p53 and p21
Most of PCV (58 patients, 81%) had no overexpression of p53, and high expression of p21 (60 patients, 83%) (Table 3). Low or undetectable expression of p21 was found in 6 of the 14 PCV patients with overexpression of p53, and of these 14, 12 were short-term survivors, 11 had early-stage disease (7 patients had stage I and 4 had stage II), and 9 had a tumor size of 4 cm or larger. Patients with overexpression of p53 were older at diagnosis than those with no overexpression (75 vs 67 years). Overexpression of p53 was more common in tumors located in the lower third or in the entire vagina than tumors located in the upper third, however, there was not a significant association (P = 0.066). Similarly, there was no association between the expression level of p53 and age at diagnosis, tumor size, or FIGO tumor stage (Table 4). Overexpression of p53 was significantly associated with short-term survival in the univariate analysis (P = 0.049, Table 3), but not in the multivariate analysis adjusted for age at diagnosis and tumor size (P = 0.058). Expression levels of p21 were not associated with survival.
Cyclin A and Ki67
High expressions of cyclin A and Ki67 were detected in most of PCV patients (71% and 83%, respectively). There was a significant association between high expression of cyclin A and poor differentiation (P = 0.007). The expression of Ki67 was significantly higher in tumors located in the lower third or in the entire vagina than tumors located only in the upper third (P = 0.044) (Table 4). There was no association between age at diagnosis, tumor size, FIGO tumor stage or survival, and the expression levels of these 2 proliferation markers.
E-Cadherin and Laminin-5γ2 Chain
Loss or underexpression of E-cadherin was found in 94% (68/72) of PCV patients. All patients with distant or inguinal node metastasis had loss of expression. Immunopositivity for the polyclonal antibody against the laminin-5γ2 chain was detected in all patients. Tumors 4 cm or larger had positive 3+ γ2 chain expression (P = 0.059). Tumors located in the lower third, or in the entire vagina had significantly lower expression of E-cadherin than tumors only located in the upper third (P = 0.038) (Table 4). No correlation was found between grade, stage, or survival and the immunoreactivity of E-cadherin and the laminin-5γ2 chain (Tables 3 and 4).
Multivariate Statistical Analysis
In the univariate analysis, there was a statistically significant association between overexpression of p53 and short-term survival, but not in the multivariate analysis adjusted for age at diagnosis and tumor size (P = 0.058). No association with survival was detected for the other molecular markers. The only factors that could independently explain the variation in survival were age at diagnosis and tumor size.
The prognostic value of biological markers in PCV has not been clearly evaluated. In this study, which to our knowledge is the largest on biological markers in PCV to date, we investigated the degree of genomic instability and proteins involved in tumor suppression, proliferation, invasion, and metastasis.
Chromosomal rearrangements and distinct DNA aneuploidy can be observed early in the carcinogenic process, making it possible to distinguish between different cellular alterations (nonspecific, premalignant, and malignant).33 In endometrial, breast, prostate, and colon cancer, DNA aneuploidy is primarily observed in clinically aggressive tumors and has been confirmed as a prognostic marker.10,26,33,34 In contrast, most of PCVs have been found to have an aneuploid distribution of DNA, with no difference seen by degree of invasiveness.12,30 In a previous study by our group, PCV was characterized by high genomic instability, with chromosomal changes bearing resemblance to those found in advanced cervical carcinoma.30 In the present study, all PCVs showed aneuploid distribution of DNA, no association between the clinicopathological variables considered, and the degree of DNA aneuploidy was observed. In other squamous cell carcinomas of the lower genital tract (eg, vulvar and cervical carcinomas), there are conflicting views concerning the value of DNA ploidy determination in clinical assessment.35 However, in a recent study on cervical carcinoma, DNA ploidy was determined to be an independent prognostic factor, and lymph node metastasis did not affect survival if the tumor was DNA diploid.36
p53 and p21
The expression of p53 in PCV has been investigated in only a few studies, and overexpression was found in 12% to 50% of PCV.11,13,30,37 In the study by Habermann et al,30 low expression of p53 was detected in 14 of 16 PCV together with high expression of p21 (WAF-1). Skomedal et al13 investigated 46 PCV and found that 50% had a TP53 gene mutation and/or protein accumulation, with a 70% correlation between mutation and immunohistochemical data. However, no significant association between TP53 alteration and survival has so far been observed in PCV.11,13 In the present study, in agreement with previous reports, we found that most of PCV patients had no overexpression of p53 (81%). However, we noticed that overexpression of p53 was significantly associated with short-term survival.
