Journal of Personalized Medicine Article Fertility-Sparing Surgery versus Radical Hysterectomy in Early Cervical Cancer: A Propensity Score Matching Analysis and Noninferiority Study Antoni Llueca 1,2, * , Maria Victoria Ibañez 3 , Aureli Torne 4 , Antonio Gil-Moreno 5 , Angel Martin-Jimenez 6 , Berta Diaz-Feijoo 4 , Anna Serra 1,2 , Maria Teresa Climent 1,2 , Blanca Gil-Ibañez 7 and on behalf of the Spain-GOG Cervical Cancer Working Group † 1 Multidisciplinary Unit of Abdominopelvic Oncology Surgery (MUAPOS), Department of Obstetrics and Gynaecology, University General Hospital of Castellon, 12004 Castellón, Spain; serraa@uji.es (A.S.); maclimen@uji.es (M.T.C.) 2 Department of Medicine, University Jaume I (UJI), 12004 Castelló de la Plana, Spain 3 Department of Mathematics, University Jaume I (UJI), 12004 Castelló de la Plana, Spain; mibanez@uji.es 4 Unit of Gynaecologic Oncology (ICGON), Endocrinology, Gynaecology and Human Reproduction (IDIBAPS), Hospital Clinic of Barcelona, 08007 Barcelona, Spain; atorne@clinic.cat (A.T.); bdiazfe@clinic.cat (B.D.-F.) 5 Gynaecological Oncology Department, Hospital Universitari Vall d’Hebron, Universitat Autonoma de Barcelona, Centro de Investigación Biomedica en Red de Cancer, 08035 Barcelona, Spain; antonioimma@yahoo.es 6 Hospital Universitario Son Llatzer, 07198 Mallorca, Spain; amartin@hsll.es 7 Gynaecological Oncology and Endoscopy Unit, Department of Obstetrics and Gynaecology, Citation: Llueca, A.; Ibañez, M.V.; University Hospital 12 de Octubre, Research Institute i+12, 28041 Madrid, Spain; blancalabacin@hotmail.com Torne, A.; Gil-Moreno, A.; * Correspondence: antonillueca@gmail.com; Tel.: +34-964726500 Martin-Jimenez, A.; Diaz-Feijoo, B.; † Members of Spain-GOG cervical cancer working group were listed in the acknowledgement. Serra, A.; Climent, M.T.; Gil-Ibañez, B.; on behalf of the Abstract: Objective: Fertility-sparing surgery (FSS) is the treatment of choice for patients with early Spain-GOG Cervical Cancer Working cervical cancer (ECC) and fertility desire, but survival rates compared to radical hysterectomy (RH) Group. Fertility-Sparing Surgery have been scarcely reported. The aim of this study was to analyse the oncological outcomes of versus Radical Hysterectomy in Early FSS compared to a balanced group of standard RH. Methods: A retrospective multicentre study of Cervical Cancer: A Propensity Score ECC patients who underwent FSS or RH was carried out in 12 tertiary hospitals in Spain between Matching Analysis and January 2005 and January 2019. The experimental group included patients who underwent a simple Noninferiority Study. J. Pers. Med. and radical trachelectomy, and the control group included patients who underwent RH. Optimal 2022, 12, 1081. https://doi.org/ 1:1 propensity score (PS) matching analysis was performed to balance the series. Results: The study 10.3390/jpm12071081 included 222 patients with ECC; 111 (50%) were treated with FSS, and 111 (50%) were treated with Academic Editor: Kuan-Hao Tsui RH. After PS matching, a total of 38 patients in the FSS group and 38 patients in the RH group were Received: 13 June 2022 analysed. In both groups, the overall survival (HR 2.5; CI 0.89, 7.41) and recurrence rates (28.9% in Accepted: 28 June 2022 the FSS group vs. 13.2% in RH group) were similar. The rate of disease-free survival at 5 years was Published: 30 June 2022 68.99% in the FSS group and 88.01% in the RH group (difference of −19.02 percentage points; 95% CI −32.08 to −5.96 for noninferiority). In the univariate analysis, only tumour size reached statistical Publisher’s Note: MDPI stays neutral significance. Conclusion: FSS offers excellent disease-free and overall survival in women with ECC with regard to jurisdictional claims in with fertility desire and is not inferior compared to RH. published maps and institutional affil- iations. Keywords: trachelectomy; early cervical cancer; fertility preservation treatment; minimally invasive surgery; radical hysterectomy; fertility-sparing surgery Copyright: © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article 1. Introduction distributed under the terms and In Spain, both the incidence and the mortality of cervical cancer are lower than conditions of the Creative Commons in neighbouring countries, at 6.8 and 2.8 cases per 100,000 women per year. Current Attribution (CC BY) license (https:// surgical treatment of cervical cancer is limited to the early stages of the disease, and radical creativecommons.org/licenses/by/ hysterectomy is the gold standard for cervical cancer treatment in these stages [1–3]. 4.0/). J. Pers. Med. 2022, 12, 1081. https://doi.org/10.3390/jpm12071081 https://www.mdpi.com/journal/jpm J. Pers. Med. 2022, 12, 1081 2 of 14 This technique provides an adequate curative treatment of the disease but in turn has a negative impact on future fertility in young patients. Novak, in 1952, and Aburel, in 1957, described a laparotomic radical trachelectomy technique to preserve the fertility of young patients with early cervical cancer. At that time, the technical difficulty of performing this technique made it unpopular [4,5]. The trachelectomy technique became more popular in the early 2000s, as described by Dargent [6], who revived the radical trachelectomy technique but used a vaginal approach with minimal invasiveness for lymphadenectomy. Different surgical approaches to early-stage uterine cancer can produce different re- sults in terms of disease-free and overall survival, as already demonstrated in the LACC (Laparoscopic Approach to Cervical Cancer) trial [7]. Although this multicentre randomised trial was focused on radical hysterectomy, it also raises a series of questions about the extrapolation of these results to any type of surgical intervention or approach for the treatment of cervical cancer. The IRTA (International Radical Trachelectomy Assessment) study [8] concluded that, for small tumours, the results and preservation of fertility are com- parable regardless of the approach. Until now, in the absence of prospective randomised studies, the recommendations regarding fertility-sparing surgeries have been based on observational studies and the decisions of tumour committees where the experience of the surgeon has prevailed. The primary objective of this study was a balanced comparison between the classical radical treatment technique and fertility-preserving surgery (FSS) in the early stages of cervical cancer in terms of oncological outcomes. A secondary objective was to evaluate the hypothesis that FSS is not inferior to radical hysterectomy with respect to the percentage of patients who are disease-free at 5 and 10 years after surgery. 