cancers Article Succinate Pathway in Head and Neck Squamous Cell Carcinoma: Potential as a Diagnostic and Prognostic Marker Ximena Terra 1,† , Victoria Ceperuelo-Mallafré 2,3,† , Carla Merma 4 , Ester Benaiges 2,3,5 , Ramon Bosch 6 , Paola Castillo 7 , Joan Carles Flores 4 , Xavier León 8 , Izaskun Valduvieco 9 , Neus Basté 10 , Marina Cámara 11 , Marylène Lejeune 6 , Josep Gumà 12 , Joan Vendrell 2,3,5 , Isabel Vilaseca 13,14 , Sonia Fernández-Veledo 2,3, *,‡ and Francesc Xavier Avilés-Jurado 3,13,14, * 1 MoBioFood Research Group, Biochemistry and Biotechnology Department, Universitat Rovira i Virgili, Campus Sescel·lades, 43007 Tarragona, Spain; ximena.terra@urv.cat 2 Department of Endocrinology and Nutrition, Institut d’Investigació Sanitària Pere Virgili (IISPV), Hospital Universitari de Tarragona Joan XXIII, 43005 Tarragona, Spain; victoria.ceperuelo@urv.cat (V.C.-M.); ester.benaiges@urv.cat (E.B.); jvortega2002@gmail.com (J.V.) 3 Centro de Investigación Biomédica en Red de Diabetes y Enfermedades Metabólicas (CIBERDEM), 28029 Madrid, Spain 4 Otorhinolaryngology Head-Neck Surgery Department, Hospital Universitari de Tarragona Joan XXIII, Insitut d’Investigació Sanitària Pere Virgili, Universitat Rovira i Virgili, 43005 Tarragona, Spain; cvmerma.hj23.ics@gencat.cat (C.M.); jcflores.hj23.ics@gencat.cat (J.C.F.) 5 School of Medicine, Universitat Rovira i Virgili, 43201 Reus, Spain 6 Pathology Department, Plataforma de Estudios Histológicos, Citológicos y de Digitalización, Hospital de Tortosa Verge de la Cinta, Institut d’Investigació Sanitària Pere Virgili (IISPV), URV,  43500 Tortosa, Spain; rbosch.ebre.ics@gencat.cat (R.B.); mlejeune.ebre.ics@gencat.cat (M.L.)  7 Pathology Department, Hospital Clínic de Barcelona, IDIBAPS, 08036 Barcelona, Spain; PCASTILL@clinic.cat 8 Otorhinolaryngology Head-Neck Surgery Department, Hospital de la Santa Creu i Sant Pau and Networking Citation: Terra, X.; Ceperuelo-Mallafré, V.; Merma, C.; Benaiges, E.; Bosch, R.; Research Center on Bioengineering, Biomaterials and Nanomedicine (CIBER-BBN, MICINN, ISCIII), Universitat Autònoma de Barcelona, 08041 Barcelona, Spain; xleon@santpau.cat Castillo, P.; Flores, J.C.; León, X.; 9 Radiation Oncology Department, Hospital Clínic de Barcelona, 08036 Barcelona, Spain; ivalduvi@clinic.cat Valduvieco, I.; Basté, N.; et al. Succinate 10 Oncology Department, IDIBAPS, Hospital Clínic de Barcelona, 08036 Barcelona, Spain; baste@clinic.cat Pathway in Head and Neck Squamous 11 Maxillofacial Department, Hospital Clínic de Barcelona, 08036 Barcelona, Spain; camara@clinic.cat Cell Carcinoma: Potential as a 12 Oncology Department, Institut d’Investigació Sanitària Pere Virgili (IISPV), Hospital Sant Joan de Reus, Diagnostic and Prognostic Marker. 43204 Reus, Spain; jguma@grupsagessa.cat 13 Otorhinolaryngology Department, UB, IDIBAPS, Hospital Clínic de Barcelona, 08036 Barcelona, Spain; Cancers 2021, 13, 1653. https:// doi.org/10.3390/cancers13071653 ivila@clinic.cat 14 Head Neck Clínic, Agència de Gestió d’Ajuts Universitaris i de Recerca (AGAUR), Academic Editor: Patrick J. Schuler 2017-SGR-01581 Barcelona, Spain * Correspondence: sonia.fernandez@iispv.cat (S.F.-V.); faviles@clinic.cat (F.X.A.-J.) † Two authors contributed equally to this work. Received: 18 February 2021 ‡ Current address: Unitat de Recerca, Institut d’Investigació Sanitària Pere Virgili, Hospital Universitari de Accepted: 24 March 2021 Tarragona Joan XXIII, C/Mallafré Guasch 4, 43007 Tarragona, Spain. Published: 1 April 2021 Simple Summary: Emerging evidence points to succinate as an important oncometabolite in cancer Publisher’s Note: MDPI stays neutral development; however, the contribution of the succinate-SUCNR1 axis to cancer progression remains with regard to jurisdictional claims in unclear. Head and neck squamous cell carcinoma (HNSCC) is associated with disease and treatment- published maps and institutional affil- related morbidity so there is an urgent need for innovation in treatment and diagnosis practices. Our iations. aim was to evaluate the potential of the succinate-related pathway as a diagnostic and prognostic biomarker in HNSCC. The circulating succinate levels are increased in HNSCC, being a potential noninvasive biomarker for HNSCC diagnosis. Moreover, the succinate receptor (SUCNR1) and genes related to succinate metabolism, which are predominantly expressed in the tumoral mucosa Copyright: © 2021 by the authors. as compared with healthy tissue, are positively associated with plasma succinate. Remarkably, we Licensee MDPI, Basel, Switzerland. found that SUCNR1 and SDHA expression levels predict prognosis. This article is an open access article distributed under the terms and conditions of the Creative Commons Abstract: Head and neck squamous cell carcinoma (HNSCC) is characterized by high rates of Attribution (CC BY) license (https:// mortality and treatment-related morbidity, underscoring the urgent need for innovative and safe creativecommons.org/licenses/by/ treatment strategies and diagnosis practices. Mitochondrial dysfunction is a hallmark of cancer and 4.0/). can lead to the accumulation of tricarboxylic acid cycle intermediates, such as succinate, which Cancers 2021, 13, 1653. https://doi.org/10.3390/cancers13071653 https://www.mdpi.com/journal/cancers Cancers 2021, 13, 1653 2 of 16 function as oncometabolites. In addition to its role in cancer development through epigenetic events, succinate is an extracellular signal transducer that modulates immune response, angiogenesis and cell invasion by activating its cognate receptor SUCNR1. Here, we explored the potential value of the circulating succinate and related genes in HNSCC diagnosis and prognosis. We determined the succinate levels in the serum of 66 pathologically confirmed, untreated patients with HNSCC and 20 healthy controls. We also surveyed the expression of the genes related to succinate metabolism and signaling in tumoral and nontumoral adjacent tissue and in normal mucosa from 50 patients. Finally, we performed immunohistochemical analysis of SUCNR1 in mucosal samples. The results showed that the circulating levels of succinate were higher in patients with HNSCC than in the healthy controls. Additionally, the expression of SUCNR1, HIF-1α, succinate dehydrogenase (SDH) A, and SDHB was higher in the tumor tissue than in the matched normal mucosa. Consistent with this, immunohistochemical analysis revealed an increase in SUCNR1 protein expression in tumoral and nontumoral adjacent tissue. High SUCNR1 and SDHA expression levels were associated with poor locoregional control, and the locoregional recurrence-free survival rate was significantly lower in patients with high SUCNR1 and SDHA expression than in their peers with lower levels (77.1% [95% CI: 48.9–100.0] vs. 16.7% [95% CI: 0.0–44.4], p = 0.018). Thus, the circulating succinate levels are elevated in HNSCC and high SUCNR1/SDHA expression predicts poor locoregional disease-free survival, identifying this oncometabolite as a potentially valuable noninvasive biomarker for HNSCC diagnosis and prognosis. Keywords: oncometabolite; succinate; succinate receptor; metabolism; head and neck cancer; treatment; prognosis 1. Introduction Head and neck squamous cell carcinoma (HNSCC) comprise a significant portion of the global cancer burden being the eighth most common cancer worldwide by both incidence and mortality [1,2]. The treatment approaches depend on the disease stage and the site of the cancer, and include radiotherapy, surgery, surgery or chemoradiotherapy, or a combination of the three [3]. The treatment modalities of patients with head and neck cancer has undergone considerable transformation in the last decade, including im- munotherapy, targeted therapy (small molecule inhibitors or antibodies), or combined modality treatments [3]. There are no prognostic factors currently available that can effi- ciently predict treatment outcomes. Accordingly, the identification of molecular markers that predict response to therapy would be an important milestone in head and neck oncol- ogy and would aid in the development of personalized treatments to maximize survival while minimizing morbidity. It is now recognized that cancer metabolism goes well beyond simple tumor prolifera- tion and survival with the identification and characterization of the so-called “oncometabo- lites”, whose abnormal and chronic accumulation