Independent risk factors for ovarian metastases in stage IA IIB cervical carcinoma

Transcription

1 Received: 10 January 2018 Revised: 8 August 2018 Accepted: 13 August 2018 DOI: /aogs ORIGINAL RESEARCH ARTICLE Independent risk factors for ovarian metastases in stage IA IIB cervical carcinoma Le Zhou Xinghui Liu 2 1,2* Chun Tang Sun 3* Lin Lin 2 Yao Xie 4 Yan Huang 1 Qiao Li 1 1 Health Management Department, West China Hospital of Sichuan University, Chengdu, China 2 Obstetrics and Gynecology Department, West China Second University Hospital, Sichuan University, Chengdu, China 3 Department of Radiation Oncology, Sichuan Cancer Hospital and Institute, Sichuan Cancer Center, Chengdu, China 4 Obstetrics and Gynecology Department, Sichuan People s Hospital, Chengdu, China Correspondence Xinghui Liu, Obstetrics and Gynecology Department, West China Second University Hospital, Sichuan University, Chengdu, China. xinghuiliu@163.com Funding information This study was funded by the Health Care Program in Sichuan Province (grant no ) and the Science Foundation of Chengdu Municipal Science and Technology Bure (grant no ofhm SF). Abstract Introduction: Cervical cancer is a common malignant tumor in women; most cervical cancer patients are premenopausal. Ovarian resection or preservation remains controversial. The purpose of this study was to discover the risk factors for ovarian metastasis in women with stage I II cervical cancer. Material and methods: A total of 3292 women with cervical carcinoma who had undergone radical hysterectomy, with pelvic lymphadenectomy and bilateral oophorectomy or wedge resection of ovaries, were included in this multicenter retrospective study. We analyzed patients demographics, International Federation of Obstetrics and Gynecology stage, and histopathologic records to determine clinicopathologic risk factors of ovarian metastasis. Results: Of the patients, 115 (3.49%) were confirmed to have ovarian metastasis. The ovarian metastasis rate was 2% (56/2794) for squamous cell carcinoma and 11.8% (59/498) for nonsquamous cell carcinoma. The risk factors independently associated with ovarian metastasis were histologic type (odds ratio [OR] 8.76, 95% CI ), lymph node metastasis (OR 2.57, 95% CI ), lymphovascular space invasion (OR 2.82, 95% CI ), and corpus invasion (OR 6.34, 95% CI ). Conclusions: The histologic type, lymph node metastasis, lymphovascular space invasion, and corpus invasion were independently associated with ovarian metastasis. Histologic type and corpus invasion were the most important risk factors. Therefore, we suggest that corpus invasion might be a strong contraindication for preservation of the ovaries. KEYWORDS cervical carcinoma, corpus invasion, ovarian metastasis, ovarian preservation, risk factor 1 INTRODUCTION Cervical cancer is the leading cancer of the female reproductive system in developing countries. 1 In recent years, both the overall survival Abbreviations: CI, confidence interval; FIGO, International Federation of Obstetrics and Gynecology; OM, ovarian metastasis; OR, odds ratio. *These authors contributed to the work equally. and tumor free survival rates have been significantly improved as a result of advances in the technologies of surgery, chemotherapy, and radiotherapy. 1,2 Currently, an increasing number of young people are suffering from cervical cancer. Approximately 40% of cervical cancers occur in women of childbearing age. 3 Ovarian function is important to these patients; unnecessary oophorectomy during a radical hysterectomy will seriously affect their long term quality of life and increase the risk of osteoporosis and cardiovascular disease. 4, Nordic Federation of Societies of wileyonlinelibrary.com/journal/aogs Acta Obstet Gynecol Scand. 2019;98: Obstetrics and Gynecology

2 19 In recent years, several studies have discussed the risk factors for ovarian metastasis (OM) and suggested the option of whether to preserve the ovaries. 6 Some researchers believe that preserving the ovaries in some patients can improve their quality of life after surgery. 7 However, there is little research in this area. Meanwhile, many of the studies are small in size and have not considered all risk factors, such as uterine corpus invasion for example. Thus, ovarian resection or preservation remains controversial. Our Key Message Histologic type and corpus invasion were the most important risk factors associated with ovarian metastasis. Corpus invasion might be a strong contraindication for preservation of the ovaries. TABLE 1 Univariate analysis of the risk factors for ovarian metastasis in cervical cancer n = 3292 n % Ovarian metastasis Positive (n = 115) Negative (n = 3177) P value Age >45 years years Stage IA IB IIA IIB Histology Squamous <0.001 Nonsquamous Tumor size 4 cm <4 cm Lymph node metastasis Involved Uninvolved Lymphovascular space invasion Involved Uninvolved Corpus invasion Involved <0.001 Uninvolved Depth of stromal invasion 1/ <1/ Parametrial invasion Involved Uninvolved Vaginal invasion Involved Uninvolved