2. Materials and Methods 2.1. Study Design A multicentre, retrospective cohort study of early cervical cancer patients who under- went FSS or radical hysterectomy was carried out in 12 tertiary hospitals in Spain between January 2005 and January 2019. Eligible patients had squamous carcinoma, adenocarci- noma, or adenosquamous carcinoma, stages IA1 with lymphovascular space invasion, IA2, and IB1 (<2 cm and 2–4 cm) (FIGO 2009) [9], and no evidence of lymph node involvement and/or other metastasis in the preoperative evaluation. Exclusion criteria included aggres- sive histologies (different from the ones described above), and where intraoperative frozen section analysis confirmed positive lymph nodes or peritoneal spread. 2.2. Methods FSS included simple trachelectomy and radical trachelectomy. The surgical approach (laparotomy, laparoscopy, robot-assisted laparoscopy, or vaginal hysterectomy), level of radicality, and the decision of whether to include sentinel lymph node mapping were based on surgeon and centre protocols. Different types of radical hysterectomy were performed in patients with no desire to preserve fertility using different surgical approaches [1,2]. Surgical outcomes were measured on the basis of surgical reports. Intraoperative complications were recorded. Postoperative complications were considered during the first 3 months following the Clavien–Dindo classification [10]. The use of adjuvant chemotherapy and/or radiation was at the discretion of the tumour boards of the different centres. Progression-free survival (PFS) was defined as the time from surgery to diagnosis of local recurrence or metastasis. Overall survival (OS) was defined as the time from surgery to the date of death or last follow-up. The appearance of cervical, paracervical, or vaginal lesions was considered local recurrence. Events involving the adnexa were considered pelvic recurrences; when the disease affected the lymph nodes, it was considered positive lymph node disease, and, when recurrences appeared in J. Pers. Med. 2022, 12, 1081 3 of 14 distant organs, they were considered metastases. This study was approved by the Clinical Research Ethics Committee of the Hospital Clinic of Barcelona (study protocol 87/2019) as the reference centre and by the Institutional Review Boards of the participating hospitals. 2.3. Statistical Analysis The variables were summarised according to their nature using means and standard deviations or frequencies and percentages. Similarly, univariate analysis was performed to analyse possible differences in the quantitative covariates of the patients treated with both surgeries. Parametric (t-test) or nonparametric tests (Mann–Whitney test) were applied depending on whether the analysed variable followed or did not follow a Gaussian distri- bution. Chi-square or Fisher’s exact tests were used for the qualitative variables. Given the low number of deaths, we did not focus our analysis on overall survival but on disease-free time, considering recurrence as the event to analyse. From now on, when talking about survival, we mean disease-free survival. Optimal 1:1 propensity score (PS) matching [11] analysis was performed to balance the series of patients who underwent radical hysterectomy or FSS, making them comparable regarding the set of covariates of interest in the study. The analysis was performed to obtain a subsample of patients verifying that covariates were balanced across both surgical groups. This score matching method is optimal in the sense that the sum of the absolute pairwise distances in the matched sample is as small as possible and provides the best match among the methods tested. Neither the optimal propensity score method nor any of the other matching methods tested preserved all the patients with recurrences in the balanced subsample. This is a problem for the posterior survival analysis, where it is important to have as many recurrence cases as possible. As a first step of PS matching, we found those patients who underwent radical hysterectomy that best matched the patients who underwent preservation surgery and suffered a recurrence. In the second step, we obtained the patients who underwent preservation surgery that best matched the patients who underwent radical hysterectomy and suffered a recurrence. Then, a third optimal 1:1 PS matching analysis was performed with the patients who were unmatched in the two previous steps. For this purpose, radical hysterectomy was considered the control arm, and FSS was considered the experimental arm. Finally, the three subsets were merged into a single matched sample. Absolute standardised mean differences <0.10 were desired in all comparisons. A noninferiority study was conducted to compare the oncological outcomes of both surgical treatments. Kaplan–Meier survival curves were compared, and a univariate Cox model was used to estimate the hazard ratios (HRs) and their 95% confidence intervals for the effect of the different surgeries on disease-free survival. In this case, the noninferiority of FSS with respect to hysterectomy was concluded using Wald’s confidence interval using the estimated confidence limits of the HR. If the upper confidence limit of the HR was less than the noninferiority margin (1.1), then noninferiority could be assessed. The assumption of proportional hazards was tested and assessed for all analyses that involved hazard ratios [12]. To check if any of the factors considered in the study were risk factors for the disease- free survival time of a patient, univariate and multivariate Cox regression models were fitted. To compare the effect of the factors considered in the study on the disease-free survival time in both surgical groups, HRs and their 95% CIs were calculated. To analyse the noninferiority of FSS with respect to hysterectomy in disease-free sur- vival rates at 2.5 and 5 years of follow-up, predictions based on Kaplan–Meier curves and confidence intervals for the difference in survival rates were calculated. In previous studies involving other types of cancer, noninferiority margins of 6–8 percentage points (absolute value) were considered to be clinically acceptable [13–15]. As described before [7], we considered a margin of 7.2 percentage points. Therefore, if the upper confidence limit of the confidence interval of the difference in the disease recurrence rates was less than the noninfe- riority margin (−7.2), then noninferiority could be rejected. Therefore, the noninferiority of J. Pers. Med. 2022, 12, 1081 4 of 14 FSS with respect to hysterectomy was assessed if the lower boundary of the 95% confidence interval of the difference in survival rates was greater than −7.2 percentage points. Statistical analysis was performed using R (R Core Team, 2021, Vienna, Austria. Available at: https://www.R-project.org/, accessed on 1 May 2022) [16]. The “MatchIt” [17] and “cobalt” [18] packages were used to perform and visualise the results of the PS matching analysis, and the “survival” [19], “EquiSurv” [20] and “survminer” [21] packages were used for the survival analysis. Unless otherwise stated, all analyses were performed with a two-sided significance level of 0.05. 