Anal Cancer

ANAL CANAL
David Marshall/Sheilaine Mabanta
** Updated 4/02**

Epidemiology

  • Majority is SCCA.
  • Median age is 55-65 years.
  • Risk factors are trauma, warts, anal fistula, anal intercourse, HPV-16 and HPV-18.
  • Any form of immunosuppression is risk factor (HIV, transplant, …).

Anatomy

  • There are 3 regions: anal margin, anal canal, and lower rectum.
  • Anal canal is 3-4 cm long from the verge to anorectal ring.
  • Tumor involving the verge is anal canal tumor.
  • Anal margin is the hair bearing skin below (exterior/distal to) the anal verge to 5 cm radius outward.

Pathology

  • SCCA, transitional carcinomas (cloacogenic), and adenocarcinomas.
  • Rare ones include sarcomas, melanomas, and lymphomas.

Clinical Presentation and Route of Spread

  • Bleeding, pain, itching, mass
  • Locally into muscle or adjacent structures
  • Regional nodal risk increases with depth of invasion:

Lymph Node Risk at Presentation per Surgical Exploration

Peri-rectal nodes 30%
Internal iliac nodes 30% (via middle hemorrhoidal artery to internal iliac/hypogastrics)
Inguinal nodes 25% (usually unilateral)
  • Lymphadenopathy clinically evident in 20% of cases at presentation
  • surgical exploration reveals 40% (30-63%) [Perez 3rd Ed].
  • In addition to 20% clinically apparent nodes, we know from late failures that additional 20% of patients have inguinal node metastasis. [((bibcite golden ))]
  • Distant metastases <5% at diagnosis
  • 10% will eventually develop distant metastasis as sole site of failure: lung, liver, bones,

Diagnostic Studies

  • H/P
  • Proctoscope with biopsy of tumor and also nodes if clinically positive
  • Radiological studies: CXR, Abdominal/pelvic CT, transanal US is optional
  • Labs: CBC, LFT Chemistries

Staging

T1  2cm N1 peri-rectal nodes
T2 >2cm but 5cm N2 unilateral internal iliac (hypogastric) or inguinal
T3 >5cm N3 peri-rectal and inguinal nodes; and/or
T4 into adjacent structures bilateral internal iliac or inguinal nodes

I T1N0 IIIA T4N0, T1-3N1
II T2-3, N0 IIIB T4N+ or anyTN2-3
IV M1

Treatment

  • Surgery
  • Abdominoperineal resection (APR)
  • remove the anus, the rectum, and part of the sigmoid colon through an incision made in the abdomen
  • colostomy.
  • Lymph nodes may be removed
  • APR results in local failures in 27-47% of cases
  • OS-5 50-70%
  • No randomized trials comparing surgery with radiation treatment
  • Surgery generally reserved for salvage and T1 lesions which can be excised without compromising sphincter function.
  • Small perianal cancers can be excised with 1 cm margin if function can be preserved.
  • Radiation Alone
  • Patients curable with organ preservation in 65-75% of cases, which compares favorably with APR data
  • 18 patient with T1 and T2 had 100 FFR and 94% OS-5 using 67 Gy (Mayo: Martenson)
  • ChemorRadiation
  • Standard therapy in US is continuous infusion 5FU + bolus Mitomycin C x 2 + pelvic RT.
  • Nigro 1974: reported 3 patients who had path negative surgical specimens after chemo/RT (30Gy/15 fx).
  • Nigro (Wayne State 1981): **19 pts tx\’d with 30Gy + 5FU/MMC then APR
  • 15/19 clinical CR
  • 7/12 path CR after APR
  • 7/7 path CR after WLE
  • Cummings (PMH 1981):** 13 pts 50Gy + 5FU/MMC; all pts controlled locally and distantly.
  • EORTC (Bartelink JCO 1997): **110 pts, T3/4 or N1-3, **randomized between RT alone or RT/5FU/MMC.
  • EBRT: Node - 45Gy/25fx, 3- or 4-field technique, 3 cm below tumor to L5/S1

Node + 45Gy/25fx AP:PA with electrons to boost nodes.

  • Chemo: 5FU via CI on days 1-5 and 29-33 + MMC 15mg/m2 on day 1.
  • Wait 6 weeks, if CR then boost (with e-, photons, or Ir-192) of 15Gy, 20Gy for PR.
  • Surgery for no response
  • Results: RT/chemo improved CR rate (54% vs 80%), LRC (50% vs. 70%), and colostomy-free survival (40% vs. 75%).

OS same 56% Chemo/RT improves organ preservation compared to RT alone.

  • UK Trial (Lancet 1996): randomized 585 pts to RT alone v. RT+5FU/MMC (25% of pts had anal margin!)
  • EBRT: 45GY/20-25 fx continuous course
  • Chemo: 5FU (1000mg/m2/d x 4) + MMC (12mg/m2 x 1 dose)
  • Results: Improved LC, CSS.

No difference in OS.
Increased acute morbidity, no difference in late toxicity.

