Outcome after the introduction of a multimodality treatment program for locally advanced rectal cancer

Outcome after the introduction of a multimodality treatment program for locally advanced rectal cancer

EJSO (2005) 31, 727–734 www.ejso.com Outcome after the introduction of a multimodality treatment program for locally advanced rectal cancer G. Palme...

148KB Sizes 0 Downloads 66 Views

EJSO (2005) 31, 727–734

www.ejso.com

Outcome after the introduction of a multimodality treatment program for locally advanced rectal cancer G. Palmera,*, A. Martlinga, L. Blomqvistb, B. Cedermarka, T. Holma a

Department of Surgery, Karolinska University Hospital and Karolinska Institute, SE-171 76 Stockholm, Sweden b Department of Diagnostic Radiology, Karolinska University Hospital and Karolinska Institute, SE-171 76 Stockholm, Sweden Accepted for publication 22 April 2005 Available online 23 June 2005

KEYWORDS Primary advanced rectal cancer; Locally recurrent rectal cancer; Multimodality treatment; Surgical outcome

Abstract Aim: This prospective study reports the results of a multimodality treatment protocol in patients with locally advanced rectal cancer and assesses outcome after curative vs non-curative surgery and in relation to primary advanced vs locally recurrent cancer. Methods: Between 1991 and 2002, 122 patients completed the protocol. Fifty-eight had primary advanced and sixty-four had locally recurrent rectal cancer. Median follow up was 82 months (5–143). Results: A potentially curative resection was achieved in 59% of the patients with primary advanced and in 34% of patients with locally recurrent cancer. After curative resection, 53 and 59%, respectively, were free from recurrence during the observation time (median 82 months) and the overall 5-year survival was 34 and 40%. Overall 5-year survival in all patients with primary advanced cancer was 29 and 16% in all patients with locally recurrent rectal cancer. Conclusion: Multimodality treatment may cure at least a third of patients with locally advanced rectal cancer provided a radical resection is performed. As the postoperative morbidity is high, an optimised patient selection for neo-adjuvant treatment and surgery is essential. However, palliative surgery may benefit the patient if local control is achieved. Future studies should focus on the problem of distant metastasis. Q 2005 Elsevier Ltd. All rights reserved.

Introduction * Corresponding author. Tel.: C46 8 517 700 00; fax: C46 8 33 15 87. E-mail address: [email protected] (G. Palmer).

Major achievements have been made in the treatment of rectal cancer in recent years. The introduction of total mesorectal excision (TME) and pre-operative radiotherapy has substantially

0748-7983/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ejso.2005.04.009

728 improved outcomes with reduced local recurrence rates and improved survival.1–3 The majority of patients with rectal cancer have tumours confined to the rectum and mesorectum and may be cured by TME-based surgery. However, 10–15% of the patients have locally more advanced tumours, i.e. tumours which are fixed to adjacent structures within the pelvis.4 In these patients and in patients with a local recurrence after previous rectal cancer treatment, surgery alone has a limited potential for cure and multidisciplinary treatment strategies are needed. Early studies of such patients reported local recurrence rates of 70% in tumours with involvement of adjacent organs5 and 5-year survival rates of 5–10% after surgery alone.6,7 The results have been improved by the addition of pre-operative radiotherapy, alone or in combination with chemotherapy, which may downsize the tumour and allow a radical resection.8,9 With extensive en bloc resections and intraoperative radiotherapy (IORT) results may improve further. With such treatment strategies 5-year survival rates of 42–48%8,10 in patients with primary advanced rectal cancer and of 33% in patients with locally recurrent rectal cancer have been reported from dedicated centers.11 In Sweden, neither standardised treatment protocols nor IORT were available for patients with locally advanced rectal cancer until 1990. In 1991, a multimodality treatment protocol was set up at the Karolinska Hospital in Stockholm with the aim to standardise pre-operative evaluation and treatment in these patients. The aim of the present study is to report the results of this multidisciplinary and multimodality protocol and to assess outcome in relation to curative vs non-curative surgery and in relation to primary advanced and locally recurrent rectal cancer.

