AKR1B1 and AKR1B10 As Prognostic Biomarkers of Endometrioid Endometrial Carcinomas

Home , AKR1B1, AKR1B10, Prostaglandin F synthase, Sorbitol dehydrogenase

Article AKR1B1 and AKR1B10 as Prognostic Biomarkers of Endometrioid Endometrial Carcinomas

Marko Hojnik 1, Snježana Frkovi´cGrazio 2, Ivan Verdenik 3 and Tea Lanišnik Rižner 1,*

1 Institute of Biochemistry, Faculty of Medicine, University of Ljubljana, 1000 Ljubljana, Slovenia; [email protected] 2 Division of Gynecology, Department of Pathology, University Medical Centre Ljubljana, 1000 Ljubljana, Slovenia; [email protected] 3 Division of Gynecology, Department of Obstetrics and Gynecology, University Medical Centre Ljubljana, 1000 Ljubljana, Slovenia; [email protected] * Correspondence: [email protected]; Tel.: +386-1-5437657; Fax: +386-1-5437641

Simple Summary: We evaluated the potential of AKR1B1 and AKR1B10 as tissue biomarkers of endometrial cancer by assessing the immunohistochemical levels of AKR1B1 and AKR1B10 in tissue paraffin sections from 101 well-characterized patients with endometrioid endometrial cancer and 12 patients with serous endometrial cancer. Significantly higher immunohistochemical levels of AKR1B1 and AKR1B10 were found in adjacent non-neoplastic endometrial tissue compared to endometrioid endometrial cancer. The group of patients with both AKR1B1 and AKR1B10 staining above the median values showed significantly better overall and disease-free survival compared to all other patients. Multivariant Cox analysis recognized a strong AKR1B1 and AKR1B10 staining as a statistically important survival prediction factor in patients with endometrioid endometrial cancer. In contrast, we observed no significant differences in AKR1B1 and AKR1B10 staining in patients with Citation: Hojnik, M.; FrkovićGrazio, serous endometrial cancer. Our results suggest that AKR1B1 and AKR1B10 have protective roles in S.; Verdenik, I.; Rižner, T.L. AKR1B1 endometrioid endometrial cancer and represent prognostic biomarker candidates. and AKR1B10 as Prognostic Biomarkers of Endometrioid Abstract: The roles of aldo-keto reductase family 1 member B1 (AKR1B1) and B10 (AKR1B10) in the Endometrial Carcinomas. Cancers 2021, 13, 3398. https://doi.org/ pathogenesis of many cancers have been widely reported but only briefly studied in endometrial 10.3390/cancers13143398 cancer. To clarify the potential of AKR1B1 and AKR1B10 as tissue biomarkers of endometrial cancer, we evaluated the immunohistochemical levels of AKR1B1 and AKR1B10 in tissue paraffin sections Academic Editors: Valerio Mais and from 101 well-characterized patients with endometrioid endometrial cancer and 12 patients with Michele Peiretti serous endometrial cancer and compared them with the clinicopathological data. Significantly higher immunohistochemical levels of AKR1B1 and AKR1B10 were found in adjacent non-neoplastic Received: 23 May 2021 endometrial tissue compared to endometrioid endometrial cancer. A trend for better survival was Accepted: 3 July 2021 observed in patients with higher immunohistochemical AKR1B1 and AKR1B10 levels. However, no Published: 7 July 2021 statistically significant differences in overall survival or disease-free survival were observed when AKR1B1 or AKR1B10 were examined individually in endometrioid endometrial cancer. However, Publisher’s Note: MDPI stays neutral analysis of AKR1B1 and AKR1B10 together revealed significantly better overall and disease-free with regard to jurisdictional claims in survival in patients with both AKR1B1 and AKR1B10 staining above the median values compared to published maps and institutional affil- all other patients. Multivariant Cox analysis identified strong AKR1B1 and AKR1B10 staining as a iations. statistically important survival prediction factor. Conversely, no significant differences were found in serous endometrial cancer. Our results suggest that AKR1B1 and AKR1B10 play protective roles in endometrioid endometrial cancer and show potential as prognostic biomarkers.

Copyright: © 2021 by the authors. Keywords: endometrial carcinoma; survival; prognosis; immunohistochemistry; biomarker; aldo- Licensee MDPI, Basel, Switzerland. keto reductase family 1 member B1 (AKR1B1); aldo-keto reductase family 1 member B10 (AKR1B10) This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Cancers 2021, 13, 3398. https://doi.org/10.3390/cancers13143398 https://www.mdpi.com/journal/cancers Cancers 2021, 13, 3398 2 of 14

1. Introduction Endometrial cancer (EC) is the most common gynecological cancer diagnosis in developed countries [1]. The prevalence of EC, which is predominantly found in postmenopausal women, is expected to increase in developed countries due to population aging [2] and the obesity epidemic [3,4]. EC is historically divided into two subgroups: type 1 includes estrogen-dependent carcinomas (80%) and type 2 includes estrogen-independent and more aggressive carcinomas (20%) [5]. The histopathological subtypes of EC include endometrioid, serous, clear cell, carcinosarcoma, mucinous, mixed cell, dedifferentiated, undifferentiated, and neuroendocrine carcinoma. The molecular classification of EC, proposed by The Cancer Genome Atlas (TCGA) Research Network, separates EC into four molecular prognostic groups: polymerase epsilon (POLE) ultramutated, microsatellite instability (MSI) hyper- mutated, copy-number-low/p53-wild-type (p53 wt), and copy-number-high/p53-mutated (p53mt) [6]. According to meta-analyses, this molecular classification is important for predicting prognoses and directing treatment strategies [6,7]. Traditionally, the precise surgical staging of EC requires removing the uterus, cervix, adnexa, and pelvic and para-aortic lymph nodes, obtaining pelvic washings, and perform- ing pathological examinations. This enables establishing a precise diagnosis, identifying the extent of the disease, predicting the prognosis, and prescribing adjuvant radiation therapy and/or chemotherapy [1]. Patients with high-risk disease have a higher frequency of para-aortic lymph node metastases, suggesting that para-aortic lymphadenectomy should be performed as a part of the surgical staging in these patients. Since 2014, sentinel lymph node mapping has been incorporated into the management of EC as a means to evaluate the lymph nodes most likely to be involved in metastatic cancer and to limit the extent of surgery and lymphadenectomy [1]. Sentinel lymph node mapping has a good diagnostic accuracy and can safely replace lymphadenectomy in the staging of EC; however, its prognostic potential regarding high-risk EC remains to be evaluated [8–10]. Diagnosing EC is straightforward in most cases. However, sometimes, the poor reproducibility of the histological type and grade classification hinders the prediction of prognosis and selection of optimal treatment. New biomarkers may provide a more accurate and prognostically relevant approach for classifying these tumors. The enzymes of the aldo-keto reductase (AKR) superfamily are involved in a plethora of important biochemical processes. The human members of AKR1B subfamily AKR1B1 and AKR1B10 catalyze NADPH-dependent reductions of carbonyl groups to hydroxyl groups and act on different endogenous and exogenous substrates. AKR1B1 catalyzes the reduction of glucose to sorbitol and plays a role in the polyol pathway, osmoregu- lation [11], prostaglandin PGF2α synthesis, and the protein kinase C pathway, which stimulates nuclear factor kappa B, inflammation, and proliferation [11–14]. AKR1B10 catalyzes the reduction of retinals to retinols, which depletes retinoic acid and leads to important pro-differentiating effects [15]. AKR1B10 is involved in the prenylation of small guanine nucleotide triphosphatases (GTPases) by reducing isoprenyl aldehydes and is thus implicated in cell proliferation [16]. AKR1B10 also regulates fatty acid biosynthesis, which plays an important role in carcinogenesis [17,18]. AKR1B1 and AKR1B10 confer resistance to numerous chemotherapeutics, including daunorubicin, idarubicin, and cisplatin [19,20]. Additionally, both AKR1B1 and AKR1B10 can exert protective actions and detoxify products of lipid peroxidation, e.g., convert cytotoxic carbonyls such as 4-hydroxynonenal to 4-hydroxynonenol [21]. In a limited number of EC samples, we previously observed a down-regulation of AKR1B1 mRNA and protein levels and up-regulation of AKR1B10 mRNA and down- regulation of AKR1B10 protein levels, compared to adjacent nontumor tissues [22]. In high-grade tumors, we observed significantly decreased ratios of AKR1B10 mRNA levels in tumor tissues versus adjacent control tissues [23–27]. To clarify the potential of AKR1B1 and AKR1B10 as tissue biomarkers of EC, we here immunohistochemically evaluated AKR1B1 Cancers 2021, 13, 3398 3 of 14

and AKR1B10 levels in tissue parafﬁn sections from a larger group of well-characterized patients with EC and a small group of patients with serous EC.

