Survival Outcomes in Patients With 2018 FIGO Stage IA2–IIA2 Cervical Cancer Treated With Laparoscopic Versus Open Radical Hysterectomy: A Propensity Score-Weighting Analysis

Objective To compare the survival and recurrence outcomes between open and laparoscopic radically hysterectomy (RH) for stage IA2-IIA2 cervical cancer based on Federation International of Gynecology and Obstetrics (FIGO) 2018. Methods Data of 1,373 early cervical cancer patients undergoing open or laparoscopic radically hysterectomy at ShengJing Hospital of China Medical University between January 1, 2013, and December 31, 2016, were retrospectively reviewed. Propensity score-based inverse probability of treatment weighting (PS-IPTW) was used to balance the covariates between the two groups. Results A total of 705 cervical cancer patients of FIGO 2009 stage IA2-IIA2 were finally enrolled in this study. After IPTW adjustment, the OS (HR = 2.095, 95% CI: 1.233-3.562, P = 0.006) and PFS (HR = 1.950, 95%CI: 1.194-3.184, P = 0.008) rates were significantly higher in the open RH (ORH) group compared with the laparoscopic RH (LRH) group. Then after re-staging according to the FIGO 2018 staging system, 561 patients still belonged to stage IA2-IIA2, 144 patients were upgraded to stage IIIC1p-IIIC2p. The ORH group had a significantly superior OS (HR = 1.977, 95%CI: 1.077-3.626, P = 0.028) and PFS (HR = 1.811, 95%CI: 1.046-3.134, P = 0.034) compared with the LRH group after PS-IPTW analysis. Furthermore, in patients with no high and intermediate risks, difference of the OS (HR = 1.386, 95%CI: 0.287-6.69, P = 0.684) and PFS (HR = 1.524, 95%CI: 0.363-6.396, P = 0.565) rates between the two groups were with no statistical meaning. Conclusions Outcomes of this retrospective cohort study were in compliance with indications for ORH recommended by the National Comprehensive Cancer Network guidelines Version 1, 2021. However, LRH showed non-inferiority for patients with no prognostic risk factors compared with ORH.


INTRODUCTION
Cervical cancer was the fourth most common cancer in women worldwide. The estimated global yearly incidence of cervical cancer in 2018 was 570,000 cases, among which China contributing approximately one fifth (1). Guidelines indicate that radical hysterectomy is standard treatment for early-stage cervical cancer. However, disputes about the prognostic outcomes of the cervical cancer patients underwent the open radically hysterectomy (ORH) or minimally invasive surgery (MIS) laparoscopic hysterectomy had persisted nearly 25 years since the introduction of the laparoscopic approach in 1992 (2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18). Unexpectedly, a phase III randomized, open-label, noninferiority clinical trial named the Laparoscopic Approach to Cervical Cancer identified that cervical cancer patients of Federation International of Gynecology and Obstetrics (FIGO) 2009 stage IA1 with lymphovascular invasion, IA2 and IB2 in the MIS group had almost four times the risk of recurrence and 6 times the risk of death compared with the women in the ORH group (7). Afterward, a meta-analysis, which included 15 highquality observational studies comprising 9,499 cervical cancer patients of FIGO 2009 stage IA1-IIA2 (stage IA1 with lymphovascular invasion), concluded that patients in the ORH group had superior overall and disease-free survival than patients in the MIS group (6). Besides, the European Society of Gynecological Oncology (ESGO), the British Gynecological Cancer Society (BGCS), and two other epidemiologic studies also had the same opinion (9,19,20).
Based on the most recent findings, the National Comprehensive Cancer Network (NCCN) Guidelines Version 1.2021 recommended that ORH was the primary treatment for FIGO 2018 stage IA2, IB1, IB2 and IIA1 cervical cancer patients. As for operable cervical cancer patients with FIGO 2018 stage IB3, IIA2, and IIICr, laparoscopic approach is absolutely prohibited. The panel had updated the guidelines according to the revised 2018 FIGO staging system. However, trial data utilized in the guidelines were all from the previous 2009 FIGO staging system. Accordingly, this study is conducted to compare the survival and recurrence outcomes between ORH and laparoscopic radically hysterectomy (LRH) for stage IA2-IIA2 cervical cancer patients comprising FIGO 2018 staging system. In addition, three subgroups of FIGO 2018 stage IA1, IB1, IB2 and IIA1, FIGO 2018 stage IB3 and IIA2, FIGO 2018 stage IIIC1p-IIIC2p referring to the NCCN guidelines Version 1.2021 were analyzed between the two surgical approaches. As squamous cell carcinomas account for approximately 75% to 80% of all cervical cancers (21), and stromal invasion pattern was considered to play better roles on predicting the prognosis of adenocarcinoma and adenosquamous carcinoma than FIGO stage system (22), the histology type of patients included in this study is only squamous cell carcinoma.