through cancer-related mutations in cellular metabolism genes can drive transformation to malignancy [4,5]. In this regard, metabolic profiling has great potential to guide diagnostics and treatments by unearthing novel biomarkers of disease and therapeutic targets [6]. One of the most studied metabolic changes in cancers cells is the uniquely high rate of glucose metabolism (aerobic glycolysis) in the presence of sufficient amounts of oxy- gen, a phenomenon known as the Warburg effect, which provides cancer cells with a metabolic advantage to meet their bioenergetic demands and maintain rapid prolifera- tion [7,8]. Metabolic reprogramming modulated by oncometabolite signaling pathways [9] is now recognized as playing a fundamental role in the malignant transformation of cells and the phenotypic evolution of tumors [10]. The upregulation of glycolysis in cancer cells is often accompanied by mitochondrial dysfunction, which can lead to the accumulation of various oncometabolites, including fumarate, succinate, L-2-hydroxyglutarate and D-2- hydroxyglutarate, with assigned pro-oncogenic features [11,12]. Succinate is an archetypal Cancers 2021, 13, 1653 3 of 16 respiratory metabolite generally accepted as an oncometabolite through its association with a breakdown in the mitochondrial tricarboxylic acid (TCA) cycle as a consequence of dys- function in succinate dehydrogenase (SDH) [10]. Succinate functions as a product inhibitor of a superfamily of enzymes known as α-ketoglutarate-dependent dioxygenases, leading to epigenetic dysregulation and induction of pseudohypoxic phenotypes via the activation of hypoxia-inducible factor (HIF)-1α [13]. Remarkably, it has become evident over the last decade that succinate also functions as an extracellular signaling metabolite via engagement with its cognate receptor succinate receptor 1, SUCNR1 (also known as GPR91). While its role in cancer has been poorly studied, SUCNR1 has been proposed as a tumor driver, as succinate-SUCNR1 signaling has been related to epithelial-to-mesenchymal transition [14], angiogenesis [15] and tumor-associated macrophage polarization [16]. Moreover, SUCNR1 expression seems to be closely related to immune status in ovarian cancer [17]. In the present study, we found markedly high levels of circulating succinate in patients with HNSCC, which prompted us to investigate the differential expression of SUCNR1 and succinate-related genes. We also assessed the potential of the succinate oncometabolite pathway as a prognostic biomarker of the local and regional control of HNSCC. 2. Materials and Methods 2.1. Study Design and Patients Sixty-six consecutive patients with pathologically confirmed, untreated HNSCC from the Hospital Universitari Joan XXIII in Tarragona (Spain) were included in the study. The tumor board evaluated all the patients, and the decision to treat with radiotherapy, chemoradiotherapy or surgery was made according to the standard protocols and guidelines of the center. In general, larynx T1-T3 tumors and oral cavity T1-T2 tumors are treated by tran- soral surgery. Primary surgical treatment is recommended for T3-T4 oral cavity and T4 laryngeal or hypopharyngeal cancers, especially when there is laryngeal cartilage invasion. Regarding the treatment of advanced oropharyngeal lesions, surgery is indicated if the employment of radiotherapy (RT) is contraindicated. Postoperative RT or chemoradiotherapy (ChRT) was administered within 6–7 weeks of surgery. Postoperative RT was administered to patients with T3-T4 tumors, resection margins with macroscopic or microscopic residual disease, perineural infiltration, lym- phatic infiltration, >1 invaded lymph node and the presence of extracapsular infiltration. Postoperative ChRT is recommended for patients with microscopic residual disease and extracapsular rupture. For those patients treated with radiotherapy, external-beam radiotherapy was admin- istered by continuous-course radiotherapy. In this cohort, all these patients were treated with conventional fractionated radiotherapy. Treatment was administered in total doses of 70 Gy to the primary site, and 54 Gy to the cervical region in all patients with N0 nodes, and 70 Gy in the case of clinical metastatic neck nodes (N+). Concomitant chemoradiotherapy included the same radiotherapy schedule combined with high-dose cisplatin (100 mg/m2 ) administered on day 1 every 3 weeks. In postoperative adjuvant RT, microscopic residual disease was irradiated with 54 Gy. High risk areas were boosted to a dose of 60–66 Gy depending on each case. Human papilloma virus (HPV) status was available for those patients that presented with an oropharyngeal carcinoma (n = 10), and was assessed by DNA detection using a multiplex polymerase chain reaction (PCR) assay. Routine follow-up included the evaluation of symptoms and locoregional examina- tions at 2-month intervals during the first year, 3-month intervals in the second year, and 4-month intervals over years 3–5. Computed tomography scans were performed every 6 months during the first two years. The mean follow-up of the patients included in the study was 3.5 years (95% confidence interval [CI] 2.5–4.6 years). The research study was reviewed and approved by the Ethical Committee of the institution and conformed to the principles outlined in the Declaration of Helsinki. All patients gave informed consent. Cancers 2021, 13, 1653 4 of 16 Plasma of 20 healthy donors (body mass index [BMI] = 22.5 ± 2.6 kg/m2 , 60% female, age = 48.6 ± 14.7) were obtained of a collection managed by the BioBank of the Pere Virgili Health Research Institute (IISPV) integrated into the platform “Red Nacional de Biobancos del ISCIII (PT17/0015/0029)”; node of the University Hospital of Tarragona “Joan XXIII” and under the approval of the Research Ethics Committee of the biobank. Inclusion criteria for this cohort were: (1) caucasian men and women; (2) body mass index (BMI) <25; (3) absence of underlying pathology on physical examination and tests; and (4) signed informed consent for participation in the study. Exclusion criteria for this cohort were: (1) serious systemic disease such as obesity, cancer, severe kidney, or liver disease; (2) systemic diseases with intrinsic inflammatory activity; (3) history of liver disease (chronic active hepatitis or cirrhosis) and/or abnormal liver function (alanine transaminase and/or aspartate transaminase three times above the upper normal value); altered renal function (creatinine >1.4 mg/dL in women and 1.5 mg/dL in men); (4) pregnancy and lactation; (5) vegetarians or subjects subjected to irregular diet; (6) patients with severe disorders of eating behavior; (7) clinical symptoms and signs of infection in the previous month; (8) anti-inflammatory chronic treatment with steroidal and/or nonsteroidal anti-inflammatory drugs; (9) prior antibiotic treatment in the last 3 months; (10) major psychiatric antecedents; and (11) uncontrolled alcoholism or drug abuse. 2.2. Measurement of Circulating Succinate Peripheral blood was collected from 66 patients after an overnight fast at the diagnostic visit, before any treatment was performed. Blood samples from 20 healthy controls were also included. Blood was drawn in 10-mL vacutainer tubes from an antecubital vein. Within 1 h of drawing, the plasma was separated by centrifugation at 1500× g for 15 min at 4 ◦ C; samples were stored at −80 ◦ C until analytical measurements were performed. Circulating plasma succinate levels were measured using the EnzyChromTM Suc- cinate Assay Kit (BioAssay Systems, Hayward, CA, USA). Assay sensitivity was 12 µM and the intra- and interassay coefficients of variance were <3.5% and 6.95%, respectively. Circulating succinate levels measured by this fluorometric assay have been previously validated by liquid chromatography-mass spectrometry and nuclear magnetic resonance analysis [18]. 