We describe a case of adenomatous polyp of the colon that harbored small foci of signet ring cell carcinoma. The patient was a 64-year-old woman with end-stage renal disease and sepsis who underwent colonoscopy to evaluate the possibility of pseudomembranous colitis. A polyp was found incidentally in the right colon and a biopsy was performed. Histologic examination of the polyp revealed typical features of tubular adenoma without evidence of high-grade dysplasia. However, 2 small foci of signet ring cell carcinoma were identified that infiltrated the lamina propria. In contrast to adenomatous epithelium, the signet ring cells were immunohistochemically positive for cytokeratin 7 and negative for cytokeratin 20, suggesting a metastasis rather than a primary tumor. Multiple random biopsies from the right and left colon, as well as the ileum, exhibited no histologic evidence of malignancy. Subsequently, signet ring cell carcinoma with similar morphology and identical immunophenotype was detected in biopsies from the endometrium, an unusual location for primary signet ring cell carcinoma. Preliminary workup excluded the breast as a possible primary site, but further investigation was not possible because of the patient's death with no autopsy granted. To the best of our knowledge, this is the first reported case of metastatic signet ring cell carcinoma to an adenomatous polyp of the colon. This case illustrates the necessity of submitting all polyps entirely and the importance of examining them carefully.
Primary signet ring cell carcinoma (SRCC) is rare in the colon and rectum compared to conventional adenocarcinoma. The reported incidence ranges from 0.01% to 2.6% among all colorectal cancers.1-3 It is defined by the presence of more than 50% of the tumor cells with prominent intracytoplasmic mucin.4 It may occur as part of mucinous adenocarcinoma or as a diffusely infiltrative process with minimal extracellular mucin. Because of its rarity, a metastasis may need to be excluded before a primary SRCC of the colon is diagnosed, particularly in the absence of typical mucinous carcinoma or conventional adenocarcinoma.1
Adenomatous polyps of the colon and rectum are common benign epithelial neoplasms. Frequently, adenomas may harbor foci of high-grade dysplasia (carcinoma in situ) or invasive adenocarcinoma, consistent with the paradigm that adenomatous lesions are the precursors for invasive colorectal carcinoma.5 Unlike conventional adenocarcinoma, however, colorectal SRCC is seldom associated with an adenoma. In fact, there have been only a few documented cases in which foci of SRCC have been observed in adenomatous polyps.2,3,6 In those cases, SRCC was considered as a colorectal primary and was believed to arise from adenomatous epithelium, because of the presence of an originating point from adenomatous epithelium with intraepithelial signet ring cells and the coexistence of separate foci of conventional adenocarcinoma in the same polyps. To our knowledge, metastatic SRCC to an adenomatous polyp of the colon has never before been reported in the literature.
REPORT OF A CASE
A 64-year-old African American woman sought medical attention for shortness of breath. She was status post-cadaveric kidney transplantation 15 years earlier for end-stage renal disease secondary to hypertension. Her posttransplant course was complicated by renal artery stenosis managed with angioplasty and stent placement. She also had a history of pancytopenia, which was resolved with discontinuation of azathioprine.
On admission, the patient was afebrile and hemodynamically stable. Physical examination revealed bibasilar crackles in the lungs and 2+ pitting edema in the lower extremities. Chest radiographs showed right upper lobe opacity and left lower lobe infiltrate. Polymerase chain reaction performed on blood samples was positive for cytomegalovirus. Her serum creatinine level was 3.1 mg/dL (normal range, 0.6-1.2 mg/dL). The patient was treated with multiple antibiotics and valganciclovir, and also received hemodialysis for renal failure. At the completion of antibiotics nnd valganciclovir regimens, her pulmonary infiltrates and cytomegalovirus viremia were resolved. However, she subsequently developed diarrhea and hematochezia. A flexible sigmoidoscopy was performed, which revealed congestion and edema in the rectum and sigmoid colon. A biopsy from the sigmoid colon at that time showed nonspecific inflammatory changes with erosions. No cytomegalovirus inclusions and no histologic evidence of pseudomembranous colitis were identified.
