Transitional cell carcinoma is the most common cancer of the urinary tract in dogs.1 Transitional cell carcinoma has been reported to metastasize to lymph nodes and distant organs in approximately 50% of dogs. Common distant sites of metastases include lung, liver, spleen, and less commonly bone and other organs.1 Transitional cell carcinoma has also been reported in the abdominal wall in dogs. In this location, the cancer is thought to have been implanted during procedures such as surgery to obtain biopsy specimens, tumor removal, tumor debulking, or cystotomy tube placement or fine-needle aspiration of TCC in the urinary bladder, urethra, and prostate.2–4
Information is limited regarding the biological behavior, possible causes other than seeding during medical procedures, and response to treatment for ABWTCC. Although most cases of ABWTCC reported to date have been associated with probable seeding through medical procedures,2–4 it is possible that ABWTCC could result from natural spread of the cancer. Regarding the response to treatment, some oncologists have the impression that ABWTCC responds poorly to medical treatment, but this has not been studied or confirmed.
Another aspect of ABWTCC that has not been defined is the potential expression of UPIII. Uroplakin III is a transmembrane protein present in the superficial transitional epithelial cells in the urinary tract.5 Uroplakin III is expressed in > 90% of TCC of dogs and has been regarded as specific for TCC, although possible expression in prostate cancer has recently been reported.5,6 Should UPIII be expressed in ABWTCC, this would be important for dogs with abdominal wall masses of unknown origin, including those with ABWTCC that do not yet have clinical signs associated with the urinary tract cancer. In these cases, finding UPIII expression in the abdominal wall mass would serve as an indication to include thorough assessment of the urinary tract in evaluation of these dogs. The goals of the study reported here were to determine the expression of UPIII in ABWTCC, identify possible etiologic factors leading to ABWTCC development, and further characterize the biological behavior of ABWTCC and its response to medical treatment.
Materials and Methods
Case selection—A search of the Purdue University Veterinary Teaching Hospital medical records system between July 1, 1985, and December 31, 2010, was conducted to identify dogs with TCC of the urinary tract. Of dogs with TCC, those with histopathologically confirmed ABWTCC that involved the subcutaneous tissues and muscle were further identified for inclusion in the study.
Medical records review—Medical records, radiographic and ultrasonographic images, and pathology reports were reviewed to obtain the following information: age of the dog at the time of diagnosis of TCC, sex, breed, TNM stage (following World Health Organization staging criteria7) at the time of diagnosis of TCC of the urinary tract and at the time of death, prostatic and urethral involvement of the cancer, grade of the primary tumor and ABWTCC lesion, location of the cancer within the abdominal wall, history of any procedures (surgery, percutaneous biopsy, or fine-needle aspiration) that could have resulted in seeding of the cancer, time from these procedures until the ABWTCC was found, time from the ABWTCC detection until death, response to any treatment given for the ABWTCC, and necropsy findings if a necropsy was performed. Evidence for the potential lymphatic spread of TCC to the abdominal wall was assessed through review of antemortem radiographic and ultrasonographic images, gross necropsy findings, and histologic examination findings. Further characteristics of ABWTCC and the biological behavior and response of ABWTCC to treatment were determined from information in the medical records and review of ultrasonographic and radiographic images.
Statements in the medical records related to comments made by pet owners regarding the effects of the ABWTCC on quality of life were also recorded. Information on tumor size included caliper measurements of external palpable lesions and measurements of internal masses from ultrasonographic and radiographic images. Over the course of the study period, urinary bladder masses would have been measured by contrast cystography or 2-D ultrasonography and tumor volume calculated as described.8,9 Tumor response was classified as follows: complete remission, complete clinical resolution of the cancer; partial remission, ≥ 50% reduction in tumor volume; stable disease, < 50% change in tumor volume; and progressive disease, ≥ 50% increase in tumor volume or the development of new tumor lesions. Available tissues from the primary tumor and ABWTCC were obtained for histologic evaluation and IHC.
IHC—Immunohistochemistry was used to detect UPIII and Prox1 expression on available tissues.5,10,11 Prospero homeobox protein 1, which is expressed in endothelial cells in lymph vessels, was used to identify lymphatic vessels.10,11 The presence of tumor cells within these vessels was considered an indication of lymphatic invasion and the potential for lymphatic spread of the cancer.
