Ann Surg Oncol (2012) 19:41864192 DOI 10.

1245/s10434-012-2485-1

ORIGINAL ARTICLE COLORECTAL CANCER

Abdominoperineal Resection for Squamous Cell Anal Carcinoma: Survival and Risk Factors for Recurrence
re mie H. Lefe ` vre, MD1, He le ` ne Corte, MD1, Emmanuel Tiret, MD1, David Boccara, MD2, Marc Chaouat, MD2, Je 3 Emmanuel Touboul, MD , Magali Svrcek, MD, PhD4, Magalie Lefrancois, MD1, Conor Shields, MD1, and Yann Parc, MD, PhD1
1 2

pital Saint-Antoine, AP-HP, Universite Pierre et Marie Curie, Paris, France; Department of Digestive Surgery, Ho pital Saint-Louis, AP-HP, Universite Diderot Paris VII, Paris, France; 3Department of Department of Plastic Surgery, Ho pital Tenon, AP-HP, Universite Pierre et Marie Curie, Paris, France; 4Department of pathology, Ho pital Radiotherapy, Ho Pierre et Marie Curie, Paris, France Saint-Antoine, AP-HP, Universite

ABSTRACT Background. Despite the results of combined chemoradiation therapy for anal canal squamous cell carcinoma (SCC), up to 30 % of patients will undergo abdominoperineal resection (APR). The aim of this study was to evaluate oncologic outcomes, survival, and recurrence, following APR for anal canal SCC performed in a single center over a 13-year period. Methods. All patients who underwent APR for anal canal SCC between 1996 and 2009 were retrospectively included. Demographic data, details on treatments, pathological report, and follow-up were noted. Survival curves were plotted using the KaplanMeier method and potential prognostic factors were evaluated using Cox proportional hazards models. Results. A total of 105 patients (77 women) were included. Indications for APR included tumor persistence (n = 42; 40 %), recurrence (n = 55; 52.4 %), or a contraindication to radiotherapy (n = 8; 7.6 %). Median follow-up was 33.3 months (range, 1.5174.3 months). Overall survival and disease-free survival were, respectively, 61 and 48 % at 5 years. In multivariate analysis, tumor stage (T3 or T4), positive margin on pathologic examination and existence of distant metastases at the time of the surgery were associated with a poor prognosis. The

indication for APR (persistent vs recurrent disease), gender, concurrent HIV infection, or performance of a VRAM ap did not inuence OS or DFS. Overall recurrence rate was 42.6 % (n = 43 of 101). The type of recurrence did not exert a signicant effect on survival (p = .4571). Conclusion. This study describes the largest single series of APR for anal carcinoma. Major prognostic factors for survival and recurrence were T status and involved margin. The 5-year overall survival was 60 %.

Society of Surgical Oncology 2012 First Received: 27 June 2011; Published Online: 24 July 2012 ` vre, MD J. H. Lefe e-mail: jeremie.lefevre@sat.aphp.fr

Since the pioneering work of Nigro et al., primary radical surgery for anal canal squamous cell carcinoma (SCC) has been superseded by combined chemotherapy and radiation therapy, resulting in 5-year survival rates of up to 80 % and a median overall survival of 7.6 years.16 Despite these results of nonsurgical therapy, up to 30 % of patients will undergo abdominoperineal resection (APR) for persistence after radiotherapy or tumoral recurrence.711 Long-term survival rates range widely from 40 to 60 %, reecting the heterogeneity in neoadjuvant therapy and the individual characteristics of both patients and tumors. The formulation of guidelines on the management of these patients is confounded by the small size of the published series, and the length of time over which patients were accrued.10,12 Some prognostic factors are well established, including an involved margin or tumor size.10,13 However, other factors, such as the indication for APR (recurrence vs persistence), are still ill dened.10,14,15 The aim of this study was to evaluate oncologic outcomes, survival, and recurrence, and to identify prognosis features following APR for anal canal SCC performed in a single center over a 13-year period.

