Abstract
The objective of this retrospective study was to evaluate the effects of surgery on outcome for dogs with naturally occurring urinary bladder transitional cell carcinoma. Forty-seven dogs met the inclusion criteria. Thirty-one dogs (Group A) were treated with partial cystectomy and adjunctive medical therapy and 16 dogs (Group B) were treated with medical therapy alone. Overall survival was greater in dogs treated with partial cystectomy and adjunctive medical therapy (498 days for Group A versus 335 days for Group B, hazard ratio 2.5; 95% confidence interval: 1.1 to 5.7; P = 0.026). Progression-free survival was not different between groups (85 days for Group A versus 83 days for Group B; P = 0.663). No prognostic factors were identified for progression-free survival. Due to the many cases in Group A that were lost to follow-up, time-to-event survival analysis was performed. No significant difference in overall survival was noted, and no prognostic factors were identified in the time-to-event analysis. Prospective, randomized studies are needed to determine the role of partial cystectomy in the treatment of transitional cell carcinoma.
Résumé
Résultats cliniques des chiens atteints d’un carcinome à cellules transitionnelles recevant un traitement médical, avec et sans cystectomie partielle. L’objectif de cette étude rétrospective était d’évaluer les effets de la chirurgie sur les résultats chez des chiens atteints d’un carcinome à cellules transitionnelles de la vessie d’origine naturelle. Quarante-sept chiens répondaient aux critères d’inclusion. Trente et un chiens (Groupe A) ont été traités par cystectomie partielle et traitement médical d’appoint et 16 chiens (Groupe B) ont été traités par thérapie médicale seule. La survie globale était plus élevée chez les chiens traités par cystectomie partielle et traitement médical d’appoint (498 jours pour le Groupe A contre 335 jours pour le Groupe B, rapport de risque de 2,5; intervalle de confiance à 95 % : 1,1 à 5,7; P = 0,026). La survie sans progression n’était pas différente entre les groupes (85 jours pour le Groupe A contre 83 jours pour le Groupe B; P = 0,663). Aucun facteur pronostique n’a été identifié pour la survie sans progression. En raison des nombreux cas dans le Groupe A qui ont été perdus de vue lors du suivi, une analyse du temps de survie a été realisée. Aucune différence significative dans la survie globale n’a été notée et aucun facteur pronostique n’a été identifié dans l’analyse du temps de survive. Des études prospectives randomisées sont nécessaires pour déterminer le rôle de la cystectomie partielle dans le traitement du carcinome à cellules transitionnelles.
(Traduit par Dr Serge Messier)
Introduction
Urinary bladder transitional cell carcinoma (TCC) is an important neoplasm in dogs, with the most common clinical signs being hematuria, dysuria, and pollakiuria (1–2). Additionally, canine urinary bladder TCC has been used as a model to study the muscle-invasive form of urothelial carcinoma in human patients (3–4). The invasive nature of TCC and its propensity to develop in the trigone of the bladder contribute to the historically poor prognosis associated with this cancer in dogs (3). Lymph node metastasis is reported in 16% of dogs with TCC at the time of diagnosis (4). Distant metastasis to the lung or liver is reported in 14% of dogs at the time of diagnosis and in 50% by the time of death (4).
Treatments for dogs with urinary bladder TCC include medical therapy, radiation therapy, and surgery (4). The most often used medical therapies for canine TCC involve cyclooxygenase (COX) inhibitors, systemic chemotherapeutic agents, or some combination thereof (4). Previous studies have evaluated overall response rates, remission duration, and survival times for various medical therapies (5–7). It is the authors’ opinion that no consensus exists as to the “gold standard” of therapy.
Although beneficial in local control, the high incidence of severe late radiation-associated toxicoses associated with conventional definitive radiation protocols has precluded radiation from being included as standard of care treatment (8–10). However, intensity-modulated radiation therapy allows for conformity of radiation dose and use of smaller treatment fields to limit exposure to normal tissue and ultimately decrease the adverse events previously noted (11). Further investigation is needed, but this has promise in the treatment of TCC in dogs.
In humans, radical cystectomy for muscle-invasive urothelial carcinoma improves outcomes but requires urinary diversion techniques (12). Although complete cystectomy provides a surgical solution to canine tumors affecting the trigone, few veterinary studies have examined the feasibility of total cystectomy with ureteral diversion, given the postoperative challenges associated with home management after complete cystectomy (13–15).
Common sequelae of complete cystectomy in the dog are permanent urinary incontinence, pyelonephritis, and azotemia (13–15). Thus, if confirmed to improve outcome, partial cystectomy and/or cytoreductive surgeries would be considered desirable options.
