Functionality of Bone Marrow Mesenchymal Stromal Cells Derived from Head and Neck Cancer Patients - A FDA-IND Enabling Study Regarding MSC-based Treatments for Radiation-Induced Xerostomia

Abstract

Purpose:

Salivary dysfunction is a significant side effect of radiation therapy for head and neck cancer (HNC). Preliminary data suggests that mesenchymal stromal cells (MSCs) can improve salivary function. Whether MSCs from HNC patients who have completed chemoradiation are functionally similar to those from healthy patients is unknown. We performed a pilot clinical study to determine whether bone marrow-derived MSCs [MSC(M)] from HNC patients could be used for the treatment of RT-induced salivary dysfunction.

Methods:

An IRB-approved pilot clinical study was undertaken on HNC patients with xerostomia who had completed treatment two or more years prior. Patients underwent iliac crest bone marrow aspirate and MSC(M) were isolated and cultured. Culture-expanded MSC(M) were stimulated with IFNγ and cryopreserved prior to reanimation and profiling for functional markers by flow cytometry and ELISA. MSC(M) were additionally injected into mice with radiation-induced xerostomia and the changes in salivary gland histology and salivary production were examined.

Results:

A total of six subjects were enrolled. MSC(M) from all subjects were culture expanded to >20 million cells in a median of 15.5 days (range 8–20 days). Flow cytometry confirmed that cultured cells from HNC patients were MSC(M). Functional flow cytometry demonstrated that these IFNγ-stimulated MSC(M) acquired an immunosuppressive phenotype. IFNγ-stimulated MSC(M) from HNC patients were found to express GDNF, WNT1, and R-spondin 1 as well as pro-angiogenesis and immunomodulatory cytokines. In mice, IFNγ-stimulated MSC(M) injection after radiation decreased the loss of acinar cells, decreased the formation of fibrosis, and increased salivary production.

Conclusions:

MSC(M) from previously treated HNC patients can be expanded for auto-transplantation and are functionally active. Furthermore, IFNγ-stimulated MSC(M) express proteins implicated in salivary gland regeneration. This study provides preliminary data supporting the feasibility of using autologous MSC(M) from HNC patients to treat RT-induced salivary dysfunction.

Introduction

Salivary dysfunction is a side effect of radiation to the head and neck region that affects more than 40% of all patients with head and neck cancer (HNC)¹. Additionally radiopharmaceuticals, such as ¹⁷⁷Lu-PSMA-617 (Pluvicto) frequently cause salivary dysfunction². Salivary dysfunction can be due to hyposalivation (decreased salivary production) or alterations in sialochemistry (pH, electrolyte, and protein content). Patients with radiation-induced salivary dysfunction often experience xerostomia (the subjective sensation of dry mouth). Salivary dysfunction in HNC patients leads to increased dental caries, impaired swallowing ability, difficulty speaking, and diminished taste. These complications can be severely detrimental to a patient’s overall quality of life. Current treatment options for radiation-induced xerostomia are supportive in nature: carrying a water bottle, consumption of specially prepared food, use of salivary substitutes, chewing gum, sugar-free mints, and pilocarpine^3–6. These supportive interventions do not reverse the causes of xerostomia and are palliative, highlighting the critical need for improved therapies.

Mesenchymal stromal cells (MSCs) have been shown to promote tissue healing and regeneration in a variety of injurious settings⁷. These healing effects may derive from the MSC secretome, which has broad immunomodulatory and trophic activity^8,9. WNT1 and glial cell line-derived neurotrophic factor (GDNF) have been shown to be key drivers of adult salivary gland stem cells, allowing for expansion and restoration of function^10,11. Several groups have demonstrated in preclinical systems, that injecting MSCs into salivary glands results in a 32%−200% increase in salivary flow^12–18. However, there have been no preclinical studies on IFNγ stimulated MSC(M) for radiation-induced xerostomia in the murine model. There has been one trial to date examining autologous adipose-derived MSCs to treat radiation-induced xerostomia in humans: the MESRIX study demonstrated increased salivary flow rates at early time points and improved patient reported salivary function¹⁹.

