Retrospective Evaluation of Adjunctive Doxorubicin for the Treatment of Feline Mammary Gland Adenocarcinoma: 67 Cases
Medical records for 67 cats with histologically confirmed mammary gland adenocarcinomas treated with adjunctive doxorubicin from June 1994 through December 2002 were reviewed. Data were examined to evaluate factors influencing disease-free interval (DFI) and survival time. The Kaplan-Meier median survival time of cats that received surgery and doxorubicin was 448 days. The Kaplan-Meier median DFI was 255 days. Significant univariate prognostic factors for DFI included histological subtype, completion of initial chemotherapy, development of metastatic disease, and location of metastatic disease. Significant univariate prognostic factors for survival included tumor volume, the development of metastatic disease, and location of metastatic disease.
Introduction
Mammary gland adenocarcinoma in cats is a common and devastating disease that is often fatal. It has been reported that between 80% and 96% of mammary gland tumors in cats are malignant.1–3 Tubular and papillary adenocarcinomas are the most common histological types.1 Feline mammary gland adenocarcinomas grow rapidly and metastasize to regional lymph nodes and lungs.1,3 In one study, 27% of cats had evidence of regional lymph node metastasis at the time of initial diagnosis.4 Numerous other studies have documented the aggressiveness of feline mammary gland adenocarcinomas, in that widespread metastasis was noted in the lungs, regional lymph nodes, liver, spleen, adrenal glands, kidneys, and diaphragm.2,3,5,6 Survival times for cats with mammary gland adenocarcinomas treated with surgery alone depend on tumor size.7 Cats with tumors >3 cm in diameter have a reported median survival time of 4 to 6 months, whereas cats with tumors <3 cm in diameter have a median survival time of 2 years. 7
Aggressive surgical excision remains the most widely utilized treatment option for feline mammary gland adenocarcinoma, but owing to the high incidence of metastasis, surgery is rarely curative. In humans, hormonal therapy is the treatment of choice after surgery for patients with hormone receptor-positive cancer.8 Unfortunately, many patients with hormone receptor-positive cancer develop recurrent and/or hormone-refractory disease.8 Chemotherapy is the preferred treatment for patients with metastatic breast cancer that has a negative hormone receptor status.8 Doxorubicina is considered to be one of the most active agents in the adjunctive treatment of high-risk metastatic breast cancer in women.8,9
Based on the high metastatic rate of feline mammary gland adenocarcinoma, additional effective therapies are needed to improve survival. Doxorubicin used alone or in combination with cyclophosphamide has resulted in partial responses in cats with nonresectable local disease or distant metastasis, and it has efficacy against macroscopic feline mammary carcinomas.10–12 To date, no conclusive studies have been published to elucidate the role of adjunctive chemotherapy in feline mammary gland adenocarcinomas, although chemotherapy has had a positive influence on both disease-free interval (DFI) and survival time in canine mammary cancer.13 The purposes of the multicenter study reported here were to retrospectively evaluate the efficacy of adjunctive doxorubicin in cats with surgically resected tumors and to identify any prognostic factors for feline mammary gland adenocarcinomas.
Materials and Methods
Criteria for Selection of Cases
Detailed questionnaires were sent to participating Veterinary Cooperative Oncology Group institutions. Responses from five private practices and four colleges of veterinary medicine were included for analysis. Cats were included in the study if complete medical records were available and the diagnosis of mammary gland adenocarcinoma was confirmed histologically. Cats were included in the statistical analysis if they were treated by surgery followed by adjunctive doxorubicin chemotherapy. Cats were staged according to a modified World Health Organization (WHO) clinical staging system [Table 1]. Cats with pulmonary or pleural metastasis at the start of chemotherapy were excluded from analysis.
