The American Journal of Surgery (2011) 201, 628 – 633

North Pacific Surgical Association

Outcomes of salvage surgery for epidermoid carcinoma of the anus following failed combined modality treatment Gareth Eeson, M.D.,a,* Marcus Foo, M.D.,b Stephen Harrow, M.D.,b Gregor McGregor, M.D.,a,c,d John Hay, M.D.b a

Division of General Surgery, University of British Columbia, Vancouver, BC, Canada; bDivision of Radiation Oncology, British Columbia Cancer Agency, Vancouver, BC, Canada; cDivision of Surgical Oncology, Vancouver General Hospital, Vancouver, BC, Canada; dDivision of Surgical Oncology, British Columbia Cancer Agency, Vancouver, BC, Canada KEYWORDS: Epidermoid anal canal cancer; Squamous cell carcinoma; Chemoradiation; Recurrence; Treatment failure; Salvage abdominoperineal resection

Abstract BACKGROUND: Chemoradiation is first-line therapy for epidermoid carcinoma of the anus (ECA).

Surgery is reserved for treatment failures. The authors report outcomes after salvage procedures for ECA. METHODS: All treatment failures managed with radical surgery between 1998 and 2006 in our institution were reviewed. The Kaplan-Meier method was used for survival analysis. Log-rank and Cox regression were used for univariate and multivariate analysis, respectively.

RESULTS: Fifty-one patients underwent salvage abdominoperineal resection for locoregional failure. Five-year overall survival after abdominoperineal resection was 29% (median, 22 months). Age, gender, human immunodeficiency virus status, tumor-node-metastasis stage, node status, and failure type did not predict survival. Negative resection margin was most strongly associated with improved overall and disease-free survival (P ⫽ .03 and P ⬍ .0001, respectively). Median survival for patients undergoing inguinal lymph node dissection for regional recurrence (n ⫽ 6) was 11 months, with freedom from cancer achieved in 2 of 6 patients.

CONCLUSIONS: Recurrent anal carcinoma after primary chemoradiotherapy carries a poor prognosis. Salvage abdominoperineal resection offers a potential for long-term survival. © 2011 Elsevier Inc. All rights reserved.

Carcinoma of the anus is an uncommon disease constituting 1.9% of all malignancies of the digestive tract.1 Epidermoid carcinoma of the anus (ECA) is responsible for 80% to 85% of these cancers and consists mainly of squamous cell carcinoma and its variants. The incidence of anal carcinoma in the general population is 20 per million and has doubled in the past 30 years, largely because of the

* Corresponding author. Tel.: 44-07876728371; fax: 001-604-8754036.

E-mail address: gareth.eeson@lshtm.ac.uk Manuscript received November 5, 2010; revised manuscript January 25, 2011

0002-9610/$ - see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjsurg.2011.01.015

human immunodeficiency virus (HIV) epidemic, but the fact remains that the disease is infrequently encountered by most physicians and surgeons.2,3 Historically, standard first-line treatment for anal carcinoma was radical abdominoperineal resection (APR).4 – 8 This approach shifted dramatically after the observation of complete remission in patients treated with chemoradiation alone, reported by Nigro et al9 in 1974. These findings have subsequently been verified and further refined in the context of randomized controlled trials.10 –12 Consequently, current standard first-line treatment includes radical combined chemoradiotherapy (CRT), which provides 5-year overall survival (OS) rates of 72% to 89% and colostomy-free survival