For IHC, briefly, 5-μm-thick paraffin sections were dewaxed and hydrated. Antigen retrieval for UPIII was done with proteinase K. For Prox1, antigen retrieval was done by use of a decloaker with slides immersed in citrate buffer at a pH of 6.0. Slides with monoclonal antibody to UPIIIa diluted at 1:50 were incubated at room temperature (approx 22°C) for 90 minutes. Slides with a rabbit polyclonal antibody to Prox1b diluted at 1:800 were incubated at room temperature for 90 minutes. An immunoperoxidase method for UPIIIc and Prox1d with diaminobenzidine as chromogen was used to demonstrate the immunologic reaction.
Positive control tissues for UPIII and Prox1 detection were normal urinary bladder and lymph node tissues, respectively. To demonstrate the specificity of the immunologic reaction, the primary antibody was replaced by a nonimmune antibody from the same species. The immunologic reaction for UPIII was graded on the basis of the approximate percentage of neoplastic positive cells with immunoreactivity as follows: 0 (no reactivity), 1+ (< 25% of cells), 2+ (25% to 50% of cells), and 3+ (> 50% of cells). For Prox1, the presence or absence of reactivity was recorded indicating the presence or absence of lymphatic vessels in the sections studied. The presence or absence of tumor cells within lymphatic vessels was also recorded. Uroplakin III was detected in the cell membrane or in the cytoplasm; Prox1-specific expression was in the nucleus.
Statistical analysis—Urinary bladder surgery as a potential etiologic risk factor for ABWTCC was investigated by analyzing the association between history of cystotomy and the development of ABWTCC. All dogs with TCC during the study period were categorized into those that had undergone a cystotomy and those that had not undergone a cystotomy. The association between history of cystotomy and the development of ABWTCC was evaluated via the χ2 test of independence.e Values of P < 0.05 were considered significant.
Results
Cases, clinical findings, and IHC characteristics—Review of medical records during the period between July 1, 1985, and December 31, 2010, resulted in identification of 544 dogs with histopathologic diagnosis of TCC of the urinary tract. Twenty-four of these dogs also had histopathologically confirmed ABWTCC.
For the 24 dogs, median age at the time of diagnosis of TCC was 11 years (range, 4 to 15 years). Dogs included 8 mixed-breed dogs, 6 Shetland Sheepdogs, 2 Scottish Terriers, and 1 each of the following breeds: Coonhound, Shih Tzu, Golden Retriever, Standard Poodle, Beagle, Lhasa Apso, German Shepherd Dog, and Portuguese Water Dog. There were 11 spayed female, 1 sexually intact female, and 12 neutered male dogs. Necropsy was performed on 15 of the 24 dogs. The ABWTCC lesions varied considerably in size. The smallest lesions were < 2 cm in diameter. The most extensive lesions were > 20 cm in diameter and covered the entire ventral aspect of the abdomen and thorax. The tumor characteristics including IHC findings were summarized (Tables 1–3). Uroplakin III was expressed in 19 of 20 primary tumors and in 17 of 17 ABWTCCs for which tissue samples were available (Figure 1). Lymphatic vessels (those with Prox1 expression) that contained TCC cells were identified in 1 of 20 primary tumors and 5 of 17 ABWTCCs for which tissue samples were available.
Tumor stage following World Health Organization staging criteria and distant site metastasis status at the time of diagnosis and at death in 24 dogs with TCC of the urinary tract that had histopathologic confirmation of ABWTCC.
| Variable | At TCC diagnosis | At death |
|---|---|---|
| TNM tumor stage | ||
| T2 | 20 | 15 |
| T3 | 4 | 9 |
| N0 | 20 | 7 |
| N1 | 4 | 17 |
| Distant site metastasis* | ||
| No | 20 | 8 |
| Yes | 4 | 16 |
Data are number of dogs.
Distant sites other than abdominal wall.
N0 = No evidence of lymph node metastasis. N1 = Lymph node metastasis detected. T2 = Tumor invading bladder wall with induration. T3 = Tumor invading neighboring organs (prostate, uterus, vagina, or pelvic canal).
Urinary tract involvement at the time of diagnosis and location of the ABWTCC in the same dogs as Table 1.
| Variable | No. of dogs |
|---|---|
| Urinary tract involvement at diagnosis | |
| Urethra | |
| Yes | 9 |
| No | 15 |
| Prostate* | |
| Yes | 5 |
| No | 7 |
| Location of ABWTCC | |
| Ventral midline | 17 |
| Away from midline | 7 |
Twelve male dogs.
