Abstract
Background/Aim: In selected patients, pelvic exenteration (PE) is curative, but morbidity and mortality are feared. Unfortunately, prerequisites for indicating PE are not generally defined. The aim of the study was to identify prognostic factors for survival after PE in advanced pelvic gynecological malignancies for finding possible prerequisites for the indication of PE. Patients and Methods: Between 2002 and 2016, 49 patients underwent pelvic exenteration for advanced pelvic malignancies apart from ovarian cancer. Progression-free survival (PFS) and overall survival (OS) were calculated based on the Kaplan-Meier method. Factors significantly affecting 5-year overall survival were identified using multivariate regression analysis. Survival distributions between the best and the worst group were compared by the log rank test. Results: Forty-nine patients with recurrent or primary pelvic gynecological malignancy (20 recurrent disease, 29 primary disease) were included. Seventeen patients had oligometastatic disease at surgical intervention. Resection margin, age, primary versus secondary exenteration and metastatic disease were independent prognostic factors in multivariate regression analysis. A significant difference was observed in 5-year overall survival regarding the best group (57.14%) and the worst group (10%) (p=0.009). Cervical cancer was the only identified risk factor for increased morbidity. Conclusion: Pelvic exenteration is a valuable therapeutic option with most long-term survivors in the group of patients below 63 years, as primary treatment, with clear microscopic margins and no distant metastases. These four factors may serve as valuable prerequisites for the indication of pelvic exenteration as survival and morbidity in this group of patients compares favorably to alternative therapeutic options.
Pelvic exenteration was introduced in 1948 for advanced pelvic malignancies with palliative intent to relieve symptoms as no other therapy remained available (1). Currently, evidence for pelvic exenteration compared to non-surgical therapies, such as chemoradiation, radiation and systemic therapies, allows no conclusions concerning equivalence or superiority of pelvic exenteration regarding overall survival and progression-free survival (2-5). Unfortunately, pelvic exenteration may be associated with significant postoperative morbidity. However, in some patients it may also be the only curative treatment (6-9). Moore et al. identified prognostic factors for the survival effect of systemic therapy in cervical cancer patients (2). Similar factors to justify pelvic exenteration have not yet been identified. Herein, we aimed to identify prognostic factors for survival after pelvic exenteration. Furthermore, risk factors for unfavorable postoperative courses were evaluated.
Patients and Methods
Data collection. This study was conducted in accordance with the standards of the ethics committee of the Faculty of Medicine at the University of Bonn, Germany (Nr: 454/20). Data of all 49 patients with pelvic exenteration apart from ovarian cancer at the University of Bonn between 2002 and 2016 were retrospectively reviewed. Age, prior therapies, date of surgery, preoperative staging, type of surgery, type of reconstructive surgery, tumor entity, tumor stage, resection margins, distant metastasis, lymphovascular invasion, number and type of transfused blood products, duration of surgery, duration of mechanical postoperative ventilation, duration of intensive care, length of in-hospital stay, the time interval between first diagnosis and relapsed disease in case of pelvic exenteration in relapsed disease, overall survival and progression-free survival were analyzed. All postoperative complications were graded according to the Memorial Sloan Kettering cancer center secondary surgical event score in grade 1 to 5 (10).
Statistical analysis. The survival analyses for progression-free survival (PFS) and overall survival (OS) are based on the Kaplan– Meier method. The time-to-event intervals were described in months from the date of surgery until the date of the event. The data were censored at the date of the last follow up if there wasn’t an event. Using multivariate regression analyses, four factors significantly influencing the 5-year overall survival were identified. Based on these four factors, the best group of patients and the worst group of patients were identified. Using log-rank test, 5-year-survival-curves were compared on a 95% confidence level. Possible risk factors for an unfavorable postoperative course were evaluated by hypothesis testing on a 95% confidence level. All statistical tests were two-sided. All statistical analyses were performed using Minitab Version 18, Minitab LLC., State College, Pennsylvania, USA.
Results
Patient characteristics. Pelvic exenteration was performed in 49 patients with a median age of 65 years (range=33-86 years) between 2002 and 2016. Two patients had synchronous cancers (vulvar and anal cancer; vulvar and cervical cancer). There were 29 primary exenterations and 20 secondary exenterations. Pelvic exenteration was performed in 17 (34.7%) patients with oligometastatic disease for local symptom control. Metastatic sites were not regional lymph nodes (n=7), bone (n=2), spleen (n=1), small intestines (n=2), liver (n=3) and lung (n=2). Microscopic peritoneal involvement was seen in 5 (29.4%) out of 17 metastasized patients. Nine patients had received radiation therapy in previous treatment lines (adjuvant chemoradiation, adjuvant radiation, chemoradiation only, radiation only). Two patients had received neoadjuvant chemotherapy before pelvic exenteration. In recurrent disease, the median time interval from the first diagnosis to recurrent disease was 16.5 months (range=5-468 month). Table I shows the different tumor entities and characteristics of patients.
