The Differential Diagnosis and Interdisciplinary Treatment of Anal Carcinoma

Abstract

Background

Anal carcinoma accounts for 2–4% of all cases of colorectal and anorectal carcinoma. Its peak incidence is from age 58 to age 64; women are affected somewhat more commonly than men. Its incidence has risen markedly in the past three decades.

Methods

This article is based on a selective review of the literature, including the guidelines of the National Comprehensive Cancer Network and the European Society of Medical Oncology.

Results

Anal carcinoma is often an incidental finding. About 85% of newly diagnosed cases are associated with an HPV infection with strain 16, 18, or 33. Radiochemotherapy with 5-fluorouracil and mitomycin C is the treatment of choice. The 5-year survival rate is 80–90%. Primary surgery with curative intent is indicated only for well-differentiated carcinoma of the anal margin (T1, N0). 10–30% of patients now undergo radical resection. The utility of endosonography and positron emission tomography for staging is debated and needs further study.

Conclusion

The treatment of patients with anal carcinoma requires a specialized multidisciplinary approach in accordance with the current evidence-based guidelines. The potential role of prophylactic vaccination against oncogenic types of HPV in the prevention of anal carcinoma merits further investigation.

Anal carcinomas account for 0.3% of all malignant tumors, 1 to 2% of all gastrointestinal tumors, and 2 to 4% of all colorectal and anorectal tumors (1, 2). A total of 7060 new cases were recorded worldwide in 2013 (3). Women are affected somewhat more frequently (0.5 to 1 per 100 000) than men (0.3 to 0.8 per 100 000) (4). The annual incidence rate worldwide is 4 to 7 per 1 million population, with frequency peaking between the ages of 58 and 64 years (5). Although incidence has increased substantially over the last 30 years—particularly in HIV patients, among whom the most recently reported incidence density rate in Germany is 34 or more per 100 000 patient-years—there are no international, globally accepted, guidelines for screening in high-risk groups (6).

Anatomy and specification

According to the definition of the Union for International Cancer Control (UICC), the surgical anal canal is a canal 3 to 4 cm in length, running from the upper edge of the internal anal sphincter (the levator, or anorectal ring) to the anal verge.

Lymph drains from the lowest portions of the rectum and the uppermost portion of the anal canal into the perirectal lymph nodes and those along the internal iliac artery. Distally to the pectinate line, lymph drains into the inguinal lymph nodes and the external iliac lymph nodes (7, 8) (Figure 1).

Figure 1.

Anatomy and lymph drainage of the anal region (7)

(from: Becker HD, Hohenberger W, Junginger T, Schlag PM. Tumoren der Analregion. In: Chirurgische Onkologie. Stuttgart, New York. Thieme 2002. Reproduced with the kind permission of Thieme publishers)

Etiology, pathogenesis, and prophylaxis

Several factors can predispose an individual to anal carcinoma. The most significant of these is infection with a strain of HPV (Box 1). Approximately 85% of newly diagnosed cases of anal carcinoma are associated with HPV infection (4). In a meta-analysis conducted by the International Agency for Research on Cancer (France) and the Fred Hutchinson Cancer Research Center (Seattle, USA) the most common strain in anal carcinomas was HPV 16 (73.4%), followed by HPV 18 (5.2%) and HPV 33 (4.8%) (9– 11).

Box 1. Risk factors for anal carcinoma.

• Human papilloma virus (HPV) (85%)

• First sexual intercourse at an early age

• Promiscuity

• Genital infections (chlamydia trachomatis, herpes simplex)

• History of sexually transmitted diseases

• Anal intercourse in men

• HIV infection

• Immunosuppression

• Radiotherapy/chemotherapy

• Smoking (particularly in premenopausal women)

• Other HPV-associated neoplasias

In a recent, large Taiwanese study, Lee et al. demonstrated that the presence of hemorrhoids, abscesses, fistulas, and other benign lesions was not associated with a long-term increase in the risk of anal carcinoma (12).

