivary glands of irradiated female mice, while controls received Kit– populations. After 90 days, transplantation of Kit+ cells resulted in glands showing a similar morphology to nonirradiated glands, including restored acinar cell populations and increased saliva production in 69% of animals (137). The Kit– transplantation only resulted in minor responses. This led researchers to investigate whether this Kit+ population of spheroids are responsible for acinar regeneration in damaged tissues. Nanduri et al. isolated Kit+ spheroids from mice and injected them into irradiated murine salivary glands, and this resulted in improved gland architecture and improved saliva production compared with controls. They continued to study these Kit+ spheres and the impact of spheres coexpressing other salivary stem cell markers like CD24 and CD49f on irradiated salivary tissues, with positive results (138, 139). This approach has also been used to isolate human Kit+ cells from salivary tissue and implant them in murine salivary glands; results demonstrated restoration of salivary gland function after irradiation in a xenotransplantation approach (140).

However, the identification of a Kit+ stem cell in the salivary gland has been disputed by other studies, including those from the Ghazizadeh group (135, 136). While this group also identified distinct Kit+ and KRT14+ cells in the salivary gland, they used lineage tracing studies to show that the KRT14+ population is the major source of regeneration (132, 135). A different study by Nanduri et al. also discovered that whether or not a spheroid is Kit+ is not critical to their effectiveness at regenerating tissue. They identified a population of CD24hiCD29hi spheroids that produced the best regenerative response to radiation regardless of Kit positivity, indicating this Kit+ cell population does not further enrich for stem cells (141). Kwak et al. concluded that Kit is not a reliable marker for salivary stem cells, and suggest that a better marker when considering clinical implications is KRT14 (136). Ninche et al. also identified KRT14+ cells as a reliable salivary stem cell marker (135). Interestingly, in their study, they demonstrated that acinar regeneration relies on methods independent of KRT14+ ductal stem cells. Their group found that in severe injury (ligation) models, KRT14+ cells contribute to generation of granular ductal cells, Kit+ intercalated duct cells, and Aqp5+ acinar cells. Using lineage tracing experiments, they showed that most of the acinar replenishment in the severe injury model is contributed by dedifferentiated KRT14+SMA+ myoepithelial cells. Upon injury, the myoepithelial cells transdifferentiate into a bipotent progenitor state capable of producing acinar cells and Kit+ intercalated duct cells. They also found that Kit+ intercalated duct cells are capable of transdifferentiation into a bipotent progenitor cell type as well, capable of producing both acinar cells and more Kit+ intercalated duct cells. These data indicate that Kit+ cells in the intercalated duct can act as a reservoir of acinar progenitors in the case of injury, but are not the main contributor (135). Despite the confounding opinions of what is and is not a salivary stem cell, these data suggest that the salivary gland maintains mechanisms capable of regeneration. Use of salivary organoids, and specifically stem cell–derived three-dimensional models, is being studied in an ongoing clinical trial, as an autologous source of transplantable material by a group in the Netherlands (140, 142).

We, and others, have utilized mesenchymal stromal cells (MSCs) to improve salivary function after radiation (143–146). MSCs are most commonly isolated from adipose tissue or bone marrow. When injected into the irradiated salivary glands of mice, they improve salivary function (147–149). This approach is being tested in ongoing clinical studies run by several groups (ClinicalTrials.gov NCT04489732, NCT04776538, NCT03876197, NCT03874572, and NCT03743155). The MESRIX study conducted by Rigshospitalet in Denmark looked at the effects of injecting autologous adipose-derived MSCs, or MSC(A), into the SMGs of patients. In this blinded randomized controlled trial, HNC patients with xerostomia had MSC(A) isolated, expanded, and injected into the submandibular gland. The delivery of MSC(A) was shown to be safe and demonstrated a significant increase in salivary flow at one and four months. Compared with baseline, symptoms of xerostomia were significantly reduced, and there was increased serous tissue in the glands (143, 145, 150). The team conducting the MESRIX study then completed a trial investigating the safety of allogeneic MSC(A) (151), which led to their opening a randomized phase II study (MESRIX-III, ClinicalTrials.gov NCT04776538) in order to evaluate the safety and efficacy of injection of MSC(A) from healthy donors (allogeneic) in patients with HNC suffering from xerostomia (152). A multidisciplinary team at the University of Wisconsin recently published a pilot study demonstrating the safety of autologous marrow–derived MSCs, or MSC(M), in patients who were two or more years out from the completion of radiation or chemoradiation (146). They recently opened a phase I study (ClinicalTrials.gov NCT05820711) utilizing autologous MSC(M) to define a phase II dose of cells (153). The optimal source of MSCs (marrow, adipose, or other; allogeneic or autologous) remains to be defined.

An adjunct to injecting stem cell populations into damaged glands is to introduce biomaterials to support regeneration of damaged glands. Several labs are testing novel biomaterial approaches to engineer implantable tissue that exhibit the regenerative potential of isolated stem or progenitor cells (154). Others hope to regenerate salivary glands by using primary murine SMG cells to build cell sheets to repair damaged regions of the tissue (155). These stem cell–based therapies show preclinical and clinical promise and may represent the next major step in the treatment or prevention of RIX.

Pharmacological intervention. In addition to gene therapy and stem cell therapy methods, many researchers are exploring pharmacological approaches to treat or prevent RIX. Minipigs receiving an injection of rapamycin, an inhibitor of mTOR signaling, one hour prior to radiotherapy had improved saliva flow rates 12 weeks following treatment (156). Another potential pharmacological intervention for salivary gland regeneration is the postirradiation delivery of ectodysplasin A receptor (EDAR)-agonist monoclonal antibodies. EDAR is a signaling molecule involved in salivary gland development. Transient activation of EDAR signaling after ionizing radiation (5 Gy) restored salivary gland function and amylase levels after 90 days in mice (157).

Regenerative treatments to restore salivary gland functions after radiation therapy require further investigation, but provide a promising outlook for relief and prevention of RIX symptoms.