2.3. Gene Expression Analysis We surveyed the gene expression pattern of 150 matched mucosa samples from 50 patients that belonged to the same initial cohort. Biopsy specimens were taken from the pri- mary site of the tumor, adjacent to the tumor and distal from the tumor. Adjacent mucosa was a macroscopically healthy tissue taken 1-cm from the tumor lesion. A sample aliquot was used for the pathologic diagnosis of the malignancy and another aliquot was immedi- ately stabilized by inclusion in RNAlater preservative (Qiagen GmbH, Hilden, Germany) and stored at −80 ◦ C until processing. Total RNA was isolated from 30 mg of tissue using the RNeasy Mini Kit (Qiagen). cDNA was prepared by reverse transcribing 1 µg of RNA with the High-Capacity cDNA Archive Kit (Applied Biosystems, Foster City, CA, USA). Real-time quantitative PCR was performed in duplicate on a 7900 HT Fast Real-Time PCR platform using commercial Taqman Assays (Applied Biosystems, Madrid, Spain). Cycle threshold (Ct) values for each sample were normalized against the reference gene RPL0 (Hs99999902_m1). Predesigned assay probes (Applied Biosystems) used for the detection of the selected genes were as follows: SUCNR1 (Hs00263701_m1), SDHA (Hs00417200_m1), SDHB (Hs01042482_m1), 2-oxoglutarate dehydrogenase OGDH (Hs01081865_m1) and HIF¬1α (Hs00153153_m1). Results were calculated using the comparative Ct method (2−∆∆Ct ) and expressed relative to a calibrator (a mix of 9 samples from normal, adjacent and tumoral mucosa). 2.4. Histology To evaluate protein expression, 10 biopsies of normal mucosa and 10 of HNSSC tumor were fixed in 70% ethanol and paraffin-embedded. Sections of 2-µm thickness were then Cancers 2021, 13, 1653 5 of 16 stained with hematoxylin and eosin to confirm the presence of sufficient representative tis- sue. For immunohistochemistry analysis, paraffin-embedded cellular blocks were sectioned at a thickness of 4 µm and each slide was deparaffinized in xylene for 20 min, rehydrated with a decreasing ethanol series and washed with phosphate buffered saline. Sections were then heated at 96 ◦ C for 20 min for antigen retrieval and then incubated for 30 min with a primary polyclonal antibody against SUCNR1/GPR91 antigen (Novus Biological, Littleton, CO, USA, cat#NLS3476), at a 1:2 dilution or against p53 antigen (DAKO, Clone DO-7, ready-to-use, cat#IR61661-2). Automatic immunodetection was performed using the END- VISIONTM FLEX method (DAKO, Carpenteria, CA, USA) using 33200 -diaminobenzidine chromogen as substrate, followed by counterstaining with hematoxylin. Slides were examined by two blinded observers and the localization of SUCNR1 was recorded in the basal, spinous and superficial layers of the normal mucosa, and in the infiltrating tumoral areas. SUCNR1 staining intensity was scored semi-quantitatively as 0+, 1+, 2+ or 3+. Score 0+ was assigned to no expression or focal weak expression, score 1+ to an intense focal or diffuse weak expression; score 2+ to moderate diffuse expression, and score 3+ an intense diffuse expression in epithelial cells. For each case, the presence or absence of high-grade dysplasia and infiltrating tumor areas was confirmed with the presence of positive p53 staining. 2.5. Statistical Analysis Differences between groups were analyzed using Student’s t test or the Mann–Whitney U test for comparisons of normally and non-normally distributed quantitative variables, as needed. One-way analysis of variance was used to compare variables in different body weight groups (lean: body mass index [BMI] <25 kg/m2 ; overweight: BMI >25 and <30 kg/m2 ; obese: BMI > 30 kg/m2 ). The paired t test and the Wilcoxon signed rank test were used for paired analysis of the different mucosae. Spearman’s rank-order correlation was used to analyze the relationship between parameters. Two-tailed p-values of 0.05 were considered statistically significant. Receiver operating characteristic (ROC) curves, in which sensitivity was plotted as a function of 1-specificity, were developed to assess the predictive value of circulating succinate. The Chi-square test was used to evaluate the relationship between categorical variables. As proposed by Chiesa et al. [19] for studies on predictive factors in HNSCC, we evaluated the outcome by the locoregional control with a follow-up of at least 2 years. In this cohort, with a mean follow-up of 3.5 years, the outcome used for our analysis was the last one registered during routine follow-up examinations, so we used time-to-event data. Locoregional disease-free survival was defined as the period from the completion of the primary treatment to any local or regional recurrence. The variable profile was defined according to the control of the locoregional disease-free survival using the Classification and Regression Tree (CRT) method. Variables to generate the regression tree included the stage, primary location, HPV status, BMI, treatment, N category, tumor differentiation and the gene expression levels in tumoral tissue. CRT analysis splits the data into segments that are as homogeneous as possible with regard to the dependent variable. The locoregional disease-free survival according to the variable profile was calculated using the Kaplan–Meier method. Differences in survival rates were compared using the log-rank test. For univariate and multivariate analysis, a Cox regression analysis was performed considering disease-free survival as the dependent variable, and location of the tumor (oropharynx-oral cavity vs. larynx-hypopharynx), Eastern Cooperative Oncology Group (ECOG) general status index, Stage (Stage I-II vs. III-IV), N category (N0 vs. N+) and the categorized variable profile (1, 2 or 3) as the independent variables. All statistical analyses were made using SPSS software v.20.0 (IBM, Madrid, Spain). 3. Results 3.1. Characteristics of the Patients Included in the Study The main characteristics of the patients and the locorregional disease-free survival rate stratified by each variable are shown in Table 1. Positive nodal status was associated with Cancers 2021, 13, 1653 6 of 16 significantly reduced locoregional disease-free survival in univariate analysis. Additionally, there were significant between-sex differences in survival, but the cohort only included three women. Treatment modality was also associated with locoregional disease-free survival. HPV status was evaluated in 10 patients with oropharyngeal cancer and most presented with HPV-negative tumors (80%). Table 1. Characteristics of the Study Cohort Including Univariate Analysis Data of Locoregional Disease-Free Survival. Characteristics Num. Patients (%) Disease-Free Survival % (95% CI) p-Value Age (years) 0.256 <50 10 (15.2) 60.0 (14.6–94.7) 50–60 23 (34.8) 52.6 (28.8–75.5) 60–70 21 (31.8) 76.0 (54.8–90.6) >70 12 (18.2) 64.7 (38.3–85.8) Sex * 0.039 Male 63 (95.5) 68.2 (55.3–79.4) Female 3 (4.5) 0.0 (0.0–70.7) Tobacco consumption 0.383 Never 9 (13.6) 55.5 (21.2–86.3) <1 pack-year 0 - >1 pack-year 57 (86.4) 66.6 (52.9–78.5) BMI (kg/m2 ) 0.966 <25 (Lean) 28 (42.4) 68.0 (46.4–85.1) 25–30 (Overweight) 28 (42.4) 64.0 (42.5–82.0) >30 (Obese) 10 (15.2) 66.7 (29.9–92.5) Alcohol consumption 0.395 Never 18 (27.3) 72.2 (46.5–90.3) Mild-moderate 6 (9.1) 83.3 (35.9–99.5) Severe 42 (63.9) 59.5 (43.2–74.3) ECOG Index 0.148 0 28 (42.4) 75.0 (55.1–89.3) >0 38 (57.6) 60.5 (43.3–75.9) Tumor location 0.131 Oral cavity-oropharynx 15 (22.7) 46.6 (21.2–73.4) Larynx-hypopharynx 51 (77.3) 70.5 (56.1–82.5) T category 0.439 T1-T2 26 (39.4) 69.2 (48.2–85.7) T3-T4 40 (60.6) 65.0 (48.3–79.3) N category * 0.039 N0 35 (53.0) 77.1 (59.8–89.5) N+ 31 (47.0) 53.3 (34.3–71.6) Stage 0.185 I–II 21 (31.8) 77.1 (59.8–89.5) III–IV 45 (68.2) 53.3 (34.3–71.6) Tumor differentiation 0.371 Good 9 (13.6) 70.0 (34.7–93.3) Moderate 49 (74.2) 63.2 (48.3–76.5) Poor 8 (12.1) 28.5 (3.6–70.9) Treatment * 0.014 RT or ChRT 41 (62.1) 61.3 (42.2–78.2) Surgery 15 (22.7) 93.8 (69.8–99.8) Surgery and RT/ChRT 10 (15.2) 45.5 (16.7–76.6) Abbreviations: BMI, body mass index; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; ChRT, chemoradiotherapy; RT, radiotherapy. *, Differences were considered statistically significant at p < 0.05. Treatment *0.014 RT or ChRT 41 (62.1) 61.3 (42.2–78.2) Surgery 15 (22.7) 93.8 (69.8–99.8) Surgery and RT/ChRT 10 (15.2) 45.5 (16.7–76.6) CancersAbbreviations: 2021, 13, 1653 BMI, body mass index; CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; ChRT, 7 of 16 chemoradiotherapy; RT, radiotherapy. *, Differences were considered statistically significant at p < 0.05. 