Furthermore, we found that most of PCV patients had high expression of p21, the major transcriptionally regulated downstream protein of p53. This finding in combination with no overexpression of p53 indicates that the activation of p21 must occur by p53-independent pathways. High p21 expression can cause antiapoptotic and mitogenic activity that may contribute to the pathogenesis of cancer.38 For example, in cervical carcinoma, p21 expression correlates with advanced stage.39 Low or undetectable levels of p21 were found in 6 of the 14 PCV patients with overexpression of p53, which might be indicative of mutational inactivation of the tumor suppressor gene TP53 in these specific patients. Cervical carcinomas have a similarly low expression of p53, and mutations in the TP53 gene are rarely detected.40,41 The function of the p53 protein is likely abolished because of viral inactivation by HPV 16 and 18 (oncoproteins E6 and E7), which are known to be the principal causative agents for cervical carcinoma.42 In vulvar cancers, the reported HPV prevalence is lower, at approximately 40%, whereas p53 overexpression is higher and detected in approximately 61% of the tumors.3,11 In PCV, HPV prevalence has been reported to be between 50% and 75%,1,3,11 whereas p53 overexpression was detected in 19%.11 The aforementioned results indicate that oncogenic HPV types and p53 gene mutations have independent carcinogenic roles in PCV.
The HPV prevalence rate has been found to be higher in tumors located in the upper sites of the genital tract, whereas p53 overexpression is more dominant in tumors occupying the lower sites. PCV has been suggested to have transitional characteristics between cervical and vulvar carcinoma, in HPV infection and p53 expression level.11 This might explain our finding of an association between p53 overexpression and tumor location in the lower part of the vagina. Moreover, this might confirm the hypothesis of different mechanisms of carcinogenesis depending on tumor location.
Cyclin A and Ki67
The proliferative activity of PCV has been studied in 2 reports.11,30 In accordance with our study, a high expression of cyclin A and Ki67 was found in most of PCV, but no association with survival was observed.
On the contrary, the expression of cyclin A and Ki67 have repeatedly been proven to be of prognostic value for survival and recurrence in several other tumors (eg, breast, colon, rectal, and anal cancer).17,21,43,44 However, in cervical and vulvar carcinoma, there are conflicting data on the prognostic value of these proliferation markers. In cervical adenocarcinoma, high levels of Ki67 staining was associated with more advanced stage at diagnosis and low grade,45 whereas no prognostic value of Ki67 was demonstrated in a study on squamous cell carcinoma of the cervix.46 However, cyclin A was found to be a poor prognostic indicator for squamous cell carcinoma of the cervix.19 In vulvar carcinoma, Ki67, but not cyclin A, has been found to be of prognostic value.47,48
E-cadherin and Laminin-5γ2 Chain
The expression level of laminin-5γ2 chain did not correlate with survival in this study; however, increased expression has been associated with more aggressive tumor behavior in other studies.26,28 Most of PCV showed loss of E-cadherin expression, suggesting that it is an aggressive malignancy with poor prognosis in accordance with other studies.31,49,50 In the present study, we also observed that reduced E-cadherin expression was related to tumor location and extension in the vagina.