2. Materials and Methods 2.1. Study Groups This retrospective study included 113 patients with EC, 101 with endometrioid EC and 12 with serous EC, who were diagnosed form 2003 and 2014 and have been enrolled in our previous studies [22,28,29]. Parafﬁn-embedded tissue samples of the primary tumors were examined, and clinical and histopathological data were collected (Table1) Tables S1–S5).

Table 1. Clinical and histopathological data of the patients with EC.

Characteristic Detail Datum Age (y) (n = 113) Mean ± SD 63.6 ± 10.1 Weight (kg) (n = 108) a Mean ± SD 82.4 ± 17.2 Height (cm) (n = 104) a Mean ± SD 162.0 ± 5.5 Body mass index (kg/m2)(n = 104) a Mean ± SD 31.5 ± 6.7 Menopausal status (n = 108) (n (%)) a Postmenopausal 94 (87.0) Parity status (n = 106) (n (%)) a Multiparous 93 (87.7) Smoking status (n = 59) (n (%)) a Smokers 5 (8.5) Histological type (n = 113) (n (%)) Endometrioid 101 (84.6) Serous 12 (9.8) Histological grade (n = 101) (n (%)) G1 65 (62.5) G2 25 (24.0) G3 11 (10.6) Myometrial invasion (n = 113) (n (%)) <50% 84 (74.3) ≥50% 29 (25.7) Lymphovascular invasion (n = 113) (n (%)) 30 (26.5) FIGO stage (n = 108) (n (%)) a I–II 97 (89.8) III–IV 11 (10.1) Surgical resection (n = 113) (n (%)) R0 108 (95.6) Lymphadenectomy (n = 113) (n (%)) 105 (92.9) Adjuvant chemotherapy (n = 113) (n (%)) Paclitaxel/Carboplatin 7 (6.2) Carboplatin 1 (0.1) Adjuvant radiotherapy (n = 113) (n (%)) 36 (31.9) Overall survival (n = 113) (y) Range 0.4–17.6 Median 7.6 Disease-free survival (n = 113) (y) Range 0.2–17.6 Median 6.95 a Cases with missing data. N: number of patients; y: years elapsed since initial diagnosis; SD: standard deviation.

We evaluated the immunohistochemical (IHC) staining of AKR1B1 and AKR1B10 in EC. The results were compared with adjacent non-neoplastic endometrial tissue from the same patients when this tissue was available (n = 70). We examined the correlations with the other clinicopathological data, including overall survival, disease-free survival, stage of the disease, lymphovascular invasion, parous status, menopausal status, and smoking.

2.2. Immunohistochemistry Immunohistochemistry was performed to visualize specific antigens on formalin- fixed, paraffin-embedded tissue samples of EC. All histopathological data and samples were collected from the archives of the University Medical Centre Ljubljana, Division of Gynecology, Department of Pathology, and were revised before inclusion in this study. From each paraffin-embedded tissue block, 3–5 µm thick paraffin sections were cut and placed onto glass slides (Superfrost Plus; Thermo Scientific, Leicestershire, UK). The sections were dehydrated in a ventilation slide-drying oven for 60 min at 60 ◦C. IHC staining for AKR1B1 and AKR1B10 was carried out with an automatic system (BenchMark Cancers 2021, 13, 3398 4 of 14

Ultra, Ventana, Basel, Switzerland) using detection kits (OptiView DAB; Ventana; Basel, Switzerland; cat. no. 760-700), following the manufacturer’s instructions. Histological sections were deparafﬁnized (EZPrep solution; Ventana, Basel, Switzer- land; cat. no. 950-102) for 4 min at 72 ◦C. Epitope retrieval was achieved using a Tris-based buffer, pH 8.5, for 24 min (for AKR1B10) and 32 min (for AKR1B1), at 95 ◦C (cell condition- ing solution CC1; Ventana, Basel, Switzerland; cat. no. 950-124). The slides were incubated with the primary antibodies anti-AKR1B1 (Abcam, Cambridge, UK; cat. no. ab62795, lot: GR64780-2) and anti-AKR1B10 (Abcam, Cambridge, UK; cat. no. ab96417, lot: GR13314-31) for 32 min (diluted 1:200 in Antibody Diluent, Dako, Agilent, Santa Clara, USA, cat. no. S080983-2). The positive and negative controls for AKR1B1 and AKR1B10 were normal liver tissue (hepatocytes, ductal epithelium, and connective tissue), high-grade serous ovarian cancer (cancer cells, stroma), and uterine tissue (see validation of antibodies in Table2). The assessment was conducted in blinded conditions and was based on the percentage of stained cells in the whole tumor area. Cytoplasmic and nuclear staining was considered as positive reaction. The stained tissue sections were evaluated by the author M.H.

Table 2. Antibody description and validation.

Antibody Information Manufacturer, Species Raised, Peptide/Protein Antibody Catalogue Number, Antigen Sequence Monoclonal, Dilution Target Lot Number Polyclonal aa 300 to the Aldo-keto C-terminus Abcam, Cambridge, UK, Polyclonal rabbit Anti-AKR1B1 reductase family 1 (conjugated to 1:200 ab62795,GR64780-2 antibody member B1 keyhole limpet hemocyanin) Aldo-keto fragment Abcam, Cambridge, UK, Polyclonal rabbit Anti-AKR1B10 reductase family 1 corresponding to 1:200 ab96417,GR13314-31 antibody member B10 aa 1-286 Antibody Validation Published validation by our research team [22] Current Validation Positive controls for AKR1B1: Kupffer cells, lymphocytes, high-grade serous ovarian cancer cells Positive controls for AKR1B10: Hepatocytes, ductal liver epithelium, lymphocytes, high-grade serous ovarian cancer cells Negative controls for AKR1B1: hepatocytes, ﬁbrous tissue Negative controls for AKR1B10: ﬁbrous tissue AKR1B1: aldo-keto reductase family 1 member B1; AKR1B10: aldo-keto reductase family 1 member B10.

2.3. Statistics Student’s t-tests were used to compare the mean expressions of AKR1B1 and AKR1B10 (as percentages of positive cancer cells) between EC and adjacent non-neoplastic endometrial tissue. Linear regression analysis was performed to evaluate the correlations between AKR1B1 and AKR1B10 IHC expression and weight, height, and body mass index. The Kaplan–Meier method and Cox regression analysis were performed to evaluate survival. The Mann–Whitney U and Kruskal–Wallis tests were used to test for correlations between percentages of AKR1B1 and AKR1B10 expression and the other clinical data. All analyses were executed using SPSS software (IBM version 22, Armonk, NY, USA).

2.4. Ethical Issues This retrospective study was approved by the National Medical Ethics Committee of the Republic of Slovenia (0120-701/2017-6). Cancers 2021, 13, 3398 5 of 14

Cancers 2021, 13, x 5 of 16 3. Results 3.1. Demographic and Histopathological Characteristics of Patients The study group (n = 113) included 101 endometrioid (65% grade I) and 12 serous 2.4.endometrial Ethical Issues carcinomas. Most patients (n = 96) were diagnosed with FIGO stage I disease (TableThis S1). retrospective The 5-year overallstudy was survival approved was 90.3%.by the National Follow-up Medical data were Ethics obtained Committee for allof the113 Republic patients andof Slovenia ranged from(0120 0.4-701/2017 to 17.6-6). years (median = 8.6 years).