The MESRIX trial did not characterize the secretome profile of the adipose-derived MSCs utilized, and no group has described the functionality of bone marrow-derived MSCs [MSC(M)] from patients treated with radiation for HNC. Multiple clinical studies have demonstrated that the radiation and chemotherapy regimens given to patients with hematological malignancies can damage the bone marrow and affect MSC function; however, no studies have been done in patients treated for HNC^20–22. Additionally, the MESRIX trial utilized allogenic MSCs, which allow for only one injection, rather than autologous MSCs which can be injected multiple times. Preliminary studies suggest that MSCs do not linger in the salivary glands, their tissue healing effects may instead be due to their secretome. Thus, multiple injections of MSCs may be needed to effectively treat chronic radiation-induced xerostomia. However, multiple injections of MSCs would only be feasible using autologous MSCs due to immune surveillance. The MESRIX II trial utilized allogenic MSCs without significant toxicity but only investigated a single injection²³. Autologous MSCs allow for multiple injections without the concern for an immune response, making autologous MSCs more feasible for potential long term therapy.

One way to accomplish multiple injections is by expanding MSCs and cryopreserving them until clinically needed. Since cryopreservation can result in cellular injury, less effective immunomodulation, or senescence^24–26, MSC(M) were stimulated with interferon gamma (IFNγ) prior to cryopreservation, which has been shown to prevent this immune dysfunctionality^25–27. We investigated this method of IFNγ stimulation for the MSC(M) from HNC patients to inform a future phase I clinical trial. We additionally investigated the secretome of MSC(M) in order to determine if IFNγ stimulated MSC(M) from HNC patients express key factors of WNT, R-spondin 1, and GDNF. Finally, we investigated the role of these IFNγ stimulated MSC(M) in the murine model of xerostomia, providing preclinical support of a future phase I clinical trial.

In preparation for a first-in-human study using IFNγ stimulated MSC(M), we performed this pilot study to verify the ability of MSC(M) obtained from previously treated HNC patients to be ex vivo expanded and to assess the functionality and secretome of IFNγ stimulated MSC(M).

Methods and Materials

MSC(M) Isolation and Culture

After institutional review board approval (UW HSB IRB#19009, ClinicalTrials.gov: NCT04007081) and informed consent, MSC(M) were isolated from 10 to 25 mL of bone marrow aspirated from the iliac crest of patients previously treated for HNC with symptomatic xerostomia. Bone marrow from healthy adult control volunteers was purchased from a commercial source (AllCells, Alameda, CA). We had originally planned to enroll 15 patients: 5 patients with HNC who underwent chemoradiation, 5 patients with HNC who underwent radiation therapy alone, and 5 HNC patients who had not yet received treatment for their disease. After enrolling 6 patients, the COVID-19 pandemic caused a pause in trial enrollment. An unplanned interim analysis of the MSC(M) from the 6 patients with HNC (4 treated with chemoradiation, 2 with radiation alone) did not reveal any significant differences between these MSC(M) and the healthy control MSC(M) and the protocol was amended to end enrollment.

MSC(M) isolation and culture were undertaken as previously described²⁸. Briefly, bone marrow aspirates were diluted 1:2 with phosphate-buffered saline and layered onto a Ficoll density gradient. Cells were centrifuged at 400 xg for 20 minutes with no acceleration or brake, and the mononuclear cells were then collected and plated in complete human MSC medium (MSC NutriStem XF Basal Medium, 5% human platelet lysate (MillCreek PLT MAX), 1% GlutaMAX, 2U/mL heparin, and 100 U/mL penicillin/streptomycin) at 100,000 to 300,000 cells/cm². Non-adherent hematopoietic cells were removed by changing the medium after 3 days of culture, and MSC(M) were allowed to expand for 7–15 days. Thereafter, cells were passaged 1–3 times weekly by treatment with TrypLE select and reseeded in fresh complete MSC medium at 3,000–5,000 cells/cm². MSC(M) were counted at each passage using a Countess^™ automated cell counter (Invitrogen^™ Grand Island, NY).

