- Email: [email protected]
Chemoradiation for adenocarcinoma of the anus
Int. J. Radiation Oncology Biol. Phys., Vol. 55, No. 3, pp. 669 – 678, 2003 Copyright © 2003 Elsevier Science Inc. Printed in the USA. All rights reserved 0360-3016/03/$–see front matter
doi:10.1016/S0360-3016(02)04118-4
CLINICAL INVESTIGATION
Anal Canal
CHEMORADIATION FOR ADENOCARCINOMA OF THE ANUS MICHAEL PAPAGIKOS, M.D.,* CHRISTOPHER H. CRANE, M.D.,† JOHN SKIBBER, M.D.,‡ NORA A. JANJAN, M.D.,† BARRY FEIG, M.D.,‡ MIGUEL A. RODRIGUEZ-BIGAS, M.D.,‡ ARTHUR HUNG, M.D.,† ROBERT A. WOLFF, M.D.,§ MARC DELCLOS, M.D.,† EDWARD LIN, M.D.,§ KAREN CLEARY, M.D.㛳
AND
*Department of Radiation Oncology, Wake Forest University School of Medicine, Winston-Salem, NC; Departments of †Radiation Oncology, ‡Surgery, §Medical Oncology, and 㛳Pathology, The University of Texas M. D. Anderson Cancer Center, Houston, TX Purpose: To assess the efficacy and limitations of definitive chemoradiation for adenocarcinoma of the anal canal and to propose a treatment strategy that addresses the limitations of treatment. Methods and Materials: Between 1976 and 1998, 16 patients with localized adenocarcinoma of the anal canal were treated with radiotherapy with or without chemotherapy with curative intent. Available histologic slides were reviewed for evidence of primary adenocarcinoma of anal duct origin. The treatment results for these patients were compared with those of a group of patients with epidermoid histologic features who were all treated with definitive chemoradiation (55 Gy with concurrent 5-fluorouracil and cisplatin, n ⴝ 92) between 1989 and 1998. The hospital records were reviewed for all patients. Patients with epidermoid carcinoma presented with more advanced primary tumors (42% vs. 19% Stage T3 or greater). All adenocarcinoma patients were treated with radiotherapy (median dose 55 Gy): 11 received concurrent 5-fluorouracil– based chemotherapy and 5 received radiotherapy alone. The initial surgical procedures included abdominoperineal resection, excisional biopsies (n ⴝ 5), and local excision (n ⴝ 1). Abdominoperineal resection was performed as salvage therapy after local recurrence in 5 patients. The Kaplan-Meier method was used to calculate 5-year actuarial pelvic control, distant disease control, disease-free survival, and overall survival. The median follow-up was 45 months (range 5–196) for patients with adenocarcinoma and 44 months (range 9 –115) for patients with epidermoid histologic features. Results: Both local and distant recurrence rates were significantly greater in the adenocarcinoma patients. Of 16 patients with adenocarcinoma, 7 (5-year actuarial rate 54%) had recurrence at the primary site compared with 16 (5-year actuarial rate 18%) of 92 patients with epidermoid histologic features (p ⴝ 0.004). Distant disease developed in more patients with adenocarcinoma (5-year actuarial rate 66%) than in patients with epidermoid carcinoma (5-year actuarial rate 10%, p <0.001). The 5-year actuarial disease-free survival and overall survival rate for adenocarcinoma patients was 19% and 64%, respectively, compared with 77% (p <0.0001) and 85% (p ⴝ 0.017) for those with epidermoid carcinoma. Conclusion: Patients with localized adenocarcinoma of the anus treated with definitive chemoradiation had high rates of pelvic failure and distant metastasis compared with comparably staged patients with epidermoid histologic features treated similarly. On the basis of these limitations, we recommend preoperative chemoradiation followed by abdominoperineal resection to maximize pelvic disease control and consideration of adjuvant chemotherapy to address the problem of micrometastatic disease. © 2003 Elsevier Science Inc. Anal cancer, Chemoradiation, Adenocarcinoma, Radiotherapy.
INTRODUCTION Adenocarcinoma of the anus is much rarer than its counterpart, epidermoid carcinoma of the anus, and the evidence in the literature, albeit limited, suggests it has a more aggressive natural history. Because reports about adenocarcinoma of the anal canal are limited to small retrospective series and case reports (1–5), the optimal therapeutic approach has not
been established and the pattern of disease recurrence after definitive chemoradiation has not been well characterized. In the 1970s and 1980s, abdominoperineal resection (APR) was the mainstay of the treatment of epidermoid carcinoma of the anal canal. By the 1990s, definitive chemoradiation had largely replaced APR as the standard initial management following the recognition that the rate of pelvic disease control with surgery alone was unsatisfactory (7–10),
Reprint requests to: Christopher H. Crane, M.D., Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 97, Houston, TX 77030. Tel: 713-792-3400; Fax: 713-794-5573; E-mail: [email protected]
Supported in part by PO1 CA-06294, T32CA77050, and P30CA16672 awarded by the National Cancer Institute, U.S. Department of Health and Human Services. Received Jun 11, 2002, and in revised form Aug 28, 2002. Accepted for publication Sep 5, 2002. 669
670
I. J. Radiation Oncology
● Biology ● Physics
that chemoradiation was effective against the disease (11– 14), and that recurrences after chemoradiation could be effectively salvaged with APR (14, 15). Historically, the primary treatment of adenocarcinoma of the anus has also been surgical, most commonly APR, requiring permanent colostomy. A population-based study from the National Cancer Data Base (NCDB) reported that approximately 75% of patients diagnosed with adenocarcinoma of the anal canal during a 5-year period underwent surgery (6). Unlike the extensive literature on epidermoid carcinoma, the rarity of adenocarcinoma of the anal canal has precluded any meaningful assessment of the efficacy of definitive chemoradiation for it. Extrapolating from the experience with epidermoid carcinoma, we have advocated a similar approach for adenocarcinoma of the anal canal with the hope that similar local control to that achieved for epidermoid disease would be possible. The purpose of this report was to compare the outcomes and pattern of failure after definitive chemoradiation for adenocarcinoma of the anal canal with those of epidermoid carcinoma of the anal canal after similar treatment. METHODS AND MATERIALS Adenocarcinoma of the anal canal The University of Texas M. D. Anderson Cancer Center Institutional Review Board reviewed and approved the concept for this study. The medical records of patients with a diagnosis of localized adenocarcinoma of the anal canal who were treated at The University of Texas M. D. Anderson Cancer Center between 1976 and 1998 were examined for inclusion in this retrospective study. The hospital records were evaluated for anatomic and histologic criteria consistent with primary anal duct origin. Only the records of patients whose tumors involved the anal canal (defined as 0 – 4 cm from the anal verge) were considered. If ⬎50% of the tumor length was proximal to this region, patients were considered to have adenocarcinoma of the rectum and therefore were excluded. Sixteen patients met the anatomic criteria for inclusion. In an attempt to eliminate rectal cancers from the group further, the histologic appearance of the hemotoxylin-eosin–stained slides was reviewed for evidence of clear rectal mucosa origin vs. clear anal ductal origin. Six cases had sufficient histologic material to support the diagnosis of primary anal duct origin. In 10 patients, the slides were either not available for review or lacked sufficient material to differentiate rectal mucosal from anal ductal origin. Of them, 2 cases had histologic findings indistinguishable from rectal adenocarcinoma, and in 8, no specimens were available for review. All 16 records were included in the study. The medical records of a homogeneously treated group of patients (n ⫽ 92) with localized epidermoid carcinoma of the anal canal who were treated with definitive chemoradiation between March 1990 and September 1998 were reviewed and served as a comparison group.
Volume 55, Number 3, 2003
Workup A surgical oncologist, radiation oncologist, and medical oncologist initially saw all patients. The initial workup consisted of a medical history, physical examination, rigid proctoscopy, routine blood work, chest X-ray, and CT of the abdomen and pelvis. Biopsy was performed by proctoscopy or colonoscopy and malignant disease confirmed by histologic examination of the specimen. Patient and tumor characteristics Demographic and staging information for all patients is listed in Table 1. Patients with adenocarcinoma were of a similar age and race and had a similar tumor size (p not significant). Six (38%) of the 16 patients with adenocarcinoma were women compared with 71 (77%) of 92 patients with epidermoid histologic features (p ⫽ 0.002). More patients with epidermoid carcinoma presented with advanced primary tumors (42% vs. 19% Stage T3 or greater). Five patients (31%) with adenocarcinoma had nodal involvement compared with 28 (30%) with epidermoid carcinoma. Patients with adenocarcinoma more often had highgrade lesions (5 [31%] of 16 vs. 6 [7%] of 92, p ⫽ 0.01). Treatment information The treatment and outcome information for adenocarcinoma patients is listed in Table 2. All patients were treated with radiotherapy (RT) as part of a definitive treatment strategy. Eleven patients received concurrent chemotherapy; in two of them, chemoradiation was an adjuvant to planned APR (one preoperatively and one postoperatively). Five patients were treated with local excision of the primary tumor followed by RT without concurrent chemotherapy. One patient was treated with definitive chemoradiation after presenting with recurrent disease after local excision. In addition to the 2 patients who underwent APR as part of their initial treatment, 5 others underwent APR as salvage therapy for locally persistent disease. The chemoradiation regimen most commonly consisted of continuous infusion 5-fluorouracil (5-FU; 250 –300 mg/m2 within 24 h Monday through Friday concurrently with 55 Gy in 30 fractions within 6 weeks). However, the treatment dose and technique varied considerably during the 22-year study interval. The median RT dose to the primary tumor volume was 55 Gy (range 53–70). The inguinal nodal regions were treated in 9 patients; 5 received a median inguinal dose of 50.4 Gy and 4 received elective inguinal irradiation to 30.6 Gy. Seven patients did not receive inguinal RT. The sequence of treatment for each patient with adenocarcinoma is summarized in Table 3. All patients with epidermoid histologic features were treated with definitive chemoradiation and received 55 Gy in 30 fractions in 6 weeks with concurrent continuous infusion 5-FU (250 mg/m2/24 h, Monday through Friday) and cisplatin (4 mg/m2/24 h, Monday through Friday). The chemotherapy was delivered through a double-lumen subclavian catheter using electromechanical pumps. Surgery (APR) was reserved for persistent or recurrent disease.