3.2. Circulating Levels of the Oncometabolite Succinate Are Elevated in Patients with HNSCC 3.2. Circulating Levels of the Oncometabolite Succinate Are Elevated in Patients with HNSCC We measured and compared the circulating levels of succinate in 66 patients with HNSCC measured We and 20 age-andandcompared sex-matched the healthy circulating levels normal of succinate weight in 66 patients control subjects, findingwith that HNSCC and 20 age- and sex-matched healthy normal weight control subjects, levels were 4-fold higher in the former (Figure 1A). As circulating succinate is associated finding that withlevels were 4-fold body weight higher [20], we in the stratified formerwith patients (Figure HNSCC1A).according As circulating succinate to their is asso- BMI (Figure 1A). ciated with body weight [20], we stratified patients with HNSCC according We found that there was a tendency for an increase in circulating succinate in patients to their BMI (Figure 1A). We found with overweight that there or obesity but thewas a tendency differences for an were notincrease in circulating statistically succinate significant, in indicating patients with overweight or obesity but the differences were not statistically that the profound changes in plasma succinate were most likely due to the tumor presence significant, indicating and not tothat the profound variations in BMI.changes We alsoin plasma succinate evaluated wereassociations the potential most likelybetween due to the the tumor presence and not to variations in BMI. We also evaluated the potential circulating levels of succinate and the clinico-pathological variables (T and N-categories,associations between the circulating levels of succinate and the clinico-pathological variables (T and N- stage, BMI and locoregional control) but no statistically significant relationships were categories, stage, BMI and locoregional control) but no statistically significant relationships found. were found. Figure 1.1. Succinate Figure Succinate asas aa diagnostic diagnostic marker marker in in HNSCC. HNSCC. (A) (A)Circulating Circulatingsuccinate succinateininpatients patientswithwith HNSCC HNSCC and control subjects. Patients were also stratified according to their body mass index and control subjects. Patients were also stratified according to their body mass index (BMI). (BMI). Results Resultsare areexpressed expressedasasmean mean(SEM). (SEM).Differences Differencesbetween betweengroups groupswere werecalculated using calculated usingStudent’s Student’st- test forfor t-test independent independent samples samples(left(left panel) or analysis panel) of variance or analysis (ANOVA, of variance (ANOVA,rightright panel). *, Differences panel). *, Differ- encesconsidered were were considered statistically statistically significant significant at p < at p < (B) 0.05. 0.05.Receiver (B) Receiver operating operating characteristic characteristic curve curve for for succinate succinate levels. levels. AUC: AUC: areaarea underunder the curve. the curve. To test To testthe thevalue valueofofthe thebasal basalsuccinate succinatelevel levelfor forHNSCC HNSCCdiagnosis, diagnosis,weweperformed performed ROC ROCcurve curveanalysis. analysis. The The area area under under the the curve curve for for succinate succinate levels levels predicting predictingthe thepresence presence of HNSCC was 0.922 (95% CI 0.86–0.98) (Figure 1B). The best cut-off value of succinate according to the Youden index calculation was 42 µM (sensitivity 89%; specificity 89%). 3.3. SUCNR1 Is Predominantly Expressed in Tumoral Mucosa and Positively Associates with Plasma Succinate To explore the potential role of succinate-related pathways in HNSCC, we first sur- veyed the gene expression of HIF-1α, TCA cycle-related genes (SDHA, SDHB, OGDH) and SUCNR1 in tumoral, nontumoral adjacent tissue and normal mucosa. With the exception of OGDH, we found that gene expression was significantly higher in tumoral mucosa than in healthy tissue (Figure 2A). Remarkably, the circulating levels of succinate correlated with the expression of succinate-related genes in tumoral tissue (Table 2) but not with their expression in the normal or adjacent mucosa (data not shown). Moreover, the correlation analysis demon- strated a strong association of expression between the examined genes in the tumor tis- sue (Figure 3); in particular, a strong correlation was found between SUCNR1 and TCA cycle-related genes, suggesting that the function of succinate as an extracellular signaling metabolite might be also related to its synthesis and oxidation. 3.3. SUCNR1 Is Predominantly Expressed in Tumoral Mucosa and Positively Associates with Plasma Succinate To explore the potential role of succinate-related pathways in HNSCC, we first sur- veyed the gene expression of HIF-1α, TCA cycle-related genes (SDHA, SDHB, OGDH) and SUCNR1 in tumoral, nontumoral adjacent tissue and normal mucosa. With the ex- Cancers 2021, 13, 1653 8 of 16 ception of OGDH, we found that gene expression was significantly higher in tumoral mu- cosa than in healthy tissue (Figure 2A). SUCNR1protein Figure2.2.SUCNR1 Figure proteinand andgene geneexpression expressionisisincreased increasedinintumors. tumors.(A) (A)HIF-1α, HIF-1α,SDHA, SDHA,SDHB, SDHB,OGDH OGDH andSUCNR1 and SUCNR1 gene expression gene in normal, expression adjacent in normal, andand adjacent tumoral tumoralmucosa. mucosa. Gene Gene expression expressiondifferences differencesbetween betweentissue tissue samples samples were cal- calculated culated using theusing the Wilcoxon Wilcoxon SignedSigned Rank Rank Test forTest for nonparametric-related nonparametric-related samples. samples. Results Results are expressed are expressed as mean as mean (SEM). (SEM). *, *, Differences Differences were considered were considered statistically statistically significantsignificant at p(B) at p < 0.05. < 0.05. (B) Representative Representative imagesimages of SUCNR1 of SUCNR1 immunostaining immunostaining in in infiltrating infiltrating tumor and normal mucosa. Upper left image: high immunohistochemical expression of SUCNR1 in infiltrating tumor and normal mucosa. Upper left image: high immunohistochemical expression of SUCNR1 in infiltrating tumor. tumor. Upper right image: low immunohistochemical expression of SUCNR1 in normal mucosa (black arrow in spinous Upper right image: low immunohistochemical expression of SUCNR1 in normal mucosa (black arrow in spinous layer). layer). Lower left image: high immunohistochemical expression of SUCNR1 in adjacent mucosa with high grade (HG) Lower left dysplasia image: (red highLower arrows). immunohistochemical expression right image: p53 staining of SUCNR1 highlight the HG in adjacent grade mucosa dysplasia with areas high grade in adjacent (HG)(red mucosa dysplasia (red arrows). Lower right image: p53 staining highlight the HG grade dysplasia areas in adjacent mucosa (red arrows). arrows). Remarkably, Table the Analysis 2. Correlation circulating levels of succinate of Circulating Succinatecorrelated with the expression with the Expression of suc- in the of Genes Involved cinate-related genes in tumoral Succinate-SUCNR1-HIF-1α tissue (Table Pathway and the2)TCA but Cycle not with their expression in Tumoral Mucosa. in the normal or adjacent mucosa (data not shown). Moreover, the correlation analysis demonstrated a strong association of expression Variables between the examined genes in the tumor tissueSuccinate (Figure 3); in particular, a strong correlation was found rho between SUCNR1 and TCA cycle-related 0.289 HIF-1α genes, suggesting p-value that the function of succinate as an extracellular signaling* metabolite 0.049 might be also related to its synthesis and oxidation.rho 0.406 SDHA p-value * 0.005 rho 0.371 SDHB p-value * 0.010 rho 0.248 OGDH p-value 0.092 rho 0.296 SUCNR1 p-value * 0.044 Associations between variables were calculated using Spearman’s rank-order correlation test for nonparametric values. *, Rho coefficients were considered statistically significant at p < 0.05. We next questioned whether the higher gene expression of SUCNR1 was reflected in protein expression in tumors. We found that infiltrating tumor exhibited robust SUCNR1 protein staining (scores 3+) (Figure 2B, upper left image), confirming that the transcriptional regulation of SUCNR1 in the infiltrating tumor is mirrored at the protein level. By contrast, SUCNR1 expression in the normal mucosa was observed only in the cells of the spinous layer of the mucosa (Figure 2B, upper right image). The mucosa adjacent to the infiltrating tumor showed a heterogenous pattern of SUCNR1 expression. We found that p53-positive areas, that highlight regions of high grade epithelial dysplasia, had stronger SUCNR1 protein staining than areas with normal histology (Figure 2B, lower left image). Additionally, the gene expression of HIF-1α, SDHA, rho 0.406 SDHA p-value * 0.005 rho 0.371 SDHB p-value * 0.010 Cancers 2021, 13, 1653 rho 0.248 9 of 16 OGDH p-value 0.092 rho 0.296 SUCNR1 p-value * 0.044 SDHB, andbetween Associations SUCNR1 in adjacent variables mucosa was were calculated higher using than in Spearman’s normal mucosa, rank-order which correlation was test for nonparametric values. *, Rho statistically significant for coefficients HIF-1α and were SDHBconsidered (Figurestatistically 2A). significant at p < 0.05. Correlationsbetween Figure3.3.Correlations Figure betweenHIF-1α, HIF-1α,SDHA, SDHA,SDHB,SDHB,OGDH OGDHand andSUCNR1 SUCNR1expression expressionin intumors. tumors.Associations Associationsbetween between variableswere variables werecalculated calculatedusing usingSpearman’s Spearman’srank-order rank-ordercorrelation correlationtest testfor for nonparametric nonparametricvalues. values.Rho Rhocoefficients coefficientswere were consideredstatistically considered statisticallysignificant atpp<<0.05. significantat 0.05. 