In Germany, unlike the USA, the Standing Vaccination Committee (STIKO, Ständige Impfkommission) currently recommends only the vaccination of girls between 9 and 14 years of age against oncogenic HPV viruses (quadrivalent vaccination) to prevent cervical cancer. To date there is no recommendation for vaccination to prevent anal carcinoma, as it has not been shown that vaccination reduces the incidence of anal carcinoma as it does for cervical cancer. A recent study showed that vaccination inhibited the development of anal intraepithelial neoplasia as a precursor of anal carcinoma in 54.9 to 78.6% of homosexual men (77.5% for HPV 6, 11, 16, or 18-associated anal intraepithelial neoplasia in the intention-to-treat population) (13). As it has been demonstrated that the efficacy of vaccination is almost 100% if HPV status is negative, versus 44% following randomization of HPV-positive women, it should be completed before the beginning of sexual activity in order to maximize efficacy (14, 15).

Classification, grading, and histopathological tumor typing

Although it is short, the anal canal can host a number of tumor types, reflecting its complex anatomical and histological structure. Interdisciplinary treatment should therefore be provided in specialized facilities. Anal canal carcinomas are classified according to the TNM system (12, 16).

Histological tumor types are classified according to the World Health Organization (WHO) system (2, 17) (Box 2).

Box 2. Histopathological tumor typing.

• Squamous cell carcinoma (approx. 75 to 80%)

• Adenocarcinoma (15 to 20%)

  • Rectal (most common)

  • Of anal gland (glandular, well differentiated mucinous)

  • Within anorectal fistulas

• Small-cell (anaplastic) carcinoma

• Undifferentiated carcinoma

• Malignant melanoma

• Other malignant tumors

Notes:

Subtypes with unfavorable prognosis:

1. Squamous cell carcinoma with mucinous microcysts (formerly known as mucoepidermoid carcinoma)

2. Small-cell, nonkeratinizing squamous cell carcinoma (anaplastic carcinoma)

Former subtypes of squamous cell carcinoma (large-cell keratinizing vs. nonkeratinizing) are now brought together under the umbrella term “squamous cell carcinoma.” Terms such as “cloacogenic,” “transitional,” and “basaloid” carcinoma have been removed from the latest editions of the WHO classification of anal carcinoma (2, 4, 16, 18, 26).

Clinical findings, symptoms

Because the early stages of anal carcinoma are relatively asymptomatic, with indurated palpable, painless skin changes, 20% of cases are discovered by chance during procedures such as histological examination following skin tag removal, scheduled colonoscopy, or anorectal examination (e1). Macroscopically, anal carcinomas are usually coarse, sometimes smooth, generally verrucous, skin-colored to reddish nodes in the anal canal or on the anal margin, which grow in circumference and depth over the course of months and years (2) (Figure 2). Anal hemorrhages are the main symptom in 45 to 50% of patients; pain on defecation and foreign body sensation (30%), pruritus, discharge of mucus, tenesmus, incontinence (mainly with sphincteral involvement), and enlarged inguinal lymph nodes are also reported (16, 18).

Figure 2.

a: Verrucous squamous cell carcinoma, cT2cN1cM0, with fistula leading to scrotum

b: Squamous epithelium with squamous cell carcinoma in situ (AIN III) with koilocytotic changes and transition to an invasive, moderately well differentiated warty squamous cell carcinoma

Grade 1 to 3 anal intraepithelial neoplasia may be either symptomatic or asymptomatic. It should initially be treated as a precancerous lesion. The symptoms—pruritus, local irritation, and hemorrhages— are similar to those of early-stage anal carcinoma. Macroscopically, anal intraepithelial neoplasia manifests as an eczematous or papillomatous area, usually with papules similar to condyloma or whitish plaques. It should be suspected in cases with a known history of HPV/HIV infection, anal intercourse, especially in men, and venereal warts (2, e2).

Diagnosis and staging

The first diagnostic steps to be taken according to current European and American guidelines are digital rectal examination including inspection of the anal canal and possibly vaginal examination, rectoscopy, and palpation of the inguinal lymph nodes.

