Abstract
Background
Toxicity of the oral and gastrointestinal mucosa induced by high-dose melphalan is a clinical challenge with no documented prophylactic interventions or predictive tests. The aim of this study was to describe molecular changes in human oral mucosa and to identify biomarkers correlated with the grade of clinical mucositis.
Methods and Findings
Ten patients with multiple myeloma (MM) were included. For each patient, we acquired three buccal biopsies, one before, one at 2 days, and one at 20 days after high-dose melphalan administration. We also acquired buccal biopsies from 10 healthy individuals that served as controls. We analyzed the biopsies for global gene expression and performed an immunohistochemical analysis to determine HLA-DRB5 expression. We evaluated associations between clinical mucositis and gene expression profiles. Compared to gene expression levels before and 20 days after therapy, at two days after melphalan treatment, we found gene regulation in the p53 and TNF pathways (MDM2, INPPD5, TIGAR), which favored anti-apoptotic defense, and upregulation of immunoregulatory genes (TREM2, LAMP3) in mucosal dendritic cells. This upregulation was independent of clinical mucositis. HLA-DRB1 and HLA-DRB5 (surface receptors on dendritic cells) were expressed at low levels in all patients with MM, in the subgroup of patients with ulcerative mucositis (UM), and in controls; in contrast, the subgroup with low-grade mucositis (NM) displayed 5–6 fold increases in HLA-DRB1 and HLA-DRB5 expression in the first two biopsies, independent of melphalan treatment. Moreover, different splice variants of HLA-DRB1 were expressed in the UM and NM subgroups.
Conclusions
Our results revealed that, among patients with MM, immunoregulatory genes and genes involved in defense against apoptosis were affected immediately after melphalan administration, independent of the presence of clinical mucositis. Furthermore, our results suggested that the expression levels of HLA-DRB1 and HLA-DRB5 may serve as potential predictive biomarkers for mucositis severity.
Introduction
For three decades, high-dose melphalan, supported with autologous stem cell transplantation (HSCT), has been a component of treatment for patients with newly diagnosed multiple myeloma (MM) [1]. However, melphalan induce adverse effects, including inflammation of the oral and gastrointestinal mucosa (mucositis) and prolonged neutropenia, which necessitates HSCT [2]. Melphalan induced mucositis occurs inconsistently, because although 80% of patients experience some degree of mucositis, only 40% are affected severely [2, 3]. Severe toxicity unfolds as a loss of mucosal integrity, severe diarrhea, and painful oral ulcers; i.e., ulcerative mucositis (UM) [3]. Complicated by bacterial or viral infections, these patients more often experience nausea, diarrhea, febrile episodes, and longer hospital stays compared to patients with mild or no mucositis (NM) [4, 5]. At present, international recommendations consist of infection control and palliative measures for pain relief [6]. Despite intense research efforts, no methods exist for preventing or reducing the duration of mucositis, and no predictive tests are available [7].
The mechanisms of action and metabolism of melphalan are well-described [8]. Melphalan alkylates DNA, which causes cross-links to form between DNA strands, and subsequently, DNA is degraded through apoptosis. The drug is administered intravenously, metabolized in the liver, and excreted through feces and urine. The degree of toxicity depends on renal function, body mass index (BMI), gender, and performance status [2, 9]. However, none of these factors are predictors of UM.
The current model of mucositis pathology is generalized across treatment regimens [10]. Initially, cancer therapy-induced DNA damage activates the intrinsic pro-apoptotic Bax/Bak and p53 pathways, and reactive oxygen species (ROS) are released [11, 12]. Simultaneously, damage to the extracellular matrix induces the release of inflammatory cytokines, which activate the extrinsic apoptotic pathway via tumor necrosis factor alpha (TNF-α) [13, 14]. This release is followed by an inflammatory response, which includes upregulation of the interleukins (IL) IL-1β, IL-6, IL-10, transforming growth factor-beta (TGF-β), nuclear factor-kappaB (NF-κB), and matrix metalloproteinases (MMPs) [15, 16]. This model is mainly based on murine studies and a few human studies, but to the best of our knowledge, no study has focused on patients with MM that were treated with melphalan.
Recent genome-wide association studies (GWAS) of patients that underwent HSCT have implied that UM development is associated with a genetic predisposition, primarily related to immune function [17, 18]. One study included 153 patients with miscellaneous malignancies that underwent HSCT, with the aim of building a predictive network for UM, based on 82 selected single nucleotide polymorphisms (SNPs) [17]. The network was subsequently tested in a cohort of 16 patients, and in the absence of any false positives, the predictive validity of the network was 81.2%. A later study included 972 patients with MM that underwent HSCT, and they identified 11 SNPs located near the matrix metalloproteinase gene that were associated with UM and several known clinical risk factors. The sensitivity of predicting UM was 52% [18]. Apart from the low sensitivity, those studies were limited by their failure to identify phenotypes or causal relationships.
Here, we present a global gene expression study on oral mucosa biopsies and peripheral blood cell samples from consecutive patients with MM that were treated with high-dose melphalan and HSCT. This study aimed to identify new molecular factors that could predict severe oral mucositis.
Materials and Methods
Patients
This study included 30 patients, aged 18 years or older, recruited from the Aalborg University Hospital, from September 1st 2010 to September 1st 2012. Patients with MM (n = 20) were recruited from the Department of Hematology. Healthy individuals (CON, n = 10) were recruited for a control group from the Department of Maxillofacial Surgery. Of the 20 MM patients, seven withdrew consent before any intervention; one was missed due to earlier start of treatment, which was not communicated to the research unit; and two withdrew after the first biopsy without giving any reason. The remaining 10 patients provided three sequential buccal biopsies and peripheral blood samples. The first biopsy was obtained immediately before they received high-dose melphalan (day0); the second was obtained after the autologous stem cell reinfusion (day2); and the third was obtained during an outpatient control visit (day21). The CON group comprised 10 healthy, non-smoking, age- and gender-matched individuals. Controls provided one buccal biopsy and peripheral blood sample. One CON individual was later diagnosed with the autoimmune disease, psoriasis (CON09), and hence, this subject was not included in the statistical analysis. The North Denmark Region Committee on Health Research Ethics approved the clinical protocol (N-20100022). Informed written consent was obtained from all patients, in accordance with the Declaration of Helsinki.
All patients with MM underwent a comprehensive, initial evaluation, including a medical history and clinical examination. Age, gender, and Eastern Cooperative Oncology Group (ECOG) performance status were recorded at baseline, in addition to the subtype of MM and the time from diagnosis to entering HSCT. The criteria for determining the level of organ involvement at diagnosis was based on the degrees of elevated calcium, renal failure, anemia, and bone lesions (CRAB criteria) [19]. Furthermore, patients were screened for dental infections and, when indicated, these infections were eradicated prior to chemotherapy.
All patients with MM received a standard induction regimen, which consisted of cyclophosphamide 500 mg/m2 delivered intravenously (i.v.) on days 1 and 8; bortezomib 1.3 mg/m2 delivered subcutaneously (s.c.) on days 1, 4, 8, and 11; and dexamethasone 20 mg, delivered orally (p.o.) on days 1–2, 4–5, 8–9, and 11–12, repeated every third week, 3 times. After this treatment, patients were primed with cyclophosphamide 2 g/m2 and recombinant granulocyte stimulating factor (rhG-CSF), before their circulating CD34+ hematopoietic stem cells were harvested with leukapheresis [1]. Only patients without progressive disease were assigned to HSCT. These patients received a high dose of melphalan (200 mg/m2), followed by infusion of autologous hematopoietic stem cells. All patients with MM had received standard antiviral, antifungal, and antibacterial treatment, according to department protocols.