Abdominoperineal Excision for Anal Canal Cancer

4187

METHODS AND PATIENTS All patients who underwent APR for anal canal SCC between January 1996 and November 2009 were included. The diagnosis of SCC was conrmed by biopsy and histology in all cases. Demographic data, details on neoadjuvant treatment, surgical procedure, signicant perioperative and postoperative events, and histology were collated and reviewed. Follow-up was obtained using clinic les for patients followed in our department or with telephone interviews by a physician (JHL and HC) for the remaining patients (calling family doctor/GP or patient himself/herself). Management of anal tumor was based upon current recommendations: after a rst clinical examination, an initial radiation dose of 4046 Gy was delivered. Patients were then evaluated 45 weeks after with regard to response and if a complete or nearly complete tumor ([50 %) regression was achieved, a complementary radiotherapy to a total dose of 6064 Gy was delivered. Conversely, if the tumor response was questionable after 4046 Gy, or in presence of radiation-induced morbidity (rectovaginal stula, major fecal incontinence) the patient underwent APR. After completion of radiotherapy, a persistent ulceration or a re-emergence of the lesion within 6 months of radiotherapy was classied as residual disease, while lesions appearing after 6 months post-therapy were classied as a recurrence. Surgical Procedure APR was performed in a manner described previously.14 Briey, extensive resection, when indicated, included posterior colpectomy for women and partial prostatectomy for men (removing a signicant portion of the gland if needed). Primary closure with omentoplasty, pedicalized on the left gastroepiploic vessels, was performed when possible. When not feasible, reconstruction was performed with a vertical rectus abdominis muscle (VRAM) ap, as described before.16 The original technique, which Taylor described, has been modied to include an oblique paddle to allow a larger area of skin to be harvested for reconstruction. The VRAM ap has been performed since 1999 in our department. Preoperative assessment for the ap includes ultrasound examination to verify the patency of the inferior epigastric vessels. No coloperineal anastomosis was performed during the study period. Patients were followed up every 6 months during the rst 5 years with a clinical exam, blood test (SCC Ag), and a radiologic exam (CT scan/ultrasound alternatively). Statistical Analysis Results are presented as mean standard deviation and were subjected to t test. Contingency tables were analyzed

by chi-square test. A p value of \.05 was regarded as signicant. Survival curves were plotted using the Kaplan Meier method.17 Potential prognostic factors were evaluated using Cox proportional hazards models.18 For each outcome, factors achieving a p value \.20 in the univariate analysis were included in a multivariable model. A backward stepwise variable selection procedure was used to remove factors with p value [.05 in the multiple model. Analyses were carried out using StatView software (Version 5, 19921998, SAS Institute Inc., Cary, NC). RESULTS Patient Cohort Between 1996 and 2009, 105 consecutive patients underwent an APR for SCC of the anal canal. No patients were excluded from the study. The clinical, pathological, and neoadjuvant treatment data are summarized in Table 1. There were 77 women (73.3 %). All but 8 patients received radiotherapy or radiochemotherapy prior to surgery. Of these 8 patients, 7 had previously undergone radiotherapy [endometrial cancer (n = 2), cervical cancer (n = 3), ovarian cancer (n = 1), seminoma (n = 1)], while the last patient was deemed unt due to systemic sclerosis. Radiotherapy was combined with 5FU (n = 70) and/or cisplatin (n = 64). Indications for APR included tumor persistence (n = 42; 40 %), recurrence (n = 55; 52.4 %), or a contraindication to radiotherapy (n = 8; 7.6 %). Details of the surgical procedure and histology are described in Table 2.

TABLE 1 Patient characteristics before abdominoperineal resection for anal cancer (n = 105) n (%) Women Body mass index (kg/m2) HIV-infection Age at the time of diagnosis TNM stage at the time of diagnosis Stage I Stage II Stage IIIA Stage IIIB Stage IV Not available Neoadjuvant treatment Radiotherapy Mean dose (Gy) Concomitant chemotherapy 99 (92.5 %) 56.1 18.4, 60 (072) 77 (73.3 %) 77 (73.3 %) 23.9 14.9, 22.7 (1639) 12 (11.2 %) 57.9 12.9, 55.5 (30.386.1) 83 (79 %) 1 (0.95 %) 37 (44 %) 26 (35.2 %) 18 (17.1 %) 1 (0.95 %) 22 (20.9 %)

