The large intestine is the site of 6–12 % of all gastrointestinal lymphomas [38–40]. Rectal lymphoma can represent a primary rectal malignancy or metastases from a nodal origin, known as secondary rectal lymphoma. The vast majority of rectal lymphomas are sequelae of systemic lymphoma. Primary rectal lymphoma is exceedingly rare, representing only 0.2–0.4 % of all primary colorectal malignant neoplasms [39, 41–43]. Almost all primary colorectal lymphomas are non-Hodgkin, B cell lymphomas [42, 44]. Very few cases of Hodgkin lymphoma have been reported in the literature and these are often associated with HIV or EBV-infection, inflammatory bowel disease or immunocompromised hosts [45–47].

Differentiation of primary from secondary colorectal lymphoma is necessary because both the survival outcomes and the therapeutic management of the two are distinct. In 1961 Dawson et al. introduced criteria to diagnosis primary gastrointestinal lymphoma [48]. In order to classify a GI malignancy as a primary lymphoma several criteria must be met. These include: no palpable peripheral lymphadenopathy, no mediastinal lymphadenopathy, normal white blood cell count on peripheral blood smear, only mesenteric lymph nodes adjacent to the tumor are involved at laparotomy, and no malignant lymphomatous disease of the liver and spleen. One detriment of this classification system is the difficulty distinguishing between secondary lymphoma and those with primary GI lymphoma who present with widely metastatic disease. There is no standardized classification or staging system for GI lymphoma and lymphoma staging systems, designed for traditional nodal-based lymphoma, fail to provide adequate prognostic treatment guidance for primary GI lesions. Several staging systems exist for GI lymphoma including the Ann Arbor staging with Musshoff modification [49], the international prognostic index (IPI) [50], the Paris staging system [51] and the WHO classification system [52]. The Ann Arbor staging system, originally designed for Hodgkin’s lymphoma, and its Musshoff modification, adopted for extranodal disease in 1977, is a very elementary system. The International Non-Hodgkin’s Lymphoma Prognostic Factors Project developed of the International Prognostic Index (IPI) for patients with a diffuse large B-cell lymphoma (DLBCL) that consisted of Ann Arbor stage, patient characteristics and simple laboratory measurements. The Paris staging system has increasingly gained significance due to its ability to distinguish primary from distant lymphoma manifestations depending on involved organ. Recently there has been greater advocacy for use of the WHO system, which characterizes lymphomas on the basis of morphology, immunophenotype, molecular genetics, and clinical features [53–55]. This system has allowed clinicians to better predict the clinical behavior of the lymphoma as well as modify management to achieve greater therapeutic success.

Secondary lymphoma of the rectum is defined as regional lymph node metastasis to the rectum. In patients with metastatic lymphoma of nodal origin, 5–46 % will have some degree of GI tract involvement [38, 79]. Presenting symptoms of metastatic lymphoma to the rectum are hematochezia and weight loss [56]. Primary therapy for metastatic lymphoma remains chemotherapy. Surgical intervention is generally of little benefit for the treatment of secondary rectal lymphoma unless it is for complications of these lesions such as intussusception, perforation, obstruction or uncontrollable hemorrhage. Lastly, prognosis is poor with overall 5-year survival of only 15 % [56].

ARM is commonly diagnosed in the sixth and seventh decades of life [93–96]. Anal bleeding is the most common symptom in those diagnosed with ARM [93, 94, 96]. The mean duration of symptoms before presentation is 5–6 months [93, 94, 96]. Multiple factors are thought to contribute to a delay in diagnosis of ARM. Frequently these lesions can be amelanotic in up to 20–25 % of cases; additionally, they are often confused with benign diseases such as hemorrhoids due to similar presenting symptomatology [88, 91, 94, 95, 97]. Two thirds of lesions are found in the anal canal or anal verge and approximately a third of cases are found in the distal rectum [92, 93]. While anal melanoma may vary slightly from rectal melanoma in initial presentation or recurrence patterns, recent data suggests there is no difference in survival outcomes by melanoma location within the anorectum [98].

Despite multiple therapeutic approaches including surgical resection, radiation, and systemic therapy alone or in combination, prognosis and survival remain dismal for ARM [95, 98, 99]. More radical surgical resection has not been shown to alter survival outcomes. Even when patients are appropriately pathologically staged after rectal resection there is no survival difference between patients who undergo APR or wide local excision (WLE). There is a higher morbidity associated with APR; however, WLE may be associated with a higher rate of local recurrence [89, 95, 97, 100, 101]. There may be some role for radiation as palliative therapy in locally advanced, recurrent, or metastatic disease, but this too has not improved prognosis.