Mucositis and diarrhea assessments
Signs of oral mucositis (OM) were recorded daily for patients with MM during the hospital stay (from administration of chemotherapy to discharge). OM signs were identified according to the WHO oral toxicity assessment worksheet [20], and they included subjective symptoms (pain and ability to eat solid food) and objective findings (erythema, ulceration) in predefined regions of the mouth (lip, check, tongue, floor of the mouth, and soft palate). Grades 0 and 1 (NM) included increasing soreness, with or without erythema, but solid food could be taken. In grades 2 to 4 (UM), food intake gradually declined, due to pain and ulcerations, and parenteral feeding might have become necessary. The maximum OM grade recorded during treatment was considered the patient’s general mucositis experience. Diarrhea was estimated according to the Common Terminology Criteria for Adverse Events (CTCAE), issued by The National Cancer Institute of the National Institutes of Health [21]. In grades 1 to 2, vomiting increases from one to two episodes in 24 h to three to five episodes in 24 h. Grade 3 included more than 6 episodes in 24 h, and grades 4 to 5 were considered life-threatening, and could gradually lead to death. Diarrhea data were gathered retrospectively, from medical records.
Biopsy
All biopsies were acquired in a standardized manner. First, the mouth was thoroughly rinsed with chlorhexidine and local anesthesia was applied (0.5 ml Citanest®: prilocain 30 mg/ml + felypressin 0.54 μg/ml; Dentsply, York, PA, US). Then, a 5-mm lens-formed biopsy of the buccal mucosa, approximately 1 cm inferior to the papilla parotidea, was taken with a scalpel. The wound was tightly sutured with resorbable vicryl 4.0 (Ethicon, Sommerville, NJ, US). Patients were instructed to rinse twice daily with chlorhexidine until suture removal, after 10 days. One-half of the biopsy was immediately immersed in RNAlater™ (Ambion, Thermofischer Scientific, Waltham, MA, US) for 24 h; then, it was frozen at -80°C until analysis. The other half of the biopsy was fixed in 10% neutral-buffered formalin, and within 1½ days, it was embedded in paraffin and maintained at room temperature until further preparation.
Peripheral blood
Within 2 h of taking the biopsy, 15 ml EDTA-mixed venous full blood was drawn. Mononuclear cells (MNCs) were isolated with an in-house standard purification protocol. This protocol follows the manufacturer’s guidelines for Ficoll-PaqueTM (GE Healthcare, Little Chalfont, Buckinghamshire, UK); density gradient centrifugation in LeukosepR tubes (Greiner Bio-One GmbH, Frickenhausen, Germany). Purified MNCs were suspended in freezing medium containing 10% dimethyl sulfoxide, in units of 5 million, vital frozen at -196°C in liquid nitrogen, and stored frozen until analysis.
Gene expression
The frozen oral mucosa samples were homogenized with TRIzolR Reagent (Invitrogen, ThermoFischer Scientific), and total RNA was isolated with the mirVanaTM miRNA Isolation Kit (Ambion/Invitrogen, ThermoFischer Scientific) according to the manufacturer’s protocol. RNA amplification was performed with the AmbionR WT Expression Kit (Applied Biosystems, ThermoFischer Scientific), according to the manufacturer’s instructions, starting with 100 ng total RNA, on a TP Basic Thermocycler, real time PCR instrument (Biometra, Göttingen, Germany). The quality of the RNA product was evaluated on the NanoDrop spectrophotometer and the 2100 Bioanalyzer with the Agilent RNA 6000 Nano Kit (Agilent Technologies, Santa Clara, CA, US). We prepared the RNA samples for hybridization to Affymetrix GeneChip Human Exon 1.0 ST Arrays with the Affymetrix GeneChip WT Terminal Labeling and Controls Kit (P/N 901524) (Affymetrix, Santa Clara, CA, US), according to the manufacturer’s instructions. CEL files were generated with Affymetrix GeneChip Command Console Software and deposited at the NCBI Gene Expression Omnibus repository, under number GSE81979. A similar procedure was applied to analyze gene expression in MNCs isolated from blood samples.
Immunohistochemistry
We cut 4-μm-thick biopsy tissue sections and mounted them on glass coverslips. Following an in-house optimized protocol, tissues were stained with an antibody against the HLA class II Histocompatibility antigen, DR beta 5 chain (HLA-DRB5 center region) with a rabbit polyclonal antibody (no. OAAB06426, Aviva Systems Biology, CA, US). Normal tonsil tissue was used as a positive control. The stained slides were then scanned on a Hamamatsu NanoZoomer slide scanner and analyzed with NDP viewer software. To estimate the number of cells that stained positive for HLA-DRB5, each stained slide was searched for a hot spot; then, this spot was framed with a 0.75×0.4 mm (0.3 mm2) rectangle; the area included approximately half lamina epithelialis and half lamina propria. We counted all cells in the frame that were distinctly stained with anti-HLA-DRB5 antibodies. The analyzer was blinded to the mucositis grade.
Statistical analysis
Power estimation of group size
To identify genes that varied more than two-fold between test points with a false discovery rate of less than 0.05% and a power of 90%, we applied the method described by Lee and Whitmore [22], implemented in the R-package, size-power (Qui 2008). We found that 10 patients in each group were sufficient for detecting significant differences.
Statistical analysis
All statistical analyses were performed with R [23] version 3.2.0 and Bioconductor packages [24]. The p-values adjusted for false discovery rates were controlled with the Benjamini-Hochberg method [25], for each of the above tests. Adjusted p-values below 0.05 were considered significant.
The CEL files produced by the Affymetrix Expression Console and the probes were preprocessed and summarized to gene level with the RMA algorithm in the Bioconductor package ‘affy’, based on custom CDF files [26]. This preprocessing resulted in the gene expression levels of 38,830 genes for each Exon array each annotated with Ensembl gene (ENSG) identifiers. Patient CON09 was included in the normalizations of the gene expression data, but excluded in the statistical analysis.
With patient ID as a cluster variable, we used the linear model for microarray data (limma package in R), a mixed linear model, and an empirical Bayes approach to test for significant differences in gene expression levels between day2 and day0, and between day21 and day0 [27]. For the peripheral blood samples, we only compared day0 and day21 to baseline, because only two blood samples were analyzable for day2. We performed an unpaired test with the limma package to test for significant differences in gene expression between patients on day0 and controls.
The patients were divided into UM or NM groups, according to their mucositis experience. We used the limma package to detect significant genes that were differentially expressed between the two groups at each time point.
We applied the Mann-Whitney test to test for the relationship between mucositis severity and duration of neutropenia, leukopenia, and thrombocytopenia. We also used the Mann-Whitney test to evaluate differences between groups in the numbers of in-hospital days and years of progression free survival (PFS).
Results
The clinical characteristics and demographics of the included patients prior to HSCT are shown in Table 1. No signs of infection at the site of biopsy were reported. The clinical data collected during the HSCT and at follow up are shown in Table 2. UM (grades 2–4) was observed in 4 patients, and NM (grades 0–1) was observed in 6 patients. The average mucositis scores were 1.5 (range 0–4) for the whole cohort, 3.3 for the UM group, and 0.3 for the NM group. The average diarrhea scores were 2.2 (range 1–4) for the whole cohort, 3.3 (range 3–4) for the UM group, and 1.5 (range 1–2) for the NM group. The average hospital stays were 22.6 days (range 16–41) for the whole cohort, 28.8 days (range 21–41) for the UM group, and 18.5 days (range 16–24) for the NM group. The difference in hospital stays was statistically significant (p = 0.020). The duration of neutropenia was not significantly different between UM and NM groups, but thrombocytopenia was significantly prolonged in the UM compared to the NM group (p = 0.047). PFS [28] was not statistically different between the UM and NM groups. The CON and MM groups were comparable in age (CON: age 58 y, range 47–78 vs. MM: age 63.5 y, range 51–69) and gender (CON: females 4/10 vs. MM: females: 6/10).