IFNγ Stimulation

Recombinant IFNγ was reconstituted to 200 μg/mL and added to complete MSC medium to a final concentration of 1,200 IU/mL. When MSC(M) reached the desired number for cryopreservation, complete MSC medium was aspirated from flasks and IFNγ-supplemented medium was added to all but the negative control flask. MSC(M) were then cryopreserved 24 +/− 2 hours later.

Cryopreservation of MSC(M)

MSC(M) were counted and resuspended in Plasma-Lyte A at a concentration of 5×10⁶ cells/mL. Cryopreservation medium was prepared (40% Plasma-Lyte A, 40% human serum albumin, 20% DMSO) and chilled. Five hundred μL of cryopreservation medium and 500 μL of MSC(M) were placed in each cryovial, and the cryovials were placed in a ThermoScientific^™ Mr. Frosty^™ Freezing container at −80 °C for 24 hours. The cryovials were then transferred to vapor-phase liquid nitrogen storage.

Phenotyping MSC(M) by Flow Cytometry

MSC(M) were thawed from cryopreservation and plated in complete MSC medium for 24 hours. MSC(M) were then harvested and resuspended at a concentration of 1 × 10⁶ cells/mL and analyzed by flow cytometry for the expression of CD73, CD90, CD105, CD14, CD20, CD34, and CD45 with the appropriate isotype controls using the Miltenyi MSC Phenotyping Kit (Miltenyi Biotec, Bergisch Gladbach, Germany). See Supplemental Table 1 for complete antibody details. All HNC MSC samples were run on an Attune NxT flow cytometer. Data were analyzed using FCS Express (De Novo Software, Pasadena, CA). The data are presented as normalized median fluorescent intensity compared with isotype control and as histogram overlay.

Differentiation of MSC(M) in adipocytes and osteocytes

MSC(M) were thawed and plated at a density of 4.2×10³ cells/cm² in two 24 well plates in MSC culture media. One plate of cells was treated with StemXVivo Adipogenic media (StemXVivo Osteogenic/Adipogenic Base Media, Cat: CCM007 + StemXVivo Adipogenic Supplement 100x, Cat: CCM011 + 1% Penicillin/Streptomycin) in order to induce adipogenesis. Adipogenic media was added 24 hours after plating and changed every 3 days. Differentiation stopped at 12 days after adipocytes were identified in the culture upon microscopic evaluation. Adipogeneis was confirmed by performing Oil Red O staining of the cells²⁹. The second plate was treated with StemXVivo Osteogenic Media (StemXVivo Osteogenic/Adipogenic Base Media, Cat: CCM007 + StemXVivo Human Osteogenic Supplement 20x, Cat: CCM008 + 1% Penicillin/Streptomycin) to induce osteogenesis. Osteogenic media was added 24 hours after plating and changed every 3 days, differentiation was stopped after 14 days as indicated by the manufacturer protocol. Osteogenesis was confirmed via Alzarian Red staining of the cells²⁹.

Functionality of MSC(M) by Flow Cytometry

Both IFNγ stimulated and non-IFNγ stimulated HNC MSC(M) were thawed from cryopreservation and plated in complete MSC media for 24 hours. MSC(M) were then harvested and resuspended at a concentration of 1 × 10⁶ cells/mL and analyzed by flow cytometry for the expression of MHC I, MHC II, IDO, ICAM-1, and PD-L1 (Miltenyi Biotec, Bergisch Gladbach, Germany; BioLegend, San Diego, CA; eBioscience, San Diego, CA), see Supplemental Table 1 for complete antibody details. All samples were run on an Attune NxT flow cytometer with the appropriate isotype controls (ThermoFischer Scientific, Waltham, MA). Data are presented as normalized median fluorescent intensity.