3.3.2. Results AKR1B1 and AKR1B10 Expression Levels in Endometrioid and Serous EC 3.1. DemographicIHC reactions and were Histopathological positive for Characteristics AKR1B1 and of AKR1B10 Patients within the cytoplasm and nucleus of the epithelial cancer cells in both endometrioid and serous EC as well as in the The study group (n = 113) included 101 endometrioid (65% grade I) and 12 serous endothelium. Positive cytoplasmic and nuclear staining of tumor cells was calculated as endometrial carcinomas. Most patients (n = 96) were diagnosed with FIGO stage I disease percentages. In all cases in which positive reaction cytoplasmic staining was seen, out of (Table S1). The 5-year overall survival was 90.3%. Follow-up data were obtained for all those cases, some cases also had positive nuclear reaction; in most cases, nuclear staining 113 patients and ranged from 0.4 to 17.6 years (median = 8.6 years). was weaker than cytoplasmic staining. The median and mean percentages of AKR1B1- positive cancer cells were 50.0% and 51.6% for endometrioid EC and 50.0% and 58.8% for 3.2. AKR1B1 and AKR1B10 Expression Levels in Endometrioid and Serous EC serous EC. The median and mean percentages of AKR1B10-positive cancer cells were 80.0% and 64.1%IHC reactions for endometrioid were positive EC and for 95.0%AKR1B1 and and 73.8% AKR1B10 for serous within EC ( Figurethe cytoplasm1 ). Adjacent and nucleusnon-neoplastic of the epithelial endometrial cancer tissue cells revealed in both endometrioid strong positive and reactions serous EC within as well the as nucleus in the enanddothelium. cytoplasm Positive of endometrial cytoplasmic glands, and withnuclear median staining and of mean tumor percentages cells was calculated of AKR1B1- as percentagepositive epithelials. In all cellscases of in 100.0% whichand positive 90.0% reaction in patients cytoplasm with endometrioidic staining was EC seen, and100.0% out of thoseand 88.3% cases, insome patients cases withalso had serous positive EC; the nuclear myometrium reaction; andin most endometrial cases, nuclear stroma stai werening wasnegative weaker (Figure than2 cytoplasm). The medianic staining. and mean The percentagesmedian and ofmean AKR1B10-positive percentages of epithelialAKR1B1- positivecells in adjacent cancer cells non-neoplastic were 50.0% endometrial and 51.6% for tissue endometrioid were 100.0% EC and and 94.1% 50.0% in and patients 58.8% with for serousendometrioid EC. The EC median and 100% and and mean 94.4% percentage in patientss of with AKR1B10 serous-positive EC; the myometrium cancer cells were and 80.0endometrial% and 64.1 stroma% for wereendometrioid negative (FigureEC and3 95.0). % and 73.8% for serous EC (Figure 1). Ad- jacent non-neoplastic endometrial tissue revealed strong positive reactions within the nu- cleus3.3. Comparison and cytoplasm of AKR1B1 of endometrial and AKR1B10 glands, Expression with Levels median in Endometrioid and mean and percentag Serous ECes and of Adjacent Non-Neoplastic Endometrial Tissue AKR1B1-positive epithelial cells of 100.0% and 90.0% in patients with endometrioid EC and 100Non-neoplastic.0% and 88.3 endometrial% in patients tissue with adjacent serous toEC the; the carcinomas myometrium was available and endometrial for IHC stromaanalysis were in 70 negative out of ( theFigure 113 2) cases.. The median Signiﬁcantly and mean higher percentage percentagess of AKR1B10 (mean ±-positiveSD) of epithelialAKR1B1-positive cells in epithelialadjacent non cells-neoplastic were observed endometri in adjacental tissue non-neoplastic were 100.0% endometrial and 94.1% tis-in patientssue (90.1% with± 20.3%endometrioid) compared EC toand endometrioid 100% and 94.4 EC% (55.8% in patients± 41.8%) with (p

Figure 1. AKR1B1 and AKR1B10 immunohistochemical staining. The percentage of positive cancer cells is shown for (a) endometrioidFigure 1. AKR1B1 EC andand (AKRb) serous1B10 EC.immunohistochemical AKR1B1: aldo-keto staining. reductase The family percentage 1 member of positive B1; AKR1B10: cancer cells aldo-keto is shown reductase for (a) familyendometrioid 1 member EC B10;and IQR:(b) serous interquartile EC. AKR1B1: range; N:aldo number-keto reductase of patients. family 1 member B1; AKR1B10: aldo-keto reductase family 1 member B10; IQR: interquartile range; N: number of patients.

Cancers 2021, 13, 3398 6 of 14 Cancers 2021, 13, x 6 of 16

Figure 2. Representative immunohistochemical staining for AKR1B1. Samples of endometrioid EC Figure(a–c), serous 2. Representative EC (d–f), non immunohistochemical-neoplastic endometrial staining tissue (g for), control AKR1B1. liver Samples tissue (h) of and endometrioid control EChigh (a–-cgrade), serous serous EC ovarian (d–f), non-neoplastic cancer (i). Upper endometrial half of panels tissue: 50× (g ),magnification control liver; tissuelower (halfh) and of panels: control high-gradethe framed serous area from ovarian the upper cancer half (i). of Upper the panel half of (200× panels: magnification 50× magniﬁcation;). Hematoxylin lower and half eosin of panels: the(HE) framed stained area sections from the (Figure upper S1). half of the panel (200× magniﬁcation). Hematoxylin and eosin (HE) stained sections (Figure S1).

Similar results were observed for AKR1B10 staining. Signiﬁcantly higher mean percentages of positive epithelial cells (mean ± SD) (94.1% ± 17.4%) were found in adjacent non-neoplastic endometrial tissue compared to endometrioid EC (65.5% ± 35.4%) (p < 0.0001; Figure5a). Similarly, as with AKR1B1, also AKR1B10 staining did not show signiﬁcant differences between adjacent non-neoplastic endometrial tissue (94.4% ± 16.7%) and serous EC (78.9% ± 39.1%) (p = 0.17; Figure5b).

Cancers 2021, 13, 3398 7 of 14 Cancers 2021, 13, x 7 of 16

FigureFigure 3. 3.Representative Representative immunohistochemical staining staining for for AKR1B10. AKR1B10. Samples Samples of endometrioid of endometrioid ECEC ( a(–ac–),c), serous serous EC EC ( d(d––f),f), non-neoplastic non-neoplastic endometrialendometrial tissuetissue ((gg),), controlcontrol liver tissue ( h)),, and control high-gradecontrol high serous-grade ovarian serous cancerovarian (i cancer). Upper (i). half Upper of panels: half of 50panels× magniﬁcation;: 50× magnification lower; halflower of half panels: of thepanels: framed the area framed from area the from upper the half upper of the half panel of the (200 panel× magniﬁcation). (200× magnification Hematoxylin). Hematoxylin and eosin and (HE) stainedeosin (HE) sections stained (Figure sections S2). (Figure S2).

3.4.3.3. The Comparison Correlation of AKR1B1 of AKR1B1 and and AKR1B10 AKR1B10 Expression Expression Levels Levels in E withndometrioid Survival and Serous EC and ToAdjacent evaluate Non AKR1B1-Neoplastic and Endometria AKR1B10l Tissue expression in relation to the clinicopathological data, endometrioidNon-neoplastic and endometrial serous EC tissue were dividedadjacent intoto the two carcinoma groups usings was theavailable median for values IHC ofanalysis AKR1B1 in or 70 AKR1B10 out of the expression 113 cases. as Significantly the threshold higher values. percentages The Kaplan–Meier (mean ± and SD) Cox of survivalAKR1B1 models-positive were epithelial used forcells the were survival observed studies. in adjacent As shown non-neoplastic by the survival endometri curvesal (Figuretissue 6(90.1), there% ± 20.3 were%) no compared statistical to differencesendometrioid in overallEC (55.8 survival% ± 41.8% between) (p < 0.0001 the; groups Figure above4a). N ando significant below the differences median percentages were observed of AKR1B1- between or adjacent AKR1B10-positive non-neoplastic cancer endome- cells in endometrioidtrial tissue (88.3 or% serous ± 33.2 EC.%) and However, serous thereEC (63.9 was% a± trend42.6%) for (p a= higher0.09; Figure survival 4b). of patients with higher percentages of AKR1B1- and AKR1B10-positive cells. When groups were separated into tertiles, quartiles, and quintiles, no signiﬁcant differences were observed.

