Fax: (507) 284-1803
Predictors of survival from urachal cancer
A mayo clinic study of 49 cases
Article first published online: 11 OCT 2007
Copyright © 2007 American Cancer Society
Volume 110, Issue 11, pages 2434–2440, 1 December 2007
How to Cite
Molina, J. R., Quevedo, J. F., Furth, A. F., Richardson, R. L., Zincke, H. and Burch, P. A. (2007), Predictors of survival from urachal cancer. Cancer, 110: 2434–2440. doi: 10.1002/cncr.23070
Fax: (507) 284-1803
- Issue published online: 19 NOV 2007
- Article first published online: 11 OCT 2007
- Manuscript Accepted: 19 JUN 2007
- Manuscript Revised: 18 JUN 2007
- Manuscript Received: 24 APR 2007
- urachal cancer;
- radiation therapy;
Outcome results of a long-term analysis of urachal cancer using a new staging system are presented.
The authors analyzed clinical outcomes from 49 patients with the diagnosis of urachal cancer who were seen at the Mayo Clinic, Rochester, Minnesota from 1950 to 2003. The TNM staging system was used to predict outcome after surgical resection.
Among 49 study patients, 33 were men, 16 were women, and their median age at presentation was 57.5 years. The vast majority of tumors were adenocarcinomas (89%), 4% were sarcomas and transitional cell carcinomas, and the rest were high-grade mixed neoplasms. Among the adenocarcinomas, 63.6% were mucin-producing tumors. Partial cystectomy with or without pelvic lymph node dissection and removal of the urachus was performed in 41 (83%) cases. Overall survival for all stages was 62 months with 17 (34%) patients still alive more than 5 years after treatment. Applying the TNM staging system, the authors demonstrated a median survival time for stage I/II patients of 10.8 years (95% CI, 6.9 years to 12.0 years) compared with a median survival of 1.3 years (95% CI, 1.1 years to 1.9 years; log-rank P<.0001) for patients with advanced disease (stages III and IV).
Stage at presentation by the TNM staging system proved to be the main predictor of outcome after surgery for urachal cancer. Better systemic modality treatments are needed for advanced stages of this disease. Cancer 2007. © 2007 American Cancer Society.
In our institution, urachal neoplasms represent a small fraction (0.22%) of primary bladder cancers or about 1 case per year.1 Currently, there are less than 350 cases of urachal cancer reported in the English-language medical literature with most of them being case descriptions. Several of these reports have associated urachal cancer with poor prognosis. However, we have reported a 5-year survival rate of 43%, which has been recently confirmed by a study from the M. D. Anderson Cancer Center (Houston, Tex).1, 2
Surgical resection in the form of partial (segmental) or radical en bloc cystoprostatectomy is the main form of treatment. However, similar results are seen with a conservative surgery that involves partial cystectomy with umbilicotomy and removal of the urachus as proposed by us.2 The role of chemotherapy and radiation therapy for the treatment of urachal cancer is unclear, although Siefker-Radtke et al. showed that chemotherapy can induce objective response in some cases.2
Because of the uncommon nature of this cancer, very few studies have investigated prognostic factors that will allow clinicians to determine outcome and risk for relapse. In 1984, Sheldon proposed a staging system for urachal cancer and reported poor outcomes after surgical resection.3 Several studies have failed to find prognostic factors of outcome using the Sheldon's staging system or biologic markers, such as histological differentiation and size and even molecular markers.4–6 For example, analysis of DNA status, proliferating cell nuclear antigen, and argyrophilic nucleolar-organizer region (AgNOR) failed to predict survival.4 On the basis of these results, this study was designed to find a potential way to stratify patients with risk of recurrence that would require closer follow-up or more aggressive treatment.
In the current study, we reviewed our experience with urachal cancer and applied the TNM (American Joint Commission on Cancer) staging system to predict outcomes after surgical resection of urachal neoplasms.