ELISAs and Cytokine Multiplex

Cryopreserved IFNγ MSC(M) were thawed and plated at a concentration of 1 × 10⁶ cells/well in 12 well plates with 1 mL complete MSC media. The supernatant from the MSC(M) was collected at 24 hours. Enzyme-linked immunosorbent assays (ELISAs) of WNT1 (MyBioSource WNT1 ELISA Kit), GDNF (Human GDNF ELISA Kit, Invitrogen ThermoFisher Scientific), and R-spondin 1 (MyBioSource R-spondin 1 ELISA Kit) were run according to the manufacturer’s instructions. The ELISAs were read on a SpectraMax i3 plate reader (Molecular Devices San Jose, CA), and the protein concentrations were interpolated from curves constructed from the protein standards and their respective median fluorescence intensity (MFI) readings. A multiplex immunoassay was used to determine the concentrations of 30 cytokines in the MSC supernatant (Human Cytokine Magnetic 30-Plex Panel, Invitrogen ThermoFisher Scientific) following the manufacturer’s instructions. The multiplex was read on the MAGPIX System (Millipore), and the protein concentrations were interpolated from curves constructed from the protein standards and their respective median fluorescence intensity (MFI) readings (Milliplex Analyst, Millipore).

Mouse Care and Irradiation

Six- to eight-week-old male C57BL/6 mice were purchased from Jackson Laboratories (Bar Harbor, ME) and the animal study was approved by the Institutional Animal Care and Use Committee (IACUC) at University of Wisconsin-Madison. Mice were irradiated on salivary glands with 15 Gy using Xstrahl Small Animal Radiation Research Platform (SARRP, Xstrahl, UK) one day before surgery. Mice were anesthetized in a chamber with 3–5% isoflurane at 1–2L/min O₂. Mice were then moved to a flatbed with a nosecone within the SARRP and maintained with 1–3% isoflurane for the duration of treatment. Using the MuriPlan software, a Cone-Beam CT image was acquired and a protocol was established for administering a total of 15 Gy split between two beams at 90 and −90 degrees to equally affect both salivary glands, with a dose rate of 2.68 Gy/min.

One day after radiation, mice were anesthetized with isoflurane and a small incision was made in the skin of each mouse followed by subcutaneous blunt dissection. The salivary glands were located and then human MSC(M) (1×10⁶ cells in 50 ul PBS) or PBS were injected into each gland, MSC(M) for the MSC(M) treated group and PBS for control and radiation groups. Submandibular salivary gland tissue was harvested at end of study, 3 months after injection.

Murine Salivary Collection

Saliva was collected from isofluorane anesthetized mice at baseline and 12 weeks after RT using a protocol adapted from Bagavant and colleagues^30,31. Briefly, two hours prior to each collection, food was removed from the cage. Saliva production was stimulated by injection of pilocarpine (1mg/kg body weight). Five minutes after pilocarpine injection, saliva was collected via absorbent swab for 15 minutes. The volume of saliva produced was determined by calculating the difference between dry and wet weight of the swabs. The ratio of saliva to body weight was calculated at each time point.

Morphology Analysis and Immunohistochemistry of Murine Salivary Glands

Salivary gland tissue was fixed in 10% neutral-buffered formalin and embedded in paraffin blocks. Five-μm sections of salivary gland were stained with hematoxylin and eosin (H&E), alcian blue (ALB), and Massons trichrome (MTC). Additionally, immunohistochemistry for α-Amylase (Ab125230, Abcam, 1:1000 dilution) was completed. All slides were imaged on an Olympus BX41 microscope (Olympus America, Inc, Waltham, MA,).

Statistical Analysis

Data are reported as mean +/− standard deviation or standard error, or median with range. Calculations were carried out using GraphPadPrism software (La Jolla, CA). Comparisons between groups were made by unpaired t-test between isotype and MSC using the False Discovery Rate approach with a two-stage step-up method and a Q=1%.

For determining potential differences in secretome cytokine expression, all calculations were done in R version 4.3.0 (2023–04-21). Data were normalized, a heatmap was generated, and based on this, pairwise scatterplots were viewed. Permutation testing to validate the results of the pairwise scatterplots was completed. Finally, a linear discriminant analysis (LDA) was performed.