G. Eeson et al.

Salvage surgery for anal carcinoma

rates of 70% to 86%.3–17 In comparison, outcomes before the era of radical CRT witnessed 40% to 70% 5-year survival, with 3% perioperative mortality from APR.6 – 8 Currently, radical resection for anal carcinoma is reserved for residual or recurrent locoregional disease and for patients with prohibitive side effects from CRT. Despite marked improvements in clinical and oncologic outcomes with combined modality therapy, treatment failures continue to present a considerable challenge. Persistent or recurrent disease is thought to occur in 20% to 40% of patients after first-line CRT.10 –12 APR is the treatment of choice for local treatment failure and is thought to provide the greatest chance of cure in this setting.18 –25 The relative efficacy of nonsurgical treatments such as local excision and salvage CRT is not well known, but outcomes from small series are generally poor.11,23 Although several reports have endeavored to describe outcomes of salvage APR, most of these studies were small, were heterogeneous, and spanned several decades,18 –25 during which time considerable progress has been made in the understanding and treatment of anal carcinoma.10 –12 This study reports the outcomes of surgical salvage after failed combined chemoradiation for anal carcinoma in a contemporary series of patients.

Methods The British Columbia Cancer Agency patient registry was used to identify all patients treated for anal carcinoma between 1998 and 2006. The British Columbia Cancer Agency is a multicenter network providing multidisciplinary cancer care for the province of British Columbia, Canada (estimated population, 4.4 million). To satisfy inclusion criteria for the study, a patient must have had (1) a pathologic diagnosis of ECA; (2) clinical documentation of tumor location within the anal canal, defined as lying between the superior border of the anorectal ring and the anal verge; (3) first-line radical CRT or radiotherapy alone, with a minimum radiation dose of 40 Gy; (4) clinical or pathologic documentation of persistent or recurrent disease after first-line therapy; and (5) curative-intent salvage resection.

Persistence was defined as pathologic evidence of residual or progressive disease within 6 months of completion of treatment, whereas recurrence was defined as evidence of disease beyond 6 months after treatment after an initial period of clinical response.

A review of patient records was performed to record clinical, radiologic, pathologic, and operative data. Clinical staging was recorded in all patients using the seventh edition of the American Joint Committee on Cancer’s tumornode-metastasis classification.26 Stage at presentation was generally recorded by the treating physician and determined using a combination of clinical, pathologic, and radiologic data. For nodal staging, either evidence of metastatic spread on biopsy or clinical palpability was considered positive.

For suspicious nodes for which neither of these features was

629 available, predefined computed tomography– based size criteria were used to determine nodal status. Patients were all restaged at the time of recurrence using the same criteria.

Survival analysis was performed using a combination of follow-up data, including outpatient clinic records, inpatient records, and vital statistics records. Survivors were categorized according to the presence or absence of disease at last followup. Patients who died were categorized with respect to cause of death and presence of disease at the time of death. Survivorship was measured from the time of first documented recurrence to last follow-up or date of death. OS and disease-free survival analysis was performed using the method of Kaplan and Meier.27 Log-rank test and Cox regression were used for univariate and multivariate analysis, respectively.

Results Patient and treatment characteristics Using the patient registry, we identified 395 consecutive patients who were treated at the British Columbia Cancer Agency between 1998 and 2006 for carcinoma of the anal canal, of whom 51 met all predefined inclusion criteria. A

Table 1 Characteristics of patients undergoing salvage APR (n ⫽ 64) Characteristic Gender Male Female Age (years) HIV positive Histology Squamous cell carcinoma Epidermoid variant (eg, basaloid) Tumor stage at presentation T1 T2 T3 T4 Nodal stage at presentation N0 N1 N2 N3 Metastases at presentation M0 M1 Type of failure Recurrence Persistence Resection margins R0 R1 R2 Not documented

Value 17 (33%) 34 (67%) 60 (35–88) 6 (12%) 38 (75%) 13 (25%) 1 (2%) 21 (41%) 16 (31%) 12 (24%) 29 (57%) 8 (16%) 5 (10%) 9 (18%) 49 (96%) 2 (4%) 31 (61%) 20 (39%) 32 (63%) 11 (22%) 4 (8%) 4 (8%)

Data are expressed as number (percentage) or as mean (range).

