Pelvic Exenteration for Primary Advanced and Recurrent Vaginal Cancer: Clinical Outcome for 37 Patients

Introduction

Primary vaginal cancer is a rare gynecologic malignancy with an incidence of 1%–3%. Five-year survival rates are ∼40%–50%, with rates of 0%–35% in stage III/IV and of 12% in recurrent cases.^1,2 No standard treatment exists due to the rarity of the disease, especially in advanced and recurrent cases.^1,3 Most patients are treated by irradiation with or without chemotherapy, but the survival rates are still low and side effects high.^4–6 Therefore, radical surgery might be an alternative to the mentioned treatment to reduce iatrogenic long-term morbidity and in patients with central recurrence after irradiation.^3–5

Since the description of pelvic exenterations by Brunschwig in 1948, ⁷ the technique of and the indication for this radical surgery as well as the perioperative care have been improved immensely.^8–10 Therefore, survival rates increased, complication rates decreased, and the rates of pelvic exenterations rose.^8,10–13 Besides exenterations in a curative intention, exenterations in palliative intention have gained more acceptance by gynecologic oncologists and progressed to a proper treatment option.^8,10,14,15

To the best of our knowledge, no data exist for the outcomes of patients with pelvic exenteration for a homogeneous group of only vaginal cancer as a malignant entity. Therefore, we retrospectively analyzed the clinical outcomes of 37 women with primary advanced and recurrent vaginal cancer treated with pelvic exenteration.

Materials and Methods

We retrospectively analyzed the medical records of 37 patients with vaginal cancer according to the Declaration of Helsinki. All patients gave their informed consent for data evaluation. No further approval from the local ethical board was needed in our institution.

Exenteration was indicated when there was no other way to achieve clear surgical margins due to tumor extent in both primary and recurrent cases. All surgical procedures were performed at the Department of Gynecologic Oncology of the University Hospital of Erlangen and at the Department of Gynecology of the Hospital of Neumarkt according to current guidelines.^7–9 All patients had a preoperative clinical examination in general anesthesia to histologically reassure the type of tumor by biopsy, to evaluate the possibility of surgical cancer removal, and a cystoscopy or rectoscopy if indicated.^1,5 Abdominal and chest computerized tomography scans were initiated to assess the status of metastasis.^1,5

A pelvic exenteration was considered curative when the resection of the tumor was achieved with clear surgical margins, no perforation of the tumor in the pouch of Douglas, no cancer cells in the peritoneal lavage, and no distant metastasis were found. Otherwise it was palliative.

An anterior exenteration included the removal of the reproductive tract with the urinary bladder, the pelvic ureters, and the urethra. A posterior exenteration included the removal of the reproductive tract with the rectosigmoid colon. A total exenteration included the removal of both pelvic urinary tract and pelvic intestine.

If ever possible, a primary reanastomosis of the colon was done. For the remaining cases, a permanent colostomy was performed. A temporary colostomy was indicated due to prior irradiation or very low located anastomosis. Reconstruction of the urinary tract occurred by formation of a continent ileocecal bladder. Either the appendix or the last 10 cm of the ileum was used as a replacement for the urethra. Otherwise, an ileal or sigmoidal conduit or an ureterouretreostomy was performed. For vaginal reconstruction, the caudal 10 cm of the remaining sigmoid colon was used under preservation of the original blood supply and was rotated by 180°. The neovagina was fixed with a Prolene mesh to the os sacrum to prevent prolapse. Attention was also placed on an additional filling of the pelvis by an omental flap to reduce small bowel-associated morbidity and mortality. Lymphadenectomy was performed according to oncologic guidelines. If necessary, the exenterations and reconstructions were enlarged by performing a vulvectomy, additional bowel resections, anus-extirpations, or cutaneous flaps.

As severe complications, we recorded relaparotomies, secondary wound healing, fistulas, sepsis, thrombosis, insufficiency of anastomosis, and pulmonary embolism.

Data analysis was performed using IBM^® SPSS Statistics^® version 25 (Chicago, IL). Continuous variables are descriptively presented using mean and standard deviation, and categorical variables are presented as frequencies and percentages.

Results

The baseline, surgical, and tumor characteristics of the 37 patients with vaginal cancer treated by pelvic exenteration are presented in Table 1.