Inclusion and Exclusion Criteria
This is a retrospective, single-center study of cervical cancer patients undergoing surgery at ShengJing Hospital of China Medical University between January 1, 2013, and January 1, 2016. The inclusion criteria were as follows: (1) the clinical diagnosis of the participants was International Federation of Gynecology and Obstetrics (FIGO) 2009 stage IA2-IIA2; (2) surgery could be accomplished via either open or laparoscopic radically hysterectomy and pelvic lymphadenectomy, with or without para-abdominal aortic lymphadenectomy; (3) the histological diagnosis was squamous cell carcinoma; (4) no neoadjuvant radiation or chemotherapy prior to surgery. The exclusion criteria were as follows: (1) pregnancy combined with cervical cancer; (2) combined with other malignant or borderline tumors at other sites; (3) cervical stump cancer. And the study was approved by the ethics review board of ShengJing Hospital of China Medical University. Inclusion condition of the participants is shown in Figure 1.

Surgical Approach and Observation Index
The radical hysterectomy includes resection of the uterus as far as possible from the uterosacral ligament, excision of the parametrial tissue as near as to the pelvic wall and removal of the upper 1/3 of the vagina. Extent of radical hysterectomy was evaluated according to Querleu and Morrow classification (23). The following data were collected from the medical records of the included participants: demographic information, body mass index (BMI), operation year, histological type, clinical stage (FIGO 2009), tumor size (pathologic tumor size), tumor grade, pathological results, and adjuvant treatment condition. And all the patients were re-staged according to the FIGO 2018 criterion.

Follow-Up
After surgery, patients were informed to come back for checkup every 3 to 6 months for the first 2 years, 6 to 12 months for 3 to 5 years, then annually thereafter. The follow-up procedures were conducted by professional gynecologists through telephone. And the last follow-up time was December 31, 2020.
The postoperative adjuvant treatments, survival status, time of death, reasons for death, recurrence time, and location were recorded. The primary endpoint was overall survival (OS) which was defined as the time (months) from initial diagnosis to death from any causes. The secondary endpoint was progression-free survival (PFS) which was defined as the time (months) from diagnosis to disease recurrence. Data of patients with no evidence of death or recurrence were censored.

Statistical Analysis
Continuous quantitative data are presented as mean ± standard deviation (SD) and analyzed through Student t test. Categorical   (24). Survival analysis was conducted through Kaplan-Meier method and compared with log-rank test (25). Univariate and multivariate Cox proportional hazards regression analyses were also applied to calculate the hazard ratio (HR) and 95% confidence internal (CI) associated with the recurrence and survival outcomes of the cervical cancer patients (26). Propensity score-based inverse probability of treatment weighting (PS-IPTW) was used to balance the covariates associated with the recurrence and survival outcomes of the cervical cancer patients between the ORH and LRH groups. And the covariates include: age, operation year, clinical stage, tumor size, tumor grade, stromal invasion, lymphovascular space invasion (LVSI), pelvic node, aortic node, parametrial invasion, vaginal margin invasion, nerve invasion, and chemoradiotherapy condition. The IPTW approach assigned in LRH group was weight = Pt/PS, and weight = (1 − Pt)/(1 − PS) in the ORH group (27). Pt was the percentage of the number of the LRH group according to the total patients. PS was the propensity score of each patient. The data analysis was conducted with Statistical Package for Social Sciences (IBM SPSS, Version 25, Armonk, NY) and R version 4.0.3 (R Foundation for Statistical Computing, Vienna, Austria). A p-value < 0.05 indicated a statistically significant difference.

Patient Enrollment
A total of 1,373 early-stage cervical cancer patients were identified accepting operation in ShengJing Hospital of China Medical University between January 1, 2013, and January 1, 2016. As shown in Figure 1

Characteristics of the Covariates
Comparison of the clinicopathological parameters of stage IA2-IIA2 before and after IPTW adjustment between the ORH and the LRH group on the basis of FIGO 2009 and FIGO 2018 criteria were listed in Tables 1 and 2, respectively. In the original sample, age at operation, operation year, clinical stage, tumor size, and stromal invasion condition were obviously discrepant between the two groups (P<0.05). After IPTW adjustment, no difference of the covariates was identified between the two groups (P>0.05). As the data of BMI were not recorded adequately, which were not included in the covariates. Similarly, the covariates between the two groups in the three subgroups were also balanced with IPTW analysis.