Results

We obtained bone marrow aspirates from six patients with HNC who had completed curative-intent therapy at least two years prior to enrollment. Four of the six patients underwent concurrent chemoradiation and two underwent radiation alone (Table 1). The patients had a median age of 65 (range 58–70). All six samples were obtained fresh (10 – 20 mL) with an average starting mononuclear cell count of 75 × 10⁶ (range 7 × 10⁶ - 180 × 10⁶ cells). The MSC(M) were expanded in culture prior to cryopreservation for a median of 16 days (range 8–20 days). All patients’ MSC(M) were able to be expanded to over 20 × 10⁶ cells, which is the number of cells needed for the subsequent clinical trial that will be injecting MSC(M) into the submandibular glands of patients with radiation-induced xerostomia. Control MSC(M) were purchased as frozen bone marrow from AllCells from 3 healthy adult donors, median age of 26. The MSC(M) were isolated from an average starting mononuclear cell count of 94 × 10⁶ cells and cultured as described above.

Table 1:

Head and neck cancer diagnoses, curative therapies, and the treatment completion dates of the 6 patients enrolled in our study.

Subject	Diagnosis	Treatment	Treatment Completion
1	Stage IV (T3N2cM0) squamous cell carcinoma of the right base of tongue, p16+	Definitive chemoradiation 70 Gy with weekly cetuximab	2/2014
2	Stage IVA (T3N2cM0) squamous cell carcinoma of the right base of tongue, p16+	Definitive chemoradiation 70 Gy with weekly cisplatin	3/2014
3	Stage IVA (T1cN2cM0) squamous cell carcinoma of the right base of tongue, p16+	Definitive chemoradiation 70 Gy with weekly cisplatin	12/2014
4	Stage IVA (T3N2bM0) squamous cell carcinoma right tonsil and retromolar trigone, p16+	Definitive chemoradiation 70 Gy with weekly cisplatin	7/2014
5	Stage II (T2N0M0) squamous cell carcinoma right buccal mucosa	Adjuvant radiation 60 Gy	9/2014
6	Stage IVA (TxN2aM0) squamous cell carcinoma of unknown primary	Definitive radiation 64.8 Gy	7/2011

The MSC(M) from HNC patients displayed the MSC phenotype (CD73⁺/CD90⁺/CD105⁺/CD14⁻/CD20⁻/CD34⁻/CD45⁻) consistent with the minimal criteria for defining human MSCs by the International Society for Cellular Therapy (ISCT), with detectable expression of CD90, CD105, and CD73 (p<0.05) and lack of expression of CD14, CD20, CD34, and CD45. (Figures 1A–1B, Supplemental Figure 1) compared to isotype control. All MSC(M) from HNC patients were able to be differentiated into adipocytes and osteocytes (Figure 1C). Time from harvest to P0 did not significantly differ between MSC(M) from HNC patients and healthy control patients (Supplemental Figure 2). After expansion, MSC(M) from HNC patients were stimulated with IFNγ and cryopreserved for at least 5 days. The cryopreserved MSC(M) were then thawed, and their growth curves were determined (Figure 1D).

Figure 1:

Characterization of bone marrow derived mesenchymal stromal cells [MSC(M)] from patients with head and neck cancer (HNC). A) Graph of the normalized median fluorescent intensity of classical markers of HNC MSC(M). There was a significant increase of CD90, CD105, CD73 as compared to isotype control (*p=<0.01), with no significant difference in the median fluorescent intensity between MSCs and isotype control for the non-MSC markers (CD14, CD20, CD34, and CD45). B) Histograms of the 6 HNC MSC(M) showing the expression of classical markers (CD90, CD105, and CD73) and no expression of non-MSC Markers (CD14, CD20, CD34, and CD45) (MSC(M) shown as black outline, isotype control shown as gray curve). C) Differentiation of the MSC(M) into adipocytes and osteocytes was achieved for all 6 patients, with healthy control shown for comparison. D) Growth curve of the 6 HNC MSCs after IFNγ stimulation, cryopreservation, and thawing, showing variable speeds of continued growth over two weeks. E) Median fluorescent intensity (MFI) of immunosuppressive functional markers of head and neck cancer patients’ MSCs with and without IFNγ stimulation. There is a significant increase in MHC I, MHC II, IDO, and ICAM-1, with a trend in increased PD-L1 expression without statistical significance, all MFIs are median with standard error bars, *p<0.05.