Chemoradiation for adenocarcinoma of the anus
● M. PAPAGIKOS et al.
671
Table 1. Demographic and staging information for patients with adenocarcinoma and epidermoid carcinoma
Number of patients Median age (y) Female gender Node positive (n) N1 N2 N3 Tx T1 T2 T3 T4 Recurrent Median maximal tumor diameter (cm) Grade unspecified Grade 1 Grade 2 Grade 3 Median follow-up (mo)
Epidermoid
Adenocarcinoma
92 57 (34–82) 71 (77) 28 (30) 7 14 7 4 (4.3) 6 (6.5) 43 (46.7) 27 (29.3) 12 (13) 0 (0) 5 (1.3–10) 0 (0) 1 (1) 85 (92) 6 (7) 44 (9–115)
16 58 (42–71) 6 (38) 5 (31) 0 2 3 4 (25) 0 (0) 8 (50) 1 (6.3) 2 (12.5) 1 (6.3) 3.8 (2–14) 4 (25) 3 (19) 4 (25) 5 (31) 45 (5–196)
p
0.001
0.09
0.01
Numbers in parentheses of percentages, except for age, median maximal tumor diameter, and median follow-up, for which they are the range.
RT technique Ten patients with adenocarcinoma and all patients with epidermoid histologic features were treated with a standardized RT technique. The initial volume included the primary disease and inguinal nodes, as well as the pararectal and lower external and internal iliac lymph nodes. Equally weighted opposed AP fields were used to a midplane dose of 30.6 Gy using 18-MV photons in node-negative patients. Node-positive patients were sometimes treated with 6-MV photons in the anterior field to ensure adequate coverage of the involved adenopathy. The superior border of the field was the bottom of the sacroiliac joints, and the inferior border was 3 cm below the primary lesion. The lateral border was placed at the lateral aspect of the acetabulum bilaterally. After the initial 30.6 Gy was delivered, patients underwent repeated simulation in the prone position using an open tabletop (“belly board”) device. A three-field technique was used with beams weighted 2:1:1 for the posterior, right lateral, and left lateral fields; 45° wedges were used in the lateral fields. An additional 19.8 Gy was delivered using this technique and was prescribed to the 95% isodose line. The superior and inferior borders were the same as the initial field. The lateral borders of the posterior field were placed 1.5 cm outside the true bony pelvis. The lateral radiation fields included the sacrum and coccyx posteriorly, and the anterior border was placed 1–2 cm posterior to the pubis. If the inguinal nodes were clinically involved at presentation, they were given a boost dose using electrons to a dose of 18 –20 Gy so that gross inguinal disease received a total dose of 55 Gy. The energy of electrons used was based on the depth to which the dose was prescribed. The final boost to the primary disease was delivered using the
same three-field belly board technique, as described above, to a dose of 4.6 Gy in two fractions. The volume treated was gross primary disease with 3-cm margins. Five patients with adenocarcinoma were treated with opposed AP fields, and the patient who received preoperative therapy was treated with a three-field belly board technique, as described above. Study end points and statistical analysis The end points for this retrospective study were the rates of local disease progression, freedom from distant disease, disease-free survival, and overall survival. Local progression was defined as persistent or recurrent tumor in the pelvis. The data were compared using the Kaplan-Meier product limit survival method (GraphPad Prism, version 3.00, for Windows, GraphPad Software, San Diego, CA). For comparison of frequencies, as well as actuarial data, between the study and comparison groups, p ⫽ 0.05 was considered statistically significant. The p values in Table 1 were determined using Fisher’s exact test. RESULTS Local progression The 5-year actuarial local control rate in patients with adenocarcinoma was 46% compared with 82% for patients with epidermoid carcinoma (p ⫽ 0.004, Fig. 1). Of the 16 patients with adenocarcinoma, 7 developed local progression or recurrence compared with 16 (17%) of the 92 patients with epidermoid carcinoma. The median time to local progression or recurrence was 20 months in patients with adenocarcinoma; with epidermoid carcinoma, it had not been reached at last follow-up. Of the 7 adenocarcinoma patients with local failure, 6 received concurrent chemora-
Table 2. Treatment and outcome data for patients with adenocarcinoma Maximal tumor diameter (cm)
N stage
Pt. No.
Age (y)
Gender
Grade
T stage
Treatment
Initial Surgery
RT (Gy)
1 2 3 4 5 6 7 8 9
68 71 51 66 63 52 70 42 52
F M M F M F F F M
3 3 3 2 3 x 2 2 2
2 2 3 2 2 2 2 4 Rec.
2.5 2.0 5.0 4.0 Unk 3.0 2.0 6.0 3.7
2 2 3 0 3 0 0 0 0
CRT CRT CRT CRT CRT CRT CRT CRT CRT
/ / / / / / / / LE
59.4 60.4 55 55 55 55 55 55 55
10 11 12 13 14
62 61 56 53 59
M M M M M
3 1 X 3 X
2 X X X 2
3.0 Unk Unk Unk 4.3
3 0 0 0 0
Post-LE Post-LE Post-LE Post-LE Post-LE
LE LE LE LE LE
60 54 63.2 70 65
15 16
57 49
F M
X 1
4 X
14.0 8.3
0 0
Chemotherapy dose (mg/m2)
Salvage
Local failure (mo)
Distant metastases (mo)
Alive
Follow-up (mo)
300 300 300 225 250 300 300 300 250
/ / / APR / APR APR APR APR
/ / 3.5 11.9 / 13.8 20.1 7.1 7.1
3.7 3.3 3.5 / 9.0 / 20.1 / /
No No No Yes Yes Yes Yes Yes Yes
16.3 5.2 10.1 34.1 20.7 25.0 28.2 60.9 60.8
/ / / / /
/ FAC 5-FU/MTX / /
16.9 / / / /
25.5 36.6 52.5 / /
No No No Yes Yes
32.6 55.4 62.8 153.0 196.4
250 300
/ /
/ /
/ 34.3
Yes Yes
55.9 58.4
Definitive chemoradiation
672
RT after excisional biopsy RT RT RT RT RT
Adjuvant chemoradiation Pre-APR CRT Post-APR CRT
APR APR
55 53
Abbreviations: Pt. No. ⫽ patient number; RT ⫽ radiotherapy; F ⫽ female; M ⫽ male; X ⫽ not specified; CRT ⫽ chemoradiation; Rec. ⫽ recurrent; Unk ⫽ unknown; LE ⫽ local excision; FAC ⫽ 5-fluorouracil, adriamycin, cyclophosphamide; / ⫽ none; 5-FU ⫽ 5-fluorouracil; MTX ⫽ methotrexate; APR ⫽ abdominoperineal resection.
Chemoradiation for adenocarcinoma of the anus
● M. PAPAGIKOS et al.
673
Table 3. Treatment sequence for patients with adenocarcinoma Pt. No.
Treatment sequence
Definitive chemoradiation 1 2 3 4 5 6 7 8 9
5-FU 300 mg/m2 and 59.4 Gy 5-FU 300 mg/m2 and 60.4 Gy 5-FU 300 mg/m2 and 55 Gy 5-FU 225 mg/m2 and 55 Gy 3 salvage 5-FU 250 mg/m2 and 55 Gy 5-FU 300 mg/m2 and 55 Gy 3 salvage 5-FU 300 mg/m2 and 55 Gy 3 salvage 5-FU 300 mg/m2 and 55 Gy 3 salvage LE 3 salvage 5-FU 250 mg/m2 and 55
RT alone after excisional biopsy
3 3 3 3 3
RT RT RT RT RT
APR APR APR Gy 3 salvage APR
10 11 12 13 14
LE LE LE LE LE
15 16
Preop 5-FU 250 mg/m2 and 55 Gy 3 APR APR 3 postop 5-FU 300 mg/m2 and 53 Gy
Adjuvant chemoradiation
postop postop postop postop postop
APR
60 Gy 54 Gy 3 salvage FAC 63.2 Gy 3 salvage 5-FU/MTX 70 Gy 65 Gy
Abbreviations as in Table 2.
diation and 1 was treated with RT alone as primary treatment. The median RT dose for patients with adenocarcinoma was 55 Gy. Local control as a function of RT dose ⬎55 Gy and ⱕ55 Gy was analyzed. Local control was achieved in 4 of 10 patients who received a median dose of ⱕ55 Gy compared with 5 of 6 patients receiving ⬎55 Gy (5-year actuarial local control rate 56% vs. 75%, p ⫽ 0.09, Fig. 2). The median dose for patients whose disease recurred locally was 55 Gy compared with 59.4 Gy in those patients in whom local control was achieved.
control of local disease at a median of 11.2 months (range 8 –54) after surgery. The patient who did not undergo salvage APR was diagnosed with liver metastases synchronously with the recurrent primary lesion and died 7 months later. Thirteen of the 16 local treatment failures in the epidermoid group occurred without distant metastases and almost all (12 of 13 patients) were salvaged with APR.
Surgical salvage of local disease Of 6 patients whose tumor recurred locally after chemoradiation, 5 underwent salvage APR at a median of 7 months (range 4 –20) after RT, and all remained alive with
Nodal control Five patients with adenocarcinoma presented with N2 or N3 nodal disease. All 5 initially node-positive patients had Grade 3 tumors. Nodal disease was controlled in 3 of the 5 patients. Both patients whose nodal disease was not controlled received 50.4 Gy to the involved nodes. In 2 of the 11 patients without clinical adenopathy at diagnosis, nodal
Fig. 1. Local control in patients with epidermoid cancer (circles) or adenocarcinoma (triangles) of the anal canal treated with chemoradiation alone.
Fig. 2. Local control among patients with adenocarcinoma of the anal canal according to dose of RT received (⬎55 Gy [triangles] or ⱕ55 Gy [circles]).