Results are expressed as the mean standard error, median (range) for continuous variables and N (%) for categorical variables

4188 TABLE 2 Surgical procedure and histological ndings (N = 105) Characteristics Age at the time of surgery (years) Delay between the diagnosis and the APR (months) Associated procedures Posterior colpectomy Partial prostatectomy Hysterectomy Ovariectomy Partial resection of the coccyx Partial resection of the bulbo-cavernous muscle Cystoprostatectomy Iliac nodes resection Hepatic metastasectomy Wound healing procedures Omentoplasty VRAM Cecoplasty Immediate perineal closure Hospital stay Histological results Stage 0 Stage I Stage II Stage IIIA Stage IIIB Stage IV Resection R0 Resection R1 14 (13.3 %) 10 (9.5 %) 36 (34.3 %) 32 (30.5 %) 11 (10.5 %) 4 (3.8 %) 86 (81.9 %) 19 (18.1 %) 46 (42.9 %) 51 (48.6 %) 4 (4.2 %) 97 (90.6 %) 26.7 14.5, 19 (8260) n (%) 59.5 12.8, 57.3 (31.688) 15.3 14.9, 11.6 (0.492.4) 60/77 (77.9 %) 4/30 (13.3 %) 4/77 (5.2 %) 2/77 (2.6 %) 1 (1 %) 1 (1 %) 1 (1 %) 2 (2 %) 3 (2.8 %)

` vre et al. J. H. Lefe

complications. Also, 35 patients had at least 1 complication (33.3 %), resulting in 21 reoperations (20 %). Morbidity was due to perineal wound problems in 50 % of the patients, leading to 11 early reinterventions. Other reasons were small bowel obstruction (n = 4) and various causes (n = 5) (i.e., cholecystitis, wound hernia, ureteral drainage, stoma refection, Bartholinitis). Overall and Disease-Free Survival Mean follow-up time, starting from the time of surgery, was 44.3 37.2 months, with a median of 33.3 months (range, 1.5174.3). It was comparable between patients who had APR for recurrence and for persistence of the disease (p = .3378). Follow-up from the time of diagnosis was either comparable between the 2 indications of APR (55.6 40.3 months for patients with a persisting tumor vs 67.0 39.3 months for patients with a tumoral recurrence, p = .1645). Overall survival (OS) and disease-free survival (DFS) curves are shown in Fig. 1a. OS and DFS were, respectively, 85 and 84 % at 1 year, 66 and 58 % at 3 years, and 61 and 48 % at 5 years. Median OS was 77.7 months, and median DFS was 40 months. There were 6 patients who died of unrelated disease processes [trauma (n = 2); stroke (n = 2), pancreatic cancer (n = 1), hepatocellular carcinoma (n = 1)]. The 5-year cancer-specic survival was 65.8 %. Univariate and multivariate analysis for OS and DFS are reported in Table 3. In the multivariate analysis, tumor size (T3 or T4), positive margin on pathologic examination and existence of distant metastases at the time of the surgery were associated with a poor prognosis. The impact of involved margins on overall survival is shown on Fig. 1b. The 5-year survival was 69 % for patients with a R0 resection, and zero for patients with an invaded margin (p \ .0001). The indication for APR (persistent vs recurrent disease), gender, concurrent HIV infection, or performance of a VRAM ap did not inuence OS or DFS. OS was not signicantly different between patient operated on for persisting tumor or a recurrent cancer even with follow-up starting at the time of diagnosis and not the time of surgery (Fig. 1c). However, the 8 patients operated on without chemoradiotherapy (RCT) had a signicant worse prognosis (p = .041). Recurrence Overall recurrence rate (after exclusion of the 4 patients with synchronous metastatic disease at the time of surgery) was 42.6 % (n = 43 of 101). Recurrences were located in local area [n = 17 (16.8 %)], inguinal nodes (n = 12; 11.9 %), or distant metastasis (n = 14; 13.9 %). The type of recurrence did not exert a signicant effect on survival as shown in Fig. 1d (p = .4571). Risk factor analysis for local or distant recurrence is detailed in Table 4.

