Article Text
Abstract
Objective To assess the feasibility, safety, oncological, and obstetric outcomes in patients with cervical tumors >2 cm treated with neoadjuvant chemotherapy in preparation for abdominal radical trachelectomy.
Methods A retrospective analysis of patients with cervical cancer >2 cm (up to 6 cm) was conducted in patients who were selected to receive neoadjuvant chemotherapy before abdominal radical trachelectomy. Surgical and clinical outcomes were examined in relation to radiological and pathological results. In addition, obstetric outcomes were described. The Mann–Whitney U test and Fisher’s exact test were performed to compare radiological findings between successful and unsuccessful abdominal radical trachelectomy procedures. International Federation of Gynecology and Obstetrics (FIGO) 2009 staging classification was used for this study.
Results A total of 19 women were treated with neoadjuvant chemotherapy for cervical tumors >2 cm at our institution between May 2006 and July 2018. The median age was 28 years (range 19–36). The distribution of FIGO stages was seven patients stage IB1 (37%), 10 patients stage IB2 (53%), and two patients (10%) stage IIA. Mean clinical tumor size was 4.4 cm (range 3.5–6.0). Histology revealed 74% cases of squamous cell carcinoma. The remaining patients had adenocarcinoma (21%) and only one patient had clear cell adenocarcinoma (5%). Chemotherapy consisted of six weekly cycles of cisplatin (70 mg/m2) and paclitaxel (70 mg/m2). In 15 of the 19 patients (74%) fertility was successfully preserved. In the four patients in whom fertility preservation failed, one patient had stable disease after three cycles and did not meet the criteria for fertility-sparing surgery and three patients had intra- or post-operative indications for adjuvant therapy. Three of the 19 patients (15.7%) had a relapse, two of whom died. One case was in the group of successful abdominal radical trachelectomy.
Conclusion Neoadjuvant chemotherapy followed by fertility-sparing surgery may be a feasible and safe option in select patients with cervical tumors >2 cm. Unfavorable prognostic factors are defined as non-responsiveness and non-squamous pathology, which can help in patient selection for fertility-sparing surgery.
- cervical cancer
- uterine cervical neoplasms
Statistics from Altmetric.com
HIGHLIGHTS
Neoadjuvant chemotherapy prior to fertility-sparing surgery in women with cervical cancer >2 cm is feasible and safe in selected women with FIGO stage IB–IIA disease
After abdominal radical trachelectomy, the local recurrence rate was 13.3% in our study
The CONTESSA/NEOCON-F study is designed to assess the feasibility of preserving fertility after neoadjuvant chemotherapy in patients with FIGO stage IB2 cervical cancer
Introduction
Cervical cancer is considered the fourth leading global cause of cancer death in women based on an estimated 570 000 cases and 311 000 deaths in 2018 worldwide.1 In all cases, it is associated with a significant socio-economic impact in life-years lost, reduced quality of life, and potentially fertility loss as patients are diagnosed at a relatively young age.2
With the introduction of the vaginal trachelectomy by Dargent et al,3 fertility-sparing surgery has been considered a safe alternative for early stage cervical cancer with a tumor diameter ≤ 2 cm in patients who wish to preserve fertility.4 However, tumor size remains a challenge for safe fertility-sparing surgery as tumors >2 cm are associated with a higher recurrence rate compared with smaller tumors, as reviewed by Bentivegna et al.5 Therefore, patients with tumors >2 cm are considered high-risk and not eligible for fertility-sparing surgery. Since the early 2000s, neoadjuvant chemotherapy preceding fertility-sparing surgery has been introduced as a treatment option to preserve fertility in cervical cancer patients with tumors >2 cm.6 However, only scarce data on clinical outcomes are available and protocols for uniform management are lacking. In 2006, the Department of Gynecology of Leiden University Medical Center introduced neoadjuvant chemotherapy followed by abdominal radical trachelectomy in patients with cervical cancer >2 cm as an alternative to a radical hysterectomy or chemoradiation. The aim of this retrospective study was to evaluate the feasibility, oncological safety, and clinical outcomes of neoadjuvant chemotherapy and abdominal radical trachelectomy in women with cervical cancer >2 cm.