Material and methods In all, 122 patients completed the treatment protocol between 1991 and 2002; 58 had a locally advanced primary rectal cancer and 64 a locally recurrent tumour. In all patients, magnetic resonance imaging (MRI) and/or computed tomography (CT) of the pelvis was performed to assess the local tumour growth and to evaluate the extension of the tumour in relation to surrounding structures. In patients with an advanced primary tumour these investigations were done to confirm the clinical diagnosis of a fixed tumour, i.e. a tumour growing

G. Palmer et al. into, or in close proximity to other structures within the pelvis. From 1993, MRI was introduced in the routine pre-operative work-up of all patients with rectal cancer. In this group of patients, some cases were first diagnosed as locally advanced by MRI. In addition, CT of the abdomen, a chest X-ray and if possible a colonoscopy was also performed to exclude distant metastases and synchronous lesions in the bowel. After the clinical and radiological assessments the patients were discussed at a multidisciplinary meeting with surgeons, pathologists, radiologists and oncologist and a treatment strategy was decided upon. Until 1997, standard therapy was pre-operative radiotherapy (28!1.8 Gy) in combination with chemotherapy (5-FU and leucovorin). From 1998, pre-operative treatment was given with either radiotherapy (25!2 Gy) alone, or in combination with chemotherapy (5-FU, leucovorin) within the protocol of a randomised multicentre trial (Nordic LARCS-A Trial). After completion of the pre-operative treatment all patients had a second pelvic MRI to assess the tumour response in terms of downsizing or tumour progression and were then scheduled for surgery with IORT availability within 6–8 weeks. The primary aim of the surgical procedure was to remove the tumour completely with safe margins. Hence, multivisceral en bloc resection, including pelvic exenteration and sacral resection was performed when necessary. The surgical procedure was classified by the surgeon as curative if the resection margins were macroscopically free from tumour, as uncertain if the tumour was close to the resection margins and as non-curative if macroscopic tumour was left after completion of the resection. The pathologist classified the resection as curative if the resection margins were free from tumour and as non-curative if the tumour was infiltrating the resection margins. The patients were considered to have had a curative resection if both the surgeon and the pathologist reported non-involved margins and to have had a noncurative resection if the pathologist reported tumour involvement of the resection margins and/or the surgeon reported residual tumour. After surgery the patients were followed by clinical evaluation for at least 5 years or until recurrence or death. MRI of the pelvis and/or CT of the abdomen and chest X-ray was performed at 3 and 12 months post-operatively and later if tumour recurrence was suspected. All patients who developed distant metastases were also evaluated for local recurrences. The median follow up was 82 months (5–143 months); 63 months for patients with

Outcome of advanced rectal cancer

729

an advanced primary cancer and 86 months for patients with a locally recurrent cancer. Data on pre-operative diagnosis and treatment, details on the surgical procedure, post-operative morbidity and mortality and follow up data on late complications, local recurrence, metastases and causes of death were recorded prospectively. In addition to the analysis of prospectively collected data, a thorough review of medical records was performed to ensure a complete registration and validation of all events (complications, recurrence or death). A cross-check with the database at the Regional Oncologic Centre in Stockholm, where data on all patients with rectal cancer in Stockholm are prospectively recorded, was also performed.

Statistical analyses Comparison of proportions was made with the c2-test. Data on local recurrence and distant metastases are presented as the total number of events (patients were considered to be at risk for the studied events until death or end of follow-up). Fiveyear survival was calculated from the date of surgery, according to Kaplan–Meier, and comparisons of survival curves were made with the log-rank test.