Table 1. Patient characteristics and demographics upon enrollment in the study.
| Patient | Age | Gender | ECOG | Weight | MM | CRAB | Induction cycles | Response induction | Standard HSCT | Diagnosis to HSCT |
|---|---|---|---|---|---|---|---|---|---|---|
| MM01 | 62 | f | 2 | 69 | IgG-κ | B | 3 | No PD | Y | 5.3 |
| MM02 | 51 | m | 0 | 110 | IgG-λ | B | 3 | No PD | Y | 3.9 |
| MM04 | 66 | f | 0 | 70 | IgG-κ | R | 3 | No PD | Y | 3.7 |
| MM05 | 67 | m | 1 | 92 | IgG-κ | B | 3 | No PD | Y | 4.6 |
| MM07 | 67 | m | 0 | 63 | IgG-κ | C | 3 | No PD | Y | 143 |
| MM09 | 63 | f | 1 | 60 | IgG-κ | B | 3 | No PD | Y | 3.9 |
| MM15 | 69 | f | 0 | 72 | IgG-κ | A | 3 | No PD | Y | 36.8 |
| MM18 | 64 | f | 2 | 52 | IgG-κ | R | 3 | No PD | Y | 5.1 |
| MM19 | 64 | m | 0 | 90 | IgG-κ | B | 3 | No PD | Y | 4.1 |
| MM20 | 62 | f | 1 | 97 | IgG-λ | B | 3 | No PD | Y | 3.8 |
Abbreviations: ECOG = Eastern Cooperative Oncology Group performance status at baseline. MM = multiple myeloma subtype. CRAB = end-organ damage at diagnosis (C = hypercalcemia, R = renal failure, A = anemia, B = bone lesions) [19]. Induction cycles = Cyclophosphamide 500 mg/m2 i.v. days 1 and 8; Bortezomib 1,3 mg/m2 s.c. days 1,4,8, and 11; Dexamethasone 20 mg p.o. days 1–2, 4–5, 8–9, and 11–12, repeated every third week. PD = progressive disease, HSCT = high dose chemotherapy (melphalan 200 mg/m2) with autologous stem cell transplantation. Diagnosis to HSCT = months between the diagnosis of MM and the HSCT procedure.
Table 2. Patient clinical data during HSCT and at follow up.
| Patients | Mucositis grade1 | Diarrhea grade2 | Neutro-penia3 days | Leuko-penia4days | Thrombo-cytopenia5 days | In-hospital days | PFS years | Status at follow up6 |
|---|---|---|---|---|---|---|---|---|
| Patients with ulcerative mucositis | ||||||||
| MM04 | 2 | 3 | 4 | 4 | 10 | 41 | 4.5 | CR |
| MM18 | 4 | 4 | 6 | 6 | 12 | 29 | 3.1 | VGPR |
| MM19 | 4 | 3 | 10 | 8 | 10 | 21 | 3.0 | VGPR |
| MM20 | 3 | 3 | 10 | 10 | 6 | 24 | 3.0 | CR |
| Patients with no/mild mucositis | ||||||||
| MM01 | 0 | 2 | 10 | 10 | 14 | 24 | 4.1 | Relapse |
| MM02 | 0 | 1 | 12 | 8 | 14 | 17 | 4.5 | CR |
| MM05 | 1 | 2 | 6 | 6 | 14 | 21 | 3.2 | Relapse |
| MM07 | 0 | 2 | 8 | 4 | 20 | 16 | 1.5 | Relapse |
| MM09 | 0 | 1 | 10 | 8 | 10 | 17 | 1.0 | Relapse |
| MM15 | 1 | 1 | 8 | 6 | 10 | 16 | 2.2 | Relapse |
Abbreviations: HSCT = high-dose chemotherapy (melphalan 200 mg/m2) and autologous stem cell transplantation; PFS = progression free survival; the surrogate marker for overall survival was defined as the time from entering HSCT to disease progression, death, or follow-up [19]; CR = complete response; VGPR = very good partial response; Relapse = clinical relapse.
1 Calculated according to WHO mucositis assessment scale [20]. Patients that experienced mucosal ulcerations during treatment were considered to have ulcerative mucositis, WHO grades 2–4; patients with only soreness or erythema were considered to have none/mild mucositis, WHO grades 0–1.
2 Calculated according to the Common Terminology Criteria for Adverse Events [21]
3 Neutropenia was defined as <0.5×106/l
4 Leukopenia was defined as <0.5×109/l
5 Thrombocytopenia was defined as <150×109/l.
6According to the International Uniform Response Criteria for multiple myeloma [19].
Analysis of gene expression in mucosa samples
All 40 biopsies (3×10 patients and 1×10 controls) provided gene expression profiles. No statistically significant differences in gene expression were found between the MM group on day0 and the CON group. Patients with MM showed no significant changes in gene expression between day0 and day21. However, 35 genes in patients with MM showed significantly different expression between day0 and day2 (Table 3). The gene expression levels were independent of clinical mucositis. The dominant gene alterations were observed in apoptosis-related genes, followed by genes related to inflammatory/immunologic response, transcription factors, and members of the Histone Cluster family. We also observed alterations in genes related to metabolism.
Table 3. Genes altered in the buccal mucosa of patients with multiple myeloma.
| Gene symbol | FC | p-value | adjusted p-value | Qualified GO term | Function | ||
|---|---|---|---|---|---|---|---|
| Upregulated genes day2 versus baseline | |||||||
| MDM2 | 2.69 | 2.37e-15 | 3.07e-11 | MDM2 oncogene, E3 ubiquitin protein ligase | Apoptosis | ||
| EDA2R | 2.63 | 1.85e-15 | 3.07e-11 | Ectodysplasin A2 receptor | Apoptosis | ||
| CUL9 | 2.25 | 1.26e-15 | 3.07e-11 | Cullin-9 | Apoptosis | ||
| INPPD5 | 2.18 | 8.39e-14 | 4.66e-10 | Inositol Polyphosphate-5-Phosphatase | Apoptosis | ||
| TIGAR | 2.17 | 7.08e-10 | 8.87e-07 | Chromosome 1 open reading frame 5 | Apoptosis | ||
| E2F7 | 2.06 | 2.60e-13 | 1.05e-09 | E2F transcription factor 7 | Apoptosis | ||
| NCR3LG1 | 2.70 | 1.14e-10 | 1.94e-07 | Natural killer cell cytotoxicity receptor 3 ligand 1 | Immune response | ||
| LAMP3 | 2.26 | 4.39e-06 | 0.0011 | lysosomal-associated membrane protein 3 | Immune response | ||
| TREM2 | 2.12 | 2.78e-09 | 2.92e-11 | Triggering receptor expressed on myeloid cells | Immune response | ||
| FKBP5 | 2.04 | 6.66e-05 | 0.0082 | FK506 Binding Protein 5 | Immune response | ||
| POLH | 2.42 | 1.60e-14 | 1.24e-10 | Polymerase; DNA directed | Transcription | ||
| ARNTL | 2.40 | 2.65e-06 | 0.00080 | Aryl hydrocarbon receptor, nuclear translocator-like | Transcription | ||
| NFIL3 | 2.20 | 4.29e-05 | 0.0011 | Nuclear factor, interleukin 3 regulated | Transcription | ||
| ABCA12 | 4.73 | 8.87e-07 | 0.00034 | ATP-binding cassette sub-family A, member 12 | Metabolism | ||
| CEL | 4.64 | 3.68e-15 | 3,58e-11 | Carboxyl ester lipase | Metabolism | ||
| CA2 | 2.57 | 9.99e-10 | 1.18e-06 | Carbonic anhydrase II | Metabolism | ||
| SLC39A6 | 2.53 | 1.16e-10 | 2.00e-07 | Solute carrier family 39 | Metabolism | ||
| SPATA18 | 2.19 | 2.16e-12 | 6.98e-09 | Spermatogenesis associated 18 | Metabolism | ||
| P3H2 | 2.10 | 2.28e-09 | 2.46e-06 | Prolyl 3-Hydroxylase 2 | Metabolism | ||
| F3 | 2.09 | 0.00037 | 0.027 | Coagulation Factor III | Metabolism | ||
| GLS2 | 2.01 | 4.79e-14 | 3.1e10 | Glutaminase 2 | Metabolism | ||
| WDR63 | 2.84 | 7.16e-11 | 1.35e-07 | WD Repeat Domain 63 | Unknown | ||
| RN7SL519P | 2.05 | 0.00061 | 0.037 | Pseudogene | Unknown | ||