3 20 TABLE 2 Multivariate analysis of the risk factors for ovarian metastasis in uterine cervical cancer Variable OR 95% CI P value Stage IA 1 IB IIA IIB Histology Squamous 1 Nonsquamous <0.001 Lymph node metastasis Involved Lymphovascular space invasion Involved Corpus invasion Involved <0.001 Parametrial invasion Involved study is a multicenter retrospective study of cervical cancer patients with stage IA IIB disease. We aim to identify the risk factors associated with OM so that these women may have the option to retain ovarian function and thereby improve their quality of life. 2 MATERIAL AND METHODS We retrospectively reviewed the demographic data and pathologic materials of women with cervical carcinoma who were treated at the Sichuan Cancer Hospital, Sichuan People s Hospital, and West China Second University Hospital, Sichuan Province, China, between January 2006 and December The inclusion criteria were as follows: (a) patients with stage IA IIB cervical carcinoma; (b) patients who had undergone surgery consisting of radical hysterectomy, pelvic lymphadenectomy, and bilateral oophorectomy or wedge resection of the ovary, with no radiotherapy or chemotherapy before surgery; and (c) patients with no history of benign or malignant ovarian tumors. To determine the risk factors for OM in cervical cancer, we reviewed the demographic data, International Federation of Obstetrics and Gynecology (FIGO) stages, and histopathologic records, which consisted of histologic types (squamous vs nonsquamous), tumor size ( 4 cm vs <4 cm), pelvic lymph node metastasis, lymphovascular space invasion, uterine corpus invasion, depth of stromal invasion ( 1/2 vs <1/2), parametrial invasion, and vaginal invasion. 2.1 Statistical analyses SPSS 23.0 statistical software (SPSS Inc., Chicago, IL, USA) was used for data analysis. Continuous variables are expressed as mean ± standard deviation and were assessed using independent sample t tests as necessary. Chi squared test was used to compare categorical variables. Fisher s exact test was used when observations in any cell of the 2 2 table was expected to be <5. Logistic regression analysis was used to control confounding variables. Odds ratio (OR) and 95% CI were used to assess the risk of OM. All statistical tests were twosided, and P values <0.05 were considered statistically significant. 2.2 Ethical approval This study was approved by clinical trials and the biomedical ethics committee, West China Hospital of Sichuan University (grant no ). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. 3 RESULTS A total of 3292 patients were included, with a mean age of 48.3 ± (range: 23 68) years. The OM rate was 3.78% (54/1429) in patients younger than 45 years and 3.27% (61/1863) in patients older than 45 years. The metastasis rate of the younger women was slightly higher than that of the older patients, but the results were not statistically significant (P = 0.435). Among the 3292 women, 115 (3.49%) were pathologically confirmed to have OM. The characteristics of the women are summarized in Table 1. The OM rate was 2% (56/2794) for squamous cell carcinoma and 11.8% (59/498) for nonsquamous cell carcinoma, among which, 49 had adenocarcinoma, 8 had adenosquamous carcinoma, and 2 had small cell carcinoma. The results were statistically significant between squamous cell carcinoma and nonsquamous cell carcinoma (P < 0.001). According to FIGO staging, the OM rate was 1.2% (2/164) in stage IA, 3.3% (63/1937) in stage IB, 4.1% (37/896) in stage IIA, and 4.4% (13/295) in stage IIB. The OM rate of stage IA was statistically lower in the univariate analysis (P = 0.013). Univariate analysis (Table 1) showed histologic types (squamous vs nonsquamous: P < 0.001), FIGO stage (IA vs IB IIB: P = 0.048), lymph node metastasis (positive vs negative: P = 0.009), lymphovascular space invasion (positive vs negative: P = 0.008), parametrial invasion (positive vs negative: P = 0.041), and uterine corpus invasion (positive vs negative: P < 0.001) to be statistically associated with the presence of OM. Other risk factors, including tumor size, depth of stromal invasion, and vaginal invasion, were not significantly associated with OM (P > 0.05). The significant variables in the univariate analysis were examined in logistic regression analyses, including histologic types, FIGO stage,