Total biopsy is recommended for lesions smaller than 1 cm or isolated mucosal findings, but punch or incision biopsies are preferable for tumors larger than 1 cm. Larger excisions for biopsy or tumor reduction often lead to damage to the sphincter and should therefore be avoided.

It should be particularly noted that incision biopsy is not permitted if relapse is suspected (see Follow-up). Clinical examination alone is not sufficient for diagnosis of metastasis to an inguinal lymph node, as 44% of all lymph node metastases are smaller than 5 mm and so cannot be identified with certainty by palpation (19). If there are clinical or radiological grounds to suspect lymph node involvement, a Tru-Cut or excision biopsy, possibly under anesthetic, should be performed in case of doubt. The pelvic lymph nodes are evaluated using computed tomography or magnetic resonance imaging of the abdomen/pelvis.

Sonography and computed tomography are suitable for both evaluating tumor extent and diagnosing distant metastases. The final procedure for staging is dual-level radiological examination of the lungs. Functional integrity of the sphincter should also be evaluated and recorded before therapy and during aftercare.

Endosonography can also be used for local staging. However, the field of vision of this procedure is restricted, and it cannot evaluate distant mesorectal lymph nodes or the inguinal lymph nodes. An advantage of endoscopic ultrasound examination over MRI is in the evaluation of small tumors on the surface of the anal canal; however, MRI is currently the first-line technique of choice for evaluation of locoregional disease dissemination and for N staging (1). Fluorodeoxyglucose positron emission tomography (FDG-PET/CT) is highly sensitive in identifying involved lymph nodes. Positron emission tomography (PET/CT) is not yet part of routine staging. According to current NCCN guidelines, PET/CT is indicated for staging confirmation in T2 to 4, N0 and all T, N+ anal carcinomas (18, 20– 22). Sentinel lymph node biopsy is currently performed only as part of clinical research, not routinely. Noorani et el. showed in a systematic review that 0 to 18.75% of sentinel node biopsy–negative patients subsequently developed inguinal metastases. They also reported that longer follow-up is needed in order to evaluate the benefit of sentinel node biopsy for staging and strategy selection in patients with anal carcinoma (23).

Serum levels of squamous cell carcinoma antigen (SCCAg), which was originally identified in cervical cancer, are also elevated in patients with other tumors, including anal carcinoma. However, the diagnostic and prognostic significance of this is still disputed (24, 25). In the authors’ facility, SCCAg levels are not routinely determined as part of either diagnosis or aftercare.

Treatment

The introduction of combined-modality therapy by Nigro in 1974 has improved prognosis. Between 1994 and 2000, five-year survival rates of 60% in men and 78% in women in the USA, 66% in central Europe, and 44% in eastern Europe were reported (2). Combined 5-fluorouracil (5-FU) and mitomycin C (MMC) chemoradiotherapy is currently the first-line therapy for all anal carcinomas (26). Abdominoperineal resection is now indicated only as salvage surgery. The current recommendations are based mainly on the results of six randomized trials (Table 1).

Table 1. Available randomized trials on radiotherapy/chemoradiotherapy for anal carcinoma.