Case Data
Medical records were examined retrospectively, and, when possible, follow-up conversations were conducted by clinicians at the participating institutions, pet owners, or referring veterinarians. Data retrieved from the medical records included breed, age, and gender; physical examination findings such as tumor volume, location of the primary tumor, and ulceration of the tumor; clinical staging information such as results of thoracic radiographs, lymph node status, and WHO stage; treatment variables such as type of initial surgery, type of definitive surgery performed after referral, completeness of surgical excision, completion of initial chemotherapy, rescue chemotherapy, and originating institution; histological variables including diagnosis, presence of vascular invasion, presence of lymphatic invasion, mitotic rate, and amount of necrosis; and outcome variables such as local recurrence, development of distant metastasis, DFI, and survival time.
Treatment and Follow-up
All cats had the primary tumor surgically excised and received adjunctive doxorubicin chemotherapy. Doxorubicin was begun at the time of suture removal (approximately 14 days after surgery) at a dose of 1 mg/kg intravenously (IV) every 3 weeks for a maximum of five treatments or until the cat developed progressive disease or concurrent illness. The dosing schedule of 1 mg/kg IV every 3 weeks was chosen by the Veterinary Cooperative Oncology Group. Treatment-related complications and toxicities associated with doxorubicin were also recorded. Each cat was evaluated at regular intervals after the last dose of doxorubicin or until the time any clinical signs of disease relating to the tumor appeared or death occurred, whichever came first. Some cats included in this study received multiple courses of treatment after receiving both surgery and doxorubicin, but follow-up clinical staging was inconsistently performed in most of these cases; therefore, individual treatments could not be evaluated, and survival was used as the final end-point of the study.
Statistical Analysis
All cases that fulfilled the inclusion criteria were included in the statistical analysis. Analysis was on an intention-to-treat basis, except where stated comparisons between numbers of chemotherapy cycles were made. Intention-to-treat gives a pragmatic estimate of the benefit of a treatment rather than of a potential benefit in patients that receive treatments exactly as planned. The DFI was defined as the time from definitive surgery until the development of local recurrence or distant metastatic disease. The DFI was calculated in 60 cats from the date of definitive surgery performed after referral at the listed institutions until development of local recurrence or distant metastatic disease. Seven cats had unilateral radical mastectomies prior to referral, and DFI was defined as the time from the date of surgery until development of local recurrence or distant metastatic disease in those cats. Survival was defined as the time from the original surgery until death from any cause. Median DFI and survival times were calculated using the Kaplan-Meier product limit method.15 Cases were censored for DFI analysis if lost to follow-up or if they were disease-free at the completion of the study. Cases were censored for survival analysis if lost to follow-up, dead from other disease, or still alive. Statistical analyses were performed with a chi-square analysis, analysis of variance (ANOVA), and paired t-tests. Univariate analysis to assess for prognostic value of the covariate was performed by the Kaplan-Meier product limit method with Log-Rank testing.15 Multivariate analysis included all univariate factors with P 0.2 using the Cox proportional hazards model.16 Statistical analyses were performed using a computerized statistical program.b Statistical significance for all testing procedures was set at P 0.05.
Results
Signalment
Sixty-seven cats with histologically confirmed mammary gland adenocarcinoma were included in the study. Cases were diagnosed between June 1994 and December 2002, with the exception of the Animal Medical Center cases, which were diagnosed May 1996 through December 2001. Ages of the cats ranged from 2.5 to 17 years (mean 11.0 years; median 8.5 years). There were eight breeds of cats represented. Domestic shorthair cats were the most common (n=46; 69%), followed by domestic longhair cats (n=8; 12%), Siamese (n=8; 12%), and one each of a Persian, Abyssinian, Balinese, domestic medium hair, and Angora. Fifty-eight (87%) cats were spayed females, six (9%) were sexually intact females, and three (4%) were castrated male cats.
Study Sites
Participating study centers included the Animal Medical Center, New York, NY (n=37); VCA West Los Angeles Animal Hospital, Los Angeles, CA (n=8); Colorado State University, Fort Collins, CO (n=7); Iowa State University, Ames, IA (n=4); Veterinary Cancer Referral Group, Venice, CA (n=3); Veterinary Specialists of South Florida, Cooper City, FL (n=3); the University of California, Davis, CA (n=2); University of Wisconsin, Madison, WI (n=2); and Pet Emergency and Specialty Center, La Mesa, CA (n=1).