674
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Table 4. Nodal status Pt. No. 1 2 3 5 10 Elective 4 8 12 15
Nodal disease at presentation
Inguinal RT dose (Gy)
Nodal control
Yes Yes Yes Yes Yes
50.3 59.4 50.3 50.3 67.5
Yes Yes No No Yes
No No No No
30.6 30.6 30.6 30.6
Yes Yes No Yes
Abbreviations as in Table 2.
disease developed at 18 and 32 months after treatment. One received no elective groin RT and the other was 1 of 4 patients who received 30.6 Gy of elective inguinal RT (Table 4). Twenty-eight patients with epidermoid carcinoma had nodal disease at diagnosis. Nodal disease was controlled in 21 (75%). All 64 epidermoid patients without clinical evidence of groin adenopathy at diagnosis received elective inguinal irradiation (30.6 Gy). Of these, none developed inguinal nodal recurrence. Metastatic disease Nine patients (56%) with adenocarcinoma developed distant metastatic disease a median of 37 months after diagnosis compared with 8 (9%) of 92 patients with epidermoid carcinoma (median time not reached). The 5-year distant disease-free survival rate was 34% for patients with adenocarcinoma and 90% for those with epidermoid carcinoma (p ⬍0.001, Fig. 3). Metastatic disease was the first site of recurrence in 3 patients with adenocarcinoma, occurred synchronously with local failure in 3 patients, and occurred after local failure in 3 patients with adenocarcinoma. The first site of metastatic disease was the liver in 4, lung in 3,
Fig. 4. Disease-free survival in patients with epidermoid cancer (circles) or adenocarcinoma (triangles) of the anal canal treated with chemoradiation alone.
soft tissue in 1, and brain in 1 patient. Distant metastases occurred ⬎34 months after therapy in 3 patients. In patients with adenocarcinoma, the 5-year actuarial survival free of distant disease rate was 17% for patients with Grade 3 tumors vs. 61% for patients with lower grade tumors (p ⫽ 0.02). Of the 8 patients with epidermoid histologic features in whom distant metastases developed, distant disease was the site of first recurrence in 4 and the successor to local relapse in 4. All 5 patients with adenocarcinoma who presented with nodal disease eventually developed distant metastases. In the epidermoid group, 8 patients developed distant metastases; 5 of those 8 initially had node-positive disease. In 1 patient whose nodes were initially negative, distant metastases developed after inguinal metastases. In 2 patients, distant disease was the first site of recurrence. Disease-free survival The median disease-free survival was 15 months in patients with adenocarcinoma and had not been reached in the epidermoid group. The 5-year actuarial disease-free survival rate was 19% and 77% for adenocarcinoma and epidermoid carcinoma, respectively (p ⬍0.0001, Fig. 4). Only 3 patients with adenocarcinoma were alive and disease free at 56, 153, and 196 months. Overall survival Ten patients with adenocarcinoma (63%) and 80 patients with epidermoid carcinoma (87%) were alive at the end of the follow-up period. The median overall survival for patients with adenocarcinoma was 63 months and had not been reached for patients with epidermoid cancer. The 5-year actuarial overall survival rate for patients with adenocarcinoma was 64% vs. 85% for those with epidermoid cancer (p ⫽ 0.017, Fig. 5). DISCUSSION
Fig. 3. Distant disease-free survival in patients with epidermoid cancer (circles) or adenocarcinoma (triangles) of the anal canal treated with chemoradiation alone.
Adenocarcinoma of the anus does not respond as well as epidermoid carcinoma to our current standard chemoradia-
Chemoradiation for adenocarcinoma of the anus
Fig. 5. Overall survival in patients with epidermoid or adenocarcinoma of the anal canal treated with chemoradiation alone.
tion regimen for epidermoid carcinoma. Treatment with definitive chemoradiation that has been very successful in treating epidermoid tumors has resulted in poor local disease control rates (less than one-half), as well as high distant recurrence rates (approximately two-thirds) in patients with adenocarcinoma (Table 5). Anal canal carcinomas represent approximately 1% of all GI tract tumors. Epidermoid (cloacogenic/basaloid and squamous cell carcinomas) represent most anal cancers. Primary adenocarcinoma of anal ductal origin has been reported in most studies to represent approximately 5–10% of all cases of anal cancer (16 –18). A recent survey of the American Society of Colon and Rectal Surgery was conducted in which only 52 cases were found, with the vast majority of surgeons reporting they had never treated an anal adenocarcinoma (19). In contrast, the NCDB Report on Carcinoma of the Anus indicated that 16.5% (212 of 1289) of anal cancer cases were primary adenocarcinoma (6). It is likely, however, that a significant number of these cases were actually low rectal cancers extending to the anal canal. Because of its rarity, information in the literature regarding the natural history, treatment options, prognosis, and outcomes is limited, with most reports consisting of case reports and retrospective studies of small cohorts.
● M. PAPAGIKOS et al.
675
It has been reported that patients with adenocarcinoma of the anal canal present with more advanced disease, have a higher rate of distant metastases, and have poorer overall survival (1, 6, 20). Data from the NCDB revealed that at the time of presentation, 9.8% of patients with anal adenocarcinoma had Stage IV disease compared with 5% for those with epidermoid carcinoma. Also more distant recurrences occurred in the adenocarcinoma patients, 28.1% vs. 11.8%. The 5-year survival rate for patients with adenocarcinoma was lower at every stage than for patients with epidermoid carcinoma, with the greatest difference seen in the 5-year survival rate of patients with Stage IV disease—13.5% for patients with adenocarcinoma and 29.0% for those with epidermoid carcinoma. All 5 patients in the present study with adenocarcinoma of the anal canal who had initial nodal disease at presentation developed distant metastases. Basik et al. (1) reported a similar finding in their study in which distant metastases developed in 4 of 10 patients with adenocarcinoma of the anus. All 4 patients developed inguinal metastasis preceding distant disease. They postulated that inguinal metastasis is a sign of disseminated disease (1). The optimal initial therapy for patients with adenocarcinoma of the anus is not known. In the present study, the overall local control rate with initial chemoradiation was unsatisfactory. Although surgical salvage was successful in 5 of 6 patients with pelvic progression, the failure to control the initial pelvic disease could have contributed to the high distant failure rate. On the basis of NCDB data, surgery has been the most frequently used modality for patients with adenocarcinoma, with 36.8% of patients undergoing surgery alone. Chemoradiotherapy without surgery was used in only 9.0% of cases. In comparison, most patients (47.4%) with epidermoid histologic features were treated with definitive chemoradiation (6). The data in the literature are very limited regarding the use of chemoradiation as primary therapy for adenocarcinoma of the anal canal. Although definitive chemoradiation has been established as the standard treatment for epidermoid carcinoma of the anal canal, it is unclear whether this approach is adequate in patients with adenocarcinoma. Tarazi and Nelson (20) reported one of the first series of
Table 5. Local progression, distant metastases, disease-free survival, and overall survival results Adenocarcinoma (n ⫽ 16)
Event
n (%)
Local recurrence 7 (44) Distant metastases 9 (56) Death 6 (38) Any 13 (81)
Epidermoid (n ⫽ 92)
Median survival free of (mo)
5-year actuarial survival free of (%)
n (%)
5-year actuarial survival free of (%)
95% CI
95% CI
p
20
46
18–74
16 (17)
82
74–90
0.004
37 63 15
34 64 19
8–60 38–90 0–36
8 (9) 12 (13) 21 (23)
90 85 77
83–97 75–95 68–85
⬍0.001 0.017 ⬍0.0001
* Median values for epidermoid population were not met at the end of follow-up. Abbreviation: CI ⫽ confidence interval.