4 21 TABLE 3 The previous studies evaluate the clinicopathologic risk factors for ovarian metastasis in cervical cancer Study FIGO clinical stage Histologic type Number Results 1 Tabata IB III Epidermoid vs adenocarcinoma 1706 This ovarian metastasis rate was significantly higher et al 15 in cases with corpus invasion IB IIIB Squamous vs adenocarcinoma 597 It is fairly safe to preserve the ovary at the time of 2 Toki et al radical operation in squamous cell carcinoma of the uterine cervix, but it may not be safe to preserve the ovary in pure adenocarcinoma of the uterine cervix 3 Aida et al 16 IB II Adenocarcinoma and adenosquamous 4 Natsume et al 14 All stages of cervical carcinoma 82 Only deep stromal invasion to be an independent risk factor for ovarian metastasis Squamous and adenocarcinoma 239 Blood vessel invasion and parametrial invasion as well as adenocarcinoma or adenosquamous carcinoma were significantly related to ovarian metastasis IB IIIB Squamous and adenocarcinoma 631 Histologic type and blood vessel invasion were 5 Sakuragi et al 17 significant independent risk factors for ovarian metastases 6 Nakanishi et al 12 All clinical stages Squamous vs adenocarcinoma 1064 Clinical stage beyond IIB was a significant variable of squamous cell carcinoma, and more than 30 mm tumor size was significant in adenocarcinoma IB IIB Squamous and adenocarcinoma 3471 Ovarian metastasis occurred more frequently 7 Yamamoto et al 13 among patients with adenocarcinoma than among those with squamous cell carcinoma IB IIB Squamous and adenocarcinoma 3471 Ovarian metastasis occurred more frequently 8 Shimada et al 18 among patients with adenocarcinoma than among those with squamous cell carcinoma IA2 IIA All histologic type 1695 Age, FIGO stage, histology, and unaffected 9 Landoni et al 19 peripheral stromal thickness (<3 mm) to be independent risk factors for ovarian metastases 10 Kim et al 20 IIB All histologic type 625 Uterine corpus involvement was the only independent risk factor for ovarian metastasis, in addition to histologic types. Nonsquamous cell carcinoma is more likely to spread to extrauterine lesions than squamous cell carcinoma with uterine corpus involvement IIB Squamous and adenocarcinoma 578 Ovarian metastatic rate was significantly higher in 11 Kasamatsu et al 21 patients with pathologic stage IIB adenocarcinoma. Positive node is a common independent prognostic factor for survival and relapse of patients with adenocarcinoma. FIGO stage I IIB patients with squamous or adenocarcinoma have similar prognosis and spread pattern, but different ovarian metastasis rates 12 Hu et al 22 IB IIB All histologic types 1889 Lymph node metastasis, corpus uteri invasion, parametrial invasion, histology, and neoadjuvant chemotherapy were associated with ovarian metastasis 13 Lu et al 23 IIA2 Adenocarcinoma 101 Lymph vascular space invasion and lymph node metastasis were independent risk factors for ovarian metastasis in all stages of cervical adenocarcinoma, but involvement of the junction of the cervix and the body of the uterus was an independent risk factor for ovarian metastasis in stage IB

5 22 lymph node metastasis, lymphovascular space invasion, parametrial invasion, and uterine corpus invasion. Stepwise regression analysis identified histologic type (OR 8.76, 95% CI ), lymph node metastasis (OR 2.57, 95% CI ), lymphovascular space invasion (OR 2.82, 95% CI ), and corpus invasion (OR 6.34, 95% CI ) to be independently associated with OM (Table 2). The OR values of the histologic type and the corpus invasion were very high, which indicated that these were the most important risk factors. 4 DISCUSSION Ovarian preservation is an important issue when deciding on surgery for cervical cancer in young women. Hormonal fluctuations greatly affect quality of life after bilateral oophorectomy. Preserving the ovaries helps to prevent menopausal symptoms, osteoporosis, and cardiovascular diseases associated with estrogen deficiency. There is an abundance of vascular anastomosis between the ovary and the uterus, and half of the ovarian blood supply is provided by the uterus, which may lead to OM in some patients with cervical cancer. 