Each HNC patient’s IFNγ-stimulated MSC(M) were compared to their non-IFNγ stimulated cryopreserved MSC(M) using flow cytometry. The IFNγ-stimulated MSC(M) acquired an immunosuppressive phenotype with increased IDO, ICAM-1, PD-L1, MHC I, and MHC II compared to unstimulated MSC(M) (supplemental Figure 3). All markers except PD-L1 were significantly increased, p<0.05.

To assess the secretome of cryopreserved IFNγ stimulated MSC(M) from HNC patients and healthy controls, we utilized ELISA assays. HNC MSC(M) were found to express GDNF (mean 117.5 pg/mL, Figure 2A), WNT1 (mean 9.4 ng/mL, Figure 2B), and R-spondin 1 (mean 65.8 pg/mL, Figure 2C) at similar levels as MSC(M) from healthy controls.

Figure 2:

Cytokine expression of IFNγ stimulated bone marrow derived mesenchymal stromal cells [MSC(M)] from patients with treated head and neck cancers (HNC) and healthy donors after cryopreservation and culture rescue. A) GDNF expression, mean GDNF for HNC MSCs was 117.5 ± 15.9 pg/mL (standard deviation 15.9 pg/mL) mean GDNF for healthy donors was 155.7 ng/mL (standard deviation 21.6 ng/mL). B) WNT1 expression, mean WNT1 was 9.4 ng/mL (standard deviation 0.8 ng/mL, mean WNT1 for healthy donors was 9.3 ng/mL (standard deviation 1.6 ng/mL). C) R-spondin 1 expression, mean R-spondin 1 was 65.8 pg/mL (standard deviation 29.7 pg/mL), mean R-spondin 1 for healthy donors was 87.8 pg/mL (standard deviation 48.1 pg/mL).

Using a multiplex ELISA assay, cryopreserved IFNγ stimulated HNC and control BM-MSC secretome was analyzed further. A heat map of the cytokines examined in the multiplex was created and showed clustering of the cytokines in two main groups, independent of origin from control versus HNC (Figure 3A). Based on this, two cytokine pairs with the possible ability to discriminate between control and HNC patients were found: FGF with IL-8 (CXCL8) and FGF with VEGF-A (Figure 3B and 3C). FGF and VEGF-A were found to have a p-value of 0.08 on permutation testing to determine the discriminant ability, which is the maximum p-value possible for this sample size. LDA confirmed the selection of FGF and VEGF-A. Concentrations of the cytokines investigated for their discriminant ability are shown in Figure 3D, other cytokine concentrations are shown in supplemental figure 4.

Figure 3:

Further analysis of secretome. A) Heat map of the correlation between each pairing of cytokines examined in the secretome, with darker colors demonstrating a stronger correlation. B) Pairwise scatter plot of the relationship between FGF and VEGF-A with each patient represented as an empty circle, showing the clustering of head and neck cancer (HNC) patients and control patients, with each group’s mean shown in a solid circle. C) Pairwise scatter plot of the relationship between FGF and IL-8 (CXCL8) with each patient represented as an empty circle, showing the clustering of HNC patients and control patients, with the group’s mean shown in a solid circle. D) Concentrations of cytokines that may be used for discriminating between HNC mesenchymal stromal cells (MSC) and healthy control MSC, means reported for all MSC: mean IL8 2464.0 pg/mL, mean FGF 24.8 pg/mL, and mean VEGF 19.7 pg/mL.