A total of 17 women (46%) received a primary and 20 women (54%) a secondary exenteration with a curative intention in 28 patients (76%). One of the patients with primary exenteration received a neoadjuvent chemotherapy before surgery because of a large tumor extent, which could not have been surgically removed without this prior treatment. The presurgical treatment of the patients with a secondary exenteration can be seen in Table 1.

A total exenteration was performed in 27 patients (72%), whereas 5 patients (14%) received an anterior or a posterior exenteration each. In 17 cases, a full reconstruction of the bowel continuity was achieved. For 20 of the 32 patients (63%) with a total or anterior exenteration, the urinary tract was reconstructed by formation of a continent ileocecal bladder with the appendix as replacement for the urethra. In the 12 remaining cases (37%), the last 10 cm of the ileum was used. In 10 cases (27%), an ileal or sigmoidal conduit was built and 2 patients (5%) received an ureterouretreostomy. A total of 23 patients (62%) also received a vaginal reconstruction. The exenteration and reconstruction were enlarged in 11 patients (30%) by performing a vulvectomy, in 7 patients (19%) by an additional bowel resection, in 6 patients (16%) by an anus extirpations, and in 5 patients (14%) by cutaneous flaps.

A complete removal of the tumor with histopathologic-free margins was achieved in 32 cases (86%), with 16 patients either in the primary or in the secondary exenteration group. In the patient with neoadjuvent chemotherapy before primary exenteration, an R0 resection could be achieved as well. A histologically proven R1 resection was present in five patients (14%), with one patient in the primary and four patients in the secondary exenteration group. An R1 resection was defined as detection of tumor cells or lymphatic space invasion at the margin of the specimen or a positive peritoneal lavage. A lymphatic space invasion within the tumor was detected in 18 patients (49%). Metastasis-free lymph nodes were found in 24 patients (65%), whereas 13 patients (35%) had metastasis as shown in Table 1.

The most common histologic type was the squamous cell carcinoma in 29 patients (78%). One patient had no tumor, but was operated due to a fistula formation after radiation therapy for vaginal carcinoma. The majority of patients had a cancer grade of 3 (65%) and a T4 stage (78%).

No perioperative or postoperative complications were reported for 17 patients (46%). One major complication was seen in 12 patients (32%) and two major complications were noted in 8 patients (22%). No patients had more than two major complications. Severe complications included 10 relaparotomies, 5 secondary wound healings, 2 rectovaginal fistulas, 2 sepsis, 2 thrombosis, 2 insufficiency of anastomosis, and 1 pulmonary embolism. Perioperatively, three patients (8%) died. Causes of death were sepsis, apoplexy, and thrombosis of the mesenterial vein once each. All of these patients were older than 62 years, and two of them had comorbidities.

The 5- and 10-year overall survival for all 37 patients was 45.3% each. The disease-free survival was likewise equal with 61.8% for 5 and 10 years, respectively. In patients with a curative exenteration, the 5-year overall survival was 57.7%, whereas that in those with a palliative exenteration was 12.6%. The group of patients with a primary exenteration had a 5-year overall survival of 69.6%, patients with a secondary exenteration had a 5-year overall survival of 29.7% instead.

The younger patients aged 26–60 years (n = 18) had a 5-year overall survival of 63.2%, whereas the older group aged 61–81 years (n = 19) had a 5-year overall survival of 32.3%. The best 5-year overall survival was found in patients with a primary exenteration in a curative intent and tumor-free lymph nodes with a rate of 88.9%. For patients with a secondary exenteration, the rate in a similar constellation was 42.6%.

At the time of evaluation, 21 out of 37 patients (57%) were still alive, 16 (43%) were deceased. Of those, seven (44%) died in succession of a local recurrence, four (25%) died because of distant metastasis, in two women (12%) the cause of death was not cancer associated, and three elderly patients with comorbidities died postoperatively, as mentioned before.

Discussion

Malignancies of the reproductive tract are very common in women. Nevertheless, the management in advanced primary or recurrent situations remains rather unclear and challenging, especially in rare gynecologic malignancies as the vaginal carcinoma. ¹⁶ Large primary or recurrent tumors are often not suitable for irradiation and have low response to chemotherapy. Therefore, pelvic exenteration in a curative or palliative approach might be the only possible option to improve patients' quality of life with good survival rates of up to 50% for mixed cohorts, although no detailed evidence about exenterative surgery is present.^5,16,17 Particularly, there is little data about homogeneous cohorts of single malignancies.

