It is widely accepted that pelvic high-grade serous carcinoma (HGSC) develops through multistep (epi)genetic mutations from precursor lesions in the distal portion of the fallopian tube 1–8. Putative precursor lesions of HGSC include secretory cell outgrowth (SCOUT), p53 signatures, and serous tubal intraepithelial carcinoma (STIC), which progress to invasive HGSC 1,9. The SCOUT is defined as a discrete expansion of at least 30 secretory cells without p53 nuclear staining 10. p53 signature is defined as benign-appearing tubal epithelial cells with p53 protein accumulation and proposed as a tubal precursor for STIC 11. STIC is an intraepithelial noninvasive lesion showing significant atypia that is formed in the distal fallopian tube epithelium (FTE) and plays a key role in the early stages of HGSC carcinogenesis 12. On the basis of the proliferative activity, STIC lesions are divided into 2 types: dormant STIC (or serous tubal intraepithelial lesion) and active STIC (or proliferative STIC) that can be characterized by differential staining of Ki-67 antigen 13. Although recent genomic and molecular data are transforming our understanding of the steps of HGSC tumorigenesis and their relationship to one another 2–7,13, the detailed sequence of (epi)genetic alterations in the carcinogenesis process remains unknown.
Epidemiological studies showed that incessant ovulation and repeated hemorrhage have been linked to the development of HGSC 9,14. Ovulation fluid and hemorrhage contain prostaglandins, hemoglobin, heme, and iron, which induce reactive oxygen species (ROS). Excessive oxidative stress can cause increased DNA damage and impaired DNA repair 9. To establish a redox balance, cells exert antioxidant defense systems that detoxify elevated ROS levels 15. Dysregulation of ROS due to reduced antioxidant defense systems may be an important etiological factor for the accumulation of (epi)genetic alterations in precancerous lesions, leading to HGSC tumorigenesis 9. Among a variety of proteins involved in the redox regulation, we focused on the expression of CD44v9, a splice variant isoform of CD44. CD44v9 stabilizes the glutamate-cystine transporter xCT [also known as SLC7A11 (solute carrier family 7 member 11)], thereby inducing de novo synthesis of glutathione antioxidant peptide inside cells and decreasing intracellular levels of ROS 16. Downregulation of CD44v9 protein may be associated with malignant transformation of endometriosis, including ovarian clear cell carcinoma and endometrioid carcinoma 17. However, there have been no reports on the spatial and temporal expression of CD44v9 during the multistage carcinogenesis or stepwise progression from precursor lesions to HGSC. The aim of this study was to evaluate the morphologic features and immunohistochemical (IHC) expression status of p53, CD44v9, and Ki-67 in HGSC tumorigenesis. We found for the first time that CD44v9 loss may be an early event during the stepwise progression from p53 signature to dormant STIC.
MATERIALS AND METHODS
Patient and Tissue Samples
Approval was obtained from the Nara Medical University Institutional Ethics Review Board (no. 2575). The study complied with the Declaration of Helsinki. Written informed consent was obtained from all patients.