3. Results The study included 222 patients with early cervical cancer; 111 (50%) were treated with radical hysterectomy, and 111 (50%) were treated with fertility-sparing surgery. 3.1. Unmatched Series The clinicopathological characteristics of the total unmatched series are shown in Table 1. Table 1. Clinicopathological characteristics of the unmatched series. Fertility-Sparing Radical Total Surgery Hysterectomy p Value n = 222 n = 111 n = 111 Age 39.5 ± 10.6 years 46.2 ± 10 years 32.7 ± 5.3 years 0.001 Follow up 62.8 ± 45.1 months 65.9 ± 46.7 months 59.6 ± 46.4 months 0.299 FIGO Classification (2009) IA1 plus LVI+ 7 5 (4.5%) 2 (1.8%) IA2 25 15 (13.5%) 10 (9%) 0.294 IB1 < 2 cm 133 67 (60.4%) 66 (59.4%) IB1 ≥ 2 cm 57 24 (21.6%) 33 (29.7%) Histology Adenocarcinoma 84 (37.8%) 45 (40.5%) 39 (35.1%) 0.4469 Squamous carcinoma 137 (61.7%) 66 (59.5%) 71 (64%) Adenosquamous 1 (0.5%) 0 1 (0.9%) carcinoma LVI Positive 30 (13.5%) 16 (14.4%) 14 (12.6%) 0.8444 Negative 192 (86.5%) 95 (85.6%) 97 (87.4%) Tumour size Less than 2 cm 155 (69.8%) 81 (73.0%) 74 (66.7%) 0.3803 Greater or equal than 2 cm 67 (30.2%) 30 (27.0%) 37 (33.3%) Nodal assessment PLD + SLN 165 80 (72.1%) 85 (76.6%) 0.6971 Only SLN 55 29 (26.1%) 26 (23.4%) Missing values 2 2 (1.8%) 0 Intraoperative complications 0.48 Yes 9 (4.1%) 6 (5.4%) 3 (2.7%) No 213 (95.9%) 105 (94.6%) 108 (97.3%) Recurrence localizations Distance 2 (12.5%) 0 (0.0%) 2 (40.0%) Adnexal 2 (12.5%) 2 (18.2%) 0 (0.0%) Cervical 3 (18.8%) 3 (27.3%) 0 (0.0%) 0.1532 Lymph nodes 3 (18.8%) 2 (18.2%) 1 (20.0%) Local 6 (37.5%) 4 (36.4%) 2 (40.0%) Surgical approach Laparoscopy/Robotic 109 (49.1%) 39 (35.1%) 70 (63.1%) <0.001 Laparotomy 4 (1.8%) 0 (0.0%) 4 (3.6%) Vaginal 109 (49.1%) 72 (64.9%) 37 (33.3%) J. Pers. Med. 2022, 12, 1081 5 of 14 Table 1. Cont. Fertility-Sparing Radical Total Surgery Hysterectomy p Value n = 222 n = 111 n = 111 Postoperative complications 0.001 Yes 7 0 (0%) 7 (6.3%) No 215 111 (100%) 104 (93.7%) Recurrence Yes 16 (7.2%) 11 (9.9%) 5 (4.5%) 0.1944 No 205 (92.8%) 100 (90.1%) 106 (95.5%) PLD: pelvic lymphadenectomy. SLN: sentinel lymph node. LVI: lymphovascular space invasion. Regarding the surgical approach, the majority of FSSs were performed vaginally (64.9%), and the minimally invasive approach was predominant in radical hysterectomies (63.1%) (p < 0.001). During FSS, intrasurgical complications occurred in 5.4% of the pa- tients (five bladder injuries and a uterine artery tear) (six patients vs. three patients in conventional surgery p = 0.48). Seven and zero patients suffered major postoperative complications (Clavien–Dindo grade III–IV) after conventional surgery and FSS, respec- tively (p = 0.001). In the analysis of recurrences, 16 patients had recurrence of their disease: 11 patients after fertility-preserving surgery (recurrence rate 9.9%) and five patients after conventional surgery (recurrence rate 4.5%). Disease-free survival in patients after FSS was 15 months (95% CI 8.5–21.5) vs. 19 months for conventional surgery (95% CI 3.9–34), and the differences were not statistically significant. The global mean follow-up time was 62.8 months, with a standard deviation (SD) of 45, while it was 59.6 (SD 43.3) and 66 (SD 46.6) months for radical hysterectomy and FSS, respectively. No significant differences were observed. 3.2. Recurrence Patients The mean time until recurrence of the patients who suffered a recurrence was 23.10 (SD 20.38) months. The minimum recurrence time was 1.63 months, and the maximum time was 64.7 months. The mean time to recurrence for patients with recurrence after hysterectomy was 26.75 (SD 24.54) months, whereas, for patients with recurrence after trachelectomy, it was 21.45 months (SD 19.29) (p = 0.6837). The clinicopathological characteristics of the patients with recurrence are described in Table 2. Table 2. Clinicopathological characteristics of the patients with recurrence. Patients with Recurrence Patients with Fertility-Sparing Surgery Radical Hysterectomy Recurrence p Value p Value n = 11 n=5 (Recurrence/Total) Age 38.2 ± 12.3 years 31.18 ± 5.7 years 53.6 ± 7.5 years 0.0008 Follow-up 66.6 ± 46.8 months 77.94 ± 51.6 months 41.8 ± 21.2 months 0.067 Time to recurrence 23.11 ± 20.38 months 21.45 ± 19.29 months 26.75 ± 24.54 months 0.68 FIGO Classification (2009) IA1 y LVI + 0/7 0 0 IA2 1/25 (4%) 0.0343 1/11 (9.1%) 0 0.294 IB1 < 2 cm 6/133 (4.5%) 5/11 (45.5%) 1/5 (20.0%) IB1 ≥ 2 cm 9/57 (15.8%) 5/11 (45.5%) 4/5 (80.0%) Histology Adenocarcinoma 8/84 (9.5%) 6/11 (54.5%) 2/5 (40.0%) 0.5669 1 Squamous carcinoma 8/137 (5.8%) 5/11 (45.5%) 3/5 (60.0%) Adenosquamous carcinoma 0/1 0 0 J. Pers. Med. 2022, 12, 1081 6 of 14 Table 2. Cont. Patients with Recurrence Patients with Fertility-Sparing Surgery Radical Hysterectomy Recurrence p Value p Value n = 11 n=5 (Recurrence/Total) LVI 0.3098 1 Positive 4/30 (13.3%) 3/11 (27.3%) 1/5 (20.0%) Negative 12/192 (6.2%) 8/11 (72.7%) 4/5 (80.0%) Tumour size Smaller than 2 cm 6/155 (3.9%) 0.0083 5/11 (45.5%) 1/5 (20.0%) 0.3803 Larger or equal than 2 cm 10/67 (14.9%) 6/11 (54.5%) 4/5 (80.0%) Intraoperative complications Yes 0/9 0 0 No 16/212 (8.3%) 11/11 (100%) 5/5 (100%) Recurrence localizations Distance 2 0 (0.0%) 2 (40.0%) Adnexal 2 2 (18.2%) 0 (0.0%) Cervical 3 3 (27.3%) 0 (0.0%) 0.1532 Lymph nodes 3 2 (18.2%) 1 (20.0%) Local 6 4 (36.4%) 2 (40.0%) Surgical approach Laparoscopy/Robotic 10/109 (9.2%) 6 (54.5%) 4 (80.0%) 0.5879 Laparotomy 0 0 (0.0%) 0 (0.0%) Vaginal 6/109 (5.5%) 5 (45.5%) 1 (20.0%) Postoperative complications Yes 0 0 0 No 16/206 11(100%) 5(100%) LVI: lymphovascular space invasion. Regarding tumour size, the percentage of patients with recurrence in lesions larger than 2 cm was significantly higher than that of recurrence in lesions smaller than 2 cm (p = 0.0083), which was also demonstrated in the FIGO classification, with a higher percent- age of recurrences in patients with FIGO 2009 stage IB1 >2 cm (p = 0.0343). Of the 109 operations performed laparoscopically, 10 (9.2% of the total) had recurrences compared to none of the laparotomic operations, and six (5.5%) of the operations were performed vaginally. Among the patients who suffered recurrence, there was no significant relationship between the approach route and the type of intervention (p = 0.5834). 