The Bethesda Handbook of Clinical Oncology

Chaoyuan Kuang and James J. Lee

INTRODUCTION

In the United States, anal cancer represents a rare malignancy and accounts for 2.5% of all gastrointestinal malignancies; 8,200 new cases are diagnosed annually in the United States. The incidence has been increasing over the past four decades. The most significant risk factors are sexually transmitted viruses, tobacco smoking, and immunosuppression. Progress has been made over the years in the management of anal cancer. In the 1970s, treatment focused on abdomino-perineal resection (APR). Initially Nigro ND and colleagues at Wayne State employed preoperative chemotherapy with 5-FU and mitomycin (MMC) with radiation therapy (30 Gy) to improve on local control. Complete pathologic responses were discovered and ushered in the concept of definitive chemoradiation, which continues to be the mainstay of therapy for localized anal canal cancer. Since then, intensity-modulated radiation therapy (IMRT) and targeted biologic therapies have been recent advances which are not yet standard of care.

EPIDEMIOLOGY

The annual age-adjusted rates in the U.S. Surveillance, Epidemiology, and End Results (SEER) registry for 2009 to 2013 had dropped to 1.5 per 100,000 for males but increased to 2.1 for females. Cancers of the anus, anal canal, and anorectum are some of the few cancers that are more common in females than in males at nearly all ages. A potential reason for the rise in females may relate to the evolving sexual practices and association with anal HPV infections. Median age at diagnosis is 61. Advancing age is a risk factor for anal canal cancer. In certain populations such as HIV positive men who have sex with other men (MSM), the rate of anal cancer can be as high as 131 per 100,000.

ETIOLOGY AND RISK FACTORS

Several risk factors have been associated with the development of anal cancer:

Sexual activity

•10 or more lifetime sexual partners

•receptive anal intercourse before the age of 30

•history of gonorrhea or syphilis or herpes simplex 2 or chlamydia

•history of cervical cancer

Human papillomavirus infection. Up to 93% of squamous cell carcinoma (SCC) of the anal canal has been associated with HPV infection, which is believed to cause tumor transformation by deregulating the cell cycle and evading immune surveillance. Women are more likely to have an HPV-associated anal cancer than men. Also HPV infection is more common in MSM. HPV 16 and 18 are the most frequently associated strains linked with anal cancer and account for 90% of anal cancers. HPV vaccination has been shown to reduce the risk of premalignant lesions (AIN 2/3).

HIV infection. The incidence of anal cancer in HIV-infected MSM was found to be as high as 131 per 100,000 males in a recent study. The effect of HIV infection on incidence is less dramatic at 46 per 100,000 in other HIV-infected men and 30 per 100,000 in HIV-infected women. It is unknown whether HIV infection directly affects the pathogenesis or if the impact is through the interaction with HPV. Loss of T cell activity likely contributes to the failure of immune clearance of tumor cells.

Cigarette Smoking. Case–control studies indicate increased risk in smokers and especially among current smokers.

There are currently no guidelines for screening high risk individuals for anal cancer or AIN due to lack of proven benefit. A phase III trial is currently underway to examine whether chemical or surgical ablation versus observation of AIN in HIV positive patients increases the time to the development of anal carcinoma (NCT02135419).

PATHOLOGY

The anal canal measures approximately 3 to 4 cm in length. It extends from the anal verge to the puborectalis muscle of the anorectal ring. The dentate line is situated within the anal canal and the histology separates depending on the location above or below the dentate line. Proximal to the dentate, the histology is columnar epithelium, and distal to the dentate, the histology becomes squamous cell epithelium. The anal margin has been arbitrarily defined as an area within 5 cm of the anal verge.

Drainage proximal to the dentate line follows the distal rectum to the internal iliac lymph nodes (pudendal, hypogastric, and obturator). Drainage from the perianal skin, anal verge, and the region distal to the dentate line follows the superficial inguinal lymph nodes with some flow to the femoral nodes and external iliac lymphatics.

The anus comprises three different histologic types: (1) glandular, (2) transitional, and (3) squamous mucosa. Cancers arising from the transitional or squamous mucosa develop into squamous cell carcinomas. The basaloid or transitional carcinomas (formerly known as cloacogenic or junctional tumors) develop from the transitional mucosa. Those cancers developing above the dentate line are nonkeratinizing squamous cell carcinomas versus those distal to the dentate line are keratinizing squamous cell carcinomas. Tumors arising from the glandular mucosa of the anal canal develop into adenocarcinomas. Anal margin tumors develop within the hair-bearing skin distal to the transitional mucosa.