676
I. J. Radiation Oncology
● Biology ● Physics
patients treated with definitive chemoradiation for adenocarcinoma as part of a non–peer review publication. Nine patients were treated with 30 – 45 Gy at standard fractionations with concomitant 5-FU and mitomycin-C before local excision. With a relatively short follow-up (range 2– 4 years), 1 of 7 assessable patients was reported to have locally recurrent disease and salvage treatment with APR was successful (20). In 3 unresectable patients treated with chemoradiation, reported by Anthony et al. (21), none had a complete response to chemoradiation and 2 of 3 died of disease. In contrast to these experiences, Joon et al. (3) reported the results of treating 6 patients who had localized T1 or T2, N0 adenocarcinoma with 50.4 Gy with or without 5-FU and mitomycin-C. These patients were selected for definitive treatment and 7 additional patients with locally advanced disease were considered incurable and treated with palliative therapy. As in the present study, the patients were considered to have adenocarcinoma of anal canal origin rather than extension of rectal cancer to the anal canal. Five patients underwent prereferral biopsy, and 1 patient underwent APR. After a median follow-up of 6.6 years, none of the 6 definitively treated patients had a local or distal recurrence and 5 patients remained alive. A response to that paper was published as a letter to the editor. Investigators from France summarized their experience with definitive chemoradiation in 22 patients with localized disease who received RT alone (50 Gy followed by a brachytherapy boost of 20 –25 Gy) who had adenocarcinoma of the anal canal and were either medically inoperable or refused APR. On the basis of a 3-year disease-free survival rate of 31%, the authors concluded that definitive RT is inadequate and APR should be the standard treatment (5). Even though no comment was made regarding the pattern of failure in that brief report and no evidence was presented that the use of APR would result in a better outcome, this seems to be a reasonable recommendation because of the lack of success with RT in controlling local disease. Historically, APR has been the standard initial approach in patients presenting with resectable adenocarcinoma of the anal canal. In the American Society of Colon and Rectal Surgery survey report by Abel et al. (19), 40 (77%) of 52 patients with adenocarcinoma underwent initial APR. Approximately one-half (52%) received adjuvant RT with or without chemotherapy. Thirty-two patients (62%) remained alive without disease at 3 years. In a retrospective analysis of 192 patients with cancer of the anal canal treated at the University of Minnesota, Klas et al. (22) described their experience with 36 patients with adenocarcinoma. The most common tumor stage was T2 (47%). Surgery alone was the primary treatment for 22 patients (61%). Surgical treatment included APR (n ⫽ 6) and local excision (n ⫽ 16). Surgery with adjuvant chemoradiotherapy was the treatment for the remaining 14 patients (39%). They reported an overall 5-year survival rate of 63% and a recurrence rate of 21%, with insufficient numbers to
Volume 55, Number 3, 2003
perform a separate analysis on the basis of treatment type. They proposed that the high survival may have been a function of small primary tumors (78% were ⬍5 cm in the greatest dimension) and that chemoradiotherapy should be used preoperatively or postoperatively for larger tumors (⬎5 cm). The results of these two surgical series are comparable to series reporting results treating rectal cancer. Because the proportion of patients with adenocarcinoma relative to epidermoid carcinoma in these studies is higher than what one would expect, it is probable that these studies included as anal cancer some rectal cancers with extension to the anal canal. Nevertheless, the results do compare favorably with our experience and support the use of initial APR in patients with adenocarcinoma, regardless of the TNM stage. In contrast, smaller, single-institution surgical series have described poor results for patients with localized adenocarcinoma treated with initial surgery. Jensen et al. (2) reported that 21 patients with advanced locoregional or metastatic adenocarcinoma treated surgically without RT during a 40-year period had a 5-year overall survival rate of 4.8%. Basik et al. (1) described 10 patients with adenocarcinoma, 7 of whom underwent APR and none of whom received RT; no patients had distant metastases and only 1 patient had nodal disease at the time of treatment. Tumor diameters ranged from 1.5 to 10 cm. Although TNM stages were not reported, the authors believed that the patients in their series benefited from a relatively early diagnosis. The overall median survival was 29 months. Five patients died of the disease at a median time of 28 months after surgery. Two patients had recurrences in the perineum, five in the inguinal nodes, and five at distant sites. Of the 7 patients who underwent APR, 3 (43%) developed distant metastases (1). This pattern of failure certainly supports the use of adjuvant chemoradiation if APR is to be the primary local strategy in this population. The organ-preserving strategy that has become the standard of care in patients with epidermoid carcinoma has been successful in part because local treatment failures have been salvaged effectively with APR. Evidence for that comes from three large randomized trials that have been conducted. Increased local failure has not translated into worse overall survival in these studies, because persistent local and recurrent disease was salvaged effectively (11, 13, 14). However, surgical salvage is ultimately only successful if local recurrence is not simultaneous with distant metastasis. In our experience, for both epidermoid and adenocarcinoma of the anal canal, local control was achieved after salvage surgery, but the distant failure rate overall was much higher in patients with adenocarcinoma. This leads to the question of whether distant recurrence would have been prevented had local disease been controlled. The question cannot be answered, but initial local control should be optimized to minimize the potential for distant spread. Our current policy is to use preoperative chemoradiation followed by APR, particularly among patients who present without nodal disease at presentation. Patients with inguinal node involve-
Chemoradiation for adenocarcinoma of the anus
ment at presentation have a particularly poor prognosis because of extremely high rates of distant recurrence not only in this disease, but also in adenocarcinoma of the rectum (23, 24). The priority in this group of patients should be initial chemoradiation (with doses ⬎55 Gy) and additional systemic chemotherapy, with surgical resection to be used selectively or for salvage of isolated local recurrences. Patients with adenocarcinoma in the present series had an alarmingly high rate of distant metastasis. It is not possible to assess the potential value of systemic chemotherapy in this or any other series, because very few patients have received adjuvant chemotherapy or even a consistently applied regimen for metastatic disease. However, the high distant metastasis rate should somehow be addressed. The administration of 5-FU and leucovorin has been shown to improve survival in colorectal cancer (25, 26). In the metastatic disease setting, randomized trials have demonstrated a benefit from the addition of either irinotecan (27, 28) or oxaliplatin (29, 30) to 5-FU and leucovorin. Until data emerge to the contrary, 5-FU and leucovorin for four to six cycles should be offered in the adjuvant setting for anal adenocarcinoma. Given the demonstrated improvement in survival duration with the addition of irinotecan (27, 28) and relapse-free survival with the addition of oxaliplatin (29, 30) to 5-FU and leucovorin in patients with metastatic colorectal cancer, it would be reasonable to include one of these
● M. PAPAGIKOS et al.
677
agents as well in the adjuvant setting in appropriately selected patients, provided a survival benefit is demonstrated in the adjuvant setting in colorectal cancer (trials are ongoing). Although the role of adjuvant chemotherapy for anal adenocarcinoma has not been investigated because of the rarity of the disease and is currently undefined, it should be recommended based on the pattern of failure. CONCLUSION Even though it is possible that disease in a subset of patients with adenocarcinoma of the anal canal can be controlled with chemoradiation, the preponderance of evidence suggests that the pelvic disease control rates are poor with definitive chemoradiation. Because we consider the optimization of pelvic control in this disease a priority, our current policy in patients without inguinal metastases at presentation is combined modality therapy with preoperative chemoradiation (45 Gy with concurrent protracted venous infusion 5-FU) followed by APR and consideration of 5-FU– based adjuvant chemotherapy. Whether the poor prognosis of these patients can be altered with the use of adjuvant chemotherapy is an open question, but our plan does address the pattern of failure. In patients with inguinal metastases at presentation, we recommend that definitive chemoradiation be given first, reserving surgery for selected cases or for salvage of isolated local recurrences.
REFERENCES 1. Basik M, Rodriguez-Bigas MA, Penetrante R, et al. Prognosis and recurrence patterns of anal adenocarcinoma. Am J Surg 1995;169:233–237. 2. Jensen SL, Shokouh-Amiri MH, Hagen K, et al. Adenocarcinoma of the anal ducts: A series of 21 cases. Dis Colon Rectum 1988;31:268 –272. 3. Joon DL, Chao MW, Ngan SY, et al. Primary adenocarcinoma of the anus: A retrospective analysis. Int J Radiat Oncol Biol Phys 1999;45:1199 –1205. 4. Behan WM, Burnett RA. Adenocarcinoma of the anal glands. J Clin Pathol 1996;49:1009 –1011. 5. Wolff P, Peiffert D. In regard to Joon et al. IJROBP 1999;45: 1199 –1205. Int J Radiat Oncol Biol Phys 2001;49:1517. 6. Myerson RJ, Karnell LH, Menck HR. The National Cancer Data Base report on carcinoma of the anus. Cancer 1997;80: 805– 815. 7. Boman BM, Moertel CG, O’Connell MJ, et al. Carcinoma of the anal canal: A clinical and pathologic study of 188 cases. Cancer 1984;54:114 –125. 8. Greenall MJ, Quan SH, DeCosse JJ. Epidermoid cancer of the anus. Br J Surg 1985;72(Suppl.):S97–S103. 9. Greenall MJ, Quan SH, Stearns MW, et al. Epidermoid cancer of the anal margin: Pathologic features, treatment, and clinical results. Am J Surg 1985;149:95–101. 10. Greenall MJ, Quan SH, Urmacher C, et al. Treatment of epidermoid carcinoma of the anal canal. Surg Gynecol Obstet 1985;161:509 –517. 11. Flam M, John M, Pajak TF, et al. Role of mitomycin in combination with fluorouracil and radiotherapy and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: Results of a phase III
12.
13.
14.
15.
16. 17. 18.
19.
randomized intergroup study. J Clin Oncol 1996;14:2527– 2539. Nigro ND, Vaitkevicius VK, Considine B Jr. Combined therapy for cancer of the anal canal: A preliminary report—1974. Dis Colon Rectum 1993;36:709 –711. Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: Results of a phase III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups. J Clin Oncol 1997;15:2040 –2049. UKCCCR Anal Cancer Trial Working Party UK Co-ordinating Committee on Cancer Research. Epidermoid anal cancer: Results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. Lancet 1996;348:1049 –1054. Bosset JF, Pavy JJ, Roelofsen F, et al., for the EORTC Radiotherapy and Gastrointestinal Cooperative Groups. Combined radiotherapy and chemotherapy for anal cancer. Lancet 1997;349:205–206. Cabrera A, Tsukada Y, Pickren JW. Adenocarcinomas of the anal canal and peri-anal tissues. Ann Surg 1966;164:152–156. Merlini M, Eckert P. Malignant tumors of the anus: A study of 106 cases. Am J Surg 1985;150:370 –372. Welch JP, Malt RA. Appraisal of the treatment of carcinoma of the anus and anal canal. Surg Gynecol Obstet 1977;145: 837– 841. Abel ME, Chiu YS, Russell TR, et al. Adenocarcinoma of the anal glands: Results of a survey. Dis Colon Rectum 1993;36: 383–387.
678
I. J. Radiation Oncology
● Biology ● Physics
20. Tarazi R, Nelson RL. Anal adenocarcinoma: A comprehensive review. Semin Surg Oncol 1994;10:235–240. 21. Anthony T, Simmang C, Lee EL, et al. Perianal mucinous adenocarcinoma. J Surg Oncol 1997;64:218 –221. 22. Klas JV, Rothenberger DA, Wong WD, et al. Malignant tumors of the anal canal: The spectrum of disease, treatment, and outcomes. Cancer 1999;85:1686 –1693. 23. Mesko TW, Rodriguez-Bigas MA, Petrelli NJ. Inguinal lymph node metastases from adenocarcinoma of the rectum. Am J Surg 1994;168:285–287. 24. Graham RA, Hohn DC. Management of inguinal lymph node metastases from adenocarcinoma of the rectum. Dis Colon Rectum 1990;33:212–216. 25. Wolmark N, Rockette H, Mamounas E, et al. Clinical trial to assess the relative efficacy of fluorouracil and leucovorin, fluorouracil and levamisole, and fluorouracil, leucovorin, and levamisole in patients with Dukes’ B and C carcinoma of the colon: Results from National Surgical Adjuvant Breast and Bowel Project C-04. J Clin Oncol 1999;17:3553–3559.