3.3. Matched Series As seen in Table 1, one of the main differences between the control and treatment groups was age. Figure 1 shows the distribution of the variable age at diagnosis in both treatment groups. At ages greater than 40 years, hysterectomy was predominant over trachelectomy, and from the age of 45 in the sample, we found only one patient treated with trachelectomy (55 years, 0.9% of the sample), compared to 63 patients treated with hysterectomy (56.76% of the sample). As seen in Table 2, this difference also remained between the patients with recurrence. As this variable was not balanced in the two surgical groups, to reduce the bias due to this possible confounding variable, we proposed a subsample with homogeneous groups before moving onto the survival study. When balancing by age, we would lose a large sample size, since we would be left mainly with women under 45 years of age, of which we had 110 in the trachelectomy group but only 48 in the hysterectomy group. Thus, looking for an age-balanced sample with the same sample size in the two surgical groups, we would be left with a maximum of 48 + 48 = 96 patients, which could now be unbalanced according to the rest of the covariates. Now, although we only found differences in both groups in the variables age and approach route (Table 1), we also considered the covariates FIGO classification 2009, histol- ogy, and tumour size in the propensity score to ensure that the resulting sample was also 3.3. Matched Series As seen in Table 1, one of the main differences between the control and treatment groups was age. Figure 1 shows the distribution of the variable age at diagnosis in both J. Pers. Med. 2022, 12, 1081 treatment groups. At ages greater than 40 years, hysterectomy was predominant over7 tra- of 14 chelectomy, and from the age of 45 in the sample, we found only one patient treated with trachelectomy (55 years, 0.9% of the sample), compared to 63 patients treated with hyster- ectomy (56.76% of the sample). As seen in Table 2, this difference also remained between balanced with respect to these variables. After the propensity score, a balanced sample of the patients with recurrence. As this variable was not balanced in the two surgical groups, 76 patients was obtained. Half of them were treated with the classical radical treatment to reduce the bias due to this possible confounding variable, we proposed a subsample technique, and the other 38 patients were treated with fertility-preserving surgery. A Love with homogeneous groups before moving onto the survival study. When balancing by plot, as shown in Figure 2, was a clean way to visually summarise balance, showing the age, we would lose a large sample size, since we would be left mainly with women under mean difference of the distribution of each variable in both surgical groups on the initial 45 years of age, of which we had 110 in the trachelectomy group but only 48 in the hyster- dataset (all) and the matched subsample (matched). As shown in Table 1, this plot again ectomy group. Thus, looking for an age-balanced sample with the same sample size in the showed that balance was quite poor prior to matching the variables age, surgical approach, two surgical groups, we would be left with a maximum of 48 + 48 = 96 patients, which and Figo Classification 2009, but the matching improved the balance across all covariates, could now be unbalanced according to the rest of the covariates. as most were within a threshold of 0.1. J. Pers. Med. 2022, 12, x FOR PEER REVIEW 8 of 16 initial dataset (all) and the matched subsample (matched). As shown in Table 1, this plot again showed that balance was quite poor prior to matching the variables age, surgical approach, and Figo Classification 2009, but the matching improved the balance across all covariates, Figure 1. Figure as most were Distribution 1. Distribution of thewithin of the a age variable variable threshold age at of 0.1. at diagnosis diagnosis in both in both treatment treatment groups. groups. Now, although we only found differences in both groups in the variables age and approach route (Table 1), we also considered the covariates FIGO classification 2009, his- tology, and tumour size in the propensity score to ensure that the resulting sample was also balanced with respect to these variables. After the propensity score, a balanced sam- ple of 76 patients was obtained. Half of them were treated with the classical radical treat- ment technique, and the other 38 patients were treated with fertility-preserving surgery. A Love plot, as shown in Figure 2, was a clean way to visually summarise balance, show- ing the mean difference of the distribution of each variable in both surgical groups on the Loveplot Figure2.2.Love Figure plotto tovisualise visualisethe thegoodness goodnessof ofthe thematch. match. For a clearest overview of the matched subsample, the clinicopathological character- istics of the included patients are described in Table 3. Table 3. Clinicopathological characteristics of the matched series. J. Pers. Med. 2022, 12, 1081 8 of 14 For a clearest overview of the matched subsample, the clinicopathological characteris- tics of the included patients are described in Table 3. Table 3. Clinicopathological characteristics of the matched series. Matched Subsample Fertility-Sparing Surgery Radical Hysterectomy p-Value n = 76 n = 38 n = 38 Age 35.9 ± 7.1 years 34.2 ± 5.5 years 37.7 ± 8.1 years 0.189 Follow-up 65.7 ± 43.1 months 61.3 ± 40.5 months 70.1 ± 45.6 months 0.3749 FIGO Classification (2009) 0 0 0 0.5435 IA1 and LVI+ IA2 9 (11.8%) 3 (7.9%) 6 (15.8%) IB1 < 2 cm 41 (53.9%) 22 (57.9%) 19 (50.0%) IB1 > 2 cm 26 (34.2%) 13 (34.2%) 13 (34.2%) Histology Adenocarcinoma 28 (36.8%) 14 (36.8%) 14 (36.8%) 1 Squamous carcinoma 48 (63.2%) 24 (63.2%) 24 (63.2%) Adenosquamous 0 0 0 carcinoma LVI Positive 14 (18.4%) 8 (21.1%) 6 (15.8%) 0.7673 Negative 62 (81.6%) 30 (78.9%) 32 (84.2%) Tumour size Smaller than 2 cm 45 (59.2%) 22 (57.9%) 23 (60.5%) 1 Larger than 2 cm 31 (40.8%) 16 (42.1%) 15 (39.5%) Intraoperative complications Yes 4 (5.3%) 3 (7.9%) 1 (2.6%) 0.6075 No 72 (94.7%) 35 (92.1%) 37 (97.4%) Recurrence localizations Distance 2 (12.5%) 0 (0.0%) 2 (40.0%) 0.1532 Adnexal 2 (12.5%) 2 (18.2%) 0 (0.0%) Cervical 3 (18.8%) 3 (27.3%) 0 (0.0%) Lymph nodes 3 (18.8%) 2 (18.2%) 1 (20.0%) Local 6 (37.5%) 4 (36.4%) 2 (40.0%) Surgical approach Laparoscopy/robotic 48 (63.2%) 23 (60.5%) 25 (65.8%) 0.812 Laparotomy 0 0 0 Vaginal 28 (36.8%) 15 (39.5%) 13 (34.2%) Recurrence Yes 16 (21.1%) 11 (28.9%) 5 (13.2%) 0.1595 No 60 (78.9%) 27 (71.1%) 33 (86.8%) All patients were negative for intraoperative nodes and, therefore, underwent cer- vical surgery. In one patient with a tumour size of 2–4 cm and FSS, micrometastasis was diagnosed at the time of ultrastaging. She underwent a hysterectomy and received adjuvant therapy. 