How Long Can a Dog Live With Pancreatic Cancer
1 INTRODUCTION
Exocrine pancreatic carcinoma is uncommon in the dog, with an incidence rate of less than 0.5%. 1 In humans, pancreatic neoplasia occurs with an estimated 53 000 new cases each yr in the U.s.a.. 2 In the overwhelming majority of these cases (78%-80%), patients will dice from their affliction, attesting to the aggressive nature of pancreatic neoplasms compared with other cancers and carcinomas arising from other anatomical locations. 2
The biologic behaviour observed in both dogs and cats has been presumed to be akin to that of humans, with only few reports describing the behaviour of exocrine pancreatic carcinomas in veterinary medicine. three-6 These tumours were described as rare in both dogs and cats, with increasing age being a predisposing risk factor. A retrospective analysis of 34 cats provided farther insight as to the behaviour of these tumours, with a high propensity for metastasis and overall poor prognosis (a reported median survival time of 97 days). 3 Those that received handling (surgery ± chemotherapy) had a longer survival time of 165 days. Both abdominal effusion and nonsteroidal antiinflammatory drug apply coincided with poorer outcomes. three The simply literature available in canines is a example serial from 1974 describing the relative frequency of this disease in a variety of domestic species and individual case reports. five-7 A more recent study provided a review of the currently known literature, too equally clinicopathologic abnormalities in 22 new patients, but little information was provided on outcome. 8 In the original study, an increased risk of developing pancreatic carcinoma was noted for older dogs and cats, with the Airedale Terrier being at the highest risk (R = 16.9). 4
In humans, the disease is usually detected at afterward stages (phase IIB or higher) correlating with its generally poor prognosis and clinical effect. 9 Additionally, the presence of metastatic affliction is mutual at the time of diagnosis. Given the shut proximity to the small-scale intestinal tract, complete surgical resection and therefore cure, are much harder to achieve. Larger tumours appear to exist more mutual, likely due to belatedly detection, and in people the 5-year survival rates are xv%. ten
Mutations in KRAS, a proto-oncogene involved in the PI3K, MAPK and RAL-GEF pathways, has been associated with constitutive activation of the gene, subsequent dysregulated proliferation and oncogenesis. 11, 12 These mutations take been extensively described in human ductal pancreatic carcinomas merely only rarely in pancreatic acinar cell carcinomas and were more recently explored in canine and feline pancreatic carcinomas. xi, xiii, 14 Mutations in exons 12 and thirteen of the KRAS gene take not been observed in canine and feline acinar cell carcinomas, which is in accordance with the human literature. 13-15 Ductal carcinomas were not previously evaluated. With this in listen, it was concluded that both canine and feline acinar pancreatic carcinomas may serve as a valuable model for studying the disease in humans. 11
Given the limited information regarding this affliction, especially in canines, the purpose of this retrospective study was to characterize the behaviour, clinical presentation and survival associated with exocrine pancreatic carcinoma in dogs.
2 MATERIALS AND METHODS
Medical records from the Colorado State University (CSU) Veterinary Didactics Hospital from 2002 to 2022 were reviewed for dogs with a confirmed diagnosis of exocrine pancreatic carcinoma. Inclusion criteria included a cytologic or histopathologic diagnosis of a carcinoma of the pancreas with no endocrine advent. If bachelor, follow-upwardly data were obtained past phone or email with the referring master care veterinarian. Tissue samples from all patients were submitted through the CSU Veterinary Diagnostic Laboratory and slides were reviewed by the board-certified pathologist assigned to the case.
Data collected included brood, age, sex, weight, date of diagnosis, clinical signs at the time of diagnosis/presentation to hospital, tumour location (established equally the limb of the pancreas afflicted) and tumour size. Tumour size and stage were correlated with the previously established tumour (T), node (N) and metastasis (1000) staging scheme validated by the American Joint Committee on Cancer (AJCC) (Table i). 10
Primary neoplasm (T) | Prognostic stage groups |
---|---|
Tis—Carcinoma in situ | 0 Tis N0 M0 |
T1—≤2 cm in greatest dimension | IA T1 N0 M0 |
T2—>2 cm and ≤ iv cm in greatest dimension | IB T2 N0 M0 |
T3—>4 cm in greatest dimension | |
T4—Involves the CA, SMA and/or CHA | IIA T3 N0 M0 |
IIB T1-3 N1 M0 | |
Regional lymph nodes (North) | Iii T4 AnyN M0 |
N0—No regional LN metastasis | IV AnyT AnyN M1 |
N1—one-three regional LN metastasis | |
N2—≥4 regional LN metastasis | |
Distant metastasis (Thou) | |
M0—No distant metastasis | |
M1—Afar metastasis |
- Abbreviations: CA, celiac axis; CHA, common hepatic artery (CHA); LN, lymph node; SMA, superior mesenteric avenue.