Results The median age and sex distribution was similar in patients with advanced primary rectal cancer and in those with locally recurrent rectal cancer (Table 1). Table 1

In one patient with a primary rectal cancer and in five with a local recurrence no preoperative treatment was given because of advanced age or symptomatic cardiovascular disease. In 17 patients with recurrent tumours irradiation had been given before the primary operation. In 10 patients with locally recurrent cancer distant metastases were present at the time of diagnosis and these patients were operated on for palliative reasons only. The proportion of en bloc resections of adjacent organs was similar in patients with an advanced primary tumour and in patients with a local recurrence; 42 vs 45 patients. However, pelvic excenteration was performed more often in patients with an advanced primary tumour than in patients with a local recurrence (15/58 vs 6/64; pZ 0.009). An irresectable tumour found at laparotomy was less common in patients with an advanced primary tumour than in patients with a local recurrence (3/58 vs 15/64; pZ0.04). IORT was given to areas possibly infiltrated by tumour and at the discretion of the responsible surgeon. 19 patients with an advanced primary tumour and 42 patients with a local recurrence received IORT (pZ0.003). Surgery was assessed as potentially curative in 34 patients with an advanced primary tumour and in 22 patients with a local recurrence (pZ0.007). In patients with an advanced primary cancer surgery was considered non-curative because of tumour at the resection margin in 6/58 patients, because of macroscopic residual tumour in 14/58

Patient and treatment characteristics

Male:female Age median (range) Median follow-up time months (range) Pre-operative treatment No treatment Radiatherapy (RT) alone Chemotherapy alone Combination RTCchemotherapy Surgical procedure Low anterior resection Abdominoperineal resection Hartman’s procedure Explorative laparotomy Other resection of recurrent disease Pelvic excenteration IORT Assessment of local tumour control Curative Non-curative

Advanced primary (nZ58)

Locally recurrent (nZ64)

36:22 65 (28–81) 63.5 (5–137)

35:29 63 (27–87) 86 (5–143)

1 28 1 28

22 10 1 31

13 36 6 3 0 15 19

3 30 9 15 7 6 42

34 24

22 42

730

G. Palmer et al.

patients and because of distant metastases but clear margins in 4/58 patients. The corresponding figures for patients with a local recurrence in whom surgery was non-curative were 5/64, 33/64 and 4/64. Thus, macroscopic residual tumour was less common in patients with an advanced primary tumour than in patients with a local recurrence (pZ0.002).

22 patients. Thirty-five patients had persistent pain necessitating regular use of analgesics. There was no significant relationship between persistent pain and the use of IORT. The rate of late complications was similar after potentially curative and noncurative surgery. In all, 22/122 patients experienced no complications or complaints after treatment.

Post-operative mortality and morbidity

Recurrence and survival

Median time in hospital was 14 days (5–120) and similar in both groups. Seventy percent of the patients could be discharged to their homes while 30% needed further care in rehabilitation hospitals. Six patients died within 3 months after surgery (Table 2). The causes of death were sepsis and cardiovascular failure. The proportion of patients with early postoperative complications (in hospital or within 30 days post-operatively) was 39/58 in patients with primary tumours and 45/64 in those with recurrent cancer. The most common surgical complications were wound infections, wound dehiscence and intraabdominal abscesses. Other post-operative infections, such as urinary tract infections and pneumonia, were also common. Complications or complaints occurring later than 30 days after surgery included defecation disturbances and urinary tract infections, fistulae, pain and intestinal obstruction. Urological problems, ranging from repeated urinary tract infections to incontinence and impotency, occurred in 37/122 patients. Post-operative fistulae were diagnosed in

The proportion of patients having any recurrence during follow up was significantly lower after a curative resection than after a non-curative resection [25/56 vs 57/66, p!0.0001] (Table 3). After a potentially curative resection, the proportion of patients having any recurrence was 16/34 in those with an advanced primary tumour and 9/22 in those with a local recurrence (pZ0.7). After curative surgery, local recurrence was detected within a median time of 22 (3.5–55) months post-operatively in 5/34 of the patients with an advanced primary cancer and in 3/22 of the patients with a locally recurrent cancer. All but one of these patients had concomitant distant metastases. In addition, 11 patients with a primary cancer and six with a recurrence developed metastases without signs of local recurrence. After non-curative surgery, local recurrence was detected in 7/24 patients with an advanced primary tumour and in 23/42 patients with a locally recurrent tumour (pZ0.05). Distant recurrence was detected in 17/24 patients and in 22/42 patients (pZ0.014), respectively.