CLINICAL PRESENTATION

Rectal bleeding and anal discomfort are the two most common symptoms, occurring in over 45% and 30% of patients, respectively. Pruritus and discharge are other symptoms. Pain can be severe. Changes in bowel habits can be a presenting symptom, especially with proximal anal canal cancers. Patients may be asymptomatic as well.

MEDICAL WORKUP

Workup should include anoscope with biopsy (incisional), digital rectal examination (DRE), inguinal lymph node evaluation (biopsy or FNA of any suspicious lymph nodes), chest CT, abdominal/pelvic CT, or MRI. Pelvic examination should be performed on women including screening for cervical cancer. Consider HIV testing and CD4 levels for patients at risk. While PET/CT has not been validated as a tool for staging compared to CT in a prospective trial, PET/CT has resulted in up- and downstaging due to the presence or absence of positive lymph nodes and should be considered if advanced radiation techniques are planned. Full chemistries and CBC should be performed as well.

Staging is based on the seventh edition of the American Joint Committee on Cancer (AJCC), which employs a TNM system for the staging of anal canal cancers. T stage is based partly upon the size of the primary lesion or the invasion of nearby structures such as the bladder, prostate, vagina, or urethra. The N stage is determined by the presence of perirectal, internal iliac, or inguinal lymph nodes. The M stage is based upon the presence or absence of distant metastases.

PROGNOSTIC FACTORS

Tumor size. The size of the primary lesion has been shown to be one of the most significant factors in predicting local control and survival for lesions confined to the pelvis.

Lymph nodes. The presence or absence of lymph nodes also has been shown to impact survival.

Metastasis. The most significant prognostic risk factor for overall survival is the presence or absence of extrapelvic metastases.

HIV status. High viral load and low CD4+ count in some series have predicted for survival and local control.

Other: Hemoglobin levels ≤10 g/L, male gender, p16+ HPV status, and ulceration of the primary lesion have impacted prognosis in some studies.

TREATMENT

Surgery

Anal Canal Lesions

In the 1970s, surgical resection with an APR was considered standard of care. APR produced local control rates of 70%, and overall survival rates from 20% to 70% (average 50%). With inguinal lymph node involvement, some series showed 5-year survivals of 10% to 20%. After the 1974 publication by Nigro et al. showing complete pathologic response in three patients treated with chemoradiation, anal cancer patients are rarely treated with upfront surgery, and definitive chemoradiation remains the current standard of care. APR is reserved for salvage after failure with definitive chemoradiation or reserved for management of radiation complications.

Surgery alone with local excision may be considered with small, localized T1N0M0 squamous cell carcinomas of the anal canal. Several small retrospective series have demonstrated good local control and 5-year survival with such an approach. The key to offering local excision is patient selection. Patients with small tumors <2 cm, well differentiated, and no involvement of the sphincter may be considered candidates. Otherwise, chemoradiation should be offered.

Anal Margin Lesions

Early anal margin cancers have traditionally been treated with local excision. Such lesions behave more like a skin cancer, although this concept has never been validated prospectively. Wide local excision has been reserved mostly for well-differentiated T1N0M0 lesions with good local control. In a retrospective review of 48 patients with squamous cell carcinoma of the anal margin, 31 patients underwent local excision, and 11 were treated by APR. Local excision provided satisfactory results with a 5-year survival of 88%. However, larger lesions T2 or > or N+ should be treated with definitive chemoradiation.

Radiation Therapy

Since the Nigro et al.’s publication in 1974, radiation therapy has become the primary curative modality for the treatment of anal cancer. Radiation techniques have evolved over time and varying types of radiation therapy have been used including external beam radiation therapy (3D and IMRT), electrons, and brachytherapy.

Radiation alone has been used to treat early anal cancer (T1–T2N0M0) with relative success; most retrospective studies have demonstrated modest local control and 5-year survivals; however, not all studies show good local control with radiation alone. Tumors <2 cm appear to have better local control in particular. NCCN guidelines, however, recommend combined modality for even small T1N0M0 squamous cell carcinomas of anal canal. For the very elderly or those with significant comorbidities, radiation alone may be a reasonable approach.