3.4. We SUCNR1 and SDHA whether next questioned Gene Expression in Tumoral the higher Tissue Is Associated gene expression of SUCNR1 with Locoregional was reflected in Disease-Free Survival protein expression in tumors. We found that infiltrating tumor exhibited robust SUCNR1 proteinTostaining evaluate(scores the discriminating 3+) (Figure 2B,potential upper left of image), the succinate-related confirming thatpathway for lo- the transcrip- coregional tional disease-free regulation of SUCNR1survival, in thewe used thetumor infiltrating CRT method, is mirroredfirstat inputting the protein the clinico- level. By pathological contrast, variables SUCNR1 (ECOG in expression index, N-category, the normal mucosastage, wasBMI, treatment, observed onlytumor in thelocation and cells of the differentiation), spinous andmucosa layer of the then the circulating (Figure succinate 2B, upper right levels image). and the tumor gene expression of HIF-1α, SDHA, SDHB, OGDH and SUCNR1 as independent The mucosa adjacent to the infiltrating tumor showed a heterogenous variables. Following patternCRT of SUCNR1 expression. We found that p53-positive areas, that highlight regions tree. analysis, only SUCNR1 and SDHA expression were selected to create the final decision of Together, these high¬grade variables epithelial correctlyhad dysplasia, discriminated 74.2% ofprotein stronger SUCNR1 the cases according staining thantoareas the locore- with gional disease-free normal survival histology (Figure 2B,achieved lower left after treatment. image). As shown Additionally, theingene Figure 4A, patients expression with of HIF- lowSDHA, 1α, levels of SUCNR1 SDHB, had betterindisease and SUCNR1 adjacent control mucosathan their was peersthan higher withinhigher normal expression mucosa, levels. In those cases, the CRT analysis revealed a cut-off which was statistically significant for HIF-1α and SDHB (Figure 2A). value of 0.859 for SUCNR1 for discriminating disease-free survival. Low SDHA expression was also associated with better survival rates (cut-off value of 2.452) in those patients with high SUCNR1 expression levels. Using this method, we generated a categorized variable “profile”, which classified the patients according to their expression levels of SUCNR1 and SDHA (Figure 4A). Profile 1 corresponded to patients with low SUCNR1 expression (<0.859); profile 2 included patients with high SUCNR1 expression (>0.859) and low SDHA expression (<2.452); and finally, profile 3 included patients with both high SUCNR1 and SDHA expression. We found higher SUCNR1 and SDHA expression in tumor tissue from patients with poor locoregional control (Figure 4B). However, in agreement with the tree created, we did not find significant differences in the plasma levels of succinate (data not shown). As shown in Figure 4B, higher SUCNR1 and SDHA expression were evident in patients with more advanced (T3-4) tumors. SUCNR1 for discriminating disease-free survival. Low SDHA expression was also associ- ated with better survival rates (cut-off value of 2.452) in those patients with high SUCNR1 expression levels. Using this method, we generated a categorized variable “profile”, which classified the patients according to their expression levels of SUCNR1 and SDHA (Figure 4A). Profile 1 corresponded to patients with low SUCNR1 expression (<0.859); Cancers 2021, 13, 1653 10 of 16 profile 2 included patients with high SUCNR1 expression (>0.859) and low SDHA expres- sion (<2.452); and finally, profile 3 included patients with both high SUCNR1 and SDHA expression. Figure Figure4. 4. SUCNR1 SUCNR1 and andSDHA SDHAexpression expressionisisassociated associated with the the locoregional locoregionalcontrol controlofofthe the HNSCC HNSCC disease. disease. (A)(A) Classifi- Classification cation and regression tree for disease-free survival rates based on the tumoral expression of SUCNR1 and and regression tree for disease-free survival rates based on the tumoral expression of SUCNR1 and SDHA. Variables included SDHA. Variables included were the were the tumoral tumoral expression expression of HIF-1α, of HIF-1α, SDHA,SDHA, SDHB,SDHB, OGDH OGDH and SUCNR1, and SUCNR1, the primary the primary location, location, the stage, the stage, T- andT-N- and N-categories, tumor differentiation degree, the treatment and body mass index categories. Pie charts represent the categories, tumor differentiation degree, the treatment and body mass index categories. Pie charts represent the proportion proportion of patients who met the locoregional disease-free survival (light grey; yes) or not (dark grey; no) at each node of patients who met the locoregional disease-free survival (light grey; yes) or not (dark grey; no) at each node of the tree; of the tree; “Yes” includes patients that are free of disease at the last follow-up; and “No” includes patients that are not “Yes” free includes of disease atpatients the last that are free(B) follow-up. of disease SUCNR1 at and the last SDHAfollow-up; and “No” expression includes according patients to the that arecontrol locoregional not freeachieved of disease at the last follow-up. (B) SUCNR1 and SDHA expression according to the locoregional control (upper panel). SUCNR1 and SDHA expression in T1-2 and T3-4 according to the locoregional control achieved (lower achieved (upper panel). SUCNR1 panel). andare Results SDHA expression expressed in T1-2(IQR). as median and T3-4 according Differences to the locoregional between groups werecontrol achieved calculated using(lower panel). ResultsUare the Mann–Whitney expressed test as median for independent (IQR). *,Differences samples. Differences between groups were were considered calculated statistically using theatMann–Whitney significant U test for independent p < 0.05. (C) Kaplan–Meier survival analysis samples.showing locoregional *, Differences wererecurrence-free survival according considered statistically significanttoatthe p 0.859) locoregional recurrence-free survival according to the categorized levels of SUCNR1 and SDHA expression in tumors.and low SDHA expres- sion (<2.452); Profile Profile 3:expression 1: SUCNR1 high SUCNR1 andProfile <0.859; SDHA2: expression. high SUCNR1 expression (>0.859) and low SDHA expression (<2.452); Profile 3: high SUCNR1 and SDHA expression. We found higher SUCNR1 and SDHA expression in tumor tissue from patients with poor 3.5.locoregional control SUCNR1/SDHA Relationship between (Figure 4B). However, Profilesin agreement and with theVariables Clinicopathological tree created, we did notThe findlocoregional significant differences in the recurrence-free plasmarates survival levels of succinate according to the(data not of location shown). As the primary tumor, the local extension of the tumor, the presence of node metastasis, the type of treatment and the BMI, relating to the profiles of SUCNR1/SDHA expression are shown in Table 3. Locoregional recurrence-free survival rate was lower in the category of patients with higher-level expression of SUCNR1/SDHA (Profile 3), reaching statistical significance for patients with N0 disease, with tumors located at the larynx-hypopharynx, lean patients or those treated with radio- or chemoradiotherapy alone. When we analyzed the subgroup of patients treated with radio-or chemoradiotherapy (n = 41), we found that SUCNR1 levels were elevated in patients with poor locoregional control (no locoregional control vs. locoregional control: 2.5 ± 0.7 vs. 1.0 ± 0.2, p = 0.046), and SDHA expression showed a trend for higher expression (2.5 ± 0.6 vs. 1.6 ± 0.3, p = 0.144). 