Effect of Cervical Conization with Positive Margins on Lymph Node Metastasis in Cervical Cancer

Radical surgery after cervical conization is a common approach for the treatment of cervical cancer. Clinically, rapid disease progression is sometimes observed in patients with positive margins at conization, but the details are unclear. This study aimed to investigate the clinical impact of positive margins at conization in cervical cancer. The medical records of patients with cervical cancer between 2010 and 2020 were reviewed retrospectively and we identied 101 eligible patients who underwent radical hysterectomy, including pelvic lymph node dissection. The association between the positive margins of the conization samples and outcomes, including subsequent lymph node metastasis, was evaluated. The rate of lymphovascular space invasion positivity at radical surgery was signicantly higher in patients with positive margins (p = 0.017) than in those with negative margins, although there was no signicant difference in the rate of pelvic lymph node metastasis (p = 0.155). Moreover, Kaplan– Meier curves showed that there were no signicant difference in overall survival or progression-free survival between the two groups (p = 0.332 and p = 0.200, respectively). A positive margin at conization presented no signicant prognostic disadvantage, and therefore, diagnostic conization is one of the reasonable options for locally advanced cervical cancer.


Introduction
Recently, the number of patients with cervical cancer has continued to increase and it remains a major health problem. In 2020, about 600,000 patients were diagnosed with cervical cancer globally, and more than 340,000 patients died due to the progression of the disease 1 . In comparison with other developed countries, the cervical cancer prevention program has been unsatisfactory in Japan, and the human papillomavirus vaccination and Pap smear rates were only 0.3% and 34%, respectively, in 2016 2 3 4 . Hence, the incidence has remained unchanged in recent years, with about 34,000 newly diagnosed patients and 3,000 deaths annually 5 6 . Moreover, in developing countries, the prevention program has not been fully provided. Therefore, cervical cancer remains a life-threatening disease.
Cervical cancer is usually diagnosed based on a combination of pelvic examination, colposcopy, and radiological ndings. Moreover, especially for early stage disease, diagnostic conization is one of the options for evaluating the spread of the disease more accurately. According to the National Comprehensive Cancer Network (NCCN) guidelines (Version 4.2019 Cervical Cancer), diagnostic conization such as "Cone biopsy (i.e., conization) is recommended if the cervical biopsy is inadequate to de ne invasiveness or if accurate assessment of microinvasive disease is required" (MS-4) 7 .
Subsequently, based on the stage and pathological ndings of conization, the treatment method is determined. Radical hysterectomy, concurrent chemoradiation (CCRT) and radiation therapy (RT) are major treatment options for cervical cancer. In the case of radical hysterectomy, it is necessary to evaluate whether pelvic lymph node dissection should be performed because adverse events like lymphedema can occur. Therefore, it is highly required that accurate preoperative risk assessment of lymph node metastasis is performed. As reported in several studies, lymphovascular space invasion (LVSI) is a risk factor for lymph node metastasis, and therefore, diagnostic conization can be useful for risk assessment 8 9 10 11 12 . However, there is a concern regarding excessive conization for locally advanced cases that may cut into the tumor itself. In fact, clinically, rapid disease progression is sometimes observed in patients with positive margins at conization, but the details are unclear.
Therefore, this study aimed to evaluate the clinical impact of cervical conization with positive margins on subsequent tumor progression and prognosis of patients with cervical cancer.

Patients And Methods
We retrospectively reviewed the records of 443 patients with cervical cancer who underwent radical surgery at the Nagoya University Hospital (Nagoya, Japan) from January 2010 to May 2020. Several gynecologic oncologists in our hospital determined the treatment strategies for each patient, and diagnostic conization was performed when biopsy ndings were insu cient. In accordance with the Japanese guidelines, if possible, radical surgery plus pelvic lymph node dissection was performed, and adjuvant CCRT or RT was considered depending on pathological ndings. Otherwise, patients basically received CCRT. Of the 443 patients, we included 101 patients who underwent conization followed by radical surgery in the present study, excluding 342 patients without conization (Fig. 1). Based on the pathological ndings of conization, we classi ed the patients into the positive margin (69 patients) and negative margin (32 patients) groups. We investigated the clinical information including age, stage, histological type, LVSI at conization, LVSI at radical surgery, pelvic lymph node metastasis, adjuvant therapy, recurrence, and outcome and compared the clinical outcomes between the two groups. This study was approved by the ethics committee of our institute (Approval No. 2019 − 0106, Nagoya University Graduate School of Medicine, Nagoya, Japan). All methods were carried out in accordance with relevant guidelines and regulations and our study strictly followed the requirements of the Declaration of Helsinki. Informed consent was obtained from all participants.
Statistical analyses were performed using SPSS version 27 (IBM Corp., Armonk, NY). Mann-Whitney U test and chi-squared tests were used for comparing the two groups. Overall survival (OS) was de ned as the time from the primary therapy to all-cause mortality, and progression-free survival (PFS) was the time from primary therapy to tumor progression, recurrence, or all-cause mortality. Kaplan-Meier curves were used for the analysis of OS and PFS, whereas the log-rank test was used to compare the survival curves.
A p value of < 0.05 was considered statistically signi cant.

