**3. Results**

Between 2014 and 2019, 48 patients with presumed early stage ovarian cancer were submitted to surgery with curative intent, the median age at the time of surgery being 43.4 years (range = 28–56 years). According to their menopausal status, there were 13 postmenopausal women. The preoperative diagnostic was suspected based on the detection of higher levels of CA 125 (the median value being 330 U/mL) in association with the imaging detection of ovarian masses/cysts with uncertain aspect. In all cases, surgery consisted of total hysterectomy en bloc with bilateral adnexectomy, pelvic and para-aortic lymph node dissection, omentectomy, serial peritoneal biopsies and also an associated resection of all suspect lesions found at the level of the abdominal cavity and peritoneal washing. Intraoperative details are presented in Table 1.

In all cases, pelvic and para-aortic lymph node dissection was performed, the borders of the lymph node dissection being represented by the origin of the epigastric artery caudally (for the pelvic lymph node dissection) and, respectively, the origin of the renal artery cranially (for the para-aortic lymph node dissection). The median number of retrieved pelvic lymph nodes was 19 while the median number of the retrieved para-aortic lymph nodes was 14. Intraoperative and histopathological details of the lymph node dissection are presented in Table 2.

The median length of the surgical procedure was 130 min (range = 90–160 min), the median estimated blood loss was 350 mL (range = 100–550 mL), while the median length of the hospital stay was 6 days (range = 4–13 days). The histopathological studies confirmed the presence of positive pelvic lymph nodes in 18% of cases and, respectively, positive para-aortic lymph nodes in 22% of cases. However, all cases presenting positive pelvic nodes also had associated positive para-aortic lymph nodes. According to these findings, all cases with positive retroperitoneal lymph nodes were upgraded to FIGO stage III of disease and were therefore confined to the oncology department to be submitted to adjuvant chemotherapy. Patients with positive lymph nodes were further classified as IIIA1 (i) if the dimension of the metastatic deposits was lower than 10 mm (in three cases), IIIA1 (ii) if the dimension of the metastatic deposits was larger than 10 mm (in five cases) and IIIB if macroscopic, lower than 2 cm, extrapelvic peritoneal metastases were encountered (in the remaining three cases). None of these cases presented macroscopic peritoneal metastases larger than 2 cm; therefore, none of them were upstaged to FIGO stage IIIC of disease.


**Table 1.** Preoperative and intraoperative characteristics of the 48 patients diagnosed with presumed early stage ovarian cancer.

**Table 2.** Intraoperative and histopathological details of the lymph node dissection.


In order to determine the risk factors for developing lymph node metastases in apparently early stage ovarian cancer, we conducted an univariate analysis in which we studied the influence of age, menopausal status, initial FIGO stage at diagnosis, laterality of the tumor, histology and degree of differentiation on the risk of developing node metastases. The univariate analysis demonstrated that the presence of positivity of the retrieved lymph nodes was significantly associated with the serous histopathological subtype as well as with the degree of differentiation. Therefore, patients diagnosed with serous ovarian carcinoma had a significantly higher rate of positive lymph nodes (when compared to the other histopathological subtypes, *p* = 0.002). In the meantime, cases diagnosed with poorly differentiated tumors also exhibited a significantly higher rate of positive lymph nodes when compared to the other degrees of differentiation (*p* = 0.004). Surprisingly, neither the laterality of the tumor nor the presumed FIGO stage at diagnosis influenced the risk of developing such metastases. Data obtained at statistical analysis is presented in the table below (Table 3).


**Table 3.** Analysis of risk factors for para-aortic lymph node metastases.

#### **4. Discussion**

The issue of lymph node metastases in presumed early stage ovarian cancer has a particular interest among surgical oncologists, gynecological oncologists and medical oncologists worldwide. It can be observed from the data presented so far that an important number of cases diagnosed with presumed early stage cancer already have, at the time of diagnosis, positive microscopic and even macroscopic lymph nodes, upstaging in this way the disease to a FIGO stage III malignancy. Therefore, all these cases, if submitted to standard treatment for early stage ovarian cancer, are at risk for developing early recurrent disease and a particularly poor long-term prognosis due to mis-staging. In the meantime, routine performance of extended lymph node dissection might predispose a significant number of cases to overtreatment and its secondary early-term and even long-term complications [2,4,13,14].