| Downregulated genes day2 versus baseline | |||||||
| SERPINB10 | -2.12 | 1.75e-06 | 0.00574 | Serpin peptidase inhibitor, clade B member 10 | Apoptosis | ||
| NR1D2 | -2.57 | 9.63e-06 | 0.00201 | Nuclear Receptor Subfamily 1, Group D, Member 2 | Transcription | ||
| NR1D1 | -2.29 | 0.00015 | 0.0142 | Nuclear Receptor Subfamily 1, Group D, Member 1 | Transcription | ||
| CIART | -2.38 | 9.14e-05 | 0.0103 | Circadian associated repressor of transcription | Transcription | ||
| HIST1H1A | -2.56 | 8.31e-07 | 4.18e-06 | Histone Cluster 1, H1a | Transcription | ||
| HIST1H1B | -2.04 | 6.70e-09 | 6.60e-06 | Histone Cluster 1, H1b | Transcription | ||
| HIST1H3J | -2.00 | 8.31e-07 | 0.00033 | Histone Cluster 1, H3j | Transcription | ||
| OXGR1 | -2.04 | 8.93e-05 | 0.010 | Oxoglutarate (Alpha-Ketoglutarate) Receptor | Cell signaling | ||
| PER3 | -2.76 | 1.77e-05 | 0.0032 | Period Circadian Clock | Metabolism | ||
| CYSLTR1 | -2.91 | 8.72e-06 | 0.0019 | Cysteinyl Leukotriene Receptor 1 | Cell structure | ||
| KIF20A | -2.05 | 4.65e-08 | 3.22e-05 | Kinesin Family Member 20A | Cell structure | ||
| PIK3C2G | -2.06 | 4.8e-06 | 0.00121 | Phosphatidylinositol-4-phosphate 3-kinase C2 domain-containing gamma polypeptide | Cell growth | ||
Abbreviations: FC = fold change; GO = gene ontology annotation
Gene expression related to mucositis grade
When we compared unsupervised gene expression profiles between NM and UM, we found that no genes were significantly differentially expressed in the blood. In contrast, in the biopsies, two genes of the major histocompatibility complex (MHC) Class II: HLA-DRB1 and HLA-DRB5 were significantly differentially expressed at the first two time points (Table 4). Patients with UM and CON expressed the same low level of HLA-DRB1 and HLA-DRB5, but patients with NM expressed significantly higher levels (Fig 1). The expression levels of HLA-DRB1 and HLA-DRB5 were independent of melphalan administration. Of the 10 CON subjects, one patient, CON09, showed high levels of HLA-DRB1 and HLA-DRB5 expression, similar to the levels observed in the NM group. We reopened the protocol and returned to the patient to reaffirm his health status. We found that subject CON09 had a mild case of psoriasis that was not reported at the baseline interview. An alternative splicing analysis revealed that patients with NM and the CON09 subject expressed a different isoform of HLA-DRB1 (NM_001243965) than that expressed by patients with UM (NM_002124.1). However, the difference in HLA-DRB5 expression between groups was not due to different isoforms.
Table 4. Genes altered in the buccal mucosa of patients with multiple myeloma that displayed mild/no mucositis (NM) compared to those that displayed ulcerative mucositis (UM).
| Gene symbol | FC | p-value | adjusted p-value | Qualified GO term | Function |
|---|---|---|---|---|---|
| NM versus UM day0, before melphalan | |||||
| HLA-DRB1 | 6.27 | 2.96e-07 | 0.00573 | Human Leukocyte / Major Histocompatibility Antigen Class II DRB1 beta chain | Immune response |
| HLA-DRB5 | 5.64 | 2.7e-07 | 0.00573 | Human Leukocyte / Major Histocompatibility Antigen Class II DRB5 beta chain | Immune response |
| NM versus UM day2 after melphalan | |||||
| HLA-DRB1 | 5.81 | 2.01e08 | 0.00039 | Human Leukocyte / Major Histocompatibility Antigen Class II DRB1 beta chain | Immune response |
| HLA-DRB5 | 5.56 | 1.98e09 | 7.69e05 | Human Leukocyte / Major Histocompatibility Antigen Class II DRB5 beta chain | Immune response |
FC = fold change; GO = gene ontology annotation.
Fig 1. Genes differentially expressed according to mucositis grade.
Expression of (a) HLA-DRB1 and (b) HLA-DRB5 genes in buccal mucosa biopsies taken at baseline (day0), two days (day2), and 21 days (day21) after high-dose melphalan therapy. Patients with mild/no mucositis (NM) express 6–8 fold more HLA-DRB1 and 4–5 fold more HLA-DRB5 than patients with ulcerative mucositis (UM). Melphalan treatment did not affect expression of HLA-DRB1 or HLA-DRB5 in either group. One healthy control (CON09) expressed the same high levels of HLA-DRB1 and HLA-DRB5 as those observed in the NM group. Subject CON09 was diagnosed with the autoimmune disease, psoriasis. Previous studies have reported that patients with psoriasis were 77% less likely to develop mucositis than patients without psoriasis [54].
Immunohistochemistry
In hematoxylin and eosin-stained specimens, no gross anatomical changes were observed in the epithelial or mesenchymal layers. In general, both the epithelium and stroma were represented in equal amounts. However, two specimens that were cut at a tangential angle that revealed only superficial layers were excluded from the histological analysis (MM15_1 and MM18_2). Generally, when present, cells that stained positive for HLA-DRB5 were localized in the lower part of the epithelial layer, near the basal membrane, around the papillae, and in the upper part of the lamina propria. Faint, diffuse HLA-DRB5 staining of the endothelium was not included in the assessment. Examples of high HLA-DRB5 expression/low-grade mucositis (MM01) and low HLA-DRB5 expression/severe-grade mucositis (MM18) are shown in Fig 2. A dotplot of the numbers of cells that stained positively for HLA-DRB5 is shown in Fig 3, and these findings supported the gene expression analysis.
Fig 2. Immunohistochemical analysis of HLA-DRB5 expression in oral mucosa biopsies.
Oral buccal mucosa (×20 magnification) staining shows HLA-DRB5 expression in the center region (a) High HLA-DRB5 expression is observed in the patient MM01 with mild mucositis. (b) Low HLA-DRB5 expression is observed in the patient MM18 with severe mucositis. Generally, when present, cells that stained positive for HLA-DRB5 are primarily localized in the lower part of the epithelium, near the basal membrane, around the papillae, and in the upper part of the submucosa in close proximity to the basal membrane. A weak, diffuse HLA-DRB5 staining of the endothelium is also visible. Normal tonsil tissue was included on the slide as a control; these appear identical in (a) and (b). The square insets highlight the morphology of one of the HLA-DRB5 expressing cells that displayed extensions, similar to those observed in dendritic cells.
Fig 3. The number of cells that stained positive for HLA-DRB5 in oral mucosa biopsies.