Combined Radiation Therapy and Chemotherapy as Standard of Care

Nigro et al. at Wayne State developed the concept of treating anal cancer patients preoperatively in order to decrease APR failures. The treatment regimen was 5-FU 1,000 m² on days 1 to 4 and 29 to 32 and mitomycin-c 10 to 15 mg/m² on day 1, combined with moderate pelvic RT dose of 30 Gy. With the discovery of three pathologic complete responses after preoperative chemoradiation, the focus changed to preserving the sphincter using chemoradiation and reserving APR for salvage. Since Nigro et al.’s original publication, definitive chemoradiation therapy has become the standard. Several retrospective series have demonstrated the success achieved with chemoradiation therapy in terms of local control and overall survival (Table 11.1).

TABLE 11.1 Randomized Phase III Trials

Trial	No. of Pts	Eligible Pts	Study Design	Treatment	5-Y LFR	CR	5-Y OS	DFS/RFS	Colostomy Rate
UKCCCR	585	T1–T4, N0–N3 M0	RT vs. CRT	MMC 12 mg/m² d 1; 5-FU 1,000 mg/m² d 1–4, 29–32 RT 45 Gy/25#; 15 Gy boost	57%-RT; 32%-CRT (P < 0.0001)	CR at 6-wk post-Tx: 30%-RT 39%-CRT	5-y OS: 53%-RT 58%-CRT 10-y OS: 34%-RT 41.5%-CRT	3-y DFS: 38%-RT 56%-CRT 5-y RFS: 34%-RT 47%-CRT	5-y CFS: 37%-RT 47%-CRT
EORTC 22861 (1987–1994)	110	T3–T4 N0–N3 M0 or T1–T2 N1–N3 M0	RT vs. CRT	MMC 15 mg/m² day 1; 5-FU 750 mg/m² days 1–5, 29–33 RT 45 Gy/25#; 20 or 15 Gy boost	50%-RT 32%-CRT at 5 y P = 0.02	54%-RT 80%-CRT at 6 wk post-therapy	54%-RT 58%-CRT (P = 0.17)	Estimated improvement in DFS by 18% at 5 y	Estimated CFS improvement of 32% at 5 y
RTOG-8704/ECOG	291	T1–T4 N0–N1 M0	5-FU/RT vs. 5-FU/MMC/RT (if biopsy is +, 5-FU/CDDP + RT)	MMC 10 mg/m² d 1 and 29; 5-FU 1,000 mg/m² d 1–4, 29–32 RT 45–50.4 Gy/25–28#; if biopsy is positive, then 9 Gy boost	4-y LRF: 16%	CR at 4-6-wk post-Tx: 86%-5-FU 92.2%-MMC	71%-5-FU 78.1%-MMC	4-y DFS: 51%-5-FU 73%-MMC (P = 0.0003)	4-y CFS 22%-5-FU 9%-MMC (P = 0.002)
RTOG 98-11	644	T2–T4 N0–N3 M0	Neoadj CDDP/5-FU then CDDP/5-FU/RT vs. 5-FU/MMC/RT	Neoadj with CDDP 75 mg/m², 5-FU 1,000 mg/m² d 1–4, then CRT with 5-FU/CDDP vs. CRT with MMC 10 mg/m² d 1–4, 29–32; and 5-FU 1,000 mg/m² d 1–4, 29–32 Rt 45 Gy/25#; T3/T4, N+, or T2 with residual- received a boost to 54–59 Gy	71.9%-MMC 65%-CDDP (P = 0.087)	CR at 6-wk post-Tx: 30%-RT 39%-CRT	78.3%-MMMC 70.7%-CDDP (P = 0.026)	5-y DFS: 67.8%-MMC 57.8%-CDDP (P = 0.006)	5-y CFS: 71.9%-RT 65%-CRT (P = 0.05)
ACCORD-03	307	T ≥ 4 cm or T < 4 cm and N1-N3 M0	Arm A: Neoadj 5-FU/CDDP + Std RT Arm B: 5-FU/ CDDP + HD RT Arm C: 5-FU/ CDDP + Std RT Arm D: 5-FU/ CDDP + HD RT	Neoadj chemo: 5-FU 800 mg/m² d 1–4 and 29–32, CDDP 80 mg/m² d 1 and 29 CRT: 5-FU and CDDP –dose same Arms A and B: 45 Gy/25# to pelvis + std dose boost of 15 Gy or brachy boost (BT) Arms C and D: 45 Gy/25# to pelvis + high-dose boost of 20–25 Gy or BT;	A: 72% B: 87.6% C: 83.7% D: 78%	A: 92% B: 97% C: 86% D: 94%	A/B: 74.5% C/D: 71% (P = 0.81) A/C: 71% B/D: 74% (P = 0.43)	3-y DFS: A: 63.8% B: 78.1% C: 66.8% D: 62.3%	5-y CFS: A: 69.6% B: 82.4% C: 77.1% D: 72.7%
ACT II	940	T1–T4 N0–N3 M0	CDDP/5-FU-CRT vs. MMC/5-FU-CRT, then 4 wk later randomized to maintenance CDDP/5-FU vs. no maintenance	CDDP 60 mg/m² d 1 and 29, 5-FU 1,000 mg/m² d 1–4, d 29–32, and CRT; vs. MMC 12 mg/m² d 1, 5-FU 100 mg/m² d 1–4 and 29–32 CRT; maintenance chemo × 2 cycles = 5-FU 1,000 mg/m² and CDDP 60 mg/m² RT 50.4 Gy/28#	11% MMC 13% CDDP	CR at 12-wk post-Tx: 94% 5-FU/MMC/RT vs. 95% 5-FU/CDDP/RT	3-y OS: 85% with maintenance; 84% without maintenance	3-y DFS: 75% MMC 75% CDDP	5% with maintenance; 4% without maintenance