Table 2

Early and late complications related to surgery in patients with locally advanced rectal cancer Advanced primary

Post-operative mortality 30 days Post-operative mortality 31–90 days Early complications None Surgical complications Ileus Infections Stoma problems Other complications Late complications None Ileus Fistulas Fractures of sacrum Urological problems Defecation problems Pain Other complications

Locally recurrent

1 1

0 4

19 16 0 23 3 15

19 29 4 25 3 14

9 6 10 3 21 8 13 18

13 13 12 3 16 13 22 20

Outcome of advanced rectal cancer

731

Table 3 Outcome after surgery of patients with advanced primary rectal cancer (median follow up 63.5 (5–137) months) and of patients with locally recurrent cancer (median follow up 86 (5–143) months)

Curative No recurrence Local only LocalCmetastases Metastases only Non-curative No recurrence Local only LocalCmetastases Metastases only

Advanced primary

Locally recurrent

(nZ34) 18 0 5 11 nZ24 5 2 5 12

(nZ22) 13 1 2 6 nZ42 4 16 16 6

The overall 5-year survival was 29% in patients with an advanced primary cancer and 16% in patients with locally recurrent disease. The survival curves in patients after curative and non-curative surgery are shown in Figs. 1 and 2. The 5-year survival after a potentially curative resection was 34% in patients with an advanced primary tumour and 40% in patients with recurrent rectal cancer (pZ0.38). The corresponding figures after a noncurative resection were 24 and 4% (pZ0.01).

Discussion In this study, local control and overall survival at 5 years was better in patients with a locally advanced primary tumour than patients with a locally recurrent rectal cancer. The rate of potentially curative resections was significantly higher in patients with an advanced primary cancer (59%) than in patients with a locally recurrent cancer (34%) because macroscopic residual tumour was more common in the latter group. When curative surgery was performed the risk of local recurrence

Figure 1 Survival in patients after potentially curative surgery. Advanced primary rectal cancer (dotted line), locally recurrent rectal cancer (full line) (pZ0.38).

and metastases was similar and the survival was not significantly different. The prognostic significance of clear resection margins has been emphasised in several studies and our data support this observation.12–15 The rate of local relapse after non-curative surgery was significantly different between patients with an advanced primary tumour (29%) and patients with a local recurrence (76%). This difference is somewhat surprising, as one would assume that the majority of patients would have progressive local tumour growth if the tumour was not completely removed. There may be several reasons explaining why not all patients developed a local relapse; the number of patients with residual tumour after surgery, may have been overestimated by the surgeon, the residual tumour may have been sterilized by the pre-operative radiochemotherapy or patients may have died from distant disease before they had signs of local recurrence. As macroscopic residual tumour as assessed by the surgeon was present less frequent in patients with an advanced primary cancer than in

Figure 2 Survival in patients after potentially noncurative outcome. Advanced primary rectal cancer (dotted line), locally recurrent rectal cancer (full line) (pZ0.01).

732 patients with a local recurrence, this may partly explain the difference in local recurrence rates between the two groups. In addition, locally recurrent tumours are likely to be more locally aggressive, prone to recur and less sensitive to radiotherapy than the majority of primary tumours.16 Other institutions, using multimodality approach to treat patients with locally advanced rectal cancer, have reported varying results concerning local control and survival.8,17–19 Thus, local control from 65 to 89% and 5-year survival from 42 to 63% has been reported. Outcome is usually better in patients with advanced primary tumours than in patients with local recurrences. The differences in local control and survival are probably mainly due to selection of patients and definitions of curative surgery. Routines and frequency of follow up and the autopsy rate may also influence the rate of local control.

