Short-course preoperative radiotherapy with delayed surgery in rectal cancer – A retrospective study

Radiotherapy and Oncology 87 (2008) 343–349 www.thegreenjournal.com

Rectal cancer

Short-course preoperative radiotherapy with delayed surgery in rectal cancer – A retrospective study ˚ ke Berglunda, Lars Pa Calin Radua,*, A ˚hlmanb, Bengt Glimeliusa,c a Department of Oncology, Radiology and Clinical Immunology, and bDepartment of Surgery, Uppsala University, Sweden, cDepartment of Pathology and Oncology, Karolinska Institutet, Stockholm, Sweden

Abstract Purpose: In the most advanced, non-resectable primary rectal cancers, conventional long-course radiotherapy (RT) (1.8–2 Gy · 25–28), frequently combined with chemotherapy, has been used since tumour regression is needed in order to allow a radical (R0) resection. In Uppsala, short-course 5 · 5 Gy with planned delayed surgery has been used in patients with contraindications to long-course RT (±chemotherapy). The aim is to describe our experience of using this approach in patients not eligible for standard treatment. Patients and methods: During 2002 and 2005, 46 patients with non-resectable rectal cancer (±synchronous distant metastases) were treated with 5 · 5 Gy and delayed surgery if possible. The clinical records were retrospectively evaluated. The first group (A) had no metastases (T4NXM0), whereas the other two groups (B + C) had metastases (T4NXM1). In group (B), the patients had predominantly loco-regional disease and were not candidates for combination chemotherapy (high age, co-morbidities), and in group (C) up-front combination chemotherapy was given, with the intention to have surgery of both the primary and the secondaries if sufficient regression at both sites were seen. Results: The patients in the first two groups (A + B) were old (median 79 and 76 years, respectively), and had several comorbidities. In group (C), median age was 63 years. The 5 · 5 Gy RT was well tolerated by most patients, but grade IV diarrhoea was recorded in three elderly patients. One patient in the group (C) died from neutropenic fever. Many patients were reported to have less local symptoms after the treatment given. Delayed surgery was performed in all but nine patients. Radical surgery (R0 + R1) was performed in 22 (92%) (group A), 4 (44%) (group B), and 6 (46%) (group C) patients, respectively. A pCR was seen in four patients (two in group A and two in group C). No postoperative deaths occurred. Conclusions: Considering the very high age and presence of co-morbidity, the 5 · 5 Gy schedule is well tolerated. Further, considering the very advanced local stage, the schedule has considerable anti-tumour activity and can result in radical surgery in a high proportion of patients. c 2007 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 87 (2008) 343–349.



Keywords: Rectal cancer; Radiotherapy; Non-resectable; 5 · 5 Gy; Delay surgery

Preoperative radiotherapy (RT), alone or with chemotherapy, is an established treatment for many patients with primary rectal cancer [1–3]. In tumours judged to be resectable, it is primarily used to lower the risk of local failure. For this purpose, short-course RT (5 · 5 Gy) followed by immediate surgery has been extensively used. Based upon the results of several randomised trials, the local failure risk is at least halved [4–6]. In the more advanced, non-resectable cancers, conventional long-course RT (1.8–2 Gy · 25–28), frequently combined with chemotherapy [7], has been used aiming at tumour regression in order to safely allow a radical (R0) resection. This long-course treatment is frequently also used in less advanced cancers at many centres [9,10].



In a few patients irradiated preoperatively with 5 · 5 Gy it was found that surgery was not possible the week after the treatment, whereas this was the case several weeks later, likely caused by tumour regression in the waiting period. This experience has, apart from a report of two cases [11], to our knowledge, not been published in a scientific article. The experience was one rationale behind the initiation of an ongoing clinical trial, the Stockholm III trial, where patients with resectable rectal cancer are randomised to preoperative 5 · 5 Gy with either immediate or delayed (4–7 weeks) surgery. The patients can also be randomised to a third group, 2 Gy · 25 with delayed surgery [8].

0167-8140/$ - see front matter c 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2007.11.025

344

Radiotherapy and surgery in rectal cancer

The present retrospective study describes our experience of using 5 · 5 Gy with delayed surgery in patients not eligible for any standard type of treatment.