Tumor Size and Stage
Twenty-seven (40%) cats had tumors <2 cm in diameter at the time of initial diagnosis. Eighteen (27%) cats had tumors 2 to 3 cm in diameter, 15 (23%) cats had tumors >3 cm in diameter, and the size was unknown in seven (10%) cats. The Kaplan-Meier median survival times were 450 days for cats with tumors <2 cm in diameter and 448 days for cats with tumors ranging from 2 to 3 cm. For cats with tumors >3 cm, median survival time was 200 days (P=0.0289) [Table 2]. Cats with tumors >3 cm in diameter had an increased hazards ratio of 2.49 at a 95% confidence interval of 1.033 to 6.024 (P=0.042) when compared to cats with tumors <3 cm; therefore, cats with tumors <3 cm were 2.4 times more likely to live than cats with tumors >3 cm. Ten (14.2%) cats had ulcerated tumors at the time of evaluation. Ten (14.2%) cats had vascular invasion, and 34 (48.5%) cats had lymphatic invasion on histological evaluation of the tumor. Twenty-seven (38.5%) cats had lymph node metastasis confirmed by histopathology. Twenty-one (31%) cats had tumors classified as WHO Stage I; 13 (20%) cats had WHO Stage II tumors; and 33 (49%) cats had WHO Stage III tumors. There were no significant differences in Kaplan-Meier median survival times or DFIs with regard to WHO stage (P=0.9893). The DFIs were 255 days for WHO Stage I cats, 448 days for WHO Stage II cats, and 442 days for WHO Stage III cats. The Kaplan-Meier median survival times were 338 days for WHO Stage I cats, 448 days for WHO Stage II cats, and 416 days for WHO Stage III cats.
Tumor Location and Surgery
Fifteen (22%) cats had tumors located in the right cranial mammary glands. Twenty-two (33%) cats had tumors of the right caudal mammary glands. Twelve (18%) cats had tumors of the left cranial mammary glands. Eighteen (27%) cats had tumors of the left caudal mammary glands. Twenty-eight (41.7%) cats had one surgical procedure performed prior to referral. Prior surgical procedures included simple mastectomy in 12 cats, regional mastectomy in nine cats, and unilateral radical mastectomy in seven cats. Five (7.4%) cats had two surgeries performed before referral. The majority of cats had definitive surgery performed after referral at the various study institutions. Thirty-five (52%) cats had one surgical procedure at the study institutions. Eighteen (26.8%) cats had two surgical procedures, and nine (13%) cats had three or more surgical procedures at the study institutions. In all, forty-one (61%) cats had a unilateral radical mastectomy. Nineteen (29%) cats had staged bilateral radical mastectomies. Six (9%) cats had regional mastectomies, and only one (1%) cat had a simple mastectomy.
Chemotherapy
Fifty-six (84%) cats completed a full course of five doses of doxorubicin chemotherapy. Eight (12%) cats received only four doses of doxorubicin, and three (4%) cats did not complete the chemotherapy. There were no treatment-related deaths from the doxorubicin. The most common reasons for discontinuing doxorubicin included progressive disease (n=5) and the development of renal disease (n=3). The Kaplan-Meier DFI for cats that received 3 doses of doxorubicin was 104 days. The Kaplan-Meier DFI for cats that received four doses of doxorubicin was 117 days, and the DFI was 442 days for cats that received five doses of doxorubicin (P=0.0002) [Table 3]. Rescue chemotherapy was used in 19 (28.3%) cats in this study. The most common reasons for rescue therapy were the development of metastatic disease (n=13) and local recurrences (n=5). Agents used for rescue therapy varied between institutions and included carboplatinc (n=8), cyclophosphamided (n=5), alternating carboplatin with mitoxantrone (n=3), mitoxantronee alone (n=2), and alternating carboplatin with doxorubicin (n=1).
