Introduction

Squamous cell carcinoma of the anus (SCCA) is a rare malignancy comprising only 2.7% of all gastrointestinal cancers in the United States.1 Although it is uncommon, the incidence of SCCA is increasing both in the United States and globally2 because of the human papillomavirus (HPV), which is identified pathologically in >90% of patients with SCCA.3 Most patients present with disease confined to the primary site (47%) or lymph nodes (33%), without distant metastases.4 Locoregional SCCA, confined to the anus and regional lymphatics, is curable in most patients using definitive, concurrent chemotherapy and radiation, sparing patients the associated morbidity of an abdominal-perineal resection (APR).5

The treatment approach to SCCA has evolved dramatically over time. Before the 1970s, SCCA was managed surgically with an APR and permanent colostomy. In the 1970s, researchers at Wayne State University in Michigan published their initial results of a preoperative chemoradiation (CRT) protocol aimed at improving treatment outcomes.6 After the first 3 patients all developed a pathologic complete response (CR) to neoadjuvant therapy (using relatively modest doses of radiotherapy [RT]), the regimen published by Nigro and colleagues (eponymously called the Nigro regimen) ignited a revolution in the treatment of this disease, ultimately resulting in the modern-day practice of sphincter-preserving CRT.7

In the current review, we consider the evolution of the management of SCCA from its inception to current practice and discuss 3 integrated components to treatment: surgery, radiation therapy, and chemotherapy (summarized in Fig. 1). The underlying technical aspects of therapy, cancer biology, clinical outcomes, toxicities, and appropriate follow-up care also are reviewed.

Timeline Depicting Advances in Surgery (Red), Systemic Therapy (Green), and Radiation Therapy (Blue) in the Management of Anal Squamous Cell Cancer Over Time. 3D indicates 3-dimensional; 5FU, 5-fluorouracil; APR, abdominal-perineal resection; CRT, chemoradiation; ECOG, Eastern Cooperative Oncology Group; EORTC, European Organization for Research and Treatment of Cancer; Gy, gray; IMRT, intensity-modulated radiation therapy; MMC, mitomycin-C; NCT, ClinicalTrials.gov identification number; RT, radiation therapy; RTOG, Radiation Therapy Oncology Group.

The Early Management of Anal Cancer

Before the 1970s, SCCA was treated with radical resection, which is described as the removal of the distal colon, rectum with intact mesorectum, anus, and anal sphincter complex through combined anterior abdominal and perineal approaches. The resulting defect is often large, requiring a tissue flap to fill the void, preventing bowel from filling the pelvis and subsequent perineal hernia development, or primarily sutured closed. As a result, a permanent colostomy is created for stool elimination for the remainder of the patient's lifetime.8 This radical surgery resulted in 5-year overall survival (OS) rates ranging from 40% to 70%, with a 3% perioperative mortality rate.9, 10

In 1974, investigators at Wayne State University experimented with a preoperative protocol for SCCA consisting of chemotherapy and radiation that propelled the management of this disease into the modern era. The initial Nigro regimen consisted of a fluoropyrimidine, 5-fluorouracil (5-FU) (1000 mg/m2 as a continuous infusion on days 1-4 and 29-32) and mitomycin-C (MMC) (10-15 mg/m2 on day 1) with concurrent radiation to a total dose of 30 gray (Gy).6 The first 3 patients treated with this protocol experienced complete regression of their tumor, prompting further investigation into sphincter-sparing therapy. In a follow-up study, patients treated with the Nigro regimen only underwent a salvage APR if there was clinical evidence of residual disease.11 Eighty-four percent of patients in that follow-up study had a clinical CR to CRT. In addition, the 5-year OS rate was 67%, and the 5-year colostomy-free survival (CFS) rate was 59% for the cohort. Interestingly, no patient with residual disease after definitive CRT who underwent salvage APR was alive at 5 years, indicating that response to CRT is an important prognostic factor.

After these promising early results, several confirmatory studies were performed in the 1980s and 1990s cementing the role of CRT as definitive therapy in the management of SCCA.12-16 The radiation dose used in subsequent studies was increased above the relatively modest 30 Gy used in the Nigro protocol. With radiation dose escalation and optimization of the chemotherapy regimen, the results demonstrated 5-year OS rates ranging from 70% to 90% and 5-year CFS rates ranging from 70% to 86%. Overall, these outcomes were superior to those previously reported after radical resection, with a significant reduction in associated morbidity, leading to a paradigm shift in the management of SCCA that replaced APR with definitive CRT.