Following institutional review board approval, patients with primary pelvic HGSC who attended Nara Medical University Hospital, Kashihara, Japan from January 2015 to December 2017 were surveyed. Exclusion criteria were patients who received preoperative chemotherapy or immunotherapy. A total of 45 formalin-fixed paraffin-embedded specimens from 16 patients with HGSC (10 ovarian, 4 peritoneal, and 2 fallopian tube carcinoma) were obtained from the surgical pathology archives. Fallopian tube lesions were categorized as follows: morphologically normal FTE (n=6 samples), SCOUT (n=5), p53 signature (n=4), dormant STIC (n=8), active STIC (n=6), and HGSC (n=16). For example, if a lesion with “p53 signature” and a lesion with “active STIC” were present on one slide, we counted the number of lesions as 2. There were only 6 lesions in which morphologically normal FTE was observed in HGSC patients. Therefore, morphologically normal FTE from 5 patients who underwent total hysterectomy and salpingo-oophorectomy for benign diseases (uterine fibroids) were included in the analysis as a control. A total of 11 normal FTE specimens were examined. These patients were diagnosed with sporadic ovarian cancer because none of them met the criteria for hereditary breast and ovarian cancer. All precursor lesions adjacent to normal-appearing FTE were discontinuous with HGSC in any slide analyzed. The definitions of SCOUT, p53 signature, and STIC have been reported by Chen and colleagues 10,11,13,18. Dormant STIC is morphologically similar to serous tubal intraepithelial lesion, characterized by cytologic atypia and low proliferative activity (Ki-67 labeling index <10%) 13. Active STIC is a pathomorphologically (significant nuclear atypia and architectural alterations) and immunohistochemically (p53 protein accumulation and high Ki-67 labeling index ≥10%) detectable lesion, located in the fimbriated end of the fallopian tube in most patients 13.
The specimens were fixed in 10% buffered formalin and embedded in paraffin. The standard tissue paraffin block was sectioned at 4 μm for IHC. Before staining, sections were deparaffinized in xylene and rehydrated in ethanol. After pretreatment using a microwave with 10 mM citrate buffer, pH 6.0, at 95°C for 20 minutes, sections were incubated with 3% hydrogen peroxide for 5 min to quench endogenous peroxidase and then incubated with the primary antibody overnight at 4°C. The primary antibodies used included mouse monoclonal anti-p53 (dilution, 1:50; clone DO-7, NCL-p53-DO7, Leica Biosystems, Newcastle, UK), mouse monoclonal anti-human Ki-67 antigen (dilution, 1:2; clone MIB-1, IS626, Dako, CA) and mouse monoclonal anti-CD44v9 antibody (dilution, 1:400; clone CD44v9, Cat No. GTX34523, GeneTex, CA). As a control for nonspecific binding of the secondary detection system, a slide in which control mouse or rabbit immunoglobulin G (Dako) was applied instead of the primary antibody was included from each block. Specimens of squamous cell carcinoma of the uterine cervix served as an equivalent positive control. Positive and negative controls were routinely included. The Envision+ solution for mouse and rabbit (Dako) was then applied for 30 min at room temperature. The reaction products were observed using 3-3’-diaminobenizidine tetrahydrochloride (Sigma Chemical Company, St. Louis, MO) and H2O2. The sections were then lightly counterstained with hematoxylin.
Assessment of IHC Staining
Protein expression was independently evaluated by 2 pathologists (S.S. and T.U.) who were blinded to all of the clinicopathologic variables. p53 immunostaining shows at least 3 staining patterns (overexpression, complete absence, normal/wild-type). The results of p53 immunostaining were interpreted according to Kobel et al. 19. Overexpression, complete absence or normal/wild-type staining predicted gain-of-function, loss-of-function (stopgain, indel, splicing) or no detectable TP53 mutations, respectively 20. Cells showing nuclear staining of p53 were considered to be positive, whereas cells without nuclear staining or with cytoplasmic staining were considered to be negative. The percentage of cells with positive immunostaining was tabulated in each lesion. The p53 positivity was defined as a distinct nuclear immunoreaction in >75% of the cells of the sample. In 3 samples, p53 expression was completely absent, which was interpreted as abnormal/mutation-type and recorded as “positive” (Supplemental Table1, Supplemental Digital Content 1, http://links.lww.com/IJGP/A114). The CD44v9 positivity was evaluated as a semiquantitative IHC staining score as described previously 21. Immunostaining intensity was classified as 0 (absent), 1 (weak, positivity observed at 400x), 2 (intermediate, positivity observed at 100×), or 3 (strong, positivity observed at 40×). The percentage of cells (from 0 to 100) was multiplied by the corresponding intensity (from 0 to 3) to obtain IHC score in the range of 0 to 300 21. The Ki-67 positivity (Ki-67 labeling index) was defined as a distinct nuclear immunoreaction in ≥10% of the cells within each lesion 22.