UKCCCR, United Kingdom Coordinating Committee on Cancer Research; RTOG, Radiation Therapy Oncology Group; ECOG, Eastern Cooperative Oncology Group; ACT, Anal Cancer Trial; RT, radiation therapy; 5-FU, 5-fluorouracil; MMC, mitomycin C; CDDP, cisplatin; CRT, chemoradiation; Std RT, standard dose radiation therapy; HD RT, high-dose radiation therapy; LRF, local-regional failure; DFS, disease-free survival; CFS, colostomy-free survival; RFS, relapse-free survival; f/u, follow-up.

Adapted from Lim F, Glynne-Jones R. Chemotherapy/chemoradiation in anal cancer: a systematic review. Cancer Treat Rev. 2011;37:522.

Radiation Therapy Alone versus Combined-Modality Therapy

Two prospective randomized trials have been conducted that have compared radiation alone versus chemoradiation (Table 11.1). The United Kingdom Coordinating Committee on Cancer Research (UKCCCR) trial enrolled patients with T1–T4, N0–N3, M0 anal cancer, and small number of anal margins and demonstrated improved local control and colostomy-free survival (CFS) with combined modality treatment (CMT) but no statistical difference in overall survival. Even for early anal cancer patients (T1–T2N0M0), there appeared to be a benefit favoring CMT on multivariate analysis.

The European Organization for Research and Treatment of Cancer (EORTC) conducted a similar trial comparing radiation alone to radiation with chemotherapy (5-FU and MMC), which also demonstrated an advantage toward CMT. This trial enrolled locally advanced anal canal cancer patients, T3–T4, N0–N3, and M0. Again, CMT showed a statistically significant advantage for CMT in terms of local control, colostomy rate, and disease-free survival (DFS) but not for overall survival. Thus, both the UKCCCR and EORTC trials established the principle that combined modality is superior to single modality radiation alone.

Value of MMC in the Combined-Modality Regimen

Due to the hematologic toxicity of MMC in CMT, two prospective phase III trials (RTOG/ECOG 8704 and RTOG 98-11) have evaluated the importance of MMC.

RTOG/ECOG 8704 was the first randomized trial to evaluate prospectively the importance of MMC in CMT and compared 5-FU and MMC with RT (standard arm) versus 5-FU and RT (experimental arm). The primary endpoint was DFS. At 4 years, DFS was 73% for 5-FU and MMC compared to 51% for 5-FU. Colostomy rate was 22% with 5-FU versus 9% with 5-FU, MMC (P = 0.002). Despite the higher toxicity of MMC, the authors concluded that MMC was still the preferred regimen, given the higher DFS and lower colostomy rate.