Chemotherapy Is Chemotherapy Necessary?

The focus of treatment in the late 1990s was on the benefit of CRT versus radiation alone. The UK Coordination Committee on Cancer Research (UKCCCR) published initial results of the first Anal Cancer Trial (ACT I) in The Lancet in 1996.17 Five-hundred eighty-five patients were randomized to receive either radiation alone (45 Gy in 20-25 fractions followed by either an external-beam or brachytherapy boost) or CRT (750-1000 mg/m2 infusional 5-FU the first week of RT and 12 mg/m2 MMC on day 1). The primary end point was locoregional failure (LRF) measured at least 6 weeks after initial therapy. Secondary end points were OS and cause-specific survival (CSS). Both local failure (36% vs 59%; P < .01) and CSS (28% vs 39%; P = .02) were improved with CRT versus radiation alone, but there was no improvement in OS. Although there was an increased rate of acute toxicity with CRT, there was no difference in late toxicity. Key findings from ACT I and other major trials for patients with nonmetastatic anal cancer are presented in Table 1.

TABLE 1. Major Trials Evaluating Therapies for Nonmetastatic Anal Cancer STUDY FOLLOW-UP, MO PATIENTS RANDOMIZATION RESULTS INTERPRETATION ACT I 42 585 Patients with T1-T4 (excluding T1N0) anal canal or anal margin cancer RT (45 Gy with or without boost) vs RT/5-FU/MMC Decreased LF with CRT (59% vs 36%); decreased risk of death from anal cancer with CRT (HR, 0.71; 95% CI, 0.53-0.95; P = .02), no difference in 3-y OS (58% vs 65%) CRT is superior to RT alone RTOG 87-04 36 310 Patients with any T or N stage RT (45 Gy with or without boost)/5-FU vs RT/5-FU/MMC Improved LC (34% vs 16%), CFS (22% vs 9%), and DFS (51% vs 73%) with MMC; similar OS; more grade 4-5 toxicity with MMC Despite increased toxicity associated with MMC, it should remain part of the standard treatment paradigm for patients with anal cancer EORTC 22861 NR 110 Patients with locally advanced ASCC (T3-T4 or N1-N3) RT (45 Gy plus boost) vs RT/5-FU/MMC Improved locoregional control at 5 y (50% vs 70%) and CFS (40% vs 70%) with CRT; similar OS CRT is superior to RT alone RTOG 98-11 30 644 Patients with T2-T4, any N stage Induction CDDP/5-FU followed by RT/5-FU/CDDP vs RT/5-FU/MMC Improved 5-y DFS (68% vs 58%), OS (78% vs 71%), and CFS (72% vs 65%) with RT/5-FU/MMC RT/5-FU/MMC without induction chemotherapy should remain standard of care ACCORD-03 50 283 Patients with primary tumor ≥4 cm or lymph node-positive disease Standard-dose RT/5-FU/CDDP vs high-dose RT/5-FU/CDDP with or without induction chemotherapy (5-FU/CDDP) Similar CFS, LC, DSS, and OS among all 4 groups There is no benefit to dose escalation beyond 60 Gy for advanced disease, and there is no benefit to induction chemotherapy ACT II 61 940 Patients with T1-T4, any N stage, SCCA RT (50.4 Gy)/5-FU/MMC vs RT/5F-U/CDDP followed by second randomization (with or without maintenance 5-FU/CDDP for 2 cycles) Similar 3-y CFS, PFS, and OS; similar complete response rate at 26 wk; similar OS with or without maintenance chemotherapy RT/5-FU/MMC without maintenance chemotherapy should remain standard of care; complete response still possible up to 26 wk RTOG 05-29 NR 63 Patients enrolled, 52 evaluated; T2-T4, any N stage, SCCA All treated with DP-IMRT prescribed per stage: T2N0, 42 Gy elective nodal and 50.4 Gy anal tumor in 28 fractions; T3-T4/N+, 45 Gy elective nodal and 54 Gy anal tumor PTVs in 30 fractions Significant reduction in acute grade ≥2 hematologic, grade ≥3 gastrointestinal, and grade ≥3 dermatologic adverse events 23% with DP-IMRT as compared to RTOG 98-11 DP-IMRT was associated with significant sparing of acute grade ≥2 hematologic and grade ≥3 dermatologic and gastrointestinal toxicity ECOG 3205 84 61 Immunocompetent patients with T2-T3N0 or N+ anal canal or anal margin RT (54 Gy to gross disease) plus CDDP/5-FU and cetuximab 3-y PFS, 68%; 3-y OS, 83% but 32% grade 4 toxicity and 5% deaths Addition of cetuximab to CRT is associated with significant toxicity and should not be used Abbreviations: +, positive; 5-FU, fluorouracil; CDDP, cisplatin; CFS, colostomy-free survival; CRT, chemoradiation; DFS, disease-free survival; DP-IMRT, dose-painted intensity-modulated radiation therapy; DSS, disease-specific survival; ECOG, Eastern Cooperative Oncology Group; EORTC, European Organization for Research and Treatment of Cancer; Gy, gray; HR, hazard ratio; LC, local control; LF, local failure; MMC, mitomycin-C; N, nodal stage; NR, not reported; OS, overall survival; PFS, progression-free survival; PTVs, planning target volumes; RT, radiation therapy; RTOG, radiation therapy oncology group; SCCA, squamous cell carcinoma of the anus; T, tumor stage.