3.4. Survival Analysis Continuing with the matched sample, Figure 3 shows the Kaplan–Meier curves to compare the probability of disease-free survival of both surgical procedures. We compared the null hypothesis that stated that there was no difference between the two survival functions with the alternative that there was (logrank test), obtaining a p-value J. Pers. Med. 2022, 12, 1081 9 of 14 J. Pers. Med. 2022, 12, x FOR PEER REVIEW 10 of 16 of 0.07. Therefore, strictly speaking, we did not detect significant differences between the functions of disease-free survival of radical hysterectomies and FSS. Figure 3. Kaplan–Meier disease-free survival curves. Tick marks indicate censored data. The vertical Figure lines 3. Kaplan–Meier indicate disease-free the upper limits of thesurvival curves. one-sided Tick CI for marks indicate noninferiority at censored 95%. data. The vertical lines indicate the upper limits of the one-sided CI for noninferiority at 95%. Additionally, to investigate the association between the disease-free survival time of We compared the null hypothesis that stated that there was no difference between patients and the two surgical procedures, Cox proportional hazards models were performed, the two survival functions with the alternative that there was (logrank test), obtaining a obtaining an HR associated with fertility-sparing surgery against hysterectomy of 2.5 (CI p-value of 0.07. Therefore, strictly speaking, we did not detect significant differences be- 0.89, 7.41). With this confidence interval, noninferiority of FSS could not be assessed. tween the functions of disease-free survival of radical hysterectomies and FSS. To check if, in addition to the surgical procedure, any of the factors considered in Additionally, to investigate the association between the disease-free survival time of the study were risk factors for the disease-free survival time of a patient, univariate and patients and the two surgical procedures, Cox proportional hazards models were per- multivariate Cox regression models were conducted. Just one of the factors considered in formed, obtaining an HR associated with fertility-sparing surgery against hysterectomy the study was significantly associated with survival in a univariate analysis, and none of of 2.5 (CI 0.89, 7.41). With this confidence interval, noninferiority of FSS could not be as- them reached statistical significance in the multivariate analysis (Table 4). Due to the clear sessed. collinearity among To check if, inthe covariates addition to the“tumour size” and “Figo surgical procedure, any ofClassification”, this lastin the factors considered variable the was not included in the Cox regression model. study were risk factors for the disease-free survival time of a patient, univariate and mul- tivariate Cox regression models were conducted. Just one of the factors considered in the Table study Factors 4. was associatedassociated significantly with disease-free survivalin(balanced with survival data). a univariate analysis, and none of them reached statistical significance in theUnivariate multivariate analysis (Table 4).Multivariate Analysis Due to theAnalysis clear col- linearity among the covariates “tumour size” and “Figo Classification”, this last variable Hazard Ratio Hazard Ratio was not included Factor in the Cox regression model. (95% CI) p-Value (95% CI) p-Value Age 1.06 (0.98–1.13) 0.101 1.05 (0.99–1.12) 0.06 Table 4. Factors associated with disease-free survival (balanced data). Tumour size larger than 2 cm vs. 2.091 (1.02–4.28) 0.04 1.99 (0.94–4.19) tumour size smaller than 2 cm Univariate Analysis Multivariate Analysis0.07 Squamous carcinoma vs. Hazard Ratio Factor adenocarcinoma Hazard Ratio (95% CI) p-Value 0.59 (0.22–1.57) 0.288 0.59 (0.20–1.76) p-Value 0.34 (95% CI) Vaginal surgical Age approach vs. 1.06 (0.98–1.13) 0.101 0.82 (0.29–2.27) 0.702 1.05 (0.99–1.12) 0.91 (0.31–2.66) 0.060.87 laparoscopy/robotic Tumour size larger LVI-positive than LVI-negative 2.0912.101 2 cm vs.vs.tumour (1.02–4.28) (0.67–6.54) 0.040.2 1.99 (0.94–4.19) 1.83 (0.47–7.01) 0.070.37 J. Pers. Med. 2022, 12, 1081 10 of 14 To compare the effect of the factors considered in this study on the disease-free survival time in both surgical groups, HRs and their 95% CIs can be seen in Table 5. Table 5. Association between different factors and the disease-free survival time in each surgical group expressed in terms of hazard ratios and their 95% CIs. Fertility-Sparing HR (Fertility-Sparing Hysterectomy Surgery Surgery vs. n = 38 n = 38 Hysterectomy) (95% CI) Matched subsample: 11/38 (28.9%) 5/38 (13.2%) 2.5 (0.89; 7.41) Tumour size: smaller 5/22 (22.72%) 1/23 (4.35%) 5.90 (0.69; 50.63) than 2 cm Tumour size: greater 6/16 (37.5%) 4/15 (26.67%) 1.71 (0.48; 6.11) than 2 cm Histology: 6/14 (42.86%) 2/14 (14.29%) 3.84 (0.77; 19.20) adenocarcinoma Histology: squamous 5/24 (20.83%) 3/24 (12.5%) 1.87 (0.44; 7.85) carcinoma Surgical approach: 6/23 (26.09%) 4/25 (16%) 1.85 (0.52; 6.61) laparoscopic/robotic Surgical approach: 5/15 (33.33%) 1/13 (7.69%) 6.33 (0.73; 54.99) vaginal Figo Classification 5/22 (22.73%) 1/19 (5.26%) 5.35 (0.62; 45.99) (2009): IB1 < 2 cm Figo Classification 5/13 (38.46%) 4/13 (30.77%) 1.50 (0.40; 5.67) (2009): IB1 > 2 cm LVI: Negative 8/30 (26.67%) 4/32 (12.5%) 2.55 (0.77; 8.53) LVI: Positive 3/8 (37.5%) 1/6 (16.67%) 2.74 (0.28; 26.68) * The second and third columns indicate the number of recurrences in each case divided by the total number of cases in the group. In none of these cases could the noninferiority of FSS over hysterectomy be concluded. Table 6 evaluates the differences in disease-free survival rates at 2.5 and 5 years af- ter surgery. The rate of disease-free survival at 5 years was 68.99% for fertility-sparing surgery and 88.01% for radical hysterectomy (difference −19.02 percentage points; 95% CI, −32.08 to −5.96 for noninferiority). The lower margin of this interval included the noninfe- riority margin of −7.2 percentage points; thus, noninferiority could not be confirmed. To affirm noninferiority, the entire confidence interval should be above −7.2 percentage points. Figure 4 shows the CI obtained for the difference in disease-free recurrence rates at 5 years. Table 6. Differences in disease-free survival rates at 2.5 and 5 years after surgery. Disease-Free Recurrence Rate Years from Surgery Difference (95% CI) (Rate (95% CI)) Fertility-Sparing Hysterectomy Surgery 2.5 years 77.46 (64.79–92.60) 91.40 (82.53–100) −13.94 (−24.84, −3.03) 5 years 68.99 (54.22–87.77) 88.01 (77.59–99.83) −19.02 (−32.08, −5.96) Table 6. Differences in disease-free survival rates at 2.5 and 5 years after surgery. Years from Disease-Free Recurrence Rate Difference (95% CI) Surgery (Rate (95% CI)) Fertility-sparing Surgery Hysterectomy J. Pers. Med. 2022, 12, 1081 2.5 years 77.46 (64.79–92.60) 91.40 (82.53–100) 11 of 14 −13.94 (−24.84, −3.03) 5 years 68.99 (54.22–87.77) 88.01 (77.59–99.83) −19.02 (−32.08, −5.96) Figure 4. Figure 4. Noninferiority Noninferiority analysis between FSS analysis between FSS and and hysterectomy hysterectomy at at 55 years. years. 