Patients and methods Patients The Department of Oncology in Uppsala is the only radiation department in the counties of Uppsala and Dalarna (total population 560,000). Besides occasional patients from Dalarna who received radiotherapy in the county of Ga ¨strikland because of waiting lists in Uppsala, all patients with rectal cancer received radiation in Uppsala. Between January 2002 and December 2005, 247 patients with rectal cancer have been irradiated, 134 received 5 · 5 Gy with immediate surgery, 70 received 5 · 5 Gy with delayed surgery and 43 received 1.8–2 Gy to about 50 Gy (±chemotherapy) (Fig. 1). Of the patients who received 5 · 5 Gy with delayed surgery, 21 patients, all with a resectable cancer, were randomised within the Stockholm III trial and 49 were treated outside the trial. Three of these had a resectable cancer (cT2-3NXM0), but were for various reasons treated outside routines (one patient had two other simultaneously detected cancers, one was an old professor who could not accept surgery prior to an important scientific event 2

Preoperatively irradiated patients 247 5 x 5 Gy with immediate surgery

134 113

1.8 - 2 Gy x 25 - 28 ± chemotherapy

43 5 Gy x 5 with delayed surgery 70

Resectable cancer, treated within the Stockholm III trial

21 49

Resectable cancer, treated outside trial (see text)

3 Non-resectable cancers 46

24 Group A non-metastatic (cT4NxMx) contraindications to RTCT

9 Group B metastatic disease (cT4NxM1) palliative intent with contraindications to RTCT and palliative systemic chemotherapy

13 Group C metastatic disease (cT4NxM1), primarily treated with combination chemotherapy, RT given before later surgery

Fig. 1. Flowchart of patients with rectal cancer treated with preoperative irradiation in Uppsala between January 2002 and December 2005.

months later and one had a psychiatric diagnosis). All three patients later had surgery and two of them had a pathological complete remission (pCR). The remaining 46 patients, constituting the population in this study, had an inextirpable primary (T4). The clinical records were retrospectively evaluated for the purpose of the study. Last follow-up was June 16, 2007. Median follow-up from diagnosis was then 26 (range 7–51) months for group A, 10 (range 2–27) months for group B and 21 (range 6–35) months for group C.

Reasons for 5 · 5 Gy with delayed surgery The reasons why the 46 patients were treated with 5 · 5 Gy plus delayed surgery fall into three groups. The first group (A, n = 24) should routinely have received long-course RT (±chemotherapy) for a non-extirpable, non-metastatic rectal cancer (T4NXM0), but the patients were for various reasons not candidates for this therapy. Of the 24 patients in group A, 13 had a very high age (chronological and biological above 80 years) with some co-morbidity, 7 had a high age and severe physical co-morbidity and 4 had a severe psychiatric and/or a social situation preventing prolonged RT (±chemotherapy). In this group, the treatment intent was curative. In the two other groups (B + C), distant metastases were present (T4NXM1). In group B (n = 9), the patients had predominantly loco-regional disease and were not candidates for combination chemotherapy, usually because of high age and co-morbidities. These patients received 5 · 5 Gy for strictly palliative reasons where surgery of the primary could possibly be done after restaging 6–8 weeks later. Reasons to do surgery were then that the tumour regression allowed local R0 surgery and that the distant metastases did not contraindicate rectal surgery, i.e. the life expectancy was reasonably long. Some of these patients could potentially have had surgery for the primary but this was not done because of an excellent local symptomatic response and a rather short expected survival time. The patients in the third group (C, n = 13) had a primary T4 tumour with synchronous distant metastases and were candidates for up-front combination chemotherapy, a few with the intention to have surgery of both the primary and the secondaries if sufficient regressions at both sites were seen. In patients, where surgery of the primary tumour was indicated, 5 · 5 Gy was given with a delay before the surgery. The radiation was then to be given at least 7 days after the last day of chemotherapy administration. In the patients where curative surgery of the metastases was potentially possible because of tumour regression, the intention was also to give two cycles of chemotherapy between the end of the RT and the surgery. The chemotherapy was usually a 5-FU/leucovorin/oxaliplatin combination (Nordic FLOX [12], or FOLFOX-4 [13]).

Staging With the exception of one patient in group A and one in group B, all patients were adequately staged with pelvic (35 MRI, 9 CT), lung (chest X-ray or computed tomography, CT) and liver imaging (usually CT), respectively, at diagnosis. Patients in group C were re-staged every two months during the chemotherapy. All patients were re-staged 4–6

C. Radu et al. / Radiotherapy and Oncology 87 (2008) 343–349

weeks after the end of the RT. For some patients in group B, this staging was mainly clinical, but in most patients imaging was re-done prior to the decision of whether surgery should be done or not. Most assessments were made at a multidisciplinary team conference.