8,9 Although the OM rate of squamous cervical cancer is lower than 2% in most of the literature, the OM rate of cervical adenocarcinoma is relatively high; according to published reports, the OM rate ranged from 6.3% to 12.9% The patients who were diagnosed with OM had a poor prognosis. Therefore, diagnosis of high risk patients with OM is very important. In the past two decades, several reports have evaluated the clinicopathologic risk factors for OM in cervical cancer. However, different researchers reached different conclusions. Table 3 lists the previous studies evaluating the risk for OM. These studies have some limitations. First, these findings are noticeably different from each other, probably owing to small or heterogeneous study samples. Second, most of the studies were aimed at stage IA IIA patients, without stage IIB. Third, few studies have analyzed the factor of corpus invasion. There has been no consensus on the criteria for ovarian preservation so far; hence, we designed a multicenter retrospective study. In our study, the metastasis rate of the younger patients was slightly higher than that of the older patients, although the results were not statistically significant. The cause of this trend may be linked to ovarian atrophy and the decline of ovarian blood supply after menopause. The OM rate for nonsquamous cell carcinoma was higher than that for squamous cell carcinoma. In our study, the risk of OM was almost 9 times higher in patients with nonsquamous cell carcinoma than in patients with squamous cell carcinoma. Nakanishi et al demonstrated that the OM rate of squamous cell carcinoma was only 1.3%, but the OM rate of nonsquamous cell carcinoma was 6.3%. 12 OM was significantly more frequent in adenocarcinoma than in squamous cell carcinoma. The Gynecologic Oncology Group reported that the OM rate of nonsquamous cell carcinoma (1.7%) was higher than that of squamous cell carcinoma (0.5%). 11 The OR value of corpus invasion was very high, which indicated that this was one of the most important risk factors in our study, but only a few studies included it. Tabata et al and Hu et al found that corpus uteri invasion was associated with OM, but in the study by Lu et al, corpus uteri invasion was a risk factor for OM only in stage IB. 15,22,23 The reason why corpus invasion easily leads to OM may be due to the abundant blood supply to the endometrium of the uterus. Most of the blood supply to the ovaries comes from the uterus, and the shedding tumor cells can easily spread to the ovarian suspension ligament along this pathway. 24 However, it is difficult to judge the corpus invasion route during a preoperative examination. Further, the factor of corpus invasion is not included in the FIGO staging. Currently, in the clinical setting, corpus invasion is rarely used as an indicator of OM. Therefore, we suggest that corpus invasion might be a strong contraindication for preservation of the ovaries. Lymph node metastasis and lymphovascular space invasion were associated with OM in the logistic analysis. Many studies have explained the reason for this finding. Tabata et al demonstrated that OM might occur through the hematogenous spread of cervical carcinoma. 15 Sakuragi et al reported that lymphatic vessel invasion (P < 0.001) was significantly related to lymph node metastasis, and blood vessel invasion (P < 0.05) was significantly related to OM. 17 The FIGO stage was not significantly associated with OM (P > 0.05) in the logistic analysis. This was probably because the FIGO stage mainly describes the extension of the lesion to the vagina and the pelvic wall, which is not directly related to OM. Parametrial invasion was not significantly associated in the logistic analysis (P = 0.219) either. When the parametrium has been invaded, the major involvement of the uterine vessels is the descending branch of the uterine vessels; the ovarian blood supply mainly comes from the ascending branch. This may be one of the factors affecting the OM rate. This retrospective study was performed in women in whom the indication to perform oophorectomy or wedge resection of the ovaries lacked uniform standards. We were also unable to obtain relapse data on women who had their ovaries preserved, which are limitations of this study. 5 CONCLUSION Based on the present data, the histologic type, lymph node metastasis, lymphovascular space invasion, and corpus invasion were independently associated with OM. Histologic type and corpus invasion were the most important risk factors. Therefore, we suggest that corpus invasion might be a strong contraindication for preservation of the ovaries. CONFLICT OF INTEREST The authors have stated explicitly that there are no conflicts of interest in connection with this article.