Diagnosis of pancreatic carcinoma was routinely made either past cytologic or histopathologic confirmation via incisional or excisional biopsy. Slides were processed as per standard protocol of the laboratory without the utilize of immunocytochemistry or immunohistochemistry. Data regarding the results of staging tests, including three-view thoracic radiographs, intestinal ultrasound and/or CT scan was reviewed. Treatment data was recorded including: surgery, chemotherapy, radiations therapy and other adjuvant treatments. The primary consequence measure assessed for each patient was overall survival time. Given the high number of patients euthanized early on in their diagnosis, and patients with metastasis already identified at diagnosis, time to metastasis was not explored. If a postmortem examination was performed, the findings and diagnoses were recorded.
ii.one Cell line validation statement
No cell lines were used in this study.
2.2 Statistical analysis
Kaplan-Meier product limit survival assay was used to judge median survival time (MST). MST was defined as the date of diagnosis to the engagement of decease from whatsoever cause. Patients that were lost to follow-upwards were censored at the time of last contact. The Cox proportional-hazards model was used to determine the upshot of various factors on survival. Several factors were individually evaluated for their impact on survival including: tumour stage, lymph node status and tumour location. A P-value of <.05 was considered significant.
2.iii Results
Search results through the CSU Veterinary Diagnostic Laboratory revealed a total of 23 dogs with a confirmed cytologic or histologic diagnosis of exocrine pancreatic carcinoma. The median age at diagnosis was 11 years (range: v-16 years) and the median weight of patients was 27.three kg (range: 5.v-fifty.ix kg). There were 9 spayed females, 10 castrated males, 2 intact males and 2 intact females. The well-nigh mutual breeds encountered were mixed breed dogs (n = vi), followed by Golden Retrievers (north = 2), English language Cocker Spaniels (northward = ii) and Labrador Retreivers (n = 2). A single Bichon Frise, Akita, Belgian Malinois, Fox Terrier, Standard Poodle, Weimaraner, Dachshund, Yorkshire terrier, Beagle, Border Collie and German Short Haired Pointer were included in this study.
Presenting clinical signs varied with the most common beingness languor (14/23; 61%), anorexia (eight/23; 35%), airsickness (8/23; 35%) and abdominal pain (i/23; iv%). Three dogs (thirteen%) had pancreatic masses that were constitute incidentally. The median duration of clinical signs prior to presentation was 7 days (range: 0-threescore days). Documented comorbidities included one each of the following: well-controlled diabetes mellitus, chronic kidney disease, cardiac arrhythmia, splenic mass, seizures, a right auricular mass and a previous diagnosis of osteosarcoma.
Diagnosis of pancreatic carcinoma was made via cytology in ix (39%), surgical biopsy in 3 (thirteen%) or necropsy in 11 (48%) patients.
The majority of patients had some form of staging performed including a complete blood count and biochemical profile in nine (39%) cases. Bloodwork abnormalities included elevated hepatic enzyme action (alanine aminotransferase and alkaline metal phosphatase) in 2 (22%) dogs, elevated pancreatic enzyme activeness (amylase and lipase) in 2 (22%) and an inflammatory leukogram in 1 (11%). Other staging tests in the 23 dogs included three-view thoracic radiographs in x dogs (43%), intestinal ultrasound in 17 (74%), thoracic CT scan in 1 (4%) or abdominal CT scan in 2 (9%). All patients who had thoracic radiographs performed also had an abdominal ultrasound. Ane patient had both thoracic and abdominal CT scans whereas one patient who had an intestinal CT scan had thoracic radiographs prior for staging. In 2 cases (9%), an abdominal exploratory surgery was performed as a diagnostic measure.
Patient tumours were sorted by size based on the same neoplasm staging scheme (adapted table listed as Table 1). 10 In total, eight (35%) patients had tumour measurements available. Measurements were made either via ultrasonography, at time of necropsy, or during a surgical procedure. Of those, 7 (88%) were classified every bit T3; the other was classified equally T1. Median tumour size was 6 cm (range 0.7-15 cm). Tumour size was not significantly associated with overall survival (P = .99).
In the 20 patients where the location of tumour was recorded, 8 (40%) patients were considered to have diffuse involvement of the pancreas, half dozen (30%) had tumours affecting the right lobe, 2 (10%) had tumours affecting the left lobe and 4 (twenty%) had tumours along the body of the pancreas. Tumour location was not significantly correlated with survival (P = .17).
Ultrasonographic records were also reviewed. Patients with masses along either the left, right, or trunk of the pancreas were typically described as heterogenous masses with distinct borders, though occasionally a cavitated mass was described. In the patients with histologically diffuse carcinoma, the pancreas was described as either grossly normal, hypoechoic and enlarged, or mildly enlarged with rounded borders. In the 8 patients with diffuse carcinoma histologically, 2 had grossly normal pancreas on ultrasound. In those patients where the pancreas was hypoechoic and enlarged, the main imaging differential was either pancreatitis or pancreatic oedema. Metastatic lymph nodes were usually described as enlarged and hypoechoic. Lesions inside the liver confirmed to be metastatic carcinoma were either described every bit heterogenous foci, hypoechoic nodules or targetoid.
Overall stage was assigned to dogs: ane patient was Phase IA (5%), four were stage IIA (17%), iii were stage IIB (xiii%) and xv were stage Iv (65%).
The metastatic rate at the time of diagnosis was 78% (xviii/23) with the most common locations being liver (12/23; 52%) or regional lymph nodes (10/23; 48%). Of these 18 patients, iii (17%) had metastasis to the lymph nodes only. Three patients (17%) had metastasis to the liver only. Ane patient (5%) had positive abdominal lymph nodes and liver metastasis and 3 patients (17%) had lymph node, liver and pulmonary metastasis. 1 patient (five%) had a negative nodal condition with distant lung metastasis observed. Other sites included the spleen in two (xi%) dogs, diaphragm in 2 (11%) and more than diffuse carcinomatosis in ii (11%). 13 patients were classified as N0 (57%), 8 were N1 (35%) and 2 were N2 (9%). Antemortem metastasis was confirmed cytologically in 8 patients (44%), and suspected in 1 patient (6%) based on imaging characteristics. Nine patients (50%) had histologic evidence of metastasis on postmortem examination. Of the patients with postmortem histologic confirmation of metastasis, two patients had previous cytologic evidence of metastasis. No patients with lung metastasis had cytologic confirmation. Ane patient had histologic evidence of pulmonary metastasis on postmortem. Lymph node status was not significantly correlated with overall survival (P = .39).
V patients underwent general anaesthesia for surgical excision or biopsy of their pancreatic mass with merely two of those patients receiving adjuvant therapy post-obit surgery. One patient undergoing surgical excision had suspected testify of local lymph node involvement, whilst a second patient had a concurrent liver mass. Two patients were euthanized at the fourth dimension of surgery: one due to the findings of a pancreatic mass, and one due to an inoperable pancreatic tumour. I patient underwent a pancreaticoduodenectomy. This patient was humanely euthanized vi days postoperatively due to complications associated with the surgical procedure and acute refuse. Another patient underwent partial pancreatectomy and survived the postoperative period; this patient was treated with adjuvant chemotherapy with a planned alternate protocol of carboplatin (300 mg/g2) and doxorubicin (thirty mg/10002), however was lost to follow-up after the kickoff dose of carboplatin, two weeks subsequently surgery. The concluding patient who received a surgical biopsy underwent a single fraction of stereotactic radiations therapy (20 Gy) to the cardinal pancreatic mass 3 weeks postbiopsy. Approximately 1 calendar month after treatment the patient was reassessed via intestinal ultrasound at another facility. Ultrasound showed stable disease compared with previous. Unfortunately, this patient was ultimately lost to follow-upwardly one month after treatment. One patient was diagnosed with a pancreatic carcinoma 1 calendar month prior to presentation to CSU. This patient had the longest reported survival in the population and was euthanized 51 days from the original diagnosis with no therapy provided other than supportive pain medications.
Near 15 dogs were euthanized within 24 hours of diagnosis due to clinical decline or every bit a consequence of their diagnosis. The remaining three dogs were euthanized due to another comorbidity prior to diagnosis of a pancreatic carcinoma. Reasons for euthanasia in these patients included chronic renal affliction, rodenticide ingestion with uncontrollable haemorrhage and uncontrollable seizures. About 11 (48%) patients in full underwent postmortem test. Three pancreatic carcinomas (27%) were considered incidental findings upon necropsy and had non been previously detected prior to euthanasia. Interestingly, two of 3 patients that were euthanized for reasons other than pancreatic carcinoma had no evidence of metastasis at necropsy. These patients were censored from survival analysis.
At necropsy, 3 patients (27%) had lengthened neoplastic involvement of the pancreas with extension into the proximal duodenum. One patient diagnosed with a liver mass was institute to have a primary diffuse pancreatic tumour with hepatic metastasis that was non detected previously. Of the patients euthanized due to their diagnosis of pancreatic carcinoma or a pancreatic mass, necropsy provided new neoplasia-related findings in comparing to original staging tests in iv (36%) dogs. These included, pulmonary metastasis that was not detected via previous thoracic radiographs in 1 (viii%) dog, thyroid carcinoma in 1 (8%), lack of a focal pulmonary mass that was present on CT browse in 1 (8%) and urothelial carcinoma of the urinary bladder in 1 (8%).
The MST for the twenty dogs diagnosed prior to necropsy was 1 day (range 0-51 days, Figure ane). The mean overall survival time was viii days.
3 DISCUSSION
Few studies of canine exocrine pancreatic carcinoma have been published to engagement. 1, 4-6 Due to the limited literature in veterinary medicine as a whole, but more specifically pertaining to dogs, the aims of this written report were to provide more upwardly-to-engagement information on survival and the clinical behaviour of exocrine pancreatic carcinoma in canines. Much of our clinical sentence is based on previous human and feline literature, with niggling ground on the affliction behaviour in their canine counterparts.
The overall MST in this patient population was 1 day. This is probable due to the big proportion of patients in this population that were euthanized either intraoperatively, or very soon after diagnosis due to the anticipated prognosis. The majority of patients (78%) had metastatic illness present in some grade (local or afar metastasis, or both) at the time of diagnosis, though it should be noted that less than half of the patients had full staging with three-view thoracic radiographs, and then distant metastasis may be underreported in this retrospective assay.
Neither tumour size nor nodal metastasis correlated with survival in this patient population. Location of the primary pancreatic mass also did not correlate with survival. Given that the majority of the patients described in this report had some form of metastatic illness, the effect of metastatic disease on overall prognosis is likely not accurately represented. As well, the limited number of cases and short median survival time may have also precluded extraction of any meaningful prognostic factors. It should also be noted that some patients suffered from diffuse effacement of the pancreatic tissue by tumour, and accurate measurements could not be obtained. The event of tumour size on survival may therefore be express by the number of cases with measurements available.
The median elapsing of clinical signs prior to diagnosis was vii days, which over again may attest to the insidious development of the disease before overt clinical progression. In humans, early on detection remains imperative for disease control. 9 Diagnostics including tumour markers in the blood using a carbohydrate antigen (CA19-9) have become widely accepted in humans. 16 If a high level of this antigen is present in high-risk grouping patients (ie, familial history, history of diabetes or pancreatitis, etc), further diagnostics including advanced imaging are pursued. 16 A circulating tumour-associated antigen for pancreatic carcinoma in animals has not been established or attempted in veterinary patients, but may warrant further investigation.
Previous medical history related to the development of pancreatic carcinomas has been discussed in feline and human literature. In a recent retrospective study of feline patients with primary exocrine pancreatic carcinoma, the illness was associated with a previous history of diabetes mellitus. 3 Diabetes has been established every bit a predisposing cistron in up to l% of human patients with pancreatic carcinoma as well. 17 In our study population, only ane dog had a history of diabetes mellitus. The lack of correlation, comparatively, could be due to the inherent limitations posed by the retrospective nature of this study and depression number of patients included in assay.
Ultrasonographically, some patients with diffuse carcinoma had described changes and imaging diagnoses of pancreatitis or pancreatic oedema. This is of import to note as at that place may exist dogs that are clinically assessed as having pancreatitis only may ultimately be diagnosed with carcinoma. Just ii dogs had elevated pancreatic enzyme activity (amylase and lipase) in this study. However, information technology should be noted that the CSU Veterinary Clinical Pathology Laboratory does not include amylase and lipase activeness in their routine biochemical profiles. Therefore, reported cases relied on referral bloodwork for evaluation and may have been underreported in this population. No patient in this population had canine pancreatic specific lipase values reported. Overall, given the reported overlap of clinical signs and possible ultrasonographic features, pancreatic carcinoma should remain a differential for patients with clinically suspect pancreatitis.
In our study population, just one canis familiaris was treated with adjuvant chemotherapy. Instead, the majority underwent surgical excision or were humanely euthanized shortly after diagnosis. The office of adjuvant chemotherapy in pancreatic carcinomas remains controversial, just given the highly reported metastatic rate in cats, people and dogs, is generally recommended. In a previous report performed in cats, there was a statistically pregnant impact on survival in those who underwent surgery and/or chemotherapy. Only ii cats that had surgery did non receive adjuvant chemotherapy and had similar survival times. 3 The patient in the electric current study who was scheduled to receive an alternate protocol of carboplatin and doxorubicin was lost to follow-upwards afterward the commencement dose of carboplatin, and so no conclusions could be made almost the consequence of chemotherapy in this patient population.
Another patient in this study was as well treated primarily with a single fraction of stereotactic radiation therapy. Primary radiations therapy for locally unresectable pancreatic carcinoma in felines and canines has not been previously reported. The single patient in this study where radiation was pursued was ultimately lost to follow-upwards, though ane month subsequently treatment was reported to have stable affliction based on a echo abdominal ultrasound performed at another facility.
Interestingly, a unmarried dog that received no treatment prior to referral lived at least 50 days postdiagnosis. Diagnostic piece of work-up on this patient performed at the CSU VTH 1 month later on the original diagnosis revealed metastasis to the mesentery, sublumbar lymph nodes and evidence of pleural/peritoneal effusion. This may once again attest to the benefits of early detection where patients may live with their affliction until overt progression where subsequent ailments ultimately lead to humane euthanasia.
Treatment for human pancreatic carcinoma is primarily catered to surgical excision via partial pancreatectomy of the affected limb, followed past aggressive chemotherapy regimens. 18 Yet, the office of both neo-adjuvant and adjuvant chemotherapy, especially in nonresectable or marginally resectable tumours is becoming more evident with various reported cytotoxic chemotherapy protocols. 18 As well, checkpoint inhibitors with PD-1/PD-L1 therapy have been gaining interest in patients with pancreatic carcinoma. 19 The part of radiation therapy in patients with nonresectable pancreatic tumours has been more than conspicuously defined every bit of late, with improvements noted in overall survival and local neoplasm control listed in several studies. 20, 21 At this time, the utility of such therapies in veterinary medicine remains largely unknown. However, given the data nowadays in humans, these therapies warrant further investigation.
Several limitations were present in this study that are non unique to similar retrospective analyses. The number of patients included in this study flow was low and many of the medical records reported lacked a standard methodology for handling and/or follow-up. The majority of the patients listed in this study were also euthanized due to the presence of illness before long after diagnosis which may have artificially lowered the reported survival time without pursuit of treatment. Nonetheless, the presence of a few outliers in this population of dogs highlights the need for further investigation through future follow-upwardly studies.
This is the offset comprehensive retrospective analysis of canine exocrine pancreatic carcinoma and our results have demonstrated that pancreatic carcinoma in canines, similar to other species, is characterized past a loftier charge per unit of metastasis and overall grave prognosis. The progression of disease is presumed to be rapid, though with early on detection and adjuvant therapy, a more favourable prognosis may be possible.
Additional studies are necessary to describe more accurate conclusions for pancreatic tumour direction and optimization for clinical command. However, the use of radiations and/or chemotherapy may be underutilized in veterinary medicine. These therapies may provide a college degree of tumour control than previously anticipated based on the aforementioned homo literature surrounding the disease.
Disharmonize OF INTEREST
The authors declare no conflicts of interest.
ACKNOWLEDGEMENTS
The authors would similar to thank the CSU Medical Records department and Veterinary Diagnostic Laboratory for their assistance in obtaining the case data utilized for this manuscript. We would as well like to give thanks Gabrielle Monteith for her assistance with statistical analyses.
Open Research
Information AVAILABILITY Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- 1 , , . Exocrine pancreatic carcinoma. Withrow and MacEwen's Small Animal Clinical Oncology. Vol 401. 5th ed. St. Louis, MO: Elsevier Saunders. 2013.
- ii , , . Cancer statistics, 2016. CA Cancer J Clin. 2016; 66(one): seven- thirty.
- iii , , , , . Feline exocrine pancreatic carcinoma: a retrospective study of 34 cases. Vet Comp Oncol. 2012; 11(3): 208- 218.
- four . Data from eleven United States and Canadian colleges of veterinary medicine on pancreatic carcinoma in domestic animals. Cancer Inquiry. 1974; 34(6): 1372- 1375.
- v , , , , , . Hyalinizing pancreatic adenocarcinoma in vi dogs. Vet Pathol. 2008; 45(4): 475- 483.
- 6 , , , , . Pancreatic acinar cell carcinoma with intracranial metastasis in a dog. J Vet Med Sci. 2007; 69(1): 91- 93.
- 7 , , , . Ultrasonographic and cytopathological diagnosis of exocrine pancreatic carcinoma in the domestic dog and cat. J Am Anim Hosp Assoc. 2001; 37(5): 466- 473.
- 8 , , , , , . Characterization of 22 canine pancreatic carcinomas and review of literature. J Comp Pathol. 2019; 173: 71- 82.
- 9 , , , et al. Early on diagnosis to improve the poor prognosis of pancreatic cancer. Cancer. 2018; 10(2): 48.
- 10 , , , et al. Multinational validation of the American joint committee on cancer 8th edition pancreatic cancer staging system in a pancreas head cancer cohort. J Hepatobiliary Pancreat Sci. 2018; 25(nine): 418- 427.
- 11 , , , et al. KRAS mutations in canine and feline pancreatic acinar prison cell carcinoma. J Comp Pathol. 2016; 155(one): 24- 28.
- 12 , , . Targeting the untargetable KRAS in cancer therapy. Acta Pharm Sinica B. 2019; 9(5): 871- 879.
- thirteen , , , et al. Impact of KRAS mutations on clinical outcomes in pancreatic cancer patients treated with beginning-line gemcitabine-based chemotherapy. Mol Cancer Ther. 2011; 10(x): 1993- 1999.
- 14 , , , . EGFR and KRAS mutational assay and their correlation to survival in pancreatic and periampullary cancer. Pancreas. 2012; 41(3): 428- 434.
- 15 , , , et al. Whole-exome sequencing of pancreatic neoplasms with acinar differentiation. J Pathol. 2014; 232(4): 428- 435.
- 16 , . The clinical utility of serum CA 19-9 in the diagnosis, prognosis and management of pancreatic adenocarcinoma: an evidence based appraisal. J Gastrointest Oncol. 2012; three(2): 105- 119.
- 17 , , , et al. Diabetes mellitus, other medical conditions and familial history of cancer as risk factors for pancreatic cancer. Br J Cancer. 1999; lxxx(eleven): 1830- 1837.
- 18 , , , , . Unresectable locally advanced pancreatic cancer: treatment with neoadjuvant leucovorin, fluorouracil, irinotecan, and oxaliplatin and assessment of surgical resectability. J Clin Oncol. 2013; 31(3): e37- e39.
- 19 , , , et al. PD-ane/PD-L1 and immunotherapy for pancreatic cancer. Cancer Lett. 2017; 407: 57- 65.
- twenty , , , , , . Stereotactic torso radiation therapy for locally advanced pancreatic cancer: a systematic review and pooled analysis of 19 trials. Int J Radiat Oncol Biol Phys. 2017; 97(ii): 313- 322.
- 21 , , , et al. One- vs. three-fraction pancreatic stereotactic body radiation therapy for pancreatic carcinoma: single establishment retrospective review. Front end Oncol. 2017; seven: 272.
Source: https://onlinelibrary.wiley.com/doi/10.1111/vco.12645
0 Response to "How Long Can a Dog Live With Pancreatic Cancer"
Post a Comment