Radiation treatment target The RT was given with 5 Gy in five fractions during five consecutive work days according to the department routines. With the exception of a few patients in groups A + B, all patients underwent dose-planning CT, and an individualized clinical target volume (CTV) was done, including the gross tumour volume with margins (2–3 cm depending upon tumour direction) and regional lymph nodes depending upon tumour location. For most patients, all lymph nodes in the dorsal pelvis up to the level of S1 were included. Lymph nodes in the groins, along the external iliac vein and the common iliac veins, were included only if at risk. A planning target volume (PTV) was drawn by adding 1 cm to CTV. Typically, the upper border of the beams was at the lower part of L5, the lower border either below or 3–4 cm above the anal verge depending upon tumour height, the lateral borders of the anterior–posterior beams at least 1 cm outside the pelvic rim, the dorsal border of the lateral beams posterior to sacrum and the ventral border at least 2.5 cm in front of the promontory. Multi-leaf collimators were usually used to shape the beams. Anterior–posterior beams alone were used in a few patients where the tumour extent ante-

345

riorly was extensive, otherwise a four-beam box technique was used.

Descriptive statistics Survival curves were constructed using the Kaplan–Meier method. The study was approved by the Research Ethics Committee, University of Uppsala, Uppsala, Sweden.

Results Patient characteristics and treatment Patient characteristics at diagnosis in the three groups are shown in Table 1. Patients in groups A + B were for obvious reasons older than in group C. In these groups, all patients under the age of 70 had definite contraindications, including in a few cases very special psychosocial situations, preventing the standard prolonged RT or adequate chemotherapy. None of the patients in groups A + B had any therapy prior to the 5 · 5 Gy, whereas this was given to all in group C. The interval from the diagnosis (date of biopsy) to the RT was thus much longer in group C. Some of those patients had several lines of chemotherapy before surgery, and thus preoperative RT was indicated. One or usually two cycles of chemotherapy were given to six of the nine operated patients in group C during the interval between the end of RT to the planned surgery, to further influence the growth of the distant metastases.

Table 1 Patient characteristics, treatment and results of surgery Group

Clinical stage Age mean (years) (range) Sex male/female Haemoglobin at diagnosis mean (range) Prior chemotherapy Oxaliplatin/FU Irinotecan/FU FU alone Interval day of biopsy to RT mean (days) (range) Chemotherapy after RT but before surgery Interval RT to surgery mean (days) (range) Surgery R0 R1 R2 None Pathology, local stage 0 (pCR) I II III–IV Max MRI length at diagnosis, mean (mm) (range) Max tumour diameter at pathology mean (mm) (range) Metastatic surgery a b c

Group A: non-metastatic, contraindications to RTCT. Group B: elderly, metastatic disease, palliative intent. Group C: metastatic disease, potentially curative intent.

A (n = 24)a

B (n = 9)b

C (n = 13)c

T4NXM0 79 (57–97) 10/14 122 (86–152) 0

T4NXM1 76 (59–87) 5/4 130 (114–159) 0

52 (28–88)

40 (21–70)

49 (30–109)

38 (31–51)

T4NXM1 63 (43–76) 7/6 121 (101–139) 13 11 4 2 216 (54–316) 6 50 (38–67)

21 1 1 1

5

6

4

3 4

2 2 12 7 68 (40–110) 4 missing 29 (0–60) 3 missing

2 3 2 70 (30–130) 4 missing 32 (10–50) 5 missing

4 3 70 (45–100) 3 missing 26 (0–70) 5 missing 2

346

Radiotherapy and surgery in rectal cancer

Surgery and follow-up

Time free of local failure

Overall Survival Complete

Censored

Cumulative Proportion Surviving

1.0 0.9 0.8 0.7 0.6 0.5 0.4

Group A (n=24) Group B (n=9) Group C (n=13)

0.3 0.2 0.1 0.0

0

6

12

18

24

30

36

42

48

54

Time, months

Fig. 2. Overall survival from the radiation therapy (5 · 5 Gy) in the three groups; A: T4NXM0 with contraindications to radiochemotherapy, (B: T4NXM1 treated for palliation with contraindications to radiochemotherapy and combination chemotherapy, and C: T4NXM1, treated up-front with combination chemotherapy and later with 5 · 5 Gy.