3.6. SUCNR1/SDHA Profiles Are Independent Predictors of HNSCC Prognosis We next analyzed the prognosis of locoregional disease-free survival using the Kaplan–Meier method. The locoregional recurrence-free survival curves according to the SUCNR1/SDHA expression profile in tumoral tissue are shown in Figure 4C. The 5-year locoregional-free survival rate for patients was 77.1% (95% CI: 48.9–100.0) in profile 1, 59.4% (95% CI: 31.5–87.3) for profile 2 and 16.7% (95% CI: 0.0–44.4) for profile 3. We then aimed to analyze the effect of the different risk factors on the period of locoregional disease-free survival. The results of the univariate and multivariate study considering locoregional disease-free survival as the dependent variable are shown in Cancers 2021, 13, 1653 11 of 16 Table 4. In the univariate study, the analysis showed that the profile was significantly related to the locoregional disease-free survival. Considering patients with profile 1 as the reference category, those with higher profiles had a higher risk of local and regional failure of the tumor after treatment. Specifically, those patients included in profile 2 had a 2.5-fold higher risk of recurrence (CI 95%: 0.5–12.2, p = 0.272), and patients with profile 3 had a 6.6-fold higher risk (CI 95%: 1.4–32.1, p = 0.019). Table 3. Local and Regional Recurrence Free-Survival According to Clinical Variables and to SUCNR1 and Succinate Dehydrogenase Profiles. SUCNR1-SDHA Expression Profiles Clinic-Pathological Variables Profile 1 (%) Profile 2 (%) Profile 3 (%) p-Value Tumor stage I–II 83.3 77.8 0.0 0.064 III–IV 86.7 63.6 37.5 0.053 N category N0 91.7 83.3 33.3 * 0.017 N+ 77.8 50.0 25.0 0.181 Tumor location Oral cavity-oropharynx 71.4 0.0 0.0 0.117 Larynx-hypopharynx 92.9 73.7 37.5 * 0.019 Treatment RT or ChRT 80.0 77.8 16.7 * 0.021 Surgery 100.0 80.0 100.0 0.466 Surgery and RT/ChRT 50.0 66.7 50.0 0.907 BMI (kg/m2 ) Lean (<25) 100.0 66.7 0.0 * 0.001 Overweight (25–30) 71.4 71.4 17.6 0.270 Obese (>30) 100.0 60.0 0.0 0.476 Abbreviations: BMI, body mass index; ChRT, chemoradiotherapy; RT, radiotherapy. Profile 1: patients with low SUCNR1 expression (<0.859); Profile 2: patients with high SUCNR1 expression (>0.859) and low SDHA expression (<2.452); Profile 3: patients with high SUCNR1 and SDHA expression. Differences between groups were calculated using the Chi-square test. * p < 0.05, statistically significant. Table 4. Prognostic Factors of Locoregional Control in Univariate and Multivariate Cox Regression Analyses. Variables Categories HR 95% CI p-Value Univariate Models SUCNR1/SDHA profile 2 vs. 1 2.46 0.50–12.20 0.272 3 vs. 1 6.63 1.37–32.07 * 0.019 Stage III–IV vs. I–II 1.79 0.65–4.90 0.258 ECOG >0 vs. 0 2.27 0.92–5.61 0.075 N category N+ vs. N0 2.45 1.01–5.92 0.047 Treatment Surgery vs. RT or ChRT 1.844 0.68–5.02 0.230 Surgery vsSurgery and RT/ChRT 0.163 0.02–1.26 0.082 Tumor location LH vs. OCO 0.53 0.21–1.32 0.171 Multivariate Mode SUCNR1-SDHA profile 2 vs. 1 2.75 0.55–13.79 0.218 3 vs. 1 7.02 1.45–33.96 * 0.015 N category N+ vs. N0 2.18 0.78–6.08 0.135 Dependent variable: Locoregional control. Abbreviations: HR, hazard ratio; CI, confidence interval; LH: larynx- hypopharynx; OCO: oral cavity-oropharynx. Profile 1: patients with low SUCNR1 expression (<0.859); Profile 2: patients with high SUCNR1 expression (>0.859) and low SDHA expression (<2.452); Profile 3: patients with high SUCNR1 and SDHA expression. * p <0.05, statistically significant. In the multivariate analysis, the binary variables included were N category (N0 vs. N+) and the profile, those that were statistically significant in the univariate analysis. Notably, patients included in profile 2 had a 2.2-fold higher risk of locoregional failure (95% CI: 0.6–13.8, p = 0.218) and patients with profile 3 had a 7-fold higher risk (95% CI: 1.5–34.0, p = 0.015) (Table 4). Cancers 2021, 13, 1653 12 of 16 4. Discussion In the present study, we show for the first time, to our knowledge, that the circulating levels of succinate are elevated in HNSCC and that SUCNR1 expression is higher in tumoral tissue than in normal mucosa. Remarkably, our data also show that high SUCNR1 and SDHA expression predict poor locoregional disease-free survival. Succinate is a key intermediate metabolite in several metabolic pathways and accu- mulates locally in extracellular spaces under pathological conditions; for instance, hyper- glycemia and hypoxia [15]. High concentrations of succinate have also been detected in the plasma of patients with metabolic diseases [21,22]. Our previous data revealed that plasma succinate plasma are dependent on body weight and that circulating succinate is elevated in obesity [18,20]. Here, we found that succinate levels are higher in patients with HNSCC than in healthy controls, with levels close to those previously detected in the higher quar- tile in people with morbid obesity [20]. No differences were found between patients with HNSCC who were lean and those who had different degrees of obesity, indicating that the greater level of circulating succinate is due to the disease. Indeed, we found that baseline circulating succinate was highly associated with tumor presence, pointing to succinate as a potential clinical tool in HNSCC. Regarding the origin of the increased succinate, several studies have reported high levels of succinate in different types of tumors [15,23]. We found close relationships be- tween the expression of genes involved in succinate metabolism in tumoral tissue, but not in the adjacent or normal mucosa. We hypothesize that plasma succinate originates from the extracellular secretion of succinate from cells in the tumoral microenvironment or the tumor cells themselves, as reported for lung tumoral cells and tumor-associated macrophages [16], and for gastric cancer tissue [15]. Several explanations for succinate accumulation have been put forward. One explanation is that in hypoxic tissue (e.g., reti- nas), the hypoxia-related induction of oncometabolites might be propagated and amplified by the oncometabolites themselves via the stabilization of HIF-1α expression [24,25]. Al- ternatively, succinate may be derived from glutamine-dependent anaplerosis, which was reported to be the main source of succinate in activated tumor-associated macrophages [26]. The increase in succinate in tumors has also been previously related to SDH dysfunction, such as papillary thyroid cancinoma, thyroid C-cell hyperplasia, pancreatic neuroendocrine tumors, paragangliomas, ovarian cancer, hepatocellular carcinoma, colorectal cancer, renal carcinomas and pituitary adenomas, among others [27,28]. There is, however, no evidence to show that SDH is dysfunctional in HNSCC and further studies are warranted to elu- cidate the precise mechanism underlying succinate accumulation and secretion in this cancer. Accumulated succinate might represent a novel noninvasive biomarker for the presence of HNSCC. Using tissue samples from patients with HNSCC, we also provide the first demonstra- tion that the expression levels of SUCNR1, HIF-1α and TCA enzymes related to succinate metabolism are elevated in tumoral mucosa. Of note, we found that SUCNR1 and SDHA expression were associated with patient outcome, with high SUCNR1 and SDHA expres- sion levels predictive of poor prognosis. This signature might be useful in identifying patients at higher risk, complementing the classical tumor grading and risk assessment system. The increased expression of SDH in patients with HNSCC and poor prognosis conflicts with the reported idea that SDH alteration, dysfunction or mutation is associ- ated with high risk of locoregional recurrence [29]. It has been suggested that reduced SDH activity could contribute to succinate accumulation and secretion [29]. By contrast, in a recent report analyzing mitochondrial function in prostate cancer, increased SDHA expression was associated with reduced respiratory capacity and a significant metabolic shift towards higher succinate oxidation, particularly in high-grade tumors [30]. SDH enzymes are characterized by fine regulatory mechanisms including the regulation of mRNA expression, post-translational modification and endogenous SDH inhibition [31]. Accordingly, while the gene expression of SDH is elevated in HNSCC tumors, its activity Cancers 2021, 13, 1653 13 of 16 might be altered, leading to succinate accumulation and/or secretion. Further studies are needed to investigate the role of this enzyme in tumorigenesis and progression of HNSCC. Data on SUCNR1 in cancer are scarce; however, it has been proposed that it might play a key role in tumor progression of some carcinomas [32]. According to the Human Protein Atlas Database (www.proteinatlas.org, accessed date: 12 January 2021), SUCNR1 mRNA is overexpressed in several types of tumors, although there does not appear to be a predominant expression in any tumor type. Moreover, SUCNR1 protein overexpression has been described in renal, urothelial and pancreatic cancers. The cancer public database Oncomine, a cancer microarray database (www.oncomine.org, accessed date: 15 January 2021) also shows high variability in SUCNR1 mRNA expression between different cancer types and in the tumoral mass. Detailed information about specific differences in the cell components of the tumor (i.e., tumor cells vs. stromal cells) is currently lacking. The functional relevance of crosstalk in the tumor microenvironment is clear [33]. Changes in nontumor adjacent mucosa not detected in healthy mucosa might be a direct response of the tumor stimuli based on physical crosstalk between the cells. This molecular communication could act through direct interactions between secreted oncometabolites and their corresponding membrane receptors. The tumor mass includes a mixture of epithelial and active stromal cells. To assess which compartment was predominantly expressing SUCNR1, we evaluated its expression in adjacent mucosa by immunohistochemistry. Of note, we found that cells in adjacent mucosa that expressed a higher SUCNR1 coincided with p53 positivity areas that morphologically showed high-grade dysplasia. It is important to highlight that p53 expression was greater in oral epithelial dysplasia with high malignant potential than in carcinoma in situ or in early stage oral squamous cell carcinoma [34]. These findings highlight the importance of p53 as an initial indicator of carcinogenesis, and its coexpression with SUCNR1 might relate to cancer development. A differential expression pattern was also observed for HIF-1α and SDHB expression as compared with normal mucosa. Thus, we cannot discard the possibility that at least some of the differences found between adjacent and normal mucosa can be explained by the presence of a preneoplastic stage. The samples we had allowed us to compare dysplasic (likely preneoplasic) and nondysplasic areas, and we conclude that adjacent normal mucosa might not be consid- ered normal. In fact, many genes were deregulated in adjacent mucosa, mimicking the tumor expression. This could have important implications for diagnosis and response to treatment due to the malignant potential of epithelial dysplasia adjacent to HNSCC. While SUCNR1 has been studied in other tumors, we are not aware of any studies performed in HNSCC or in the adjacent mucosa. Considering the three-dimensional anatomy of the organs affected and the resection-dependent morbidity in terms of breathing, swallowing and aesthetics, we will need to clearly establish the role that SUCNR1 plays in adjacent mucosa malignancy. Radiotherapy is widely applied in HNSCC, either alone or in multimodal therapy strategies comprising surgery, radiotherapy and chemotherapy. Despite the many tech- nical innovations in recent years, there remains a significant rate of radioresistance in patients [35,36], highlighting an urgent medical need for new concepts in radiotherapy practice. In this regard, it has been suggested that the succinate pathway is related to pa- tient outcome or response to cancer therapy [37]. Interestingly, beyond its association with locoregional control, we found that the SUCNR1-SDHA profiles can predict the outcome in patients treated with radio- or chemotherapy. Recent studies have described that succinate and other oncometabolites not only modulate cell signaling, but also impact cancer cell response to radio- and chemotherapy, presumably by the epigenetic modulation of DNA repair [38]. Our results suggest that the succinate pathway might be useful in clinical research as a marker of radiotherapy response. There are some limitations in our study that need to be considered. The elevation of the succinate levels in the plasma prompted us to analyze the expression of the TCA enzymes that synthesize the succinate precursor, succinyl-CoA, from α-ketoglutarate (OGDH) and the enzyme that converts it into fumarate (SDH). However, we did not quantify their Cancers 2021, 13, 1653 14 of 16 protein expression, activity or fumarate levels, and so the involvement of these enzymes in HNSCC progression needs further study. Another limitation is the relatively small number of patients, as well as the heterogeneity of the sample, which limits the statistical power of our analysis. 5. Conclusions In conclusion, our data suggest an important role of the succinate-related pathway in tumor development and response to treatment in patients with HNSCC. These novel find- ings identify succinate as a potentially valuable noninvasive biomarker for the diagnosis of HNSCC. Future studies could consider developing novel therapeutic strategies that target the succinate-SUCNR1 axis. Author Contributions: Conceptualization, X.T., V.C.-M., R.B., P.C., I.V. (Isabel Vilaseca), J.G., J.V., I.V. (Izascun Valduvieco), S.F.-V. and F.X.A.-J.; data curation, X.T., C.M., E.B., R.B., J.C.F., I.V. (Isabel Vilaseca) and M.C.; formal analysis, X.T., V.C.-M., E.B., M.C., M.L., I.V. (Izascun Valduvieco), S.F.-V. and F.X.A.-J.; funding acquisition, F.X.A.-J.; investigation, X.T. and C.M.; methodology, X.T., C.M., N.B., R.B., P.C., I.V. (Izascun Valduvieco), M.L., S.F.-V. and F.X.A.-J.; project administration, J.V., S.F.-V. and F.X.A.-J.; supervision, J.C.F.; writing—original draft, X.T. and V.C.-M.; writing—review and editing, X.L., E.B., J.G., J.V., I.V. (Isabel Vilaseca), S.F.-V. and F.X.A.-J. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by a grant from Plan Estatal de I+D+I of the Instituto de Salud Carlos III (FIS PI15/02047 and FIS PI18/0844 to FX A-J and IV, and FIS PI19/01661 to XL). Fondo Europeo de Desarrollo Regional (FEDER), A Way to Build Europe. XT is a Serra-Húnter fellow. Asociación Española contra el Cáncer (LABAE18025AVIL). RTI2018-093919 to S.F.V. cofinanced by the European Regional Development Fund (ERDF). The Spanish Biomedical Research Center in Diabetes and Associated Metabolic Disorders (CIBERDEM) (CB07708/0012) is an initiative of the Instituto de Salud Carlos III. S.F.V. acknowledges support from the Miguel Servet tenure-track program (CP10/00438 and CPII16/00008) from the Fondo de Investigación Sanitaria, cofinanced by the ERDF. Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the Institutional Ethics Committee of Hospital Joan XXIII (PI15/02047). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Acknowledgments: The Fundación Cientifica AECC is the entity responsible for the financial sup- port, with the project code LABAE18025AVIL.This work was supported by a grant from Plan Estatal de I+D+I of the Instituto de Salud Carlos III (FIS PI15/02047 and FIS PI18/0844 to FX A-J and IV, and FIS PI19/01661 to XL). Fondo Europeo de Desarrollo Regional (FEDER), A Way to Build Europe. XT is a Serra-Húnter fellow. This study was also funded by Spanish Ministry of Economy and Competitiveness RTI2018-093919 to S.F.V., cofinanced by FEDER. The Spanish Biomedical Research Center in Diabetes and Associated Metabolic Disorders (CIBERDEM) (CB07708/0012) is an initiative of the Instituto de Salud Carlos III. S.F.V. acknowledges support from the Miguel Servet tenure-track program (CP10/00438 and CPII16/00008) from the Fondo de Investigación Sanitaria, cofinanced by the ERDF. We wish to acknowledge all the patients and volunteers involved in this study for their collaboration. We also acknowledge the BioBank IISPV (PT17/0015/0029) integrated into the Spanish National Biobanks Network. Conflicts of Interest: The authors declare no conflict of interest. References 1. Bray, F.; Ferlay, J.; Soerjomataram, I.; Siegel, R.L.; Torre, L.A.; Jemal, A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 2018, 68, 394–424. [CrossRef] 2. Johnson, D.E.; Burtness, B.; Leemans, C.R.; Lui, V.W.Y.; Bauman, J.E.; Grandis, J.R. Head and neck squamous cell carcinoma. Nat Rev. Dis. Prim. 2020, 6, 92. [CrossRef] [PubMed] 3. Orit, K.