The Kaplan-Meier median survival time was 448 days for cats that received surgery and adjunctive doxorubicin [Figure 1]. The percentage of cats alive at 1 year was 58.9%, while 37.2% were alive at 2 years, 33.5% were alive at 3 and 4 years, and 16.7% were alive at 5 years. The Kaplan-Meier median DFI was 255 days, which corresponded to a disease-free state in 38.9% of the cats at 2 and 3 years, in 31.1% of the cats at 4 years, and in 10.4% of the cats at 5 years [Figure 2]. At the completion of the study, 17 cats were censored from the Kaplan-Meier survival analysis (13 cats died from other causes, and four cats were lost to follow-up), whereas 26 cats were censored from the Kaplan-Meier disease-free analysis (23 cats were disease-free, and three cats were lost to follow-up). The median follow-up time for cases included in the statistical analysis was 514 days.
Prognostic Factors
The Kaplan-Meier DFI was 183 days for cats that developed metastatic disease, and the DFI for cats without metastasis had not been reached at the conclusion of this study (>1700 days; P=0.0004) [Figure 3]. The Kaplan-Meier median survival time was 331 days for cats that developed metastatic disease, whereas the Kaplan-Meier median survival time for cats without metastasis had also not been reached at the conclusion of this study (>2100 days; P=0.0001) [Figure 4]. Thirty-five (52%) cats developed metastatic disease. Twenty-four (35.8%) cats had pulmonary metastases, five (7.4%) cats had metastases to the regional lymph nodes, five (7.4%) cats had pleural metastases, and two (2.9%) cats developed body wall metastases. The Kaplan-Meier DFIs for cats with nodal, pulmonary, or pleural metastasis were 1122 days, 183 days, and 115 days, respectively (P=0.0115) [Figure 5]. The Kaplan-Meier median survival times for cats with nodal, pulmonary, or pleural metastasis were 1543 days, 331 days, and 188 days, respectively (P=0.0049) [Figure 6].
Cats with anaplastic mammary gland tumors were significantly more likely to have incomplete margins following surgical excision (P=0.0008). Cats with ductular mammary gland adenocarcinomas had a Kaplan-Meier DFI of 306 days, whereas cats with papillary or tubular mammary gland adenocarcinomas had a Kaplan-Meier DFI of >1131 days (median not reached by the end of this study). Anaplastic or high-grade mammary gland adenocarcinomas had a Kaplan-Meier DFI of 95 days (P=0.0057). Vascular invasion was noted in 50% of the cats with incomplete surgical margins. There was a significant correlation between the presence of vascular invasion and the use of additional rescue chemotherapy (P=0.0099). Lymphatic invasion was noted in all cases of anaplastic and high-grade mammary gland adenocarcinomas.
Regarding tumor variables, cats with ulcerated mammary gland adenocarcinomas were more likely to have an advanced WHO clinical stage (P=0.003). Fifty percent of ulcerated mammary gland adenocarcinomas had evidence of vascular invasion, 87.5% had lymphatic invasion, and 75% had lymph node metastasis. Histopathologically confirmed lymph node metastasis was present in 7% of the cats with no evidence of lymphatic invasion and in 71% of cats with lymphatic invasion observed within the tumors. Both lymphatic invasion and lymph node metastasis were positively correlated with advanced WHO clinical stage (P<0.001). No statistically significant prognostic factors were identified with multivariate analysis.
No significant differences were identified between the originating institution and Kaplan-Meier median survival or DFI times; however, the small sample size from various institutions likely reduced statistical power.