Results
Patient characteristics are shown in Table 1, and age, stage, and histology were well-balanced between the two groups. In particular, the margin positivity and LVSI positivity at conization were independent of each other (p = 0.417). First, we investigated whether positive margins at conization contributed to cancer progression. The LVSI positive rate at radical surgery was signi cantly higher in the positive margin group than in the negative margin group (the positive and negative margin groups, 21.7% and 3.1%, respectively; p = 0.017; Table 2). Similarly, pelvic lymph node metastasis tended to increase in the positive margin group; however, there was no signi cant difference between the groups (the positive and negative margin groups, 11.6% and 3.1%, respectively; p = 0.155). In addition, we focused on patients with stage IB1 squamous cell carcinoma, the main subpopulation of our cohort. Similar to the above results, LVSI at radical surgery was also signi cantly higher in patients in the positive margin group than in those in the negative margin group (the positive and negative margin groups, 27.3% and 0%, respectively; p = 0.005; Table 3). Moreover, a high frequency of lymph node metastasis was observed in the positive margin group, with no signi cant difference (the positive and negative margin groups, 11.4 % and 4.8 %, respectively; p = 0.362).  Subsequently, we evaluated the difference between the LVSI status at conization and at radical surgery (Fig. 2a). The LVSI statuses at conization and radical surgery were consistent in most patients. However, negative LVSI at conization became positive at radical surgery in four cases, three of whom were in the positive margin group. The detailed clinical information about one of the cases is shown in Fig. 2b-d. She underwent diagnostic conization because both Pap smear and biopsy were negative, despite suspicions of cervical cancer by transvaginal ultrasound. In conization, the tumor was resected to the extent possible, while cutting through the tumor wall. Hematoxylin and eosin staining revealed that it was a pT1b1 adenocarcinoma (AC) with negative LVSI and a positive surgical margin (Fig. 2b). However, after radical surgery, the disease up-staged to pT2a1N1 and LVSI turned highly positive (Fig. 2c). Moreover, obvious enlargement of the left common iliac lymph node was identi ed using computed tomography, although there were no swelling before conization (Fig. 2d).
Finally, we analyzed the PFS and OS in the positive and negative margin groups. The median follow-up period was 56 months (range: 4-124 months). The 5-year PFS rates in the positive and negative margin groups were 92.3% and 96.8%, respectively (Fig. 3a). The 5-year OS rates in the positive and negative margin groups were 96.6 % and 100 %, respectively (Fig. 3b). Kaplan-Meier curves indicated that there were no statistically signi cant differences between the two groups in PFS (p = 0.200; log-rank test) and OS (p = 0.332; log-rank test).

Discussion
According to the Japanese and NCCN guidelines, diagnostic conization is an important method for determining the exact stage and the most accurate treatment strategy for patients with cervical cancer 7 13 14 . However, as depicted by the results of the present study, there is a potential risk of cutting into a tumor in some cases and resulting in a positive margin at conization. Until now, there is no consensus on whether diagnostic conization is appropriate even when positive margins are predicted. Therefore, we evaluated the association between diagnostic conization with positive margins and patient outcomes, including lymph node metastasis in uterine cervical cancer. Hence, the method has also been applied for pancreatic and liver cancers 17 18 . For uterine cervical cancer, patients who underwent minimally invasive radical hysterectomy could have shorter survival than those who underwent conventional abdominal radical hysterectomy. A possible explanation is the use of an intrauterine manipulator in the minimally invasive surgery, which might increase intrauterine pressure and spread cancer cells to the lymphovascular space 15 19 20 21 . However, other reports showed that the use of manipulators is not associated with worse prognosis, and its clinical signi cance is still debated 22 23 . Therefore, the basic principles are important, but they are not absolute factors, and other factors may have a great in uence.
Diagnostic conization does not meet these principles in case of a positive margin. The risk of positive margins at conization was associated with several factors, such as menopausal status, grade and size of disease, devices used for conization, and purpose of conization (diagnostic or therapeutic) 24 25 26 . According to previous studies, a positive margin at conization would mean residual disease; however, it was not associated with parametrial invasion at the time of hysterectomy 27 28 29 . In the present study, we showed that positive margins may increase the rate of LVSI positivity in hysterectomy samples, although there was no statistical signi cance regarding pelvic lymph node metastasis. Moreover, there was no signi cant difference in PFS and OS between the positive and negative margin groups. Consistent with our results, a recent report also indicated that there were no signi cant differences in lymph node metastasis, LVSI positivity, recurrence, and death between the patients with positive and negative margins 30 . Therefore, the prognostic impact of a positive margin at conization was considered to be limited.
However, in some cases, obvious lymph node enlargement and strong LVSI positivity can occur after diagnostic conization with a positive margin. Needless to say, lymph node metastasis is one of the worst prognostic factors in cervical cancer 31 32 . Therefore, regardless of the lack of statistical signi cance, clinicians should keep in mind that there are cases with rapid progression after conization. Moreover, due to the present relatively small-scale retrospective study, it was di cult to statistically evaluate such rare cases. Therefore, further studies are needed to evaluate the potential risk of a positive margin at conization.
In conclusion, the present study provided important evidence for the association between cervical conization with positive margins and subsequent cancer progression. We showed that there were no signi cant differences in the lymph node metastasis rate and patient prognosis between the positive and negative margin groups. Therefore, diagnostic conization is considered acceptable even if a positive margin will occur.
Declarations Figure 1 Flowchart of patient selection Of the 443 patients who underwent radical surgery, including pelvic lymph node dissection, for cervical cancer from January 2010 to May 2020, we excluded 342 patients without conization and included 101 patients with conization. Of the 101 patients, 69 had positive margins at conization and 32 had negative margins