Cancers 2021, 13, x 8 of 16

Cancers 2021, 13, x 8 of 16 Cancers 2021, 13, 3398 8 of 14

Figure 4. AKR1B1 immunohistochemical expression in EC and the adjacent non-neoplastic endometrial tissue. Before and after graphs show the expression of AKR1B1 in endometrial cancer tissue and its paired adjacent non-neoplastic endometrial tissue. (a) Adjacent non-neoplastic endometrial tissue (n = 61, mean = 90.1, median = 100, IQR = 10) compared to endometrioid EC (n = 61, mean = 55.8, median = 70.0, IQR = 90.0). (b) Adjacent non-neoplastic endometrial tissue (n = 9, mean = 88.3, median = 100, IQR = 2.5) compared to serous EC (n = 9, mean = 63.9, median = 90.0, IQR = 85.0). AKR1B1: aldo-keto reductase family 1 member B1; IQR: interquartile range; N: number of patients.

Figure 4. AKR1B1 immunohistochemical expression in EC and the adjacent non-neoplastic endome- Figuretrial 4. Similar tissue.AKR1B1 Before results immunohistochemical and were after observed graphs show expression for the AKR1B10 expression in EC staining and of AKR1B1the adjacent. Significantly in endometrial non-neoplastic higher cancer meanendo- tissue per- and metrial tissue. Before and after graphs show the expression of AKR1B1 in endometrial cancer tis- centageits paireds of adjacent positive non-neoplastic epithelial cells endometrial (mean tissue.± SD) ((94.1a) Adjacent% ± 17.4 non-neoplastic%) were found endometrial in adjacent tissue sue and its paired adjacent non-neoplastic endometrial tissue. (a) Adjacent non-neoplastic endo- non(n =-neoplastic61, mean = endometrial90.1, median = tissue 100, IQR compared = 10) compared to endometrioid to endometrioid EC (65.5 EC (n% = ±61, 35.4mean%) = (p 55.8 < , metri0.0001al tissue; Figure (n = 61, 5a mean). Similarly, = 90.1, median as with = 100, AKR1B1 IQR = ,10) also compared AKR1B10 to endometrioid staining did EC not (n show = 61, sig- meanmedian = 55.8, median = 70.0, IQR= 70.0, = IQR 90.0). = 90.0). (b) Adjacent (b) Adjacent non-neoplastic non-neoplastic endometrial endometri tissueal tissue (n = (n9, = mean 9, = 88.3, nificant differences between adjacent non-neoplastic endometrial tissue (94.4% ± 16.7%) meanmedian = 88.3, median = 100, IQR = 100 =, 2.5)IQR compared = 2.5) compared to serous to serous EC (n EC = 9, (n mean = 9, mean = 63.9, = 63.9, median median = 90.0, = 90.0, IQR = 85.0). IQRand =AKR1B1: 85.0). serous AKR1B1: aldo-keto EC (78.9 aldo reductase%-keto ± 39.1 reductase% family) (p = 1family 0.17 member; Figure 1 member B1; IQR:5b). B1; interquartile IQR: interquartile range; N range;: number N: num- of patients. ber of patients.

Similar results were observed for AKR1B10 staining. Significantly higher mean percentages of positive epithelial cells (mean ± SD) (94.1% ± 17.4%) were found in adjacent non-neoplastic endometrial tissue compared to endometrioid EC (65.5% ± 35.4%) (p < 0.0001; Figure 5a). Similarly, as with AKR1B1, also AKR1B10 staining did not show significant differences between adjacent non-neoplastic endometrial tissue (94.4% ± 16.7%) and serous EC (78.9% ± 39.1%) (p = 0.17; Figure 5b).

Figure 5. AKR1B10 immunohistochemical expression in EC and the adjacent non-neoplastic endome- Figuretrial tissue. 5. AKR1B10 Before andimmunohistochemical after graphs show theexpression expression in EC of AKR1B1 and the inadjacent endometrial non-neoplastic cancer tissue endo- and metriits pairedal tissue adjacent. Before non-neoplastic and after graphs endometrial show the tissue. expression (a) Adjacent of AKR1B1 non-neoplastic in endometrial endometrial cancer tis- tissue sue and its paired adjacent non-neoplastic endometrial tissue. (a) Adjacent non-neoplastic endo- (n = 61, mean = 94.1, median = 100, IQR = 0) compared to endometrioid EC (n = 61, mean = 65.5, metrial tissue (n = 61, mean = 94.1, median = 100, IQR = 0) compared to endometrioid EC (n = 61, median = 80.0, IQR = 75.0). (b) Adjacent non-neoplastic endometrial tissue (n = 9, mean = 94.4, mean = 65.5, median = 80.0, IQR = 75.0). (b) Adjacent non-neoplastic endometrial tissue (n = 9, meanmedian = 94.4, = 100, median IQR = = 100, 0) compared IQR = 0) compared to serous to EC serous (n = 9,EC mean (n = 9, = mean 78.9, median= 78.9, median = 100, = IQR 100, = IQR 47.5). = AKR1B10:47.5). AKR1B10: aldo-keto aldo reductase-keto reductase family 1family member 1 member B10; IQR: B10; interquartile IQR: interquartile range; N range;: number N: number of patients. of patients. Similarly, the disease-free survival curves did not differ significantly between groups Figure3.4.above 5. The AKR1B10 andCorrelation below immunohistochemical theof AKR1B1 median and percentages AKR1B10 expression of Expression AKR1B1-in EC and Levelsthe or AKR1B10-positiveadjacent with Survival non-neoplastic cancer endo- cells in metriendometrioidal tissue. Before or and serous after graphs EC (Figure show7 ).the expression of AKR1B1 in endometrial cancer tis- To evaluate AKR1B1 and AKR1B10 expression in relation to the clinicopathological sue and itsFor paired endometrioid adjacent non- EC,neoplastic we also endometrial performed tissue. survival (a) Adjacent analysis, non-neoplastic which included endo- IHC data, endometrioid and serous EC were divided into two groups using the median values metristainingal tissue ( forn = both61, mean AKR1B1 = 94.1, andmedian AKR1B10. = 100, IQR Cases = 0) compared were stratified to endometrio into twoid groups EC (n = according61, of AKR1B1 or AKR1B10 expression as the threshold values. The Kaplan–Meier and Cox meanto = the65.5, median median values= 80.0, IQR of AKR1B1 = 75.0). (b and) Adjacent AKR1B10. non-neoplastic The first endometri group includedal tissue cases (n = 9, with both meanAKR1B1 = 94.4, median and AKR1B10 = 100, IQR staining= 0) compared above to theserous median. EC (n = The 9, mean second = 78.9, group median included = 100, IQR all other = 47.5).cases AKR1B10: with either aldo-keto or bothreductase AKR1B1 family and 1 member AKR1B10 B10; stainingIQR: interquartile below the range; median. N: number Wefound of patients.significantly better overall survival (p = 0.029) and disease-free survival (p = 0.022) in

the ﬁrst group compared to the second group (Figure8). In addition, when we limited 3.4. The Correlation of AKR1B1 and AKR1B10 Expression Levels with Survival analysis only to low-risk endometrioid EC (grade 1 and 2) and to low FIGO stage (FIGO I–II),To evaluate we found AKR1B1 signiﬁcantly and AKR1B10 better overall express survivalion in relation (p = 0.023 to the and clinicopathologicalp = 0.020) and disease- data,free endometrioid survival (p and= 0.022 serous and ECp =were 0.014) divided in the into group two with groups AKR1B1 using andthe median AKR1B10 value stainings of AKR1B1above theor AKR1B10 median compared expression to as group the threshold that included value alls. The other Kaplan cases– (FiguresMeier and S3 Cox and S4, Tables S6 and S7).

Cancers 2021, 13, x 9 of 16

survival models were used for the survival studies. As shown by the survival curves (Fig- ure 6), there were no statistical differences in overall survival between the groups above and below the median percentages of AKR1B1- or AKR1B10-positive cancer cells in endo- Cancers 2021, 13, 3398 metrioid or serous EC. However, there was a trend for a higher survival of patients with 9 of 14 higher percentages of AKR1B1- and AKR1B10-positive cells. When groups were separated into tertiles, quartiles, and quintiles, no significant differences were observed.

