Chapter 25: Anal Cancer

Carcinoma of the anal canal is a rare malignancy representing approximately 2.5% of all gastrointestinal malignancies. It is estimated in 2015 that over 7,200 patients will be diagnosed with carcinoma of the anal canal in the United States, resulting in greater than 1,000 deaths (¹). The incidence of this disease continues to rise steadily. A practicing oncologist will evaluate and treat less than one such patient per year. The majority of anal carcinoma arises within the mucosa of the anus and is of squamous cell histology (²). Traditionally, 74% to 90% of carcinomas of the anal canal are cured with the combined modalities of chemoradiation, reserving an abdominoperineal resection (APR) for salvage therapy of persistent or recurrent disease (³). This chapter focuses on treatment of squamous cell carcinoma of the anal canal and the potential innovative strategies that lie ahead.

The anal canal is approximately 4 cm wide and is composed of the region extending from the proximal anorectal ring to the distal anal verge (margin) (Fig. 25-1). Because various definitions of the normal anal canal anatomy exist, classifying these tumors by a histologic definition based on the lining mucosa offers a more consistent approach to guide diagnosis and treatment (²).

Malignancies of the anal margin are treated as primary skin cancers and are often surgically excised. The rectal mucosa adjacent to the anorectal ring is composed of columnar epithelium. A transition zone of both cuboidal and columnar epithelium (6-12 mm in length) extends from the distal rectum to the dentate line. The dentate line separates the columnar epithelium (columns of Morgagni) of the proximal anal canal and the squamous epithelium of the distal canal, which extends to the anal verge. The anal verge is the convergence of squamous epithelium and the anal margin. The anal margin comprises the dermis, located within 5 cm of the anal verge.

The mucosa of the transition zone, formally referred to as the cloacogenic mucosa, represents 66% of the lesions now commonly referred to as nonkeratinizing squamous cell carcinoma (SCC) (Figs. 25-2 and 25-3) (⁴). Tumors distal to the dentate line are usually keratinizing SCC (Figs. 25-4 and 25-5).

The vascular supply of the anal canal consists of the superior, middle, and inferior rectal vessels that originate from the inferior mesenteric, internal iliac, and internal pudendal arteries, respectively. Lymphatic drainage superior to the dentate line is identical to rectal carcinomas flowing to the perirectal and paravertebral nodes. Tumors located inferior to the dentate line drain to the inguinal and femoral lymph nodes. A complete physical examination should include examination of the lymph nodes of the groin.

Multiple risk factors have been associated with the development of carcinoma of the anal canal. Benign conditions such as hemorrhoids, fissures, and anal fistulas have not been determined to be causal factors (⁵). Rather, it has been postulated that these benign conditions may instead represent the initial symptoms of anal cancer (⁵).

Sexual Activity

The pathogenesis of developing anal cancer is largely related to infection with specific subtypes of human papillomavirus (HPV), most commonly with HPV-16 or HPV-18. Common risk factors associated with anal cancer include a history of more than 10 sexual partners; receptive anal intercourse before the age of 30; and sexually transmitted diseases, including condyloma acuminata (genital warts, attributed to HPV), gonorrhea, herpes virus, hepatitis, Chlamydia trachomatis, or a history of infection with human immunodeficiency virus (HIV) (^6,7,8). Women with a history of cervical, vaginal, or vulvar cancer are three to five times more likely to develop anal cancer as opposed to stomach or colon cancer, demonstrating the link between sexual activity and likely field cancerization effects from prior high-risk HPV infection (⁹).

Human Papillomavirus

Human papillomavirus is the most common sexually transmitted disease in the United States and has been strongly associated with the development of anal carcinoma (¹⁰). It is estimated that 75% of men and women of reproductive age have been infected with genital HPV.

High-risk subtypes HPV-16 and HPV-18 are associated with anal cancer—and also cervical dysplasia—and may result in anal intraepithelial neoplasia (AIN). Human papillomavirus subtype 16 reportedly results in a greater incidence of high-grade AIN (¹¹). However, unlike cervical dysplasia, AIN is a premalignant condition for which standard screening methods currently have not been universally recommended and have been limited to select high-risk individuals.