A strength of our study is that outcomes for only vaginal cancer are provided. There is no bias in outcomes regarding the impact of other cancer entities. Nevertheless, a limitation of the study is the retrospective character of the study and the relatively small cohort number. However, our cohort has, to the best of our knowledge, the largest number of patients published with advanced primary or recurrent vaginal cancer treated with pelvic exenteration in one center.

As stated before, one major issue of most of the yet published data regarding pelvic exenteration is the heterogeneity of patients in the studies. Mainly, patients with different gynecologic cancer types were analyzed together in the same cohort, so that these studies need to be interpreted with caution. Therefore, the differences in survival rates may be due to different cancer entities.

Therefore, our group recently published data for outcomes of pelvic exenteration for cervical and endometrial cancer separately. In patients with exenteration for endometrial cancer, the 5- and 10-year survival rates in cases of histologically free lymph nodes were 63.3% and 57%, respectively. ¹⁸ The overall survival rate of women with exenteration for endometrial cancer was 61.4% at 5 years and 51.1% at 10 years. ¹⁸ In patients with exenteration for cervical cancer, the 5- and 10-year overall survival rates were 41% and 36.9%, respectively. ¹⁹ For those patients with cervical cancer, who underwent a palliative exenteration, the rates were 19.1% and 17.7%, respectively. ¹⁹

For patients with vaginal cancer, the 5- and 10-year overall survival was 45.3% each in our cohort and, therefore, better than women with cervical cancer but worse than those with endometrial cancer. However, in patients with exenteration for vaginal cancer in a curative intention, the 5-year overall survival was 57.7%, whereas that in those with a palliative exenteration was 12.6%. The group of patients with a primary exenteration even had a 5-year overall survival of 69.6%, those with a secondary exenteration had a 5-year overall survival of 29.7% instead. The best subgroup of patients (primary exenteration in a curative intent and tumor-free lymph nodes) even had a 5-year overall survival of 88.9%.

As our rates mentioned show, there are big differences in survival rates according to the cancer stage and the type of exenteration (primary vs. secondary and curative vs. palliative). Therefore, mixed cohorts regarding these parameters are hardly compared and more studies for separated tumor entities and treatment cohorts are needed. Many patients who are nowadays treated with irradiation are still in early cancer stages. The disadvantage of such a procedure is that in cases of recurrence, there is no further possibility of another irradiation or at least in less extent. Therefore, a great advantage of primarily operated patients is the fact that irradiation is still available in cases of recurrence. In addition, as we could show, survival rates are much higher in cases of primary exenteration than in cases of secondary exenteration. Therefore, primary exenteration should always be discussed with patients as the first treatment option.

In the case of anterior exenteration, the entire lower urinary tract is removed together with the vagina by definition. There are cases in whom the tumor has a more favorable position with regard to less radical surgery, for example, at the level of the hymenal hem. Tumors located further caudally count as vulva carcinomas. Less radical surgery is also possible for tumors that grow rather exophytically, as the lower infiltration more likely enables an R0 resection with preservation of the adjacent organs. A partial resection of the caudal part of the urethra of up to 5 mm is possible in most cases without loss of function. In contrast, a wider resection, both of the urethra and of the trigonum or of the lower part of the bladder, results in significant functional loss, which significantly reduces the patient's quality of life and endangers her from recurrent infections. In such cases, complete anterior exenteration with reconstruction should be considered. However, patients with partial bladder or urethra resections were not considered in our report. The least radical type of exenteration is a posterior exenteration. In this study, reconstruction can often be performed using a colorectal or coloanal anastomosis. Especially in primary cases without prior radiation in this area, the postoperative course is rather uncomplicated. In contrast, patients after radiotherapy have a high rate of anastomotic insufficiency, so they always receive a protective ileostomy in first line, which can be removed though after 6–8 weeks in most cases.