Methods
All patients with cervical cancer >2 cm who were selected for neoadjuvant chemotherapy preceding fertility-sparing surgery at our department between May 2006 and July 2018 were included in this retrospective cohort study. All records were extracted from the electronic patient database. Clinical characteristics, treatment outcomes, and survival data were evaluated. Patients with cervical cancer who had fertility-sparing surgery without receiving neoadjuvant chemotherapy were excluded. Patients who were pregnant at the time of treatment were also excluded
Neoadjuvant chemotherapy was considered in patients presenting with cervical carcinoma International Federation of Gynecology and Obstetrics (FIGO) stage IB–IIA (FIGO 2009 classification) with radiologic or clinical evidence of a tumor >2 cm. All patients were discussed in a multidisciplinary team and patients were extensively counseled. Neoadjuvant chemotherapy involved six cycles of weekly paclitaxel 70 mg/m2 and cisplatin 70 mg/m2. The toxicity of chemotherapy was also registered and analyzed. After three cycles the tumor response was clinically and radiologically evaluated according to Response Evaluation Criteria in Solid Tumors (RECIST) criteria. In cases of partial response, chemotherapy was continued with another three cycles followed by surgery. In cases of stable disease (no changes in size) or progressive disease (increasing size), fertility-sparing treatment was discontinued and patients were submitted to standard treatment (radiotherapy with concomitant chemotherapy or radical hysterectomy) depending on the stage of the disease.
Fertility-sparing surgery consisted of a nerve-sparing abdominal radical trachelectomy with pelvic lymphadenectomy. We performed nerve-sparing abdominal trachelectomy from 2005 onwards, similar to the Swift procedure for nerve-sparing radical hysterectomy as described by van Gent et al.7 Intra-operative frozen sections of the pelvic lymph nodes as well as the isthmus were evaluated. In cases of tumor-positive lymph nodes or isthmus, surgery was converted into radical hysterectomy or the uterus was left in situ, depending on the stage of the operation. If post-operative pathology results showed positive lymph nodes, positive surgical margins, or positive parametria, adjuvant radiotherapy and/or chemoradiation was administered. For all treatment decisions, patients were discussed during multidisciplinary consultation.
The primary endpoint was the rate of abdominal radical trachelectomy. Secondary endpoints were recurrence-free survival after fertility-sparing surgery or radical hysterectomy. In cases where only chemoradiation or post-operative (chemo)radiation was administered, recurrence-free survival was calculated from the date (chemo)radiotherapy treatment was completed and the patient had no evidence of disease. Patients were censored in the survival analysis if follow-up ended without recurrence of disease or if they were lost to follow-up.
Pathological responses were determined post-operatively from surgical specimens according to the SNAP01 trial criteria.8 An optimal response included complete pathological response (no detectable tumor in the surgical specimen) or residual disease ≤3 mm stromal invasion. A sub-optimal response consisted of persistent residual disease with more than 3 mm stromal invasion. Obstetric data were retrieved via the electronic patient database.
To investigate if tumor response could be calculated more accurately compared with existing diameter-based measurements, tumor volume reduction in percentage was calculated based on tumor volume of the baseline MRI and the follow-up MRI, performed after three courses of chemotherapy. Therefore, these measurements were computed as an ellipsoid (V=dcc × dap× dl × Π/6).9 The craniocaudal diameter (dcc) along the long axis of the endometrial cavity and the anteroposterior diameter (dap) were measured on the sagittal images. The largest lateral diameter (dl) was measured on the axial images. Tumor volume reduction between successful and unsuccessful abdominal radical trachelectomy procedures were compared. Unsuccessful abdominal radical trachelectomy was defined as abdominal radical trachelectomy converted pre- or intra-operatively to radical hysterectomy or (chemo)radiation or requiring post-operative (chemo)radiation. This study was approved by the local ethical committee.
The survival curves were plotted by the Kaplan–Meier method and compared using the log rank test. The Mann–Whitney U test and Fisher’s xact test were performed to compare radiological findings between successful and unsuccessful abdominal radical trachelectomy procedures. P values <0.05 were considered statistically significant. All statistical analyses were performed using IBM SPSS Statistics 23.
Results
In total, 19 patients with cervical cancer >2 cm received neoadjuvant chemotherapy prior to attempted abdominal radical trachelectomy between May 2006 and July 2018 and were included in the study. Patient and tumor characteristics are summarized in Table 1. The median age was 28 years (range 19–36). The distribution of FIGO stages was: 7 patients with stage 1B1 (37%), 10 patients with stage 1B2 (53%), and 2 patients (10%) with stage IIA. Mean clinical tumor size was 4.4 cm (range 3.5–6). Pre-treatment histology was retrieved by loop electrosurgical excision or biopsy. Histology revealed 74% of patients with squamous cell carcinoma, 21% with adenocarcinoma, and 5% with clear cell adenocarcinoma.