Epithelial-mesenchymal transition with loss of cell adhesion is seen as a key event in the cascade leading to tumor invasion and metastasis.23 The suppression of E-cadherin is central in epithelial-mesenchymal transition and is regulated by various intracellular signaling pathways activated not only by multiple extracellular signals, for example, growth factors, wnt and estrogens,23,51 but also by hypoxia and inflammation. Furthermore, it has been shown that inhibition of p53 down-regulates E-cadherin, which induces invasion of serous borderline ovarian tumor cells.52 Our finding of no overexpression of p53 in most of PCVs might likewise indicate involvement of this pathway in the down-regulation of E-cadherin.
There is also growing evidence of a strong connection between estrogen signaling and E-cadherin expression.53,54 In elderly patients with vulvar carcinoma, decreased expression of estrogen receptors (ERα was lost and ERβ was decreased) was associated with significantly reduced levels of E-cadherin compared to normal vulvar epithelium, suggesting the ER signaling pathway plays an important role in vulvar carcinogenesis.54 Given the pronounced down-regulation of E-cadherin in almost all PCV cases in the present study, disturbances in the signaling pathways involved in the control of the E-cadherin-mediated cell adhesion described previously is likely also important in vaginal tumorigenesis.
Moreover, our results indicate that the biological behavior of vaginal tumors may differ depending on their location in the vagina. Tumors in the lower third and in the entire vagina seemed to be more aggressive in a higher proliferative activity, p53 overexpression, and more pronounced deregulation of E-cadherin, compared to tumors confined to the upper third of the vagina. Indeed, tumors in the upper third of the vagina are more often seen in patients who have been treated earlier for cervical dysplasia, or who have undergone hysterectomy, and these patients also have better prognosis.4 According to earlier studies, as mentioned previously, there are probably different HPV prevalence rates in the upper and lower genital tract.11 The various expressions of the biological markers according to tumor location discovered in this study might thus be due to different etiologies. As with vulvar carcinoma, there are probably 2 types of vaginal carcinoma, one that is HPV-related like cervical carcinoma, located in the upper vagina, and another that is not HPV-related, more often located in the lower part of the vagina.
As PCV preferentially occurs in older postclimacteric women, the etiology might also be associated with hormonal factors and the estrogen signaling pathway, as has been suggested for vulvar carcinoma. Estrogens have been found to be a cofactor for HPV-induced cervical and vaginal carcinogenesis; however, ERα has been shown to be down-regulated in both HPV-positive and HPV-negative cervical carcinoma.55 Although estrogen signaling may be of importance for vaginal carcinogenesis, the tumor development itself is probably complex, and like in other malignancies, involves multiple extracellular and intracellular pathways that may be initiated not only by viruses or estrogens, but also by other factors like gene mutations and chronic inflammation due to trauma. An earlier proteomic study on PCV has shown alterations and interactions of several proteins from different pathways, such as proteins involved in inactivation of p53 (deadbox, the ubiquitin-proteasome pathway, and 14-3-3 prot), activation of the RAS pathway (erbB3 binding protein), and proteins implicated in apoptosis (Dj-1, Gelsolin, and hnRNPH1).56
In summary, the results from the present study revealed that PCV is characterized by high genomic instability, highly aneuploid distribution of the nuclear DNA content, high proliferative activity, and pronounced down-regulation of E-cadherin, indicating either that it is an aggressive type of tumor, or that PCV is discovered in the late stages of disease progression. However, none of the biological markers investigated could explain the variation in survival. Interestingly, we found different expression of some molecular markers depending on location and extension of the vaginal tumor, indicating that tumorigenesis and tumor etiology might differ between tumors located in the upper and lower part of the vagina. However, further studies are needed to substantiate this hypothesis and to elucidate the molecular biology of this disease to find markers that might predict tumor aggressiveness and prognosis.
The authors thank Associate Professor Claes Silfversward (CS), Department of Pathology at Karolinska University Hospital for valuable help with re-evaluating the histopathological diagnosis.
1. Fuste V, del Pino M, Perez A, et al.. Primary squamous cell carcinoma of the vagina: human papillomavirus detection, p16(INK4A) overexpression and clinicopathological correlations. Histopathology
. 2010; 57: 907–916.