Immunohistochemical findings in the same dogs as Table 1.
| Variable | Primary TCC | ABWTCC |
|---|---|---|
| Tumor grade | ||
| 1 | 0 | 0 |
| 2 | 3 | 0 |
| 3 | 20 | 16 |
| NA | 1 | 8 |
| UPIII expression | ||
| 1+ | 0 | 5 |
| 2+ | 1 | 3 |
| 3+ | 18 | 9 |
| Negative | 1 | 0 |
| NA | 4 | 7 |
| TCC cells in Prox1-positive vessels* | ||
| Positive | 1 | 5 |
| Negative | 19 | 12 |
| NA | 4 | 7 |
Data are number of dogs.
Transitional cell carcinoma cells detected in lymphatic vessels.
NA = Tissues or slides not available.
Representative photomicrographs of sections of various tissues from a dog with TCC of the urinary tract and ABWTCC. Immunohistochemistry was performed for the detection of UPIII and Prox1 in affected tissues. Immunoperoxidase-diaminobenzidine stain with hematoxylin counterstain. A—Photomicrograph of a section of the urinary bladder. Intravascular neoplastic cells (t) are positive for UPIII. Bar = 80 μm. B—Photomicrograph of a section of the abdominal wall. Clusters of neoplastic cells (t) embedded in the abdominal wall strongly express UPIII. Bar = 800 μm. C—Photomicrograph of the urinary bladder, which is in the same field as in panel A. The nuclei (arrowheads) of endothelial cells lining a vessel are positive for Prox1, identifying this as a lymphatic vessel. Neoplastic cells (t) are not stained. Bar = 80 μm. D—Photomicrograph of a lymph node. The nuclei of endothelial cells lining the subcapsular sinus are positive for Prox1 (arrowheads). Neoplastic cells (t) are not stained. L = lymphoid tissue. Bar = 80 μm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Representative photomicrographs of sections of various tissues from a dog with TCC of the urinary tract and ABWTCC. Immunohistochemistry was performed for the detection of UPIII and Prox1 in affected tissues. Immunoperoxidase-diaminobenzidine stain with hematoxylin counterstain. A—Photomicrograph of a section of the urinary bladder. Intravascular neoplastic cells (t) are positive for UPIII. Bar = 80 μm. B—Photomicrograph of a section of the abdominal wall. Clusters of neoplastic cells (t) embedded in the abdominal wall strongly express UPIII. Bar = 800 μm. C—Photomicrograph of the urinary bladder, which is in the same field as in panel A. The nuclei (arrowheads) of endothelial cells lining a vessel are positive for Prox1, identifying this as a lymphatic vessel. Neoplastic cells (t) are not stained. Bar = 80 μm. D—Photomicrograph of a lymph node. The nuclei of endothelial cells lining the subcapsular sinus are positive for Prox1 (arrowheads). Neoplastic cells (t) are not stained. L = lymphoid tissue. Bar = 80 μm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Representative photomicrographs of sections of various tissues from a dog with TCC of the urinary tract and ABWTCC. Immunohistochemistry was performed for the detection of UPIII and Prox1 in affected tissues. Immunoperoxidase-diaminobenzidine stain with hematoxylin counterstain. A—Photomicrograph of a section of the urinary bladder. Intravascular neoplastic cells (t) are positive for UPIII. Bar = 80 μm. B—Photomicrograph of a section of the abdominal wall. Clusters of neoplastic cells (t) embedded in the abdominal wall strongly express UPIII. Bar = 800 μm. C—Photomicrograph of the urinary bladder, which is in the same field as in panel A. The nuclei (arrowheads) of endothelial cells lining a vessel are positive for Prox1, identifying this as a lymphatic vessel. Neoplastic cells (t) are not stained. Bar = 80 μm. D—Photomicrograph of a lymph node. The nuclei of endothelial cells lining the subcapsular sinus are positive for Prox1 (arrowheads). Neoplastic cells (t) are not stained. L = lymphoid tissue. Bar = 80 μm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Possible etiologic events—Possible events that could have allowed seeding of the TCC in the abdominal wall were known for 20 of 24 dogs. Possible inciting events included the following: laparotomy and cystotomy (17 dogs), limited laparotomy and cystotomy tube placement (1), ultrasound-guided fine-needle aspiration of the prostate (1), and ultrasound-guided fine-needle aspiration of the urinary bladder (1). Of the 18 dogs that had cystotomy performed, the urinary bladder had ruptured prior to surgery in 2 dogs, and free urine was present in the abdominal cavity at the time of surgery. In 1 dog in which a cystotomy tube had been placed, there was leakage of urine around the tube. During the study period, 177 dogs with TCC had undergone cystotomy, and 18 (10.2%) of these dogs developed ABWTCC. During the study period, 367 dogs with TCC had not undergone cystotomy, and 6 (1.6%) of these dogs developed ABWTCC. Significantly (P < 0.001) more dogs that had surgery, compared with those that did not have surgery, developed ABWTCC. Transitional cell carcinoma of abdominal wall developed in 5 of 74 (6.8%) dogs undergoing cystotomy between July 1, 1985, and December 31, 1999. Transitional cell carcinoma of the abdominal wall developed in 13 of 103 (12.6%) dogs undergoing cystotomy between the later period of January 1, 2000, to December 31, 2010. Four dogs with ABWTCC had no known inciting event.