-P.; Ziv, G.; Salem, B. Precision medicine in head and neck cancer. Drug Resist. Updates 2018, 40, 13–16. 4. Dando, I.; Pozza, E.D.; Ambrosini, G.; Torrens-Mas, M.; Butera, G.; Mullappilly, N.; Pacchiana, R.; Palmieri, M.; Donadelli, M. Oncometabolites in cancer aggressiveness and tumour repopulation. Biol. Rev. 2019, 94, 1530–1546. [CrossRef] [PubMed] Cancers 2021, 13, 1653 15 of 16 5. Wang, Y.P.; Li, J.T.; Qu, J.; Yin, M.; Lei, Q.Y. Metabolite sensing and signaling in cancer. J. Biol. Chem. 2020, 295, 11938–11946. [CrossRef] 6. Eijkelenkamp, K.; Osinga, T.E.; Links, T.P.; van der Horst-Schrivers, A.N.A. Clinical implications of the oncometabolite succinate in SDHx-mutation carriers. Clin. Genet. 2020, 97, 39–53. [CrossRef] 7. Vaupel, P.; Schmidberger, H.; Mayer, A. The Warburg effect: Essential part of metabolic reprogramming and central contributor to cancer progression. Int. J. Radiat. Biol. 2019, 95, 912–919. [CrossRef] 8. La Vecchia, S.; Sebastián, C. Metabolic pathways regulating colorectal cancer initiation and progression. Semin. Cell Dev. Biol. 2020, 98, 63–70. [CrossRef] 9. Nanda, C.; Venkateswaran, S.V.; Patani, N.; Yuneva, M. Defining a metabolic landscape of tumours: Genome meets metabolism. Br. J. Cancer 2020, 122, 136–149. [CrossRef] [PubMed] 10. Yong, C.; Stewart, G.D.; Frezza, C. Oncometabolites in renal cancer. Nat. Rev. Nephrol. 2020, 16, 156–172. [CrossRef] [PubMed] 11. Zhao, Y.; Feng, F.; Guo, Q.H.; Wang, Y.-P.; Zhao, R. Role of succinate dehydrogenase deficiency and oncometabolites in gastrointestinal stromal tumors. World J Gastroenterol. 2020, 26, 5074–5089. [CrossRef] [PubMed] 12. Mikawa, T.; Lleonart, M.E.; Takaori-Kondo, A.; Inagaki, N.; Yokode, M.; Kondoh, H. Dysregulated glycolysis as an oncogenic event. Cell Mol. Life Sci. 2015, 72, 1881–1892. [CrossRef] 13. Schaefer, I.M.; Hornick, J.L.; Bovée, J.V.M.G. The role of metabolic enzymes in mesenchymal tumors and tumor syndromes: Genetics, pathology, and molecular mechanisms. Lab. Investig. 2018, 98, 414–426. [CrossRef] [PubMed] 14. Ortiz-Masiá, D.; Gisbert-Ferrándiz, L.; Bauset, C.; Coll, S.; Mamie, C.; Scharl, M.; Esplugues, J.V.; Alós, R.; Navarro, F.; Cosín-Roger, J.; et al. Succinate Activates EMT in Intestinal Epithelial Cells through SUCNR1: A Novel Protagonist in Fistula Development. Cells 2020, 9, 1104. [CrossRef] [PubMed] 15. Mu, X.; Zhao, T.; Xu, C.; Shi, W.; Geng, B.; Shen, J.; Zhang, C.; Pan, J.; Yang, J.; Hu, S.; et al. Oncometabolite succinate promotes angiogenesis by upregulating VEGF expression through GPR91-mediated STAT3 and ERK activation. Oncotarget 2017, 8, 13174–13185. [CrossRef] [PubMed] 16. Wu, J.Y.; Huang, T.W.; Hsieh, Y.T.; Wang, Y.F.; Yen, C.C.; Lee, G.L.; Yeh, C.C.; Peng, Y.J.; Kuo, Y.Y.; Wen, H.T.; et al. Cancer-Derived Succinate Promotes Macrophage Polarization and Cancer Metastasis via Succinate Receptor. Mol. Cell 2020, 77, 213–227.e5. [CrossRef] 17. Zhang, J.; Zhang, Q.; Yang, Y.; Wang, Q. Association Between Succinate Receptor SUCNR1 Expression and Immune Infiltrates in Ovarian Cancer. Front. Mol. Biosci. 2020, 7, 150. [CrossRef] 18. Serena, C.; Ceperuelo-Mallafré, V.; Keiran, N.; Queipo-Ortuño, M.I.; Bernal, R.; Gomez-Huelgas, R.; Urpi-Sarda, M.; Sabater, M.; Pérez-Brocal, V.; Andrés-Lacueva, C.; et al. Elevated circulating levels of succinate in human obesity are linked to specific gut microbiota. ISME J. 2018, 12, 1642–1657. [CrossRef] 19. Chiesa, F.; Tradati, N.; Mauri, S.; Calabrese, L.; Grigolato, R.; Giugliano, G.; Zurrida, S.; Chiesa, E.; Squadrelli, M.; Viale, G.; et al. Prognostic factors in head and neck oncology: A critical appraisal for use in clinical practice. Anticancer Res. 1998, 18, 4769–4776. 20. Ceperuelo-Mallafre, V.; Llaurado, G.; Keiran, N.; Benaiges, E.; Astiarraga, B.; Martínez, L.; Pellitero, S.; González-Clemente, J.M.; Rodríguez, A.; Fernández-Real, J.M.; et al. Preoperative circulating succinate levels as a biomarker for diabetes remission after bariatric surgery. Diabetes Care 2019, 42, 1956–1965. [CrossRef] 21. Fernández-Veledo, S.; Vendrell, J. Gut microbiota-derived succinate: Friend or foe in human metabolic diseases? Rev. Endocr. Metab. Disord. 2019, 20, 439–447. [CrossRef] 22. Li, X.; Xie, L.; Qu, X.; Zhao, B.; Fu, W.; Wu, B.; Wu, J. GPR91, a critical signaling mechanism in modulating pathophysiologic processes in chronic illnesses. FASEB J. 2020, 34, 13091–13105. [CrossRef] [PubMed] 23. Lussey-Lepoutre, C.; Bellucci, A.; Morin, A.; Buffet, A.; Amar, L.; Janin, M.; Ottolenghi, C.; Zinzindohoué, F.; Autret, G.; Burnichon, N.; et al. In vivo detection of succinate by magnetic resonance spectroscopy as a hallmark of SDHx mutations in paraganglioma. Clin. Cancer Res. 2016, 22, 1120–1129. [CrossRef] [PubMed] 24. Hayashi, Y.; Yokota, A.; Harada, H.; Huang, G. Hypoxia/pseudohypoxia-mediated activation of hypoxia-inducible factor-1α in cancer. Cancer Sci. 2019, 110, 1510–1517. [CrossRef] [PubMed] 25. Hu, J.; Li, T.; Du, X.; Wu, Q.; Le, Y.Z. G protein-coupled receptor 91 signaling in diabetic retinopathy and hypoxic retinal diseases. Vis. Res. 2017, 139, 59–64. [CrossRef] [PubMed] 26. Tannahill, G.M.; Curtis, A.M.; Adamik, J.; Palsson-McDermott, E.M.; McGettrick, A.F.; Goel, G.; Frezza, C.; Bernard, N.J.; Kelly, B.; Foley, N.H.; et al. Succinate is an inflammatory signal that induces IL-1β through HIF-1α. Nature 2013, 496, 238–242. [CrossRef] 27. Moreno, C.; Santos, R.M.; Burns, R.; Zhang, W.C. Succinate dehydrogenase and ribonucleic acid networks in cancer and other diseases. Cancers 2020, 12, 3237. [CrossRef] [PubMed] 28. MacFarlane, J.; Seong, K.C.; Bisambar, C.; Madhu, B.; Allinson, K.; Marker, A.; Warren, A.; Park, S.M.; Giger, O.; Challis, B.G.; et al. A review of the tumour spectrum of germline succinate dehydrogenase gene mutations: Beyond phaeochromocytoma and paraganglioma. Clin. Endocrinol. 2020, 93, 528–538. [CrossRef] 29. Li, J.; Liang, N.; Long, X.; Zhao, J.; Yang, J.; Du, X.; Yang, T.; Yuan, P.; Huang, X.; Zhang, J.; et al. SDHC-related deficiency of SDH complex activity promotes growth and metastasis of hepatocellular carcinoma via ROS/NFκB signaling. Cancer Lett. 2019, 461, 44–55. [CrossRef] Cancers 2021, 13, 1653 16 of 16 30. Schöpf, B.; Weissensteiner, H.; Schäfer, G.; Fazzini, F.; Charoentong, P.; Naschberger, A.; Rupp, B.; Fendt, L.; Bukur, V.; Giese, I.; et al. OXPHOS remodeling in high-grade prostate cancer involves mtDNA mutations and increased succinate oxidation. Nat. Commun. 2020. [CrossRef] 31. Dalla Pozza, E.; Dando, I.; Pacchiana, R.; Liboi, E.; Scupoli, M.; Donadelli, M.; Palmieri, M. Regulation of succinate dehydrogenase and role of succinate in cancer. Semin. Cell Dev. Biol. 2020, 98, 4–14. [CrossRef] [PubMed] 32. Ristic, B.; Bhutia, Y.D.; Ganapathy, V. Cell-surface G-protein-coupled receptors for tumor-associated metabolites: A direct link to mitochondrial dysfunction in cancer. Biochim. Biophys. Acta Rev. Cancer 2017, 1868, 246–257. [CrossRef] [PubMed] 33. Sanz-Pamplona, R.; Berenguer, A.; Cordero, D.; Molleví, D.G.; Crous-Bou, M.; Sole, X.; Paré-Brunet, L.; Guino, E.; Salazar, R.; Santos, C.; et al. Aberrant gene expression in mucosa adjacent to tumor reveals a molecular crosstalk in colon cancer. Mol. Cancer 2008, 13, 28–30. [CrossRef] 34. Ikeda, Y.; Suzuki, T.; Saitou, H.; Ogane, S.; Hashimoto, K.; Takano, N.; Nomura, T. Usefulness of fluorescence visualization-guided surgery for early-stage tongue squamous cell carcinoma compared to iodine vital staining. Int. J. Clin. Oncol. 2020, 25, 1604–1611. [CrossRef] [PubMed] 35. Wozny, A.S.; Alphonse, G.; Cassard, A.; Malésys, C.; Louati, S.; Beuve, M.; Lalle, P.; Ardail, D.; Nakajima, T.; Rodriguez-Lafrasse, C. Impact of hypoxia on the double-strand break repair after photon and carbon ion irradiation of radioresistant HNSCC cells. Sci. Rep. 2020, 10, 21357. [CrossRef] [PubMed] 36. Pardo-Reoyo, S.; Roig-Lopez, J.L.; Yang, E.S. Potential biomarkers for radiosensitivity in head and neck cancers. Ann. Transl. Med. 2016, 4, 24. [CrossRef] [PubMed] 37. Jiang, S.; Yan, W. Succinate in the cancer–immune cycle. Cancer Lett. 2017, 390, 45–47. [CrossRef] 38. Xiang, K.; Jendrossek, V.; Matschke, J. Oncometabolites and the response to radiotherapy. Radiat. Oncol. 2020, 15, 197. [CrossRef] [PubMed]