In order to increase the rates of preoperative detection of potential positive lymph nodes, certain authors proposed routine association of positron emission computed tomography. In the study conducted by Signorelli et al. published in 2013, the authors included 68 patients with presumed early stage ovarian cancer in which routine positron emission tomography, as well as systematic lymph node dissection, was performed. The authors underlined the fact that among the 12 cases who finally presented lymph node metastases at the histopathological studies, 10 cases had been correctly previously identified at the imaging studies. Therefore, the authors concluded that this imaging tool could be safely used in order to identify cases in which systematic lymph node dissection could be avoided, especially based on the high negative predictive value of the method [8]. Another promising method which might provide a more accurate identification of patients who present ovarian cancer lymph node metastases even in apparently early stages of the disease is represented by the sentinel node detection [4]. The method, which has been widely implemented in cases diagnosed with early stage breast cancer, melanoma and even gynecological cancers (such as endometrial cancer or cervical cancer), is still under evaluation in patients with presumed early stage ovarian cancer, further studies being still needed before introducing it as part of the standard therapeutic protocol [4,15–18].

One of the first studies, which designed a nomogram-based analysis in order to identify cases at risk of developing para-aortic lymph node metastases, was conducted by Bogani et al., on 290 patients with presumed early stage disease. According to their study, the authors demonstrated that bilateral lesions as well as high-grade serous histology represent the strongest predictors for para-aortic lymph node metastases even in cases with presumed early stage disease [2].

A similar conclusion was also presented by the study conducted by Zhou et al. [19]. In the paper published in 2016, the Chinese authors came to demonstrate that systematic lymph node dissection should be performed in cases diagnosed with poorly differentiated tumors, with serous histology and higher values of CA125 at the time of diagnosis. Therefore, cases in which the preoperative levels of CA125 surpass 740 U/mL seem to have a higher risk of associated para-aortic lymph node metastases [19].

An interesting study which investigated the effectiveness of surgical staging in cases with apparent early stage ovarian cancer has been recently published by Hengeveld in 2019. The study included all patients submitted to surgery with presumed early stage disease between 2005 and 2017 in Danish and Dutch hospitals [20]. Finally, there were 1234 cases that had been preoperatively presumed to be classified as FIGO stage I disease; in all cases, omentectomy, pelvic and para-aortic lymph node sampling or lymph node dissection, as well as multiple peritoneal biopsies, were retrieved. After analyzing the specimens, the histopathological studies revealed the fact that 20 patients were finally upstaged due to the presence of positive pelvic lymph nodes (in seven cases), positive para-aortic lymph nodes (in 12 cases) and both pelvic and para-aortic lymph nodes (in one case). Moreover, the authors underlined the fact that a total of 207 cases were upstaged after applying this protocol, with other sites of involvement being represented by the omentum, peritoneum or positive cytology. However, the authors underlined the fact that in another 50 cases, the malignant process was down-staged after applying this protocol, as the histopathological analysis of the macroscopically suspect lesions was not able to confirm the disease. This fact was rather explained by the absence of bilateral lesions and the absence of capsular invasion, respectively. Similarly to our study, the presence of serous histology as well as the poorer degree of differentiation significantly impacted on the risk of further upstaging. Other factors that were significantly associated with upstaging were represented by higher age, the postmenopausal status as well as the endometroid histology. Therefore, when it comes to the type of histology that is mainly associated with a poorer prognosis, according to this study, serous and endometroid histology versus any other type of tumor were associated with higher rates of upstaging. Moreover, the authors underlined the fact that upstaging was responsible for changing the plan of treatment in 35.1% of cases. Therefore, the importance of an adequate staging was underlined once again; in the meantime, the study came to demonstrate that a significant proportion of cases that were finally upstaged originated from cases with serous or endometrial histology in association with a lower degree of differentiation [20].

Another extremely interesting study that came to demonstrate the effect of the upstaging of presumed early stage ovarian cancer has been recently published in the *New England Journal of Medicine* in 2019. In this paper, the authors came to demonstrate that seven patients out of the 15 cases that were included in the presumed early stage ovarian cancer group presented in fact para-aortic lymph node metastases and were therefore upstaged to FIGO stage IIIC of disease. Moreover, the authors underlined the fact that performing para-aortic lymph node dissection in presumed early stages of the disease will probably prolong the surgical procedure (which is otherwise a short one) by almost an hour and will not predispose to such important complications when compared to cases diagnosed in advanced stages of the disease. In the meantime, routine association of this procedure in cases with advanced stages will also prolong a more demanding and laborious surgical procedure by another hour and will predispose to significant postoperative complications, such as a larger amount of ascites and lymphorrhea. Moreover, the authors also demonstrated that performing systematic lymph node dissection in patients with clinically negative lymph nodes increases the risk of perioperative complications without improving the long-term outcome [21].