RTOG 98-11 was the second randomized trial to evaluate prospectively the role of MMC in CMT and compared 5-FU and MMC with concurrent radiotherapy (standard arm) versus induction 5-FU and CDDP followed by concurrent 5-FU and CDDP with radiotherapy (experimental arm). The primary endpoint was again DFS. With a median of 2.51 years, the initial report revealed no statistically significant difference in 5-year DFS (60% MMC vs. 54% CDDP), 5-year OS (75% MMC vs. 70% CDDP), 5-year local-regional control rates (25% MMC vs. 33% CDDP), or 5-year distant metastasis (DM) rates (15% MMC vs. 19% CDDP). The cumulative colostomy rate proved to be statistically significantly higher with CDDP, 19% versus MMC, 10%.

However, the updated RTOG 98-11 examined the long-term impact of treatment on survival (DFS, OS, CFS), as well as colostomy failure (CF), and locoregional failure (LRF), and DM. With longer follow-up, 5-FU/MMC regimen produced statistically significant improvement in DFS and OS for RT + 5-FU/MMC versus RT + 5-FU/CDDP (5-year DFS; 67.8% vs. 57.8%; P = 0.006; 5-year OS, 78.3% vs. 70.7%; P = 0.026). There was a trend toward statistical significance for CFS, LRF, and CF. Thus, the authors conclude that RT + 5-FU/MMC remains the preferred standard of care.

ACT II addressed two questions: (1) whether substituting CDDP for MMC improves the complete response rate and (2) whether two cycles of maintenance chemotherapy (5-FU/CDDP) reduce recurrences. The randomization employed a 2 × 2 factorial, and patients were randomized to 5-FU/MMC with concurrent radiotherapy (standard arm) or CDDP/5-FU with concurrent radiotherapy. Patients were then randomized to receive either two cycles of maintenance chemotherapy (CDDP/5-FU) or no maintenance therapy. At 26 weeks, the complete response rate was 90.5% for MMC and 89.6% for CDDP (P =0.64). No statistically significant differences in terms of 3-year recurrence-free survival and overall survival were noted between CDDP and MMC regimens, nor between the maintenance and the no maintenance arms. The authors again concluded that 5-FU/MMC with RT should remain the standard of care.

The data from ACT II and the updated RTOG 98-11 may seem to contradict one another: the CDDP arm in RTOG 98-11 appears to have a detrimental effect on DFS and OS, whereas the ACT II shows that CDDP may be at least equivalent to that of MMC. However, the trial designs were quite different. RTOG employed the use of neoadjuvant chemotherapy prior to the start of concurrent chemoradiotherapy. The prolongation of the overall treatment time may account for the inferior results of the CDDP arm. However, given the current data available, 5-FU/MMC should remain the standard of care.

In a single institution retrospective study comparing the efficacy of one cycle of MMC versus two cycles of MMC in combination with radiation, there was no significant difference in PFS (78% vs. 85%; P = 0.39) or OS (84% vs. 91%; P = 0.16). There were statistically significant fewer incidences of grade 2 or higher hematologic, dermatologic, and gastrointestinal toxicity in the single cycle versus the two cycles. While prospective multicenter data are lacking, this study suggests that a single cycle of MMC may be as efficacious as two cycles with less acute treatment-related toxicity.

Replacing 5-FU with Capecitabine

Capecitabine is a readily absorbed oral precursor of 5-FU with demonstrated efficacy and favorable side effect profile in the treatment of early stage or metastatic colorectal adenocarcinoma. Its theoretical benefit is that oral dosing will allow continuous therapeutic levels of drug during a treatment cycle, and preferential activation of capecitabine in tumor tissue over normal tissue enhances the therapeutic ratio. Clinical benefit of capecitabine has not yet been demonstrated with equal rigor in anal carcinoma as it has in colorectal cancer. Two phase II trials have demonstrated CR rates of 86% and 90% at 6 months, with almost no grade 4 toxicity in patient with stage I-IIIB anal carcinoma. NCCN guidelines currently recommend CMT with capecitabine/MMC as an alternative to 5-FU/MMC.

Acute Toxicity

Patients typically experience moderate to severe acute toxicities from the combination of both chemotherapy (5-FU and MMC) and radiation therapy. Side effects include nonhematologic toxicities (nausea/vomiting, abdominal pain, increased frequency of stool, diarrhea, skin irritation, fatigue, and weight loss) and hematologic toxicities (neutropenia, thrombocytopenia, anemia).