Statistical analysis was conducted using SPSS software (version 22.0, SPSS Inc., Chicago, IL). Not all of IHC scores in groups, categorized fallopian tube lesions and HGSC, show normal distribution by using Shapiro-Wilk analysis. The significance of the difference of the IHC expression status among groups were analyzed by Kruskal-Wallis test. For significant Kruskal-Wallis tests, pairwise comparisons utilized the Steel-Dwass test. Further, Mann-Whitney U test was used comparing between 2 groups. The optimal cut-off value for CD44v9 IHC staining score were determined using the receiver operating characteristic (ROC) curve. For χ2 test, Pearson χ2 test or Fisher exact test were applied. P-value of <0.05 was considered statistically significant.
Baseline Characteristics of Study Population
Forty-five paraffin-embedded specimens were obtained from 16 patients with pelvic HGSC (10 ovarian, 4 peritoneal, and 2 fallopian tube carcinoma) who underwent surgery in the hospital. Furthermore, normal FTE tissue from 5 fibroid patients was included. Detailed information for all patients is provided in Supplemental Table1 (Supplemental Digital Content 1, http://links.lww.com/IJGP/A114). The age of the patients with fibroids ranged from 38 to 65 yr, with a median of 57 yr. The age of the HGSC patients ranged from 41 to 79 yr, with a mean of 60 and a median of 59 yr. At the time of diagnosis, 81.3% of patients had stage III or worse state of HGSC. Fallopian tube lesions analyzed were located closest to or within 10 mm from the fimbriated end of the fallopian tube. The demographic and clinicopathologic data of the study population are presented in Table 1.
TABLE 1 -
Baseline characteristics of study subjects
|Age at diagnosis, median (range) (yr)
|BMI, median (range), kg/m2
|FIGO stage, patients number (%)
BMI indicates body mass index; HGSC, high-grade serous carcinoma.
IHC Expression Status of p53, CD44v9 and Ki-67
We investigated the expression of the p53 protein, CD44v9 protein, and Ki-67 antigen in fallopian tube lesions and HGSC lesions using IHC and described the correlation of the expression of these markers. Representative images of IHC staining are shown in Figure 1, and the results are numerically shown in Table 2. We confirmed that there was no difference in the IHC expression status of p53, CD44v9, and Ki-67 between morphologically normal FTE (n=6) from patients with HGSC and FTE (n=5) from patients with benign diseases (data not shown).
TABLE 2 -
Immunohistochemical expression status in p53 protein, CD44v9 IHC score, and Ki-67 labeling index in fallopian tube lesions and HGSC
|Total samples, n
|p53 positivity, n (%), median (range)
||0 (0%), 20.0 (5–40)
||0 (0%), 20.0 (10–25)
||4 (100%), 92.5 (90–95)
||8 (100%), 95.0 (80–95)
||6 (100%), 95.0 (95–95)
||16 (100%), 95.0 (0–95)
|CD44v9, IHC score, median (range)
|Ki-67 positivity, n (%), median (range)
||0 (0%), 2.0 (1–5)
||0 (0%), 0.0 (0–2)
||0 (0%), 1.0 (0–5)
||0 (0%), 3.5 (0–5)
||6 (100%), 35.0 (30–50)
||16 (100%), 40.0 (20–70)
The significance of the difference of the immunohistochemical expression status obtained in paired samples was estimated according to the Kruskal-Wallis test followed by the Steel-Dwass test. CD44v9 IHC score: normal versus dormant STIC, P=0.022; normal versus active STIC, P=0.039; normal versus HGSC, P=0.010; SCOUT versus dormant STIC, P=0.034; SCOUT versus HGSC, P=0.011; and P53 signature versus HGSC, P=0.028. p53 protein: normal versus dormant STIC, P=0.018; normal versus active STIC, P=0.025; SCOUT versus dormant STIC, P=0.029; and SCOUT versus active STIC, P=0.032. Ki-67 labeling index: normal versus active STIC, P=0.040; normal versus HGSC, P=0.005; SCOUT versus HGSC, P=0.010; p53 signature versus HGSC, P=0.026; dormant STIC versus active STIC, P=0.019; and Dormant STIC versus HGSC, P=0.001.