4. Discussion Principal Principal Findings Findings In In this this multicentre multicentreretrospective retrospectivestudy, study, wewefound foundnonodifference in terms difference of disease-free in terms of disease- rates free rates at 5 years after surgery between FSS and radical hysterectomy, evenadjusting at 5 years after surgery between FSS and radical hysterectomy, even after after ad- for potential justing confounding for potential variables confounding due to unbalanced variables groups (p due to unbalanced = 0.07). groups (pIn= the study 0.07). of In the the noninferiority study of FSS (experimental of the noninferiority group) versus of FSS (experimental hysterectomy group) (control group), versus hysterectomy we (control could not affirm that FSS is inferior to radical hysterectomy in the disease-free group), we could not affirm that FSS is inferior to radical hysterectomy in the disease-free survival of early cervical cancer; however, as we can see in Figure 4, FSS could be survival of early cervical cancer; however, as we can see in Figure 4, FSS could be up to up to 23 percentage points lower than 23 percentage thelower points classicthan treatment. the classic treatment. In 2018, Tseng et al. [22] evaluated In 2018, Tseng et al. [22] evaluated 2717 patients 2717 patientswith withcervical cancer cervical in FIGO cancer 20092009 in FIGO 1B1 stages from the SEER database (Surveillance, Epidemiology, and End Results). All patients 1B1 stages from the SEER database (Surveillance, Epidemiology, and End Results). All underwent lymphadenectomy, but only 125 women had preserved fertility and underwent patients underwent lymphadenectomy, but only 125 women had preserved fertility and conisation/trachelectomy, while the remaining 2592 women underwent hysterectomy. underwent conisation/trachelectomy, while the remaining 2592 women underwent hys- There were no differences in disease-specific survival at 10 years between the two cohorts. terectomy. There were no differences in disease-specific survival at 10 years between the This is consistent with our results. two cohorts. This is consistent with our results. Despite the rising trend of delaying pregnancy in a person’s early 30s in Spain, fertility Despite the rising trend of delaying pregnancy in a person’s early 30s in Spain, fertil- preservation surgery is not very often offered after a diagnosis of early cervical cancer [23]. ity preservation surgery is not very often offered after a diagnosis of early cervical cancer Although fertility-sparing surgery for cervical cancer should exclusively be undertaken [23]. in specific gynaecologic oncology centres with highly trained surgeons, every young Although fertility-sparing surgery for cervical cancer should exclusively be under- woman with a desire to preserve fertility should be counselled about the possibility of taken in specific gynaecologic oncology centres with highly trained surgeons, every fertility-sparing surgery and be referred to a tertiary care hospital with comprehensive young woman expertise in thiswith kinda ofdesire to preserve oncologic therapyfertility [24]. should be counselled about the possibility of fertility-sparing Nevertheless, the oncological outcomestoofa Spanish surgery and be referred tertiary women care hospital with comprehensive with cervical cancer and a expertise in this kind of oncologic therapy [24]. tumour size <2 cm are consistent and in concordance with those published in the literature [25]. Nevertheless, In our study, the theonly oncological independent outcomes factor of thatSpanish impactedwomen with cervical the oncological cancerafter outcomes and a tumour size <2 cm are consistent and in concordance with those published fertility-sparing surgery was tumour size. Patients who underwent trachelectomy with a in the litera- ture [25].size of 2–4 cm had an almost sixfold increased risk of recurrence compared to those tumour with tumours <2 cm, even with a higher rate of adjuvant treatment. Publications focusing on oncological outcomes in tumours larger than 2 cm are heterogeneous. The reported recurrence rates for patients with tumours larger than 2 cm vary tremendously, ranging from 0% to 38% [15]. An essential criterion for performing FSS is that the nodes are negative. Park et al. [26] suggested that lymph node assessment with sentinel lymph node biopsy and/or full lymph node dissection should be performed in this population. Although some authors [27] advo- cate for the evaluation of the nodes after the administration of neoadjuvant chemotherapy, J. Pers. Med. 2022, 12, 1081 12 of 14 considering that neoadjuvant chemotherapy can eliminate the possibility of micrometas- tases and, thus, is able to preserve fertility in most cases, all the patients in this series were negative for intraoperative nodes and, therefore, underwent surgery. Most of the teams represented in the study considered that patients with lymph node dissemination should be treated with adjuvant chemoradiotherapy and, therefore, have no surgical indication. This is in accordance with current guidelines from the National Comprehensive Cancer Network and the European Society for Medical Oncology [3,28]. Tumour size (and, thus, disease stage) is the second main criterion when considering the indication and the type of fertility-sparing surgery. In our data, tumours >2 cm recurred in 14.8% of cases compared to those <2 cm, which only recurred in 3.9% of cases (p < 0.0083) for both classical and experimental techniques. This was demonstrated with an HR value of 2.09 for tumours >2 cm. This is in agreement with other authors who affirmed the same [26] This adverse result observed in our series is independent of histology, lymphovascular space invasion, previous conisation, or surgical approach. Regarding the surgical approach, we found no differences when comparing vaginal or minimal invasion routes, contrary to what was already published in the LACC trial [7]. It is noteworthy that in our matched series, 60.5% and 65.8% of FSSs and radical hysterectomies, respectively, were performed with a minimally invasive approach. However, a tendency is observed when comparing relapses from laparoscopic radical hysterectomy with vaginal radical hysterectomy (16% versus 7.69%). Nevertheless, in a recent randomised clini- cal trial [29], a comparison between open and laparoscopic radical trachelectomy was performed, and they found no differences in terms of DFS or OS. Nevertheless, as we can observe in Figure 3, there a tendency for the experimental technique to be inferior in terms of DFS. Therefore, although strictly speaking we could not affirm or deny that FSS is inferior to the conventional technique for the treatment of initial cervical cancer, we cannot forget that there was a clear trend toward a decrease in the DFS when we applied the experimental technique, especially for tumour sizes of more than 2 cm. In other words, FSS may have a certain risk of recurrence and should only be offered if there is a significant desire to preserve fertility. 