On her subsequent hospital course, the patient continued to have diarrhea. She also developed abdominal pain, low-grade fevers, and uterine bleeding. Multiple blood cultures grew coagulase-negative Staphylococcus species, and multiple stool cultures were positive for vancomycin-resistant Enterococcus species. Stool assays for Clostridium difficile toxin, which were negative before, became positive. A computed tomographic scan revealed diffuse thickening of the colonic wall, consistent with pseudomembranous colitis. Repeat colonoscopy demonstrated erythematous mucosa in the entire colon. In addition, a pedunculated polyp, measuring less than 1 cm, was identified in the right colon. The terminal ileum was endoscopically normal. Biopsies of the polyp and random colonic and ileal mucosa were performed. Endometrial biopsies were also performed 12 days later.
The patient's final admission occurred 2 months later, when she became hypotensive during hemodialysis. She died shortly thereafter with a clinical diagnosis of sepsis. An autopsy was not obtained.
The colonoscopic biopsies from the right colon polyp consisted of 3 fragments of gray-tan soft tissue, each measuring 0.2 cm in maximal dimension. Multiple random colonoscopic biopsies from the right and left colon, and a single biopsy from the terminal ileum were also received. Histologic sections of the right colon polyp showed typical features of tubular adenoma (Figure 1). Although the number of lamina propria inflammatory cells was increased, no ulceration or surface erosion was appreciated. No stalk and no submucosa were present in these biopsies. One of the biopsies exhibited 2 small foci (each
The random colonoscopic biopsies from the right and left colon showed inflammatory changes similar to those seen in the previous sigmoid biopsies. Careful review of these and previous sigmoid biopsies revealed no evidence of malignancy. No histologic features of pseudomembranous colitis were appreciated. The biopsy from the terminal ileum was histologically unremarkable.
Immunohistochemical stains demonstrated the signet ring cells identified in the tubular adenoma to be strongly positive for pancytokeratin (AE1/AE3, 1:2000 dilution; Chemicon, Temecula, Calif; Figure 4) and negative for CD68 (Dako Corporation, Carpinteria, Calif). Mucicarmine stain was positive for intracytoplasmic mucin. Further immunohistochemical studies revealed that, in contrast to the adenomatous epithelium, the signet ring cells were positive for cytokeratin 7 (1:1000 dilution; Dako; Figure 5) and negative for cytokeratin 20 (Dako).
The endometrial biopsies consisted of multiple minute tissue fragments admixed with blood clots, measuring 0.5 cm^sup 3^ in aggregate. Microscopic examination showed inactive endometrial glands and stroma with areas of coagulative necrosis. No hyperplasia or cytologic atypia was observed in the endometrial glands. However, the endometrial stroma was infiltrated by poorly differentiated adenocarcinoma, with many of the tumor cells exhibiting signet ring cell morphology (Figure 6), histologically similar to those detected in the tubular adenoma of the right colon. In addition, some of the tumor cells showed rudimentary gland formation. Immunohistochemically, the tumor cells were also positive for cytokeratin 7 and negative for cytokeratin 20, phenotypically identical to adjacent benign endometrial glands.