The long-term results of the ACT I trial confirmed the superiority of CRT for SCCA versus radiation alone.18 At 12 years, CRT was associated with a reduction in the risk of LRF (hazard ratio [HR], 0.46; 95% CI, 0.35-0.60; P < .001). CSS was also improved in the CRT arm (HR, 0.67; 95% CI, 0.51-0.88; P < .004). The absolute reduction in the risk of death from SCCA was 12.5% at 12 years in favor of CRT. OS was similar between the 2 groups, which was attributed to an increased risk of non-SCCA–related deaths in the first 5 years that disappeared by year 10.

A second trial led by the European Organization for the Research and Treatment of Cancer again confirmed the superiority of CRT over RT alone.19 One-hundred ten patients were randomized to receive either radiation alone (45 Gy in 25 fractions followed by a boost) or CRT (750 mg/m2 infusional 5-FU in week 1 and 15 mg/m2 MMC on day 1). Locoregional control was 18% higher in the CRT arm (P = .02), and CFS was improved by 32% (P = .002). There was no difference in 5-year OS (56%) or toxicity between groups. The results of these 2 randomized studies affirmed the need for combined CRT in treating patients with SCCA.

What Is the Ideal Concurrent Chemotherapy Regimen?

At the turn of the 21st century, investigators attempted to define the optimal chemotherapy regimen for patients with SCCA. MMC is a highly toxic chemotherapeutic agent that traditionally had minimal antitumor activity against squamous cell cancers, leading investigators to question its necessity for patients with SCCA. A phase 3 Intergroup randomized controlled trial (RCT) led by the Radiation Therapy Oncology Group (RTOG), RTOG 87-04, randomized 310 patients to receive either RT plus 5-FU (1000 mg/m2 daily in week 1) or RT plus 5-FU and MMC (10 mg/m2 on days 1 and 29).20 Although acute toxicity was higher in the MMC arm (23% vs 7%; P < .01), benefit was observed for both CFS (71% vs 59%; P = .014) and disease-free survival (DFS) (75% vs 51%; P < .01) with the addition of MMC. The results of that study made it evident that 5-FU alone was insufficient for patients with SCCA.

Investigators next questioned whether MMC could be replaced with a less toxic agent to decrease treatment-related morbidity without compromising outcomes. Some of the early phase 1 and 2 trials investigating the safety and efficacy of CRT for SCCA used cisplatin in place of MMC with comparable outcomes.14-16, 21 This prompted investigators to evaluate the efficacy of replacing MMC with cisplatin. RTOG 98-11 (ClinicalTrials.gov identifier NCT00003596) was a multicenter, phase 3 RCT that compared RT (45-59 Gy) and concurrent 5-FU plus MMC versus RT and concurrent 5-FU plus cisplatin after induction chemotherapy (cisplatin and 5-FU).22 The primary end point of that trial was an improvement in DFS in the cisplatin arm. The long-term update of the trial showed that, of the 649 analyzable patients, there was a significant improvement in 5-year DFS (68% vs 58%; P < .01) and OS (78% vs 71%; P = .03) and a trend toward improvement in CFS (72% vs 65%; P = .05) in patients who received MMC.23 Hematologic toxicity was increased in the MMC arm, but overall toxicity severity was similar between arms. The results of RTOG 98-11 support the continued use of MMC in combination with 5-FU and RT for patients with SCCA. A notable limitation of the study was that patients received induction chemotherapy in the cisplatin arm, thus direct comparisons between concurrent MMC and cisplatin regimens are challenging.