5. Strengths and Weaknesses One of the main strengths of this study is that it offers a vision in terms of the status of the conservative treatment of cervical cancer in early stages over a wide period in the Spanish state. It also collects the results of tertiary hospitals in Spain dedicated to the treatment of gynaecological cancer. At the same time, this work allowed the creation of a network of centres dedicated to cervical cancer in our country from which numerous collaborations have arisen (Spain GOG cervical cancer taskforce). Another strength of this study is the considerable amount of statistical work carried out to be able to balance and extract reliable conclusions from the data. It is well known that a retrospective study has multiple biases that limit its clinical validity, but it is also true that well-performed propensity score matching makes the results obtained through this technique resemble a prospective study by equally balancing the arms of the study. In addition, with the introduction of noninferiority studies in daily clinical practice, the applicability of an experimental treatment with respect to the standard can be compared with great efficiency. Regarding the weaknesses, one of the principal weak points of this study is its retro- spective nature and the long duration of the analysis, nearly 14 years, during which some modifications were introduced progressively in the standard of care for the management of cervical cancer, including the introduction of sentinel node biopsy [30]. Another weakness of this study lies in the low number of cases resulting after balancing the study arms, which had an even greater effect, if possible, on the low frequency at which FSS was performed in our environment. In conclusion, in this study, none of the analyses allowed us to demonstrate the inferiority of the experimental method compared to the standard method. Patients with tumours larger than 2 cm had a higher risk of recurrence regardless of the technique used. J. Pers. Med. 2022, 12, 1081 13 of 14 Therefore, according to our results, fertility preservation surgery in the initial stages of cervical cancer can be just as effective and safe as radical treatment. Author Contributions: B.G.-I., B.D.-F., A.L. and M.V.I.: conception and design, data acquisition, analysis and interpretation, writing of the manuscript; A.T., A.G.-M., A.M.-J., B.G.-I., A.S. and M.T.C.: acquisition of data and revision for important intellectual content. Spain-GOG cervical cancer working group: acquisition of data. All authors have read and agreed to the published version of the manuscript. Funding: This work received financial support from the Medtronic University Chair for Training and Surgical Research, University Jaume I (UJI), Castellon, Spain. Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the Institutional Review Board (or Ethics Committee) of Comité de Ética de Investigación (CEIm) del Hospital Clínic de Barcelona (No Registro HCB/2019/0734) for studies involving humans. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Data Availability Statement: The data that support the findings of this study are available from the corresponding author upon reasonable request. Acknowledgments: The authors would like to thank all the investigators from the twelve sites of the CEFER study. Spain-GOG cervical cancer working group † : Berta Díaz-Feijoo, Aureli Torné, Blanca Gil-Ibáñez, Antonio Llueca, Antonio Gil-Moreno, Maite Climent, Anna Serra, Pluvio Coronado, Gemma Mancebo, Marina Martin, Martinez Canto, Dolores Piquer, Cabrera Diaz, Angel Martin. Conflicts of Interest: The authors declare no conflict of interest. References 1. Querleu, D.; Morrow, C.P. Classification of Radical Hysterectomy. Lancet Oncol. 2008, 9, 297–303. Available online: https: //pubmed.ncbi.nlm.nih.gov/18308255/ (accessed on 4 January 2022). [CrossRef] 2. Sant, M.; Lopez, M.D.C.; Agresti, R.; Pérez, M.J.S.; Holleczek, B.; Bielska-Lasota, M.; Dimitrova, N.; Innos, K.; Katalinic, A.; Langseth, H.; et al. Survival of women with cancers of breast and genital organs in Europe 1999–2007: Results of the EUROCARE-5 study. Eur. J. Cancer 2015, 51, 2191–2205. [CrossRef] [PubMed] 3. Koh, W.J.; Abu-Rustum, N.R.; Bean, S.; Bradley, K.; Campos, S.M.; Cho, K.R.; Chon, H.S.; Chu, C.; Clark, R.; Cohn, D.; et al. Cervical Cancer, Version 3.2019, NCCN Clinical Practice Guidelines in Oncology. J. Natl. Compr. Cancer Netw. JNCCN 2019, 17, 64–84. Available online: https://pubmed.ncbi.nlm.nih.gov/30659131/ (accessed on 5 January 2022). [CrossRef] [PubMed] 4. Novak, F. Radical abdominal subcorporeal extirpation of the cervix with bilateral pelvic lymph nodes dissection in cancer in situ of the cervix uteri. Acta Med. Iugosl. 1952, 6, 59–71. [PubMed] 5. Aburel, E. Sub-corporeal extended colpohysterectomy in therapy of incipient cancer of cervix. Comptes R. Soc. Fr. Gyncol. 1957, 27, 237–243. 6. Dargent, D.; Martin, X.; Sacchetoni, A.; Mathevet, P. Laparoscopic vaginal radical trachelectomy: A treatment to preserve the fertility of cervical carcinoma patients. Cancer 2000, 88, 1877–1882. [CrossRef] 7. Ramirez, P.T.; Frumovitz, M.; Pareja, R.; Lopez, A.; Vieira, M.; Ribeiro, R.; Buda, A.; Yan, X.; Shuzhong, Y.; Chetty, N.; et al. Minimally Invasive versus Abdominal Radical Hysterectomy for Cervical Cancer. N. Engl. J. Med. 2018, 379, 1895–1904. [CrossRef] 8. Salvo, G.; Ramirez, P.T.; Leitao, M.; Cibula, D.; Fotopoulou, C.; Kucukmetin, A.; Rendon, G.; Perrotta, M.; Ribeiro, R.; Vieira, M.; et al. International Radical Trachelectomy Assessment: IRTA Study. Int. J. Gynecol. Cancer Off. J. Int. Gynecol. Cancer Soc. 2019, 29, 635–638. Available online: https://pubmed.ncbi.nlm.nih.gov/30765489/ (accessed on 14 January 2022). [CrossRef] 9. Pecorelli, S.; Zigliani, L.; Odicino, F. Revised FIGO staging for carcinoma of the cervix. Int. J. Gynaecol. Obstet. Off. Organ Int. Fed. Gynaecol. Obstet. 