Results are expressed as the mean standard error, median (range) for continuous variables and N (%) for categorical variables VRAM vertical rectus abdominis muscle

The VRAM technique was mainly used in patients with T3 or T4 stage disease [35 of 51 (68.6 %) vs 16 of 51 (31.4 %); p = .005]. The incidence of involved margins on histological examination was not statistically different between the 2 perineal closure techniques: 23.5 % (n = 12 of 51) in the VRAM group and 13.2 % (n = 7 of 53) in the remaining patients (p = .06). There was a trend toward a higher rate of histological margin involvement in the T3 T4 cohort (14 of 57, 24.6 %) than in the T0T2 group (5 of 48, 10.4 %), (p = .0607). Mortality and Morbidity The mortality rate 2 months after surgery was 2.1 % (n = 2). The 2 patients died at 45 and 51 days of septic

Abdominoperineal Excision for Anal Canal Cancer FIG. 1 a Overall survival (OS) and disease-free survival (DFS) curves. b Impact of tumoral margin on overall survival (p \ .0001). c Overall survival from the time of diagnosis and depending on the indication of abdominoperineal resection (p = .041). d Overall survival depending on the site of recurrence, log-rank test (p = .4571)

4189

a
Percent survival
1.00

b
Percent survival OS DFS
1.00

R0 R1

0.75

0.75

0.50

0.50

0.25

0.25

20

40

60

80

100

20

40

60

80

100

Time (months)

Follow-up (months)

c
Percent survival
1.00

d
Percent survival Persistance Recurrence No RCT
1.00

Inguinal Local Metastatic

0.75

0.75

0.50

0.50

0.25

0.25

20

40

60

80

100

20

40

60

80

100

Follow-up (months)

Follow-up (months)

DISCUSSION In the present work, we report the oncologic results of patients undergoing APR in a large cohort of 105 patients over a 13-year period. This series is one of the largest ever published on the surgical results of APR for anal canal SCC. Overall survival was 60 % at 5 year with a major inuence of T stage, histological margins, and presence of distant metastases. The indication of APR (i.e., recurrence or persisting tumor) was no longer a prognostic factor. Recurrence rate was 40 %, and the location of the recurrence did not inuence survival. SCC of the anal canal accounts for only 12 % of all lower gastrointestinal cancers and nearly 4 % of anorectal carcinomas (included anal and low rectal cancer).19 The gold standard treatment is chemoradiotherapy, but mutilating surgery such as abdominoperineal resection (APR) is still required in about 30 % of patients for persistent or recurrent disease.7,10,20 Because of the relative infrequency of this carcinoma, many studies examining APR for anal canal

cancer contain small numbers or include patients accrued over a very long period of time, inducing a certain degree of bias. Initial studies even described different conclusions regarding the impact of the indication (i.e., recurrence or persistence) of the APR.14,15 We have already published on the inuence of the VRAM ap on perineal healing after APR.21 The present series allowed analysis of inuencing factors on survival and recurrence following APR. Regarding overall survival in this study, 60 % at 5 years compares favorably with recent studies that report a range of 5264 % (Table 5).10,12,15 However, in other series, patients with rT3 or rT4 tumors comprised only between 13 and 35 %. In this series, such patients represent more than half of the cohort (n = 57; 54.3 %) indicating favorable survival despite the tumor size. Comparison on the basis of histology is not easy, as in many series, the details of the postoperative pathology are not provided.12,15,22,23 In the present study, the main factors inuencing survival and recurrence were T status, margin status, and the presence of metastases at the time of surgery. These factors

4190 TABLE 3 Univariate and multivariate analysis of overall and disease free survival (N = 105) Variable Univariate analysis p value Multivariate analysis p value 95 % CI Local recurrence .65 .3477 .7019 .9636 .5286 .0005 .0798 .03 \.0001 .4913 .3345 .2754 .6463 .4291 .0001 .0164 \.0001 \.0001 .028 NS \.0001 \.0001 10.6 (3.531.7) 4.1 (2.17.7) 2.5 (1.44.6) .0056 NS .0006 .0003 9.2 (2.632.7) 3.9 (1.98.0) 2.76 (1.35.6) Gender Age [ 56 VIH No VRAM Tumoral recurrence yp T3T4 N? R1 Metastatic recurrence Gender Age [ 56 VIH No VRAM Tumoral recurrence yp T3-T4 N? R1 Inguinal recurrence Gender Age [ 56 VIH No VRAM Tumoral recurrence yp T3T4 N? R1 .4311 .1118 .4215 .1650 .4028 .0001 .0072 .003 .0029 .0285 NS NS NS .5049 .3235 NA .0189 .7233 .1932 .1379 .2805 NS NS NS .9198 .1683 .5257 .4562 .4994 .0011 .0010 \.0001 .0142 NS \.0001 NS