A dotplot shows the cell count performed on oral mucosa biopsies that had been immunostained to detect HLA-DRB5. Each biopsy was examined to determine hot spots of expression; then, this section was framed with a rectangle of 0.75×0.4 mm (0.3 mm2), and all distinctly stained cells within the rectangle were counted.
Gene expression in peripheral blood
Of the blood samples drawn from 10 patients with MM, we successfully performed gene expression profiles in 8 out of 10 drawn on day0 and day21, but only 2 out of 10 drawn on day2 (MM20 and MM05). Consequently, we performed an unsupervised global gene analysis of peripheral MNCs by comparing the MM day0 samples to CON samples (disease vs. healthy) and MM day0 samples versus MM day21 samples (before vs. after treatment). We found that two genes that encoded B-cell surface markers (CD22, CD200) were downregulated in MM day0 samples compared to CON samples, and these genes were further downregulated on day21, though the differences from day0 were not significant. The levels of CD22 and CD200 expression were independent of the mucositis grade.
Discussion
This study described the gene signature of buccal mucosa samples from patients with MM during HSCT. We found that this signature was dominated by altered expression of inflammatory and anti-apoptotic genes, but expression was independent of the presence of clinical mucositis. Furthermore, we identified a specific isoform of the immunomodulatory gene, HLA-DRB1, which may serve as a biomarker for mucositis severity.
The process that leads to mucositis is triggered immediately upon initiation of cancer therapy and before any visible macroscopic damage [10]. Moreover, eventual UM coincides with neutropenia, within 7–10 days of starting chemotherapy. Therefore, to avoid compromising patients with neutropenia, we acquired the second biopsy before the onset of neutropenia. With this approach, the biopsy was unlikely to comprise disintegrated tissue that, presumably, would be dominated by inflammatory mediators. Instead, we aimed to gain insight on the cellular processes that gave rise to the inflammatory state. Mucositis lasts for approximately 7 days, and then, it spontaneously resolves. We acquired the third biopsy on day 21, when the mucosa was fully restored.
Little is known about the effects of melphalan on normal epithelium, but in cancer cells, melphalan induces oxidative stress and upregulates a wide range of apoptosis-related genes [8, 29], consistent with our findings of EDA2R upregulation. EDA2R encodes a TNF receptor that mediates the activation of NF-κB and jun-N-terminal (JNK) pathways, which lead to caspase-initiated apoptosis. Previous studies reported that these pathways were activated in buccal mucosa of patients with various cancers that received HSCT [30–32], and in gastrointestinal mucosa of patients treated with 5-fluoruracil [33]. In contrast, we found upregulated expression of INPP5D, which encodes a membrane protein in hematopoietic cells. The INPP5D protein negatively regulates JNK signaling and limits Fas-FasL-induced apoptosis in T-cells found at mucosal surfaces [34]. In addition, we identified five genes involved in suppressing the pivotal p53 apoptotic pathway. Four of these were upregulated: MDM2, CUL9, E2F7, and TIGAR; and one was downregulated: SERPINB10. The MDM2 gene encodes a protein ligase that ubiquitinates p53, and thus, inhibits p53-mediated cell cycle arrest and apoptosis. A previous gene expression study on three patients with acute myeloid leukemia used the same time intervals between biopsies that we used, and they found MDM2 upregulation [35]. In an array of studies, p53 has been identified as a key regulator of apoptosis, which leads to mucositis [11, 12, 36]. However, the gene alterations associated with apoptosis observed in our study, including EDA2R, did not depend on the level of clinical mucositis.
We found several genes related to transcription that were altered to favor DNA repair. For example, POLH was upregulated; POLH encodes a specialized DNA polymerase that accurately replicates UV-damaged DNA. Conversely, members of the histone cluster family (e.g., HIST1H1A), NR1D1, and NR1D2 were downregulated. These results implied that defense against apoptosis and DNA damage was a central objective in the initial stage of mucositis. Importantly, this objective was independent of the mucositis grade, which implies that other factors must be involved in distinguishing UM and NM.
The immune response was activated at an early stage, through the upregulation of TREM2 and LAMP3 on day2. Both these genes encode DC membrane proteins that contribute to T-cell activation and mucosal inflammation. The LAMP3 gene is specifically expressed in mature DCs [37, 38]. TREM2, which is expressed on both DCs and macrophages, can bind and phagocytose yeasts, Gram positive bacteria, and Gram negative bacteria [39], which are commonly present in the oral cavity [40]. Generally, DCs are potent antigen-presenting cells that respond to microbial exposure by secreting abundant cytokines; e.g., IL-12 and type I interferon. In turn, IL-12 mobilizes natural killer (NK) cells. The NCR3LG1 gene, which encodes a ligand that triggers NK cells, was also upregulated [41]. Several studies have shown that an important aspect of mucositis pathology is the thinning of the epithelium, in combination with changes in the composition and concentration of the oral and gastrointestinal microbiota [13, 42, 43]. Our results confirmed the notion that the host immune response towards the microbiome played a dominant role, early in mucositis pathogenesis; however, these responses were not associated with mucositis severity.
Among several genes associated with metabolism, ABCA12 and CEL were upregulated 4.7-fold and 4.6-fold, respectively, on day2. ABCA12 encodes a membrane transporter protein primarily involved in the keratinocyte lipid-barrier that maintains homeostasis in the epidermis [44]. To the best of our knowledge, no previous study has described a role for ABCA12 in the mucosa, but it most likely performs a similar function of barrier protection. CEL encodes a lipase with multiple functions in lipid metabolism; it is also expressed in macrophages [45]. The expression levels of both these genes were unaffected by clinical mucositis.
Gene alterations associated with clinical mucositis grade
When we compared the gene expression profiles between the six patients with NM and the four patients with UM, we found two genes that were more highly expressed in NM patients: HLA-DRB1 and HLA-DRB5 (Fig 1). The HLA-DRB1 and HLA-DRB5 genes are related members of the MHC Class II family, located on chromosome 6p21.32. They encode surface proteins that are almost exclusively expressed on specialized antigen presenting cells, including macrophages, B-cells, and DCs or Langerhans cells [46, 47]. These surface receptors function as a ligand for the T-cell receptor, and their primary function is to capture potentially foreign antigens on the cell surface and to present them for recognition by CD4+ T-cells [48]. Thus, they form a communication between the innate and adaptive immune systems, and determine whether to bring forth resistance or tolerance, in addition to taking up and processing dying cells [49].
Several pharmacogenomic GWAS studies were recently performed on drug toxicity, which showed that HLA Class I and II paralogs were associated with toxicity [50] or inflammatory mucosal conditions [51, 52]. Even more interestingly, certain HLA-DRB1 alleles (HLA-DRB1*15) have been detected in patients with MM that were exposed to bisphosphonates and developed osteonecrosis of the jaw [53].
We conducted a search for alternatively spliced variants of HLA-DRB1 and HLA-DRB5 and found two isoforms of HLA-DRB1 (NM_001243965 and NM_002124). According to the UCSC genome browser, NM_001243965 harbors six exons, and NM_002124 harbors an extended isoform within seven exons. We found that patients with NM expressed the longer transcript variant of HLA-DRB1 (NM_002124), and patients with UM and healthy subjects (CON) expressed the shorter variant (NM_001243965). No splice variant was found for HLA-DRB5; however, that gene was expressed at different levels. We found that HLA-DRB5 was expressed 4.5 to 5 times more frequently in the NM group than in the UM group, in the first two of three biopsies. These findings were confirmed in an immunohistochemical analysis of HLA-DRB5 in the biopsies (Fig 3). Furthermore, cells that stained positively for HLA-DRB5 were localized primarily in the epithelium and submucosa, relatively close to the basal membrane, and these cells displayed a morphology similar to DCs (Langerhans cells).