Commonly, the disease is advanced at initial presentation with regional nodal spread in up to 20 % of patients and systemic metastases in up to 40 % [87, 88, 90–93, 96]. Five-year overall survival is estimated at 20–22 % with a disease-free survival of 16–17 % [87–90, 92, 93, 100, 102]. Recurrence is often systemic and fatal. A predictor of poor outcome is thought to be the presence of perineural invasion in the primary tumor; however, unlike cutaneous melanoma, lymph node status has not shown an impact on survival [95, 100, 103].

Due to the uniformly poor prognosis of ARM, efforts combining surgical resection with effective systemic therapy will be required for the successful management of this disease. Recently survival outcomes have improved in the treatment of cutaneous metastatic melanoma through advances in both immunotherapy and targeted therapy. Ipilimumab, a monoclonal antibody targeting cytotoxic T-lymphocyte antigen 4 (CTLA-4), was the first agent to improve overall survival in patients with metastatic melanoma in a phase III, randomized, control trial [104, 105]. Additionally, vemurafenib, an inhibitor of mutant BRAF, has increased both progression-free and overall survival in a phase III trial in patients with melanoma containing the V600E BRAF mutation [106]. Lastly, various inhibitors of KIT, such as imatinib mesylate, sunitinib, nilotinib and dasatinib, have also demonstrated survival benefit in a subset of patients with metastatic melanoma harboring mutant KIT [107]. BRAF and KIT mutation rates vary in mucosal melanomas according to anatomic site [108, 109]. Thus, it can be hypothesized that those ARM which are more immunogenic may represent better targets for drugs such as ipilimumab which exert its effect through immune system modulation, while those which possess KIT or BRAF mutations may demonstrate a therapeutic response to targeted inhibitors, such as imatinib and vemurafenib. While no current study has specifically examined ARM treatment with these new drugs, new trials utilizing these agents, including ipilimumab, BRAF- and KIT-inhibitors, for the treatment of all mucosal melanomas, are underway.

Most patients diagnosed with NETs of rectum are in their 50s–60s [5, 9–12, 16]. There is a slight male predominance and higher rates of rectal NETs among black and Asian populations [3, 5]. The incidence of colonic and rectal NETs may be increasing, perhaps due to the more frequent use of endoscopy for diagnosis and screening purposes [3, 5–7]. Most patients with rectal NETs are asymptomatic and are diagnosed incidentally in patients undergoing screening or endoscopic testing for unrelated reasons [10, 17].

Abdominal imaging (CT or MRI) combined with lower endoscopy provides significant information to aid in the staging and diagnosis of colorectal NETs. Endoscopic Ultrasound (EUS) has also played a key role in staging and treatment of rectal NETs; by providing data on size, depth of invasion, and lymph node involvement EUS helps determines the feasibility of conservative (endoscopic or transanal excision) management or the necessity of radical surgery (LAR or APR) [19, 25]. Electron microscopy and immunohistochemistry, are often required however for definitive diagnosis. Immunohistochemistry provides information on NE differentiation and bioactive amines. Rectal NE carcinomas, like rectal carcinoids, seem relatively incapable of producing carcinoid syndrome (even with liver metastases). Lastly, when small cell NEC of the rectum is diagnosed on rectal biopsy, which is histologically identical to small cell carcinoma of the lung, a search to rule out a pulmonary primary tumor must be performed.

High-grade colorectal NETs, unlike low or intermediate grade carcinoids, are extremely aggressive tumors with poor prognosis. This tumor is associated with a 58 % 6-month survival rate, and a 6 % 5-year survival rate in some series [16, 110]. All rectal NETs, including well-differentiated tumors, have an overall 5-year survival of 88.3 %, with localized disease having a rate of 90.8 %, regional disease at 48.9 % and those with distant metastases 32.2 % [3]. This finding reflects that most of rectal carcinoid tumors (82 %) are localized at diagnosis, with a median size of only 0.6 cm [123]. Colon NECs proximal to the rectum are more aggressive on average, with a 5-year survival of only 62 % across all stages [3]. These tumors have a high propensity for nodal and distant metastasis. As high as 65–80 % of patients will have nodal or distant metastasis at presentation [110, 126, 127]. Tumor grade, tumor size, depth of invasion, lymphovascular invasion, elevated mitotic rate and lymph node involvement significantly predict malignant behavior in localized rectal NETs [128]. According to one analysis of the literature, metastases were observed in 2 % of patients with rectal NETs measuring less than 1.0 cm, 10–15 % of tumors measuring 1.0–2.0 cm, and 60–80 % in patients with tumors measuring greater than 2.0 cm [129]. Patients appear to have a marginally better prognosis if they present without metastatic disease, have an adenocarcinoma component within their tumor, or respond to chemotherapy. Surgery, particularly in the presence of metastatic disease, may not offer a survival benefit for the majority of patients [127].