Staging. The pelvic and abdominal preoperative work-up included a cystoscopy in 42 patients, a rectoscopy in 34 patients, an abdominal CT scan in 40 patients, magnetic resonance imaging in 9 patients and an abdominal CT scan and pelvic magnetic resonance imaging in 14 patients. Two patients received a positron emission tomography-computed tomography scanning (PET-CT). All patients had a preoperative CT scan of the thorax.
Surgery details. Table II shows the surgical details. Two small intestine resections were necessary due to fistulas between the uterus and the small intestines after prior radiation therapy. Lymph node resection was omitted in 9 patients. There were 28 anterior, 7 posterior and 14 complete exenterations. Nineteen additional perineal resections and only 1 lateral extended endopelvic resection (LEER).
Final pathology report. Table III shows the details of the final pathology report. There were three patients with pT1 stages in final pathology. One patient had a large vaginal carcinoma, one patient had a rhabdomyosarcoma after neoadjuvant chemotherapy, and one patient had a synchronous pT3 anal cancer. The 19 final pT2 stages were cervical and vaginal cancers infiltrating in the bladder without mucosal infiltration.
Postoperative complications. Postoperative complications were graded according to the Memorial Sloan Kettering Cancer center – secondary surgical event score (MSKCC-SSE Score) (11). Severe complications (grade 3-5) occurred in 16 patients (32.7%). Table IV shows all postoperative complications by grade. Cervical cancer was the only significant risk factor for severe complications (p=0.006). Age (p=0.478), a higher risk classification according to the American Society of Anesthesiologists (ASA) (p=0.335), duration of mechanical ventilation (p= 0.774), duration of surgery (p=0.463), transfusion of blood products (p=0.509), suspected pelvic wall infiltration (p=0.593), neoadjuvant chemotherapy (p=0.245), and type of surgery (p=0.335) were not significant concerning the incidence of severe complications.
The following parameters were not suitable for evaluation as they were less common in patients affected by postoperative complications: smoking, relapsed disease, radiation therapy, M1, L1, higher T-Stage, cardiovascular risk factors, duration of intensive care. Severe postoperative complications in detail were 4 fistulations followed by abdominal abscesses in two cases. One out of those four patients died in the postoperative course. Further severe complications were 1 necrosis of the sigma conduit, 1 sepsis, 1 ureteral stenosis, 4 wound infections with secondary dehiscence, 1 lung embolism, 1 deep vein thrombosis, and 1 ileus.
Survival analysis. The 5-year overall survival was 31.5% and the 5-year over all progression-free survival was 28.7%. All parameters considered for the analysis regarding median overall and median progression-free survival are shown in Table V. Multivariate regression analysis, using a 95% confidence interval, identified 4 factors significantly affecting the 5-year overall survival: Resection margins, age, primary versus relapsed disease, and distant versus no distant metastatic disease. The group of patients with histologically confirmed tumor-free margins (R0), aged under 63 years without metastatic disease (M0), whenever PE was the primary treatment had a 5-year overall survival of 57.1%. This group showed the best 5-year overall survival and was therefore considered as the best group. The group of patients with histologically affected margins (R1), aged above 63 years with metastatic disease (M1), whenever PE was the secondary treatment had a 5-year overall survival of only 10%. This group showed the worst 5-year overall survival and was therefore considered as the worst group. The analysis of both survival curves by the log-rank test identified this difference to be significant (p=0.009). Table VI shows this survival analysis in groups. Figure 1 shows the survival curves of the best and the worst group.
Discussion
In selected cases, pelvic exenteration will be the only curative treatment. As randomized trials are missing there is no evidence to favor pelvic exenteration over nonsurgical treatments such as chemoradiation, radiation and systemic therapy in advanced gynecologic pelvic malignancies (12). Furthermore, the curative potential of pelvic exenteration is often weighted against mortality, morbidity, and mutilation. While mortality has declined in recent years from 23% to 2%, severe morbidity will still occur in 21% to 34% of all patients undergoing pelvic exenteration (1, 4, 6, 7, 9, 13, 14). However, early postoperative morbidity is not necessarily a permanent condition. Surgical intervention, antibiotics and anticoagulation as well as physical therapy substantially improved the long-term outcome in our cohort. Our data on morbidity of 32% and mortality of 2% are in accordance with the literature. The only identified significant risk factor for morbidity was cervical cancer. We identified four positive prognostic factors for PE in our cohort: primary exenteration, clear microscopic margins, no distant metastases and an age below 63 years.