` vre et al. J. H. Lefe TABLE 4 Multivariate analysis of risk factors of recurrences after APR Variable Univariate analysis p value Multivariate analysis p value 95 % CI

Overall survival Sex Recurrence Age [ 56 VIH VRAM yp T3T4 N? M1 R1 Sex Recurrence Age [ 56 VIH VRAM yp T3T4 N? M1 R1

4.2 (1.413.3) 7.2 (2.917.9)

Disease-free survival

NS not signicant, 95 % CI 95 % condence interval

are also identied in other series.13,2325 The impact of an involved margin is signicant with a 5-year survival of 0 % in this study. This observation emphasizes the importance of optimal surgical technique and the requirement for aggressive resection (if necessary, larger tissue resection, colpectomy, or prostatectomy). The performance of large resections facilitated by ap reconstruction may explain the absence of macroscopic residual tumor tissue (R2) in this cohort with advanced tumors. Multivariable analysis of factors inuencing tumor recurrence revealed that resection margin (R1) and the size of the tumor (T3 or T4) had a signicant impact. Despite larger tumors and a greater degree of R1 margin involvement in the VRAM ap cohort, overall survival did not differ from the primary closure cohort. With regard to the inuence of HIV infection on overall survival and disease-free survival, we did not nd a signicant inuence, in contrast to other studies.26 The indication for APR (i.e., recurrence or persisting disease), previously described as a signicant prognostic factor, was not found in the present study to demonstrate an impact upon survival.14,15 This result is probably explained by the larger number of patients included in our cohort. Indeed, Mariani et al. included 83 patients and did not nd that the indication for APR impacted upon survival.10 Even with a

6.9 (1.924.4) 2.9 (1.17.6)

NS not signicant, 95 % CI 95 % condence interval

follow-up starting from the time of diagnosis, the survival was similar between patients with a recurrence or a persisting tumor, while one could have expected a longer survival for patients with a recurrent tumor. In the present series, patients with an APR without neoadjuvant radiochemotherapy had a signicantly worse survival than the remaining patients. Moreover, the indication of APR had no inuence on the localization of the recurrence. T status and involved margins are denitively the major prognosis factors. Accordingly, further studies are needed to assess the role of preoperative treatment and eventually the indications for adjuvant chemotherapy after APR if histology reveals involved margins. To improve survival and recurrence results for these patients, a therapeutic option could be to improve number of R0 resection, with enlarging resection margins (which is easier with the ap reconstruction). Another possibility could be a better

Abdominoperineal Excision for Anal Canal Cancer TABLE 5 Comparison of the former series on APR for anal squamous cell cancer Study, year Period n T3T4 at initial treatment NA NA NA 37.1 % NA 6% 58.3 % NA 13 % 35 % NA NA 35 % 59.2 % Median of follow-up (months) 47 40 53 33 24.2 16 67 45 29 18 NA 66 104 33 Median of 5-year OS 44 % 33 % 47 % 52 % 33 % 30 % 69.4 % 40 % 64 % 39 % 69 % 53 % 56.5 % 60 % VRAM

4191

Recurrence rate 60.5 % 38.1 % NA 42.8 % 42.3 % 89 % 64 % NA 38.7 % 52 % NA NA 20.7 % 40.9 %

Ellenhorn 199427 Pocard 199814 Van der Wal 200128 Nilsson 200215 Akbari 200413 Ferenschild 200529 Ghouti 200522 Renehan 200523 Mullen 200712 Schiller 200725 Vorobev 200830 Rouquie 200831 Mariani 200810 Present study