Complete

Cumulative Proportion Surviving

All patients were assessed for potential surgery about 4 – 6 weeks after the last RT fraction. This was made only clinically in a few patients in groups B and C because of obvious progression of distant metastases with no local symptoms prompting surgery. In group A, where patients had no distant metastases at diagnosis, surgery was considered indicated in all but one patient. The exceptional patient had a still non-resectable tumour and she died after 15 months, with no prominent local symptoms. Of the 23 operated patients, 10 had an abdomino-perineal resection (APR), 9 had a low anterior resection (AR) and 4 had a Hartmann procedure. An R0 resection was possible in 21 patients and a pCR was seen in 2 (9%) patients. The stage distribution is shown in Table 1 in the group of R0-resected patients. One patient has had a clinically verified local recurrence. Eight patients have died, but only two of rectal cancer. Overall survival is shown in Fig. 2 and time free from failing locally in Fig. 3. In group B, where all patients were treated with palliative intent, surgery was considered possible and indicated in five patients (one APR, one AR and three Hartmann). All these patients had locally an R0 resection. One of them had symptoms of local failure before death. All nine patients have died from rectal cancer with a symptomatology indicating progression of systemic disease in most of them. Only one received palliative chemotherapy. Median survival time was 10 (range 2–27) months (Fig. 2). In group C, where all patients had metastatic disease and were treated with up-front chemotherapy, surgery was considered possible and indicated in nine patients, whereas three patients did not qualify (because of no local symptoms and progressive/extensive systemic disease) at the restaging 4–6 weeks after the RT. Of the nine operated patients (one APR, five AR and two Hartmann), an R0 resection was possible in six. None of the three patients with an R2 resection had extensive, mutilating local surgery because of the

Censored

1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3

Group A (n=24) Group B (n=9) Group C (n=13)

0.2 0.1 0

6

12

18

24

30

36

42

48

54

60

Time (months)

Fig. 3. Time free of failing locally in the three groups; A: nonmetastatic with contraindications to radiochemotherapy, B: elderly, metastatic disease, palliative intent, and C: metastatic disease, potentially curative intent. No resection or an R2 resection indicates that the patient did not become local tumour-free.

distant metastases. Two of the six R0-resected tumours showed a pCR. Subsequent metastatic surgery was possible in two patients, one of which is recurrence-free 31 months after diagnosis. In this group, 10 patients have died after median 19 (range 6–35) months from diagnosis, and three are alive (one recurrence-free), all without local symptoms after median 31 (range 27–35) months. Overall survival is shown in Fig. 2 and time free of failing locally in Fig. 3.

Toxicity to 5 · 5 Gy In group A, 1/24 patients was admitted to hospital because of diarrhoea. This 84-year-old lady had previously been irradiated for a cervical cancer and had diarrhoea already at the start of her RT. In group B, 2/9 patients also had grade IV diarrhoea. They were both above 80 years of age. In these patients, the diarrhoea resolved within 7–12 days. Diarrhoea grade III was recorded in one patient. In group C, one patient died from neutropenic fever 4 days after the last RT fraction. She had received 10 courses of Nordic FLOX with little toxicity, but was neutropenic for the first time after her last treatment prior to the RT. The RT was given without problems, but she complained of abdominal pain 2 days after the last RT. She comes to the hospital on day 3 and has subnormal leucocytes (2.6 · 109/ L), fever and dies on day 4 in what is interpreted as abdominal sepsis, but was not given any specific treatment. Apart from the toxicity described above, no other information about adverse events was noted in the clinical records. Rather, many patients were reported to have less local symptoms (pain, bleeding, stenosis and mucous production) from their rectal cancer. The outcome of the surgery was good in all cases, although, as can be expected in an elderly population with several co-morbidities, the postoperative period contained several complications [one myocardial infarction, one unstable angina, one stroke, three paralytic ileus, three reoperations (bleeding, stoma revision, small bowel perforation)]. The median postoperative hospital stay period

C. Radu et al. / Radiotherapy and Oncology 87 (2008) 343–349

was 16 (range 6–28) days. No postoperative deaths occurred.