Volume 55, Number 3, 2003
26. O’Connell MJ, Mailliard JA, Kahn MJ, et al. Controlled trial of fluorouracil and low-dose leucovorin given for 6 months as postoperative adjuvant therapy for colon cancer. J Clin Oncol 1997;15:246 –250. 27. Saltz LB, Cox JV, Blanke C, et al., for the Irinotecan Study Group. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 2000;343:905–914. 28. Douillard JY, Cunningham D, Roth AD, et al. Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: A multicentre randomised trial. Lancet 2000;355:1041–1047. 29. de Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000;18:2938 – 2947. 30. Giacchetti S, Perpoint B, Zidani R, et al. Phase III multicenter randomized trial of oxaliplatin added to chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2000;18:136 –147.
doi:10.1016/S0360-3016(02)04118-4
CLINICAL INVESTIGATION
Anal Canal
CHEMORADIATION FOR ADENOCARCINOMA OF THE ANUS MICHAEL PAPAGIKOS, M.D.,* CHRISTOPHER H. CRANE, M.D.,† JOHN SKIBBER, M.D.,‡ NORA A. JANJAN, M.D.,† BARRY FEIG, M.D.,‡ MIGUEL A. RODRIGUEZ-BIGAS, M.D.,‡ ARTHUR HUNG, M.D.,† ROBERT A. WOLFF, M.D.,§ MARC DELCLOS, M.D.,† EDWARD LIN, M.D.,§ KAREN CLEARY, M.D.㛳
AND
*Department of Radiation Oncology, Wake Forest University School of Medicine, Winston-Salem, NC; Departments of †Radiation Oncology, ‡Surgery, §Medical Oncology, and 㛳Pathology, The University of Texas M. D. Anderson Cancer Center, Houston, TX Purpose: To assess the efficacy and limitations of definitive chemoradiation for adenocarcinoma of the anal canal and to propose a treatment strategy that addresses the limitations of treatment. Methods and Materials: Between 1976 and 1998, 16 patients with localized adenocarcinoma of the anal canal were treated with radiotherapy with or without chemotherapy with curative intent. Available histologic slides were reviewed for evidence of primary adenocarcinoma of anal duct origin. The treatment results for these patients were compared with those of a group of patients with epidermoid histologic features who were all treated with definitive chemoradiation (55 Gy with concurrent 5-fluorouracil and cisplatin, n ⴝ 92) between 1989 and 1998. The hospital records were reviewed for all patients. Patients with epidermoid carcinoma presented with more advanced primary tumors (42% vs. 19% Stage T3 or greater). All adenocarcinoma patients were treated with radiotherapy (median dose 55 Gy): 11 received concurrent 5-fluorouracil– based chemotherapy and 5 received radiotherapy alone. The initial surgical procedures included abdominoperineal resection, excisional biopsies (n ⴝ 5), and local excision (n ⴝ 1). Abdominoperineal resection was performed as salvage therapy after local recurrence in 5 patients. The Kaplan-Meier method was used to calculate 5-year actuarial pelvic control, distant disease control, disease-free survival, and overall survival. The median follow-up was 45 months (range 5–196) for patients with adenocarcinoma and 44 months (range 9 –115) for patients with epidermoid histologic features. Results: Both local and distant recurrence rates were significantly greater in the adenocarcinoma patients. Of 16 patients with adenocarcinoma, 7 (5-year actuarial rate 54%) had recurrence at the primary site compared with 16 (5-year actuarial rate 18%) of 92 patients with epidermoid histologic features (p ⴝ 0.004). Distant disease developed in more patients with adenocarcinoma (5-year actuarial rate 66%) than in patients with epidermoid carcinoma (5-year actuarial rate 10%, p <0.001). The 5-year actuarial disease-free survival and overall survival rate for adenocarcinoma patients was 19% and 64%, respectively, compared with 77% (p <0.0001) and 85% (p ⴝ 0.017) for those with epidermoid carcinoma. Conclusion: Patients with localized adenocarcinoma of the anus treated with definitive chemoradiation had high rates of pelvic failure and distant metastasis compared with comparably staged patients with epidermoid histologic features treated similarly. On the basis of these limitations, we recommend preoperative chemoradiation followed by abdominoperineal resection to maximize pelvic disease control and consideration of adjuvant chemotherapy to address the problem of micrometastatic disease. © 2003 Elsevier Science Inc. Anal cancer, Chemoradiation, Adenocarcinoma, Radiotherapy.
INTRODUCTION Adenocarcinoma of the anus is much rarer than its counterpart, epidermoid carcinoma of the anus, and the evidence in the literature, albeit limited, suggests it has a more aggressive natural history. Because reports about adenocarcinoma of the anal canal are limited to small retrospective series and case reports (1–5), the optimal therapeutic approach has not
been established and the pattern of disease recurrence after definitive chemoradiation has not been well characterized. In the 1970s and 1980s, abdominoperineal resection (APR) was the mainstay of the treatment of epidermoid carcinoma of the anal canal. By the 1990s, definitive chemoradiation had largely replaced APR as the standard initial management following the recognition that the rate of pelvic disease control with surgery alone was unsatisfactory (7–10),
Reprint requests to: Christopher H. Crane, M.D., Department of Radiation Oncology, The University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 97, Houston, TX 77030. Tel: 713-792-3400; Fax: 713-794-5573; E-mail: [email protected]
Supported in part by PO1 CA-06294, T32CA77050, and P30CA16672 awarded by the National Cancer Institute, U.S. Department of Health and Human Services. Received Jun 11, 2002, and in revised form Aug 28, 2002. Accepted for publication Sep 5, 2002. 669
670
I. J. Radiation Oncology
● Biology ● Physics
that chemoradiation was effective against the disease (11– 14), and that recurrences after chemoradiation could be effectively salvaged with APR (14, 15). Historically, the primary treatment of adenocarcinoma of the anus has also been surgical, most commonly APR, requiring permanent colostomy. A population-based study from the National Cancer Data Base (NCDB) reported that approximately 75% of patients diagnosed with adenocarcinoma of the anal canal during a 5-year period underwent surgery (6). Unlike the extensive literature on epidermoid carcinoma, the rarity of adenocarcinoma of the anal canal has precluded any meaningful assessment of the efficacy of definitive chemoradiation for it. Extrapolating from the experience with epidermoid carcinoma, we have advocated a similar approach for adenocarcinoma of the anal canal with the hope that similar local control to that achieved for epidermoid disease would be possible. The purpose of this report was to compare the outcomes and pattern of failure after definitive chemoradiation for adenocarcinoma of the anal canal with those of epidermoid carcinoma of the anal canal after similar treatment. METHODS AND MATERIALS Adenocarcinoma of the anal canal The University of Texas M. D. Anderson Cancer Center Institutional Review Board reviewed and approved the concept for this study. The medical records of patients with a diagnosis of localized adenocarcinoma of the anal canal who were treated at The University of Texas M. D. Anderson Cancer Center between 1976 and 1998 were examined for inclusion in this retrospective study. The hospital records were evaluated for anatomic and histologic criteria consistent with primary anal duct origin. Only the records of patients whose tumors involved the anal canal (defined as 0 – 4 cm from the anal verge) were considered. If ⬎50% of the tumor length was proximal to this region, patients were considered to have adenocarcinoma of the rectum and therefore were excluded. Sixteen patients met the anatomic criteria for inclusion. In an attempt to eliminate rectal cancers from the group further, the histologic appearance of the hemotoxylin-eosin–stained slides was reviewed for evidence of clear rectal mucosa origin vs. clear anal ductal origin. Six cases had sufficient histologic material to support the diagnosis of primary anal duct origin. In 10 patients, the slides were either not available for review or lacked sufficient material to differentiate rectal mucosal from anal ductal origin. Of them, 2 cases had histologic findings indistinguishable from rectal adenocarcinoma, and in 8, no specimens were available for review. All 16 records were included in the study. The medical records of a homogeneously treated group of patients (n ⫽ 92) with localized epidermoid carcinoma of the anal canal who were treated with definitive chemoradiation between March 1990 and September 1998 were reviewed and served as a comparison group.
Volume 55, Number 3, 2003
Workup A surgical oncologist, radiation oncologist, and medical oncologist initially saw all patients. The initial workup consisted of a medical history, physical examination, rigid proctoscopy, routine blood work, chest X-ray, and CT of the abdomen and pelvis. Biopsy was performed by proctoscopy or colonoscopy and malignant disease confirmed by histologic examination of the specimen. Patient and tumor characteristics Demographic and staging information for all patients is listed in Table 1. Patients with adenocarcinoma were of a similar age and race and had a similar tumor size (p not significant). Six (38%) of the 16 patients with adenocarcinoma were women compared with 71 (77%) of 92 patients with epidermoid histologic features (p ⫽ 0.002). More patients with epidermoid carcinoma presented with advanced primary tumors (42% vs. 19% Stage T3 or greater). Five patients (31%) with adenocarcinoma had nodal involvement compared with 28 (30%) with epidermoid carcinoma. Patients with adenocarcinoma more often had highgrade lesions (5 [31%] of 16 vs. 6 [7%] of 92, p ⫽ 0.01). Treatment information The treatment and outcome information for adenocarcinoma patients is listed in Table 2. All patients were treated with radiotherapy (RT) as part of a definitive treatment strategy. Eleven patients received concurrent chemotherapy; in two of them, chemoradiation was an adjuvant to planned APR (one preoperatively and one postoperatively). Five patients were treated with local excision of the primary tumor followed by RT without concurrent chemotherapy. One patient was treated with definitive chemoradiation after presenting with recurrent disease after local excision. In addition to the 2 patients who underwent APR as part of their initial treatment, 5 others underwent APR as salvage therapy for locally persistent disease. The chemoradiation regimen most commonly consisted of continuous infusion 5-fluorouracil (5-FU; 250 –300 mg/m2 within 24 h Monday through Friday concurrently with 55 Gy in 30 fractions within 6 weeks). However, the treatment dose and technique varied considerably during the 22-year study interval. The median RT dose to the primary tumor volume was 55 Gy (range 53–70). The inguinal nodal regions were treated in 9 patients; 5 received a median inguinal dose of 50.4 Gy and 4 received elective inguinal irradiation to 30.6 Gy. Seven patients did not receive inguinal RT. The sequence of treatment for each patient with adenocarcinoma is summarized in Table 3. All patients with epidermoid histologic features were treated with definitive chemoradiation and received 55 Gy in 30 fractions in 6 weeks with concurrent continuous infusion 5-FU (250 mg/m2/24 h, Monday through Friday) and cisplatin (4 mg/m2/24 h, Monday through Friday). The chemotherapy was delivered through a double-lumen subclavian catheter using electromechanical pumps. Surgery (APR) was reserved for persistent or recurrent disease.
Chemoradiation for adenocarcinoma of the anus
● M. PAPAGIKOS et al.
