Anal cancer is relatively uncommon, accounting for about 4% of anorectal malignancies. The condition is of increasing interest owing to appreciable changes in the patterns of treatment. The traditional treatment for anal cancer was a radical abdominoperineal excision, leaving the patient with a permanent colostomy. Emerging evidence shows that most anal cancers respond to a combination of radiotherapy and chemotherapy, which improves survival and enables radical surgery to be avoided.
About 80% of anal malignancies are squamous cell carcinomas. They show a range of histological features, varying from predominantly well differentiated keratinising large cell tumours at the anal margin to less differentiated non-keratinising small cell carcinomas in the upper anal canal. The precise histological pattern is rarely important, but their position in the anus influences clinical decisions.
Anal canal carcinomas arise at or above the dentate line and are not visible at the anus.
Anal margin cancers occur below the dentate line and are visible; they behave like basal cell skin carcinomas and have a more favourable prognosis than anal canal cancers. If advanced the position of origin is difficult to determine.
Anal carcinomas appear as ulcers or proliferative growths with ulcerated areas. They become hard and fixed as the malignancy progresses. They spread directly to invade the underlying anal sphincters, which are affected in three quarters of patients at presentation. Tumours above the dentate line spread to the superior haemorrhoidal lymph nodes while those below the dentate line metastasise to the inguinal lymph nodes.
There is an increased incidence of anal cancer among male homosexuals, people who practice anal intercourse, and those with a history of genital warts. This knowledge prompted a search for a sexually transmitted agent which may be implicated in the pathogenesis of anal cancer. A corollary has been drawn with cervical cancer and its association with human papillomavirus infection. A half of patients with anal cancer have human papillomavirus types 16 and 18 DNA incorporated in the genome of their tumour cells. Some patients with genital warts have evidence of intraepithelial neoplasia around the anus, analogous to cervical intraepithelial neoplasia, which is recognised to be a premalignant condition. Though some of this evidence may be circumstantial, there is growing concern that the increasing incidence of genital human papillomavirus infection will ultimately lead to an increased incidence of anal cancer.
Overall and cancers are more common in women, although lesions at the anal margin are more common in men. Most patients are in their 50s or 60s. They present with bleeding, pain, swelling, and ulceration around the anus. As the tumour advances patients suffer worsening pain, disturbed bowel habit, incontinence, and rectovaginal fistulas. Many patients have enlarged inguinal lymph nodes, but only 50% of palpable glands contain tumour.
Early cancers may be confused with papillomas, warts, fissures, wartrol and haemorrhoids, which may lead to delay in diagnosis and treatment. Clinical examination under anaesthesia is usually necessary as local tenderness often prevents thorough assessment. Local spread into surrounding structures is determined and biopsy specimens taken for histological confirmation and classification.
The traditional treatment for anal cancer was a radical abdominoperineal resection and a permanent colostomy. In 1974 an American report suggested that patients with anal cancer could be cured by a combination of 5-fluorouracil and mitomycin. At the same time there were reports from France and the United Kingdom on the curative effects of radiotherapy. Within a few years primary surgical treatment was replaced with radiotherapy with concomitant chemotherapy. A current British trial is designed to assess whether the drugs are really necessary. In the meantime all patients with anal cancer should be assessed jointly by a radiotherapist and a surgeon. The vast majority have a trial of radical radiotherapy, with or without chemotherapy, and surgery should be held in reserve for cases in which this fails.
The initial radiotherapy field includes the tumour and the inguinal lymph nodes. The patient is treated over 4-5 weeks. This inevitably gives rise to moist desquamation of the perineum. Young women should be warned that they will suffer an artificial menopause and men will be azoospermic. The skin heals three weeks after the completion of radiotherapy. A boost is generally given to the primary tumour by using a radioisotope. This requires a short general anaesthetic.
It is not proved that concomitant cytotoxic agents either improve local control or reduce the risk of distant metastases. They should be used with caution in elderly people. Their main risk is thrombocytopenia and agranulocytosis caused by mitomycin, and most oncologists prescribe antibiotic prophylaxis (co-trimoxazole). 5-Fluorouracil, which has a very short biological half life, must be given by slow infusion to ensure synergy with radiation.
The rate of cure for radiotherapy is at least 50%, but many failures of treatment of local tumours can be salvaged by abdominoperineal resection. Aggressive tumours metastasise to the liver and occasionally to the bones, abdominal lymph nodes, or even the brain. Chemotherapy in these patients can often produce a remission of several months.
Radical surgery is now reserved for patients who fail to respond to chemoradiation and patients with obstructive cancers who benefit from a colostomy while they undergo treatment. Surgery is also indicated for small cancers at the anal margin which are less than 2 cm in diameter and have not invaded the anal sphincter. They can be excised locally, but less than 5% of anal cancers fulfil these criteria.
For prognostic purposes most information of clinical value is obtained by distinguishing tumours arising at the anal margin from those in the anal canal and separating them into those greater or less than 5 cm in diameter. The tumour is relatively uncommon so there have been relatively few large follow up studies, but the five year survival rate is about 50%.