MATERIALS AND METHODS
Our article presents the results of our experience with urachal cancer during the last 53 years (1950- 2003). This study has been reviewed and approved by the Mayo Foundation Institutional Review Board. We performed a retrospective review of all cases of urachal cancer (n = 49) seen at our institution since 1950. For each case, we abstracted demographic information as well as information on symptoms at presentation, diagnosis, histological features, treatment, and follow-up. For cases that presented after the establishment of the Mayo Clinic Tumor Registry in 1975, we searched for the diagnosis of urachal cancer. For the years prior to 1975, the diagnosis of bladder cancer was screened for features of urachal cancer on the basis of modifications of criteria initially described by Mostofi7 and associates as follows: 1) tumor location at the dome of the bladder or in the path of the urachus, 2) urachal remnants in association with tumor in the pathology specimen, 3) pathology report of predominant invasion of the muscularis or deeper tissue with sharp demarcation between the tumor and the surface of the epithelium, 4) tumor extension into bladder, urachus, anterior abdominal wall, or Retzius space and, finally, for us, the finding of adenocarcinoma with mucinous features was considered highly suggestive of urachal origin. For tumors located outside the bladder, the diagnostic requirements included 1) tumor location at the dome of the bladder or in the path of the urachus, 2) urachal remnants in association with tumor in the pathology specimen.
The TNM staging system was used to determine prognosis. The TNM staging system was applied as follows: Tis is a tumor localized to the urachal mucosa with no invasion to the basal membrane (carcinoma in situ). T1 is a tumor with invasion through the basal membrane. Because of findings that urachal tumors can present in any place along the path of the urachus from the umbilicus to the bladder and that these tumors are actually extravesical cancers with a tendency to invade into the bladder, T2 is divided into pT2a, tumor invades deep muscle (outer half); pT2b, tumor invades the superficial muscle of the bladder (inner half); T3, tumor invades perivesical fat, abdominal wall muscle (in cases of extravesical urachal tumors). Involvement of regional lymph nodes followed the traditional TNM staging system. Peritoneal implants were considered metastasis or stage IV. The TNM staging system differs from the one described by Sheldon et al.3 and was used for this study because of its consistency and universal application. Survival data were analyzed by the Kaplan-Meier method with the log-rank univariate test.
Review of medical records of our 49 cases of urachal cancer showed that the median age of presentation was 57.5 years (range, 42 years to 76 years). Men represented the majority of cases (68%). The vast majority of our cases presented in white patients (94%), although this may represent referral biases. Tobacco exposure was documented in 63% of these cases (Table1).
|Age, y, [mean]||[57.5]|
Symptoms and Findings
Hematuria that required the patient to seek medical attention was the predominant finding on presentation (85%), followed by mucinuria (17%), a palpable mass in the lower abdomen (17%), and bacteriuria and pain with 8.5% each (Table 2). Findings described in the past, such as umbilical discharge, were not seen in any of our cases.8 All patients underwent cystoscopy to assess local extension to the bladder. The main cystoscopic finding was a mass seen at the dome of the bladder or anterior wall (91.5%). This mass was often described in appearance as a polypoid or ulcerated lesion. Location of the tumor in the dome or anterior wall of the bladder was considered as one of the critical factors in determining urachal origin. Imaging techniques included plain films and cystograms that were performed to look for filling defects in the bladder or calcification at the bladder dome, findings described by others but not seen in our series.9 After they became available, ultrasound (US) and computed tomography (CT) have proven to be good tests for diagnosing urachal tumors (Figs. 1 and 2).
|Mean duration of symptoms||13.3 wk|
|Mass at dome of bladder||91.5%|
Review of pathology reports revealed adenocarcinoma as the predominant type of tumor (89.7%). Most (63.6%) of these adenocarcinomas are mucin-producing. Small groups of patients had features of adenocarcinoma and transitional cell carcinoma (TCC) (4.5%), or adenocarcinoma plus TCC (4.5%), and small cell carcinoma (2.2%). Pure sarcomas and TCC each represented 8% of the cases with the remaining 2% of tumors being highly undifferentiated neoplasms (Table 3).
|Histology||No. of cases||Percentage|
|Adeno without mucin||13||29.7|
Sites of local recurrences included pelvic lymph nodes and peritoneum or omentum. Distant metastasis included lung (50%), lymph nodes (46%), bone (30%), intestine (30%), brain (20%), and liver (16%). Metastasis to multiple organs was a common feature for advanced stages.