Several studies have suggested cytoreductive surgeries may benefit dogs with TCC (16–18). In 1 report, a benefit was noted following complete excision of non-metastatic bladder tumors, with a median survival time (MST) of 365 d following surgery alone (16). Extent of tumor excision was prognostic, with an MST of 120 d when > 50% but < 100% of the tumor was removed, and 75 d when < 50% of the tumor was removed (16). A separate study noted 6/10 dogs with urinary bladder TCC treated with partial cystectomy, with or without ureteral re-implantation, survived over 1 y with or without chemotherapy and/or radiation therapy (17). A retrospective study evaluating the combination of doxorubicin and piroxicam for canine TCC noted that the subset of dogs first treated with partial cystectomy had aimproved MST (217 d with surgery compared to 133 d without surgery, P = 0.02) but not progression-free survival (PFS) (18). A more recent study by Marvel et al (19) evaluated 37 dogs with TCC treated with partial cystectomy and COX inhibitors, with or without other medical therapies. The MST for all dogs was 348 d, but dogs with non-trigonal bladder TCC treated with full-thickness partial cystectomy and daily piroxicam (± chemotherapy) had improved survival (772 d, P ≤ 0.0001) (19). Although these latter results are promising, it is still unclear whether initiating surgery before medical therapy improves disease-free survival compared to medical therapy alone.
The goal of the current retrospective study was to compare the outcomes of dogs with urinary bladder TCC treated with partial cystectomy and adjuvant chemotherapy and COX inhibitors versus those treated with chemotherapy and COX inhibitors alone. A secondary aim was to identify prognostic factors for PFS and overall survival (OS) for the same patient groups.
Materials and methods
Case selection
Medical records of dogs that were presented over a 9-year interval (September 2003 through November 2012) to 3 multidisciplinary referral hospitals (Red Bank Veterinary Hospital, Tinton Falls, New Jersey, USA; The Animal Medical Center, New York, New York, USA; and The Koret School of Veterinary Medicine of the Hebrew University of Jerusalem, Israel) were retrospectively reviewed to identify dogs diagnosed with urinary bladder TCC. Dogs were included if they were diagnosed with TCC by either histopathology (surgical or endoscopic biopsies) or by cytology (obtained via traumatic catheterization or fine-needle aspiration) and were treated with systemic injectable chemotherapy and COX inhibitors. Dogs were then grouped to compare those that were also treated with partial cystectomy to those that received medical therapy alone. Partial cystectomy was defined as removal of any portion of the urinary bladder with the intent of tumor cytoreduction. This was further categorized into removal of all visible disease or partial resection of visible disease as noted at the time of surgery. Dogs were excluded if they were not followed with sequential abdominal ultrasound after the initial 2 cycles of chemotherapy and then every 2 to 4 cycles thereafter. Dogs were also excluded if information in regard to abdominal ultrasonography findings, type and/or dose of chemotherapy, and type of COX inhibitor used was missing. Data collected from the medical records included breed age, gender, weight, date of diagnosis, staging tests performed, date of surgery, tumor location in bladder based on abdominal imaging, surgical procedure (removal of all visible disease versus partial resection of visible disease), histopathology results, adjuvant therapy (chemotherapy, COX inhibitors), date of recurrence, date and cause of death, or loss to follow-up. Tumor location was based on the site of attachment or origin of the tumor and was assigned to 1 of 4 anatomic regions of the bladder: apex, body, trigone, or neck. The apex was defined as the region immediately surrounding the previous urachal attachment site at the cranial/proximal 25% of the bladder. The body was defined as the region between the apex and the trigione. The trigione was defined as the triangular area, near the neck, on the dorsal surface of the bladder immediately surrounding the implantation site of the ureters. The neck was defined as the region ventral and distal to the trigione and was site of bladder termination and urethral origin. Ureteral involvement was defined as visualization of tumor within one or both ureters at the time of initial ultrasonographic examination or at the time of surgery. Prostatic involvement was determined based on ultrasonographic evidence of tumor extension into the prostate, with prostate enlargement and mineralization. After initiation of treatment, dogs were periodically reassessed with physical examinations, clinical laboratory data, and imaging. Urinary tract ultrasonography was employed to determine tumor response and remission status, using World Health Organization (WHO) criteria. Tumor size was calculated as the product of the longest diameter and the greatest perpendicular diameter. Complete remission was defined as disappearance of all measurable disease, partial remission was defined as > 50% decrease in tumor size from baseline measurements, stable disease was defined as < 25% increase in tumor size from baseline measurement, and progressive disease was defined as > 25% increase in tumor size, based on baseline measurement or the appearance of new lesions (either locally or at a distant site).
Dogs were assigned to 1 of 2 groups: Group A — dogs that underwent a partial cystectomy, followed by treatment with systemic chemotherapy and COX inhibitors; and Group B — dogs receiving only medical therapy with chemotherapy and COX inhibitors. Median PFS and median OS were compared between the 2 groups. Data were also evaluated for prognostic factors across both groups. Dogs in Group A were restaged using ultrasonography for detection of tumor recurrence following surgery and chemotherapy initiation; whereas, dogs in Group B were monitored for change in tumor status from baseline following initiation of chemotherapy. Ultrasonography was done after the initial 2 cycles of chemotherapy, then every 2 to 4 cycles thereafter. Progressive disease (PD) was determined in both Groups A and B by ultrasonographic evaluation.
Histology
All biopsy samples were reviewed and the diagnosis of TCC confirmed by Board-certified anatomic pathologists. Histologic margins were classified as complete or incomplete.
Cytology
All cytology samples were reviewed and the diagnosis of TCC was confirmed by Board-certified clinical pathologists.