671
Table 1. Demographic and staging information for patients with adenocarcinoma and epidermoid carcinoma
Number of patients Median age (y) Female gender Node positive (n) N1 N2 N3 Tx T1 T2 T3 T4 Recurrent Median maximal tumor diameter (cm) Grade unspecified Grade 1 Grade 2 Grade 3 Median follow-up (mo)
Epidermoid
Adenocarcinoma
92 57 (34–82) 71 (77) 28 (30) 7 14 7 4 (4.3) 6 (6.5) 43 (46.7) 27 (29.3) 12 (13) 0 (0) 5 (1.3–10) 0 (0) 1 (1) 85 (92) 6 (7) 44 (9–115)
16 58 (42–71) 6 (38) 5 (31) 0 2 3 4 (25) 0 (0) 8 (50) 1 (6.3) 2 (12.5) 1 (6.3) 3.8 (2–14) 4 (25) 3 (19) 4 (25) 5 (31) 45 (5–196)
p
0.001
0.09
0.01
Numbers in parentheses of percentages, except for age, median maximal tumor diameter, and median follow-up, for which they are the range.
RT technique Ten patients with adenocarcinoma and all patients with epidermoid histologic features were treated with a standardized RT technique. The initial volume included the primary disease and inguinal nodes, as well as the pararectal and lower external and internal iliac lymph nodes. Equally weighted opposed AP fields were used to a midplane dose of 30.6 Gy using 18-MV photons in node-negative patients. Node-positive patients were sometimes treated with 6-MV photons in the anterior field to ensure adequate coverage of the involved adenopathy. The superior border of the field was the bottom of the sacroiliac joints, and the inferior border was 3 cm below the primary lesion. The lateral border was placed at the lateral aspect of the acetabulum bilaterally. After the initial 30.6 Gy was delivered, patients underwent repeated simulation in the prone position using an open tabletop (“belly board”) device. A three-field technique was used with beams weighted 2:1:1 for the posterior, right lateral, and left lateral fields; 45° wedges were used in the lateral fields. An additional 19.8 Gy was delivered using this technique and was prescribed to the 95% isodose line. The superior and inferior borders were the same as the initial field. The lateral borders of the posterior field were placed 1.5 cm outside the true bony pelvis. The lateral radiation fields included the sacrum and coccyx posteriorly, and the anterior border was placed 1–2 cm posterior to the pubis. If the inguinal nodes were clinically involved at presentation, they were given a boost dose using electrons to a dose of 18 –20 Gy so that gross inguinal disease received a total dose of 55 Gy. The energy of electrons used was based on the depth to which the dose was prescribed. The final boost to the primary disease was delivered using the
same three-field belly board technique, as described above, to a dose of 4.6 Gy in two fractions. The volume treated was gross primary disease with 3-cm margins. Five patients with adenocarcinoma were treated with opposed AP fields, and the patient who received preoperative therapy was treated with a three-field belly board technique, as described above. Study end points and statistical analysis The end points for this retrospective study were the rates of local disease progression, freedom from distant disease, disease-free survival, and overall survival. Local progression was defined as persistent or recurrent tumor in the pelvis. The data were compared using the Kaplan-Meier product limit survival method (GraphPad Prism, version 3.00, for Windows, GraphPad Software, San Diego, CA). For comparison of frequencies, as well as actuarial data, between the study and comparison groups, p ⫽ 0.05 was considered statistically significant. The p values in Table 1 were determined using Fisher’s exact test. RESULTS Local progression The 5-year actuarial local control rate in patients with adenocarcinoma was 46% compared with 82% for patients with epidermoid carcinoma (p ⫽ 0.004, Fig. 1). Of the 16 patients with adenocarcinoma, 7 developed local progression or recurrence compared with 16 (17%) of the 92 patients with epidermoid carcinoma. The median time to local progression or recurrence was 20 months in patients with adenocarcinoma; with epidermoid carcinoma, it had not been reached at last follow-up. Of the 7 adenocarcinoma patients with local failure, 6 received concurrent chemora-
Table 2. Treatment and outcome data for patients with adenocarcinoma Maximal tumor diameter (cm)
N stage
Pt. No.
Age (y)
Gender
Grade
T stage
Treatment
Initial Surgery
RT (Gy)
1 2 3 4 5 6 7 8 9
68 71 51 66 63 52 70 42 52
F M M F M F F F M
3 3 3 2 3 x 2 2 2
2 2 3 2 2 2 2 4 Rec.
2.5 2.0 5.0 4.0 Unk 3.0 2.0 6.0 3.7
2 2 3 0 3 0 0 0 0
CRT CRT CRT CRT CRT CRT CRT CRT CRT
/ / / / / / / / LE
59.4 60.4 55 55 55 55 55 55 55
10 11 12 13 14
62 61 56 53 59
M M M M M
3 1 X 3 X
2 X X X 2
3.0 Unk Unk Unk 4.3
3 0 0 0 0
Post-LE Post-LE Post-LE Post-LE Post-LE
LE LE LE LE LE
60 54 63.2 70 65
15 16
57 49
F M
X 1
4 X
14.0 8.3
0 0
Chemotherapy dose (mg/m2)
Salvage
Local failure (mo)
Distant metastases (mo)
Alive
Follow-up (mo)
300 300 300 225 250 300 300 300 250
/ / / APR / APR APR APR APR
/ / 3.5 11.9 / 13.8 20.1 7.1 7.1
3.7 3.3 3.5 / 9.0 / 20.1 / /
No No No Yes Yes Yes Yes Yes Yes
16.3 5.2 10.1 34.1 20.7 25.0 28.2 60.9 60.8
/ / / / /
/ FAC 5-FU/MTX / /
16.9 / / / /
25.5 36.6 52.5 / /
No No No Yes Yes
32.6 55.4 62.8 153.0 196.4
250 300
/ /
/ /
/ 34.3
Yes Yes
55.9 58.4
Definitive chemoradiation
672
RT after excisional biopsy RT RT RT RT RT
Adjuvant chemoradiation Pre-APR CRT Post-APR CRT
APR APR
55 53
Abbreviations: Pt. No. ⫽ patient number; RT ⫽ radiotherapy; F ⫽ female; M ⫽ male; X ⫽ not specified; CRT ⫽ chemoradiation; Rec. ⫽ recurrent; Unk ⫽ unknown; LE ⫽ local excision; FAC ⫽ 5-fluorouracil, adriamycin, cyclophosphamide; / ⫽ none; 5-FU ⫽ 5-fluorouracil; MTX ⫽ methotrexate; APR ⫽ abdominoperineal resection.
Chemoradiation for adenocarcinoma of the anus
● M. PAPAGIKOS et al.
673
Table 3. Treatment sequence for patients with adenocarcinoma Pt. No.
Treatment sequence
Definitive chemoradiation 1 2 3 4 5 6 7 8 9
5-FU 300 mg/m2 and 59.4 Gy 5-FU 300 mg/m2 and 60.4 Gy 5-FU 300 mg/m2 and 55 Gy 5-FU 225 mg/m2 and 55 Gy 3 salvage 5-FU 250 mg/m2 and 55 Gy 5-FU 300 mg/m2 and 55 Gy 3 salvage 5-FU 300 mg/m2 and 55 Gy 3 salvage 5-FU 300 mg/m2 and 55 Gy 3 salvage LE 3 salvage 5-FU 250 mg/m2 and 55
RT alone after excisional biopsy
3 3 3 3 3
RT RT RT RT RT
APR APR APR Gy 3 salvage APR
10 11 12 13 14
LE LE LE LE LE
15 16
Preop 5-FU 250 mg/m2 and 55 Gy 3 APR APR 3 postop 5-FU 300 mg/m2 and 53 Gy
Adjuvant chemoradiation
postop postop postop postop postop
APR
60 Gy 54 Gy 3 salvage FAC 63.2 Gy 3 salvage 5-FU/MTX 70 Gy 65 Gy
Abbreviations as in Table 2.
diation and 1 was treated with RT alone as primary treatment. The median RT dose for patients with adenocarcinoma was 55 Gy. Local control as a function of RT dose ⬎55 Gy and ⱕ55 Gy was analyzed. Local control was achieved in 4 of 10 patients who received a median dose of ⱕ55 Gy compared with 5 of 6 patients receiving ⬎55 Gy (5-year actuarial local control rate 56% vs. 75%, p ⫽ 0.09, Fig. 2). The median dose for patients whose disease recurred locally was 55 Gy compared with 59.4 Gy in those patients in whom local control was achieved.
control of local disease at a median of 11.2 months (range 8 –54) after surgery. The patient who did not undergo salvage APR was diagnosed with liver metastases synchronously with the recurrent primary lesion and died 7 months later. Thirteen of the 16 local treatment failures in the epidermoid group occurred without distant metastases and almost all (12 of 13 patients) were salvaged with APR.
Surgical salvage of local disease Of 6 patients whose tumor recurred locally after chemoradiation, 5 underwent salvage APR at a median of 7 months (range 4 –20) after RT, and all remained alive with
Nodal control Five patients with adenocarcinoma presented with N2 or N3 nodal disease. All 5 initially node-positive patients had Grade 3 tumors. Nodal disease was controlled in 3 of the 5 patients. Both patients whose nodal disease was not controlled received 50.4 Gy to the involved nodes. In 2 of the 11 patients without clinical adenopathy at diagnosis, nodal
Fig. 1. Local control in patients with epidermoid cancer (circles) or adenocarcinoma (triangles) of the anal canal treated with chemoradiation alone.
Fig. 2. Local control among patients with adenocarcinoma of the anal canal according to dose of RT received (⬎55 Gy [triangles] or ⱕ55 Gy [circles]).