FTE indicates fallopian tube epithelium; HGSC, high-grade serous carcinoma; IHC, immunohistochemical; SCOUT, secretory cell outgrowth; STIC, serous tubal intraepithelial carcinoma; STIL, serous tubal intraepithelial lesion.
First, 3 cases with p53 complete absence were included in 16 patients with HGSC (Supplemental Table1, Supplemental Digital Content 1, http://links.lww.com/IJGP/A114, case No. 7, 10, and 15). The p53 positivity was identified in HGSC and concomitant p53 signature, dormant STIC and active STIC, but only scattered positive cells were present in normal FTE and SCOUT (Fig. 1, upper panel, Table 2). p53 signature, dormant STIC, active STIC and HGSC had a significantly higher p53 protein accumulation than normal FTE and SCOUT (P<0.001). There was no significant difference in the p53 protein accumulation among p53 signature, dormant STIC, active STIC, and HGSC (P=0.342).
Second, there was a significant lower level of CD44v9 membrane staining and IHC score in dormant STIC, active STIC, and HGSC than in normal FTE, SCOUT, and p53 signature (P<0.001) (Fig. 1, middle panel, Table 2). Furthermore, the loss of CD44v9 staining coincided with the appearance of malignant features such as significant nuclear atypia and architectural alterations.
Third, both active STIC and HGSC had a significantly higher Ki-67 index than normal FTE, SCOUT, p53 signature, and dormant STIC (P<0.001) (Fig. 1, lower l, Table 2). The Ki-67 index in HGSC ranged from 20.0% to 70.0% (median was 40.0%) and the mean±SD was 43.4%±13.5%. Active STIC had a Ki-67 index ranging from 30.0% to 50.0%, with a median index of 35.0% and a mean±SD of 36.7%±8.17%. There was no significant difference in the Ki-67 labeling index between active STIC and HGSC (P=0.255). Dormant STIC had a low Ki-67 index, ranging from 0% to 5.0%, with a median index of 1.0% and a mean±SD of 1.75%±2.36%.
Finally, our data indicate that dual immunostaining for p53 and CD44v9 can reliably identify p53 signature in fallopian tube lesions. Positive p53/negative CD44v9 immunostaining is a highly specific marker for differentiating dormant STIC, active STIC, and HGSC from p53 signature. Twenty-nine (96.7%) of the 30 cases, including 14 STICs and 16 HGSCs, was negative for CD44v9 immunostaining (Supplemental Table1, Supplemental Digital Content 1, http://links.lww.com/IJGP/A114, only case 6 is positive for CD44v9). The combined effects of p53 mutations and CD44v9 loss contributes to the progression of p53 signature to dormant STIC. Elevated Ki-67 labeling index is involved in the stepwise evolution from dormant STIC to active STIC, in association with p53 mutations and CD44v9 loss.