Statistical analysis
Continuous data were expressed as median and range, and categorical data as frequencies and percentages. Progression-free survival and OS were calculated from the date of first chemotherapy to the date of PD or death from any cause, and death, respectively. Dogs were censored if they had not developed PD or died at the time of data analysis, or if they were lost to follow-up. The cases that were lost to follow-up were then uncensored and a time-to-event (TTE) comparison was done for OS. Continuous variables were compared between groups of patients using a 2-sample t-test or Mann-Whitney test, depending on normality of the data. Categorical variables were compared between cohorts using a Fisher’s Exact test. The Kaplan-Meier method was used to estimate and display the distribution of PFS and OS. Differences between potential prognostic subsets and between study arms were compared using the log-rank test. All reported P-values are 2-sided and P < 0.05 was used to determine statistical significance. All statistical analyses were completed using a commercial software package (Prism v. 6.0b; GraphPad Software, La Jolla, California, USA).
Results
Forty-seven dogs met the inclusion criteria (Group A; n = 31 and Group B; n = 16). Information regarding signalment is presented in Table 1. Clinical signs included; hematuria (n = 32; 68%), pollakiuria (n = 23; 49%), stranguria (n = 21; 45%), dysuria (n = 10; 21%), polydipsia (n = 3; 6%), 2 cases each of lethargy, urine dribbling, and polyuria, and 1 case each of urinary incontinence, tenesmus, kyphosis, and recurrent lower urinary tract infection. In 3 dogs (Group A, n = 2; Group B, n = 1) the diagnosis was incidental, and the dogs had no clinical signs.
Table 1.
Characteristics of dogs in Groups A and B.
| Parameter | Group A (31 dogs) | Group B (16 dogs) | P-value |
|---|---|---|---|
| Age (y) | |||
| Median (range) | 10.3 (5.3 to 15.3) | 11.1 (5.6 to 15.6) | 0.256a |
| Gender | |||
| Spayed female | 18 (58%) | 8 (50%) | 0.718b |
| Neutered male | 9 (29%) | 7 (44%) | |
| Intact male | 3 (10%) | 0 (0%) | |
| Intact female | 1 (3%) | 1 (6%) | |
| Weight (kg) | |||
| Median (range) | 14.2 (2.3 to 80) | 12.7 (7.1 to 46.4) | 0.953a |
| Breed | |||
| Mixed breed | 7 (23%) | 3 (19%) | 0.761b |
| Beagle | 6 (19%) | 3 (19%) | |
| West highland white terrier | 2 (6%) | 1 (6%) | |
| Other (n = 1 each breed) | 16 (52%) | 9 (56%) | |
t-test.
Chi-square test.
Twenty-four (51%) dogs had a urinalysis and 11 (23%) dogs had a urine culture completed before initiation of treatment. Urinalysis revealed atypical transitional cells in 6 of the 24 dogs and bacterial culture was positive in 5 of the 11 dogs. Bacteria isolated included Escherichia coli, Proteus mirabilis, and Staphylococcus intermedius. All dogs underwent an initial abdominal ultrasonographic examination that identified a bladder mass and 91% (n = 43) had a detailed description of the mass location available for review. The most common location was the trigone (Group A, n = 8; Group B, n = 13), followed by apical (Group A, n = 11; Group B, n = 1), and neck (Group A, n = 2; Group B, n = 1). Ten dogs (Group A, n = 7; Group B, n = 3) had multifocal disease. Three dogs (Group A, n = 1; Group B, n = 2) had ultrasonographic evidence of ureteral involvement, 10 dogs (Group B, n = 10) had evidence of urethral involvement, and 5 dogs (Group A, n = 2; Group B, n = 3) had evidence of prostatic involvement. No dogs had ultrasonographic evidence of metastasis to abdominal organs outside the urinary tract. No dogs had evidence of regional lymph node involvement at the time of initial diagnosis. Baseline thoracic radiographs were done for 60% of the dogs (n = 28). All dogs in Group A and 3 dogs (19%) in Group B had diagnosis confirmed via histopathology (surgically retrieved samples for Group A and cystoscopic biopsies for Group B). The diagnosis of TCC was confirmed in the remaining Group B dogs (n =13; 81%) by evaluation of cytology samples collected via traumatic catheterization (n = 10; 63%) or fine-needle aspiration (n = 3; 19%). Five dogs (38%) diagnosed via cytology had evidence of concurrent lower urinary tract infection.
Thirty-one dogs (66%; Group A) were treated with partial cystectomy, followed by systemic chemotherapy and COX inhibitors. Sixteen dogs (34%; Group B) were treated with systemic chemotherapy and COX inhibitors alone. One dog had a left neocystoureterostomy, due to left ureteral involvement. No dogs with urethral involvement were in Group A. Except for 1 case, dogs with prostatic extension and those with trigonal tumors had partial thickness cytoreductive surgeries that did not disrupt the ureterovesicular junction. Based upon the surgical report, complete excision of all gross disease was achieved in 16 dogs (52%) and partial excision of gross disease achieved in 15 (48%). Histopathology reports were available for review for all 31 cases and margins were deemed complete in 6 dogs (19%), incomplete in 21 dogs (68%), and histologic margins were not noted in the histopathology report in 4 cases (13%). One dog had regional lymph node biopsy performed at the time of cystectomy and no evidence of metastatic disease was detected histologically. Specific information on peri-operative surgical complications was not available, but all dogs survived to discharge. One dog (2.7%) developed tumor seeding to the abdominal wall associated with the surgical scar 95 d after surgery, following a diagnosis via fine-needle aspiration and cytology.