674
I. J. Radiation Oncology
● Biology ● Physics
Volume 55, Number 3, 2003
Table 4. Nodal status Pt. No. 1 2 3 5 10 Elective 4 8 12 15
Nodal disease at presentation
Inguinal RT dose (Gy)
Nodal control
Yes Yes Yes Yes Yes
50.3 59.4 50.3 50.3 67.5
Yes Yes No No Yes
No No No No
30.6 30.6 30.6 30.6
Yes Yes No Yes
Abbreviations as in Table 2.
disease developed at 18 and 32 months after treatment. One received no elective groin RT and the other was 1 of 4 patients who received 30.6 Gy of elective inguinal RT (Table 4). Twenty-eight patients with epidermoid carcinoma had nodal disease at diagnosis. Nodal disease was controlled in 21 (75%). All 64 epidermoid patients without clinical evidence of groin adenopathy at diagnosis received elective inguinal irradiation (30.6 Gy). Of these, none developed inguinal nodal recurrence. Metastatic disease Nine patients (56%) with adenocarcinoma developed distant metastatic disease a median of 37 months after diagnosis compared with 8 (9%) of 92 patients with epidermoid carcinoma (median time not reached). The 5-year distant disease-free survival rate was 34% for patients with adenocarcinoma and 90% for those with epidermoid carcinoma (p ⬍0.001, Fig. 3). Metastatic disease was the first site of recurrence in 3 patients with adenocarcinoma, occurred synchronously with local failure in 3 patients, and occurred after local failure in 3 patients with adenocarcinoma. The first site of metastatic disease was the liver in 4, lung in 3,
Fig. 4. Disease-free survival in patients with epidermoid cancer (circles) or adenocarcinoma (triangles) of the anal canal treated with chemoradiation alone.
soft tissue in 1, and brain in 1 patient. Distant metastases occurred ⬎34 months after therapy in 3 patients. In patients with adenocarcinoma, the 5-year actuarial survival free of distant disease rate was 17% for patients with Grade 3 tumors vs. 61% for patients with lower grade tumors (p ⫽ 0.02). Of the 8 patients with epidermoid histologic features in whom distant metastases developed, distant disease was the site of first recurrence in 4 and the successor to local relapse in 4. All 5 patients with adenocarcinoma who presented with nodal disease eventually developed distant metastases. In the epidermoid group, 8 patients developed distant metastases; 5 of those 8 initially had node-positive disease. In 1 patient whose nodes were initially negative, distant metastases developed after inguinal metastases. In 2 patients, distant disease was the first site of recurrence. Disease-free survival The median disease-free survival was 15 months in patients with adenocarcinoma and had not been reached in the epidermoid group. The 5-year actuarial disease-free survival rate was 19% and 77% for adenocarcinoma and epidermoid carcinoma, respectively (p ⬍0.0001, Fig. 4). Only 3 patients with adenocarcinoma were alive and disease free at 56, 153, and 196 months. Overall survival Ten patients with adenocarcinoma (63%) and 80 patients with epidermoid carcinoma (87%) were alive at the end of the follow-up period. The median overall survival for patients with adenocarcinoma was 63 months and had not been reached for patients with epidermoid cancer. The 5-year actuarial overall survival rate for patients with adenocarcinoma was 64% vs. 85% for those with epidermoid cancer (p ⫽ 0.017, Fig. 5). DISCUSSION
Fig. 3. Distant disease-free survival in patients with epidermoid cancer (circles) or adenocarcinoma (triangles) of the anal canal treated with chemoradiation alone.
Adenocarcinoma of the anus does not respond as well as epidermoid carcinoma to our current standard chemoradia-
Chemoradiation for adenocarcinoma of the anus
Fig. 5. Overall survival in patients with epidermoid or adenocarcinoma of the anal canal treated with chemoradiation alone.
tion regimen for epidermoid carcinoma. Treatment with definitive chemoradiation that has been very successful in treating epidermoid tumors has resulted in poor local disease control rates (less than one-half), as well as high distant recurrence rates (approximately two-thirds) in patients with adenocarcinoma (Table 5). Anal canal carcinomas represent approximately 1% of all GI tract tumors. Epidermoid (cloacogenic/basaloid and squamous cell carcinomas) represent most anal cancers. Primary adenocarcinoma of anal ductal origin has been reported in most studies to represent approximately 5–10% of all cases of anal cancer (16 –18). A recent survey of the American Society of Colon and Rectal Surgery was conducted in which only 52 cases were found, with the vast majority of surgeons reporting they had never treated an anal adenocarcinoma (19). In contrast, the NCDB Report on Carcinoma of the Anus indicated that 16.5% (212 of 1289) of anal cancer cases were primary adenocarcinoma (6). It is likely, however, that a significant number of these cases were actually low rectal cancers extending to the anal canal. Because of its rarity, information in the literature regarding the natural history, treatment options, prognosis, and outcomes is limited, with most reports consisting of case reports and retrospective studies of small cohorts.
● M. PAPAGIKOS et al.
675
It has been reported that patients with adenocarcinoma of the anal canal present with more advanced disease, have a higher rate of distant metastases, and have poorer overall survival (1, 6, 20). Data from the NCDB revealed that at the time of presentation, 9.8% of patients with anal adenocarcinoma had Stage IV disease compared with 5% for those with epidermoid carcinoma. Also more distant recurrences occurred in the adenocarcinoma patients, 28.1% vs. 11.8%. The 5-year survival rate for patients with adenocarcinoma was lower at every stage than for patients with epidermoid carcinoma, with the greatest difference seen in the 5-year survival rate of patients with Stage IV disease—13.5% for patients with adenocarcinoma and 29.0% for those with epidermoid carcinoma. All 5 patients in the present study with adenocarcinoma of the anal canal who had initial nodal disease at presentation developed distant metastases. Basik et al. (1) reported a similar finding in their study in which distant metastases developed in 4 of 10 patients with adenocarcinoma of the anus. All 4 patients developed inguinal metastasis preceding distant disease. They postulated that inguinal metastasis is a sign of disseminated disease (1). The optimal initial therapy for patients with adenocarcinoma of the anus is not known. In the present study, the overall local control rate with initial chemoradiation was unsatisfactory. Although surgical salvage was successful in 5 of 6 patients with pelvic progression, the failure to control the initial pelvic disease could have contributed to the high distant failure rate. On the basis of NCDB data, surgery has been the most frequently used modality for patients with adenocarcinoma, with 36.8% of patients undergoing surgery alone. Chemoradiotherapy without surgery was used in only 9.0% of cases. In comparison, most patients (47.4%) with epidermoid histologic features were treated with definitive chemoradiation (6). The data in the literature are very limited regarding the use of chemoradiation as primary therapy for adenocarcinoma of the anal canal. Although definitive chemoradiation has been established as the standard treatment for epidermoid carcinoma of the anal canal, it is unclear whether this approach is adequate in patients with adenocarcinoma. Tarazi and Nelson (20) reported one of the first series of
Table 5. Local progression, distant metastases, disease-free survival, and overall survival results Adenocarcinoma (n ⫽ 16)
Event
n (%)
Local recurrence 7 (44) Distant metastases 9 (56) Death 6 (38) Any 13 (81)
Epidermoid (n ⫽ 92)
Median survival free of (mo)
5-year actuarial survival free of (%)
n (%)
5-year actuarial survival free of (%)
95% CI
95% CI
p
20
46
18–74
16 (17)
82
74–90
0.004
37 63 15
34 64 19
8–60 38–90 0–36
8 (9) 12 (13) 21 (23)
90 85 77
83–97 75–95 68–85
⬍0.001 0.017 ⬍0.0001
* Median values for epidermoid population were not met at the end of follow-up. Abbreviation: CI ⫽ confidence interval.
676
I. J. Radiation Oncology
● Biology ● Physics
patients treated with definitive chemoradiation for adenocarcinoma as part of a non–peer review publication. Nine patients were treated with 30 – 45 Gy at standard fractionations with concomitant 5-FU and mitomycin-C before local excision. With a relatively short follow-up (range 2– 4 years), 1 of 7 assessable patients was reported to have locally recurrent disease and salvage treatment with APR was successful (20). In 3 unresectable patients treated with chemoradiation, reported by Anthony et al. (21), none had a complete response to chemoradiation and 2 of 3 died of disease. In contrast to these experiences, Joon et al. (3) reported the results of treating 6 patients who had localized T1 or T2, N0 adenocarcinoma with 50.4 Gy with or without 5-FU and mitomycin-C. These patients were selected for definitive treatment and 7 additional patients with locally advanced disease were considered incurable and treated with palliative therapy. As in the present study, the patients were considered to have adenocarcinoma of anal canal origin rather than extension of rectal cancer to the anal canal. Five patients underwent prereferral biopsy, and 1 patient underwent APR. After a median follow-up of 6.6 years, none of the 6 definitively treated patients had a local or distal recurrence and 5 patients remained alive. A response to that paper was published as a letter to the editor. Investigators from France summarized their experience with definitive chemoradiation in 22 patients with localized disease who received RT alone (50 Gy followed by a brachytherapy boost of 20 –25 Gy) who had adenocarcinoma of the anal canal and were either medically inoperable or refused APR. On the basis of a 3-year disease-free survival rate of 31%, the authors concluded that definitive RT is inadequate and APR should be the standard treatment (5). Even though no comment was made regarding the pattern of failure in that brief report and no evidence was presented that the use of APR would result in a better outcome, this seems to be a reasonable recommendation because of the lack of success with RT in controlling local disease. Historically, APR has been the standard initial approach in patients presenting with resectable adenocarcinoma of the anal canal. In the American Society of Colon and Rectal Surgery survey report by Abel et al. (19), 40 (77%) of 52 patients with adenocarcinoma underwent initial APR. Approximately one-half (52%) received adjuvant RT with or without chemotherapy. Thirty-two patients (62%) remained alive without disease at 3 years. In a retrospective analysis of 192 patients with cancer of the anal canal treated at the University of Minnesota, Klas et al. (22) described their experience with 36 patients with adenocarcinoma. The most common tumor stage was T2 (47%). Surgery alone was the primary treatment for 22 patients (61%). Surgical treatment included APR (n ⫽ 6) and local excision (n ⫽ 16). Surgery with adjuvant chemoradiotherapy was the treatment for the remaining 14 patients (39%). They reported an overall 5-year survival rate of 63% and a recurrence rate of 21%, with insufficient numbers to