Animal Medical Center Subgroup
There were several significant prognostic factors identified for cases included from the Animal Medical Center when the cases were evaluated as a single group [Tables 4, 5]. The Kaplan-Meier DFI for cats with vascular invasion was 222 days, whereas for cats without vascular invasion it was 428 days (P=0.0492). The Kaplan-Meier DFI for cats that completed five doses of doxorubicin was 442 days compared to 87 days for those that did not complete the chemotherapy (P<0.0001). The Kaplan-Meier median survival time for cats that completed five doses of doxorubicin was 641 days versus 125 days (P<0.0001) for those that did not. The Kaplan-Meier DFI for cats that developed metastatic disease was significantly shorter than the DFI for cats without metastatic disease (i.e., 183 days versus >2107 days; median not reached; P=0.0001). The Kaplan-Meier median survival time for cats that developed metastatic disease was also significantly shorter than the median survival time for cats without metastasis (354 days versus 917 days; P=0.0429). The Kaplan-Meier DFI for cats with lymphatic invasion was trending toward significance when compared to DFIs for cats without lymphatic invasion (222 days versus 466 days; P=0.0924). Cats with tumors classified as WHO Stage I had significantly longer Kaplan-Meier DFIs when compared to DFIs for all other stages (1672 days versus 222 days; P=0.0188). The Kaplan-Meier median survival times for cats that underwent a bilateral radical mastectomy, unilateral radical mastectomy, or regional mastectomy were 917 days, 348 days, and 428 days, respectively (P=0.0271). The location of metastatic disease was also prognostic for Kaplan-Meier median survival time (248 days for pulmonary metastases versus 641 days for lymph node metastases; P=0.0359). No significant prognostic variables were identified with multivariate analysis of the Animal Medical Center’s subpopulation of cats.
Discussion
Eighty-three percent of the cats in this study completed five total doses of doxorubicin. The reasons for discontinuing doxorubicin chemotherapy varied, but the most common contributing factors included progressive disease and the development of renal disease. The failure to complete a total of five doses of doxorubicin had a significant impact on both the general population and the Animal Medical Center subgroup [Tables 3, 4, 5]. The level of toxicity from doxorubicin may be under-reported here because of the retrospective nature of the study, but the findings were consistent with previous reports in cats using doxorubicin at a dose of 1 mg/kg.10,11 Nineteen cats received rescue chemotherapy, but specific responses to individual treatments could not be determined because of the wide variation in rescue agents used between institutions.
The mean age (11.0 years) and breeds of cats diagnosed with mammary gland adenocarcinoma in this study were in accordance with earlier reports.1,7 Ovariohysterectomy status among female cats was also consistent with previous reports, in that 86.5% of cats included in this study were spayed females. It is interesting to note that there were three castrated male cats included in this study. No significant differences in survival (792 days, 1092 days, still alive >1500 days) were found for castrated males compared to females. While this lack of a statistical difference may have related to poor statistical power, the actual survival times for the three male cats provided strong evidence that there may not be an actual difference in survival of males compared to females. This finding was consistent with human medicine, where overall survival rates of men with breast cancer are similar to those of women.8
Previous studies have identified various prognostic factors for feline mammary gland adenocarcinomas, such as tumor volume, WHO stage, extent of previous surgery, lymph node involvement, histological grade, and proliferation status (as defined by argyrophilic nucleolar organizer regions [AgNORs] and a nonhistone nuclear protein [Ki-67]).1,7,17,18 The high frequency of cyclin A amplification in feline mammary carcinomas has also been suggested to play an important role in tumorigenesis.19 Significant prognostic factors for survival in this study, based upon univariate analysis, included tumor volume, the development of metastatic disease, and the location of metastatic disease [Table 2]. Significant univariate prognostic factors for DFI included histological subtype, completion of initial chemotherapy, development of metastatic disease, and location of metastatic disease [Table 3]. There were no significant prognostic factors based upon multivariate analysis.