All study dogs were treated with a COX inhibitor at standard anti-inflammatory doses. The COX inhibitors used for both treatment groups are summarized in Table 2. Twenty-two dogs (47%) received deracoxib only, 16 (34%) received piroxicam only, 1 (2%) received carprofen only, and 1 (2%) received meloxicam only. Seven dogs (15%) were switched from one COX inhibitor to another at varying times throughout therapy, due to patient intolerance to a specific drug. The most commonly intolerable drug was piroxicam and the most common undesirable side effects were vomiting, diarrhea, and azotemia. Three dogs (6%) received sequential piroxicam then deracoxib, and 1 (2%) each received sequential piroxicam then meloxicam, piroxicam then carprofen, piroxicam then firocoxib, and piroxicam then carprofen then deraxocib.
Table 2.
Summary of medical therapy protocols for dogs in Groups A and B.
| Medical therapy protocols | Group A (31 dogs) | Group B (16 dogs) |
|---|---|---|
| Single agent | 20 (65%) | 9 (56%) |
| Doxorubicin | 13 (42%) | 4 (25%) |
| Mitoxantrone | 4 (13%) | 2 (13%) |
| Vinblastine | 3 (10%) | 1 (6%) |
| Carboplatin | 0 | 2 (13%) |
| Multi agent | 11 (35%) | 7 (44%) |
| Vinblastine/doxorubicin | 3 (10%) | 2 (13%) |
| Mitoxantrone/doxorubicin/carboplatin | 3 (10%) | 1 (6%) |
| Mitoxantrone/doxorubicin | 2 (6%) | 3 (19%) |
| Vinblastine/mitoxatrone | 1 (3%) | 0 |
| Doxorubicin/carboplatin | 0 | 1 (6%) |
| Doxorubicin/vinblastine/mitoxantrone/carboplatin | 1 (3%) | 0 |
| Cisplatin/carboplatin/gemcitabine/cyclophosphamide | 1 (3%) | 0 |
| Rescue protocol | 2 (6%) | 1 (6%) |
| Cyclophosphamide | 2 (6%) | 0 |
| Cyclophosphamide/gemcitabine | 0 | 1 (6%) |
| COX inhibitor | ||
| derocoxib | 16 (52%) | 6 (38%) |
| piroxicam | 9 (29%) | 7 (44%) |
| meloxicam | 1 (3%) | 0 |
| carprofen | 0 | 1 (6%) |
| multiple | 5 (16%) 2 | (13%) |
All study dogs received chemotherapy using various agents and protocols. The chemotherapy protocols used for both treatment groups are summarized in Table 2. First-line protocols were recorded for all 47 dogs. Twenty-nine dogs (62%) received a single agent protocol consisting of either doxorubicin given once every 21 d for 4 to 6 treatments (n = 17; 36%), mitoxantrone given once every 21 d for 4 to 6 treatments (n = 6; 13%), vinblastine given once every 7 d for 5 or 6 treatments (n = 4; 9%), or carboplatin given once every 21 d for 5 or 6 treatments (n = 2; 4%). Eighteen dogs (38%) received multiagent therapy. Combination mitoxantrone/doxorubicin was given as alternating doses every 21 d for 2 or 3 cycles (n = 5; 11%). Combination vinblastine/doxorubicin with vinblastine was given on week 1, doxorubicin on week 2, and week 3 was off. This cycle was repeated for a total of 4 to 6 cycles (n = 5; 11%). Combination mitoxantrone/doxorubicin/carboplatin was given as sequential doses every 21 d for 2 or 3 cycles (n = 4; 9%). Vinblastine/mitoxantrone with vinblastine was given on week 1, mitoxantrone on week 2, and week 3 was off. This cycle was repeated for 2 cycles (n = 1, 2%). Doxorubicin/carboplatin was given as sequential doses every 21 d for 3 cycles (n = 1, 2%). Doxorubicin/vinblastine/carboplatin/mitoxantrone was given as sequential doses every 21 d for 6 cycles (n = 1, 2%). Cisplatin, carboplatin, and gemcitabine/cyclophosphamide were administered with cisplatin and carboplatin given sequentially, then gemcitabine and oral cyclophosphamide given concurrently. These were administered every 21 d for 3 cycles (n = 1, 2%). Upon confirmation of progressive disease, rescue protocols utilizing oral cyclophosphamide (n = 2, 4%) and combination gemcitabine oral cyclophosphamide (n = 1, 2%) were instituted.