A systematic literature review on HPV type distribution in anal cancer showed a combined prevalence of HPV-16 and/or HPV-18 of 72% in invasive anal cancer, with the prevalence of HPV-16 being the highest in these cases, as in cervical cancer (¹²). A cohort of patients with metastatic SCC of the anal canal at the University of Texas MD Anderson Cancer Center (MDACC) revealed the presence of HPV, via detectable HPV DNA and/or expression of the protein p16, in 68 (95%) of 72 tumor samples analyzed (¹³). Therefore, HPV appears to be found in the vast majority of anal cancers.

The presence of HPV also has been reported to be a positive prognostic biomarker for patients with nonmetastatic SCC of the anal canal. In one study of patients with stages I to III anal cancer, HPV was detected in 120 (88%) of 137 tumors analyzed. In a multivariate analysis, p16 expression was determined to be associated with an improvement both in overall survival and disease-specific survival relative to patients with HPV-negative tumors (¹⁴).

Introduction in 2006 and 2009 of the prophylactic vaccines Gardasil and Cervarix, respectively, directed against primary infection by HPV has demonstrated efficacy in reducing precancerous anogenital lesions caused by the targeted subtypes 6, 11, 16, and 18 (^15,16,17). In late 2014, the US Food and Drug Administration (FDA) approved the introduction of a nonavalent vaccine targeting nine HPV subtypes (6, 11, 16, 18, 31, 33, 45, 52, and 58). A large, double-blind placebo-controlled study of men who have sex with men between 16 and 26 years of age demonstrated that the use of a quadrivalent vaccine against HPV not only was safe and well tolerated but also decreased the incidence of precancerous AIN (¹⁸). These findings generate early optimism that this preventative approach may decrease the incidence of this disease in the future. Even though initially approved for the vaccination of adolescent females, Gardasil has subsequently received extended FDA approval for males of the same age category after it was shown to be efficacious and may offer a promising primary prevention strategy in both genders (¹⁹).

Human Immunodeficiency Virus

Although a direct relationship between HIV and carcinoma of the anal canal has not been clearly established, a strong correlation exists between HIV and HPV. Compared with HIV-negative patients, HIV-positive patients are two to six times more likely to be diagnosed with HPV regardless of sexual practices and are also more likely to have a persistent infection (^20,21). Human immunodeficiency virus–positive men and women exposed to HIV are less likely to clear the virus and become HPV negative (^21,22). For patients who are infected with HIV, the prevalence of anal carcinoma is greater and cancer presents at a younger age of onset than in HIV-negative patients (²³).

Chronic Immunosuppression

Solid organ transplantation has been associated with a 10-fold increased risk of developing anal cancer and a 20-fold increased risk for vulvar and vaginal cancers (²⁴). A recent population-based cohort study conducted using the Danish National Patient Registry and the Danish Cancer Registry (DCR) from 1978 to 2005 found that HIV infection, solid organ transplantation, hematologic malignancies, and a range of specific autoimmune diseases were strongly associated with increased risk of anal SCC (²⁵).

Tobacco Use

Prior case-control studies have indicated that chronic tobacco use may result in a two- to five-fold increased likelihood of developing anal cancer (²⁶). Moreover, tobacco smoking appears to be associated with recurrence of anal carcinoma and is related to increased mortality; thus, smoking cessation should be encouraged once a diagnosis of anal carcinoma is made (²⁷).

The mean age of diagnosis is approximately 62 years (²⁸). The most common presenting complaint is rectal bleeding. Other symptoms may include tenesmus, pain, local irritation, discharge, or a change in bowel habits. Clinically enlarged lymph nodes are present in 15% to 25% of patients at presentation (²⁹). Extreme case presentations may include a fungating perianal mass or a verrucous mass, as seen in Figs. 25-6 and 25-7.

A diagnostic evaluation should consist of a complete physical examination including examination of the inguinal lymph nodes, a digital rectal examination (DRE), and evaluation of the surrounding mucosa of the anus. Diagnostic studies should include proctosigmoidoscopy or anoscopy, chest x-ray, and computed tomography (CT) of the chest, abdomen, or pelvis or magnetic resonance imaging (MRI) of the abdomen and pelvis to rule out distant disease. A transrectal or transvaginal ultrasound may be of added benefit in accurate disease staging (^30,31,32,33). Histologic confirmation is recommended, with a tissue biopsy of the suspected area and/or fine-needle aspiration of any palpable inguinal lymph nodes because this may impact the radiation fields. An HIV test should be considered in all patients, and a one-time test for hepatitis C infection may also be considered for patients born before the year 1965 given the disproportionately high risk of incidence of viral infection for patients in this age range (³⁴).