There is no standardized management of rectal NETs; as with carcinoids, treatment has been guided by tumor size. Because of their low risk of metastatic spread, localized disease or tumors that are small (<1 cm) and confined to the mucosa or submucosa (T1) can be managed with endoscopic resection or transanal excision. In lesions of 1–1.9 cm, transanal endoscopic microsurgery (TEMs) should be considered, which allows for deeper, full-thickness excision [130]. Larger tumors and those with adverse features should undergo APR or LAR. Palliative resection in advanced disease may be offered particularly where debulking may improve symptoms or relieve obstruction.

There are no good treatment outcomes for patients with metastatic colorectal NETs [16]. Metastatic hindgut NETs are incurable and optimal management requires a multidisciplinary approach with chemotherapy [131]. Palliative chemotherapy with or without radiation therapy has been used for adjuvant therapy and for treatment of metastatic disease without a clear survival benefit. For those few hindgut patients with functional tumors, somatostatin analogs are effective in the management of carcinoid syndrome and may delay disease progression. Liver-directed therapy and surgical debulking can improve the quality of life for some patients.

Due to the high rates of recurrence in colonic and rectal NETs, they require surveillance for recurrence even after successful complete resection at surgery except where the risk of recurrence is very low i.e. pT1a lesions <1 cm. There is little common consensus as to the best surveillance modality, interval period or length of surveillance but the European Neuroendocrine Tumour Society (ENETS) has published guidelines based upon tumor size [120]. Lesions <1 cm that are well differentiated (G1) with no invasion of the muscularis or lymphovascular invasion require no follow-up if resection is complete. Lesions that are between 1 and 2 cm or are higher grade should undergo annual follow-up with endoscopy, EUS and MRI. Lesions that are >2 cm: one endoscopy, CT or MRI scan, and serum marker within the first year. For high grade (G3) tumors: surveillance endoscopy, CT scan and serum marker every 4–6 months in the first year, and thereafter at least annually. ENETS recommends follow-up for at least 10 years [120], while NANETS guidelines recommend surveillance for up to 7 years [16].

Diffuse cavernous malformation (previously diffuse cavernous hemangioma) of the rectum is a rare condition albeit an important differential diagnosis in rectal bleeding. Approximately 130 cases have been described in the literature [133]. The first case was described in 1839 by Buie and Nesselrod in their paper “Erectile tumor of the anus.” Diffuse cavernous malformations comprise approximately 20 % of intestinal angiomas [134]. They can be found anywhere in the GI tract but occur most commonly in the rectosigmoid area, or in 50–70 % of cases [135].

Rectal lymphangiomas are exceedingly rare with only a few reported cases to date [152]. These lesions consist of abnormal dilatation and mass-like proliferation of lymphatic channels. They have been incidentally detected during colonoscopy, but when symptomatic, present with pelvic pain and rectal bleeding [153, 154]. On colonoscopy, they frequently appear pedunculated or as cystic submucosal nodules with a smooth, translucent surface that is easily compressible. Lymphangiomas less than 20 mm in size can be removed by snare polypectomy but larger, sessile, and infiltrating lesions may require surgical resection [155].

Hemangiopericytoma is a rare tumor first described in 1942 by Stout and Murray [156]. It is comprised of profuse proliferation of capillaries surrounded by sheets of rounded, sometimes elongated pericytes [157]. They can become quite large and cause compressive symptoms, obstruction, intussusception and gastrointestinal bleeding [158]. These tumors are variable in their behavior and occur in all age groups without a gender predilection. While some tumors remain localized for decades, some are malignant, with approximately a third developing aggressive invasion or metastases, more commonly in older patients [159, 160].

Metastases can be detected years after removal of the original tumor. Out of 106 cases reported in 1976, 26 were in the retroperitoneum and pelvis [161]. Seven cases of colorectal hemangiopericytoma were reported in 1959 [162]. Their size ranged from 3.5 to 26 cm in greatest dimension. Local recurrence and distal metastases were common.

Treatment consists of complete surgical excision. Due to the rarity of the tumor and unpredictable biological behavior, there remains controversy about their management. Neoadjuvant chemotherapy and adjuvant radiation may have a role in larger and unresectable tumors [163–165].

Squamous cell carcinoma of the colon and rectum is extremely rare with approximately 70 cases reported in the literature [166], comprising approximately 0.1 % of all colorectal neoplasms [167]. The rectum harbors nearly half of these tumors. Williams et al., in 1979 set forth criteria for the diagnosis of primary colorectal squamous cell carcinoma: (1) The lesion may not be a secondary metastasis from another primary lesion. (2) There should be no squamous-lined fistula track, where squamous carcinoma is known to originate. (3) A rectal squamous cell carcinoma should not be an extension from an anal squamous cell primary [168].