2. Insinga RP, Liaw KL, Johnson LG, et al.. A systematic review of the prevalence and attribution of human papillomavirus types among cervical, vaginal, and vulvar precancers and cancers in the United States. Cancer Epidemiol Biomarkers Prev
. 2008; 17: 1611–1622.
3. Smith JS, Backes DM, Hoots BE, et al.. Human papillomavirus type-distribution in vulvar and vaginal cancers and their associated precursors. Obstet Gynecol
. 2009; 113: 917–924.
4. Hellman K, Lundell M, Silfversward C, et al.. Clinical and histopathologic factors related to prognosis
in primary squamous cell carcinoma of the vagina. Int J Gynecol Cancer
. 2006; 16: 1201–1211.
5. Lian J, Dundas G, Carlone M, et al.. Twenty-year review of radiotherapy for vaginal cancer
: an institutional experience. Gynecol Oncol
. 2008; 111: 298–306.
6. Creasman WT, Phillips JL, Menck HR. The national cancer data base report on cancer of the vagina. Cancer
. 1998; 83: 1033–1040.
7. Nashiro T, Yagi C, Hirakawa M, et al.. Concurrent chemoradiation for locally advanced squamous cell carcinoma of the vagina: case series and literature review. Int J Clin Oncol
. 2008; 13: 335–339.
8. Dalrymple JL, Russell AH, Lee SW, et al.. Chemoradiation for primary invasive squamous carcinoma of the vagina. Int J Gynecol Cancer
. 2004; 14: 110–117.
9. Beller U, Benedet JL, Creasman WT, et al.. Carcinoma of the vagina. FIGO 26th Annual Report on the Results of Treatment in Gynecological Cancer. Int J Gynaecol Obstet
. 2006; 95: (suppl 1): S29–S42.
10. Habermann J, Lenander C, Roblick UJ, et al.. Ulcerative colitis and colorectal carcinoma: DNA-profile, laminin-5 gamma2 chain and cyclin A expression as early markers for risk assessment. Scand J Gastroenterol
. 2001; 36: 751–758.
11. Koyamatsu Y, Yokoyama M, Nakao Y, et al.. A comparative analysis of human papillomavirus types 16 and 18 and expression of p53 gene and Ki-67 in cervical, vaginal, and vulvar carcinomas. Gynecol Oncol
. 2003; 90: 547–551.
12. Punnonen R, Kallioniemi OP, Mattila J, et al.. Primary invasive and in situ vaginal carcinoma. Flow cytometric analysis of DNA aneuploidy and cell proliferation from archival paraffin-embedded tissue. Eur J Obstet Gynecol Reprod Biol
. 1989; 32: 247–251.
13. Skomedal H, Kristensen G, Helland A, et al.. TP53 gene mutations and protein accumulation in primary vaginal carcinomas. Br J Cancer
. 1995; 72: 129–133.
14. Cadwell C, Zambetti GP. The effects of wild-type p53 tumor suppressor activity and mutant p53 gain-of-function on cell growth. Gene
. 2001; 277: 15–30.
15. Hahn WC, Weinberg RA. Rules for making human tumor cells. N Engl J Med
. 2002; 347: 1593–1603.
16. Yam CH, Fung TK, Poon RY. Cyclin A in cell cycle control and cancer. Cell Mol Life Sci
. 2002; 59: 1317–1326.
17. Li L, Mu K, Zhou G, et al.. Genomic instability and proliferative activity as risk factors for distant metastases in breast cancer. Br J Cancer
. 2008; 99: 513–519.
18. Kim YT, Zhao M. Aberrant cell cycle regulation in cervical carcinoma. Yonsei Med J
. 2005; 46: 597–613.
19. Shiohara S, Shiozawa T, Miyamoto T, et al.. Expression of cyclins, p53, and Ki-67 in cervical squamous cell carcinomas: overexpression of cyclin A is a poor prognostic factor in stage Ib and II disease. Virchows Arch
. 2005; 446: 626–633.