The staging classification system for anal cancer was adopted by the American Joint Committee on Cancer. The T stage, unlike most gastrointestinal malignancies, is not dependent on the degree of tumor tissue penetration but rather on the size of the primary tumor site.

Carcinomas of the anal margin are commonly excised with complete resolution of the tumor. Independent poor prognostic features include tumor size (T stage) with a clear distinction in prognosis between T2 and T3 tumors (^23,35,36). Patients with T1 to T2 tumors have an expected 5-year survival of 80%, whereas patient with T3 to T4 tumors have an expected median 5-year survival of <20% (²⁸). Inguinal lymph node involvement may reduce the cure rate by 50%, with increased nodal stage also being a significant prognostic factor (^37,38,39,40). Multivariate analysis of the results from the Radiation Therapy Oncology Group (RTOG) 98-11 trial indicates that tumor-related prognosticators for poorer overall survival included node-positive status (hazard ratio [HR], 1.88), large tumor diameter >5 cm (HR, 1.30), and male sex (HR, 1.38; P = .031) (⁴¹).

A pivotal approach led to the anecdotal finding that surgery may not be necessary for curative intent in the treatment of SCC of the anal canal (⁴²). The benefits of combined chemoradiation in other gastrointestinal malignancies prompted Nigro and colleagues to consider the use of chemotherapy as a radiation sensitizer. Patients received concomitant 5-fluorouracil (5-FU) and mitomycin C, which was administered along with external beam radiation therapy (EBRT) (30 Gy). The pathologic specimens in three of three patients failed to demonstrate any viable tumor. This observation culminated in the use of chemoradiation as the primary treatment modality for the treatment of anal cancer and revolutionized the approach to its treatment. The Nigro approach of chemoradiation has subsequently been evaluated in several other small phase II studies with radiation doses ranging from 30 to 60 Gy.

A retrospective analysis from the Princess Margaret Hospital reviewed the outcomes of patients who had been treated with (1) radiation alone, (2) 5-FU/mitomycin C, (3) split-course 5-FU/mitomycin C, or (4) split-course 5-FU (⁴³). The 5-year disease-free survival (DFS) was significantly improved in the 5-FU/mitomycin C arms versus 5-FU alone (76% vs 64%). Although the combination of 5-FU/mitomycin C was superior in locoregional control (LRC) with the addition of mitomycin C (86% vs 60%), the use of split-course radiation resulted in decreased morbidity, notably acute skin toxicities.

United Kingdom Coordinating Committee on Cancer Research

The UK Coordinating Committee on Cancer Research (UKCCCR) created the phase III anal cancer trial (ACT I). Patients were randomized to radiation alone (45 Gy) versus continuous infusion 5-FU (1,000 mg/m² on days 1-4 or 750 mg/m² on days 1-5) during the first and last weeks of radiation (45 Gy) with mitomycin C (12 mg/m² on day 1) (⁴⁴). After a median follow-up time of 42 months, the 3-year local failure rate was significantly reduced in the chemoradiation arm versus the radiation-alone arm (39% vs 61%, P < .0001; Table 25-1).

Notably, the 3-year mortality rate in the radiation-only arm was greater than that of the chemoradiation arm (39% vs 28%). Twenty patients (3%) required a palliative colostomy or anorectal excision due to treatment-related morbidities. Early morbidity was significant in the chemoradiation arm. Subsequently, dose reduction of mitomycin C was recommended if the patient was ≥70 years old or if deemed medically necessary. The authors concluded that chemoradiation provided a 46% reduction in local recurrence compared with radiation alone.

European Organization for Research and Treatment of Cancer

A smaller study completed by the European Organization for Research and Treatment of Cancer (EORTC) explored the role of chemoradiation and its potential benefit in LRC and colostomy-free interval (⁴⁵). A total of 110 patients were randomized to radiation (45 Gy) with or without continuous infusion 5-FU (750 mg/m² on days 1-5 and 29-33) and mitomycin C (15 mg/m² on day 1).