The current study on HGSC carcinogenesis, based on analyses of precursor lesions, termed SCOUT, p53 signature, dormant STIC, and active STIC lesions, offers the following possibility: HGSC carcinogenesis is a progressive multistep process involving the sequential accumulation of genetic and epigenetic alterations; p53 genetic mutations, followed by CD44v9 loss, then genetic alterations associated with increased Ki-67 labeling index. That is, each alteration in the evolutionary trajectory of HGSC progression may be acquired independently and sequentially. The expression of p53 protein, CD44v9 protein, and Ki-67 antigen was determined by IHC analysis in morphologically normal FTE and precursor lesions of the fallopian tubes from 16 patients with HGSC. We show that the fallopian tube lesions in HGSC patients displayed distinctive immunophenotypes: (1) normal FTE and SCOUT demonstrated the p53−/CD44v9+/Ki-67− expression; (2) p53 signature showed the p53+/CD44v9+/Ki-67- expression; (3) dormant STIC was the p53+/CD44v9−/Ki-67− expression; and (4) active STIC and HGSC were the p53+/CD44v9−/Ki-67+ expression (Fig. 1). These immunostaining patterns can reliably identify malignant transformation during the sequential progression from normal FTE and SCOUT to p53 signature, dormant STIC, active STIC, and then HGSC, thereby narrowing the differential diagnosis and elucidating the mechanism of HGSC tumorigenesis. Figure 2 shows a schematic putative model of HGSC progression based primarily on the current IHC model. The new finding of this study is that CD44v9 loss may be involved in the progression from p53 signature to dormant STIC.
First, the molecular alterations accumulate in a stepwise manner along the malignant transformation process from normal FTE through precursor lesions to HGSC 1,23, but the initial molecular events that lead to transformation from p53 signature to dormant STIC and then active STIC remain poorly understood 13. The majority of HGSC exhibited high levels of p53 immunoreactivity, and harbored deleterious TP53 mutations 24. Nakamura et al. 25 reported that the SCOUT with TP53 mutations progresses to HGSC via the p53 signature and STIC, indicating that TP53 mutations are associated with early events in carcinogenesis of HGSC. Our study also confirmed that the p53 protein accumulation increased with the progression from SCOUT to p53 signature and then to STIC and invasive cancer.
Second, we explore factors involved in the progression from p53 signature to STIC, because alterations in the p53 protein accumulation or p53 mutations alone do not cause significant nuclear atypia and architectural abnormalities 13. Our group recently reported that CD44v9 loss may be involved in the progression of endometriosis to type 1 ovarian cancer 17. Malignant transformation of endometriosis occurs mainly through oxidative stress caused by hemorrhage. Similarly, since the fallopian tube epithelial cells are also exposed to the recurrent reflux of menstrual shedding or incessant ovulation, we speculated that redox imbalance is involved in HGSC carcinogenesis. Therefore, we focused on CD44v9 protein as a marker for a sequential progression from normal FTE to precursor lesions and then HGSC. The expression of CD44v9 was lost or significantly reduced in dormant STIC, active STIC and HGSC, suggesting that CD44v9 loss takes part in the progression of p53 signature to dormant STIC (Fig. 1). As a result of positive clonal selection, the p53+/CD44v9− phenotype may be advantageous for survival (Fig. 2).
We discuss why CD44v9 loss causes nuclear atypia and structural complexity. CD44v9 has been identified as one of the cancer stem cell markers and contributes to ROS defense through upregulation of the intracellular antioxidant 8,16,26. The activation of antioxidant capacities provides protection against oxidative stress and associated diseases, including cancer 27. Thus, CD44v9 loss may promote tumorigenesis by enhancing the ROS-induced DNA mutations, which may play an important role in the evolution of STIC 9,28. STIC is considered to be susceptible to impaired repair of DNA double-strand breaks, exhibits DNA replication stress and increases genomic instability 9,28. The accumulation of genomic instability, including chromosomal instability and copy-number alternations, causes large-scale genomic alterations, leading to HGSC tumorigenesis 29.
Why does CD44v9 loss occur in the early stages of STIC lesions? Many regions of the genome exhibit loss of heterozygosity in HGSC 30. Chromosome arms 11p13, 11p15.5, 11q24, 17p and 17q21 are considered to be regions of frequent loss of heterozygosity 30. Since the CD44 gene was mapped to chromosome arm 11p13, CD44v9 loss may be due likely to loss of heterozygosity at 11p13. Another possibility is that CD44v9 downregulation may be attributed to promoter hypermethylation 31,32. Verkaik et al. 33 reported that CD44 was hypermethylated and transcriptionally silenced in prostate cancer, suggesting an important role in tumor progression and metastasis. However, hypermethylation of CpG islands of CD44v9 has never been reported in HGSC.