For dogs in which a possible seeding event was known, the median time between the event and the detection of ABWTCC was 121 days (range, 11 to 358 days). The median time from detection of ABWTCC until death in all dogs was 57 days (range, 0 to 324 days). The median time between diagnosis of TCC of the urinary tract and death in all dogs was 231 days (range, 0 to 486 days). In some dogs, the inciting event occurred after the diagnosis of TCC of the urinary tract had been made.
Four dogs had no known events that may have resulted in tumor seeding. In one of these dogs, review of ultrasonographic images made on multiple hospital visits revealed a cord of tissue extending from the apex of the urinary bladder to the ventral portion of the abdominal wall. Necropsy of this dog revealed transmural spread of the TCC at the apex of the urinary bladder and TCC extending down the median ligament of the urinary bladder into the abdominal wall (Figure 2). The median ligament attached just caudal to the umbilicus in this dog. There were multiple nests of neoplastic cells surrounded by a marked scirrhous response down the length of the ligament connecting with the mass in the abdominal wall (Figure 3). It was not possible to determine whether the TCC cells were within a lymphatic vessel associated with the ligament or whether they were migrating along the ligament. Neoplastic cells within Prox1-positive vessels were not observed in the primary tumor or ABWTCC in this dog.
Photographs of gross lesions of a dog with TCC in the urinary tract and ABWTCC. A—Photograph of the ventral serosal surface of the urinary bladder (*) immediately after euthanasia. Transitional cell carcinoma has penetrated the serosal surface, spread into the median ligament (arrow), and infiltrated the abdominal wall musculature (arrowheads). B—Photograph of the luminal surface of the opened urinary bladder (*) and the adjacent abdominal wall at necropsy. The luminal surface of the urinary bladder is effaced by TCC. The median ligament of the bladder (arrow) is infiltrated by TCC, which has extended into the adjacent abdominal wall musculature. Multiple nodules of TCC are within the abdominal wall (arrowheads). Bar = 1 cm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Photographs of gross lesions of a dog with TCC in the urinary tract and ABWTCC. A—Photograph of the ventral serosal surface of the urinary bladder (*) immediately after euthanasia. Transitional cell carcinoma has penetrated the serosal surface, spread into the median ligament (arrow), and infiltrated the abdominal wall musculature (arrowheads). B—Photograph of the luminal surface of the opened urinary bladder (*) and the adjacent abdominal wall at necropsy. The luminal surface of the urinary bladder is effaced by TCC. The median ligament of the bladder (arrow) is infiltrated by TCC, which has extended into the adjacent abdominal wall musculature. Multiple nodules of TCC are within the abdominal wall (arrowheads). Bar = 1 cm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Photographs of gross lesions of a dog with TCC in the urinary tract and ABWTCC. A—Photograph of the ventral serosal surface of the urinary bladder (*) immediately after euthanasia. Transitional cell carcinoma has penetrated the serosal surface, spread into the median ligament (arrow), and infiltrated the abdominal wall musculature (arrowheads). B—Photograph of the luminal surface of the opened urinary bladder (*) and the adjacent abdominal wall at necropsy. The luminal surface of the urinary bladder is effaced by TCC. The median ligament of the bladder (arrow) is infiltrated by TCC, which has extended into the adjacent abdominal wall musculature. Multiple nodules of TCC are within the abdominal wall (arrowheads). Bar = 1 cm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Photomicrographs of sections of tissues from the dog in Figure 2 with TCC in the urinary tract and ABWTCC. Histologic preparations reveal the connection between the urinary bladder wall (uw) and the abdominal wall (aw) via the median ligament (l). Neoplastic cells are indicated (asterisk). H&E stain; bar = 1 cm. Inset—Photomicrograph of neoplastic cells at higher magnification. Bar = 80 μm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Photomicrographs of sections of tissues from the dog in Figure 2 with TCC in the urinary tract and ABWTCC. Histologic preparations reveal the connection between the urinary bladder wall (uw) and the abdominal wall (aw) via the median ligament (l). Neoplastic cells are indicated (asterisk). H&E stain; bar = 1 cm. Inset—Photomicrograph of neoplastic cells at higher magnification. Bar = 80 μm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
Photomicrographs of sections of tissues from the dog in Figure 2 with TCC in the urinary tract and ABWTCC. Histologic preparations reveal the connection between the urinary bladder wall (uw) and the abdominal wall (aw) via the median ligament (l). Neoplastic cells are indicated (asterisk). H&E stain; bar = 1 cm. Inset—Photomicrograph of neoplastic cells at higher magnification. Bar = 80 μm.
