To the best of our knowledge, this is the first report of implantation metastasis to an anal fistula after neoadjuvant therapy for rectal cancer in a patient without a history of anal fistula. Persistent anal fistulas with recurrent inflammation are known risk factors for primary cancer [13]. Furthermore, when colorectal cancer spreads to the anal fistula, cancer cells shedding into the intestinal lumen are implanted in the injured mucosa [14]. Wound healing or an inflammatory response to a stimulus activates cancer cell growth, presumably resulting in cancer implantation into an anal fistula [7].
In most cases, the cause of anal fistula is considered to be non-specific cryptoglandular infection, and to a lesser extent, it is associated with inflammatory bowel disease, infections (such as actinomycosis, tuberculosis, lymphogranuloma venereum, human immunodeficiency virus), trauma, surgery, malignancy, and irradiation [15]. This patient did not have any signs of perianal disease, and the pretreatment images did not show any findings of an anal fistula. After completing neoadjuvant therapy, he developed persistent anal pain, and post-treatment images showed the presence of an anal fistula for the first time. Therefore, we believe that the anal fistula formed when he felt anal pain after completing consolidation chemotherapy, but the minimal histological changes in anal glands might have been initiated during neoadjuvant therapy.
In this patient, the pathological findings suggest that the anal fistula metastasis developed through a transluminal process during or after the neoadjuvant therapy. A follow-up colonoscopy 4 weeks after the completion of consolidation chemotherapy showed complete tumor remission. However, the patient presented with persistent anal pain 2 weeks after the last chemotherapy session: this was slightly earlier than the imaging studies that showed complete response. Therefore, the transluminal metastatic process of cancer cells might have begun before tumor response to neoadjuvant therapy.
Chronic radiation proctitis is characterized by mucosal fragility, pallor, spontaneous bleeding, telangiectasias, mucosal edema, strictures, fistulas, and ulceration [16]. Although we could not find any previous reports regarding anal fistula associated with neoadjuvant CRT, radiation therapy can lead to fistula formation as a late complication [17], and a previous report has shown that preoperative CRT is significantly associated with postoperative fistulous complications [18]. We assumed that radiation therapy might induce chronic injury to the corresponding anorectal mucosa, and malignant cells could have been delivered from the proximal large intestine and implanted into the anal fistula. Therefore, the implantation metastasis may have grown in the setting of chronic proctitis. We suggest the following three explanations for the etiology of anal fistula metastasis. First, long-term inflammation caused tumor hypoxia associated with increased radioresistance and a tendency to metastasize [19]. Second, although the location of anal fistula metastasis was within the range of irradiation, CRT and the following chemotherapy were not sufficiently effective to prevent transluminal implantation of cancer cells. Third, an anal fistula is an infectious condition; therefore, it may have become apparent when the patient was chronically immunocompromised by CRT or chemotherapy.
Recently, expectant management may be chosen for patients with rectal cancer, who showed a complete response after neoadjuvant therapy [20]. In our patient, colonoscopy revealed the complete response of the tumor after CRT and consolidation chemotherapy. Due to the presence of the anal fistula, expectant management was not offered. If an anal fistula is suspected after neoadjuvant therapy for rectal cancer in a patient without a history of anal fistula, the treatment strategy needs to be modified. The differential diagnosis of a benign anal fistula or implantation metastasis and the anatomical relationship between the primary tumor and the anal fistula determine the surgical approach, which includes abdominoperineal resection or sphincter-saving surgery [2,3,4,5,6,7,8,9]. In our patient, the primary tumor was originally located 5 cm from the anal verge; therefore, abdominoperineal resection was selected considering the sufficient resection margin of both primary and metastatic lesions. In addition, colorectal anastomosis would have been very proximal to the resection margin of anal fistula metastasis if sphincter-saving surgery had been performed.
When synchronous anal fistula metastasis is suspected, the timing of metastasectomy is controversial. Incisional or excisional biopsy before radical surgery is crucial for planning an appropriate treatment strategy for rectal cancer. However, the decision may be complicated, because several reports have suggested that anal metastasis can occur after injury to the mucosa or previous surgical site [2, 21, 22]. Currently, there is no established surgical management of anal fistula metastasis in rectal cancer. In our patient, abdominoperineal resection was performed as a radical surgery with complete tumor removal. Conversely, local resection with sphincter-preserving surgery and close surveillance can be an option for selected patients, and the surgical margin must be free of tumor cells to prevent local recurrence [5].