Trial	Cases (n)	Treatment regimen	FU (y)	LRR	DSS (%)	CR (%)	OAS (%)
EORTC 22861 (27)	110	A: RT 45 Gy + 15 to 20 Gy boost B: CRT 45 Gy + 15 to 20 Gy boost; 5-FU (750 mg/m2, d1 to 5 and d29 to 33); MMC (15 mg/m2 , d1)	3.5	A: 48% (25/52) B: 29% (15/51) At 5 years: 18% improvement in arm B (p = 0.02)	N/A	At 5 years: 32% improvement in arm B (p = 0.02)	At 5 years: 56% for all patients (p= ins.)
UKCCCR/ACT I (28. 29)	577	A: RT 45 Gy, possibly 15 to 25 Gy boost B: CRT 45 Gy, possibly 15 to 25 Gy boost; 5-FU (750 mg/m 2 d1 to 5 and d29 to 33 or 1000 mg/m2, d1 to 4 and d29 to 32); MMC (12 mg/m2 , d1)	13.1	A: 54% (153/285) B: 29% (86/292) (p <0.001)	At 12 years: 12.5% improvement in arm B (p=0.004)	At 12 years: approx. 10% improvement in arm B (p = 0.004)	At 12 years: 5.1% better in arm B (p = ins.)
RTOG 87-04 (30)a	291	A: RT 45 to 50.4 Gy; 5-FU (1000 mg/m2, d1 to 4 and d28 to 31) B: CRT: 45 to 50.4 Gy; 5-FU (1000 mg/m2, d1 to 4 and d28 to 31); MMC (10 mg/m2, d1 + 28) N.B.: Salvage CRT (9 Gy; cisplatin) if positive biopsy following induction in both arms	3	N/A	N/A	A: 22% (32/145) B: 9% (13/146) (p = 0.002)	A: 71% (103/145) B: 78% (114/146) (p = ins.)
RTOG 98-11 (31. 32)	649	A: CRT 45 to 59 Gy; 5-FU (1000 mg/m2, d1 to 4 and d29 to 32); MMC (10 mg/m2, d1 + 29) B: CRT 45 to 59 Gy; 5-FU (1000 mg/m2, d1 to 4, d29 to 32, d57 to 60 and d85 to 88); cisplatin (75 mg/m2, d1, 29, 57, 85)	5	A: 21% (67/325) B: 27% (86/324) (p = ins.)	N/A	A: 33% (106/325) B: 41% (132/324) (p = 0.05)	A: 78% (238/325) B: 71% (209/324) (p = 0.026)
ACT II (33)	940	A: CRT 50.4 Gy; 5-FU (1000 mg/m2, d1 to 4 and d29 to 32); MMC (12 mg/m2, d1) B: CRT 50.4 Gy; 5-FU (1000 mg/m2, d1 to 4 and d29 to 32); cisplatin (60 mg/m2, d1, 29) N.B.: Second randomization to 5-FU and cisplatin maintenance chemotherapy in both groups	5.1	N/A	N/A.	A: 11% (52/472) B: 13% (60/468) (p = ins.)	A: 22% (103/472) B: 23% (108/468) (p = ins.)
UNICANCER ACCORD 03 (34)	307	A: CRT 45 Gy + 15 Gy boost; 5-FU (800 mg/m2, d1 to 4 and d29 to 32); cisplatin (80 mg/m2, d1 + 29) B: CRT 45 Gy + 20 to 25 Gy boost; 5-FU (800 mg/m2, d1 to 4 and d29 to 32); cisplatin (80 mg/m2, d1 + 29) N.B.: Second randomization to 5-FU and cisplatin induction chemotherapy in both groups	4.2	At 5 years: 4.9% better in arm B (p = ins.) No difference in terms of induction	At 5 years: 4.1% better in arm B (p = ins.) No difference in terms of induction	At 5 years: 4.1% better in arm B (p = ins.) No difference in terms of induction	N/A

FU, follow-up time, years; LRR, locoregional relapse rate; DSS, disease-specific survival; CR, colostomy rate; OAS, overall survival; A, standard treatment arm; B, experimental arm; RT, radiotherapy; CRT, chemoradiotherapy; N/A, not available

Simultaneous chemoradiotherapy

Two European phase III trials have shown a clear benefit for simultaneous 5-fluorouracil (5-FU) and mitomycin C (MMC) chemoradiotherapy over radiation alone: the EORTC 22861 trial examined data from patients with T3 to 4, N0 to 3 and T1 to 2, N1 to 3 anal carcinomas (27). It found a clear benefit for chemoradiotherapy in terms of complete remission, local control, and absence of colostomy after five years. There was no significant difference in terms of overall survival. The UKCCCR trial (ACT I) investigated all stages other than T1N0. Long-term analysis after 12 years showed sustained improvement in locoregional control in the chemoradiotherapy arm of the trial. Local control, absence of colostomy, disease-specific survival, and overall survival were also better. There was no difference in terms of late toxicity (28, 29).