Upon analysis of cases from the Animal Medical Center, several prognostic factors were found that were not detected in the total study population. Even though there were fewer cases (n=37), the subpopulation may have been more uniform because of fewer treatment variations. In this subpopulation, the Kaplan-Meier DFI for cats with vascular invasion was significantly shorter than the DFI for cats without vascular invasion. In addition, the Kaplan-Meier DFI was significantly longer for cats classified with WHO Stage I disease than for cats with all other stages of disease. There was also a significant difference in survival times for cats that received different types of surgery. Cats that received bilateral radical mastectomies had the longest survival times, followed by cats with regional mastectomies and unilateral radical mastectomies. This latter finding is interesting, as earlier studies have demonstrated that the type of surgery had no effect on survival time but did impact DFI.7
The location of metastatic disease was a significant prognostic factor with regard to both DFI and survival. The presence of tumor-positive lymph nodes at initial diagnosis has been described as an independent negative prognostic factor for survival.2 The long-term survival associated with lymph node metastasis in the current study was an unexpected finding. The reason for the difference in survival times based on metastatic location was unknown, but it was theorized that adjunctive chemotherapy may have improved survival times in cats with lymph node metastasis. The long-term survival associated with lymph node metastasis in the cats of this study paralleled results in people, where there is substantial data indicating that local regional lymph node metastasis does not automatically imply further systemic involvement and decreased survival.8 The finding of late nodal metastasis may also suggest that chemotherapy influences the overall metastatic pattern. Consistent with earlier studies, tumor volume was significantly correlated with median survival.7 Cats with tumors <3 cm in diameter had longer Kaplan-Meier median survival times than cats with tumors >3 cm.
Several limitations arose in this study from the retrospective, multicenter nature of the study design. The first limitation was that all cats did not receive the same type of surgery. The original intent of the study was to enroll cats that underwent either a unilateral or bilateral mastectomy. Because of poor case accrual, the criteria were changed by the Veterinary Cooperative Oncology Group in 2000 to include cats that underwent any type of surgery to remove macroscopic disease. By changing the surgical criteria, more cases were eligible for entry into the study, but the decision regarding the type of surgical procedure to be applied to each cat varied between institutions. In addition, all of the histopathological samples were not reviewed by a single pathologist. Unfortunately, many tissue samples could not be located for a second review because of the lapse of time from case accrual to data reporting. Histological diagnosis and classification were based solely on the original histopathology reports submitted upon tumor removal. Another limitation of the current study was the absence of a control population. The comparison of current data to historical controls is affected by shifts in diagnosis and treatment, such as stage migration and improvement in surgical techniques. Nonetheless, survival times following surgical treatment of mammary gland adenocarcinomas have been well documented in cats.1,7
The Kaplan-Meier median survival time for cats treated with surgery and adjunctive doxorubicin chemotherapy was 448 days. Forty-seven percent of cats in the current study were classified as WHO Stage III and had a Kaplan-Meier median survival time of 416 days. Based on historical controls, WHO Stage III cats would be expected to have shorter survival times (approximately 4 to 6 months) with surgery as the sole treatment option.7 Therefore, despite the inherent limitations of this study, the combination of surgery and doxorubicin appeared to provide longer survival times in cats with advanced-stage disease than (historically reported) surgery alone. Additional research into the pathogenesis and behavior of mammary gland adenocarcinomas and randomized, prospective clinical trials are ultimately necessary to define the role of chemotherapy in the treatment of this disease. Additional prognostic factors for feline mammary gland adenocarcinoma must also be identified to determine which cats will benefit the most from adjunctive chemotherapy. Unfortunately, this study found few univariate factors and no significant multivariate factors to be prognostic.
Conclusion
When compared to historically reported outcomes with surgery alone, the study reported here suggested that the combination of surgery and adjunctive doxorubicin chemotherapy resulted in improved long-term survival of cats with mammary gland adenocarcinomas. Significant univariate prognostic factors for DFI included histological subtype, completion of initial chemotherapy, development of metastatic disease, and location of metastatic disease. Significant univariate prognostic factors for survival included tumor volume, the development of metastatic disease, and location of metastatic disease. These results strongly warrant a prospective, randomized, and controlled trial to better define the efficacy of doxorubicin in the treatment of feline mammary gland adenocarcinomas.