Toxic deaths from CMT have ranged from 0% to 5%. In the UKCCCR study, 6/116 (2%) experienced toxic death, mostly due to septicemia. The EORTC trial reported on 1 toxic death out 110 patients. In the RTOG 8704 study, four patients (3%) experienced death in the MMC arm. More recently, there were no reported toxic deaths in both RTOG 98-11 and ACT II trials. The ACCORD 03 trial, a four-arm randomized trial, showed similar toxic deaths across all four arms (A = 1 [1%], B = 2 [2.6%], C = 3 [3%], D = 1 [1%]). No patient required an APR for acute toxicity in any of the arms during the induction phase or the concurrent chemoradiation phase.

Late Toxicity

Late effects have not been well documented within the randomized trials. Part of the challenge in evaluating late effects is the differing toxicity scales used in the various trials. Early toxicity criteria used in the randomized trials did not allow for characterizing radiation-induced side effects.

Both the early EORTC and UKCCCR trials did not demonstrate a difference in long-term complications between those receiving CMT versus RT alone. RTOG 8704 similarly revealed no significance difference in long-term toxicity between RT/5-FU and RT/5-FU/MMC, although two patients from each arm required stoma secondary to RT-related complications. In the RTOG 9811 update, the most common types of late grade 3 or 4 toxicity included skin, small/large intestine, subcutaneous tissue, or other. There did not appear to be a difference between the MMC or CDDP arms for grade 3/4 toxicity (13.1% vs. 10.7%; P = 0.35). In the ACCORD 03 trial, late toxicities were primarily of grade 1 or 2. However, nine patients experienced grade 4 toxicities including necrosis, fistula, bleeding, or pain of whom five were treated with an APR and four underwent colostomy alone.

Chemoradiation in HIV-Positive Patients

The majority of retrospective studies suggest that HIV-positive patients do just as well with CMT as the general population. Those with CD4 counts of <200, however, may require a modification in their treatment regimen such as omission of MMC or a reduction in the RT field and/or dose.

Dose of Radiation

Local-regional failures occur in 20% to 30% after definitive chemoradiation. Because of such local-regional failures, RTOG 92-08, a phase II, dose escalation trial was designed to escalate dose to 59.4 Gy, with a mandatory treatment break of 2 weeks after the initial 36 Gy. The initial study included 47 patients with a mandatory break. The update of RTOG 92-08 analyzed not only the original 47 patients with the mandatory break but also analyzed 20 additional patients who did not have a planned break. Both groups of patients showed no difference in OS or LRF when compared historically to patients on RTOG 87-04 MMC arm. The higher dose likely did not result in improved outcomes because of the treatment break, which may have allowed for tumor repopulation and/or repair of sublethal damage.

The ACCORD-03 trial evaluated both the value of treatment intensification by induction chemotherapy (two cycles of 5-FU and CDDP) and radiation dose escalation by incorporating a 20 to 25 Gy boost in patients with locally advanced anal cancer patients (T2 > 4 cm or T3–T4Nx or any T, N1–N3, M0). The trial was conducted as a factorial 2 × 2 study (A = ICT; B = ICT +HDRT; C = reference arm = pelvic RT 45 Gy per 25 fractions with two cycles of 5-FU-CDDP + boost of 15 Gy; D = HDRT). High-dose RT (HDRT) incorporated a boost of 20 to 25 Gy. Thus, arms A and C received 60 Gy total and arms B and D received 65 to 75 Gy total. There appeared to be no difference in their primary endpoint of colostomy-free survival (CFS) at 3 years, or in any secondary endpoints such as response rate, toxicity, local control, or overall survival.

Although controversy exists regarding the optimal dose, a reasonable approach is to treat between 55 to 59 Gy (RTOG 98-11) if 3D conformal radiation is being contemplated and 54 Gy if dose painting IMRT (DP IMRT) is being employed per RTOG 0529.

Tumor Regression after Chemoradiation

After patients have completed definitive chemoradiation therapy, patients should be followed up clinically in 8 to 12 weeks after therapy. Cummings demonstrated that mean time for tumor regression was 3 months but regression of a tumor could occur for up to 12 months. Thus, if there is persistent disease at 8 to 12 weeks, patients should be followed up closely (every month) to document regression. As long as there is documented regression on serial examinations, patients may continue to be monitored. However, at any point if there is progression, then biopsy followed by salvage APR should be considered.