Primary SRCC has been described in virtually every organ that can produce mucin, but is most frequently seen in the stomach.7 In the colon, SRCC differs from conventional colorectal adenocarcinoma in that the former is only rarely associated with a precursor lesion adenoma.2,3,6 In 2 well-characterized cases reported in the literature, SRCC was believed to arise from an adenoma.3,6 The case described by Nakamura et al6 in 1983 was a 4.5-cm rectal tubulovillous adenoma removed surgically that contained 4 foci of SRCC and 4 foci of conventional adenocarcinoma. Serial sections revealed an originating point that connected 1 of the SRCC foci to the adenomatous epithelium. The case reported by Tandon et al3 in 1999 was a 1.5-cm tubular adenoma removed endoscopically from the sigmoid colon that harbored a focus of invasive SRCC. Foci of more differentiated invasive adenocarcinoma and high-grade dysplasia were also present. This case was also considered to be a colonic primary because a thorough workup did not reveal any other sites of involvement. The pathologic observations from these 2 cases3,6 are generally consistent with the findings of Mai et al.2 In that report, foci of SRCC were found in 2 additional cases of adenomatous polyp, one of which also contained a focus of conventional adenocarcinoma. Intraepithelial signet ring cells were identified in both cases. In addition, among 15 cases of adenocarcinoma with a signet ring cell component greater than 50%, 6 cases contained focal areas suggestive of preexisting tubular or villous adenomas.2
We believe that the case described in this report represents the first reported case of SRCC metastasizing to an adenomatous polyp in the colon. The findings that support a metastasis rather than a primary arising from the adenoma include the lack of intraepithelial signet ring cells or an originating point, the absence of high-grade dysplasia in the adenomatous epithelium or concurrent conventional adenocarcinoma, and distinct immunoreactive pattern for cytokeratins 7 and 20. Preliminary workup had excluded the breast as a possible primary by a negative mammography. In addition, computed tomography, chest radiography, and gallbladder sonography performed prior to the discovery of SRCC did not detect any evidence of malignancy in the lungs, pancreas, stomach, gallbladder, common bile duct, or urinary bladder. The patient's poor condition and subsequent death precluded further investigation, such as upper gastrointestinal endoscopy, to rule out a gastric primary.
The primary site of SRCC in our case remains undetermined because of incomplete workup, although an endometrial origin is suspected. The tumor cells seen in the endometrial biopsies were not purely signet ring cells; some of them also formed vague glandular structures. However, a metastasis may not necessarily reproduce the entire morphologic spectrum. In addition, the immunostaining pattern for cytokeratins 7 and 20 was compatible with an endometrial origin. However, this possibility is intriguing because most SRCCs of the endometrium are also believed to be metastatic from extragenital sites, particularly from the breast. The literature contains only 1 documented case of primary endometrial SRCC.8 Interestingly, the tumor growth pattern observed in that case was similar to that seen in our case, in that the tumor cells diffusely infiltrated the endometrial stroma between the benign, inactive endometrial glands. Poorly formed glandular and solid areas were also described in that case.8
Another possibility is that the SRCCs detected in the tubular adenoma of the colon and the endometrium are both metastases from an unknown primary. In a 64-year-old woman, breast carcinoma appears to be a likely consideration. Mammograms can be falsely negative in a substantial number of cases. This is particularly true for infiltrating lobular carcinoma, which may not form a discrete mass and may not have a well-demarcated border.9 Furthermore, it is not uncommon for metastasis to be the initial clinical manifestation of an occult breast carcinoma.9 Lobular carcinoma may also present with distinctive patterns of systemic metastases, preferentially to locations such as peritoneum, retroperitoneum, gastrointestinal tract, and gynecologic organs.9 Studies have shown that disseminated disease is associated with loss of expression of intercellular adhesion molecule E-cadherin, which results in poor cohesion of the tumor cells.10 Whether long-term immunosuppression, as seen in this patient, plays a role in promoting tumor dissemination remains to be investigated.9
Despite the uncertainty of the primary origin, our case illustrates the necessity to submit all adenomatous polyps of the colon entirely and the importance for pathologists to examine them in detail. The metastatic foci of SRCC detected incidentally in the adenoma in this case are tiny and can be easily confused with muciphages. Another diagnostic pitfall is benign signet ring cell change, which may occasionally be encountered in colorectal biopsies, particularly in the setting of pseudomembranous colitis.11-13 Interestingly, benign signet ring cell aggregates that simulate SRCC have also been observed in an adenomatous polyp.11 In all these cases, ulceration of the mucosal surface was always present and the benign signet ring cells were always confined to the mucosal surface admixed with pseudomembranes or within the crypts without breaking the basement membrane. Infiltration of the lamina propria of the mucosa has not been observed.11-13 Non-neoplastic signet ring cells have also been described in endometrial specimens.14 They are detected within the endometrial stroma and are believed to represent vacuolated decidual cells or histiocytes.