In contrast to the other healthy subjects, CON09 (a patient with psoriasis) expressed the long transcript variant of HLA-DRB1 and a high level of HLA-DRB5 protein, similar to patients in the NM group. Previous reports have indicated that patients with psoriasis are 77% less likely to develop mucositis [10, 54]. Psoriasis is an auto-inflammatory skin disorder with reduced apoptosis. It is known that patients with psoriasis express certain HLA-DRB1 alleles [55]. Our results suggested that mechanisms related to inflammatory and/or apoptotic pathways may be common in psoriasis and low-grade mucositis in MM. In addition, ABCA12 expression was upregulated in CON09 compared to the other healthy subjects. Previous gene expression studies on patients with psoriasis confirmed this finding [56].
The two recent GWAS studies on patients that received HSCT identified SNPs near the locus of MMP and other genes related to inflammation, but none related to HLA-DR [17, 18]. Other recent studies found a major role for MMPs in mucositis pathology [57]. We did not find any changes of that nature in our material. The phenotypes described in our study may provide additional information to guide future GWAS studies [50]. Recent studies have shown that induction therapy with immune modulating agents reduced the frequency and severity of mucositis [58]; our results may provide additional knowledge to elucidate the development of those therapies.
In peripheral blood, we did not find any differences in gene expression between NM and UM groups, at any time point. However, among all patients with MM, CD22 and CD200 were downregulated on day0 compared to controls. Both these genes encode cell membrane glycoproteins of the immunoglobulin superfamily. CD200 is expressed in multiple cell types, including B-cells, T-cell subsets, DCs and endothelial cells. In contrast, CD22 is exclusively expressed on mature B-cell lineages [59, 60]. Low CD200 expression has been linked to prolonged survival among patients with MM [61]. In our cohort, the lowest CD200 expression levels were observed among patients with UM; however, the levels were not significantly different between UM and NM groups.
Study strengths and limitations
There was some concern that breaking the mucosal barrier by taking a biopsy during chemotherapy might lead to potential fatal infections. We could reject this concern, because none of our patients experienced any infection related to the biopsy; only mild discomfort was reported. This finding was also reported in previous studies [15, 35, 36, 62]. Therefore, we concluded that our method would be safe for patients undergoing HSCT, provided that the second biopsy is taken before the onset of neutropenia. The major limitation of the study was the low number of subjects. We designed the study to identify genes that were altered by more than 2-fold between time points, with a false discovery rate of less than 0.05% and a power of 90%. However, because our method of harvesting human mucosa during high-dose melphalan treatment was controversial, we sought to include the least possible number of patients required to draw valid conclusions. However, we recognize that the power of this study was set to estimate any fold-changes above two, and false negative findings may be concealed. Consequently, we did not expect to elucidate the full, true picture; nevertheless, we brought to light some important biological associations, which have motivated us to continue this research and confirm the results in a larger cohort.
Conclusions
Currently, there is a great need to develop a clinically applicable method for identifying potential susceptibility to toxicity among patients before treatment initiation. We found that patients with NM displayed upregulations of HLA-DR1 and HLA-DRB5 compared to patients with UM and healthy individuals. These genes encode proteins expressed on the surface of antigen presenting cells in mucosa, which suggested that the immune response might play a major role as a primary effector in UM. Indeed, the results suggested that expression of a certain isoform of HLA-DRB1 might diminish the inflammatory response to melphalan toxicity. However, because the levels of HLA-DRB1 and HLA-DRB5 expression were constant throughout treatment, this isoform may serve as a predictor of UM. The findings in this study were based on a small number of samples, and thus, our results must be validated in a larger patient cohort.
Acknowledgments
We acknowledge the laboratory assistance of the laboratory technicians Ann-Maria Jensen and Helle Hoeholt, Department of Hematology, Aalborg University Hospital, Denmark. We acknowledge the staff of the Department of Maxillofacial Surgery, Aalborg University Hospital, Denmark for assistance in the collection of biologic material; in particular the nurses Birthe Kaas Soerensen, Annette Holt and Karina Vinther Olesen for assistance during the bedside harvesting of mucosa biopsies.
Data Availability
All CEL files are available NCBI Gene Expression Omnibus (GEO) repository under the accession number GSE81979. The following link has been created to allow review, while it remains in private status: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=mfaryscwprqvnmx&acc=GSE81979.
Funding Statement
Support was provided by: Det Obelske Familiefond [http://www.obel.com/]; The Health Scientific Research Foundation of The North Denmark Region [http://www.aalborguh.rn.dk/Forskning/Forskning-Nyhedsliste-Aalborg/Nyhed?id=c0b95ee0-4e0b-4fdd-bd92-5da0cb92a63e]; Herta Christensen Foundation [http://cvrapi.dk/virksomhed/dk/herta-christensens-fond/11681409]; Doctor Henrik Kopps Grant [http://www.laeger.dk/portal/page/portal/LAEGERDK/Laegerdk/Om%20L%C3%A6geforeningen/L%C3%A6geforeningen%20Nordjylland/Kontakt%20os/Speciall%C3%A6ge%20Heinrich%20Kopps%20Legat]. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References
- 1.Moreau P, Attal M, Facon T. Frontline therapy of multiple myeloma. Blood 2015;125:3076–84. 10.1182/blood-2014-09-568915 [DOI] [PubMed] [Google Scholar]
- 2.Blijlevens N, Schwenkglenks M, Bacon P, D´Addio A, Einsele H, Maertens J, et al. Prospective oral mucositis audit: oral mucositis in patients receiving high-dose melphalan or BEAM conditioning chemotherapy. European Blood and Marrow Transplantation Mucositis Advisory Group. J Clin Oncol 2008;26:1519–25. 10.1200/JCO.2007.13.6028 [DOI] [PubMed] [Google Scholar]
- 3.Elting LS, Cooksley C, Chambers M, Cantor SB, Manzullo E, Rubenstein EB. The burdens of cancer therapy. Clinical and economic outcomes of chemotherapy-induced mucositis. Cancer 2003;98:1531–9. 10.1002/cncr.11671 [DOI] [PubMed] [Google Scholar]
- 4.Chen Y-K, Hou H, Chow J-M, Chen Y-C, Hsueh P-R, Tien H-F. The impact of oral herpes simplex virus infection and candidiasis on chemotherapy-induced oral mucositis among patients with hematological malignancies. Eur J Clin Microbiol Infect Dis 2011;30:753–9. 10.1007/s10096-010-1148-z [DOI] [PubMed] [Google Scholar]