Citation: Journal of the American Veterinary Medical Association 242, 4; 10.2460/javma.242.4.499
The presence of the ABWTCC was not recorded until the time of death in 6 dogs. On their first visit to the veterinary teaching hospital, 4 of these 6 dogs had advanced cancer, including a large ABWTCC, and the dog owners elected to have the dogs euthanized at that time. In 2 of the 6 dogs, the ABWTCC had not been detected prior to death and was found at necropsy.
Response to treatment and outcome—Attempts were made to treat the clinically detectable ABWTCC masses in 18 dogs; 17 dogs received medical treatment alone and 1 dog underwent medical treatment and surgical resection of the ABWTCC. In the dog that underwent surgical resection of the ABWTCC and medical treatment, a 4 × 3 × 3-cm mass was removed from the incision site of a previous surgery after the mass had been progressing despite carboplatin and piroxicam treatment. No residual ABWTCC was found in this dog at necropsy. All of the remaining dogs had ABWTCC present at the time of death. Various combinations of 4 cyclooxygenase inhibitors and 7 chemotherapy protocols were given to 18 dogs with clinically detected ABWTCC (Table 4). Of the 18 dogs, 11 received > 1 treatment protocol. No dog had complete or partial remission of the ABWTCC, and only 3 of 18 dogs that received medical treatment had stable disease in terms of the clinically detected ABWTCC. At the time the ABWTCC was treated in 18 dogs, the responses in the primary tumor included stable disease in 8 dogs and progressive disease in 9 dogs; 1 dog was not evaluable for treatment response at that site. Responses in distant metastases at that time included stable disease in 5 dogs and progressive disease in 7 dogs.
Response of TCC to medical treatment in 18 dogs with detectable ABWTCC.
| Treatment protocol | No. of dogs* | Primary tumor | Distant metastases | ABWTCC | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CR | PR | SD | PD | NA | CR | PR | SD | PD | NA | CR | PR | SD | PD | NA | ||
| Cyclooxygenase inhibitor† | 12 | 0 | 0 | 4 | 7 | 1 | 0 | 0 | 3 | 4 | 5 | 0 | 0 | 1 | 11 | 0 |
| Carboplatin | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 |
| Carboplatin plus cyclooxygenase inhibitor | 3 | 0 | 0 | 2 | 1 | 0 | 0 | 0 | 0 | 1 | 2 | 0 | 0 | 0 | 3 | 0 |
| Mitoxantrone plus cyclooxygenase inhibitor | 4 | 0 | 0 | 0 | 4 | 0 | 0 | 0 | 0 | 4 | 0 | 0 | 0 | 0 | 4 | 0 |
| Gemcitabine | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 |
| Chlorambucil | 1 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 |
| Chlorambucil plus cyclooxygenase inhibitor | 4 | 0 | 0 | 3 | 1 | 0 | 0 | 0 | 0 | 1 | 3 | 0 | 0 | 0 | 4 | 0 |
| Vinblastine | 2 | 0 | 0 | 2 | 0 | 0 | 0 | 0 | 2 | 0 | 0 | 0 | 0 | 2 | 0 | 0 |
Data are number of dogs.
Of the 18 dogs, 11 underwent > 1 treatment protocol.