1. Connelly JH, Robey-Cafferty SS, el-Naggar AK, Cleary KR. Exophytic signet-ring cell carcinoma of the colorectum. Arch Pathol Lab Mad. 1991;115:134-136.
2. Mai KT, Isotalo PA, Guindi M, Burns BF, Parks W. Intestinal epithelial lesions associated with signet ring cell carcinoma of the colon and small intestine. Pathology. 2002;34:51-56.
3. Tandon M, Sostek M, Klein MA. Focus of signet ring cell carcinoma in an adenoma of the sigmoid colon. Arch Pathol Lab Med. 1999;123:957-959.
4. Hamilton SR, Vogelstein B, Kudo S, et al. Carcinoma of the colon and rectum. In: Hamilton SR, Aaltonen LA, eds. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Digestive System. Lyon, France: IARC Press; 2000:105-119.
5. Vogelstein B, Kinzler KW. The multistep nature of cancer. Trends Genet. 1993;9:138-141.
6. Nakamura T, Nakano G, Sakamoto K. Adenoma of the rectum with multiple foci of signet-ring cell carcinoma: report of a case. Dis Colon Rectum. 1983;26: 529-532.
7. Maehara Y, Sakaguchi Y, Moriguchi S, et al. Signet ring cell carcinoma of the stomach. Cancer. 1992;69:1645-1650.
8. Mooney EE, Robboy SJ, Hammond CB, Berchuck A, Bentley RC. Signet-ring cell carcinoma of the endometrium: a primary tumor masquerading as a metastasis, Int J Gynecol Pathol. 1997;16:169-172.
9. Rosen PP. Rosen's Breast Pathology. 2nd ed. Philadelphia, Pa: Lippincott Williams & Wilkins; 2001:627-687.
10. Jiang WG, Mansel RE. E-cadherin complex and its abnormalities in human breast cancer. Surg Oncol. 2000;9:151-171.
11. Michal M, Chlumska A, Mukensnabl P. Signet-ring cell aggregates simulating carcinoma in colon and gallbladder mucosa. Pathol Res Pract. 1998;194: 197-200.
12. Schiffman R. Signet-ring cells associated with pseudomembranous colitis. Am J Surg Pathol. 1996;20:599-602.
13. Sidhu JS, Liu D. Signet-ring cells associated with pseudomembranous colitis [letter]. Am J Surg Pathol. 2001;25:542-543.
14. Iezzoni JC, Mills SE. Nonneoplastic endometrial signet-ring cells: vacuolated decidual cells and stromal histiocytes mimicking adenocarcinoma. Am J Clin Pathol. 2001;115:249-255.
Tarek A. Bismar, MD; Horacio Maluf, MD; Hanlin L. Wang, MD, PhD
Accepted for publication June 13, 2003.
From the Department of Pathology and Immunology, Lauren V. Ackerman Laboratory of Surgical Pathology, Washington University School of Medicine, St Louis, Mo. Dr Bismar is now with the Department of Pathology, Harper Hospital, Detroit, Mich.
Reprints: Hanlin L. Wang, MD, PhD, Department of Pathology and Immunology, Lauren V. Ackerman Laboratory of Surgical Pathology, Washington University School of Medicine, Campus Box 8118, 660 S Euclid Ave, St Louis, MO 63110-1093 (e-mail: firstname.lastname@example.org).
Copyright College of American Pathologists Nov 2003
Provided by ProQuest Information and Learning Company. All rights Reserved