Twelve dogs (39%) in Group A and 2 dogs (13%) in Group B were censored in OS analysis due to being lost to follow-up. Only 1 dog in Group A (3%) and no dogs in Group B were lost to follow-up with respect to date of progression. The mean and median times to censoring for dogs lost to follow-up in Group A were 491 d (mean) and 364 d (median); and for Group B were 447 d (mean) and 447 d (median). Twenty-seven dogs (87%) in Group A and all dogs in Group B died or were euthanized due to progression of disease. Dogs that were lost to follow-up were presumed to have died from progressive disease if tumor progression was confirmed at the date of last contact. Four dogs (13%) in Group A and no dogs in Group B died or were euthanized due to causes unrelated to TCC. Causes of death in these 4 Group A dogs included acute collapse and cardiopulmonary arrest during exercise, progression of dilated cardiomyopathy, euthanasia due to progression of chronic renal failure and abscessation of a cutaneous plasmocytoma, and severe cervical pain suspected to be secondary to intervertebral disc disease. The dog that died of acute cardiopulmonary arrest was treated with doxorubicin; however, the total cumulative dose was 150 mm/m2. The dog that died of progressive dilated cardiomyopathy had this diagnosis before being diagnosed with TCC but was managed medically and was stable at the time of TCC diagnosis. This dog was not treated with doxorubicin but rather vinblastine. The dog that was euthanized due to progressive chronic renal failure and abscessation of cutaneous plasmocytoma was treated with the COX inhibitor piroxicam. At the time the cutaneous abscess was diagnosed, the owner of this dog elected euthanasia rather than surgical treatment of the cutaneous abscess. All 6 dogs that had histologically complete tumor excision died or were euthanized due to progression of disease.
Outcome and prognostic factors
Disease progression was documented in all 47 dogs. Progression was confirmed via abdominal ultrasonography and characterized by recurrence within the bladder lumen (n = 41), evidence of metastasis to sub-lumbar lymph nodes (n = 1), local extension into the prostate or urethra (n = 1), and abdominal wall seeding (n = 1).
The outcomes and potential prognostic variables for both treatment groups are summarized in Table 3. The median PFS for dogs undergoing cytoreductive surgery (Group A) and dogs not undergoing surgery (Group B) were not different (85 d versus 83 d, P = 0.663; Figure 1). No variables were prognostic for PFS (Table 2). The median OS for dogs undergoing cytoreductive surgery (Group A) was greater than that for dogs not undergoing surgery (Group B) when lost to follow-up cases were censored (498 d versus 335 d, hazard ratio 2.5; 95% CI: 1.1 to 5.7; P = 0.026; Figure 2). There was no difference in PFS or OS between dogs with histologically complete versus histologically incomplete tumor resection (PFS 70 d for complete versus 57 d for incomplete, P = 0.558 and OS 329 d for complete versus 484 d for incomplete, P = 0.340; Table 3). No other prognostic factors were identified.
Table 3.
Prognostic factors evaluated for progression-free survival (PFS) and overall survival (OS) for dogs treated with partial cystectomy and medical therapy or medical therapy alone.
| Variable | PFS (days) | P-value | OS (days) | P-value | OS TTE (days) | P-value |
|---|---|---|---|---|---|---|
| Surgery | 0.026 | |||||
| Yes | 85 | 0.663 | 498 | HR 2.5 | 276 | 0.200 |
| No | 83 | 335 | 95% CI (1.1–5.7) | 306 | ||
| Age | ||||||
| Above median | 121 | 0.499 | 363 | 0.341 | 306 | 0.434 |
| Below median | 79 | 424 | 302 | |||
| Gender | ||||||
| Male | 122 | 0.117 | 484 | 0.738 | 329 | 0.620 |
| Female | 61 | 424 | 210 | |||
| Body weight | ||||||
| Above median | 85 | 0.331 | 424 | 0.784 | 363 | 0.989 |
| Below median | 61 | 276 | 216 | |||
| Multifocal | ||||||
| Yes | 67 | 0.414 | 535 | 0.134 | 461 | 0.112 |
| No | 85 | 306 | 220 | |||
| Trigonal | ||||||
| Yes | 57 | 0.149 | 363 | 0.360 | 215 | 0.267 |
| No | 104 | 374 | 329 | |||
| Diffuse | 60 | 0.204 | 484 | 0.115 | 309 | 0.117 |
| Polypoid | 146 | 616 | 498 | |||
| Ureteral involvement | ||||||
| Yes | 168 | 168 | ||||
| No | 79 | 0.819 | 432 | 0.266 | ||
| Urethral involvement | ||||||
| Yes | 125 | 363 | 363 | |||
| No | 77 | 0.895 | 484 | 0.066 | 306 | 0.407 |
| Prostate involvement | ||||||
| Yes | 125 | 329 | 0.257 | 270 | ||
| No | 77 | 0.873 | 484 | 363 | 0.377 | |
| Complete gross excision | 85 | 424 | 276 | |||
| Gross cytoreduction | 72 | 0.379 | 616 | 0.487 | 287 | 0.948 |
| Margins | ||||||
| Complete | 70 | 0.558 | 329 | 0.340 | 248 | 0.296 |
| Incomplete | 57 | 484 | 242 | |||
| Deracoxib | 56 | 276 | 226 | 0.465 | ||
| Piroxicam | 104 | 0.521 | 424 | 0.793 | 364 |
HR — Hazard ratio; CI — Confidence interval.