Event-free survival was superior in the combined-modality arm (P = .03). The overall 5-year survival was 56% in this patient population. In contrast to the UKCCCR study, this clinical trial was limited to T3 to T4 or node-positive tumors. Chemoradiation contributed 18% actuarial improvement in 5-year LRC and an increase in colostomy-free survival (CFS) (36%). In summary, the phase III UKCCCR and EORTC studies established combined chemoradiation as superior to radiation alone for LRC and CFS.

Despite the evident benefits of mitomycin C in the treatment of anal cancer, potential treatment-related toxicities may include leukopenia, thrombocytopenia, and, rarely, hemolytic uremic syndrome (HUS) and leukemia. Although a review of prior 5-FU/mitomycin C clinical trials in SCC of the anal canal suggests the potential for superior results, these often come at the cost of treatment-related morbidity and mortality. Therefore, other agents that may reduce treatment-related toxicities without compromising efficacy would be preferable.

The Eastern Cooperative Group (ECOG) 4292 trial attempted to evaluate the combination of 5-FU/cisplatin (Table 25-2) (⁴⁶). The overall response rate was 95%. Thirteen patients (68%) were determined to have a complete response (CR), and five patients (26%) had partial response. Fifteen patients (79%) experienced grade ≥3 toxicities. Locoregional control could not be achieved in approximately one-third of patients accrued. This treatment schedule with split-course radiation was identical to that reported in RTOG 92-08. The authors concluded that the delay in radiation treatment likely accounted for the inferior LRC rate and recommended examining this combination further.

A retrospective analysis in 197 patients with TxNxM0 SCC of the anal canal was completed at MDACC (⁴⁷). Patients received continuous infusion of 5-FU and cisplatin for the duration of radiation therapy (55 Gy), and median follow-up was 8.6 years. Complete responses after chemoradiation were observed in 185 patients (94%). The local recurrence rate was 11% with the use of 5-FU and cisplatin, and all patients with a local recurrence underwent salvage APR or diverting colostomies. Only 16 patients (8%) developed distant metastases. Overall survival at 5 years was 86%. Grade 4 acute toxicities (diarrhea, dehydration, and skin ulceration) were infrequent. These studies suggest that the use of cisplatin and 5-FU with concurrent radiation is safe and effective in patients with locally advanced anal cancer and may be an acceptable alternative to the traditional 5-FU/mitomycin C regimen.

Neoadjuvant (Induction) Chemotherapy

In place of standard chemoradiation therapy, could the addition of systemic induction treatment provide an improvement in overall survival and decrease the risk of distant disease development?

The phase III Intergroup/ACCORD 03 trial created a 2 × 2 factorial design to compare standard-dose (45 Gy/25 fractions + boost of 15 Gy) versus high-dose radiation therapy (boost of 20-25 Gy) and the potential benefits of induction chemotherapy with 5-FU (800 mg/m² on days 1-4) and cisplatin (80 mg/m² on day 1) for two cycles in locally advanced SCC of the anal canal (⁴⁸). Patients were required to have T2 >4 cm or node-positive disease. Three hundred six patients were allocated to one of four treatment arms: (1) induction with standard-dose radiation therapy, (2) induction with high-dose radiation therapy, (3) control arm of 5-FU/cisplatin plus standard boost, and (4) control arm of 5-FU/cisplatin plus high-dose boost. The results were compared in terms of CFS. After a median follow-up of 43 months, no difference in CFS was noted for the induction arm (P = not significant) or the higher dose radiation therapy arm (P = .67) versus the control arm. Overall, no statistical differences were noted across all arms for local control, CFS, event-free survival, or overall survival.