A patient flow chart illustrating treatment outcomes is shown in Figure 1. A summary of clinical, histopathological, and radiological characteristics before and after fertility-sparing surgery preceded by neoadjuvant chemotherapy is given in Table 1. Six patients underwent laparoscopic lymphadenectomy prior to neoadjuvant chemotherapy. The main reasons were: suspected lymph nodes on pre-operative imaging (patients 2 and 16), tumor size exceeding 6 cm (patient 17), and FIGO stage IIA (patients 1 and 10). One case (patient 14) underwent a lymphadenectomy prior to neoadjuvant chemotherapy in another hospital and had an abdominal radical trachelectomy at our hospital. In all six patients, none had positive lymph nodes.
All patients received weekly cisplatin-paclitaxel as induction chemotherapy. Eleven patients (58%) finished the total number of six cycles of chemotherapy. Three patients (16%) did not complete their last cycle of chemotherapy due to acute renal failure (patients 7 and 13) and bone marrow depression (patient 12). In one patient (patient 14), chemotherapy was discontinued after two cycles because she developed a stroke. Four patients (21%) discontinued chemotherapy after only three cycles, either due to inadequate response of the tumor (patients 5 and 16) or complete regression of the tumor (patients 15 and 17).
One patient showed stable disease during neoadjuvant chemotherapy. Therefore, this patient underwent a radical hysterectomy (patient 16). Eighteen of the 19 patients (95%) underwent surgery with the intention of fertility-sparing surgery. Fifteen of the 18 patients (83%) resulted in a successful abdominal radical trachelectomy and three resulted in unsuccessful abdominal radical trachelectomy (Figure 1). No intra-operative complications occurred. The median number of lymph nodes removed was 21 (range 10–35). Abdominal radical trachelectomy was converted to hysterectomy intra-operatively in two cases due to positive frozen section of the uterine isthmus (patient 17) and positive pelvic lymph node (patient 18). In one case (patient 4), post-operative pathology results showed parametrium involvement and therefore this patient received adjuvant radiotherapy.
The recurrence-free survival probability of the successful abdominal radical trachelectomy group was 0.92 (95% CI 0.78 to 1.07) and 0.82 (95% CI 0.59 to 1.05) for 2 and 5 years, respectively. The recurrence-free survival probability of the unsuccessful abdominal radical trachelectomy group was 0.75 (95% CI 0.32 to 1.17) for both 2 and 5 years. Log rank test was 0.58, showing no significant difference between the two groups.
After neoadjuvant chemotherapy, nine of the 19 patients (47%) had an optimal pathological response, five of whom had a complete pathological response (no detectable lesion) and four patients had residual disease with ≤3 mm stromal invasion (Table 1). A sub-optimal pathological response (>3 mm stromal invasion) was observed in seven patients and stable disease (<50% reduction) in three cases. Among the five patients with complete pathological responses, two had an adenocarcinoma and three had squamous cell carcinoma (online supplementary Tables S1 and S2). Three of the five patients showed no lymphovascular space invasion and, in two patients, lymphovascular space invasion status was unknown.
Supplemental material
An example of MRI findings before and after three cycles of chemotherapy is shown in online supplementary Figure S1. For this study we examined tumor volume reduction as a parameter to calculate response on neoadjuvant chemotherapy. Tumor volume reduction could not be evaluated in three patients because of inability to distinguish tumor tissue from surrounding tissue (patients 3 and 12) or a missing MRI after three cycles of chemotherapy (patient 17). The overall mean tumor volume before treatment in these 16 patients was 18.8 cm3. No significant difference was seen in pre-treatment mean tumor volume or mean cranial extent between successful (n=13, mean=18.6 cm3) and unsuccessful abdominal radical trachelectomies (n=3, mean=19.8 cm3). However, mean tumor volume reduction on MRI after three courses was significantly higher in patients who underwent successful abdominal radical trachelectomy compared with patients in whom abdominal radical trachelectomy was unsuccessful (84% and 38.6%, respectively; p=0.036). These results are shown in Table 2.
Interestingly, a complete clinical response was associated with large tumor volume reduction on MRI (TVR range 76.3–100%) whereas stable disease was associated with lower tumor volume reduction (TVR range 41.5–62%). One patient (patient 10) showed a discordant clinical and radiological response. In addition, six of eight patients with a complete clinical response showed an optimal pathological response whereas two of eight patients displayed a suboptimal pathological response. However, definitive conclusions on the agreement between clinical radiological and pathological responses cannot be drawn because of the small sample size of our cohort.