630

The American Journal of Surgery, Vol 201, No 5, May 2011

Figure 1

Overall survival from time of recurrence for patients undergoing salvage APR.

total of 51 patients (17 men, 34 women; mean age, 60 years; range, 35– 88 years; HIV prevalence, 12%) were analyzed (Table 1).

During this time period, the standard treatment for anal carcinoma was radiation therapy combined with concurrent chemotherapy. Split-course radiotherapy with 50 Gy in 24 daily fractions (⫾10 –15 Gy boost) with a 2.5-week break was most common in the early study period. Toward the end of the study period, continuous course treatment was more usual. The vast majority of patients received 2 cycles of concurrent 5-fluorouracil and mitomycin C. A small minority received cisplatin instead of mitomycin C.

Abdominoperineal resection Local treatment failure was managed with APR in 51 patients, 31 of whom were for recurrence and 20 for persistent disease. All patients had previously received radiotherapy (mean dose, 51.8 Gy; minimum dose, 48 Gy) and all but 2 received concomitant chemotherapy. Median follow-up from the time of diagnosis was 34 months overall. For patients who died and did not die during the study period, median follow-up was 31 and 53 months, respectively.

After the completion of therapy, median time to local treatment failure was 8.3 months, with the latest recurrence occurring at 62.0 months. Five-year OS for patients undergoing salvage APR was 29% (median, 22 months; 95%

confidence interval, 13.1–39.7 months; Fig. 1). Disease-free survival at 5 years was 24.7%. One patient died within 30 days of surgery (cause unknown). Microscopically, negative margins (R0) were achieved in 32 of 51 patients (63%).

Microscopically (R1) and grossly (R2) positive margins were seen in 11 (22%) and 4 (8%) patients, respectively.

Margin status was unavailable in 4 patients.

APR achieved durable freedom from cancer in 18 of 51 patients (35%) at the time of last follow-up. The remaining patients developed secondary recurrence of cancer after APR or had residual disease after surgery. With regard to patients who had potentially curative operations, 18 of 32 (53%) developed secondary recurrence within the study follow-up. Ten of 18 secondary recurrences after potentially curative surgery had evidence of locoregional disease at the time of second recurrence, whereas 6 had evidence of distant disease but no locoregional burden.

Resection margin status at APR was the most powerful indicator of OS (P ⫽ .03) and disease-free survival (P ⬍ .0001). Negative margins were associated with improved overall 5-year (41.7% vs .0%) and median survival compared with positive margins (33.0 vs 14.3 months) (log-rank test P ⫽ .03; Fig. 2). Resection margin status was the only indicator of decreased OS after multiple regression. After adjusting for resection margin, there appeared to be a trend toward higher rates of post-APR recurrence in HIV-positive patients (hazard ratio, 2.98; 95% confidence interval, .85–10.40; P ⫽ .08). Age

G. Eeson et al.

Salvage surgery for anal carcinoma

631

Figure 2 OS from time of recurrence by resection margin status (R0 ⫽ microscopically negative resection margin; R1/R2 ⫽ positive resection margin).

⬎ 55 years, gender, tumor size, nodal status at diagnosis, American Joint Committee on Cancer stage, nodal status at time of recurrence, and failure type had no statistically significant effect on survival by multiple regression analysis.

Inguinal lymph node dissection (ILND) Seven ILNDs were performed in 6 patients for isolated inguinal metastases. At the time of last follow-up, 3 patients had died of progressive distant disease at 4, 11, and 55 months, while 1 patient was alive with extensive metastatic burden at 25 months. One patient died at 11 months without cancer. One patient was alive without disease at 55 months after bilateral ILND. Local inguinal failure after ILND was not documented in any patient.

Four of 6 patients had received groin irradiation at the time of diagnosis. Median OS was 11 months for patients undergoing ILND for regional recurrence.