Figure 1.
Kaplan–Meier curve comparing the progression-free survival for dogs treated with partial cystectomy and medical therapy (Surgery — solid line) versus those treated with medical therapy alone (No surgery — dashed line). Vertical tick marks indicate censored cases.
Figure 2.
Kaplan–Meier curve comparing the overall survival for dogs treated with partial cystectomy and medical therapy (Surgery — solid line) versus those treated with medical therapy alone (No surgery — dashed line). Hazard ratio (HR); 95% CI (confidence interval). Vertical tick marks indicate censored cases.
Table 3 also reports OS time to event (TTE) with lost to follow-up (LTF) cases uncensored due to the high numbers of LTF in dogs in the surgery group (Group A). In this scenario, the median OS for dogs undergoing cytoreductive surgery (Group A) was not different from that for dogs not undergoing surgery (Group B) (276 d versus 306 d, P = 0.200; Figure 2).
In Group B dogs, stable disease was noted in 10 dogs (63%) and progressive disease in 6 dogs (38%). No dogs in Group B had a complete or partial response to therapy.
Discussion
Previous veterinary studies have formally evaluated effects of surgery on outcomes for dogs with naturally occurring urinary bladder TCC (13–19). However, to the authors’ knowledge, there are no studies comparing the outcomes of canine patients with urinary bladder TCC receiving surgery and adjunctive medical therapy, compared to those receiving medical therapy alone, with case controls from the same time interval and institutions. Our study population was similar with respect to breed, gender, and clinical signs to previously reported populations of dogs with urinary bladder TCC (2,3,19,20).
In the study reported here, the median OS for dogs treated with partial cystectomy and adjuvant systemic chemotherapy/COX inhibitors (Group A), was 498 d compared to 335 d for dogs treated with systemic chemotherapy/COX inhibitors alone (Group B) (P = 0.026). The median PFS of 83 d and median OS of 335 d for dogs in Group B (medical therapy only) were comparable to previously published survival times for dogs treated with medical therapies alone (2,5,7,21). The OS of 498 d for dogs in Group A, seemed greater than the reported median OS of approximately 1 y for dogs undergoing partial cystectomy, with or without adjuvant chemotherapy in other reports (16,18,19).
Knowing the substantial number of LTF cases in Group A may have influenced outcomes, a TTE analysis was performed. In this analysis, median OS was not significantly different between the 2 groups. While censoring LTF cases likely artificially prolonged the median OS for Group A, uncensoring these cases for TTE calculation may have underestimated the median OS calculation. Although the exact dates of death were unknown, 11 (92%) of the dogs that were LTF in Group A had ultrasonographic evidence of progressive disease and were alive at the date of last contact, with median time for LTF of 491 d. This approaches the censored median OS of 498 d and it is reasonable to presume these patients subsequently died or were euthanized due to their disease shortly thereafter.
In our study, the impact of extent of surgery was unclear, as surgical dose (complete visual excision versus gross debulking) and histologic completeness of excision were not significantly associated with outcome. This finding was in contrast to that of Marvel et al (19), in which dogs that underwent full thickness cystectomy had significantly improved PFI and median OS compared to dogs that had partial thickness tumor excision. That surgical dose is prognostic is also suggested by data reported by Knapp et al (4), who noted improved median OS for patients undergoing cytoreductive surgery and adjuvant medical therapy (4). The lack of significance of surgical dose in the current study could be attributed to inadequate power due to low case numbers.
Tumor location away from the trigone was not associated with an improvement in median OS (374 d, P = 0.360) (using both the censored and uncensored data, Table 3). This was in contrast to the study by Marvel et al (19) in which tumor location away from the trigone was a prognostic factor that significantly improved survival. In our study, dogs with non-trigonal tumors, were in fact, more likely to undergo surgery prior to medical therapy (74% versus 38%). The propensity for patients with non-trigonal tumors to undergo surgery more commonly than patients with trigonal tumors was likely explained by surgical decision-making, as full-thickness partial cystectomy or total cystectomy for trigonal tumors is a more challenging surgical procedure with higher morbidity than partial cystectomy for tumors at non-trigonal locations. In a pilot study, total cystectomy with ureterocolonic anastomosis in 10 dogs with TCC was associated with a poor prognosis and adverse effects related to systemic uremia (15). Total cystectomy with vaginal or preputial ureteral diversion in 10 dogs yielded an MST of 385 d and cutaneous ureteral diversion in 4 dogs was associated with an MST of 279 d (13,14). These patients were persistently incontinent, making at home management more difficult and it is the authors’ opinion that these surgical approaches would be less acceptable to owners.
An alternative explanation would be that surgery does not improve outcome, rather it is purely a result of anatomic and functional impact differences in where the tumor arises (and recurs). This latter alternative is supported by the fact that surgery did not improve PFS in our study. This implies, and is intuitive, that progression away from the trigone (and therefore away from ureters and urethra) is less likely to affect quality of life and indeed, life-threatening outflow problems. In such a scenario, PFS duration that was not improved by surgery may not have as grave an impact on survival based on an anatomically more favorable location and therefore, that patient would live longer despite, not because of, surgery.