The cyclooxygenase inhibitor included piroxicam in 7 dogs, deracoxib in 3 dogs, firocoxib in 1 dog, and carprofen in 1 dog.
CR = Complete remission. NA = Response not available. PD = Progressive disease. PR = Partial remission. SD = Stable disease.
Tumor response was classified as follows: complete remission, complete clinical resolution of the cancer; partial remission, ≥ 50% reduction in tumor volume; stable disease, < 50% change in tumor volume; and progressive disease, ≥ 50% increase in tumor volume or the development of new tumor lesions.
All 24 dogs with ABWTCC were euthanized because of poor quality of life related to their cancer. Owners of 7 of the 24 dogs commented that the ABWTCC was a major factor in reducing quality of life, leading to the decision to euthanize the dog.
Discussion
At least 3 important findings from the study should be noted. The first important finding was that all of the dogs with tissues available for study had UPIII expression in the ABWTCC as well as in the primary tumor. This was not unexpected, considering that in an earlier study,5 UPIII expression was detected in 50 of 55 primary TCC lesions and in 4 of 5 metastatic TCC lesions. Finding UPIII in the ABWTCC is important because an abdominal wall mass may be detected by a veterinarian before any urinary tract masses have been detected. The results of this study indicate that UPIII can serve as a marker of urothelial origin if the mass is an ABWTCC. As in the case of primary tumors, the absence of UPIII may not rule out urothelial origin, but the presence of UPIII certainly would indicate the need to search for cancer in the urinary tract. Finding UPIII in the tumors is also important as targeted treatments are developed. Any treatments aimed at this protein would be expected to be effective against the abdominal wall tumor as well as in the primary TCC that expresses UPIII.
The second important finding from the study was the information on possible etiologic factors for ABWTCC. It was not surprising that in 20 of the 24 cases, there were known events that could have resulted in tumor seeding in the abdominal wall, given that seeding of TCC in medical procedures has been reported.2–4 In a case series reported by Gilson and Stone,2 8 dogs had seeding of cancer following surgical procedures, including 5 dogs with carcinoma of the urinary tract. Many dogs in the present study were euthanized because of the seeded tumors. In other reports,3,4 TCC seeding has been reported from ultrasound-guided fine-needle aspiration as well as from surgery. Similarly, tumor implantation at the time of urinary bladder cancer surgery has been reported in humans, albeit rarely, following open and laparoscopic surgery approaches, especially if urinary bladder perforation has occurred.12–16 The location of the ABWTCC in the area of the previous incision site in several dogs in the present study also provides evidence for the probable role of surgery in the abdominal wall growth of the cancer. Reported mechanisms that could be involved in the growth of TCC in incision sites include direct implantation of tumor cells into the incision area at the time of surgery, the microenvironment in a surgical wound that would support tumor cell proliferation should TCC cells arrive there at surgery or through lymphatic or hematogenous spread, impaired immune response in the perioperative period, and increased vascularity in surgical wounds.17,18 Even in the last half of the study period, when clinicians should have been well aware of the risk of TCC seeding at surgery, > 1 in 10 dogs with TCC that underwent surgery developed ABWTCC. It is crucial for clinicians to be diligent when performing surgery in dogs with TCC to minimize the risk of tumor seeding. Furthermore, it is important to avoid percutaneous or transabdominal procedures such as needle biopsy and fine-needle aspiration in dogs that potentially have TCC. Although not formally investigated yet, there is also concern that performing cystocentesis could serve as a means to spread TCC.
This study also provided a key piece of information regarding another route of spread of TCC to the abdominal wall. There was clear documentation of TCC spread down the median ligament of the urinary bladder in 1 dog. It was interesting that the ligament attached cranially near the umbilicus in this dog, whereas it is more typical for the median ligament to attach ventrally much closer to the pubic bone.19 Although lymphatic vessels run along the median ligament, it could not be determined whether the cancer cells were traveling in lymphatic vessels or whether they were following the scaffold provided by the ligament. The primary care veterinarian in this case was certain that no procedure (surgery, percutaneous biopsy, fine-needle aspiration, or cystocentesis) had been performed in the dog. Similarly, none of these procedures was performed in this dog at the Veterinary Teaching Hospital. Findings in this dog indicate that although seeding through medical procedures remains a common cause of ABWTCC, natural processes can result in growth of the TCC in the abdominal wall as well.