- 5.McCann S, Schwenkglenks M, Bacon P, D´Addio A, Einsele H, Maertens J, et al. The Prospective Oral Mucositis Audit: relationship of severe oral mucositis with clinical and medical resource use outcomes in patients receiving high-dose melphalan or BEAM-conditioning chemotherapy and autologous SCT. Bone Marrow Transplant 2009;43:141–7. 10.1038/bmt.2008.299 [DOI] [PubMed] [Google Scholar]
- 6.Lalla RV, Bowen J, Barasch A, Elting L, Epstein J, Keefe DM, et al. MASCC/ISOO clinical practice guidelines for the management of mucositis secondary to cancer therapy. Cancer 2014;120:1453–61. 10.1002/cncr.28592 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gibson RJ, Keefe DMK, Lalla RV, Bateman E, Blijevens N, Fijlstra M, et al. Systematic review of agents for the management of gastrointestinal mucositis in cancer patients. Support Care Cancer 2013; 21:313–26. 10.1007/s00520-012-1644-z [DOI] [PubMed] [Google Scholar]
- 8.Gkotzamanidou M, Sfikakis PP, Kyrtopoulos SA, Bamia C, Dimopoulos MA, Soulitis VL. Cromatin structure, transcriptional activity and DNA repair efficiency affects the outcome of chemotherapy in multiple myeloma. Br J Cancer 2014;23:1293–304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Nath CE, Shaw PJ, Trotman J, Zeng L, Duffull SB, Hegarty G, et al. Population pharmacokinetics of melphalan in patients with multiple myeloma undergoing high dose therapy. Br J Clin Pharmacol 2010;69:484–97. 10.1111/j.1365-2125.2010.03638.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sonis ST. The pathobiology of mucositis. Nat Rev Cancer 2004;4:277–84. 10.1038/nrc1318 [DOI] [PubMed] [Google Scholar]
- 11.Bowen JM, Gibson RJ, Keefe DM, Cummins AG. Cytotoxic chemotherapy upregulates pro-apoptotic Bax and Bak in the small intestine of rats and humans. Pathology 2005;37:56–62. [DOI] [PubMed] [Google Scholar]
- 12.Bowen JM, Gibson RJ, Cummins AG, Keefe DMK. Intestinal mucositis: The role of the Bcl-2 family, p53 and caspases in chemotherapy-induced damage. Support Care Cancer 2006;14:713–31. 10.1007/s00520-005-0004-7 [DOI] [PubMed] [Google Scholar]
- 13.Logan RM, Gibson RJ, Bowen JM, Stringer AM, Sonis ST, Keefe DMK. Characterisation of mucosal changes in the alimentary tract following administration of irinotecan: implications for the pathobiology of mucositis. Cancer Chemother Pharmacol 2008;62:33–41. 10.1007/s00280-007-0570-0 [DOI] [PubMed] [Google Scholar]
- 14.Bowen JM, Gibson RJ, Tsykin A, Stringer AM, Logan RM, Keefe DMK. Gene expression analysis of multiple gastrointestinal regions reveals activation of common cell regulatory pathways following cytotoxic chemotherapy. Int J Cancer 2007;121:1847–56. 10.1002/ijc.22895 [DOI] [PubMed] [Google Scholar]
- 15.Logan RM, Gibson RJ, Sonis ST, Keefe DMK. Nuclear factor-kappaB (NF-kappaB) and cyclooxygenase-2 (COX-2) expression in the oral mucosa following cancer chemotherapy. Oral Oncol 2007;43:395–401. 10.1016/j.oraloncology.2006.04.011 [DOI] [PubMed] [Google Scholar]
- 16.Stringer AM, Al-Dasooqi N, Bowen JM, Tan TH, Radzuan M, Logan RM, et al. Biomarkers of chemotherapy-induced diarrhoea: A clinical study of intestinal microbiome alterations, inflammation and circulating matrix metalloproteinases. Support Care Cancer 2013;21:1843–52. 10.1007/s00520-013-1741-7 [DOI] [PubMed] [Google Scholar]
- 17.Sonis S, Antin J, Tedaldi M, Alterovitz G. SNP-based Bayesian networks can predict oral mucositis risk in autologous stem cell transplant recipients. Oral Dis 2013;19:721–7. 10.1111/odi.12146 [DOI] [PubMed] [Google Scholar]
- 18.Coleman EA, Lee JY, Erickson SW, Goodwin JA, Sanathkumar N, Raj VR, et al. GWAS of 972 autologous stem cell recipients with multiple myeloma identifies 11 genetic variants associated with chemotherapy-induced oral mucositis. Support Care Cancer 2014;23:841–9. 10.1007/s00520-014-2406-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rajkumar SV, Dimopoulos MA, Palumbo A, Blade J, Merlini G, Mateos MV, et al. International myeloma group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol 2014;15:e538–48. 10.1016/S1470-2045(14)70442-5 [DOI] [PubMed] [Google Scholar]
- 20.Quinn B, Stone R, Uhlenhopp M, McCann S, Blijlevens N. Ensuring accurate oral mucositis assessment in the European Group for Blood and Marrow Transplantation Prospective Oral Mucositis Audit. Eur J Oncol Nurs 2007;11:s10–11. 10.1016/S1462-3889(07)70003-0 [DOI] [PubMed] [Google Scholar]
- 21.U:S: Department of Health and Human Services, National Institutes of health, National Cancer Institute; Common terminology criteria for adverse events (CTCAE)http://evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03_2010-06-14_QuickReference_5x7.pdf [Google Scholar]
- 22.Lee MLT, Whitmore GA Power and sample size for DNA microarray studies. Statist Med 2002;21:3543–70. [DOI] [PubMed] [Google Scholar]
- 23.R Core Team. A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, 2012,Vienna, Austria: http://R-project.org [Google Scholar]
- 24.Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, et al. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 2004;5:R80 10.1186/gb-2004-5-10-r80 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc 1995;57:289–300. [Google Scholar]
- 26.Dai M, Wang P, Boyd AD, Kostov G, Athey B, Jones EG, et al. Evolving gene/transcript definitions significantly alter the interpretation of GeneChip data. Nucleic Acids Res 2005;33:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Smyth G. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol 2004;3:1–25. [DOI] [PubMed] [Google Scholar]
- 28.Durie BGM, Harousseau JL, Miguel JS, Bladé J, Barlogie B, Anderson K, et al. International uniform response criteria for multiple myeloma. Leukemia 2006;20:1467–73. 10.1038/sj.leu.2404284 [DOI] [PubMed] [Google Scholar]
- 29.Podar K, Raab MS, Tonon G, Sattler M, Barilà D, Zhang J, et al. Up-regulation of c-Jun inhibits proliferation and induces apoptosis via caspase-triggered c-Abl cleavage in human multiple myeloma. Cancer Res 2007;67:1680–8. 10.1158/0008-5472.CAN-06-1863 [DOI] [PubMed] [Google Scholar]
- 30.Fall-Dickson JM, Ramsay ES, Castro K, Woltz P, Sportés C. Oral mucositis-related oropharyngeal pain and correlative tumor necrosis factor-alpha expression in adult oncology patients undergoing hematopoietic stem cell transplantation. Clin Ther 2007;29 Suppl:2547–61. [DOI] [PubMed] [Google Scholar]
- 31.Lalla R V, Pilbeam CC, Walsh SJ, Sonis ST, Keefe DMK, Peterson DE. Role of the cyclooxygenase pathway in chemotherapy-induced oral mucositis: a pilot study. Support Care Cancer 2010;18:95–103. 10.1007/s00520-009-0635-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Logan RM, Gibson RJ, Sonis ST, Keefe DMK. Nuclear factor-kappaB (NF-kappaB) and cyclooxygenase-2 (COX-2) expression in the oral mucosa following cancer chemotherapy. Oral Oncol 2007;43:395–401. 10.1016/j.oraloncology.2006.04.011 [DOI] [PubMed] [Google Scholar]
- 33.Yeoh ASJ, Bowen JM, Gibson RJ, Keefe DMK. Nuclear factor kappaB (NFkappaB) and cyclooxygenase-2 (Cox-2) expression in the irradiated colorectum is associated with subsequent histopathological changes. Int J Radiat Oncol Biol Phys 2005;63:1295–1303. 10.1016/j.ijrobp.2005.04.041 [DOI] [PubMed] [Google Scholar]