Multimodality treatment The use of multimodality treatment strategies, including extensive surgery, are most likely to improve the outcome in patients with locally advanced rectal cancer. In a previous study on patients with local recurrence, included in the first Stockholm Trial,20 we found that only 72 of 156 patients with local recurrence had surgery and that only 12 of these had a macroscopic complete removal of the pelvic recurrence. There was no uniform treatment protocols for patients with locally recurrent rectal cancer established at that time with the result that treatments were decided at the discretion of the surgeon. The 5-year overall survival was a dismal 3% in all patients and 7% in operated patients.21 Multimodality regimens in locally advanced rectal cancer may include pre- or post-operative radiotherapy, pre- and/or post-operative chemotherapy, extensive surgery and IORT. The optimal regimen is not yet established and may differ between patients with advanced primary tumours and patients with local recurrences.10,22– 24 To control local disease, a combination of preoperative radio- and chemotherapy followed by extensive surgery and possibly IORT in selected cases is probably the best option. However, if the patients have previously received radiotherapy they can usually not have any additional external irradiation and the regimens have to be refined. In Sweden, this is often the case in patients with a local recurrence. Thus, in this study 34% of these patients did not receive any neo-adjuvant

G. Palmer et al. treatment and recommendations for pre-operative treatment strategies in these patients are still lacking. In patients with an advanced primary tumour a combination of pre-operative radio- and chemotherapy is current standard. In our study 48% of these patients had only radiotherapy, as they were included in a randomised trial assessing radiotherapy alone vs radio-chemotherapy. The results of that trial will hopefully prove whether a combination treatment improves the rate of curative resections and survival. The role of IORT is still controversial. It may improve local control but the effect on survival is unclear.8,10,25,26 In this study we could not evaluate the effect of IORT since it was given to most patients in whom the local clearance was uncertain or negative and there was no control group.

Downsizing and surgery The necessity to perform extensive surgery in order to achieve tumour free margins is uncontroversial. However, the extent of surgery in relation to downsizing of the tumour after neo-adjuvant treatment is still debatable. Most authors argue that the surgical procedure should be planned according to the tumour extension before the preoperative treatment whilst others claim that a less extensive organ resection including a sphincter preservation approach may be justified if the tumour is significantly reduced after radio-chemotherapy or shows a complete response after radio-chemotherapy.27,28 In this study the preoperative treatment downsized 70% of the primary advanced cancers according to evaluation with MRI, yet the resections were all planned according to the tumour extension on the primary MRI. Even if local control is improved with multimodality treatment the risk of distant metastasis in patients with locally advanced rectal cancer remains a significant problem. In this study, 52% of the patients developed distant disease during follow-up which probably reflects the aggressive properties of locally advanced rectal tumours.16 Methods to identify patients with high or low risk of developing distant metastases may allow a better selection of patients for extensive treatment and thus avoid treatment affecting the quality of life negatively. More effective systemic treatments have to be developed to improve prognosis.

Complications Although a multimodality treatment strategy may improve the cure rate the treatment related

Outcome of advanced rectal cancer morbidity is substantial. In this study, the 30-day post-operative mortality was only one in 122, but more than two thirds of the patients experienced early complications. In addition, late post-operative complications and complaints were common and only 18% of the patients had no treatment related problems. Similar findings have been reported previously and in a recent publication the post-operative complication rate was 72%.29 Damage to normal tissues with nerve injuries and subsequent functional disturbances are probably inevitable when extensive pelvic surgery and radiotherapy are performed.30 However, surgery related problems like leakage, bleeding and infections may be reduced if the patients are managed in dedicated centres by experienced surgeons.

Time trend In this study, there was an obvious change in the referral patterns during the study period between 1991 and 2002. Initially, the majority of patients had locally recurrent disease but in the latter part of the period the number of patients with a local recurrence decreased while the number of patients with an advanced primary tumour increased. This probably reflects the reduced risk of local relapse after the increased use of pre-operative radiotherapy and the introduction of TME based surgery but also the increased awareness of pre-operative local staging in rectal cancer with CT scan and MRI, detecting more patients with locally advanced primary cancers. We also noted an increased proportion of curative resections of primary advanced rectal cancers during the study period. As two surgeons have performed the majority of the resections, increased surgical skill and high volume surgery may explain the improved outcome.31 However, the outcome in patients with local recurrences was unchanged.