Another study conducted by the UKCCR, ACT II (ClinicalTrials.gov identifier NCT00025090), also examined the substitution of MMC with cisplatin. ACT II was a 2 × 2 factorial trial in which patients were first randomized to receive CRT with either MMC or cisplatin, and this was followed by randomization to receive either adjuvant therapy with 5-FU and cisplatin or observation.24 For the MMC versus cisplatin comparison, the primary end point was improved clinical CR at 6 months with the use of cisplatin. For the comparison of maintenance versus no maintenance, the primary end point was improved progression-free survival (PFS), with further maintenance chemotherapy after CRT completion. Neither experimental arm showed a benefit with respect to the primary end point studied in that superiority-designed trial (the clinical CR rate was 91% for MMC vs 90% for cisplatin; PFS was 74% for maintenance chemotherapy at 3 years vs 73% for no maintenance). The results of these 2 trials (the superiority of MMC seen in RTOG 98-11 and the failure for cisplatin to show superiority in ACT II) have led to MMC remaining the standard of care for managing patients with SCCA.

In recent years, oral fluoropyrimidine chemotherapy has also been used in combination with MMC because of its superior tolerability and ease of administration; however, this was based only on phase 2 data. Glynn-Jones and colleagues examined radiation (50.4 Gy in 28 fractions) with concurrent capecitabine (825 mg/m2 twice daily on RT days) and MMC (12 mg/m2 on day 1) and showed a 68% rate of compliance with chemotherapy, a 77% clinical CR rate, and no locoregional recurrences at a median follow-up of 14 months.25 Two additional studies of capecitabine for patients with SCCA yielded similar results.26, 27 A recent national cohort evaluation from the United Kingdom showed lower rates of hematologic toxicity when capecitabine was used instead of 5-FU, with other toxicity rates similar between the 2 groups.28 Capecitabine with MMC or with 5-FU is listed as an option in the National Comprehensive Cancer Network (NCCN) guidelines for SCCA.29

Is There a Role for Induction or Maintenance Chemotherapy?

Although most patients with SCCA are cured with CRT alone, patients with bulky or extensive nodal disease have poor outcomes. The role of induction chemotherapy to improve outcomes in patients with a poor prognosis was studied in the phase 2 trial conducted by the Cancer and Leukemia Group B.30 Forty-five patients with T3-T4 tumors or bulky nodal disease were enrolled. Patients received 2 cycles of induction infusional FU plus cisplatin, followed by CRT. At 4 years of follow-up, 68% of patients were alive, 61% were disease-free, and 50% were colostomy-free and disease-free, leading the authors to conclude that induction chemotherapy is effective in offering long-term disease control to patients with poor-prognosis SCCA.

Two subsequent randomized phase 3 trials failed to show a benefit to induction chemotherapy for patients with SCCA, however. The first, RTOG 98-11, failed to adequately address the role of induction chemotherapy because the study arm, which included induction chemotherapy, also used experimental concurrent chemotherapy with 5-FU/cisplatin.23 Therefore, it is unclear whether the inferior outcomes seen in the induction arm were caused by the induction chemotherapy and the subsequent delay of CRT or by the use of concurrent 5-FU/cisplatin in place of 5-FU/MMC. A subsequent trial, ACCORD 03 (ClinicalTrials.gov identifier NCT00003652), included a 2 × 2 factorial design in which treatment with 2 varying doses of radiation with or without induction chemotherapy were compared with one another.31 The use of induction chemotherapy did not improve the primary end point, CFS. Taken together, the results of RTOG 98-11 and ACCORD 03 indicate that there is no role for induction chemotherapy in the management of patients with SCCA. Similarly, ACT II showed no benefit to adjuvant chemotherapy after standard CRT (with 5-FU/MMC).