The RTOG 87–04 trial investigated the additional benefit of MMC as part of 5-FU-based chemoradiotherapy. All disease stages with a measurable tumor in the anal canal were included. All patients received chemoradiotherapy and simultaneously either 5-FU/MMC (5-fluorouracil/mitomycin C) or 5-FU alone. After four years’ follow-up there were lower colostomy rates, better colostomy-free survival, and higher disease-free survival in the 5-FU/MMC arm. There was no difference in overall survival. Hematological acute toxicity and total late toxicity were higher in the MMC arm (30).

The RTOG 98–11 trial investigated the value of a cisplatin-based regimen versus MMC-based standard chemoradiotherapy. The patients included had been diagnosed with T2 to 4 anal carcinoma. The standard treatment (5-FU/MMC) arm was superior in terms of colostomy-free survival. There were also better outcomes for five-year disease-free survival and five-year overall survival. However, the MMC arm had significantly higher acute hemotoxicity (31, 32).

Another British trial (ACT II) investigated whether cisplatin-based chemoradiotherapy was superior to standard 5-FU/MMC therapy, either with or without 5-FU and cisplatin maintenance chemotherapy. After five years there were no differences in the clinical endpoints or toxicity. No significant beneficial effect was found for maintenance chemotherapy. The authors concluded that 5-FU/MMC chemoradiotherapy with no maintenance chemotherapy should remain the standard treatment (33).

The last available trial (UNICANCER ACCORD 03) was conducted in France. It investigated the value of induction chemotherapy and a local increase in radiation dose (boost). This was a four-armed study. It found no superiority in any arm; in other words, no benefit for either induction chemotherapy or boost dose escalation (34).

In conclusion, risk-adapted chemoradiotherapy up to a total dose of 50.4 to 59 Gy, combined with simultaneous 5-FU and MMC, should be the standard treatment in the future. There is currently no rationale for additional induction or maintenance chemotherapy except within clinical trials (Table 1).

Indications for surgery

According to current US and European guidelines, primary surgery with curative intent is now indicated only for well differentiated T1, N0 carcinomas of the anal margin. Carcinomas of the anal margin represent 25 to 33% of all squamous cell carcinomas of the anal region (35, 36). According to the rules of the American Society of Colon and Rectal Surgeons, excision for such cancers should be performed with a 1 cm surgical margin (recommendation of category 2A); however, the current guidelines of the European Society of Medical Oncology state that a surgical margin of more than 5 mm is sufficient for effective tumor-free excision (level of evidence IV, grade of recommendation C) (2, 18).

There are currently specific indications for radical surgical resection (salvage surgery) (Box 3).

Box 3. Indications for radical resection (salvage surgery).

• Histologically confirmed tumor remnant following neoadjuvant combined chemoradiotherapy

• Locoregional tumor relapse following chemoradiotherapy

• Advanced tumors with infiltration of neighboring organs after completion of combined chemoradiotherapy for downstaging

• Persistent fistulas and/or ulcerations affecting quality of life after completion of neoadjuvant radiotherapy/chemotherapy

Cylindrical abdominoperineal rectal extirpation or multivisceral resection with end colostomy and primary transabdominal transposition of a vertical rectus abdominis myocutaneous (VRAM) flap to cover the perineal resection defect can now be recommended as a surgical method for complete removal of persistent, relapsing, or advanced tumors (37, 38). Rectal extirpation is required as salvage surgery in approximately 10 to 30% of cases; it is performed as a result of local relapse in approximately 19% of cases (16, 35, 36).

Colostomy can be considered for primary relief of the anal canal before chemoradiotherapy (in cases of persistent fistulas or high-grade stenosis) or as part of palliative care. Only 50% of colostomies are later reversed (2).