Figure 6. OverallFigure survival 6. Overall curves survival for AKR1B1 curves andfor AKR1B1 AKR1B10 and inAKR1B10 patients in patients with endometrioid with endometrioid and serousand EC. AKR1B1 in serous EC. AKR1B1 in patients with (a) endometrioid EC (p = 0.27) and (b) serous EC (p = 0.38). patients with (a) endometrioid EC (p = 0.27) and (b) serous EC (p = 0.38). AKR1B10 in patients with (c) endometrioid EC AKR1B10 in patients with (c) endometrioid EC (p = 0.20) and (d) serous EC (p = 0.36). The groups (p = 0.20) and (d) serousare separated EC (p =according 0.36). The to the groupsir median are separated values. Time according on x-axis torepresents their median time elapsed values. since Time ini- on x-axis represents Cancers 2021, 13,time x elapsed sincetial initial diagnosis. diagnosis. AKR1B1: AKR1B1: aldo-keto aldo-keto reductase family reductase 1 member family B1; 1 AKR1B10: member aldo B1;- AKR1B10:keto reductase10 of aldo-keto 16 reductase

family 1 member B10.family 1 member B10.

Similarly, the disease-free survival curves did not differ significantly between groups above and below the median percentages of AKR1B1- or AKR1B10-positive cancer cells in endometrioid or serous EC (Figure 7).

Figure 7. Disease-free survival curves for AKR1B1 and AKR1B10 in endometrioid and serous EC. AKR1B1 in patients Figure 7. Disease-free survival curves for AKR1B1 and AKR1B10 in endometrioid and serous EC. with (a) endometrioidAKR1B1 EC in (patients = 0.38) with and (a ()b endometrioid) serous EC EC (p = (p0.86). = 0.38) AKR1B10 and (b) serous in patients EC (p = 0.86 with). AKR1B10 (c) endometrioid in EC (p = 0.14) and (d) serous ECpatients (p = 0.08). with (c The) endometrioid groups are EC separated (p = 0.14) and according (d) serous to EC their (p = median 0.08). The values. groups Timeare separated on x axis represents time elapsed since initialaccording diagnosis. to their AKR1B1: median values. aldo-keto Time on reductase x axis represents family 1time member elapsed B1; since AKR1B10: initial diagnosis. aldo-keto reductase family 1 member B10. AKR1B1: aldo-keto reductase family 1 member B1; AKR1B10: aldo-keto reductase family 1 member B10.

For endometrioid EC, we also performed survival analysis, which included IHC staining for both AKR1B1 and AKR1B10. Cases were stratified into two groups according to the median values of AKR1B1 and AKR1B10. The first group included cases with both AKR1B1 and AKR1B10 staining above the median. The second group included all other cases with either or both AKR1B1 and AKR1B10 staining below the median. We found significantly better overall survival (p = 0.029) and disease-free survival (p = 0.022) in the first group compared to the second group (Figure 8). In addition, when we limited analysis only to low-risk endometrioid EC (grade 1 and 2) and to low FIGO stage (FIGO I–II), we found significantly better overall survival (p = 0.023 and p = 0.020) and disease-free survival (p = 0.022 and p = 0.014) in the group with AKR1B1 and AKR1B10 staining above the median compared to group that included all other cases (Figures S3 and S4, Tables S6 and S7).

Cancers 2021, 13, 3398 10 of 14

Table 3. Data on multivariant analysis of independent factors predictive of survival.

Overall Survival Signiﬁcance Hazard Ratio Conﬁdence Interval Lymphovascular invasion p = 0.001 3.8 1.8–8.2 Risk group (EC grade I–II vs. EC III or SC) p = 0.005 2.9 1.4–6.2 AKR1B1 and AKR1B10 expression above the median values p = 0.036 0.4 0.1–0.9 FIGO (I–II vs. III–IV) p = 0.103 2.2 0.9–5.5 Disease-free survival Lymphovascular invasion p = 0.003 3.2 1.5–6.7 Risk group (EC grade I–II vs. EC III or SC) p = 0.004 2.9 1.4–6.1 Cancers 2021, 13, x AKR1B1 and AKR1B10 expression above the median values p = 0.023 0.311 of 16 0.1–0.9 FIGO (I–II vs. III–IV) p = 0.126 2.0 0.8–5.0

Figure 8. OverallFigure survival 8. Overall and survival disease-free and disease survival-free in relationsurvival toin AKR1B1relation to and AKR1B1 AKR1B10. and OneAKR1B10. group One includes cases with both AKR1B1group and includes AKR1B10 cases staining with above both AKR1B1 the median and values, AKR1B10 and staining the other above group the includes median cases value withs, and either the AKR1B1 or AKR1B10 belowother the group median includes values cases in endometrioid with either AKR1B1 EC. (a) Overallor AKR1B10 survival below (p = the 0.029) median and (bvalues) disease-free in endome- survival curves trioid EC. (a) Overall survival (p = 0.029) and (b) disease-free survival curves (p = 0.022). Time on x (p = 0.022). Time on x axis represents time elapsed since initial diagnosis. AKR1B1: aldo-keto reductase family 1 member B1; axis represents time elapsed since initial diagnosis. AKR1B1: aldo-keto reductase family 1 member AKR1B10: aldo-keto reductase family 1 member B10. Multivariant Cox analyses of independent prediction factors of overall B1; AKR1B10: aldo-keto reductase family 1 member B10. Multivariant Cox analyses of independ- and disease-freeent prediction survival werefactors performed of overall (Table and disease3). For- overallfree survival survival, were the performed significant (Table prediction 3). For factors, overall in descending order, weresurvival lymphovascular, the significant invasion prediction (p = 0.001), factors risk,group in descending (EC grade order I–II, vs.were EC lymphovascular III or SC) (p = 0.005), invasion and ( AKR1B1p and AKR1B10 expression= 0.001), risk above group the (EC median grade values I–II vs (p. =EC 0.036). III or ForSC) disease-free(p = 0.005), and survival, AKR1B1 the significantand AKR1B10 prediction expres- factors were lymphovascularsion above invasion the (medianp = 0.003), values risk ( groupp = 0.036 (EC). For grade diseas I–IIe vs.-free EC survival III or SC), the ( psignificant= 0.004), andprediction AKR1B1 fac- and AKR1B10 expression abovetors were the medianlymphovascular values (p invasion= 0.023). (p = 0.003), risk group (EC grade I–II vs. EC III or SC) (p = 0.004), and AKR1B1 and AKR1B10 expression above the median values (p = 0.023). 3.5. The Correlation of AKR1B1 and AKR1B10 Expression Levels with Other Clinical Data Table 3. Data on multivariant analysis of independent factors predictive of survival. No statistically significant correlations were found between the percentages of AKR1B1- Overall Survival or AKR1B10-positive cellsSignificance and the FIGO Hazard stage (I–II Ratio or III–IV)Confidence for endometrioid Interval EC (p = 0.96 Lymphovascular invasionfor AKR1B1 and p = 0.83 forp AKR1B10;= 0.001 Mann–Whitney3.8 U tests)1.8– or8.2 serous EC (p = 0.60 for Risk group (EC grade I–II vs. ECAKR1B1 III or SC) and p = 0.37 for AKR1B10;p = 0.005 Mann–Whitney2.9 U tests).1.4 Grade–6.2 III endometrioid EC AKR1B1 and AKR1B10 expression abovecompared the median to grade values I–II didp not = 0.036 show significantly0.4 different levels0.1–0.9 of AKR1B1 (p = 0.96) or FIGO (I–II vs. III–IV)AKR1B10 (p = 0.29) (Mann–Whitneyp = 0.103 U test).2.2 In patients with0.9 endometrioid–5.5 and serous EC, Disease-free survivalthere were no significant correlations between AKR1B1 or AKR1B10 expression and age, height, weight, body mass index, parous status, menopausal status, smoking status, inva- Lymphovascular invasion p = 0.003 3.2 1.5–6.7 sion in lymph vessels, lymph nodes, or myometrium, or cervix or parametria (Chi-squared Risk group (EC grade I–II vs. EC III or SC) p = 0.004 2.9 1.4–6.1 test, T-test analysis; Tables S8 and S9). AKR1B1 and AKR1B10 expression above the median values p = 0.023 0.3 0.1–0.9 FIGO (I–II vs. III–IV)4. Discussion p = 0.126 2.0 0.8–5.0 In the present study, we assessed IHC AKR1B1 and AKR1B10 levels in a larger cohort 3.5. The Correlationof patients of AKR1B1 with and endometrioid AKR1B10 Expression (n = 101) Levels EC and, with forOther the Clinical first time, Data in a small cohort of No statistically significant correlations were found between the percentages of AKR1B1- or AKR1B10-positive cells and the FIGO stage (I–II or III–IV) for endometrioid EC (p = 0.96 for AKR1B1 and p = 0.83 for AKR1B10; Mann–Whitney U tests) or serous EC (p = 0.60 for AKR1B1 and p = 0.37 for AKR1B10; Mann–Whitney U tests). Grade III endometrioid EC compared to grade I–II did not show significantly different levels of AKR1B1 (p = 0.96) or AKR1B10 (p = 0.29) (Mann–Whitney U test). In patients with endometrioid and serous EC, there were no significant correlations between AKR1B1 or AKR1B10 expression and age, height, weight, body mass index, parous status, menopausal status, smoking status, invasion in lymph vessels, lymph nodes, or myometrium, or cervix or parametria (Chi-squared test, T-test analysis; Tables S8 and S9).