20. Youssef RF, Shariat SF, Kapur P, et al.. Expression of cell cycle-related molecular markers
in patients treated with radical cystectomy for squamous cell carcinoma of the bladder. Hum Pathol
. 2011; 42: 347–355.
21. Scholzen T, Gerdes J. The Ki-67 protein: from the known and the unknown. J Cell Physiol
. 2000; 182: 311–322.
22. Perl AK, Wilgenbus P, Dahl U, et al.. A causal role for E-cadherin in the transition from adenoma to carcinoma. Nature
. 1998; 392: 190–193.
23. Beavon IR. The E-cadherin-catenin complex in tumour metastasis: structure, function and regulation. Eur J Cancer
. 2000; 36: 1607–1620.
24. Salo S, Haakana H, Kontusaari S, et al.. Laminin-5 promotes adhesion and migration of epithelial cells: identification of a migration-related element in the gamma2 chain gene (lamc2) with activity in transgenic mice in process citation. Matrix Biol
. 1999; 18: 197–210.
25. Tryggvason K. The laminin family. Curr Opin Cell Biol
. 1993; 5: 877–882.
26. Lundgren C, Frankendal B, Silfversward C, et al.. Laminin-5 gamma2-chain expression and DNA ploidy
as predictors of prognosis
in endometrial carcinoma. Med Oncol
. 2003; 20: 147–156.
27. Hellman K, Hellstrom AC, Silfversward C, et al.. Cancer of the vagina: laminin-5gamma2 chain expression and prognosis
. Int J Gynecol Cancer
. 2000; 10: 391–396.
28. Lenander C, Habermann JK, Ost A, et al.. Laminin-5 gamma 2 chain expression correlates with unfavorable prognosis
in colon carcinomas. Anal Cell Pathol
. 2001; 22: 201–209.
29. Auer G, Steinbeck R, Zetterberg A. Molecular markers
in diagnostic pathology. The Computerized Cytology and Histology Laboratory. 1994;129–142.
30. Habermann JK, Hellman K, Freitag S, et al.. A recurrent gain of chromosome arm 3q in primary squamous carcinoma of the vagina. Cancer Genet Cytogenet
. 2004; 148: 7–13.
31. Li C, Berx G, Larsson C, et al.. Distinct deleted regions on chromosome segment 16q23-24 associated with metastases in prostate cancer. Genes Chromosomes Cancer
. 1999; 24: 175–182.
32. Umbas R, Schalken JA, Aalders TW, et al.. Expression of the cellular adhesion molecule E-cadherin is reduced or absent in high-grade prostate cancer. Cancer Res
. 1992; 52: 5104–5109.
33. Auer GU, Heselmeyer KM, Steinbeck RG, et al.. The relationship between aneuploidy and p53 overexpression during genesis of colorectal adenocarcinoma. Virchows Arch
. 1994; 424: 343–347.
34. Forsslund G, Esposti PL, Nilsson B, et al.. The prognostic significance of nuclear DNA content in prostatic carcinoma. Cancer
. 1992; 69: 1432–1439.
35. Fox H. Ploidy in gynaecological cancers. Histopathology
. 2005; 46: 121–129.
36. Susini T, Olivieri S, Molino C, et al.. DNA ploidy
is stronger than lymph node metastasis as prognostic factor in cervical carcinoma: 10-year results of a prospective study. Int J Gynecol Cancer
. 2011; 21: 678–684.
37. Berchuck A, Kohler MF, Marks JR, et al.. The p53 tumor suppressor gene frequently is altered in gynecologic cancers. Am J Obstet Gynecol
. 1994; 170: 246–252.
38. Chang BD, Watanabe K, Broude EV, et al.. Effects of p21Waf1/Cip1/Sdi1 on cellular gene expression: implications for carcinogenesis, senescence, and age-related diseases. Proc Natl Acad Sci U S A
. 2000; 97: 4291–4296.
39. Abbas T, Dutta A. p21 in cancer: intricate networks and multiple activities. Nat Rev Cancer
. 2009; 9: 400–414.