Similar to other reports (18,19), complete histologic excision was not prognostic in our study. This repeatable finding may be explained by the field cancerization effect within the bladder lining and/or the likelihood of TCC cell exfoliation and implantation in other locations within the urinary tract.
Comparison of benefit achieved with the various chemotherapy protocols and COX inhibitors used in this study was not possible, as the agents used, dose intensity, and duration were not uniform. Given the findings supporting the theory of field cancerization effect, the exfoliative nature of this tumor, locoregionally aggressive behavior, and considerable risk of metastasis with TCC, adjuvant chemotherapy should be considered, even in cases of complete excision with no evidence of metastasis (4). This recommendation is supported by the fact that 43 dogs (91%) in this study died or were euthanized due to progressive disease.
Serial abdominal ultrasound was used to monitor response to therapy and assess for progressive disease. In a recent study, experienced observers using a standardized protocol (WHO or RECIST criteria) reliably determined urinary bladder TCC dimensions using 2D ultrasonography (22).
Intra-operative complications were not specifically reported in this study, but all dogs survived to discharge, and no dogs died due to peri-operative surgical or anesthetic complications. One dog developed tumor seeding to the abdominal wall associated with the surgical scar; this represented 2.7% of cases undergoing partial cystectomy in our study, considerably lower than the 10 to 11% incidence reported by others (19,23)
Limitations of this study were primarily due to its retrospective nature. These include the relatively small number of cases, a protracted case accrual period, lack of review by one pathologist, non-standardized follow-up schedule, and non-uniform reporting of data in the medical record precluding accurate determination of TNM stage and verification of tumor response using the cRECIST criteria. The retrospective nature also did not allow for standardization of systemic therapy protocols. A further limitation was the potential for case selection bias towards surgery for dogs with non-trigonal tumors. This potential bias could mean these cases with non-trigonal tumors benefited the most from surgical treatment. Given the anatomy of the trigone region, elimination of this bias in future studies would require investigation of non-trigonal tumors separate from trigonal tumors, or further development of surgical techniques for neobladder in the dog.
In summary, dogs treated with partial cystectomy, systemic chemotherapeutics, and daily COX inhibitors had an improved median OS of 498 d compared to 335 d for dogs receiving medical therapy alone [hazard ratio: 2.5; 95% CI: 1.1 to 5.7; P = 0.026]. No prognostic factors were identified for PFS. The findings of our study supported the recommendation for partial cystectomy with systemic chemotherapeutics and daily COX inhibitors in the treatment of this disease. The current study also supported the need for prospective, randomized controlled trials evaluating dogs with primary urinary bladder TCC to determine the most favorable protocol for adjuvant chemotherapy and the role of partial cystectomy for the treatment of TCC. CVJ
Footnotes
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (hbroughton@cvma-acmv.org) for additional copies or permission to use this material elsewhere.
References
- 1.Henry CJ. Management of transitional cell carcinoma. Vet Clin North Am Small Anim Pract. 2003;33:597–613. doi: 10.1016/s0195-5616(03)00032-9. [DOI] [PubMed] [Google Scholar]
- 2.Mutsaers AJ, Widmer WR, Knapp DW. Canine transitional cell carcinoma. J Vet Intern Med. 2003;17:136–144. doi: 10.1892/0891-6640(2003)017<0136:ctcc>2.3.co;2. [DOI] [PubMed] [Google Scholar]
- 3.Knapp DW, Ramos-Vara JA, Moore GE, Dhawan D, Bonney PL, Young KE. Urinary bladder cancer in dogs, a naturally occurring model for cancer biology and drug development. ILAR Journal. 2014;55:100–118. doi: 10.1093/ilar/ilu018. [DOI] [PubMed] [Google Scholar]
- 4.Knapp DW, Glickman NW, DeNicola DB, Bonney PL, Lin TL, Glickman LT. Naturally-occurring canine transitional cell carcinoma of the urinary bladder: A relevant model of human invasive bladder cancer. Uro Oncol. 2000;5:47–59. doi: 10.1016/s1078-1439(99)00006-x. [DOI] [PubMed] [Google Scholar]