Radiation Therapy Oncology Group 98-11

The RTOG 98-11 trial was a large phase III multi-institutional randomized trial for locally advanced SCC of the anal canal (⁴¹) that randomized 682 patients with T2 to T4 tumors and any nodal status to receive 5-FU/mitomycin C and concurrent radiation (control arm) or induction 5-FU/cisplatin followed by concurrent 5-FU/cisplatin and radiation. All patients received 45 to 50 Gy of radiation, with an additional 10 to 14 Gy in 2-Gy fractions given to patients with residual evidence of disease, tumors >5 cm, or tumor invasion of adjacent organs. From a recent long-term update, improved outcomes were noted in the group randomized to 5-FU/mitomycin C and concurrent radiation. Five-year DFS was better in this group (68% vs 58%, P = .026) when compared to those receiving induction 5-FU/cisplatin followed by concurrent 5-FU/cisplatin and radiation, as was 5-year overall survival (78% vs 71%, P = .026). Trends in CFS, locoregional failure, and colostomy failure all favored the 5-FU/mitomycin C arm as well, although none reached statistical significance. Grade 3 or 4 hematologic toxicity was observed more frequently in patients randomized to the 5-FU/mitomycin C arm (62% vs 42%, P < .0001).

Despite the initial appearance that 5-FU/mitomycin C treatment is superior for patients with nonmetastatic SCC of the anal cancer, these findings must be interpreted with caution. This study has since been criticized for the inequality of the two treatment schedules, making a direct comparison of mitomycin C and cisplatin difficult due to the confounding factor of induction chemotherapy in the investigational arm (⁴⁹). Additionally, administration of induction chemotherapy delayed the time to initiation of curative chemoradiation and thereby prolonged the overall treatment time, a metric that, when extended, has been associated with worse clinical outcomes in clinical trials for anal cancer. Nonetheless, induction chemotherapy in patients with early-stage SCC of the anal canal is not routinely administered, based on these findings, at MDACC.

Mitomycin Versus Cisplatin and the Role of Adjuvant Therapy

The UK ACT II trial is the largest phase III trial conducted in SCC of the anal cancer and is the first direct analysis of 5-FU/mitomycin C versus 5-FU/cisplatin with concurrent radiation therapy. The ACT II trial also evaluated whether maintenance (adjuvant) chemotherapy following completion of chemoradiotherapy reduces recurrence-free survival (RFS) (⁵⁰). Using a 2 × 2 factorial design, 940 patients with T1 to T4 node-negative and -positive disease were randomly assigned to either 5-FU/mitomycin C or 5-FU/cisplatin administered concurrently with continuous radiotherapy of 50.4 Gy. The second randomization was to two courses of 5-FU/cisplatin (same schedule) consolidation chemotherapy or no further treatment. There was a greater incidence of acute grade 3 or 4 hematologic toxicity on the mitomycin C arm (26% vs 16%, P = .001) but no statistical differences in grade 3 or 4 nonhematologic toxicities. Results after a median follow-up of 5 years demonstrated no statistically significant difference for the end point of 6-month CR rate for concurrent chemoradiation with 5-FU/mitomycin C versus 5-FU/cisplatin (90.5% vs 89.6%, respectively), no difference in RFS, and no difference in CFS, respectively. The role of maintenance (adjuvant) 5-FU/cisplatin chemotherapy showed no added benefit for RFS or overall survival.

Although the investigators failed to fulfill the primary end point of superiority for the cisplatin-based regimen, the final results indicate that 5-FU/cisplatin is noninferior to 5-FU/mitomycin C in achieving a CR and is associated with fewer significant hematologic toxicities. Based on these findings, at MDACC, 5-FU/cisplatin remains our preferred regimen due to its efficacy and decreased myelosuppression relative to 5-FU/mitomycin C. This also allows for safer treatment of immunocompromised and elderly patients who otherwise might not tolerate myelosuppressive combinations.

Currently, the most commonly used approach in the combined-modality treatment of carcinoma of the anal canal is continuous-course radiation (45 Gy in 1.8-Gy fractions using opposed anterior and posterior treatment fields with a boost to the primary tumor to 5.4 Gy) plus two cycles of concurrent continuous infusion 5-FU plus mitomycin C. This regimen is considered by most to be the standard of care. Review of the literature has shown the gradual increase in radiation from 30 to 59.4 Gy in recent studies. Prior studies have indicated that there is a dose-response relationship between treatment and outcome. An analysis completed at MDACC from 1979 to 1987 revealed a dramatic difference in LRC for patients who received 45 to 49 Gy (50%) versus ≥55 Gy (90%) (⁵¹).