Most of patients were treated with surgery (95%). In the remaining 5%, the tumor was found to be unresectable. Partial cystectomy with umbilicotomy and removal of the urachus was the main surgical approach (91.5%). Cystoprostatectomy was performed in 8.5% of cases (Table 4).
|Partial cystectomy (PC)||87|
|PC+umbilectomy with removal of the urachus||59|
|PC+umbilectomy with removal of the urachus+lymph node sampling||25|
Ten of 49 patients received treatment with chemotherapy for locally advanced or metastatic disease. Several chemotherapy regimens were used during this period of time (1950–2003). These regimens included single-agent 5-fluorouracil (5-FU) (2 cases), 5-FU and cisplatin (1 case), 5-FU, lomustine and vincristine (1 case), taxol and cisplatin (2 cases), platinum and etoposide,1 and MVAC (methotrexate, vinblastine, doxorubicin, cisplatin) alone or plus radiation therapy in 3 cases (Table 4). The median survival after chemotherapy was 20.4 months (range, 4 months to 36 months). Progressive disease was seen when single-agent 5-FU was used. Conversely, platinum-containing regimens produced stable disease or partial response in 71% (5 of 7 cases) of those treated with a platinum-containing regimen. The combination of 5-FU, CCNU (lomustine), and vincristine resulted in stable disease for 22 months in 1 patient.
Radiation therapy was given as adjuvant treatment in 10 of 49 cases, as treatment for local recurrence in 1 case, and for control of metastatic disease in 1 case (Table 4). Median survival after radiation was 19.5 months (range, 4 months to 28.5 months). When radiation was given with chemotherapy, the median survival was 21 months (range, 6 months to 32.5 months).
By using the TNM staging system, we were able to analyze differences in survival between local disease and stages I and II and between locally advanced and metastatic disease in stages III and IV. Our results demonstrated that the median survival time for stage I/II patients was 10.8 years (95% CI, 6.9 years to 12.0 years) compared with a median survival of 1.3 years (95% CI, 1.1 years to 1.9 years) (log-rank P < .0001) for patients with metastatic disease (stage IV).(Table 5 and Figs. 3 and 4).
|Stage||Survival time, y|
The urachus is present in 100% of children at birth and can be seen as a closed epithelial canal in about 30% of adults.10–11 The urachus is the embryonic remnant of the involution of the allantois and the ventral cloaca.10–13 During early embryonic development (Days 16 to 24), the allantois, which is derived from the vitelline sac, opens into the upper part of the ventral or anterior cloaca forming the allantois channel.11 By Week 5, the bladder is formed from the medial part of the allantois, and the rest, or the cloaca, narrows into a channel forming the urachus. Although some controversy existed about the embryologic origin of the urachus as being derived from the allantois or the cloaca, recent evidence supports its origin in the cloaca.11, 12
In this series, most urachal cancers were adenocarcinomas (89%; Table 3) However, it should be noted that not all adenocarcinomas of the bladder are urachal cancers, in fact, only 20% to 40% of bladder adenocarcinomas are of urachal origin.14, 15 Other histological types of tumors are also seen involving the urachus, including sarcomas (leiomyosarcoma, rhabdomyosarcoma, and malignant fibrous histiocytoma [MFH]), small cell carcinomas, transitional cell cancer, and mixed neoplasias.15 In the current article, these tumors represent 10% of our cases. Tumors with mixed histology, such as transitional cell and adenocarcinoma, and adenocarcinoma and small cell were also seen (Table 3). To make a definite diagnosis of urachal cancer when a tumor is outside of the bladder, we must emphasize the importance of the anatomic location of the tumor and the discovery of remnants of the urachus. Moreover, the finding of an adenocarcinoma with mucinous features in the trajectory of the urachus, once metastasis from another organ is excluded, should suggest the diagnosis of urachal cancer. For tumors located within the bladder, the main criteria for determining urachal origin are tumor location at the dome or anterior wall of the bladder and the discovery of remnants of the urachus. Other features, such as finding a sharp demarcation between the tumor and the surface epithelium free of glandular or polypoid proliferation, are also suggestive but not always required for diagnosis.