To assess the effects of IFNγ-stimulated MSC(M) on radiated salivary glands, we injected IFNγ-stimulated MSC(M) or PBS into the salivary glands of mice. There were three groups of mice with 12 mice per group: the control group who did not undergo radiation but had PBS injection, the radiation group who underwent radiation and PBS injection, and the MSC(M) group who underwent radiation and MSC(M) injection. As expected, we found the salivary glands of mice injected with PBS after irradiation to have less amylase and mucin as well as increased fibrosis (Figure 4A). In mice who were injected with IFNγ-stimulated MSC(M) there was less radiation damage observed by H&E and decreased MTC staining of collagen fibrosis, as well as increased levels of amylase and mucin expression – more similar to unirradiated controls (Figure 4A). Additionally, we found injection of IFNγ-stimulated MSC(M) resulted in increased salivary production (Figure 4B). Mean saliva to weight ratio (S:W) in the control group at baseline was 2.50 (standard error of the mean, SEM, 0.65) and 3.59 (SEM 0.67) at 12 weeks post RT, in the irradiated group S:W was 3.64 (SEM 0.62) at baseline and 0.85 (SEM 0.26) at 12 weeks post RT, and in the irradiated group injected with MSC(M) S:W was 3.89 (SEM 0.84) at baseline and 2.26 (SEM 0.81) at 12 weeks.

Figure 4:

Treatment of irradiated mice with MSC(M) allows for salivary gland recovery. A) Murine submandibular glands at 20x magnification show changes after radiation and MSC(M) injection. Radiation (RT) resulted in histopathologic changes including decreased mucin, decreased amylase, and increased fibrosis (collagen). In irradiated mice injected with IFNγ-stimulated MSC(M) there was restoration of amylase and mucin production and decreased fibrosis, similar to unirradiated controls. ALB – Alician blue stains mucin, MTC – Massons trichrome stains collage fibers. B) Bar graph of the ratio of salivary production (in grams) to mouse body weight (in grams) showing MSC(M) injection allows for recovery of salivary production. Graph shows mean and SEM error bars. **p=0.0015

Discussion

In this study we verified that MSC(M) of patients with HNC who underwent radiation (+/− chemotherapy) at least two years prior could be isolated, culture-expanded, cryopreserved, and culture-rescued. In addition, we confirmed that these MSC(M) expressed key salivary stem cell morphogens and immunomodulatory cytokines consistent with a potential role in supporting salivary gland regeneration or downregulating a harmful inflammatory response. Our data demonstrate for the first time that functionally immunosuppressive MSC(M) can be expanded from the bone marrow of patients with HNC who were treated with radiation (+/− chemotherapy), despite the potentially myelosuppressive effects of these treatments and the relatively advanced age of the average HNC patient.

We performed this study because several in vitro studies have demonstrated the damaging effects of chemotherapy and radiation on MSC(M)^32,33. Clinical studies have also shown the radiation and chemotherapy regimens given to patients with hematological malignancies can damage the bone marrow and affect MSC(M)^20–22. In our study, we did not identify any negative effects of HNC treatment on MSC(M). This is likely due to the nature of the therapies for our patients. Radiation would have been directed at the head and neck regions, avoiding the majority of the MSC-producing marrow and the chemotherapy regimen was either cisplatin or cetuximab, both of which are much less myelosuppressive than the regimens typically used for patients with hematological malignancies. Some studies have found cisplatin to affect the viability and differentiation of MSCs, especially in the context of concurrent radiation^34,35. However, the extent of cisplatin effects on MSCs is debated, with other studies finding MSCs relatively unaffected by cisplatin³⁶. We found MSC(M) from HNC patients were able to proliferate at a rate like those of healthy volunteers, despite having previously undergone curative treatments for HNC.

We then undertook assays to examine the growth and function of these MSC(M) after cryopreservation. Cryopreservation is a useful tool to allow the expansion of large numbers of MSC(M) that can be frozen and thawed on an as-needed basis for clinical trials – particularly those that might require repeated MSC injections. There are some potential drawbacks to cryopreservation; the freeze/thaw cycle can result in cellular injury or senescence impairing the functional abilities of MSC(M)^25,26,37. Prior studies have shown stimulation of MSC(M) with IFNγ prior to cryopreservation can prevent cellular dysfunction^25–27. In our study we demonstrated that MSC(M) from HNC patients can be successfully cryopreserved and culture-rescued while retaining appropriate function. MSC(M) can exert their beneficial effects through modulation of the immune microenvironment and/or through the provision of tissue-promoting morphogens.