Volume 55, Number 3, 2003
perform a separate analysis on the basis of treatment type. They proposed that the high survival may have been a function of small primary tumors (78% were ⬍5 cm in the greatest dimension) and that chemoradiotherapy should be used preoperatively or postoperatively for larger tumors (⬎5 cm). The results of these two surgical series are comparable to series reporting results treating rectal cancer. Because the proportion of patients with adenocarcinoma relative to epidermoid carcinoma in these studies is higher than what one would expect, it is probable that these studies included as anal cancer some rectal cancers with extension to the anal canal. Nevertheless, the results do compare favorably with our experience and support the use of initial APR in patients with adenocarcinoma, regardless of the TNM stage. In contrast, smaller, single-institution surgical series have described poor results for patients with localized adenocarcinoma treated with initial surgery. Jensen et al. (2) reported that 21 patients with advanced locoregional or metastatic adenocarcinoma treated surgically without RT during a 40-year period had a 5-year overall survival rate of 4.8%. Basik et al. (1) described 10 patients with adenocarcinoma, 7 of whom underwent APR and none of whom received RT; no patients had distant metastases and only 1 patient had nodal disease at the time of treatment. Tumor diameters ranged from 1.5 to 10 cm. Although TNM stages were not reported, the authors believed that the patients in their series benefited from a relatively early diagnosis. The overall median survival was 29 months. Five patients died of the disease at a median time of 28 months after surgery. Two patients had recurrences in the perineum, five in the inguinal nodes, and five at distant sites. Of the 7 patients who underwent APR, 3 (43%) developed distant metastases (1). This pattern of failure certainly supports the use of adjuvant chemoradiation if APR is to be the primary local strategy in this population. The organ-preserving strategy that has become the standard of care in patients with epidermoid carcinoma has been successful in part because local treatment failures have been salvaged effectively with APR. Evidence for that comes from three large randomized trials that have been conducted. Increased local failure has not translated into worse overall survival in these studies, because persistent local and recurrent disease was salvaged effectively (11, 13, 14). However, surgical salvage is ultimately only successful if local recurrence is not simultaneous with distant metastasis. In our experience, for both epidermoid and adenocarcinoma of the anal canal, local control was achieved after salvage surgery, but the distant failure rate overall was much higher in patients with adenocarcinoma. This leads to the question of whether distant recurrence would have been prevented had local disease been controlled. The question cannot be answered, but initial local control should be optimized to minimize the potential for distant spread. Our current policy is to use preoperative chemoradiation followed by APR, particularly among patients who present without nodal disease at presentation. Patients with inguinal node involve-
Chemoradiation for adenocarcinoma of the anus
ment at presentation have a particularly poor prognosis because of extremely high rates of distant recurrence not only in this disease, but also in adenocarcinoma of the rectum (23, 24). The priority in this group of patients should be initial chemoradiation (with doses ⬎55 Gy) and additional systemic chemotherapy, with surgical resection to be used selectively or for salvage of isolated local recurrences. Patients with adenocarcinoma in the present series had an alarmingly high rate of distant metastasis. It is not possible to assess the potential value of systemic chemotherapy in this or any other series, because very few patients have received adjuvant chemotherapy or even a consistently applied regimen for metastatic disease. However, the high distant metastasis rate should somehow be addressed. The administration of 5-FU and leucovorin has been shown to improve survival in colorectal cancer (25, 26). In the metastatic disease setting, randomized trials have demonstrated a benefit from the addition of either irinotecan (27, 28) or oxaliplatin (29, 30) to 5-FU and leucovorin. Until data emerge to the contrary, 5-FU and leucovorin for four to six cycles should be offered in the adjuvant setting for anal adenocarcinoma. Given the demonstrated improvement in survival duration with the addition of irinotecan (27, 28) and relapse-free survival with the addition of oxaliplatin (29, 30) to 5-FU and leucovorin in patients with metastatic colorectal cancer, it would be reasonable to include one of these
● M. PAPAGIKOS et al.
677
agents as well in the adjuvant setting in appropriately selected patients, provided a survival benefit is demonstrated in the adjuvant setting in colorectal cancer (trials are ongoing). Although the role of adjuvant chemotherapy for anal adenocarcinoma has not been investigated because of the rarity of the disease and is currently undefined, it should be recommended based on the pattern of failure. CONCLUSION Even though it is possible that disease in a subset of patients with adenocarcinoma of the anal canal can be controlled with chemoradiation, the preponderance of evidence suggests that the pelvic disease control rates are poor with definitive chemoradiation. Because we consider the optimization of pelvic control in this disease a priority, our current policy in patients without inguinal metastases at presentation is combined modality therapy with preoperative chemoradiation (45 Gy with concurrent protracted venous infusion 5-FU) followed by APR and consideration of 5-FU– based adjuvant chemotherapy. Whether the poor prognosis of these patients can be altered with the use of adjuvant chemotherapy is an open question, but our plan does address the pattern of failure. In patients with inguinal metastases at presentation, we recommend that definitive chemoradiation be given first, reserving surgery for selected cases or for salvage of isolated local recurrences.
REFERENCES 1. Basik M, Rodriguez-Bigas MA, Penetrante R, et al. Prognosis and recurrence patterns of anal adenocarcinoma. Am J Surg 1995;169:233–237. 2. Jensen SL, Shokouh-Amiri MH, Hagen K, et al. Adenocarcinoma of the anal ducts: A series of 21 cases. Dis Colon Rectum 1988;31:268 –272. 3. Joon DL, Chao MW, Ngan SY, et al. Primary adenocarcinoma of the anus: A retrospective analysis. Int J Radiat Oncol Biol Phys 1999;45:1199 –1205. 4. Behan WM, Burnett RA. Adenocarcinoma of the anal glands. J Clin Pathol 1996;49:1009 –1011. 5. Wolff P, Peiffert D. In regard to Joon et al. IJROBP 1999;45: 1199 –1205. Int J Radiat Oncol Biol Phys 2001;49:1517. 6. Myerson RJ, Karnell LH, Menck HR. The National Cancer Data Base report on carcinoma of the anus. Cancer 1997;80: 805– 815. 7. Boman BM, Moertel CG, O’Connell MJ, et al. Carcinoma of the anal canal: A clinical and pathologic study of 188 cases. Cancer 1984;54:114 –125. 8. Greenall MJ, Quan SH, DeCosse JJ. Epidermoid cancer of the anus. Br J Surg 1985;72(Suppl.):S97–S103. 9. Greenall MJ, Quan SH, Stearns MW, et al. Epidermoid cancer of the anal margin: Pathologic features, treatment, and clinical results. Am J Surg 1985;149:95–101. 10. Greenall MJ, Quan SH, Urmacher C, et al. Treatment of epidermoid carcinoma of the anal canal. Surg Gynecol Obstet 1985;161:509 –517. 11. Flam M, John M, Pajak TF, et al. Role of mitomycin in combination with fluorouracil and radiotherapy and of salvage chemoradiation in the definitive nonsurgical treatment of epidermoid carcinoma of the anal canal: Results of a phase III
12.
13.
14.
15.
16. 17. 18.
19.
randomized intergroup study. J Clin Oncol 1996;14:2527– 2539. Nigro ND, Vaitkevicius VK, Considine B Jr. Combined therapy for cancer of the anal canal: A preliminary report—1974. Dis Colon Rectum 1993;36:709 –711. Bartelink H, Roelofsen F, Eschwege F, et al. Concomitant radiotherapy and chemotherapy is superior to radiotherapy alone in the treatment of locally advanced anal cancer: Results of a phase III randomized trial of the European Organization for Research and Treatment of Cancer Radiotherapy and Gastrointestinal Cooperative Groups. J Clin Oncol 1997;15:2040 –2049. UKCCCR Anal Cancer Trial Working Party UK Co-ordinating Committee on Cancer Research. Epidermoid anal cancer: Results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. Lancet 1996;348:1049 –1054. Bosset JF, Pavy JJ, Roelofsen F, et al., for the EORTC Radiotherapy and Gastrointestinal Cooperative Groups. Combined radiotherapy and chemotherapy for anal cancer. Lancet 1997;349:205–206. Cabrera A, Tsukada Y, Pickren JW. Adenocarcinomas of the anal canal and peri-anal tissues. Ann Surg 1966;164:152–156. Merlini M, Eckert P. Malignant tumors of the anus: A study of 106 cases. Am J Surg 1985;150:370 –372. Welch JP, Malt RA. Appraisal of the treatment of carcinoma of the anus and anal canal. Surg Gynecol Obstet 1977;145: 837– 841. Abel ME, Chiu YS, Russell TR, et al. Adenocarcinoma of the anal glands: Results of a survey. Dis Colon Rectum 1993;36: 383–387.
678
I. J. Radiation Oncology
● Biology ● Physics
20. Tarazi R, Nelson RL. Anal adenocarcinoma: A comprehensive review. Semin Surg Oncol 1994;10:235–240. 21. Anthony T, Simmang C, Lee EL, et al. Perianal mucinous adenocarcinoma. J Surg Oncol 1997;64:218 –221. 22. Klas JV, Rothenberger DA, Wong WD, et al. Malignant tumors of the anal canal: The spectrum of disease, treatment, and outcomes. Cancer 1999;85:1686 –1693. 23. Mesko TW, Rodriguez-Bigas MA, Petrelli NJ. Inguinal lymph node metastases from adenocarcinoma of the rectum. Am J Surg 1994;168:285–287. 24. Graham RA, Hohn DC. Management of inguinal lymph node metastases from adenocarcinoma of the rectum. Dis Colon Rectum 1990;33:212–216. 25. Wolmark N, Rockette H, Mamounas E, et al. Clinical trial to assess the relative efficacy of fluorouracil and leucovorin, fluorouracil and levamisole, and fluorouracil, leucovorin, and levamisole in patients with Dukes’ B and C carcinoma of the colon: Results from National Surgical Adjuvant Breast and Bowel Project C-04. J Clin Oncol 1999;17:3553–3559.
Volume 55, Number 3, 2003
26. O’Connell MJ, Mailliard JA, Kahn MJ, et al. Controlled trial of fluorouracil and low-dose leucovorin given for 6 months as postoperative adjuvant therapy for colon cancer. J Clin Oncol 1997;15:246 –250. 27. Saltz LB, Cox JV, Blanke C, et al., for the Irinotecan Study Group. Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 2000;343:905–914. 28. Douillard JY, Cunningham D, Roth AD, et al. Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: A multicentre randomised trial. Lancet 2000;355:1041–1047. 29. de Gramont A, Figer A, Seymour M, et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000;18:2938 – 2947. 30. Giacchetti S, Perpoint B, Zidani R, et al. Phase III multicenter randomized trial of oxaliplatin added to chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2000;18:136 –147.