However, patients have to be informed about the side effects and complication rates of a pelvic exenteration. In the first reports of Brunschwig about pelvic exenteration, mortality rates were as high as 23% and long-term survival was low. ⁷ In times of improved patient selection, surgical and reconstructive techniques, and perioperative care, mortality and morbidity with the procedure have immensely decreased, with mortality rates from 0% to 16% and a 5-year overall survival of 35%–70%.^{8,10,11,13,14,18–30} Nevertheless, the rate of perioperative morbidity, major postoperative complications, and long-term complications are still ∼40%–70%, 22%–32%, and 20%–48%, respectively.^{8,10,11,13,14,19,21,22,24,25,29,30} One reason for these still high rates of complications is the change in patient characteristics over time toward increasing rates of exenterations in older and more obese women with higher comorbidities. ²¹ In our cohort of women with only vaginal cancer and with 86% of these with an advanced cancer stage III or IV, we found a 5-year overall survival rate of 45%. The best subgroup had a 5-year survival rate of even 88.9%.

The most frequent postoperative complications after exenteration described in the literature are infections with a rate of 19%–86%, followed by anastomotic leaks and fistulas with 8%–36% and small bowel and ureteral obstructions with 5%–10%.^{9,13,24,28,30} In our cohort, 17 women (46%) had no perioperative or postoperative complications. The remaining 20 women (54%) suffered from 1 (12 patients, 32%) or 2 major complications (8 patients, 22%), including 10 relaparotomies, 5 secondary wound healings, 2 rectovaginal fistulas, 2 sepsis, 2 thrombosis, 2 insufficiency of anastomosis, and 1 pulmonary embolism. Three patients (8%) died perioperatively of sepsis, apoplexy, or thrombosis of the mesenterial vein, but all were elderly and with comorbidities.

Besides, quality of life after pelvic exenteration is an important issue, with different satisfaction rates and subjective outcomes of patients.^31,32 For sure, the outcome for palliative exenterations is worse, with higher morbidity and markedly lower survival rates.^18–20 However, even with this approach, quality of life is improved in 50% of patients and it might be their only potential chance for improved survival.^11,20 In our cohort, patients with palliative exenteration had a long-term survival rate of 12.6%.

Other therapy options, such as standard irradiation or chemotherapy or a combination of them, merely have overall survival rates of a few months combined with severe adverse side effects.^3,8 However, the treatment approach of external beam irradiation and intravaginal brachytherapy seems to be more promising for the last years, even in advanced or recurrent cases with 5-year survival rates from 34% to 58% but partly severe side effects.^2,3,33,34 Nevertheless, there are no studies comparing pelvic exenteration with irradiation or chemotherapy regimens in advanced or recurrent cases. As we have already shown, the survival rates after a primary exenteration are significantly better than after a secondary exenteration, that is, after radiochemotherapy. Therefore, we would prefer the surgical approach in the primary setting, given the patient characteristics already mentioned. Nevertheless, there might be cases in which surgery is declined by the patient or appears to be too risky due to comorbidities. For these patients, we would recommend primary radiochemotherapy. In addition, in cases of very extensive tumor findings, where the surgical approach might be less promising in first line but who meet the criteria for surgery in second line, neoadjuvant chemotherapy can also be considered.

For sure, it might be worth to carefully select patients for pelvic exenteration in cases of vaginal cancer. Unfortunately, little is known about prognostic factors for women with that type of cancer. We are not able to determine any in our cohort due to the small number of patients. Some negative prognostic factors are known from studies with cohorts of different cancer entities. These factors are elderly patients, tumor size >4–5 cm, higher cancer stage, positive histopathologic resection margins, positive lymph nodes (especially para-aortic nodes) or transperitoneal spread, lymphovascular invasion, histologic type, and total pelvic exenteration compared with solely anterior or posterior one.^{2,5,6,8,23–27,35} Most likely, these factors are also suitable for vaginal cancer. In our study, women with a primary exenteration in a curative intent and tumor-free lymph nodes had the best 5-year overall survival with a rate of 88.9%.

Conclusion

In cases of advanced or recurrent vaginal cancer, a surgical approach in form of a pelvic exenteration seems to be a justifiable treatment option, especially in cases wherein an R0 resection can be achieved, or in recurrent cancers, wherein another irradiation would cause major side effects or is impossible after former irradiation. With the current surgical and reconstruction techniques and the improvements of perioperative care, outcomes of patients has improved tremendously. However, complication rates of exenterations are still high and not negligible and need to be intensively discussed with patients.