To our knowledge, studies from several research groups have pointed out an opposite role of CD44v9. Overexpression of CD44v9 contributes to the aggressive nature of many cancers, including gastric 34,35, esophageal 36, cholangiocarcinoma 37, hepatocellular 8 and prostate 38 cancers, and significantly correlated with the upregulation of p53 and Ki-67 expression 34. The expression level of CD44v9 in tumor tissue is significantly associated with the proliferation, invasion, epithelial-mesenchymal transition, aggressiveness, and poor prognosis 8,34–38. In contrast, our study revealed CD44v9 loss in active STIC. This discrepancy may be due to differences in the function of CD44v9 in cancer and precancerous lesions. CD44v9 contributes to ROS defenses. In cancer cells, the upregulation of CD44v9 expression is associated with increased aggressive tumor phenotypes. On the other hand, in the early stage of malignant transformation such as active STIC, CD44v9 silencing results in tubal epithelial cell damage due to ovulation and hemorrhage-induced oxidative stress, causing the accumulation of additional genetic alterations, which may be associated with an increased risk of future developing HGSC. However, there are few reports on the importance of CD44v9 expression in precancerous lesions.
Finally, the Ki-67 antigen expression was involved in the stepwise fashion from dormant STIC to active STIC. This finding is consistent with the previous study 1. Ki-67 is one of the molecules critical for evaluating the proliferation status and regulating cell cycle progression. Our study identified two immunophenotypes of STIC: the p53+/CD44v9−/Ki-67− subtype (mostly dormant STIC) and the p53+/CD44v9−/Ki-67+ subtype (mostly active STIC). Active STIC had significantly higher Ki-67 indices than dormant STIC (P<0.01), reflecting the existence of tumor subclones with higher proliferative activity and neoplastic progression. Concurrent alterations of p53 mutations, CD44v9 loss and high mitotic index may be associated with a more proliferative immunophenotype in STIC. However, it is unknown at this time which genetic alterations cause the malignant behavior of the dormant STIC to the active STIC, but there are several possibilities. Few studies have analyzed the genetic and molecular alterations from dormant STIC to active STIC 13. Subsequent to CD44v9 loss, deletion of the tumor suppressor genes located within and in close proximity to known deleted area at chromosome arm 11p13 may cause malignant behavior of dormant STIC to active STIC. A potential candidate gene at this locus is Wilms’ tumor 1 (WT1), a proposed tumor suppressor gene, but this gene did not reveal any abnormalities in ovarian cancer 39. Allelic imbalance for other tumor suppressor genes involving chromosome 11p13 may be an early event in active STIC lesion formation. Alternatively, CD44v9 loss itself may cause an accumulation of mutations in other genes and subsequent carcinogenesis, which may influence preneoplastic changes in gene expression relevant to nuclear atypia, architectural alterations, and tumor development.
In conclusion, STIC lesions may progress through the acquisition of sequential changes in p53 mutations, CD44v9 loss, and elevated Ki-67 expression. We propose the following hypothesis for multistage carcinogenesis of HGSC: the progression from p53 signature to dormant STIC is caused by CD44v9 loss following p53 gene mutation. Subsequently, progression from dormant STIC to active STIC may require additional genetic mutations that can trigger Ki-67 overexpression. Future studies are needed to assess the underlying mechanism of CD44v9 loss and to identify the genetic mutations involved in the transition from dormant STIC to active STIC during the process of carcinogenesis.
The authors are grateful to Masako Nakata from Laboratory of Diagnostic Pathology at Nara Medical University for technical assistance with immunohistostaining.
1. Mittal N, Srinivasan R, Gupta N, et al. Secretory cell outgrowths, p53 signatures, and serous tubal intraepithelial carcinoma in the fallopian tubes of patients with sporadic pelvic serous carcinoma. Indian J Pathol Microbiol 2016;59:481–8.