Comments Chemoradiation is the standard of care for anal carcinoma and provides durable cure with sphincter preservation in a majority of patients.10 –12,21 Despite improvements in

multimodal therapy over the past 30 years, treatment failure still presents a considerable challenge. A review of our institutional network revealed a recurrence rate of 32.4% in patients treated with radical CRT.28 Of patients in whom CRT fails, roughly 60% recur locoregionally and are considered for radical salvage, most commonly via APR.28 In our series of 51 patients who underwent salvage APR, 5-year OS was 29%. By crude comparison, OS at our institution after local failure in patients not undergoing APR was 0% at 2 years, with a median of 6 months, underscoring the poor outcomes in this population as a whole.28 Our results coincide with other single-institution studies reporting 5-year OS after APR ranging between 24% and 65%.18 –25,29 We evaluated a relatively contemporary cohort in the era after the major randomized clinical trials evaluating CRT,10 –12 serving to illustrate the lack of significant improvement in surgical outcomes within this challenging subgroup.

Achieving negative resection margins was the only significant indicator of improved survival in our study.

Our study was underpowered to detect more moderate effects of other potential clinical indicators, but our data suggest a trend toward higher recurrence rates in HIVpositive patients (hazard ratio, 2.98; P ⫽ .08). A host of other potential predictive factors have been identified by

632

The American Journal of Surgery, Vol 201, No 5, May 2011

other reports but are inconsistent and lend little to clinical decision making.18,19,22–24 Clearly, larger prospective studies are required to address these hypotheses if we hope to be able to base patient selection on accurate estimates of benefit.

The importance of clear margins is intuitive to surgeons and clear from this and other studies.18,22 Our own series noted a considerable rate of positive resection margins compared with some single-center series (30% vs 11%–17%).18,22,24 To some degree, this may be accounted for by aggressive biology and overrepresentation of large tumors (55% T3 and above), but certainly these findings encourage efforts aimed at improving marginnegative resection rates within our institutional network.

Advanced preoperative imaging, aggressive multivisceral resection, subspecialty support, and early recognition of treatment failures may drive improved regional outcomes. A very real challenge in these patients is the identification of treatment failures after radiotherapy. Efforts aimed at early recognition of recurrence after CRT are justified by their potential to increase rates of curative salvage resection. Furthermore, identifying tumors at high risk for treatment failure at the time of diagnosis may allow us to determine a subgroup that may benefit from planned APR. Inguinal recurrence after primary treatment occurs in ⬍5% of patients treated with elective inguinal irradiation, posing an uncommon but challenging problem.30 Our series of 6 patients undergoing ILND demonstrated generally poor outcomes, reflected by the median survival of 11 months and disease- clearance achieved in 2 of 6 patients. Although this was a very small experience, the literature is deficient in describing this subgroup. In a series of 19 patients with metachronous inguinal disease, durable cure was achieved after ILND plus radiation therapy in 9 of 19 patients, with 5-year OS of 41.4%, but none had received previous inguinal irradiation.30 Akbari et al18 reported a small series of 5 patients with recurrent disease after groin irradiation, of whom 3 were rendered disease-free after ILND. Although durable cure after ILND is achieved in a minority of patients, it remains a viable option in carefully selected patients.

The outlook for patients with recurrent anal cancer after primary chemoradiation remains poor. Fewer than a third of patients treated with APR after failed CRT can expect to survive 5 years and must endure considerable treatment-related morbidity. A host of factors, including postradiation changes, HIV infection, and aggressive biology, make these patients particularly challenging to surgeons, who continue to play a key role in both the curative and palliative arena. Salvage APR provides the best opportunity for cure in patients with locoregional failure on the basis of the available knowledge. Marginnegative resection is of critical importance and should be the objective for any curative-intent resection.

References 1. Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2009. CA Cancer J Clin 2009;59:225– 49.

2. Johnson LG, Madeleine MM, Newcomer LM, et al. Anal cancer incidence and survival: the surveillance, epidemiology, and end results experience, 1973-2000. Cancer 2004;101:281.