Conclusion Locally advanced rectal cancer still constitutes a considerable challenge for both surgeons and oncologists. A meticulous patient selection and a multimodality program for pre-operative evaluation and adjuvant treatment in combination with radical surgery within dedicated centres should improve the outcome. However, distant metastases still remains the most significant problem to address in future studies.

733

Acknowledgements The study was supported by the Cancer Society, Stockholm.

References 1. Kapiteijn E, Marijnen CA, Nagtegaal ID, Putter H, Steup WH, Wiggers T, et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer. N Engl J Med 2001;345:638–46. 2. Heald RJ, Moran BJ, Ryall RD, Sexton R, MacFarlane JK. Rectal cancer: the Basingstoke experience of total mesorectal excision, 1978–1997. Arch Surg 1998;133:894–9. 3. Martling AL, Holm T, Rutqvist LE, Moran BJ, Heald RJ, Cedemark B. Effect of a surgical training programme on outcome of rectal cancer in the County of Stockholm. Stockholm Colorectal Cancer Study Group, Basingstoke Bowel Cancer Research Project. Lancet 2000;356:93–6. 4. Heslov SF, Frost DB. Extended resection for primary colorectal carcinoma involving adjacent organs or structures. Cancer 1988;62:1637–40. 5. Rich T, Gunderson LL, Lew R, Galdibini JJ, Cohen AM, Donaldson G. Patterns of recurrence of rectal cancer after potentially curative surgery. Cancer 1983;52:1317–29. 6. Polk Jr HC, Spratt Jr JS. The results of treatment of perineal recurrence of cancer of the rectum. Cancer 1979;43:952–5. 7. Yeung RS, Moffat FL, Falk RE. Pelvic exenteration for recurrent and extensive primary colorectal adenocarcinoma. Cancer 1993;72:1853–8. 8. Farouk R, Nelson H, Gunderson LL. Aggressive multimodality treatment for locally advanced irresectable rectal cancer. Br J Surg 1997;84:741–9. 9. Frykholm GJ, Pahlman L, Glimelius B. Combined chemo- and radiotherapy vs. radiotherapy alone in the treatment of primary, nonresectable adenocarcinoma of the rectum. Int J Radiat Oncol Biol Phys 2001;50:427–34. 10. Tveit KM, Wiig JN, Olsen DR, Storaas A, Poulsen JP, Giercksky KE. Combined modality treatment including intraoperative radiotherapy in locally advanced and recurrent rectal cancer. Radiother Oncol 1997;44:277–82. 11. Rutten HJ, Mannaerts GH, Martijn H, Wiggers T. Intraoperative radiotherapy for locally recurrent rectal cancer in The Netherlands. Eur J Surg Oncol 2000;26(Suppl A):S16–S20. 12. Quirke P, Durdey P, Dixon MF, Williams NS. Local recurrence of rectal adenocarcinoma due to inadequate surgical resection. Histopathological study of lateral tumour spread and surgical excision. Lancet 1986;2:996–9. 13. Hall NR, Finan PJ, Al-Jaberi T, Tsang CS, Brown SR, Dixon MF, et al. Circumferential margin involvement after mesorectal excision of rectal cancer with curative intent. Predictor of survival but not local recurrence? Dis Colon Rectum 1998;41: 979–83. 14. Wibe A, Rendal PR, Svensson E, Norstein J, Eide TJ, Myrvold HE, et al. Prognostic significance of the circumferential resection margin following total mesorectal excision for rectal cancer. Br J Surg 2002;89:327–34. 15. Cawthorn SJ, Parums DV, Gibbs NM, A’Hern RP, Caffarey SM, Broughton CI, et al. Extent of mesorectal spread and involvement of lateral resection margin as prognostic factors after surgery for rectal cancer. Lancet 1990;335:1055–9. 16. van den Brink M, Stiggelbout AM, van den Hout WB, Kievit J, Klein Kranenbarg E, Marijnen CA, et al. Clinical nature and

734

17.