The lymphatic spread of TCC remains to be completely understood. In a study20 of near-infrared imaging to track lymphatic drainage from the urinary bladder, it was found that lymphatic drainage from the urinary bladder can vary considerably from dog to dog, and lymphatic drainage does not always follow expectations based on anatomy. Documenting lymphatic spread of TCC at the cellular level remains a challenge as well. In the dogs with ABWTCC in the present study, lymphatic vessels containing TCC cells were rarely observed in the primary tumor and were uncommonly found in the ABWTCC samples. This may not rule out lymphatic spread of the cancer. In small biopsy specimens, it is possible that insufficient tumor tissue may be available to adequately observe lymphatic vessels. Lymphatic vessels are usually near the periphery of tumor masses, and biopsy specimens obtained from the middle of a TCC mass may not contain lymphatic vessels. In addition, TCC cells would not be expected to be observed in every microscopic slice through a lymphatic vessel even if TCC cells were present in the vessel. The absence of TCC cells in 1 part of a lymphatic vessel would not rule out their presence in other areas of the lymphatic vessel.
The third important finding of the study was the poor response of ABWTCC to medical treatment once masses were clinically detectable. This was not thought to be the result of particularly advanced cancer stage or grade at the time of diagnosis of TCC because the stage and grade of cancer in dogs with ABWTCC were similar to those in dogs in other reports1 of TCC treatment. At the time of diagnosis of TCC, distant spread of the cancer (in addition to the cancer in the abdominal wall) was detected in < 20% of dogs in the present study. The medical treatment of dogs with ABWTCC included 11 protocols given in 28 courses of treatment. No dog had remission of the ABWTCC, and only 3 of 18 dogs that received medical treatment had stable disease in terms of the ABWTCC. Progressive disease in the ABWTCC occurred following the other 25 courses of treatment. At the time the ABWTCC was treated in 18 dogs, the best response in the primary tumor was stable disease, which occurred in 8 dogs. The treatment response of ABWTCC appeared much less favorable than response rates reported for the medical treatment of TCC.1 The remission rate has typically been reported to range from 18% to 35%, and the stable disease rate has been reported to be approximately 50%.1 In most dogs with TCC of the urinary tract and ABWTCC, the mass in the abdominal wall did not become detectable for several weeks (median, 17 weeks) after diagnosis of TCC and a possible inciting event; therefore, the treatments the dogs received between diagnosis and the emergence of clinically detectable ABWTCC likely had some effect in controlling the cancer at that site for a while. It is possible that treatment may be initially beneficial. Once TCC becomes established in the abdominal wall and clinically detectable at that site, it appears that medical treatment is rarely effective. In humans, extravesical TCC resulting from surgical seeding also responds poorly to treatment.16
On the basis of pet owner comments, ABWTCC was a major factor leading to euthanasia in several of the dogs. Transitional cell carcinoma of abdominal wall can become quite massive and can cover the entire ventral portion of the abdomen and thorax. The only dog that did not have ABWTCC at the time of death was the dog that had the mass surgically removed. Although 1 case is not sufficient to draw firm conclusions, the absence of the ABWTCC in this dog and the failure of drugs to control the tumor at this site in other dogs suggest that surgical removal of ABWTCC before it can progress could be warranted.
In this study, 24 of 544 (4.4%) dogs with TCC of the urinary tract evaluated at our institution had ABWTCC. It is possible that this is an underestimation of the number of dogs actually affected with ABWTCC. In this study, histopathologic confirmation of ABWTCC was required. This was done to avoid incorrectly including dogs with inflammation, scar tissue, or other tumors in the abdominal wall. It is recognized that not all dogs with TCC and ABWTCC underwent biopsy of the abdominal wall, and this may have resulted in an underestimation of the number of dogs that actually had ABWTCC. Regardless, the finding of 4.4% serves as an important reminder of the potential for this cancer to spread to the abdominal wall.
In conclusion, ABWTCC is not common but carries a poor prognosis once TCC becomes established in the abdominal wall and clinically detectable at that site. It is crucial to undertake measures to prevent ABWTCC, such as careful surgical technique and avoiding percutaneous and transabdominal sampling of TCC. The response of clinically detected ABWTCC to medical treatment is poor, and therefore, early surgical removal of ABWTCC should be considered. Uroplakin III is commonly expressed in ABWTCC; finding UPIII in abdominal wall masses of unknown origin would serve as an indication to carefully evaluate the urinary tract.