3. Chiao EY, Krown SE, Stier EA, et al. A population-based analysis of temporal trends in the incidence of squamous anal canal cancer in relation to the HIV epidemic. J Acquir Immune Defic Syndr 2005;40:451–5.

4. Golden GT, Horsley JS III. Surgical management of epidermoid carcinoma of the anus. Am J Surg 1976;131:275– 80.

5. Salmon RJ, Fenton J, Asselain B, et al. Treatment of epidermoid anal canal cancer. Am J Surg 1984;147:43– 8.

6. Pintor MP, Northover JM, Nicholls RJ. Squamous cell carcinoma of the anus at one hospital from 1948 to 1984. Br J Surg 1989;76:806 –10.

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 –25.

8. 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.

9. Nigro ND, Vaitkevicius VK, Considine B Jr. Combined therapy for cancer of the anal canal: a preliminary report. Dis Colon Rectum 1974;17:354 – 6.

10. 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 –9.

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 randomized intergroup study. J Clin Oncol 1996;14:2527–39.

12. UKCCCR Anal Cancer Trial Working Party. Epidermoid anal cancer: results from the UKCCCR Randomised Trial of radiotherapy alone versus radiotherapy, 5-fluorouracil, and mitomycin. Lancet 1996;348:1049 –54.

13. Ajani JA, Winter KA, Gunderson LL, et al. US Intergroup Anal Carcinoma Trial: tumor diameter predicts for colostomy. J Clin Oncol 2009;27:1116 –21.

14. Doci R, Zucali R, La Monica G, et al. Primary chemoradiation therapy with fluorouracil and cisplatin for cancer of the anus: results in 35 consecutive patients. J Clin Oncol 1996;14:3121–5.

15. Peiffert D, Seitz JF, Rougier P, et al. Preliminary results of a phase II study of high-dose radiation therapy and neoadjuvant plus concomitant 5-fluorouracil with CDDP chemotherapy for patients with anal canal cancer: a French cooperative study. Ann Oncol 1997;8:575– 81.

16. Gerard JP, Ayzac L, Hun D, et al. Treatment of anal canal carcinoma with high dose radiation therapy and concomitant fluorouracil-cisplatinum.

Long-term results in 95 patients. Radiother Oncol 1998;46:249 –56.

17. Martenson JA, Lipsitz SR, Lefkopoulou M, et al. Results of combined modality therapy for patients with anal cancer (E7283). An Eastern Cooperative Oncology Group study. Cancer 1995;76:1731– 6.

18. Akbari RP, Paty PB, Guillem JG, et al. Oncologic outcomes of salvage surgery for epidermoid carcinoma of the anus initially managed with combined modality therapy. Dis Colon Rectum 2004;47:1136 – 44.

19. Mullen JT, Rodriguez-Bigas MA, Chang GJ, et al. Results of surgical salvage after failed chemoradiation therapy for epidermoid carcinoma of the anal canal. Ann Surg Oncol 2007;14:478 – 83.

20. Stewart D, Yan Y, Kodner IJ, et al. Salvage surgery after failed chemoradiation for anal canal cancer: should the paradigm be changed for high-risk tumors? J Gastrointest Surg 2007;11:1744 –51.

21. Blumetti J, Bastawrous AL. Epidermoid cancers of the anal canal: current treatment. Clin Colon Rectal Surg 2009;22:77– 83.

G. Eeson et al.

Salvage surgery for anal carcinoma

22. Schiller D, Cummings B, Rai S, et al. Outcomes of salvage surgery for squamous cell carcinoma of the anal canal. Ann Surg Onc 2007;14:2780–9.

23. Longo WE, Vernava AM, Wade TP, et al. Recurrent squamous cell carcinoma of the anal canal. Predictors of initial treatment failure and results of salvage therapy. Ann Surg 1994;220:40 –9.

24. Renehan AG, Saunders MP, Schofield PF, et al. Patterns of local disease failure and outcome after salvage surgery in patients with anal cancer. Br J Surg 2005;92:605–14.