Survival Analysis
Median follow-up was 61 months (range, 23-94 months) in the ORH group versus 57.5 months (range, 26-88 months) in the LRH group. And among all the enrolled patients, 66 patients died and 78 patients had recurrence in the ORH group, 22 patients died and 27 patients had recurrence in the LRH group up to December 31, 2020. The 5-year OS rates were 88.4% in the ORH group and 83.7% in the LRH group (Log-Rank P=0.202, Figure 2A), respectively. The 5-year PFS rates were 86% in the ORH group and 81.2% in the LRH group (Log-Rank P=0.143, Figure 2C), respectively. After IPTW adjustment, the OS (HR = 2.095, 95% CI: 1.233-3.562, P = 0.006, Adjust Log-Rank P= 0.001, Figure 2B) and PFS (HR=1.950, 95%CI: 1.194-3.184, P=0.008, Adjust Log-Rank P= 0.002, Figure 2D) rates were significantly higher in the ORH group compared with the LRH group.

Analysis of Survival Factors
Multivariate Cox analysis was further applied to identify the survival factors associated with the PFS and OS of FIGO 2018 stage IA2-IIA2 patients. As exhibited in Figure 4A, before adjustment, group, age, tumor size, LVSI, and parametrial invasion were proven to be associated with the survival condition of the cervical cancer patients. Then group, tumor size, LVSI, and parametrial invasion were identified to be connected with the recurrence of patients before adjustment ( Figure 4C). After IPTW adjustment, group, tumor size, LVSI were confirmed to be significantly associated PFS and OS of cervical cancer patients of FIGO stage IA2-IIA2 consistently ( Figures 4B, D). Besides, age and parametrial invasion also showed critically significant P-value.  undergoing LRH suffered from obviously poor 5-year OS (72.2% vs 78.6%) and 5-year PFS (69.6% vs 77%) than patients undergoing ORH, which was not statistically significant. In general, the survival outcomes in our study were in accordance with several population-based and high-volume institutional retrospective studies, confirming the opinion of LACC trial, which concluded that MIS increased recurrence and death for early cervical cancer patients, as the Table 2 in the research of Yang et al. has shown (35). At the same time, multivariate Cox analysis was also conducted to discover the prognostic factors associated with the recurrence and survival outcomes of cervical cancer with FIGO 2018 stage IA2-IIA2. Surgical approach, tumor size, LVSI, and parametrial invasion were found to be independent prognostic factors. The finding was essentially in agreement with the pathologic risk factors of cervical cancer. Besides, patients in the LRH group were proven to have almost two times of death and recurrence than the ORH group, which further supported the conclusion of the LACC Trial (7).
Overall, results of our study supported the indications of ORH for early-stage cervical cancer patients based on the FIGO 2018 staging system, which was in compliance with the NCCN guidelines Version 1.2021. However, in patients of FIGO 2018 stage IA2, IB1-IB2, and IIA1, LRH showed non-inferiority compared with ORH even after IPTW adjustment. This finding reminded us that the advantages of laparoscopic surgery in radical hysterectomy of cervical cancer could not be completely denied. Patients in this subgroup all had no "highrisk" prognostic factors, presenting negative nodes, negative margins, and negative parametria. According to the "intermediate-risk" Sedlis Criteria: 1) greater than 1/3 stromal invasion; 2) LVSI; or 3) cervical tumor diameters more than 4 cm (36,37), several patients showed deep stromal invasion and positive LVSI. So LRH might be applicable for some specific cervical cancer patients. In the LACC trial, it emphasized that the results cannot be generalized to patients with "low-risk" cervical cancer (tumor size < 2 cm; no lymphovascular invasion; depth of invasion < 10 mm; and no lymph-node involvement) (7). Many previous studies also identified that LRH was non-inferior to ORH for early cervical cancer (5,8,(28)(29)(30)(31)(32)(33), especially in patients with conization before surgery and no visible tumor on the final pathology (38). In view of the currently negative point to LRH, we supposed that LRH might be applicable for early-stage cervical cancer patients without high and intermediate risk factors, including negative nodes, negative margins, negative parametria, less than 1/3 stromal invasion, negative LVSI, and tumor size less than 4 cm. And the data from our center also supposed this standpoint. Furthermore, as some studies have shown that patients with tumor size <2 cm have better prognosis, the tumor size could be set as < 2 cm as selection criterion more carefully. On this basis, patients should receive conization before LRH during clinical work (18,(39)(40)(41)(42)(43). Besides, histology should also be taken into consideration (22,44,45). Patients of FIGO stage IIIC1-IIIC2 in the LRH group have obviously worse PFS and OS than the ORH group, but which was not statistically significant. The reasons may be as follows: first, number of patients in the subgroup was limited, especially in the LRH group; second, though lymph node metastasis is an important factor associated with the prognosis of cervical cancer patients, stratifying the clinical stage according to the lymph node status individually is still controversial (46). If the local extent of the disease between the groups was not comparable, outcomes of the surgery might also be affected. However, individualized chemoradiotherapy is recommended for the FIGO 2018 IIIC1p-IIIC2p in the NCCN guidelines Version 1.2021. So the rationality of the clinical staging is better worth discussing than the operation way.