19811993 19861995 19801998 19852000 19802001 19852000 19872002 19882000 19902002 19882006 19952007 19872007 19692003 19962009

38 21 17 35 55 18 36 70 31 40 39 95 83 105

0 0 9 (52.9 %) 0 0 4 (22.2 %) 10 (27.8 %) 2 (2.8 %) 14 (45.2 %) 18 (45 %) 0 2 (2.1 %) 0 51 (48.7 %)

OS overall survival, VRAM vertical rectus abdominis muscle

preoperative staging, today better with the use of MRI, to adapt preoperative treatment and extent of resection. In conclusion, this study describes the largest single series of APR for anal carcinoma. Major prognostic factors for survival and recurrence were T status and involved margin. The 5-year overall survival was 60 %. The use of a VRAM ap was not associated with a reduction in local recurrence, but facilitated a more extensive resection. REFERENCES
1. Nigro ND, Vaitkevicius VK, Considine B Jr. Combined therapy for cancer of the anal canal: a preliminary report. Dis Colon Rectum. 1974;17:3546. 2. Epidermoid anal cancer: results from the UKCCCR randomised trial of radiotherapy alone versus radiotherapy, 5-uorouracil, and mitomycin. UKCCCR Anal Cancer Trial Working Party. UK Co-ordinating Committee on Cancer Research. Lancet. 1996; 348:104954. 3. Flam M, John M, Pajak TF, Petrelli N, Myerson R, Doggett S, et al. Role of mitomycin in combination with uorouracil and radiotherapy, and of salvage chemoradiation in the denitive nonsurgical treatment of epidermoid carcinoma of the anal canal: results of a phase III randomized intergroup study. J Clin Oncol. 1996;14:252739. 4. Longo WE, Vernava AM 3rd, Wade, TP, Coplin, MA, Virgo, KS, Johnson, FE. Recurrent squamous cell carcinoma of the anal canal. Predictors of initial treatment failure and results of salvage therapy. Ann Surg. 1994;220:409. 5. Bartelink H, Roelofsen F, Eschwege F, Rougier P, Bosset JF, Gonzalez DG, 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;15:20409.

6. Northover J, Glynne-Jones R, Sebag-Monteore D, James R, Meadows H, Wan S, et al. Chemoradiation for the treatment of epidermoid anal cancer: 13-year follow-up of the rst randomised UKCCCR Anal Cancer Trial (ACT I). Br J Cancer. 2010;102: 11238. 7. Peiffert D, Giovannini M, Ducreux M, Michel P, Franc ois E, Lemanski C, et al. High-dose radiation therapy and neoadjuvant plus concomitant chemotherapy with 5-uorouracil and cisplatin in patients with locally advanced squamous-cell anal canal cancer: nal results of a phase II study. Ann Oncol. 2001;12:397404. 8. Staib L, Gottwald T, Lehnert T, Ruf G, Sturm J, Becker HD, et al. Sphincter-saving treatment in epidermoid anal cancer: cooperative analysis of 142 patients in ve German university surgical centers. Int J Colorectal Dis. 2000;15:28290. 9. Deniaud-Alexandre E, Touboul E, Tiret E, Sezeur A, Houry S, Gallot D, et al. [Epidermoid carcinomas of the anal canal treated with denitive radiation therapy in a series of 305 patients]. Cancer Radiother. 2003;7:23753. 10. Mariani P, Ghanneme A, De la Rochefordiere A, Girodet J, Falcou MC, Salmon RJ. Abdominoperineal resection for anal cancer. Dis Colon Rectum. 2008;51:1495501. 11. Roohipour R, Patil S, Goodman KA, Minsky BD, Wong WD, Guillem JG, et al. Squamous-cell carcinoma of the anal canal: predictors of treatment outcome. Dis Colon Rectum. 2008;51: 14753. 12. Mullen JT, Rodriguez-Bigas MA, Chang GJ, Barcenas CH, Crane CH, Skibber JM, et al. Results of surgical salvage after failed chemoradiation therapy for epidermoid carcinoma of the anal canal. Ann Surg Oncol. 2007;14:47883. 13. Akbari RP, Paty PB, Guillem JG, Weiser MR, Temple LK, Minsky BD, et al. Oncologic outcomes of salvage surgery for epidermoid carcinoma of the anus initially managed with combined modality therapy. Dis Colon Rectum. 2004;47:113644. 14. Pocard M, Tiret E, Nugent K, Dehni N, Parc R. Results of salvage abdominoperineal resection for anal cancer after radiotherapy. Dis Colon Rectum. 1998;41:148893. 15. Nilsson PJ, Svensson C, Goldman S, Glimelius B. Salvage abdominoperineal resection in anal epidermoid cancer. Br J Surg. 2002;89:14259.