Interestingly, several studies on patients who underwent myeloablative chemotherapies for multiple myeloma found their MSCs to have impaired immunomodulatory activity^38,39. Additionally, cryopreservation has been shown to result in less effective immunomodulation^24–26. Here we demonstrated that IFNγ stimulated MSC(M) from HNC patients have an immunosuppressive phenotype with robust growth after thawing. We additionally found that injection of MSC(M) into the irradiated salivary glands of mice resulted in increased salivary production and decreased fibrosis – supporting our hypothesis that these IFNγ-stimulated MSC(M) may restore salivary gland function after radiation.

MSCs can also limit cell death and encourage repair and expansion of the tissue’s resident stem cells through the release of tissue-promoting morphogens⁷. We examined the paracrine factors expressed by MSC(M) to determine if the MSC(M) from treated HNC patients secrete restorative cytokines. We found production of GDNF, WNT1, and R-spondin 1 by the IFNγ stimulated MSC(M) from HNC patients, suggesting these MSC(M) can provide factors which are crucial to the function of adult stem cells in glandular tissues^10,11,40. The stem cell niche in gut epithelium utilizes WNT and R-spondin 1 signaling to allow for the renewal of adult stem cells^40–42. Maimets et al. has also shown WNT signaling to be a key driver of adult stem cells in the salivary glands¹⁰. Further, GDNF has also been identified as an important chemokine in the adult salivary stem cell¹¹. Furthermore, we demonstrated that the secretum of IFNγ stimulated MSC(M) from HNC patients contains key promoters for adult salivary stem cells, at similar levels as MSCs from healthy controls. Based on our further investigation of the secretome, FGF and VEGF-A may be used as a discriminatory cytokine pair for differentiating between MSCs of healthy control patients and HNC patients. While this pair was not statistically significant, this may be largely driven by the small number of subjects in this study. This potential difference in secretome, particularly in FGF and VEGF-A is intriguing, as MSCs are known to influence the immune microenvironment to be pro-regeneration and anti-inflammatory^43–45. Further investigation into the secretome of IFNγ stimulated MSC(M) is needed to determine the levels of various cytokines, and their clinical relevance.

This study was limited by the small number of patients enrolled. Our goal was to ensure that the MSC(M) from HNC patients met minimal criteria for phenotype and function per ISCT criteria, which was able to be met with our six biological replicates²⁴. These pre-clinical data inform a clinical trial (NCT04489732) investigating the injection of autologous MSC(M) into the salivary glands after radiation (+/− chemotherapy) as an innovative remedy to treat xerostomia and restore quality of life. This clinical trial will include a larger number of HNC patients which will allow us to better examine the immunosuppressive phenotype of IFNγ stimulated MSC(M) and the secretome that these cells produce.

Conclusion

Based on these data we conclude that MSC(M) from patients with HNC who underwent curative radiation (+/− chemotherapy) at least 2 years prior have similar expansion capabilities as MSC(M) from healthy volunteers, express an immunosuppressive phenotype with IFNγ stimulation, and have an anti-inflammatory, pro-regeneration secretome. This work serves as a foundation for further studies examining the use of MSC(M) from patients who underwent chemoradiation for HNC, including as a possible therapy for RT-induced salivary dysfunction.

Significant Statement:

Salivary dysfunction is a significant side effect of radiation therapy for head and neck cancer patients. Preliminary data suggests that mesenchymal stromal cells can improve salivary function by regenerating adult salivary gland stem cells. We showed that cryopreserved bone marrow derived mesenchymal stromal cells from head and neck cancer patients treated with chemotherapy and radiation are functionally active after stimulation with interferon gamma and produce proteins needed for salivary gland regeneration. This study provides preliminary data supporting the feasibility of using interferon gamma stimulated mesenchymal stromal cells to treat radiation-induced salivary dysfunction.

Data Availability:

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.