4. Discussion In the present study, we assessed IHC AKR1B1 and AKR1B10 levels in a larger cohort of patients with endometrioid (n = 101) EC and, for the first time, in a small cohort of

Cancers 2021, 13, 3398 11 of 14

patients with serous (n = 12) EC and correlated AKR1B1 and AKR1B10 expression with clinicopathological data. There is an emerging interest in studying the role of AKR in the pathogenesis of different pathologies [30], including uterine diseases [31,32]. Many studies demonstrated that AKR1B1 and AKR1B10 are involved in different cancers [12,27,33–36]. In some types of cancer, the published studies associated AKR1B1 and AKR1B10 with poor survival of patients, while in other types of cancer, the protective effects were seen. The overexpression of AKR1B1 has been reported in breast, ovarian, cervical, lung, hepatocellular, and rectal cancer [27,37–40], but down-regulation of AKR1B1 has been observed in adenocarcinoma samples compared to adjacent nontumor tissue in endometrial and colorectal cancer [22,24–26]. In gastric carcinoma, oral squamous cell and lung adenocarcinoma AKR1B10 overexpression was associated with significantly poorer prognosis [34,35,41,42], but AKR1B10 was downregulated in colorectal cancer cells compared to the adjacent normal colorectal tissues, and the survival of AKR1B10 negative patients was significantly worse [43]. So far, only three studies examined AKR1B1 or AKR1B10 expression in EC. Yoshitake et al. [44] reported that AKR1B10 is expressed in EC but found no correlation with the clinicopathological features [44]. The other two studies were performed by our group. We first examined AKR1B1 and AKR1B10 mRNA and protein expression in 47 patients with EC [22]. We found lower AKR1B1 mRNA and protein levels in cancerous tissue compared to adjacent control tissue. Furthermore, we found higher AKR1B10 mRNA levels but lower protein levels in cancerous tissue compared to adjacent control tissue, which might be explained by the different protein translation efficiencies in cancer tissues [22]. Further IHC analysis revealed positive staining for AKR1B1 and AKR1B10 in all 22 EC and adjacent control tissue samples [22]. In the second published study, we examined the ratio of AKR1B1 and AKR1B10 mRNA levels in tumor tissues versus adjacent control tissues in relation to the clinicopathological data of 51 patients [23]. We found that body weight and BMI were negatively correlated with AKR1B1 and positively correlated with AKR1B10. Furthermore, AKR1B10 mRNA levels were significantly lower in high-grade compared to low-grade EC. The same study also revealed that age, myometrial invasion, FIGO stage, and lymphovascular invasion were not associated with AKR1B1 or AKR1B10 expression [23]. These previous findings are not completely in line with the present study. As observed previously [22], results at the mRNA level do not always correlate with those at the protein level, which is possibly due to different factors that affect translation. At the protein level, we here confirmed our published Western blot analysis of 30 paired tissue samples [22], as we found higher AKR1B1 and AKR1B10 IHC levels in control endometrium versus adjacent cancerous tissue from 61 patients with endometrioid EC. In the current study, we thus demonstrated that AKR1B1 and AKR1B10 protein levels are significantly lower in endometrioid EC compared to adjacent control endometrial tissues, while no significant differences were observed in serous EC. Although there was a trend of better survival in patients with higher AKR1B1 and AKR1B10 IHC levels, survival studies of endometrioid EC did not show significant differences in overall and disease-free survival when AKR1B1 and AKR1B10 were examined individually. However, the group with both AKR1B1 and AKR1B10 IHC levels above the median values had significantly better overall and disease-free survival compared to other patients. These results indicate that higher AKR1B1 and AKR1B10 levels in cancer tissues correlate with better prognoses of patients with endometrioid EC, suggesting a protective role of AKR1B1 and AKR1B10. Additionally, multivariant Cox analysis identified high AKR1B1 and AKR1B10 expression as an important prediction factor for both overall and disease-free survival. AKR1B1 and AKR1B10 exert a plethora of physiological actions and are involved in prostaglandin synthesis, retinoid metabolism, prenylation, lipid synthesis, and detoxifi- cation of unsaturated carbonyl products of lipid peroxidation [23,31]. Thus, AKR1B1 and Cancers 2021, 13, 3398 12 of 14

AKR1B10 can stimulate inflammation and proliferation and attenuate differentiation and apoptosis. Conversely, the protective role of AKR1B1 and AKR1B10 can be explained by their involvement in detoxifying different products (e.g., 4-hydroxynonenal) of lipid peroxidation, which is a consequence of oxidative stress [45]. Oxidative stress and electrophilic stress are recognized by nuclear erythroid 2-related factor 2 (NRF2), which is a master regulator of numerous antioxidant/detoxifying genes, such as the AKR genes AKR1B1 and AKR1B10. NRF2 binds to and up-regulates the expression of the antioxidant response elements of these genes. This is supported by studies showing that NRF2 inducers elevate AKR1B1 and AKR1B10 expression and that NRF2 signaling is activated by chemicals that produce reactive oxygen species [46–48]. AKR1B1 and AKR1B10 play similar roles in detoxifying lipid peroxidation products but exhibit different catalytic efficiencies for reducing 4-hydroxynonenale [49]. In this study, better survival was observed for patients with high IHC levels (staining above the median values) of both AKR1B1 and AKR1B10. These high levels of AKR1B1 and AKR1B10 were also identified as significant predictive markers for both overall and disease-free survival. This implies that the combined action of these enzymes is needed to exert protective effects. Our data suggest that AKR1B1 and AKR1B10 are involved in the pathogenesis of endometrioid EC; however, their exact roles still need to be determined. In serous EC, no significant differences were found; however, due to a very limited number of cases (n = 12), these results should be considered with caution.

5. Conclusions In summary, our results indicate that AKR1B1 and AKR1B10 may play protective roles in the pathogenesis of endometrioid EC and show prognostic potential, as high levels of both AKR1B1 and AKR1B10 predict better patient survival. These findings are important because examining AKR1B1 and AKR1B10 expression might improve prognostic accuracy and influence on treatment planning for patients with endometrioid EC. The importance of AKR1B1 and AKR1B10 was not confirmed for serous EC, which was most probably due to the limited number of cases. To clarify the complex roles of AKR1B1 and AKR1B10 in the pathogenesis of endometrioid and serous EC, further studies are needed.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/cancers13143398/s1. Table S1: IHC data and tumor data; Table S2: Survival data; Table S3: Pathological data; Table S4: Clinical data 1; Table S5: Clinical data 2; Table S6: Survival analysis for low/high-risk endometrioid EC; Table S7: Survival analysis for FIGO I–II stage and for FIGO III–IV EC; Table S8: Associations between clinical and histopathological data and AKR1B1 and AKR1B10 IHC levels in patients with EC—Part 1; Table S9: Associations between clinical and histopathological data and AKR1B1 and AKR1B10 IHC levels in patients with EC—Part 2; Figure S1: Representative hematoxylin and eosin (HE) stained sections of the same samples as shown in Figure2 in the main text of the manuscript; Figure S2: Representative hematoxylin and eosin (HE) stained sections of the same samples as shown in Figure3 in the main text of the manuscript, Figure S3: Overall survival and disease-free survival in relation to AKR1B1 and AKR1B10 for patients with low/high-risk endometrioid EC. Figure S4: Overall survival and disease-free survival in relation to AKR1B1 and AKR1B10 for patients with FIGO I–II/FIGO III–IV disease. Author Contributions: Conceptualization, T.L.R.; methodology, T.L.R. and S.F.G.; software, M.H. and I.V.; validation, T.L.R., S.F.G., and M.H.; formal analysis, I.V.; investigation, M.H.; resources, T.L.R. and S.F.G.; data curation, S.F.G. and M.H.; writing—original draft preparation, M.H. and T.L.R.; writing—review and editing, T.L.R. and S.F.G.; supervision, T.L.R. and S.F.G.; project administration, M.H.; funding acquisition, T.L.R. All authors have read and agreed to the published version of the manuscript. Funding: This study was supported by the grants J3-8212 and J3-2535 to T.L.R. from the Slovenian Research Agency and a scholarship to M.H. for co-funding of doctoral studies from the Republic of Slovenia. Cancers 2021, 13, 3398 13 of 14

Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki and was approved by the National Medical Ethics Committee of the Republic of Slovenia (0120-701/2017-6). Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data presented in this study are available in supplementary material. Acknowledgments: The authors thank Eva Lasic for editing a draft of this manuscript and Nik Krajnc, Ajda Zemljiˇcand Špela Smrkolj for their assistance in data collection. Conﬂicts of Interest: The authors declare no conﬂict of interest.