  • What is optimal RT dose when combined with chemo?
  • RTOG 87-04: 45-50 Gy + 5FU + bolus MMC. If PR then chemotherapy + RT boost. If NR in 6-8weeks then APR.
  • Results: local control 70-80% with 45-50Gy continuous course + MMC
  • RTOG 92-08: Dose escalation with chemo/RT split course. 47 patients, T2.
  • Chemo: 5FU CI x 4d + MMC 10 mg/m2
  • EBRT: 59.6Gy/1.8 Gy/fx with 2 week split, 50.6 Gy to involved inguinal nodes.
  • Results: Increased colostomy rate @ 2 yr (30 vs 7%)with split course higher dose RT compared to 87-04 data.
  • Concluded continuous course needed for dose escalation
  • current RTOG protocol (RTOG 98-11) : The trial is to compare different Chemotherapy agents (5-FU/ MMC vs 5-FU / Cisplatin)with concurrent RT (30.6 Gy — 45 Gy — then Boost to 55 or 59 Gy) in order to reduce the toxicity of the MMC
  • MDAH (Lori Hughes): Retrospective study showing improved outcomes for doses 55Gy with 5FU.
  • Is Mitomycin necessary?
  • **Intergroup study (Flam JCO 1996): **310 pts, randomized to RT/5FU or RT/5FU + MMC
  • EBRT: 45-50.4 Gy
  • Chemo: 5FU (1000mg/m2/d x 4d)  MMC (10mg/m2 x 2 doses)
  • Results: MMC lowered positive re-biopsies, LF, and need for colostomy, and better DFS

Overall survival not improved. Colostomy free survival 71% v. 59%.
However, there were greater complications (23% v. 7% grade 4 and 5).

  • Cummings (PMH 1991): 192 pts, non-randomized, RT alone v. RT/5FU v. RT/5FU/MMC
  • Improved LC with RT/5FU/MMC (86% v. 60% v. 56%) although no improvement for T1
  • How about 5FU and Cisplatin instead of Mitomycin-C?
  • Gerard 1998: retrospective, 95 pts, most advanced stage, all treated with 5FU/Cisplatin + RT
  • EBRT: Total dose \~50Gy. 30 Gy/10 fx (400 cGy given dose) from a perineal field + 18Gy/6 fx sacral field via arcs to tumor and peri-rectal nodes only (total EBRT 48Gy). Highest tumors or large peri-rectal nodes were treated with 39Gy/13 fx  4FP boost of 9-12 Gy/3-4 fx (Total EBRT 39-51Gy)
  • Chemo with EBRT:5FU 96 hr continuous infusion + cisplatin 25mg/m2 days 1-4
  • Implant 1r-192: 19 Gy at \~100cGy/hr (EBRT tumor boost 10-16Gy if unable to give implant)
  • Surgery: Inguinal node dissection done first for N2-3 disease
  • Results: Local control 80%, Ultimate Local control 93%, Salvage rate \~70%

5yr overall survival 84%, for T1, T2, T4 90%, for T3 =70%
82% anal preservation
6% colostomy for severe bleeding or necrosis
1 treatment related death

Treatment Recommendations:

  • Dose
  • 54-59.4 for T1/T2 and T3/T4 respectively

Fun Facts
UF technique of EBRT + Implant per Papillion (Lyon, 1989):

  • EBRT to pelvis (4Field) and inguinal nodes (QS with AP electrons) to 45Gy and wait 2 weeks
  • boost tumor to 15Gy with either

1) implant with single plane Ir-192, 100 cGy/ hr at 0.5cm from plane if
<2/3 circumference of anus, <4-6 cm long, otherwise
2) perineal Co-60,(5-year survival of 67%) If >2/3 circumference, within 3-4cm of verge, otherwise
3) 3-field prone boost if >6cm, starts beyond 3-4 cm from verge.

  • Always electively treat inguinals: Incidence of inguinal nodal failure after RT is 0-5% and if no RT 7-22%.

Chemo: Use concurrent RT and 5FU/MMC or 5FU/Cis for: 1) T3, 2) N+, 3) cloacogenic, if medically fit.

Surgery:

  • If incontinent of bowel, 30Gy + chemo followed by APR.
  • APR for salvage

Outcome

  • Surgery(APR): OS = 32-71%
  • If inguinal node positive then survival after APR is 10-20% at 5 years.
  • CMT: __
  • LC rates are 80-90%
  • 5-year survival 75%.
  • From Gerard 1998 using RT/5FU/Cisplatin:__
  • Local control 80%
  • Ultimate Local control 93%
  • Salvage rate \~70%
  • 5 yr overall survival 84%,
  • 5 yr OS T1, T2, T4 = 90%

T3 =70%

  • 82% anal preservation
  • 6% colostomy for severe bleeding or necrosis
Bibliography
1. Golden GT, Horsley JS III. Surgical management of epidermoid carcinoma of the anus. Am J Surg 1976;131:275-280.
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-Share Alike 2.5 License.