Is There a Role for Monoclonal Antibodies?

As the biology of SCCA became better understood, investigators from the International Rare Cancer Initiative, together with the Eastern Cooperative Oncology Group and the AIDS Malignancy Consortium, began experimenting with novel treatment approaches. After the success of antiepidermal growth factor receptor (anti-EGFR) therapies for HPV-mediated cancers of the head and neck, trials assessing the addition of cetuximab, a monoclonal antibody that blocks EGFR, were performed in both immunocompetent and HIV-infected patients. Eastern Cooperative Oncology Group trial E3205 (ClinicalTrials.gov identifier NCT00316888) enrolled 61 immunocompetent patients with stage I through III SCCA and treated them with cisplatin, 5-FU, radiation, and 8 once-weekly doses of concurrent cetuximab. Although the addition of cetuximab resulted in lower LRF rates than historical controls (3-year LRF rate, 23%; P = .03), toxicity was unacceptably high, with grade 4 toxicity occurring in 32% of patients and a 5% rate of treatment-related deaths.32 In a similarly designed trial enrolling HIV-infected patients conducted by the AIDS Malignancy Consortium, there were also unacceptably high rates of grade 4 (26%) and grade 5 (4%) toxicity.33 Because of these high rates of toxicity, anti-EGFR therapy is not recommended to be further pursued in this setting as a radiosensitizer.34

Can Any Patient Avoid Chemotherapy?

There are no RCTs that assess the omission of chemotherapy for patients with early stage disease. There are limited data showing excellent OS and anal sphincter-preservation rates for tumors ≤4 cm in size treated with high-dose radiation alone using a combination of external-beam RT and brachytherapy.35 If limited to T1 tumors (≤2 cm in size), treatment with radiation alone achieves a 100% local control rate.36 It is important to note that both RCTs that established the use of CRT as the standard of care for SCCA, ACT I and European Organization for the Research and Treatment of Cancer trial 22861, excluded patients with T1N0 disease.17, 19 The majority of subsequent trials after the establishment of CRT as standard of care also excluded patients who had T1N0 disease, with the exception of ACT II, in which 10% of patients had T1 disease. Because of the paucity of data, the NCCN guidelines universally recommend CRT for all patients with locoregional disease. Salvage therapy for patients with recurrent disease is morbid, typically requiring APR, thus chemotherapy should not be omitted for patients receiving RT.

Future Directions: Immunotherapy

Immunotherapy is an active area of research in nearly all malignancies, including SCCA. The efficacy of immunotherapeutic agents, specifically those targeted against the programmed cell death receptor 1 (PD-1), has been shown for other HPV-mediated cancers such as squamous cell cancer of the head and neck.37 Owing to this success, anti-PD-1 therapy has been studied in patients with metastatic and surgically unresectable recurrent SCCA, with promising results.38, 39 A phase 2 National Cancer Institute (NCI) Experimental Therapeutics Clinical Trials Network trial enrolled patients with treatment-refractory, metastatic SCCA who were treated with the anti–PD-1 agent nivolumab (NCI9673: Part A) and reported a 24% objective response rate with no serious adverse events reported.38 Pooled results from the phase 1B KEYNOTE-028 and phase 2 KEYNOTE-158 trials (ClinicalTrials.gov identifiers NCT02054806 and NCT02628067, respectively) of pembrolizumab for previously treated, advanced SCCA indicated an objective response rate of 11% in the group overall and 14% among patients with tumors expressing PD-L1. The median OS was 11.7 months, with 18% of patients experiencing grade 3 and 4 adverse events and no grade 5 adverse events reported.40

Favorable outcomes among patients with advanced SCCA have led to clinical trials investigating the role of immunotherapeutic agents for patients with locally advanced SCCA. A randomized phase 2 study investigating the role of maintenance nivolumab after CRT for patients with high-risk, recurrent stage IIB or III disease has completed enrollment (ClinicalTrials.gov identifier NCT03233711). Pembrolizumab is also being studied as salvage therapy for patients with advanced SCCA who failed at least one prior therapy (ClinicalTrials.gov identifier NCT02628067).

Radiation Therapy What Is the Optimal Radiation Dose?