References 1. Murali, R.; Delair, D.F.; Bean, S.M.; Abu-Rustum, N.R.; Soslow, R.A. Evolving Roles of Histologic Evaluation and Molecu- lar/Genomic Proﬁling in the Management of Endometrial Cancer. J. Natl. Compr. Cancer Netw. 2018, 16, 201–209. [CrossRef] 2. Moore, K.; Brewer, M.A. Endometrial Cancer: Is This a New Disease? Am. Soc. Clin. Oncol. Educ. Book 2017, 37, 435–442. [CrossRef] 3. Shaw, E.; Farris, M.; McNeil, J.; Friedenreich, C. Obesity and Endometrial Cancer. Recent Results Cancer Res. 2016, 208, 107–136. [CrossRef] 4. Blüher, M. Obesity: Global epidemiology and pathogenesis. Nat. Rev. Endocrinol. 2019, 15, 288–298. [CrossRef][PubMed] 5. Malik, T.Y.; Chishti, U.; Aziz, A.B.; Sheikh, I. Comparison of Risk Factors and survival of Type 1 and Type II Endometrial Cancers. Pak. J. Med. Sci. 2016, 32, 886–890. [CrossRef] 6. Kandoth, C.; Schultz, N.; Cherniack, A.D.; Akbani, R.; Liu, Y.; Shen, H.; Robertson, A.G.; Pashtan, I.; Shen, R.; Benz, C.C.; et al. Integrated genomic characterization of endometrial carcinoma. Nature 2013, 497, 67–73. [CrossRef][PubMed] 7. 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] 8. Ji, Q.; Wang, X.; Jiang, J.; Chen, L. Sentinel lymph node mapping in high-risk endometrial cancer: A systematic review and meta-analysis. Gland. Surg. 2020, 9, 2091–2105. [CrossRef][PubMed] 9. Rossi, E.C.; Kowalski, L.D.; Scalici, J.; Cantrell, L.; Schuler, K.; Hanna, R.K.; Method, M.; Ade, M.; Ivanova, A.; Boggess, J.F. A comparison of sentinel lymph node biopsy to lymphadenectomy for endometrial cancer staging (FIRES trial): A multicentre, prospective, cohort study. Lancet Oncol. 2017, 18, 384–392. [CrossRef] 10. Obermair, A.; Abu-Rustum, N.R. Sentinel lymph node mapping in endometrial cancer—areas where further research is needed. Int. J. Gynecol. Cancer 2020, 30, 283–284. [CrossRef] 11. Schwab, A.; Siddiqui, A.; Vazakidou, M.E.; Napoli, F.; Böttcher, M.; Menchicchi, B.; Raza, U.; Saatci, Ö.; Krebs, A.M.; Ferrazzi, F.; et al. Polyol Pathway Links Glucose Metabolism to the Aggressiveness of Cancer Cells. Cancer Res. 2018, 78, 1604–1618. [CrossRef][PubMed] 12. Khayami, R.; Hashemi, S.R.; Kerachian, M.A. Role of aldo-keto reductase family 1 member B1 (AKR1B1) in the cancer process and its therapeutic potential. J. Cell Mol. Med. 2020, 24, 8890–8902. [CrossRef] 13. Alzamil, H.A.; Pawade, J.; Fortier, M.A.; Bernal, A.L. Expression of the prostaglandin F synthase AKR1B1 and the prostaglandin transporter SLCO2A1 in human fetal membranes in relation to spontaneous term and preterm labor. Front. Physiol. 2014, 5, 272. [CrossRef] 14. Kang, E.S.; Kim, G.H.; Kim, H.J.; Woo, I.S.; Ham, S.A.; Jin, H.; Kim, M.Y.; Lee, J.H.; Chang, K.C.; Seo, H.G.; et al. Nrf2 regulates curcumin-induced aldose reductase expression indirectly via nuclear factor-kappaB. Pharmacol. Res. 2008, 58, 15–21. [CrossRef] 15. Gallego, O.; Ruiz, F.X.; Ardèvol, A.; Domínguez, M.; Alvarez, R.; de Lera, A.R.; Rovira, C.; Farrés, J.; Fita, I.; Parés, X. Structural basis for the high all-trans-retinaldehyde reductase activity of the tumor marker AKR1B10. Proc. Natl. Acad. Sci. USA 2007, 104, 20764–20769. [CrossRef][PubMed] 16. Matsunaga, T.; Wada, Y.; Endo, S.; Soda, M.; El-Kabbani, O.; Hara, A. Aldo-Keto Reductase 1B10 and Its Role in Proliferation Capacity of Drug-Resistant Cancers. Front. Pharmacol. 2012, 3, 5. [CrossRef][PubMed] 17. Ma, J.; Yan, R.; Zu, X.; Cheng, J.M.; Rao, K.; Liao, D.F.; Cao, D. Aldo-keto reductase family 1 B10 affects fatty acid synthesis by regulating the stability of acetyl-CoA carboxylase-alpha in breast cancer cells. J. Biol. Chem. 2008, 283, 3418–3423. [CrossRef] 18. Mounier, C.; Bouraoui, L.; Rassart, E. Lipogenesis in cancer progression (review). Int. J. Oncol. 2014, 45, 485–492. [CrossRef] 19. Zhong, L.; Shen, H.; Huang, C.; Jing, H.; Cao, D. AKR1B10 induces cell resistance to daunorubicin and idarubicin by reducing C13 ketonic group. Toxicol. Appl. Pharmacol. 2011, 255, 40–47. [CrossRef] 20. Matsunaga, T.; Suzuki, A.; Kezuka, C.; Okumura, N.; Iguchi, K.; Inoue, I.; Soda, M.; Endo, S.; El-Kabbani, O.; Hara, A.; et al. Aldo-keto reductase 1B10 promotes development of cisplatin resistance in gastrointestinal cancer cells through down-regulating peroxisome proliferator-activated receptor-γ-dependent mechanism. Chem. Biol. Interact. 2016, 256, 142–153. [CrossRef] 21. Martin, H.J.; Breyer-Pfaff, U.; Wsol, V.; Venz, S.; Block, S.; Maser, E. Puriﬁcation and characterization of akr1b10 from human liver: Role in carbonyl reduction of xenobiotics. Drug Metab. Dispos. 2006, 34, 464–470. [CrossRef][PubMed] 22. Hevir, N.; Sinkovec, J.; Lanišnik Rižner, T. Decreased levels of AKR1B1 and AKR1B10 in cancerous endometrium compared to adjacent non-cancerous tissue. Chem. Biol. Interact. 2013, 202, 226–233. [CrossRef] Cancers 2021, 13, 3398 14 of 14