4192 16. Bell SW, Dehni N, Chaouat M, Lifante JC, Parc R, Tiret E. Primary rectus abdominis myocutaneous ap for repair of perineal and vaginal defects after extended abdominoperineal resection. Br J Surg. 2005;92:4826. 17. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;53:45781. 18. Cox DR. Regression models and life-tables. J R Stat Soc B. 1972;34:187220. 19. Fuchshuber PR, Rodriguez-Bigas M, Weber T, Petrelli NJ. Anal canal and perianal epidermoid cancers. J Am Coll Surg. 1997; 185:494505. 20. Ryan DP, Compton CC, Mayer RJ. Carcinoma of the anal canal. N Engl J Med. 2000;16:792800. 21. Lefevre JH, Parc Y, Kerneis S, Shields C, Touboul E, Chaouat M, et al. Abdomino-perineal resection for anal cancer: impact of a vertical rectus abdominis myocutaneous ap on survival, recurrence, morbidity, and wound healing. Ann Surg. 2009;250: 70711. 22. Ghouti L, Houvenaeghel G, Moutardier V, Giovannini M, Magnin V, Lelong B, et al. Salvage abdominoperineal resection after failure of conservative treatment in anal epidermoid cancer. Dis Colon Rectum. 2005;48:1622. 23. Renehan AG, Saunders MP, Schoeld PF, ODwyer ST. Patterns of local disease failure and outcome after salvage surgery in patients with anal cancer. Br J Surg. 2005;92:60514. 24. Hill J, Meadows H, Haboubi N, Talbot IC, Northover JM. Pathological staging of epidermoid anal carcinoma for the new era. Colorectal Dis. 2003;5:206-13.

` vre et al. J. H. Lefe 25. Schiller DE, Cummings BJ, Rai S, Le LW, Last L, Davey P, et al. Outcomes of salvage surgery for squamous cell carcinoma of the anal canal. Ann Surg Oncol. 2007;14:27809. 26. Wexler A, Berson AM, Goldstone SE, Waltzman R, Penzer J, Maisonet OG, et al. Invasive anal squamous-cell carcinoma in the HIV-positive patient: outcome in the era of highly active antiretroviral therapy. Dis Colon Rectum. 2007;51:7381. 27. Ellenhorn JD, Enker WE, Quan SH, Salvage abdominoperineal resection following combined chemotherapy and radiotherapy for epidermoid carcinoma of the anus. Ann Surg Oncol. 1994;1: 10510. 28. van der Wal BC, Cleffken BI, Gulec B, Kaufman HS, Choti MA. Results of salvage abdominoperineal resection for recurrent anal carcinoma following combined chemoradiation therapy. J Gastrointest Surg. 2001;5:3837. 29. Ferenschild FT, Vermaas M, Hofer SO, Verhoef C, Eggermont AM, de Wilt JH. Salvage abdominoperineal resection and perineal wound healing in local recurrent or persistent anal cancer. World J Surg. 2005;29:14527. 30. Vorobev GI, Shelygin IuA, Nechushkin MI, Rybakov EG. Results of surgical treatment of residual and recurrent anal tumors. Khirurgiia (Mosk). 2008;(8):49. In Russian. re D, Pignon JP, 31. Rouquie D, Lasser P, Castaing M, Boige V, Goe et al. Complete (R0) resection is the only valid prognostic factor in abdominoperineal resection for recurrent cancer of the anal canal (a consecutive series of 95 patients). J Chir (Paris). 2008;145:33540 (In French).

Copyright of Annals of Surgical Oncology: An Oncology Journal for Surgeons is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.