The optimal dose of RT for patients with SCCA has been debated since the original reported 30 Gy used in the Nigro regimen. Early studies attempting to escalate the dose above 50 Gy failed to show an improvement in disease control; however, split-course therapy with a 2-week break after 36 Gy was mandated because of accumulated toxicities.41, 42 Delaying the total treatment time because of breaks in therapy has been shown to adversely affect local control and CFS.43, 44 The inferior DFS and OS rates noted in patients with large tumors and nodal disease on RTOG 98-11, who were treated to 45 Gy (up to 59 Gy if they had residual disease),23, 45 prompted later investigation into radiation dose escalation without mandated breaks. ACCORD 03 evaluated the role of a high-dose boost (20-25 Gy) after 45 Gy versus a standard boost of 15 Gy.31 There was no difference in the primary end point of CFS for standard versus high-dose boost RT (78% vs 74%, respectively; P = .067). However, there was a small nonsignificant improvement in 5-year local control for higher dose boost RT compared with standard boost RT (83% vs 78%, respectively; P = .28). A recently published post-hoc analysis of ACT II found that patients who did not receive the total planned dose of radiation (50.4 Gy) had significantly worse PFS and OS, suggesting that there is a dose-response relationship with regard to radiation dose in the treatment of SCCA.44

What Is the Optimal Radiation Technique?

Given the significant acute morbidity rates seen in several of the CRT trials for SCCA, including RTOG 98-11, the modality of RT delivery has undergone significant improvement since the original Nigro regimen, which used 2 opposed anterior and posterior radiation fields. The publication of RTOG 0529 (ClinicalTrials.gov identifier NCT00423293) in 2014 supported the use of intensity-modulated radiation therapy (IMRT) over previous radiation techniques because of a likely benefit with respect to hematologic, gastrointestinal, and dermatologic toxicities.46 Compared with RTOG 98-11, which used nonconformal radiation techniques, the use of IMRT in RTOG 0529 resulted in a 12% improvement in grade ≥2 hematologic toxicity (P = .03), a 15% improvement in grade ≥3 gastrointestinal toxicity, and a 26% improvement in grade ≥3 dermatologic toxicity (P < .01). The long-term results of this study were recently reported.47 At a median follow-up of 7.9 years, the rates of LRF, distant metastases, CFS, DFS, and OS were comparable to those reported for cohorts of patients treated with more conventional radiation techniques, indicating that the reduced toxicity of IMRT does not come at the expense of efficacy. A national audit on the use of RT for patients with SCCA in the United Kingdom showed a significant improvement in toxicity rates for patients treated with IMRT versus those treated with previous techniques.48 As a result of these trials, IMRT has become the standard treatment technique in the management of SCCA.

Administration of an additional boost dose of radiation to the gross disease using brachytherapy has also been investigated.36, 49-52 Although some data sets have shown a benefit to the brachytherapy boost,50, 51 those results are not universal.49 The lack of RCT data and the mixed results from retrospective series have resulted in external-beam RT without brachytherapy as the primary treatment modality for SCCA.

What Should Be Targeted by Radiation Therapy?

Because RT for SCCA has moved toward a more conformal treatment approach with IMRT, a detailed understanding of the structures that should be targeted is critical to avoid LRF. The RTOG Gastrointestinal Committee established a task force to develop an atlas of the elective target volume definitions that should be used for IMRT planning.53 The Australasian Gastrointestinal Trials Group and a group of specialists from the United Kingdom54, 55 developed detailed contouring atlases to outline the elective volumes that should be treated in patients undergoing RT. All of these guidelines recommend inclusion of the mesorectal, presacral, internal iliac, external iliac, obturator, and inguinal lymph nodes in the elective target volume. The borders of the inguinal region were variable among these atlases, causing the target volume for inguinal lymph node irradiation to remain uncertain. A study comparing the patterns of inguinal nodal metastases in a cohort of patients with the nodal volumes outlined in published guidelines helped to refine the delineation of inguinal lymph node elective volumes.56 Careful coverage of the inguinal region is critical given data showing the high rates of locoregional recurrence in the inguinal regions of patients who have not received elective groin irradiation.57 Long-term results of RTOG 0529 showed no isolated nodal failures in elective nodal volumes.47 Strict RT quality assurance of radiation plans, particularly with respect to contouring, was performed in that trial.46, 58