6 23 ORCID Le Zhou REFERENCES 1. Small W, Bacon MA, Bajaj A, et al. Cervical cancer: a global health crisis. Cancer. 2017;123: Karimi ZM, Akhavan A, Fallahzadeh H, Gholami H, Dehghani A, Teimoori S. Outcome of cervical cancer in Iranian patients according to tumor histology, stage of disease and therapy. Asian Pac J Cancer Prev. 2010;11: Plante M. Evolution in fertility preserving options for early stage cervical cancer: radical trachelectomy, simple trachelectomy, neoadjuvant chemotherapy. Int J Gynecol Cancer. 2013;23: Bukovsky I, Halperin R, Schneider D, Golan A, Hertzianu I, Herman A. Ovarian function following abdominal hysterectomy with and without unilateral oophorectomy. Eur J Obstet Gynecol Reprod Biol. 1995;58: Hée P. Compliance to estrogen treatment one to three years after hysterectomy and bilateral salpingooophorectomy. The cohort s lifestyle, knowledge of ERT, benefits etc. Acta Obstet Gynecol Scand. 1999;78: Makar AP, Tropé C. Fertility preservation in gynecologic cancer. Acta Obstet Gynecol Scand. 2001;80: Kontoravdis A, Kalogirou D, Antoniou G, Kontoravdis N, Karakitsos P, Zourlas PA. Prophylactic oophorectomy in ovarian cancer prevention. Int J Gynaecol Obstet. 1996;54: Lamond DR, Drost M. Blood supply to the bovine ovary. J Anim Sci. 1974;38: Kelekci S, Yilmaz B, Yasar L, Savan K, Sonmez S, Kart C. Ovarian reserve and ovarian stromal blood supply after tubal ligation by the Pomeroy technique: comparison with controls. Gynecol Endocrinol. 2005;20: Toki N, Tsukamoto N, Kaku T, et al. Microscopic ovarian metastasis of the uterine cervical cancer. Gynecol Oncol. 1991;41: Sutton GP, Bundy BN, Delgado G, et al. Ovarian metastases in stage IB carcinoma of the cervix: a Gynecologic Oncology Group study. Am J Obstet Gynecol. 1992;166: Nakanishi T, Wakai K, Ishikawa H, et al. A comparison of ovarian metastasis between squamous cell carcinoma and adenocarcinoma of the uterine cervix. Gynecol Oncol. 2001;82: Yamamoto R, Okamoto K, Yukiharu T, Kaneuchi M, Negishi H, Sakuragi N, et al. A study of risk factors for ovarian metastases in stage Ib IIIb cervical carcinoma and analysis of ovarian function after a transposition. Gynecol Oncol. 2001;82: Natsume N, Aoki Y, Kase H, Kashima K, Sugaya S, Tanaka K. Ovarian metastasis in stage IB and II cervical adenocarcinoma. Gynecol Oncol. 1999;74: Tabata M, Ichinoe K, Sakuragi N, Shiina Y, Yamaguchi T, Mabuchi Y. Incidence of ovarian metastasis in patients with cancer of the uterine cervix. Gynecol Oncol. 1987;28: Aida H, Kodama S, Aoki Y, et al. [The study of ovarian metastasis in uterine cancer]. Nihon Sanka Fujinka Gakkai Zasshi. 1992;44: Sakuragi N, Takeda N, Hareyama H, et al. A multivariate analysis of blood vessel and lymph vessel invasion as predictors of ovarian and lymph node metastases in patients with cervical carcinoma. Cancer. 2000;88: Shimada M, Kigawa J, Nishimura R, et al. Ovarian metastasis in carcinoma of the uterine cervix. Gynecol Oncol. 2006;101: Landoni F, Zanagnolo V, Lovato Diaz L, et al. Ovarian metastases in early stage cervical cancer (IA2 IIA): a multicenter retrospective study of 1965 patients (a Cooperative Task Force study). Int J Gynecol Cancer. 2007;17: Kim MJ, Chung HH, Kim JW, Park NH, Song YS, Kang SB. Uterine corpus involvement as well as histologic type is an independent predictor of ovarian metastasis in uterine cervical cancer. J Gynecol Oncol. 2008;19: Kasamatsu T, Onda T, Sawada M, et al. Radical hysterectomy for FIGO stage I IIB adenocarcinoma of the uterine cervix. Br J Cancer. 2009;100: Hu T, Wu L, Xing H, et al. Development of criteria for ovarian preservation in cervical cancer patients treated with radical surgery with or without neoadjuvant chemotherapy: a multicenter retrospective study and meta analysis. Ann Surg Oncol. 2013;20: Lu H, Li J, Wang L, et al. Is ovarian preservation feasible in early stage adenocarcinoma of the cervix. Med Sci Monit. 2016;22: Hossain MI, O Shea JD. The vascular anatomy of the ovary and the relative contribution of the ovarian and uterine arteries to the blood supply of the ovary in the guinea pig. J Anat. 1983;137(Pt 3): How to cite this article: Zhou L, Sun C T, Lin L, et al. Independent risk factors for ovarian metastases in stage IA IIB cervical carcinoma. Acta Obstet Gynecol Scand. 2019;98: