Colorectal Plasmacytomas: A Retrospective Study of Nine Dogs
Nine cases of colorectal plasmacytomas diagnosed between 1998 and 2001 were reviewed. Treatment consisted of complete surgical resection when possible. Two dogs had multiple plasmacytomas. Two dogs had local recurrence of the tumor at 5 and 8 months after resection. Two dogs were alive at 20 and 23 months with no recurrences at the time of follow-up. The median survival time was 15 months (range 5 to 33 months). Colorectal plasmacytomas are similar to mucocutaneous plasmacytomas, in that they tend to progress slowly and do not recur with complete excision.
Introduction
Plasmacytomas are extramedullary tumors of the B-lymphocyte cell line. Other tumors of this cell line include solitary plasmacytomas of bone and multiple myeloma.1 Studies have reported that plasmacytomas of the skin and mucosa tend to be benign and rarely recur with complete surgical excision.2,3 These studies consisted mainly of plasmacytomas of the skin and oral mucosa, with very few tumors originating at other sites. In a review of 75 cases of canine mucocutaneous plasmacytomas, only three (4%) cases were rectal plasmacytomas.2 In another study of 131 dogs, only two cases occurred in the rectum, and there were no colonic plasmacytomas reported.3
The purposes of this study are to report the period prevalence of colorectal plasmacytomas from January 1998 to December 2001 at the University of Pennsylvania and to describe the presentation, diagnosis, treatment, and tumor behavior of nine dogs that were available for follow-up.
Materials and Methods
The surgical pathology records of the Laboratory of Pathology at the University of Pennsylvania, School of Veterinary Medicine, were searched for all cases of plasmacytomas that had been diagnosed between January 1998 and December 2001. The histopathology reports for all cases of rectal and colonic plasmacytomas were reviewed by one pathologist (Goldschmidt) before inclusion in the study. Cases were included in the study if the diagnosis of rectal or colonic plasmacytoma was confirmed and follow-up information was available from either hospital records or the owner.
Histopathological evaluations of the surgical margins were noted in the original biopsy reports for four cases. The surgical margins for the remaining cases could not be evaluated, as the entire mass had not been submitted for histopathological examination. The presence or absence of amyloid within the neoplasm was noted and confirmed by staining with Congo red and examination using polariscopy for birefringence. No attempt was made to further categorize these tumors according to the subclassification system reported by Platz et al.4 Immunohistochemistry was not performed, as cell morphology was considered to be diagnostic in all cases.
The age, gender, and breed of each dog were determined from the information provided by the referring veterinarian on the histopathology submission forms. The tumor location, biopsy margins, and presence of amyloid were determined by the histopathological appearance. Additional information (e.g., clinical signs, radiographic findings, abdominal ultrasound findings, endoscopy findings, complete blood counts, serum biochemical analyses, type of surgical procedure performed, additional histopathology results, tumor recurrence, time to recurrence, and date and cause of death or reason for euthanasia) was retrieved from the medical records obtained from the primary veterinarian (n=7) or by telephone conversation with the owner (n=4).
The time to recurrence was defined as the time from removal of the original tumor to the time the tumor was found on physical examination. Tumor recurrence was confirmed with histopathology. The survival time was defined as the time from the date of the earliest histopathology report to the death of the animal, or, if alive, to the time of follow-up evaluation.
The mean age of dogs with colorectal plasmacytomas was compared to the mean age of dogs with plasmacytomas at other sites using an unpaired t-test.a The Fisher’s exact test was used for breed predilection.a
Results
Between January 1998 and December 2001, a total of 56,175 biopsies from dogs were accessioned by the Laboratory of Pathology at the University of Pennsylvania, School of Veterinary Medicine. Of these accessions, 55% were neoplasms and 751 were plasmacytomas, which represented 1.3% of all canine submissions and 2.4% of the canine neoplasms. Plasmacytomas arising from the rectum and colon (n=30) represented 4% of all plasmacytomas. The most common sites of origin for plasmacytomas were the skin (86%) and the mucous membranes of the oral cavity and lips (9%), with 1% arising at other sites (e.g., stomach, small intestine, genital system, eye, spleen).
With the small number of cases (n=30) to evaluate, there was no statistically significant breed or gender predisposition for developing plasmacytomas of the colon and rectum. Several breeds had a predisposition for developing plasmacytomas in general, including the American cocker spaniel (odds ratio [OR] 3.6; 95% confidence interval [CI] 3.0 to 4.5; P<0.0001), English cocker spaniel (OR 10.5; 95% CI 5 to 22; P<0.0001), and the West Highland white terrier (OR 2.6; 95% CI 1.6 to 4.2; P<0.0001).
A significant difference (P=0.01) was seen in the mean age of 10.7±1.9 years for dogs with colorectal plasmacytomas (n=28; median 10.5 years; range 5.5 to 13.5 years) versus the mean age of 9.3±2.8 years for dogs with plasmacytomas at other sites (n=695; median 9.5 years; range 1 to 18 years).
Of the nine dogs diagnosed with colorectal plasmacytomas that were available for follow-up, five had solitary rectal tumors, two had solitary colonic tumors, and two had both rectal and colonic plasmacytomas. The signalment data and presenting clinical signs for these nine dogs are presented in Table 1. Information on tumor size and the duration of clinical signs prior to diagnosis was not available. Diagnosis of the plasmacytomas was by digital rectal examination (n=9), abdominal ultrasonography (n=1), and endoscopic biopsy (n=1). No significant abnormalities were found in the complete blood counts (n=6) or serum biochemical profiles (n=6). Specifically, hyperproteinemia was not noted in any of the dogs. Serum protein electrophoresis was performed in one dog (case no. 2) and was normal.
The histopathological appearance of the plasmacytomas in this study was the same as that previously described for cutaneous and mucocutaneous plasmacytomas.5 The mucosa overlying the mass was often eroded or ulcerated and covered by a fibrinonecrotic exudate. Where the mucosa was present, there were areas of infiltration of the lamina propria by the neoplastic plasma cells. However, many of the masses were located only within the submucosa, and they infiltrated the outer muscle layers. The neoplastic cells formed sheets or nests that were surrounded by a fine, fibrovascular stroma. The neoplastic cells were monomorphic or pleomorphic with a peripherally located hyperchromatic nucleus, clumped chromatin at the periphery of the nucleus, and a central nucleolus. The cells had variable amounts of amphophilic cytoplasm, and cell margins were distinct. The neoplastic cells showed considerable nuclear pleomorphism with binucleate and multinucleate cells. When amyloid was associated with the neoplastic cells, it was found as small islands surrounded by the neoplastic cells or as large sheets of amyloid and occasional multinucleated, foreign-body giant cells. Amyloid was present in three tumors in this study (case nos. 4, 5, 7).
In all dogs, treatment consisted of surgical removal of the mass(es) whenever possible [Table 2]. Removal was by superficial, submucosal resection (n=8); 360° full-thickness resection and anastomosis via a rectal pull-through technique (n=2); and colonic resection and anastomosis via a laparotomy (n=1). One dog (case no. 6) had multiple colonic masses diagnosed via colonoscopy, which were not removed.
The minimum follow-up time was 12 months. At the time of last follow-up, the median survival time was 15 months (range 5 to 33 months). Seven of the dogs were euthanized or died at a median of 15 months (range 5 to 33 months) after diagnosis. Two dogs were alive with no recurrences at 20 and 23 months.
Discussion
Plasma cell tumors can occur as intraosseous tumors, extramedullary tumors, multiple myeloma, or immunoglobulin M macroglobuminemia.1 Diagnosis of plasmacytoma is based on histopathology or cytology. Extramedullary plasmacytomas have been diagnosed in the oral cavity, gastrointestinal tract, trachea, skin, brain stem, spinal cord, lymph nodes, and abdominal viscera.2,3,6–19 Canine rectal and colonic plasmacytomas occur infrequently.
Colorectal plasmacytomas tended to occur in older animals in this report. This finding was consistent with previous studies on extramedullary plasmacytomas, cutaneous plasmacytomas, and mucocutaneous plasmacytomas in which the mean age range was 9.4 to 9.7 years.2,3,20 In the study reported here, there was no breed predisposition for colorectal plasmacytomas. Other studies have shown the American cocker spaniel to have a predisposition toward plasmacytomas of extramedullary origin, mucocutaneous origin, and cutaneous origin.2,3,5,20 No significant hematological or biochemical abnormalities were found in any of the dogs in this study, which was consistent with a retrospective study by Rakich et al. involving 75 cases of canine mucocutaneous plasmacytomas.2
Amyloid deposits in plasmacytomas have been previously evaluated and are of immunoglobulin origin (primary amyloid) rather than reactive (secondary) amyloid.16 The significance of the presence of amyloid in plasmacytomas is unknown.16 It has been reported that amyloid occurs in 3% to 4% of cutaneous plasmacytomas.5,20 In a study of mucocutaneous plasmacytomas, only two (3%) of 75 cases had amyloid.2 Amyloid was present in four (13%) of the 30 cases of colorectal plasmacytomas that were diagnosed at the University of Pennsylvania from 1998 to 2001.
Two dogs (case nos. 2, 6) had multiple plasmacytomas. Both dogs had rectal masses diagnosed by digital rectal examination, and further diagnostic testing was performed to evaluate the extent of the disease process. A colonic mass was found via abdominal ultrasonography in case no. 2. Multiple colonic masses were found upon colonoscopy in case no. 6. It was not possible to determine if the multiple masses in these dogs represented metastasis or if these dogs had multiple primary masses. Case no. 2 was the only dog to have a lymph node biopsy. Histopathology revealed increased numbers of plasma cells, suggesting immune stimulation or metastasis. Other dogs in this report may have had other undiagnosed tumors, as abdominal ultrasonography and/or colonoscopy were not performed in every case. In a study of cutaneous plasmacytomas, 10 of 75 dogs had multiple tumors diagnosed.2
Only one dog in this study was euthanized for progression of the disease process, and this dog had multiple masses that were not completely removed. This finding was consistent with previous studies in which only four of 57 dogs were euthanized for reasons related to plasmacytomas; the implication is that plasmacytomas are not aggressive tumors.2
Plasmacytomas recur if incompletely excised.3 In one study of 131 dogs with extramedullary plasmacytomas, there were neoplastic cells at the surgical margins of the excised tumors in eight of the nine dogs that had local recurrence.3 Similarly, in the study reported here, two of eight dogs in which the tumors were removed, had recurrences. One dog had neoplastic cells at the surgical margin, and the other dog had very narrow, clean surgical margins. While the goal of surgery is to obtain clean margins to decrease the chances of recurrence, one dog in this study was euthanized because of surgical complications. The risks of aggressive surgical procedures should be considered when treating dogs for plasmacytomas, as the long survival times of the dogs in this study (including one dog with multiple, nonresected masses) indicate that the disease may progress slowly.
In people, extramedullary plasmacytomas are rare and account for 4% of all plasma cell tumors; 869 human cases were reported from 1905 to 1997.21 Most of these extramedullary plasmacytomas occurred in the upper digestive tract (n=714; 82.2%), and only 15 (1.7%) cases occurred in the colon or rectum.21 Treatment for a majority of the cases consisted of surgery with or without radiation therapy.21 Outcomes for these patients showed 61.2% (n=536) with no recurrence, 21.9% (n=190) with recurrence, and 15.8% (n=137) developing multiple myeloma.21 Information is limited regarding the behavior of plasmacytomas of the colon and rectum in humans.
In this study, conclusions were difficult to make because of the small number of cases and the inherent weaknesses of retrospective studies (e.g., incomplete staging procedures, incomplete observations in the medical records, no scheduled follow-up, and follow-up information based on the owner’s observations).
Conclusion
Based on the nine dogs in this study, it appeared that canine colorectal plasmacytomas behave similarly to other extramedullary plasmacytomas, in that they recur with incomplete excision and are not associated with systemic disease. The long survival times of the dogs in this study indicated that colorectal plasmacytomas can progress slowly even if surgical excision is not possible. However, complete surgical excision was curative for many dogs. Based on the results of the current study, further investigations regarding the use of chemotherapy and radiation therapy for inoperable, recurrent, and metastatic plasmacytomas are warranted.
Statexact; Cytel Software, Cambridge, MA 02139
Acknowledgment
The authors thank Fran Shofer for her assistance in statistical analysis.