Future Directions for Radiation Therapy

In the era of IMRT, dose escalation is being reexamined for those patients at highest risk of local relapse. A review on the use of IMRT for anal cancer found that, based upon the linear quadratic model, a >5 Gy increase in radiation dose may result in >10% improvement in local control.59 The PLATO trial (Personalizing Anal Cancer Radiotherapy Dose) (International Standard Randomized Controlled Trial [ISRCT] number ISRCTN88455282) is an integrated protocol comprising 3 separate trials (ACT3, ACT4, and ACT5) that aim to optimize the radiation dose for low-risk, intermediate-risk, and high-risk SCCA (Cancer Research UK trial number CRUK/15/007). ACT5 is a randomized phase 2/3 trial for patients with locally advanced SCCA, comparing standard-dose CRT (53.2 Gy in 28 fractions) with two higher doses of CRT (58.8 and 61.6 Gy, both in 28 fractions). However, a major concern for dose escalation is the increased risk of fecal incontinence.60 This risk must be balanced with the potential for improved local control, and results from these prospective trials will hopefully elucidate the benefits and risks associated with this approach.

The role of dose de-escalation is also being assessed. A clinical study in patients who had T1-T2 tumors with close or involved margins and received CRT to radiation doses as low as 30 Gy demonstrated excellent local control (90%).11 Three ongoing clinical trials, the UK ACT3 and ACT4 trials and the US DECREASE (ClinicalTrials.gov identifier NCT04166318) study, are evaluating the strategy of dose de-escalation in patients with T1-T2 tumors. ACT3 is a nonrandomized phase 2 trial for patients with early stage tumors who have undergone surgery (local excision) to determine whether these patients can be observed, for margins >1 mm, or offered low-dose radiation with chemotherapy, for margins >1 mm, with acceptably low local recurrence rates. ACT4 is a randomized phase 2 trial for patients with intermediate-risk disease comparing standard-dose CRT (50.4 Gy in 28 fractions) with reduced-dose CRT (41.4 Gy in 23 fractions). Finally, the DECREASE trial in the United States is investigating dose de-escalation for patients with early stage SCCA. Researchers are also investigating different RT modalities to reduce toxicity further, such as proton therapy (ClinicalTrials.gov identifier NCT03018418).

Surgery Is There a Role for Surgery?

Although radical surgery with APR has been replaced by definitive CRT for the vast majority of patients with nonmetastatic SCCA, there remains a role for surgery in certain clinical scenarios. One such scenario is the selective utilization of a wide local excision (WLE) in patients with superficially invasive SCCA (SISCCA). SISCCA is defined as anal cancer that has been completely excised to negative margins with ≤3 mm of basement membrane invasion and ≤7 mm of horizontal spread.61 It is typically encountered incidentally when an excisional biopsy is performed for a lesion suspicious for dysplasia. The frequency of SISCCA is increasing because of the increasing frequency of anal cancer screening among high-risk populations. Retrospective data have shown no difference in OS for these patients treated with WLE versus CRT.62 Because of the increasing frequency of SISCCA and the comparable outcomes seen with WLE without the toxicity of CRT, the use of WLE nearly doubled over the decade from 2004 to 2012 (17% vs 31%). It should be noted, however, that the data are mixed with regard to what constitutes an adequate margin for WLE, thus WLE should be used with caution in carefully selected patients; moreover, multidisciplinary evaluation and discussion are recommended for patients with SCCA who are being considered for WLE.

Other indications for surgery are in the post-CRT setting, either as salvage or to manage long-term toxicity from CRT. Surgery with an APR as salvage therapy for patients failing definitive CRT (with recurrent or persistent disease) is associated with 5-year OS rates ranging from 25% to 50%.63-71 Patients who have recurrent disease within the groin after CRT should undergo an inguinal node dissection, with or without an APR, because this is potentially curative in the salvage setting.72 In addition, surgery may be needed to manage the toxicity after definitive CRT. A secondary analysis of RTOG 98-11 showed that 12% of patients required permanent colostomy because of treatment-related complications from non-IMRT CRT.73 It should be noted that this rate is substantially lower using modern-day RT techniques, and no patients required a permanent colostomy in RTOG 0529.47, 58

Metastatic Disease How Should Patients With Metastatic Disease Be Managed?