40. Choo KB, Chong KY. Absence of mutation in the p53 and the retinoblastoma susceptibility genes in primary cervical carcinomas. Virology
. 1993; 193: 1042–1046.
41. Tommasino M, Accardi R, Caldeira S, et al.. The role of TP53 in cervical carcinogenesis. Hum Mutat
. 2003; 21: 307–312.
42. Zur Hausen H. Papillomavirus causing cancer: evasion from host-cell control in early events in carcinogenesis. J Natl Cancer Inst
. 2000; 92: 690–698.
43. Aamodt R, Jonsdottir K, Andersen SN, et al.. Differences in protein expression and gene amplification of cyclins between colon and rectal adenocarcinomas. Gastroenterol Res Pract
. 2009; 2009: 285830.
44. Nilsson PJ, Lenander C, Rubio C, et al.. Prognostic significance of Cyclin A in epidermoid anal cancer. Oncol Rep
. 2006; 16: 443–449.
45. Muller S, Flores-Staino C, Skyldberg B, et al.. Expression of p16INK4a and MIB-1 in relation to histopathology and HPV types in cervical adenocarcinoma. Int J Oncol
. 2008; 32: 333–340.
46. Oka K, Arai T. MIB1 growth fraction is not related to prognosis
in cervical squamous cell carcinoma treated with radiotherapy. Int J Gynecol Pathol
. 1996; 15: 23–27.
47. Hantschmann P, Lampe B, Beysiegel S, et al.. Tumor proliferation in squamous cell carcinoma of the vulva. Int J Gynecol Pathol
. 2000; 19: 361–368.
48. Knopp S, Bjorge T, Nesland JM, et al.. Cyclins D1, D3, E, and A in vulvar carcinoma patients. Gynecol Oncol
. 2005; 97: 733–739.
49. Delektorskaya VV, Perevoshchikov AG, Golovkov DA, et al.. Expression of E-cadherin, beta-catenin, and CD-44v6 cell adhesion molecules in primary tumors and metastases of colorectal adenocarcinoma. Bull Exp Biol Med
. 2005; 139: 706–710.
50. Galera-Ruiz H, Rios MJ, Gonzalez-Campora R, et al.. The cadherin-catenin complex in nasopharyngeal carcinoma. Eur Arch Otorhinolaryngol
. 2011; 268: 1335–1341.
51. Shin SY, Rath O, Zebisch A, et al.. Functional roles of multiple feedback loops in extracellular signal-regulated kinase and Wnt signaling pathways that regulate epithelial-mesenchymal transition. Cancer Res
. 2010; 70: 6715–6724.
52. Cheng JC, Auersperg N, Leung PC. Inhibition of p53 represses E-cadherin expression by increasing DNA methyltransferase-1 and promoter methylation in serous borderline ovarian tumor cells. Oncogene
. 2011; 30: 3930–3942.
53. Cardamone MD, Bardella C, Gutierrez A, et al.. ERalpha as ligand-independent activator of CDH-1 regulates determination and maintenance of epithelial morphology in breast cancer cells. Proc Natl Acad Sci U S A
. 2009; 106: 7420–7425.
54. Zannoni GF, Prisco MG, Vellone VG, et al.. Changes in the expression of oestrogen receptors and E-cadherin as molecular markers
of progression from normal epithelium to invasive cancer in elderly patients with vulvar squamous cell carcinoma. Histopathology
. 2011; 58: 265–275.
55. Kwasniewska A, Postawski K, Gozdzicka-Jozefiak A, et al.. Estrogen and progesterone receptor expression in HPV-positive and HPV-negative cervical carcinomas. Oncol Rep
. 2011; 26: 153–160.
56. Hellman K, Alaiya A, Becker S, et al.. Differential tissue-specific protein markers of vaginal carcinoma. Br J Cancer
. 2009; 100: 1303–1314.
Keywords:Copyright © 2013 by IGCS and ESGO
Vaginal cancer; Prognosis; Molecular markers; DNA ploidy