Median follow-up time was 50 months (range 3–144). Of the 19 patients, 17 (89%) were without any evidence of disease at July 2018, and in 13 patients fertility is still preserved. Three patients relapsed. Two had undergone a successful abdominal radical trachelectomy (patients 14 and 15, Table 1). Patient 14 had a squamous cell carcinoma and underwent a lymphadenectomy prior to abdominal radical trachelectomy. She had a loco-regional relapse 17 months after abdominal radical trachelectomy and received chemoradiation but died 3 months after the recurrence. Patient 15 had an adenocarcinoma and developed a local recurrence 45 months after abdominal radical trachelectomy and lymphadenectomy. A salvage hysterectomy was performed. Final pathology showed a focus of adenocarcinoma (2 mm) which corresponded to the initial tumor. Currently this patient is without evidence of disease. Therefore, in the group of patients undergoing lymphadenectomy during abdominal radical trachelectomy, one of nine patients developed a local recurrence compared with one of six patients who developed a local-regional recurrence in the group undergoing lymphadenectomy prior to neoadjuvant chemotherapy. One patient (patient 18) was diagnosed with recurrence after an unsuccessful abdominal radical trachelectomy. This patient had an unsuccessful abdominal radical trachelectomy due to positive pelvic lymph node intra-operatively. A hysterectomy was performed followed by adjuvant radiotherapy. However, 3 months after finishing radiotherapy she had a regional recurrent disease and died of the disease.
Three of 15 patients with a successful abdominal radical trachelectomy became pregnant. Eight spontaneous pregnancies in these three women were established, which in six of the eight pregnancies led to the birth of healthy babies. All delivered at full term via cesarian section without complications. One patient terminated two pregnancies at 4 and 5 weeks due to personal reasons. One patient (patient 15) needed assisted reproductive treatment because of primary infertility but developed a local recurrence and was treated with a hysterectomy. One patient was diagnosed with Asherman syndrome, for which she is receiving treatment currently.
Discussion
Fertility-sparing surgery for small cervical tumors (≤2 cm) has become a widely adopted treatment modality with good oncologic outcomes. However, data on feasibility and clinical outcomes of fertility-sparing surgery in cervical tumors >2 cm preceded by neoadjuvant chemotherapy are scarce.
The combination of neoadjuvant chemotherapy with fertility-sparing surgery in comparison with standard treatment of cervical cancers >2 cm remains a topic for ongoing discussion in terms of safety in relation to pregnancy. In our study, two of 19 patients died. However, according to tumor size it is more realistic to consider this group of patients as a group with approximately similar prognosis to patients with FIGO 1B2 cervical tumors, since mean tumor size was 4.4 cm in this cohort. The average overall survival rate of this group reached 70–75%, which is lower compared with our cohort.10 This study shows that neoadjuvant chemotherapy prior to fertility-sparing surgery in women with cervical cancer >2 cm is feasible and safe in those with stage IB–IIA disease without lymph node metastasis. In our study, 15 of 19 (74%) patients underwent successful fertility-sparing surgery. Two of the 15 patients developed a recurrence, of which one died. The extent of reduction in tumor volume on MRI after three cycles of chemotherapy was associated with successful fertility-sparing surgery (p=0.036). Eight pregnancies in three women resulted in the birth of six babies, all at term.
There is a wide difference in chemotherapy regimens between the few studies that have reported on the treatment of cervical cancer >2 cm with neoadjuvant chemotherapy and fertility-sparing surgery (Table 3). Most studies describe the combination of paclitaxel, cisplatin, and ifosfamide/epirubicin (TIP/TEP).6 11–17 The interval between these cycles varies between 21 and 28 days compared with 7–14 days when a doublet regimen is administered. In our institution, cisplatin and paclitaxel with a 7-day interval were used. These dose-dense schemes make it possible to adjust treatment in a short time in case of non-responsiveness of the tumor. The most common neoadjuvant chemotherapy-related toxicity observed was hair loss (alopecia). Gastrointestinal side effects such as anorexia and nausea were also frequently described. Other side effects such as renal failure or bone marrow depression were rare.