Although most patients with SCCA present with localized disease, the incidence of distant-stage disease has tripled in recent decades.74 First-line therapy for patients with metastatic disease is platinum-based doublet therapy with either carboplatin or cisplatin and paclitaxel or fluorouracil.75-79 The phase 2 InterAACT trial (ClinicalTrials.gov identifier NCT02051868) randomized patients with advanced SCCA to treatment with either carboplatin and weekly paclitaxel or cisplatin and infusional fluorouracil.77 Not only was there an improvement in median OS (20 vs 12 months; P = .01) with carboplatin/paclitaxel, but the toxicity profile was more favorable (36% vs 62% rate of adverse events; P = .02), establishing carboplatin/paclitaxel as the preferred regimen for patients with metastatic disease. Other regimens, such as DCF (docetaxel, cisplatin, and fluorouracil) and FOLFCIS (fluorouracil, leucovorin, and cisplatin) have been studied,80, 81 but carboplatin/paclitaxel remains the preferred regimen in the United States for managing patients with metastatic disease because it is the only regimen supported by randomized data.

As discussed above, immunotherapy has been studied in patients with advanced or metastatic SCCA because of its success in treating other HPV-related malignancies. The previously discussed immune checkpoint inhibitors targeting the PD-1 pathway, nivolumab and pembrolizumab, are currently approved in the NCCN guidelines for patients with metastatic SCCA who have progressed on first-line chemotherapy given data showing response rates of 20% to 24%.38, 39, 82

Future Directions for Metastatic Disease

The NCI9673 study (Part B; ClinicalTrials.gov identifier NCT02314169) has been expanded to a randomized phase 2 trial of nivolumab with or without ipilimumab in refractory metastatic anal cancer and is expected to complete enrollment by December 2021. Ongoing trials are investigating the role of immunotherapy as part of first-line treatment for patients with metastatic disease. POD1UM-303/InterAACT2 (ClinicalTrials.gov identifier NCT04472429) is phase 3 trial studying the addition of retifanlimab, an anti–PD-1 antibody, to standard chemotherapy for patients with previously untreated, metastatic SCCA.83 Similarly, the phase 3 clinical trial EA2176 is underway to test the addition of nivolumab to chemotherapy versus chemotherapy alone (2:1 randomization) for patients with metastatic SCCA (ClinicalTrials.gov identifier NCT04444921). It is also evaluating the role of HPV circulating tumor DNA (ctDNA) with regard to response and survival.84 ctDNA levels before and after treatment with chemotherapy for advanced SCCA have been shown to correlate with patient outcomes.85 Thus the use of ctDNA both for prognostic and therapeutic purposes is being investigated in clinical trials to guide the management of these patients. In addition to HPV ctDNA, comprehensive genomic profiling is being investigated to provide novel therapeutic approaches for patients with metastatic SCCA.86

Follow-Up How Has Follow-Up Evolved?

A post-hoc analysis of ACT II revealed that SCCA continues to respond to CRT up to 26 weeks after treatment.87 An astonishing 72% of patients who did not have a clinical CR to treatment at 11 weeks had achieved one by 26 weeks. As a result, the NCCN recommends performing a digital anorectal examination for patients at 3 and 6 months to assess for CR. If, after 6 months, patients have persistent or recurrent SCCA, they are considered for salvage APR.82 Ideally, however, it would be important to have examinations performed by surgeons at the 3-month and 6-month visits to accurately assess the scar or persistent disease for appropriate surgical planning because reconstruction typically involves a plastic surgery reconstructive service and may take several weeks to plan.

For patients who have recurrent or persistent disease, most (approximately 70%) experience disease recurrence/persistence at the site of the primary disease and within 24 months from their time of treatment, highlighting the continued importance of optimal local control for patients with SCCA.88 SCCA is best assessed through clinical examination after therapy. Lesions within the anal canal are not well visualized on computed tomography, and positron emission tomography/computed tomography evaluation can reveal fluorodeoxyglucose avidity within the anal canal as a result of inflammation from treatment, resulting in a false assessment. It is important to emphasize the importance of in-person clinical assessments, even during the time of distancing recommendations because of the COVID-19 pandemic, to identify local recurrences early.

HIV-Positive Patients How Should HIV-Positive Patients Be Managed?

Data show that patients with HIV have outcomes equivalent to those of patients without the disease and thus should be managed no differently. Although older data preceding the use of modern-day highly active antiretroviral therapy showed