2009, 105, 107–108. Available online: https://pubmed.ncbi.nlm.nih.gov/19342051/ (accessed on 14 January 2022). [CrossRef] [PubMed] 10. Clavien, P.A.; Sanabria, J.R.; Strasberg, S.M. Proposed classification of complications of surgery with examples of utility in cholecystectomy. Surgery 1992, 111, 518–526. [PubMed] 11. Anderson-Cook, C.M. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. J. Am. Stat. Assoc. 2005, 100, 708. [CrossRef] 12. Harrell, F.E.; Lee, K.L. Verifying assumptions of the Cox proportional hazards model. In Proceedings of the Eleventh Annual SAS Users Group International Conference, Atlanta, GA, USA, 9–12 February 1986; pp. 823–828. 13. Fleshman, J.; Branda, M.; Sargent, D.J.; Boller, A.M.; George, V.; Abbas, M.; Peters, W.R.; Maun, D.; Chang, G.; Herline, A.; et al. Effect of Laparoscopic-Assisted Resection vs. Open Resection of Stage II or III Rectal Cancer on Pathologic Outcomes: The J. Pers. Med. 2022, 12, 1081 14 of 14 ACOSOG Z6051 Randomized Clinical Trial. JAMA 2015, 314, 1346–1355. Available online: https://pubmed.ncbi.nlm.nih.gov/26 441179/ (accessed on 14 January 2022). [CrossRef] [PubMed] 14. Stevenson, A.R.; Solomon, M.J.; Lumley, J.W.; Hewett, P.; Clouston, A.D.; Gebski, V.J.; Davies, L.; Wilson, K.; Hague, W.; Simes, J.; et al. Effect of Laparoscopic-Assisted Resection vs. Open Resection on Pathological Outcomes in Rectal Cancer: The ALaCaRT Randomized Clinical Trial. JAMA 2015, 314, 1356–1363. Available online: https://pubmed.ncbi.nlm.nih.gov/26441180/ (accessed on 14 January 2022). [CrossRef] 15. Janda, M.; Gebski, V.; Davies, L.C.; Forder, P.; Brand, A.; Hogg, R.; Jobling, T.W.; Land, R.; Manolitsas, T.; Nascimento, M.; et al. Effect of Total Laparoscopic Hysterectomy vs. Total Abdominal Hysterectomy on Disease-Free Survival among Women with Stage I Endometrial Cancer: A Randomized Clinical Trial. JAMA 2017, 317, 1224–1233. Available online: https://pubmed.ncbi. nlm.nih.gov/28350928/ (accessed on 14 January 2022). [CrossRef] 16. R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2021. 17. Ho, D.E.; Imai, K.; King, G.; Stuart, E.A. MatchIt: Nonparametric Preprocessing for Parametric Causal Inference. J. Stat. Softw. 2011, 42, 1–28. Available online: https://www.jstatsoft.org/index.php/jss/article/view/v042i08 (accessed on 16 January 2022). [CrossRef] 18. Noah Greifer. Covariate Balance Tables and Plots. R Package Version 4.3.1. 2021. Available online: https://CRAN.R-project.org/ package=cobalt (accessed on 1 May 2022). 19. Therneau, T.M.; Grambsch, P.M. A Package for Survival Analysis in S, Version 2.38. Modeling Survival Data: Extending the Cox Model. 2000. Available online: http://cran.r-project.org/package=survival (accessed on 16 January 2022). 20. Möllenhoff, K.; Tresch, A. Survival Analysis under Non-Proportional Hazards: Investigating Non-Inferiority or Equivalence in Time-to-Event Data. 2020. Available online: http://arxiv.org/abs/2009.06699 (accessed on 16 January 2022). 21. Kassambara, A. Drawing Survival Curves Using “ggplot2” [R Package Survminer Version 0.4.9]. 2021. Available online: https://CRAN.R-project.org/package=survminer (accessed on 16 January 2022). 22. Tseng, J.H.; Aloisi, A.; Sonoda, Y.; Gardner, G.J.; Zivanovic, O.; Abu-Rustum, N.R.; Leitao, M.M., Jr. Less Versus More Radical Surgery in Stage IB1 Cervical Cancer: A Population-Based Study of Long-Term Survival. Gynecol. Oncol. 2018, 150, 44–49. Available online: https://pubmed.ncbi.nlm.nih.gov/29776598/ (accessed on 21 January 2022). [CrossRef] [PubMed] 23. Gorostidi, M.; Gil-Ibañez, B.; Alonso, S.; Gil-Moreno, A.; Hernandez, A.; Torné, A.; Zapardiel, I. Fertility Preservation Treatment of Gynecological Cancer Patients in Spain: A National Survey (GOFER Study). Arch. Gynecol. Obstet. 2020, 301, 793–800. Available online: https://pubmed.ncbi.nlm.nih.gov/32124016/ (accessed on 21 January 2022). [CrossRef] 24. Cibula, D.; Pötter, R.; Planchamp, F.; Avall-Lundqvist, E.; Fischerova, D.; Haie-Meder, C.; Köhler, C.; Landoni, F.; Lax, S.; Lindegaard, J.C.; et al. The European Society of Gynaecological Oncology/European Society for Radiotherapy and Oncol- ogy/European Society of Pathology Guidelines for the Management of Patients with Cervical Cancer. Int. J. Gynecol. Cancer Off. J. Int. Gynecol. Cancer Soc. 2018, 28, 641–655. Available online: https://pubmed.ncbi.nlm.nih.gov/29688967/ (accessed on 5 January 2022). [CrossRef] 25. Bentivegna, E.; Gouy, S.; Maulard, A.; Chargari, C.; Leary, A.; Morice, P. Oncological Outcomes after Fertility-Sparing Surgery for Cervical Cancer: A Systematic Review. Lancet Oncol. 2016, 17, e240–e253. Available online: https://pubmed.ncbi.nlm.nih.gov/27 299280/ (accessed on 5 January 2022). [CrossRef] 26. Park, J.Y.; Kim, D.Y.; Kim, J.H.; Kim, Y.M.; Kim, Y.T.; Nam, J.H. Outcomes after Radical Hysterectomy according to Tumor Size Divided by 2-cm Interval in Patients with Early Cervical Cancer. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol. 2011, 22, 59–67. Available online: https://pubmed.ncbi.nlm.nih.gov/20595451/ (accessed on 5 January 2022). [CrossRef] 27. Cao, D.Y.; Yang, J.X.; Wu, X.H.; Chen, Y.L.; Li, L.; Liu, K.J.; Cui, M.H.; Xie, X.; Wu, Y.M.; Kong, B.H.; et al. Comparisons of Vaginal and Abdominal Radical Trachelectomy for Early-Stage Cervical Cancer: Preliminary Results of A Multi-Center Research in China. Br. J. Cancer 2013, 109, 2778–2782. Available online: https://pubmed.ncbi.nlm.nih.gov/24169350/ (accessed on 5 January 2022). [CrossRef] [PubMed] 28. Marth, C.; Landoni, F.; Mahner, S.; McCormack, M.; Gonzalez-Martin, A.; Colombo, N. Cervical Cancer: ESMO Clinical Practice Guidelines for Diagnosis, Treatment and Follow-Up. Ann. Oncol. Off. J. Eur. Soc. Med. Oncol. 2017, 28, iv72–iv83. Available online: https://pubmed.ncbi.nlm.nih.gov/28881916/ (accessed on 5 January 2022). [CrossRef] [PubMed] 29. Salvo, G.; Ramirez, P.T.; Leitao, M.M.; Cibula, D.; Wu, X.; Falconer, H.; Persson, J.; Perrotta, M.; Mosgaard, B.J.; Kucukmetin, A.; et al. Open vs minimally invasive radical trachelectomy in early-stage cervical cancer: International Radical Trachelectomy Assessment Study. Am. J. Obstet. Gynecol. 2022, 226, 97.e1–97.e16. [CrossRef] [PubMed] 30. Salvo, G.; Ramirez, P.T.; Levenback, C.F.; Munsell, M.F.; Euscher, E.D.; Soliman, P.T.; Frumovitz, M. Sensitivity and Negative Predictive Value for Sentinel Lymph Node Biopsy in Women with Early-Stage Cervical Cancer. Gynecol. Oncol. 2017, 145, 96–101. Available online: https://pubmed.ncbi.nlm.nih.gov/28188015/ (accessed on 21 January 2022). [CrossRef] [PubMed]