Several causes were proposed to explain the high risk of recurrence and poor survival in patients undergoing LRH, including the establishment of pneumoperitoneum through CO 2 insufflation, application of uterine manipulators and the method of colpotomy. The continuously perfusing and flowing CO 2 in the abdominopelvic cavity could lead to spread of the detached tumor cells (47)(48)(49). The compression effects of the uterine manipulator on the upper vagina might increase the risk of tumor cell detachment, especially bringing about the distant dissemination and metastasis of intra-luminal tumors in patients with positive LVSI (50,51). The way of colpotomy was also considered to increase likely exposure of the tumor to the abdominal cavity at the end of the surgery (52,53). These three reasons might also explain the results of our research. Besides, several studies identified that the most effective way to reduce the recurrence rate during the LRH is to avoid tumor dissemination, especially during the vaginal colpotomy (54)(55)(56)(57). Kanao et al. have identified that the no-look no-touch technique may be useful to reduce recurrence risks through preventing intraoperative tumor spillage during LRH for early cervical cancer patients. The technique incorporates four specific measures: 1) creation of a vaginal cuff, 2) avoidance of a uterine manipulator, 3) minimal handling of the uterine cervix, and 4) bagging of the specimen (57).
Overall, the leading strength of this study was that it compared the efficacy of the ORH and LRH for the early cervical cancer patients comprising the new FIGO staging criterion. Then the study adopted the IPTW analysis to balance the prognostic risk factors between the two groups, even for the subgroup analysis. However, there are still several limitations of our study. In essence, this is a retrospectively single-center analysis. Moreover, variation of the surgeon's operative experience on the rates of OS and PFS was not explored. Besides, since the robotic surgery was not adopted in our institution, we did not include patients accepting robotic radically hysterectomy. A recent meta-analysis conducted by Shazly et al. identified that laparoscopy and robotic RH are equivalent in terms of perioperative outcomes (5,58). Furthermore, Gallotta et al. discovered that robotic RH and LRH had comparable perioperative, postoperative and survival outcomes for early cervical cancer patients through a large case matched control study. Therefore, robotic RH might also not play better roles than ORH for early stage cervical cancer. But to better identify the role of robotic RH, an ongoing prospective, international, multi-institutional, open-label randomized controlled Robot-assisted Approach to Cervical Cancer (RACC) trial is performed (15). It is worth mentioning that manipulator is prohibited, and patients with tumor size more than 4 cm were excluded in that study. So the research results were very worth expecting. Of course, none of these concerns alter the results of our primary outcome of interest.
In conclusion, this study confirms the superiority of open surgery on overall and progression free survival for early cervical cancer patients, no matter under FIGO 2009 or FIGO 2018 staging system. However, in subset of FIGO 2018 IA2, IB1-IB2, and IIA1, laparoscopic surgery showed non-inferiority, especially in patients with no prognostic risks. Considering the advantages and popularity of the minimally invasive surgery, laparoscopic radically hysterectomy should not be completely prohibited in early cervical cancer patients. In brief, ORH was recommended for early stage cervical cancer patients under 2018 FIGO staging system. But LRH might be suitable for early-stage cervical cancer patients without high and intermediate risk factors, including negative nodes, negative margins, negative parametria, less than 1/3 stromal invasion, negative LVSI and tumor size less than 2 cm.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

AUTHOR CONTRIBUTIONS
QY and FB designed and edited this study. YX and WZ enrolled and followed up the patients. WCZ analyzed the data and wrote the manuscript. All authors contributed to the article and approved the submitted version.