- 34.Srivastava N, Kerr WG. Role of inositol poly-phosphatases and their targets in T cell biology. Front Immunol 2013;4:1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Mougeot JLC, Bahrani-Mougeot FK, Lockhart PB, Brennan MT. Microarray analyses of oral punch biopsies from acute myeloid leukemia (AML) patients treated with chemotherapy. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2013;112:446–52. [DOI] [PubMed] [Google Scholar]
- 36.Keefe DM, Brealey J, Goland GJ, Cummins AG. Chemotherapy for cancer causes apoptosis that precedes hypoplasia in crypts of the small intestine in humans. Gut 2000;47:632–7. 10.1136/gut.47.5.632 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Correale C, Genua M, Vetrano S, Mazzini E, Martinoli C, Spinelli A, et al. Bacterial sensor triggering receptor expressed on myeloid cells-2 regulates the mucosal inflammatory response. Gastroenterol 2013;144:346–56.e3. [DOI] [PubMed] [Google Scholar]
- 38.Saint-Vis B de, Vincent J, Vandenabeele S, Vanbervliet B, Pin JJ, Ait-Yahia S, et al. A novel lysosome-associated membrane glycoprotein, DC-LAMP, induced upon DC Maturation, is transiently expressed in MHC Class II compartment. Immunity 1998;9:325–36. [DOI] [PubMed] [Google Scholar]
- 39.N’Diaye EN, Branda CS, Branda SS, Nevarez L, Colonna M, Lowell C, et al. TREM-2 (triggering receptor expressed on myeloid cells 2) is a phagocytic receptor for bacteria. J Cell Biol 2009;184:215–23. 10.1083/jcb.200808080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Aas JA, Paster BJ, Stokes LN, Olsen I, Dewhirst FE. Defining the Normal Bacterial Flora of the Oral Cavity. J Clin Microbiol 2005;43:5721–32. 10.1128/JCM.43.11.5721-5732.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Schlecker E, Fiegler N, Arnold A, Altevogt P, Rose-John S, Moldenhauer G, et al. Metalloprotease-mediated tumor cell shedding of B7-H6, the ligand of the natural killer cell-activating receptor NKp30. Cancer Res 2014;74:3429–40. 10.1158/0008-5472.CAN-13-3017 [DOI] [PubMed] [Google Scholar]
- 42.Fijlstra M, Ferdous M, Koning AM, Rings EHHM, Harmsen HJM, Tissing WJE. Substantial decreases in the number and diversity of microbiota during chemotherapy-induced gastrointestinal mucositis in a rat model. Support Care Cancer 2014;23:1513–22. 10.1007/s00520-014-2487-6 [DOI] [PubMed] [Google Scholar]
- 43.Bergmann OJ. Alterations in oral microflora and pathogenesis of acute oral infections during remission-induction therapy in patients with acute myeloid leukaemia. Scand J Infect Dis 1991;23:355–66. [DOI] [PubMed] [Google Scholar]
- 44.Akiyama M. The roles of ABCA12 in epidermal lipid barrier formation and keratinocyte differentiation. Biochim Biophys Acta 2014;1841:435–40. 10.1016/j.bbalip.2013.08.009 [DOI] [PubMed] [Google Scholar]
- 45.Hui DY, Howles PN. Carboxyl ester lipase: structure-function relationship and physiological role in lipoprotein metabolism and atherosclerosis. J Lipid Res2002;43:2017–30. [DOI] [PubMed] [Google Scholar]
- 46.Neefjes J, Jongsma MLM, Paul P, Bakke O. Towards a systems understanding of MHC class I and MHC class II antigen presentation. Nat Rev Immunol 2011;11:823–36. 10.1038/nri3084 [DOI] [PubMed] [Google Scholar]
- 47.Hershberg RM, Framson PE, Cho DH, Lee LY, Kovats S, Beitz J, et al. Intestinal epithelial cells use two distinct pathways for HLA class II antigen processing. J Clin Invest 1997;100:204–15. 10.1172/JCI119514 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Parkin J, Cohen B. An overview of the immune system. Lancet 2001;357:1777–89. 10.1016/S0140-6736(00)04904-7 [DOI] [PubMed] [Google Scholar]
- 49.Steinman RM, Hemmi H. Dendritic cells: translating innate to adaptive immunity. Curr Top Microbiol Immunol 2006;311:17–58. In: Pulendran B, Ahmed R, editors. From Innate Immunity to immunological memory. Springer Verlag http://www.springer.com/978-3-540-32635-9 [DOI] [PubMed]
- 50.Low SK, Takahashi A, Mushiroda T, Kubo M. Genome-wide association study: a usefull tool to identify common genetic variants associated with drug toxicity and efficacy in cancer pharmacogenomics. Clin Cancer Res 2014;15:2541–52. [DOI] [PubMed] [Google Scholar]
- 51.Goyette P, Boucher G, Mallon D, Ellinghaus E, Jostins L, Huang H, et al. High-density mapping of the MHC identifies a shared role for HLA-DRB1*01:03 in inflammatory bowel diseases and heterozygous advantage in ulcerative colitis. Nat Genet 2015;47:172–9. 10.1038/ng.3176 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Parkes M, Cortes A, van Heel DA, Brown MA. Genetic insights into common pathways and complex relationships among immune-mediated diseases. Nat Rev Genet 2013;14:661–73. 10.1038/nrg3502 [DOI] [PubMed] [Google Scholar]
- 53.Stockmann P, Nkenke E, Englbrecht M, Schlittenbauer T, Wehrhan F, Rauh C, et al. Major histocompatibility complex class II polymorphisms are associated with the development of anti-resorptive agent-induced osteonecrosis of the jaw. J Cranio-Maxillofacial Surg 2013;41:71–75. [DOI] [PubMed] [Google Scholar]
- 54.Chen E. Impact of Addisons disease and psoriasis on the frequency of oral mucositis. Masters thesis, Harward School of Dental Medicine 2003, private copy.
- 55.Vasku A, Bienertova-Vasku J, Izakovicova-Holla L, Goldbergova MP, Kozacikova Z, Splichal Z, et al. Polymorphisms in HLA-related genes and psoriasis heredity in patients with psoriasis. Int J Dermatol 2013;52:960–5. 10.1111/ijd.12213 [DOI] [PubMed] [Google Scholar]
- 56.ABCA12 –Psoriasis leasion and non-lesional skin http://www.ncbi.nlm.nih.gov/geoprofiles/100696959
- 57.Al-Dasooqi N, Gibson RJ, Bowen JM, Keefe DM. Matrix metalloproteinases: key regulators in the pathogenesis of chemotherapy-induced mucositis? Cancer Chemother Pharmacol 2009;64:1–9. 10.1007/s00280-009-0984-y [DOI] [PubMed] [Google Scholar]
- 58.Fleming S, Harrison SJ, Blombery P, Joyce T, Stokes K, Seymour JF, et al. The choice of multiple myeloma induction therapy affects the frequency and severity of oral mucositis after melphalan-based autologous stem cell transplantation. Clin Lymphoma, Myeloma Leuk 2014;14:291–6. [DOI] [PubMed] [Google Scholar]
- 59.Aref S, Azmy E, El-Gilany AH. Upregulation of CD200 is associated with regulatory T cell expansion and disease progression in multiple myeloma. Hematol Oncol 2007;25:127–131. 10.1002/hon.2206 [DOI] [PubMed] [Google Scholar]
- 60.Dörner T, Shock A, Smith KGC. CD22 and Autoimmune Disease. Int Rev Immunol 2012;31:363–378. 10.3109/08830185.2012.709890 [DOI] [PubMed] [Google Scholar]
- 61.Douds JJ, Long DJ, Kim AS, Li S. Diagnostic and prognostic significance of CD200 expression and its stability in plasma cell myeloma. J Clin Pathol 2014;67:792–6. 10.1136/jclinpath-2014-202421 [DOI] [PubMed] [Google Scholar]
- 62.Lalla R V, Pilbeam CC, Walsh SJ, Sonis ST, Keefe DMK, Peterson DE. Role of the cyclooxygenase pathway in chemotherapy-induced oral mucositis: a pilot study. Support Care Cancer 2014;18:95–103. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All CEL files are available NCBI Gene Expression Omnibus (GEO) repository under the accession number GSE81979. The following link has been created to allow review, while it remains in private status: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?token=mfaryscwprqvnmx&acc=GSE81979.