2. Saini U, Suarez AA, Naidu S, et al. STAT3/PIAS3 levels serve as “Early Signature” genes in the development of high-grade serous carcinoma from the fallopian tube. Cancer Res 2018;78:1739–50.
3. Eckert MA, Pan S, Hernandez KM, et al. Genomics of ovarian cancer progression reveals diverse metastatic trajectories including intraepithelial metastasis to the fallopian tube. Cancer Discov 2016;6:1342–51.
4. McDaniel AS, Stall JN, Hovelson DH, et al. Next-generation sequencing of tubal intraepithelial carcinomas. JAMA Oncol 2015;1:1128–32.
5. Kuhn E, Kurman RJ, Vang R, et al. TP53 mutations in serous tubal intraepithelial carcinoma and concurrent pelvic high-grade serous carcinoma—evidence supporting the clonal relationship of the two lesions. J Pathol 2012;226:421–6.
6. Vang R, Visvanathan K, Gross A, et al. Validation of an algorithm for the diagnosis of serous tubal intraepithelial carcinoma. Int J Gynecol Pathol 2012;31:243–53.
7. Visvanathan K, Vang R, Shaw P, et al. Diagnosis of serous tubal intraepithelial carcinoma based on morphologic and immunohistochemical features: a reproducibility study. Am J Surg Pathol 2011;35:1766–75.
8. Wada F, Koga H, Akiba J, et al. High expression of CD44v9 and xCT in chemoresistant hepatocellular carcinoma: potential targets by sulfasalazine. Cancer Sci 2018;109:2801–10.
9. Kobayashi H, Ogawa K, Kawahara N, et al. Sequential molecular changes and dynamic oxidative stress in high-grade serous ovarian carcinogenesis. Free Radic Res 2017;51:755–64.
10. Chen EY, Mehra K, Mehrad M, et al. Secretory cell outgrowth, PAX2 and serous carcinogenesis in the fallopian tube. J Pathol 2010;222:110–6.
11. Kindelberger DW, Lee Y, Miron A, et al. Intraepithelial carcinoma of the fimbria and pelvic serous carcinoma: evidence for a causal relationship. Am J Surg Pathol 2007;31:161–9.
12. Crum CP. Intercepting pelvic cancer in the distal fallopian tube: theories and realities. Mol Oncol 2009;3:165–70.
13. Wu RC, Wang P, Lin SF, et al. Genomic landscape and evolutionary trajectories of ovarian cancer precursor lesions. J Pathol 2019;248:41–50.
14. Fathalla MF. Incessant ovulation and ovarian cancer—a hypothesis re-visited. Facts Views Vis Obgyn 2013;5:292–7.
15. Iwabuchi T, Yoshimoto C, Shigetomi H, et al. Oxidative stress and antioxidant defense in endometriosis and its malignant transformation. Oxid Med Cell Longev 2015;2015:848595.
16. Mvunta DH, Miyamoto T, Asaka R, et al. SIRT1 regulates the chemoresistance and invasiveness of ovarian carcinoma cells. Transl Oncol 2017;10:621–31.
17. Niiro E, Kawahara N, Yamada Y, et al. Immunohistochemical expression of CD44v9 and 8-OHdG in ovarian endometrioma and the benign endometriotic lesions adjacent to clear cell carcinoma. J Obstet Gynaecol Res 2019;45:2260–6.
18. Kurman RJ, Shih IeM. The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. Am J Surg Pathol 2010;34:433–43.
19. Kobel M, Reuss A, du Bois A, et al. The biological and clinical value of p53 expression in pelvic high-grade serous carcinomas. J Pathol 2010;222:191–8.
20. Kobel M, Piskorz AM, Lee S, et al. Optimized p53 immunohistochemistry is an accurate predictor of TP53 mutation in ovarian carcinoma. J Pathol Clin Res 2016;2:247–58.