Targeted Therapy

Outcomes of several other types of solid tumors have benefited from targeted biologic therapy. However, use of targeted therapy has yet to reach phase III trials in anal carcinoma. Cetuximab is a monoclonal antibody, which blocks epidermal growth factor receptor (EGFR) from sending mitogenic signals via a KRAS dependent pathway. KRAS mutations are rare in anal carcinoma, making this pathway targetable in anal cancer. Reports of the phase II ECOG 3205 and AMC045 trial data evaluating safety and efficacy of cetuximab in addition to CMT with 5-FU/CDDP in both immunocompetent and HIV-positive patient with stage II–III anal carcinoma suggest improved rates of LRF compared to historical data, but also had high rates of grade 4 toxicity at 32% and 26% in EGOC 3205 and AMC045, respectively. A similar phase II trial, ACCORD 16, as well as a phase I trial both looking at safety and efficacy of a 5-FU/CDDP based regimen with cetuximab were terminated prematurely due to unacceptably high rates of grade 3/4 toxicity. Panitumumab is another EGFR antagonist. Preliminary results of a phase II trial (VITAL/GEMCAD 09-02) in which 36 patients with nonmetastatic anal carcinoma received panitumumab in addition to standard of care suggested an acceptable safety profile, with no treatment-related deaths, 8% grade 4 toxicity, and 56% CR rate at 8 weeks and 24 weeks.

Intratumoral HPV oncoproteins upregulate immune checkpoint proteins such as PD-L1 and promote immune resistance. Morris et al. reported the result of a phase II trial of nivolumab monotherapy in previously treated and immune therapy naïve patients (NCT02314169). Thirty-seven patients were enrolled and received at least one dose of nivolumab. Among the 37 patients, nine patients had responses (2 CR and 7 PR; ORR, 24%). Grade 3 adverse events were anemia (N = 2), fatigue (N = 1), rash (N = 1), and hypothyroidism (N = 1). No serious adverse events were reported.

TREATMENT OPTIONS ACCORDING TO STAGE

Stage 0

Surgical resection is the treatment of choice for the lesions of the perianal area that does not involve the anal sphincter.

Stage I

Small, well-differentiated tumors of the anal margin not involving the anal sphincter can be treated with wide local excision.

All the other stage I tumors of the anal margin and anal canal are treated with chemoradiation with 5-FU/MMC or capecitabine/MMC.

Patients who cannot tolerate chemotherapy, such as the very elderly or those with multiple comorbid conditions, may be treated with radiation alone.

Surgical salvage with APR is reserved for residual cancer in the anal canal after chemoradiation.

Stages II to IIIB

Chemoradiation with 5-FU/MMC or capecitabine/MMC is the recommended initial approach.

Patients who cannot tolerate chemotherapy may be treated with radiation alone.

Surgical salvage with APR is reserved for residual disease in the anal canal after chemoradiation.

Stage IV

There are limited data regarding the treatment of metastatic disease given the overall rarity of the disease. There are no available phase III data and only very limited phase II prospective data are available. The most widely used regimen is cisplatin plus 5-FU, which is the recommended as first-line therapy by NCCN. Response rates have been as high as 50% to 66% and median survivals of 12 to 34.5 months. Clinical trials should be encouraged. Palliative efforts remain an important component of care.

PERSISTENT OR RECURRENT ANAL CANCER

Surgery with APR is considered the treatment of choice for either persistent or recurrent disease and 20% to 40% of patients may achieve long-term control. For persistent disease, RTOG 8704 treated 22 patients with a 9 Gy boost with 5-FU and CDDP as salvage. Ultimately, 12 of 22 remained disease-free after surgical intervention. Given the limited data, surgery should remain the standard for chemoradiation failures.

Follow-Up

There are no prospective data regarding the optimal follow-up regimen. The ACT II trial did show that 29% of patient who did not achieve CR at 11 weeks achieved CR at 26 weeks. Based on these data, NCCN recommends an initial follow-up DRE in 8 to 12 weeks following completion of therapy. Patients with regression or no progression may be observed for up to 6 months to see if CR can be achieved and APR spared. If CR is achieved, then DRE, anoscopy, and inguinal lymph node palpation should be repeated every 3 to 6 months for 5 years. For T3–T4 or positive inguinal lymph nodes at diagnosis, one should consider chest/abd/pelvic imaging annually for 3 years. Patients with clear progression or relapse at any point require biopsy and restaging with CT or PET/CT and if progression or recurrence is proven, they should be considered for salvage APR.