25. Ellenhorn JD, Enker WE, Quan SH. Salvage abdominoperineal resection following combined chemotherapy and radiotherapy for epidermoid carcinoma of the anus. Ann Surg Oncol 1994;1:105–10.

26. Edge SB, Byrd DR, Compton CC, et al. AJCC (American Joint Committee on Cancer) cancer staging manual. 7th ed. New York: Springer; 2010.

27. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Statist Assoc 1958;53:457– 81.

28. Foo M, Harrow S, Hay J, et al. Outcomes and treatment following recurrence of anal carcinoma after radiotherapy: the British Columbia experience. Unpublished data, 2010.

29. Zelnick RS, Haas PA, Ajlouni M, et al. Results of abdominoperineal resections for failure after combination chemotherapy and radiation therapy for anal canal cancers. Dis Colon Rectum 1992; 35:574 – 8.

30. Gerard JP, Chapet O, Samiei F, et al. Management of inguinal lymph node metastases in patients with carcinoma of the anal canal: experience in a series of 270 patients treated in Lyon and review of the literature. Cancer 2001;92:77– 84.

Discussion Scott Browning, M.D. (Portland, OR): The work presented by Eeson and colleagues updates our understanding of a difficult and underappreciated problem. Anal cancer is uncommon, but not quite rare. I suspect that most colorectal surgeons see no more than about one new case each year. Combine this low rate with the fact that most of these patients will be cured by primary chemoradiation therapy, and it is easy for us to forget that this once primarily surgical disease still may yet require our intervention. This paper is a sobering reminder that more of these patients than our individual anecdotal experiences would suggest do go on to fail multimodality therapy. There are many reports in the literature addressing salvage therapy for recurrent or persistent anal cancer, but as the authors point out, these studies tend to be small and/or span decades of changing therapy. The result is a collection of papers which largely contradict each other with regard to which factors do or do not impact recurrence and survival following salvage resection.

633 In an attempt to clarify the data with a sizable contemporary cohort, the authors have queried the substantial registry of the British Columbia Cancer Agency to identify the evaluable records of patients with persistent or recurrent anal cancer who underwent curative intent salvage surgery between 1998 and 2006, following failure of first-line chemoradiation therapy or radiation therapy alone. The main group developed included 51 such patients who underwent salvage APR with curative intent. Clear margins (R0) were achieved in only 63% of these patients, and 53% of those went on to secondary failure. Five-year actuarial and disease-free survival were 29% and 24.7%, respectively.

This paper satisfies the goal of its title, describing the outcomes of salvage surgery for epidermoid cancer of the anal canal following combined modality therapy with a truly contemporary cohort free of some of the deficiencies of prior studies. The conclusion that despite poor outcomes and high morbidity, APR remains the procedure of choice for local treatment failure is justified. However, the utility of this report is, like those that precede it, limited by its sample size. Beyond the purely descriptive aspect of this study, the authors evaluated 9 factors considered potentially predictive of survival following salvage APR. Such predictive factors would be of considerable value in guiding surgeons and patients when making decisions as to whether or not to proceed with a potentially morbid procedure in the name of salvage. Though there appear to have been meaningful differences in median survival following salvage APR according to such preoperatively identifiable factors as age, gender, tumor size, nodal status and AJCC stage, none of these factors reached statistical significance. This is most likely a function of the small size of the patient group studied. Unfortunately, the only factor which significantly predicted survival following APR for persistent or recurrent anal cancer was a negative resection margin, which of course is known only after the procedure has been completed. While this finding lends further importance to achieving an R0 resection, it is of no value in determining which patients will benefit from an operation unless an R0 resection itself can be predicted.

I thank the authors for their contribution and careful analysis of this contemporary cohort and for reminding us that the surgical management of failed multimodality therapy for anal cancer is indeed still a difficult and relevant problem.