23. Sinreih, M.; Štupar, S.; Cemažar,ˇ L.; Verdenik, I.; FrkovićGrazio, S.; Smrkolj, Š.; Rižner, T.L. STAR and AKR1B10 are downregulated in high-grade endometrial cancer. J. Steroid Biochem. Mol. Biol. 2017, 171, 43–53. [CrossRef] 24. Uzozie, A.C.; Selevsek, N.; Wahlander, A.; Nanni, P.; Grossmann, J.; Weber, A.; Buffoli, F.; Marra, G. Targeted Proteomics for Multiplexed Verification of Markers of Colorectal Tumorigenesis. Mol. Cell Proteom. 2017, 16, 407–427. [CrossRef][PubMed] 25. Uzozie, A.; Nanni, P.; Staiano, T.; Grossmann, J.; Barkow-Oesterreicher, S.; Shay, J.W.; Tiwari, A.; Buffoli, F.; Laczko, E.; Marra, G. Sorbitol dehydrogenase overexpression and other aspects of dysregulated protein expression in human precancerous colorectal neoplasms: A quantitative proteomics study. Mol. Cell Proteom. 2014, 13, 1198–1218. [CrossRef][PubMed] 26. Kropotova, E.S.; Tychko, R.A.; Zinov’eva, O.L.; Zyrianova, A.F.; Khankin, S.L.; Cherkes, V.L.; Aliev, V.A.; Beresten, S.F.; Oparina, N.I.U.; Mashkova, T.D. Downregulation of AKR1B10 gene expression in colorectal cancer. Mol. Biol. 2010, 44, 243–250. [CrossRef] 27. Reddy, K.A.; Kumar, P.U.; Srinivasulu, M.; Triveni, B.; Sharada, K.; Ismail, A.; Reddy, G.B. Overexpression and enhanced specific activity of aldoketo reductases (AKR1B1 & AKR1B10) in human breast cancers. Breast 2017, 31, 137–143. [CrossRef] 28. Knific, T.; Osredkar, J.; Smrkolj, Š.; Tonin, I.; Vouk, K.; Blejec, A.; FrkovićGrazio, S.; Rižner, T.L. Novel algorithm including CA-125, HE4 and body mass index in the diagnosis of endometrial cancer. Gynecol. Oncol. 2017, 147, 126–132. [CrossRef] 29. Sinreih, M.; Knific, T.; Anko, M.; Hevir, N.; Vouk, K.; Jerin, A.; FrkovićGrazio, S.; Rižner, T.L. The Significance of the Sulfatase Pathway for Local Estrogen Formation in Endometrial Cancer. Front. Pharmacol. 2017, 8, 368. [CrossRef] 30. Penning, T.M. The aldo-keto reductases (AKRs): Overview. Chem. Biol. Interact. 2015, 234, 236–246. [CrossRef] 31. Rižner, T.L. Enzymes of the AKR1B and AKR1C Subfamilies and Uterine Diseases. Front. Pharmacol. 2012, 3, 34. [CrossRef] [PubMed] 32. Hojnik, M.; Kenda Šuster, N.; Smrkolj, Š.; FrkovićGrazio, S.; Verdenik, I.; Rižner, T.L. AKR1C3 Is Associated with Better Survival of Patients with Endometrial Carcinomas. J. Clin. Med. 2020, 9, 4105. [CrossRef] 33. Taskoparan, B.; Seza, E.G.; Demirkol, S.; Tuncer, S.; Stefek, M.; Gure, A.O.; Banerjee, S. Opposing roles of the aldo-keto reductases AKR1B1 and AKR1B10 in colorectal cancer. Cell Oncol. 2017, 40, 563–578. [CrossRef] 34. Fang, C.Y.; Lin, Y.H.; Chen, C.L. Overexpression of AKR1B10 predicts tumor recurrence and short survival in oral squamous cell carcinoma patients. J. Oral Pathol. Med. 2019, 48, 712–719. [CrossRef] 35. Ahmed, S.M.U.; Jiang, Z.N.; Zheng, Z.H.; Li, Y.; Wang, X.J.; Tang, X. AKR1B10 expression predicts response of gastric cancer to neoadjuvant chemotherapy. Oncol. Lett. 2019, 17, 773–780. [CrossRef] 36. DiStefano, J.K.; Davis, B. Diagnostic and Prognostic Potential of AKR1B10 in Human Hepatocellular Carcinoma. Cancers 2019, 11, 486. [CrossRef] 37. Saraswat, M.; Mrudula, T.; Kumar, P.U.; Suneetha, A.; Rao Rao, T.S.; Srinivasulu, M.; Reddy, B. Overexpression of aldose reductase in human cancer tissues. Med. Sci. Monit. 2006, 12, CR525–CR529. [PubMed] 38. Liu, J.; Zhong, X.; Li, J.; Liu, B.; Guo, S.; Chen, J.; Tan, Q.; Wang, Q.; Ma, W.; Wu, Z.; et al. Screening and identification of lung cancer metastasis-related genes by suppression subtractive hybridization. Thorac. Cancer 2012, 3, 207–216. [CrossRef][PubMed] 39. Torres-Mena, J.E.; Salazar-Villegas, K.N.; Sánchez-Rodríguez, R.; López-Gabiño, B.; Del Pozo-Yauner, L.; Arellanes-Robledo, J.; Villa-Treviño, S.; Gutiérrez-Nava, M.A.; Pérez-Carreón, J.I. Aldo-Keto Reductases as Early Biomarkers of Hepatocellular Carcinoma: A Comparison Between Animal Models and Human HCC. Dig. Dis. Sci. 2018, 63, 934–944. [CrossRef][PubMed] 40. Demirkol Canlı, S.; Seza, E.G.; Sheraj, I.; Gömçeli, I.; Turhan, N.; Carberry, S.; Prehn, J.H.M.; Güre, A.O.; Banerjee, S. Evaluation of an aldo-keto reductase gene signature with prognostic significance in colon cancer via activation of epithelial to mesenchymal transition and the p70S6K pathway. Carcinogenesis 2020, 41, 1219–1228. [CrossRef][PubMed] 41. Hung, J.J.; Yeh, Y.C.; Hsu, W.H. Prognostic significance of AKR1B10 in patients with resected lung adenocarcinoma. Thorac. Cancer 2018, 9, 1492–1499. [CrossRef][PubMed] 42. Ko, H.H.; Cheng, S.L.; Lee, J.J.; Chen, H.M.; Kuo, M.Y.; Cheng, S.J. Expression of AKR1B10 as an independent marker for poor prognosis in human oral squamous cell carcinoma. Head Neck 2017, 39, 1327–1332. [CrossRef][PubMed] 43. Yao, Y.; Wang, X.; Zhou, D.; Li, H.; Qian, H.; Zhang, J.; Jiang, L.; Wang, B.; Lin, Q.; Zhu, X. Loss of AKR1B10 promotes colorectal cancer cells proliferation and migration via regulating FGF1-dependent pathway. Aging 2020, 12, 13059–13075. [CrossRef] 44. Yoshitake, H.; Takahashi, M.; Ishikawa, H.; Nojima, M.; Iwanari, H.; Watanabe, A.; Aburatani, H.; Yoshida, K.; Ishi, K.; Takamori, K.; et al. Aldo-keto reductase family 1, member B10 in uterine carcinomas: A potential risk factor of recurrence after surgical therapy in cervical cancer. Int. J. Gynecol. Cancer 2007, 17, 1300–1306. [CrossRef][PubMed] 45. Singh, M.; Kapoor, A.; Bhatnagar, A. Oxidative and reductive metabolism of lipid-peroxidation derived carbonyls. Chem. Biol. Interact. 2015, 234, 261–273. [CrossRef][PubMed] 46. Jung, K.A.; Choi, B.H.; Nam, C.W.; Song, M.; Kim, S.T.; Lee, J.Y.; Kwak, M.K. Identification of aldo-keto reductases as NRF2-target marker genes in human cells. Toxicol. Lett. 2013, 218, 39–49. [CrossRef][PubMed] 47. Penning, T.M. Aldo-Keto Reductase Regulation by the Nrf2 System: Implications for Stress Response, Chemotherapy Drug Resistance, and Carcinogenesis. Chem. Res. Toxicol. 2017, 30, 162–176. [CrossRef][PubMed] 48. Asangani, I.; Blair, I.A.; Van Duyne, G.; Hilser, V.J.; Moiseenkova-Bell, V.; Plymate, S.; Sprenger, C.; Wand, A.J.; Penning, T.M. Using Biochemistry & Biophysics to Extinguish Androgen Receptor Signaling in Prostate Cancer. J. Biol. Chem. 2021, 100240. [CrossRef] 49. Shen, Y.; Zhong, L.; Johnson, S.; Cao, D. Human aldo-keto reductases 1B1 and 1B10: A comparative study on their enzyme activity toward electrophilic carbonyl compounds. Chem. Biol. Interact. 2011, 191, 192–198. [CrossRef]