Primary pelvic exenteration. There is a lack of data regarding the rationale for the choice of therapy in advanced pelvic malignancies. Therefore, primary exenteration in cervical, vulvar and vaginal cancer remains a matter of debate despite its proven curative effect (4, 6, 7, 13, 15). Nevertheless, the indication for any therapy in advanced pelvic malignancies has to be carefully assessed and patients thoroughly counseled as morbidity and therapeutic failures are possible in nonsurgical alternatives as well. Definite chemoradiation in patients diagnosed with FIGO stage VIA cervical cancer harbors the risk of fistulation in 22% to 47.8% (16, 17). Predictive factors for locoregional failure of definite chemoradiation in cervical cancer are: tumor size exceeding 5 cm, young age, non-squamous histology, positive lymph nodes, and being diagnosed with FIGO stage III or IV cervical cancer (16-22). Regarding chemotherapy and antiangiogenic therapy, the Moore criteria (pelvic disease, prior radiotherapy, age, race and ECOG above 1) are prospectively validated prognostic factors for survival in cervical cancer patients receiving chemotherapy and bevacizumab. It is important to consider that survival will decrease substantially the more of these factors accumulate (2, 3).
Neoadjuvant chemotherapy in vulvar cancer has shown favorable response rates of 80%. Unfortunately, optimal resection margins in order to omit mutilation are not defined yet as quick recurrences are possible (23, 24). Definite chemoradiation only shows low clinical response rates in vulvar cancer patients eligible for pelvic exenteration and a high morbidity (10, 25-29).
In FIGO stage III and IVA vaginal cancers, radiation therapy may lead to a 5-year disease-specific survival of 58%. Major morbidities will be seen in 21%, including bladder dysfunction, urethral stricture, and severe proctitis. However, the therapeutic success decreases significantly at tumor sizes above 4 cm (30).
In endometrial cancer distant disease will be seen more frequently than central recurrences only or a primary adjunct organ infiltrating tumor. Therefore, primary exenteration is rarely indicated in endometrial cancer patients. In selected patients, recurrent endometrial cancer treated with pelvic exenteration shows 5-year survival rates of 60% to 72% and a morbidity of 30% (7, 14).
Taken together, our median survival data of 36 months in the group of patients with primary pelvic exenteration is confirmed by others (4, 14, 15). Primary pelvic exenteration seems a reasonable consideration especially in cervical and vaginal cancer whenever the pelvic tumor size exceeds 4 to 5 cm (22, 30).
Clear microscopic margins and no distant metastases. The rate of tumor-free microscopic margins after pelvic exenteration ranges between 45% and 83% (4, 6, 7, 13, 14, 31). In lateral pelvic wall involvement this rate drops below 43% (13, 31). Whenever resection margins are affected by tumor the two-year survival rate decreases to 10% and less (4, 6, 15). In our cohort the median overall survival for patients with tumor affected resection margins was as low as 16 months. Unfortunately, the preoperative prediction of negative margins may fail in the clinical workup especially in cases of prior radiation therapy and lateral pelvic wall involvement (6, 13, 31-33). Many studies agree that tumor involved resection margins of the surgical specimen and distant metastases are negative prognostic factors for survival after pelvic exenteration (4, 6, 13, 31, 34, 35). Nevertheless, the ethical discussion about the value of pelvic exenteration in these situations is still ongoing as pelvic exenteration may relieve symptoms in the palliative setting (4, 6, 13, 31, 35). Next to clear microscopic margins, two studies proposed no lymphovascular space invasion as a significant prognostic factor for survival in pelvic exenteration for cervical cancer (6, 15). Further factors mentioned in these studies were no nodal involvement concerning an age of more than 43 years (15), and a curative intent (6).
Age. In general, age above 75 years is a risk factor for morbidity and mortality in extensive surgery (36). However, in advanced cervical cancer patients, young age is a risk factor for decreased survival. This may reflect an aggressive tumor biology in younger advanced cervical cancer patients (6). An age above 60 years may be considered as a risk factor for an increased perioperative morbidity but seems to have no influence on survival rates after pelvic exenteration (4, 15, 37, 38). In our cohort an age below 63 years was a significant factor with positive influence on survival.
Despite several limitations of our study such as its retrospective nature and the inclusion of several tumor entities, we were able to identify four prognostic factors favoring pelvic exenteration in advanced pelvic gynecologic malignancies with long-term survivorship. These factors may serve as prerequisites for a safe indication for pelvic exenteration.
Conclusion
The therapeutic potential of pelvic exenteration in advanced pelvic gynecologic malignancies is seen especially in patients under 63 years, whenever it is the primary treatment, in the absence of metastatic disease and whenever clear microscopic margins are accomplished. These four factors may serve as valuable prerequisites for the indication of pelvic exenteration as survival and morbidity in this group of patients compares favorably to alternative therapeutic options.
Footnotes
Authors’ Contributions
Study conception: Egger EK, Alexander Mustea. Study design: Egger EK, Alexander Mustea. Data acquisition: Hanna Liesenfeld, Florian Recker, Anna Döser. Quality control of data and algorithms: Dominique Könsgen. Data analysis and interpretation: Dominique Könsgen, Milka Marinova. Statistical analysis: Egger EK, T Hilpert. Manuscript preparation: Egger EK. Manuscript editing: Matthias Stope. Manuscript review: Jörg Ellinger, Milka Marinova, Alexander Mustea.
Conflicts of Interest
The Authors declare that no conflicts of interest exist with regard to the present study.
- Received March 26, 2021.
- Revision received April 11, 2021.
- Accepted April 30, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.