In our study, relapse was diagnosed in three of 19 patients, two of whom died.In the successful abdominal radical trachelectomy group there were two of 15 recurrences (13.3%) and these results are in accordance with recurrence rates of a recent review by Marchiole et al, which was reported as 10.5%17 (Table 3). We extended this aforementioned review with our series, with case series including more than 15 patients with cervical cancer >2 cm treated with neoadjuvant chemotherapy and fertility-sparing surgery (Table 3). Notably, our study describes patients with the largest mean tumor size (44 mm) of all published studies in this review.
All patients in our group who underwent surgery in the context of fertility-sparing surgery underwent abdominal radical trachelectomy. Compared with a vaginal radical trachelectomy, the important advantage of abdominal radical trachelectomy is a wider parametrial resection. Einstein et al18 showed a 50% wider parametrial resection in abdominal radical trachelectomy compared with vaginal radical trachelectomy. Although the value of parametrial resection is debatable in lymph-vascular invasion negative ≤2 cm tumors,2 the minimum width of parametria is still a concern when treating larger tumors. Therefore, wider resection of the parametria is important to rule out any tumor involvement in the parametria, indicating adjuvant therapy, and to prevent local recurrence. Some studies show comparable recurrence rates to vaginal radical trachelectomy,19–21 but Cao et al22 compared both approaches and showed a significant difference in the case of >2 cm tumors in favor of abdominal radical trachelectomy, indicating a possible benefit for larger cervical tumors as in the cases of our cohort, when abdominal radical trachelectomy is performed.
Although there is a consensus that rare tumor sub-types (eg, clear cell adenocarcinoma or neuroendocrine tumors) have a higher risk of recurrence, it remains still undecided whether this holds true for adenocarcinoma compared with squamous cell carcinoma.23–29 Five of 19 (26%) patients of our cohort had non-squamous histology, involving four patients with adenocarcinoma and one clear cell adenocarcinoma. In only one case of adenocarcinoma was fertility preserved (25%), compared with 12 of 14 cases (86%) es of squamous cell carcinoma. Furthermore, this adenocarcinoma case had the shortest follow-up time of our cohort (9 months). Some authors, such as Mabuchi et al, support the view of an inferior prognosis of adenocarcinoma, concluding that adenocarcinoma is an independent unfavorable prognostic factor in surgically treated early stage cervical carcinoma.30 Furthermore, Zusterzeel et al29 reported a significantly higher recurrence rate in adenocarcinoma in a cohort of only vaginal radical trachelectomy. Although this sample size only consists of 19 patients and data comparing both histology sub-types remain limited, our data suggest that, when fertility-sparing surgery is considered, extra caution should be taken in cases of adenocarcinoma.
Crucial in treating tumors >2 cm with fertility-sparing surgery is the selection of eligibility for induction chemotherapy, since the risk of non-responsiveness and consequently tumor progression delay treatment and impair prognosis. This is reflected in the tumor volume reduction, which was significantly lower in the unsuccessful abdominal radical trachelectomy group (as shown in Table 2). The only patient with an excellent response, but relapsed, had adenocarcinoma. Our series indicates that stable disease after three courses of chemotherapy is an unfavorable prognostic indicator for survival.
This study has several limitations. First, the use of retrospective data for analysis. Although we collected data on 19 women with cervical cancers >2 cm who were treated by fertility-sparing surgery following neoadjuvant chemotherapy, which is, to our knowledge, one of the largest numbers to date (Table 3), the number is limited. Second, we examined abdominal radical trachelectomy as surgical treatment. A comparison between vaginal and abdominal radical trachelectomy could not be assessed and therefore future studies should be aimed at examining what surgical approach is preferable, taking into account the complications and oncological and fertility outcomes. It is therefore of vital importance to pool international data prospectively in order to improve treatment for these patients.
Our results show that fertility-sparing surgery is feasible and safe in a selected group of patients and recurrence rates are comparable with current literature. Possible unfavorable prognostic factors are radiological or clinically-confirmed stable or progressive disease on chemotherapy and non-squamous histology, but confirmation in larger prospective studies is necessary. Therefore, we are looking forward to the results of the recently proposed NEOCON-F/CONTESSA study (NCT 04016389).31 The aim of this study is to examine the safety and efficacy of neoadjuvant chemotherapy and fertility-sparing surgery by simple trachelectomy/conization in women with node negative FIGO 2018 stage IB2 cervical cancer.
References
Footnotes
Contributors FMT: collecting data, analyzing, writing. RvL: providing data, reviewing. JJB, MIEvP, JRK, BT, KG: analyzing, reviewing and writing. CdK, RAN, VS: reviewing data.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.