21. Hsu CP, Lee LY, Hsu JT, et al. CD44 predicts early recurrence in pancreatic cancer patients undergoing radical surgery. In Vivo 2018;32:1533–40.
22. Kuhn E, Kurman RJ, Sehdev AS, et al. Ki-67 labeling index as an adjunct in the diagnosis of serous tubal intraepithelial carcinoma. Int J Gynecol Pathol 2012;31:416–22.
23. Nakamura K, Nakayama K, Ishikawa N, et al. Reconstitution of high-grade serous ovarian carcinoma from primary fallopian tube secretory epithelial cells. Oncotarget 2018;9:12609–19.
24. Chien J, Sicotte H, Fan JB, et al. TP53 mutations, tetraploidy and homologous recombination repair defects in early stage high-grade serous ovarian cancer. Nucleic Acids Res 2015;43:6945–58.
25. Nakamura M, Obata T, Daikoku T, et al. The association and significance of p53 in gynecologic cancers: the potential of targeted therapy. Int J Mol Sci 2019;20:5482–97.
26. Nagano O, Okazaki S, Saya H. Redox regulation in stem-like cancer cells by CD44 variant isoforms. Oncogene 2013;32:5191–8.
27. Yang Y, Yee D. IGF-I regulates redox status in breast cancer cells by activating the amino acid transport molecule xC. Cancer Res 2014;74:2295–305.
28. Liu Z, Liu J, Segura MF, et al. MiR-182 overexpression in tumourigenesis of high-grade serous ovarian carcinoma. J Pathol 2012;228:204–15.
29. Zhang H, Liu T, Zhang Z, et al. Integrated proteogenomic characterization of human high-grade serous ovarian cancer. Cell 2016;166:755–65.
30. Pedersen BS, Konstantinopoulos PA, Spillman MA, et al. Copy neutral loss of heterozygosity is more frequent in older ovarian cancer patients. Genes Chromosomes Cancer 2013;52:794–801.
31. Kito H, Suzuki H, Ichikawa T, et al. Hypermethylation of the CD44 gene is associated with progression and metastasis of human prostate cancer. Prostate 2001;49:110–5.
32. Sato S, Yokozaki H, Yasui W, et al. Silencing of the CD44 gene by CpG methylation in a human gastric carcinoma cell line. Jpn J Cancer Res 1999;90:485–9.
33. Verkaik NS, Trapman J, Romijn JC, et al. Down-regulation of CD44 expression in human prostatic carcinoma cell lines is correlated with DNA hypermethylation. Int J Cancer 1999;80:439–43.
34. Yasui W, Kudo Y, Naka K, et al. Expression of CD44 containing variant exon 9 (CD44v9) in gastric adenomas and adenocarcinomas: relation to the proliferation and progression. Int J Oncol 1998;12:1253–8.
35. Jang BI, Li Y, Graham DY, et al. The role of CD44 in the pathogenesis, diagnosis, and therapy of gastric cancer. Gut Liver 2011;5:397–405.
36. Taniguchi D, Saeki H, Nakashima Y, et al. CD44v9 is associated with epithelial-mesenchymal transition and poor outcomes in esophageal squamous cell carcinoma. Cancer Med 2018;7:6258–68.
37. Suwannakul N, Ma N, Thanan R, et al. Overexpression of CD44 variant 9: a novel cancer stem cell marker in human cholangiocarcinoma in relation to inflammation. Mediators Inflamm 2018;2018:4867234.
38. Ghatak S, Hascall VC, Markwald RR, et al. Stromal hyaluronan interaction with epithelial CD44 variants promotes prostate cancer invasiveness by augmenting expression and function of hepatocyte growth factor and androgen receptor. J Biol Chem 2010;285:19821–32.
39. Viel A, Giannini F, Capozzi E, et al. Molecular mechanisms possibly affecting WT1 function in human ovarian tumors. Int J Cancer 1994;57:515–21.