Adriamycin; Bedford Laboratories, Bedford, OH 44146
Statview Statistical Software; SAS Institute, Inc., Cary, NC 27513
Paraplatin; Bristol-Myers Squibb, Princeton, NJ 08543
Cytoxan; Bristol-Myers Squibb, Princeton, NJ 08543
Novantrone; Serono Inc., Rockland, MA 02370
Acknowledgments
This report is based on a collaborative study performed by the Veterinary Cooperative Oncology Group. The authors thank the Veterinary Cooperative Oncology Group for its support and the following collaborators for their case contributions:
From Colorado State University: Victoria S. Larson, DVM, MS, Diplomate ACVIM; William S. Dernell, DVM, MS, Diplomate ACVS; Stephen J. Withrow, DVM, Diplomate ACVS, Diplomate ACVIM
From the Veterinary Cancer Referral Group: Sue Downing, DVM, Diplomate ACVIM
From University of Wisconsin-Madison: David M. Vail, DVM, Diplomate ACVIM
From Veterinary Oncology Services: Brenda S. Phillips, DVM, Diplomate ACVIM
Citation: Journal of the American Animal Hospital Association 42, 2; 10.5326/0420110
Citation: Journal of the American Animal Hospital Association 42, 2; 10.5326/0420110
Citation: Journal of the American Animal Hospital Association 42, 2; 10.5326/0420110
Citation: Journal of the American Animal Hospital Association 42, 2; 10.5326/0420110
Citation: Journal of the American Animal Hospital Association 42, 2; 10.5326/0420110
Citation: Journal of the American Animal Hospital Association 42, 2; 10.5326/0420110
Kaplan-Meier curve for overall survival in 67 cats with mammary gland adenocarcinomas treated with a combination of surgery and adjunctive doxorubicin chemotherapy. The green circles denote cats that were censored from statistical analysis (see text for explanation). The red circles represent events (i.e., death secondary to the tumor). The numbers on the vertical axis represent the proportion of cats surviving. The days on the horizontal axis represent the time from surgery until death.
Kaplan-Meier curve for disease-free interval in 67 cats with mammary gland adenocarcinomas treated with a combination of surgery and adjunctive doxorubicin chemotherapy. The green circles denote cats that were censored from statistical analysis (see text for explanation). The red circles represent events (i.e., cats that are not disease-free). The numbers on the vertical axis represent the proportion of cats with no local recurrence. The days on the horizontal axis represent the time from surgery to local recurrence or distant metastasis.
Kaplan-Meier curve for disease-free interval in 67 cats with mammary gland adenocarcinomas treated with a combination of surgery and adjunctive doxorubicin chemotherapy. The cases are grouped according to the presence or absence of metastatic disease. The blue line represents 32 cats that were free of metastatic disease, and the green line represents 35 cats with metastatic disease (P=0.0004). The yellow circles denote cats that were censored from statistical analysis (see text for explanation). The red circles represent events (i.e., cats that are not disease-free). The numbers on the vertical axis represent the proportion of cats free of disease. The days on the horizontal axis represent the time from surgery to local recurrence or distant metastasis.
Kaplan-Meier survival curve for 67 cats with mammary gland adenocarcinomas treated with a combination of surgery and adjunctive doxorubicin chemotherapy. The cases are grouped according to the presence or absence of metastatic disease. The blue line represents 32 cats that were free of metastatic disease, and the green line represents 35 cats with metastatic disease (P<0.0001). The yellow circles denote cats that were censored from statistical analysis (see text for explanation). The red circles represent events (i.e., death secondary to the tumor). The numbers on the vertical axis represent the proportion of cats surviving. The days on the horizontal axis represent the time from surgery until death.
Kaplan-Meier curve for disease-free interval in 67 cats with mammary gland adenocarcinomas treated with a combination of surgery and adjunctive doxorubicin chemotherapy. The cases are grouped according to the location of metastatic disease (P=0.0115). The colored circles on each line represent events (i.e., cats that are not disease-free). The numbers on the vertical axis represent the proportion of cats free of disease. The days on the horizontal axis represent the time from surgery to local recurrence or distant metastasis.
Kaplan-Meier survival curve for 67 cats with mammary gland adenocarcinomas treated with a combination of surgery and adjunctive doxorubicin chemotherapy. The cases are grouped according to the location of metastatic disease (P=0.0049). The colored circles on each line represent events (i.e., death secondary to the tumor). The numbers on the vertical axis represent the proportion of cats surviving. The days on the horizontal axis represent the time from surgery until death.
