Carcinoid tumour associated with enterovirus infection
- Correspondence to Dr John K Chia, EV Med Research LLC, 25332 Lomita Avenue #170, Lomita, CA 90717, USA;
- Accepted 21 December 2010
- Published Online First 28 January 2011
Enteroviruses commonly infect the gastrointestinal tract, and replication of enteroviruses has been well documented in the Peyer patches of the small bowel. Chronic enterovirus infection has been found in the stomach and terminal ileum of patients with myalgic encephalomyelitis/chronic fatigue syndrome. The authors report the unexpected finding of enterovirus VP1 protein, by immunoperoxidase staining, in carcinoid tumours found in one patient with myalgic encephalomyelitis/chronic fatigue syndrome and another patient with chronic lower quadrant abdominal pain, and suggest a possible association between enteroviruses and tumorigenesis.
Carcinoids are well-differentiated neuroendocrine tumours, derived from primitive stem cells in the gut wall, most commonly found in the distal small bowel and appendix.1–3 Enteroviruses replicate in Peyer patches of the small bowel,4 and persistent enterovirus infection has been demonstrated in the terminal ileum of a patient with myalgic encephalomyelitis/chronic fatigue syndrome.5 In this report, one myalgic encephalomyelitis (ME)/chronic fatigue syndrome (CFS) patient with chronic right-lower-quadrant pain was found to have a carcinoid tumour in the terminal ileum, and enterovirus protein was demonstrated in the proximal margin of the carcinoid tumour by immunoperoxidase staining. A second patient was found to have enterovirus protein in the carcinoid tumour found in the appendix removed at the time of prophylactic gynaecological surgery for right-lower-quadrant pain.
A 44-year-old white female 20 years ago developed bedridden fatigue, headache, myalgia, sore throat, tender cervical lymphadenopathy, cognitive dysfunction and postexertional malaise fulfilling all the necessary criteria for the diagnosis of ME/CFS.6 She continued to have debilitating fatigue and intermittent nausea, right lower abdominal pain despite a total abdominal hysterectomy/right salpingo-oophorectomy. An upper gastrointestinal endoscopy showed mild antral gastritis, and a colonoscopic examination showed a carcinoid tumour in the distal terminal ileum. She underwent an ileocolectomy and left salpingo-oophorectomy. The remaining carcinoid tumour was present in the mucosa and extended to the submucosa and muscularis mucosa; 3/9 lymph nodes were positive for metastasis. Immunoperoxidase staining reproducibly demonstrated enterovirus protein in the proximal margin of the tumour (figure 1A–C), but no staining was seen with CMV-specific antibody (figure 1D). The staining was markedly positive in the superficial nests of tumour cells, whereas more distal or deeper tumour cell nests had minimal or no staining (figure 1E, F). A section of the caecum and stomach biopsies also showed extensive enterovirus protein (data not shown). Terminal ileum specimens taken from five patients who underwent ileocolectomy performed for colon cancer and biopsies of terminal ileum of 10 patients with ME/CFS fulfilling the CDC criteria6 did not show any evidence of enterovirus protein in the mucosa. Enterovirus RNA was not detected in an extract of the paraffin-embedded carcinoid tumour by quantitative RT-PCR using procedures described previously.4
A 45-year-old female with a family history of ovarian cancer underwent prophylactic total abdominal hysterectomy/bilateral salpingo-oophorectomy and appendectomy for chronic right lower abdominal pain and prevention of ovarian cancer. The pathology of the appendix tip revealed incidental carcinoid tumour, which repeatedly stained positive for enterovirus protein (figure 1G,H) but stained negative for CMV protein in the mucosa (not shown). Three specimens of appendices from patient who underwent TAH/BSO and appendectomy for other gynaecological disorder did not show any enteroviral staining (not shown).
In the intestinal tract, carcinoid tumour deriving from primitive stem cells develops deep in the mucosa, growing slowly and extending into the underlying submucosa and mucosal surface. In patient 1, the stainable enterovirus protein was observed in nests of carcinoid cells located in intestinal mucosa of proximal tumour margin but quickly disappeared as more distal sections and the deeper layer of the tumours were examined. The positive finding is unlikely to be non-specific staining of an unknown antigen of the carcinoid tumour, since only a small area stained. In the second patient, the epithelium of the appendiceal mucosa immediately adjacent to the carcinoid tumour was found to have enterovirus protein. Enterovirus RNA was not found in the two specimens, but the areas of viral involvement were quite small.
The significance of these findings is unknown. The first patient had ME/CFS, which has been strongly associated with chronic enterovirus infections, evidenced by enterovirus VP1 protein in her stomach biopsy and colon epithelium, but the presence of viral protein in the carcinoid tumour was unexpected. The second patient has unexplained chronic right lower quadrant pain which led to a prophylactic TAH/BSO and appendectomy. Whether the presence of carcinoid tumour was related to the pain was unclear, but the intermittent pain did improve after surgery.
We recently demonstrated enteroviral protein, RNA and non-cytopathic viruses in the stomach biopsies, and high levels of enterovirus RNA in the Peyer patches of terminal ileum long after the onset of ME/CFS.4 Unlike gastric and colonic mucosa, immunoperoxidase staining of the small bowel, including the terminal ileum, has not demonstrated enteroviral protein in the mucosa but only in the muscle layers.5 The staining clearly showed that the viral infection started at one end of the tumour. Although speculative, one of the neuroendocrine stem cells could be persistently infected with enterovirus and grew without inhibition or other regulatory mechanisms. However, the lack of stainable enterovirus protein in the distal and deeper parts of the tumour suggests that further growth of the tumour may not depend on the presence of viral infection. An alternative explanation would be that observation merely represents a chronic enterovirus infection of the proximal part of the tumour, which has not been reported previously, and has no causative relationship with tumorigenesis.
The aetiology of carcinoid tumours is not known, but numerous chromosomal abnormalities have been observed7 8: losses of bands 18q22-qter (67%), 11q22-q23 (33%) and 16q21-qter (22%) with a gain of band 4p14-qter (22%). Following Coxsackievirus B-3 (CVB-3) infection and lysis of highly susceptible Buffalo green monkey kidney cells, the regrown cells were resistant to CBV-3 infection and found to have chromosomes alteration.9 Furthermore, viruses are known to manipulate the host ubiquitin-proteosome system for their own benefits. CVB-3 infection results in downregulation of several host proteins, such as cell-cycle protein cyclin D1, tumour suppressor p53, and transcription activator b-catenin in infected HeLa cells.10 Whether enteroviruses could trigger tumorigenesis in the intestinal tract remains speculative and will need further investigation of larger cohorts and confirmation.
Carcinoid tumour was detected in the terminal ileum of an ME/CFS patient with evidence of persistent enterovirus infection of the stomach and colon.
Enterovirus VP1 protein was detected in the proximal edge of the carcinoid tumour consistent with an associated enterovirus infection.
Whether enterovirus infection of intestinal tissues can trigger tumorigenesis will require further study.
Funding EV Med Research.
Correction notice This article has been corrected since it was published Online First. It is now a Short report.
Competing interests None.
Patient consent Obtained.
Ethics approval Ethics approval was provided by the Providence Little Company of Mary Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.