18.

19.

20.

21.

22. 23.

24.

prognosis of locally recurrent rectal cancer after total mesorectal excision with or without preoperative radiotherapy. J Clin Oncol 2004;22:3958–64. Mannaerts GH, Rutten HJ, Martijn H, Hanssens PE, Wiggers T. Comparison of intraoperative radiation therapycontaining multimodality treatment with historical treatment modalities for locally recurrent rectal cancer. Dis Colon Rectum 2001;44:1749–58. Tveit KM, Wiig JN, Olsen DR, Storaas A, Poulsen JP, Giercksky KE. Combined modality treatment including IORT in locally advanced and recurrent rectal cancer: results from a prospective Norwegian study. Front Radiat Ther Oncol 1997;31:221–3. Nakfoor BM, Willett CG, Shellito PC, Kaufman DS, Daly WJ. The impact of 5-fluorouracil and intraoperative electron beam radiation therapy on the outcome of patients with locally advanced primary rectal and rectosigmoid cancer. Ann Surg 1998;228:194–200. Stockholm Rectal Cancer Study Group. Preoperative shortterm radiation therapy in operable rectal carcinoma. A prospective randomized trial. Cancer 1990;66:49–55. Holm T, Cedermark B, Rutqvist LE. Local recurrence of rectal adenocarcinoma after ‘curative’ surgery with and without preoperative radiotherapy. Br J Surg 1994;81:452–5. Glimelius B. Chemoradiotherapy for rectal cancer—is there an optimal combination? Ann Oncol 2001;12:1039–45. Glimelius B. Recurrent rectal cancer. The pre-irradiated primary tumour: can more radiotherapy be given? Colorectal Dis 2003;5:501–3. Wiggers T, Mannaerts GH, Marinelli AW, Martijn H,

G. Palmer et al.

25.

26.

27.

28.

29.

30.

31.

Rutten HJ. Surgery for locally recurrent rectal cancer. Colorectal Dis 2003;5:504–7. Ratto C, Valentini V, Morganti AG, Barbaro B, Coco C, Sofo L, et al. Combined-modality therapy in locally advanced primary rectal cancer. Dis Colon Rectum 2003;46:59–67. Sadahiro S, Suzuki T, Ishikawa K, Fukasawa M, Saguchi T, Yasuda S, et al. Preoperative radio/chemo-radiotherapy in combination with intraoperative radiotherapy for T3-4Nx rectal cancer. Eur J Surg Oncol 2004;30:750–8. Habr-Gama A, Perez RO, Nadalin W, Sabbaga J, Ribeiro Jr U, Silva e Sousa Jr AH, et al. Operative versus nonoperative treatment for stage 0 distal rectal cancer following chemoradiation therapy: long-term results. [discussion 717–8] Ann Surg 2004;240:711–7. Habr-Gama A, Perez RO, Kiss DR, Rawet V, Scanavini A, Santinho PM, et al. Preoperative chemoradiation therapy for low rectal cancer. Impact on downstaging and sphinctersaving operations. Hepatogastroenterology 2004;51:1703–7. Bakx R, van Tinteren H, van Lanschot JJ, Zoetmulder FA. Surgical treatment of locally recurrent rectal cancer. Eur J Surg Oncol 2004;30:857–63. Mannaerts GH, Schijven MP, Hendrikx A, Martijn H, Rutten HJ, Wiggers T. Urologic and sexual morbidity following multimodality treatment for locally advanced primary and locally recurrent rectal cancer. Eur J Surg Oncol 2001;27:265–72. Martling A, Cedermark B, Johansson H, Rutqvist LE, Holm T. The surgeon as a prognostic factor after the introduction of total mesorectal excision in the treatment of rectal cancer. Br J Surg 2002;89:1008–13.