Adenocarcinoma with mucinous features was a very common finding in our cases of adenocarcinoma of the urachus (63.6%). The so-called signet-ring type, in which a cell contains a mucin-filled vacuole that displaces the hyperchromic nucleus to one side, was also seen in the group of mucin-producing tumors. The histologic features of these tumors support the theory that urachal adenocarcinomas are derived from glandular intestinal epithelium derived from remnants of cloaca or enteric inclusions. To further support this theory, a study by Pantuch et al. showed that the monoclonal antibody 7E12H12 developed against a colonic epithelial protein and reacted positively in immunoperoxidase assays of urachal cancer samples.17 The similarity to the colonic epithelium as well as the clinical behavior of these tumors also suggest that urachal cancers are histologically and embryologically more closely related to colon cancer than to the typical transitional cell bladder cancer.
Urachal tumors seem to spread by local extension to involve pelvic structures and then by way of hematogenous dissemination to other organs. In this series, the presence of peritoneal implants was associated with poor prognosis. When resecting these tumors, surgeons should perform detailed exploration of the abdomen looking for peritoneal implants and include sampling of lymph nodes to allow a better staging of the case. Sites of regional recurrence after surgery included pelvic lymph nodes, peritoneum, and omentum. Distant metastasis included lung (50%), lymph nodes (46%), bone (30%), intestine (30%), brain (20%), and liver (16%).
In 1984, Sheldon proposed a staging system for urachal cancers.3 However, the Sheldon staging system does not account for the finding that urachal tumors can present in any place along the path of the urachus from the umbilicus to the bladder and that these tumors are actually extravesical cancers with a tendency to invade the bladder.3, 7, 15, 18, 19 When this tumor is found close to the umbilicus, involvement of the abdominal wall is likely (Fig. 1), conversely, if the tumor is close to the bladder, invasion of the bladder is a common event (Fig. 2). The Sheldon system also assumes that involvement of regional lymph nodes is equivalent to distant metastasis. We implemented the TNM staging system and proved that anatomic location can be extrapolated to prognosis after surgical resection. Our data show that if the stage of the tumor at the time of surgery is III or less, then these patients have an overall good prognosis with resection alone (Fig. 4 and Table 5).
Survival data were available in 42 of 49 patients. Our overall median survival for all stages was 62 months (Fig. 4). Determination of prognosis based on the TNM staging system shows that when the tumor is confined to the urachus, stage I, survival is more that 10 years, and survival is 7.5 years for stage II (Table 5). There is, however, a significant drop in life expectancy if the patient has involvement of regional lymph nodes, peritoneal implants, or extension to local viscera, (Fig. 4, Table 5). Overall survival for distant metastatic disease is less than 1 year. Patients with early disease, stages I through II, have a statistically significant difference in survival (P < .0001) compared with late disease, stage IV (Fig. 4). This difference is more impressive when we look at the life expectancy in years for early disease, more than 12 years, compared with advanced disease, 1.7 years. The significant difference in survival between early and advanced disease is an indication that patients with advanced disease on presentation, stage IV, should be treated more aggressively.
At our institution, patients with localized urachal cancers undergo partial cystectomy with umbilicotomy and resection of the urachus. Data from our institution, confirmed by others, have shown that partial cystectomy with umbilicotomy is associated with the same prognosis and better quality of life as more radical surgery.1, 2 Most cases presented in our series underwent surgery, 83%, with 91.5% of them having partial cystectomy with umbilicotomy and resection of the urachus. Our 5-year survival of 62 months validates this approach and is similar to another series.2
Systemic therapy in the form of chemotherapy has been indicated for advanced or metastatic urachal tumors.2, 19 In our series, chemotherapy was given to 20% of the cases, usually for recurrent or metastatic disease with a median overall survival of 20.4 months. Partial response was seen associated with platinum-based regimes. Radiation therapy was given to 26% of the cases, usually for pelvic disease or for palliation of metastasis with an overall median survival of 19.5 months.
We have shown that the TNM staging system is a good predictor of survival for urachal cancer. Patients who present with early disease confined to the urachus, stages I and II, have a good survival when treated with partial cystectomy, umbilicotomy, and urachal resection. However, the prognosis drastically changes for patients that present with advanced stages of this disease.
- 13Diseases of the Umbilicus. Philadelphia, Pa: WB Saunders; 1916..