The rectum normally does not contain squamous cells, but several theories have been proposed to explain the development of squamous cell carcinoma in this location: (1) Proliferation of pluripotent stem cells, which exist in the mucosal crypts following mucosal injury. (2) Squamous metaplasia resulting form chronic irritation [169], although squamous metaplasia is rarely seen; similarly squamous cell carcinoma is rarely seen. (3) Persistent embryonal nests of committed or uncommitted ectodermal cells remaining in an ectopic site after embryogenesis. This could explain the lower rectal squamous carcinomas, but is unlikely to explain colonic tumors. (4) Squamous differentiation arising in an adenoma. Rare adenomas have been found to have squamous differentiation, which could indicate the parallel adenoma-carcinoma sequence of squamous cells [168, 170]. Squamous carcinomas have clinicopathological features in common with adenocarcinomas; the age and sex distribution of affected patients are similar for adenocarcinoma and squamous cell carcinoma; the anatomic distribution of adenomatous and squamous tumors is similar within the large bowel, although the number of squamous cell tumors is too low to allow for a meaningful statistical correlation. No association has been found between HPV subtypes 6, 11, 16 and 18 and squamous or adenosquamous carcinoma [171]. Immunohistochemical staining of keratin suggests a pluripotent endodermal stem cell origin for both squamous call carcinoma and adenocarcinoma of the rectum [172].

Adenosquamous carcinoma is an extremely rare malignancy with an incidence of 0.025–0.1 % of all colorectal cancers [181]. As its name implies, it has both glandular and squamous histologic components, both of which are malignant and can metastasize. The mean age at diagnosis of adenosquamous cell carcinoma is in the sixth and seventh decades with an equal gender distribution [167, 182]. Although Cagir et al. found the rectum and distal colon to be affected most often [181], others have found adenosquamous carcinoma more frequently in the right colon [166, 182].

The cause of adenosquamous lesions is unknown, but the theories for histogenesis mirror the ones for squamous colorectal cancer. There may be an association with ulcerative colitis [183], polyposis, schistosomiasis, ovarian adenocarcinoma and endometrial adenocarcinoma [184]. The squamous component seems to metastasize more frequently and is more aggressive than the glandular component [184].

Adenosquamous carcinoma seems to be more aggressive than adenocarcinoma and have a worse prognosis stage for stage. Cagir et al. also found that 85 % of their patients presented with regional or metastatic disease [181].

The primary treatment modality for these tumors is surgery. Adjuvant chemotherapy is frequently used but the benefit is unknown in this uncommon disease [185]. The Nigro regime has also been used as an adjunct after surgery [181].

In a recent population based study using the California SEER database, Masoomi et al. identified 99 cases of adenosquamous carcinoma [182]. They found that adenosquamous tumors present with more advanced disease and more poorly differentiated tumors. The 5-year survival was worse compared to adenocarcinoma with an increased overall mortality. They concluded hat adenosquamous tumors should be considered a poor prognostic factor in patients with colorectal cancer. The overall 5-year survival is 30 % or less [182] and the mean survival is 12 months [181].

Sarcomatous tumors of the colon and rectum are rare and include tumors such as fibrosarcoma, angiosarcoma, leiomyosarcoma and GIST. Before the discovery that most leiomyosarcomas are in fact gastrointestinal stomal tumors, 95 % of colorectal sarcomas were attributed to leiomyosarcomas. Their earlier described incidence of 0.07–0.1 % of all rectal malignant tumors is in fact much lower [186]. They are most common in the lower third of the rectum [186, 187]. They remain difficult to diagnose and are very aggressive with a poor prognosis.

While GISTs arise from interstitial cells of Cajal and express KIT, leiomyosarcomas have a distinct lineage from smooth muscle cells and do not express KIT. Our previously held notions on leiomyosarcoma need to be redrafted with newer studies using modern diagnostic criteria, based on findings from true leiomyosarcomas rather than findings from GISTs.

Gastrointestinal stromal tumors (GISTs) have been recognized as the most common mesenchymal tumors of the GI tract. Prior to the advent of immunohistochemical diagnostics, most GI stromal tumors were identified as leiomyosarcoma. GISTs occur in 1.1–2 persons per 100,000 [197]. The rectum is the third most common site for GIST, accounting for 4–5 % of all GIST tumors [190, 198]. The tumor occurs more commonly in males and usually between the fifth and seventh decades of life [190].