At MDACC, our initial experience of 55 Gy for all patients indicated that the local failure rate was 25% to 30% in patients with T3-4 tumors (⁵²). In 1999, based on this data, the dose was increased to 59 Gy for T3-4 tumors. Definitive evidence that the higher dose improves local tumor control will eventually be determined. It should be noted that radiation dosages >60 Gy result in an unacceptably high risk of anal canal and urethral stricture, ulceration, and fistula formation.

Studies using CT imaging are often relatively insensitive for detecting the total burden of disease in patients during evaluation for anal cancer, and traditionally evaluation for response following definitive treatment for nonmetastatic anal cancer has been based on clinical assessment. Although the literature is scant regarding use of MRI in evaluation of response to anal cancer, MRI may be used to evaluate a concerning lesion of interest noted on CT with relation to the adjacent anatomic structures.

The question arises regarding the duration of time to wait for a complete clinical response before referral to a surgeon for APR. Review of outcomes from ACT II revealed that 29% of patients without CR by week 11 following completion of definitive chemoradiation did achieve a CR by week 26 (⁵³). Based on these findings, in the presence of improving symptoms and/or continued tumor regression on evaluation, at MDACC patients are allowed 26 weeks to achieve a CR before referral for APR. Should clinical progression become apparent prior to this point, then patients are sent promptly for salvage evaluation. Response is largely based on clinical examination. A tissue biopsy should only be pursued if there is clinical evidence of progression or significant residual disease despite an adequate surveillance period. Because anal cancer tends to remain locoregionally confined, the opportunity will not typically be missed for salvage APR with curative intent even if a clinically concerning area is followed closely without biopsy. Treatment guidelines suggest that a tumor biopsy be performed only if recurrent or persistent disease is suspected after serial DRE (Fig. 25-8) (⁵⁴).

Salvage of Recurrent Disease

Abdominoperineal resection is the only effective treatment option for localized recurrent or residual primary disease following chemoradiation for SCC of the anal canal. Fifty to 70% of patients are cured who undergo APR for recurrence at the primary site (^55,56). There is no role for definitive reirradiation for an in-field recurrence, but neoadjuvant chemoradiation (39 Gy in 26 fractions twice daily with chemotherapy) may be used in cases where there are concerns about the radial margin.

The management of nodal recurrences should be individualized based on the extent of disease, prior radiation delivered to the area of recurrence, and performance status of the patient. In cases where patients have been referred to MDACC with nodal recurrence outside of or at the margin of a prior radiation field, salvage chemoradiation has been effective provided that a full dose of reirradiation is possible. For in-field nodal recurrences, we typically use preoperative chemoradiation to 39 Gy in 26 fractions twice daily followed by surgical resection.

Adenocarcinomas of the anal canal occur less frequently than SCCs (²⁸). Nonetheless, most of the reports overestimate the incidence of adenocarcinoma of the anal canal because they do not exclude contamination by the more common distal rectal cancer. The true incidence is likely less than 10%. Its etiology remains unclear, but like its squamous cell counterpart, it has been linked in the past to chronic inflammatory conditions and HPV (^37,55). The most appropriate management remains to be defined, with no large prospective studies completed to date. The most striking difference between adenocarcinoma and SCC of the anal canal is the high distant metastasis rate, which tends to undermine the impact of local tumor control. Retrospective analysis from the MDACC experience with adenocarcinoma of the anal canal has suggested a benefit of neoadjuvant chemoradiation followed by APR with the consideration of adjuvant chemotherapy analogous to the treatment of rectal cancer (⁵⁷).

Although the majority of patients with SCC of the anal canal will be cured with chemoradiation, a minority of patients will develop distant metastatic disease. Overall, 5% of patients initially present with extrapelvic metastases, and 10% to 20% of patients treated with curative intent will develop metastatic disease. Due to the rarity of this disease, a universally accepted treatment paradigm has not been established, with choice and duration of therapy largely based on individual case studies and small case series (⁵⁸).