Groin dissection should not be performed prophylactically. It should only be performed as salvage surgery in cases of confirmed inguinal relapse or for persistent inguinal lymph node metastases following chemoradiotherapy. This is because the risk of lymphoedema or a lymphatic fistula is very high, and radiotherapy of the affected region leads to better outcomes (26, 39).

The most common nonpelvic distant metastases are to the liver, lungs, and extrapelvic lymph nodes. Because only 10 to 20% of patients with anal carcinomas develop distant metastases, the amount of data that has been published is extremely limited. There is some evidence that systemic cisplatin/5-FU chemotherapy and radiotherapy for local tumor control is associated with better outcomes. Current guidelines do not recommend resection of distant metastases, but in individual cases resection with curative intent, especially for single metachronous distant metastases, may be indicated following discussion within the interdisciplinary tumor board (18).

Follow-up

The current aftercare guidelines have been revised since 2010. According to the most recent recommendations, patients with complete remission should be re-evaluated using clinical and digital rectal examination and palpation of the inguinal lymph nodes eight weeks after the end of chemoradiotherapy and then every three to six months (18). Locoregional relapse is more common than distant metastasis, so annual imaging of the chest, abdomen, and pelvis is recommended for a total of three years (2, 18, e3– e7). Resection of distant metastases does not improve survival rates. It should be noted that Tru-Cut punch or forceps biopsy is indicated for suspected local relapse. Incision biopsy should be avoided, as the risk of incontinence or severely delayed wound healing is extremely high. Aftercare ends after five years. Data from the ACT II trial shows that only 1% of relapses occur after more than three years (2, 18).

Prognosis, prognostic factors

After combined chemoradiotherapy was introduced as standard treatment, a five-year survival rate of 80 to 90% was reported; this compares to 47 to 71% following radical surgery (26, 40). Between 2004 and 2010, 49% of patients with anal carcinomas in the U.S.A. were initially diagnosed at stage N0, M0 with a five-year survival rate of 79.7%, versus 58.1% for N+, M0 carcinomas and 32% for patients with distant metastases at diagnosis (18, e8). Salvage surgery can achieve local control in the pelvis minor in 60% of cases and a five-year survival rate of 30 to 60% (2).

According to the results of the phase III trial (RTOG 98–11), male sex and N+ stage were independent factors for tumor-free survival following combined 5-FU and mitomycin C or cisplatin chemoradiotherapy (Table 2). Male sex, N+ stage, and tumor size greater than 5 cm were reported as independent prognostic factors for overall survival (32). Secondary analysis showed that tumor size was also associated with colostomy rate (31). The previously published EORTC 22861 trial showed male sex, N+ stage, and ulceration as adverse prognostic factors for overall survival and locoregional control (27). Nutz, in a review of the literature, reported a decrease in the five-year survival rate from 75% for well differentiated tumors to 25% for poorly differentiated tumors (26). Inguinal lymph node metastases are found in up to 25% of cases and according to multiple studies are independent factors for poorer prognosis. In older patients, anal carcinomas are most commonly initially diagnosed at an advanced stage, so age is considered to be only an indirect prognostic factor (26).

Table 2. Factors in prognosis (source).

	OAS	DFS	LRR	CR	ACD
Male sex (27–32)	+		+		+
N stage (27–32)	+		+		+
Tumor size (>5 cm) (30–32)	+	+		+
Ulceration (27)	+		+
Low hemoglobin (28, 29)					+
Degree of differentiation (26)	+

OAS, overall survival; DFS, disease-free survival; LRR, locoregional relapse rate; CR, colostomy rate; ACD, anal carcinoma–associated death

Key Messages.

• Anal carcinoma should be suspected in cases of chronic, nonhealing processes.

• Approximately 85% of cases are associated with HPV infection; 20% are discovered by chance during procedures such as histological examination following skin tag removal.

• Combined 5-FU and mitomycin C chemoradiotherapy is currently first-line treatment.

• Radical resection is indicated only as salvage surgery.

• The five-year survival rate for non-metastasized carcinomas is 80 to 90%