- 5.Arnold EJ, Childress MO, Fourez LM, et al. Clinical trail of vinblastine in dogs with transitional cell carcinoma of the urinary bladder. J Vet Intern Med. 2011;25:1385–1390. doi: 10.1111/j.1939-1676.2011.00796.x. [DOI] [PubMed] [Google Scholar]
- 6.Boria PA, Glickman NW, Schmidt BR, et al. Carboplatin and piroxicam therapy in 31 dogs with transitional cell carcinoma of the urinary bladder. Vet Comp Oncol. 2005;3:73–80. doi: 10.1111/j.1476-5810.2005.00070.x. [DOI] [PubMed] [Google Scholar]
- 7.Henry CJ, McCaw DL, Turnquist SE, et al. Clinical evaluation of mitoxantrone and piroxicam in a canine model of human invasive urinary bladder carcinoma. Clin Cancer Res. 2003;9:906–911. [PubMed] [Google Scholar]
- 8.Anderson CR, McNeil EA, Gillette EL, Powers BE, LaRue SM. Late complications of pelvic irradiation in 16 dogs. Vet Radiol Ultrasound. 2002;43:187–192. doi: 10.1111/j.1740-8261.2002.tb01668.x. [DOI] [PubMed] [Google Scholar]
- 9.Walker M, Breider M. Intraoperative radiotherapy of canine bladder cancer. Vet Radiol Ultrasound. 1987;28:200–204. [Google Scholar]
- 10.Withrow SJ, Gillette EL, Hoopes PJ, McChesney SL. Intraoperative irradiation of 16 spontaneously occurring canine neoplasms. Vet Surg. 1989;18:7–11. doi: 10.1111/j.1532-950x.1989.tb01035.x. [DOI] [PubMed] [Google Scholar]
- 11.Nolan MW, Kogan L, Griffin LR, et al. Intensity-modulated and image-guided radiation therapy of treatment of genitourinary carcinomas in dogs. J Vet Intern Med. 2012;26:987–995. doi: 10.1111/j.1939-1676.2012.00946.x. [DOI] [PubMed] [Google Scholar]
- 12.Lee RK, Abol-Enein H, Artibani W, et al. Urinary diversion after radical cystectomy for bladder cancer: Options, patient selection, and outcomes. BJU Int. 2014;113:11–23. doi: 10.1111/bju.12121. [DOI] [PubMed] [Google Scholar]
- 13.Ricardo Huppes R, Crivellenti LZ, Barboza De Nardi A, et al. Radical cystectomy and cutaneous ureterostomy in 4 dogs with trigonal transitional cell carcinoma: Description of technique and case series. Vet Surg. 2016;46:111–119. doi: 10.1111/vsu.12583. [DOI] [PubMed] [Google Scholar]
- 14.Saeki K, Fujita A, Fujita N, Nakagawa T, Nishimura R. Total cystectomy and subsequent urinary diversion to the prepuce or vagina in dogs with transitional cell carcinoma of the trigone area: A report of 10 cases (2005–2011) Can Vet J. 2015;56:73–80. [PMC free article] [PubMed] [Google Scholar]
- 15.Stone EA, Withrow SJ, Page RL, Schwarz PD, Wheeler SL, Seim HB., 3rd Ureterocolonic anastomosis in ten dogs with transitional cell carcinoma. Vet Surg. 1988;17:147–153. doi: 10.1111/j.1532-950x.1988.tb00293.x. [DOI] [PubMed] [Google Scholar]
- 16.Norris AM, Laing EJ, Valli VE, et al. Canine bladder and urethral tumors: A retrospective study of 115 cases (1980–1985) J Vet Intern Med. 1992;6:145–153. doi: 10.1111/j.1939-1676.1992.tb00330.x. [DOI] [PubMed] [Google Scholar]
- 17.Stone EA, George TF, Gilson SD, Page RL. Partial cystectomy for urinary bladder neoplasia: Surgical technique and outcome in 11 dogs. J Small Anim Pract. 1996;37:1–6. doi: 10.1111/j.1748-5827.1996.tb01745.x. [DOI] [PubMed] [Google Scholar]
- 18.Robat C, Burton J, Thamm D, Vail D. Retrospective evaluation of doxorubicin-piroxicam combination for treatment of transitional cell carcinoma in dogs. J Small Anim Pract. 2013;54:67–74. doi: 10.1111/jsap.12009. [DOI] [PubMed] [Google Scholar]
- 19.Marvel SJ, Séguin B, Dailey DD, Thamm DH. Clinical outcome of partial cystectomy for transitional cell carcinoma of the canine bladder. Vet Comp Oncol. 2017;15:1417–1427. doi: 10.1111/vco.12286. [DOI] [PubMed] [Google Scholar]
- 20.Knapp DW, McMillian SK. Tumors of the urinary system. In: Withrow SJ, Vail DM, Page RL, editors. Withrow and MacEwen’s Small Animal Clinical Oncology. 5th ed. St. Louis, Missouri: Elsevier; 2013. pp. 572–582. [Google Scholar]
- 21.Mcmillan SK, Boria P, Moore GE, Widmer WR, Bonney PL, Knapp DW. Antitumor effects of deracoxib treatment in 26 dogs with transitional cell carcinoma of the urinary bladder. J Am Vet Med Assoc. 2011;239:13–18. doi: 10.2460/javma.239.8.1084. [DOI] [PubMed] [Google Scholar]
- 22.Honkisz SI, Naughton JF, Weng HY, Fourez LM, Knapp DW. Evaluation of two-dimensional ultrasonography and computed tomography in the mapping and measuring of canine urinary bladder tumors. Vet J. 2018;232:23–26. doi: 10.1016/j.tvjl.2017.12.008. [DOI] [PubMed] [Google Scholar]
- 23.Higuchi T, Burcham GN, Childress MO, et al. Characterization and treatment of transitional cell carcinoma of the abdominal wall in dogs: 24 cases (1985–2010) J Am Vet Med Assoc. 2013;242:499–506. doi: 10.2460/javma.242.4.499. [DOI] [PubMed] [Google Scholar]