One recent retrospective analysis of our experience at MDACC, the largest series published to date of 72 patients with metastatic anal cancer treated between 2000 and 2012, demonstrated that the majority of patients (55%) with metastatic anal carcinoma receive a 5-FU– and cisplatin-based regimen as first-line therapy (⁵⁹). Patients received a median of two lines of cytotoxic chemotherapy, with carboplatin and paclitaxel being the most popular second-line treatment for patients still able to tolerate further therapy following progression on frontline chemotherapy. With a median follow-up period of 42 months, the median overall survival was 22 months. Systemic treatment should be considered in any patient demonstrating a good performance status, with duration of therapy continued indefinitely for maximal outcome if tolerated well. A subset of patients in this cohort (43%) with oligometastatic disease was able to proceed to surgical resection of distant metastases or chemoradiation to affected regions that had not received prior radiotherapy. For these patients, overall survival was prolonged (53 months vs 17 months, P < .001), which was likely a reflection of a lower total burden of disease and favorable clinical characteristics that enabled this group to withstand multidisciplinary management of their metastatic disease. Consideration of surgical resection with curative intent should be encouraged for patients if surgically resectable or borderline resectable.

When patients are treated with chemoradiation therapy with curative intent, patients should be evaluated every 6 to 8 weeks until a maximal clinical response is achieved. Biopsy should not be performed before 26 weeks following chemoradiation therapy unless there is clear evidence of residual disease or progression is suspected. If clinical CR is achieved, then patients should be evaluated every 3 months for 2 years after diagnosis. Physical examination must include a DRE and assessment for any palpable inguinal lymph nodes. Vaginal dilators may be used three times a week if needed. The critical time for the prevention of vaginal stenosis is 3 to 6 months following completion of chemoradiation therapy. Vaginal hormonal creams and suppositories are also useful for treatment of vaginal dryness and dyspareunia. Proctosigmoidoscopy should be performed biannually following a CR for 2 years. Chest x-ray and CT of the chest, abdomen, and pelvis or MRI of the abdomen and pelvis should be completed annually for 2 years. Pap smears should continue to be performed annually.

It is likely that additional chemotherapy agents other than 5-FU, mitomycin C, and cisplatin may provide benefit to patients with SCC of the anal canal. A prospective, randomized trial (InterAACT) comparing cisplatin/5-FU with carboplatin/paclitaxel as frontline therapy for patients with metastatic anal cancer is currently under way and will elucidate the optimal choice of cytotoxic agents to be used in patients with nonresectable disease. Given the tight association between HPV infection and anal cancer, there is interest that these immune checkpoint blockade agents may have efficacy for this virally associated tumor. A phase II trial of nivolumab, a monoclonal antibody against PD-1, is under way for patients with refractory metastatic anal cancer, and these results will be important in understanding the safety and efficacy of this class of agents in patients with metastatic disease.

Biologic Agents

The role of the anti–vascular endothelial growth factor (VEGF) agent bevacizumab in SCC of the anal canal has not been well defined. However, although this drug has received recent approval by the FDA in combination with paclitaxel and either cisplatin or topotecan in patients with recurrent/metastatic cervical cancer (⁶⁰), further investigation in the coming years will clarify the benefit of this anti-VEGF agent in advanced anal cancer.

Anti–epidermal growth factor receptor monoclonal antibodies such as cetuximab have demonstrated efficacy in patients with KRAS wild-type metastatic colorectal cancer (^61,62). This oncogene appears to be mutated infrequently in anal cancer. Data published from the ACCORD 16 phase II trial studying cetuximab in combination with cisplatin/5-FU in patients receiving concomitant radiation for locally advanced anal cancer revealed an unacceptably high frequency of significant adverse events (⁶³).

Carcinoma of the anal canal is a unique malignancy where chemoradiation therapy is provided with curative intent or failure to respond to therapy will result in an APR. Hence, it is recommended that all patients diagnosed be initially evaluated at a tertiary cancer center or the equivalent given the rarity of the disease and the potential for permanent loss of sphincter preservation. We highly recommend a multidisciplinary team discussion with significant expertise for the most appropriate treatment in this rare malignancy.

The authors would like to acknowledge Amir Mehdizadeh for editorial input and Dr. Stanley Hamilton and the anal carcinoma patients at the University of Texas MD Anderson Cancer Center who allowed their photographs to be taken as a contribution to this chapter.