ABBREVIATIONS
| ABWTCC | Transitional cell carcinoma in the abdominal wall |
| IHC | Immunohistochemistry |
| Prox1 | Prospero homeobox protein 1 |
| TCC | Transitional cell carcinoma |
| UPIII | Uroplakin III |
Research Diagnostics, Concord, Md.
AngioBio Corp, Del Mar, Calif.
EnVision+, Dako Corp, Carpinteria, Calif.
LSAB+, Dako Corp, Carpinteria, Calif.
STATA SE, version 11.1, StataCorp, College Station, Tex.
References
1. Knapp DW. Tumors of the urinary system. In: Withrow SJ, Vail DM, eds. Withrow and MacEwen's small animal clinical oncology. 4th ed. St Louis: WB Saunders Co, 2007;649–658.
2. Gilson SD, Stone EA. Surgically induced tumor seeding in eight dogs and two cats. J Am Vet Med Assoc 1990; 196:1811–1815.
3. Liptak JM, Brutscher SP, Monnet E, et al. Transurethral resection in the management of urethral and prostatic neoplasia in 6 dogs. Vet Surg 2004; 33:505–516.
4. Nyland TG, Wallack ST, Wisner ER. Needle-tract implantation following us-guided fine-needle aspiration biopsy of transitional cell carcinoma of the bladder, urethra, and prostate. Vet Radiol Ultrasound 2002; 43:50–53.
5. Ramos-Vara JA, Miller MA, Boucher M, et al. Immunohistochemical detection of uroplakin III, cytokeratin 7, and cytokeratin 20 in canine urothelial tumors. Vet Pathol 2003; 40:55–62.
6. Lai CL, van den Ham R, van Leenders G, et al. Histopathological and immunohistochemical characterization of canine prostate cancer. Prostate 2008; 68:477–488.
7. Owens L. TNM classification of tumors in domestic animals. Geneva: World Health Organization, 1980;34.
8. Chun R, Knapp DW, Widmer WR, et al. Phase II clinical trial of carboplatin in canine transitional cell carcinoma of the urinary bladder. J Vet Intern Med 1997; 11:279–283.
9. McMillan SK, Boria P, Moore GE, et al. Antitumor effects of deracoxib treatment in 26 dogs with transitional cell carcinoma of the urinary bladder. J Am Vet Med Assoc 2011; 239:1084–1089.
10. Wilting J, Papoutsi M, Christ B, et al. The transcription factor Prox1 is a marker for lymphatic endothelial cells in normal and diseased human tissues. FASEB J 2002; 16:1271–1273.
11. Martin A, Gasse H, Staszyk C. Absence of lymphatic vessels in the dog dental pulp: an immunohistochemical study. J Anat 2010; 217:609–615.
12. Via KJ, Burns KM, Lamm DL. Tumor implantation: a rare but potentially preventable cause of death in cystectomy patients. Can J Urol 2010; 17:5216–5218.
13. Huang J, Lin T, Liu H, et al. Laparoscopic radical cystectomy with orthotopic ileal neobladder for bladder cancer: oncologic results of 171 cases with a median 3-year follow-up. Eur Urol 2010; 58:442–449.
14. Rouprět M, Smyth G, Irani J, et al. Oncological risk of laparoscopic surgery in urothelial carcinomas. World J Urol 2009; 27:81–88.
15. Chakavarti A, Day DW, MacDermott S. Extravesical transitional cell carcinoma as a result of implantation after perforation of the bladder. BJU Int 2000; 85:1150–1151.
16. Abaza R, Miocinovic R, Keck RW, et al. Prevention and treatment of transitional cell carcinomatosis with intraperitoneal chemotherapy in a rat model. J Urol 2009; 181:1901–1906.
17. Tsivian A, Sidi AA. Port site metastases in urological laparoscopic surgery. J Urol 2003; 169:1213–1218.
18. Anderson WI, Dunham BM, King JM, et al. Presumptive subcutaneous surgical transplantation of a urinary bladder transitional cell carcinoma in a dog. Cornell Vet 1989; 79:263–266.
19. Evans HE, Christensen GC. The urogenital system. In: Evans HE, ed. Miller's anatomy of the dog. 3rd ed. Philadelphia: WB Saunders Co, 1993;494–558.
20. Knapp DW, Adams LG, Degrand AM, et al. Sentinel lymph node mapping of invasive urinary bladder cancer in animal models using invisible light. Eur Urol 2007; 52:1700–1708.