Discussion This retrospective evaluation shows that the short-course 5 · 5 Gy schedule can result in substantial down-staging in patients with a locally advanced non-resectable T4 rectal cancer and an R0 resection can be achieved in a majority of the patients at surgery median 6–7 weeks after the RT. Standard treatment for these patients is long-course radiochemotherapy (RTCT) with delayed surgery, but this preoperative treatment was contraindicated in all patients. In this group of very high-risk patients, 5 · 5 Gy is thus an attractive alternative, to be used in routine care. In the light of its apparently high efficacy and potentially very low toxicity, it may, however, also be an attractive alternative to explore further in more patients with these most locally advanced tumours. A tumour infiltrating an adjacent organ or structure (T4) so that it is considered ‘non-resectable’ constitutes about 10–15% of all newly diagnosed rectal cancers, half of which do not have any detectable distant metastases at diagnosis. A much larger fraction of the rectal cancers are ‘locally advanced’, generally meaning clinical stage T3 or T4 or node positive (cT3-4, any cTN+), although this proportion is not precisely known. Better imaging, particularly using MRI, will in the future more objectively stage the rectal cancers [19]. The use of 5 · 5 Gy with immediate surgery in ‘locally advanced, but resectable’ rectal cancer is an established treatment at many centres world-wide, whereas other centres in these patients use RTCT (1.8–2 Gy · 25-29 with 5FU alone, or more lately also with another drug such as oxaliplatin). It is not the scope of this study to discuss the relative merits of the short-course or the long-course schedules, although the only study directly comparing them preoperatively found the same overall and disease-free survival, a tendency to better local control and less toxicity using the short-course schedule compared with RTCT [14]. The use of 5 · 5 Gy with immediate surgery in patients with a nonresectable cancer is not recommended since tumour regression is per definition required in order to allow subsequent R0 surgery, even if it could be argued that more extensive surgery will result in an R0 resection in some of the patients. The morbidity of such a procedure can, however, be high. The experience of using 5 · 5 Gy with delayed surgery, allowing time for tumour regression, as reported here, is based upon a retrospective study, having a low level of scientific evidence (level IV according to ASCO criteria.[15]). Since previously only case reports have been available (level V [11]) it could be claimed that the scientific evidence for its use has increased. Attaining an R0 resection in 21/24 patients not possible to treat with conventional RTCT indicates to us that in this group, 5 · 5 Gy with delayed surgery is a new and valid treatment for routine use. Since the treatment efficacy appears comparable to what RTCT can achieve in similar [16] or less advanced cancers, a randomised study comparing these two options also in patients suitable for RTCT is attractive. The advantage of 5 · 5 Gy is its ease and low toxicity. A disadvantage is that it cannot be

347

combined with concomitant chemotherapy. However, the incremental effect of adding 5FU to long-course RT in rectal cancer is limited, and has at least not up to now improved survival [9,10,16]. The addition of other drugs may result in further improvements, but this is not yet evidence-based. Further, if the relative cell kill effects of 5 · 5 Gy are higher than those of 25 · 2 Gy, which data from the trials indirectly show [17,18], the addition of chemotherapy only compensates for a less efficient radiation schedule in rectal cancer. The use of 5 · 5 Gy with delayed surgery in the group of patients with non-resectable primary and distant metastases not suitable for any intensive anti-tumour therapy also appears attractive, even if this retrospective evaluation only included nine patients. However, these patients had an excellent response both symptomatically and objectively. The survival in this group is short, prompting short treatments. Concerns of late toxicity are also irrelevant. Some of these patients had severe diarrhoea, albeit temporary. This can likely be reduced if the radiation target is limited to the gross tumour volume, excluding any prophylactic radiation to regional lymph nodes. 5 · 5 Gy with delayed surgery is attractive also in the third group of patients with non-resectable primary tumour and distant metastases, but without contraindications to intensive tumour therapy. The treatment of these patients is controversial, and our approach will likely not be immediately accepted by many clinicians. The group is clinically very heterogeneous. Some may have a rather asymptomatic primary whereas others may have an extensive primary with severe symptoms of pain, bleeding, soiling and stenosis. This can be combined with either limited systemic disease burden with one or a few small liver or lung metastases, or extensive metastatic disease with weight loss, night sweats and general deterioration. Depending upon this pattern, these patients will likely be treated either with upfront combination chemotherapy and then, dependent upon tumour response, with the loco-regional treatment or with RTCT first, surgery if indicated and then chemotherapy for palliation or for cure if the metastases can be resected after tumour regression. All patients in this third group were treated with up-front chemotherapy, either because the systemic disease dominated the symptomatology or because the systemic disease, even if limited, constituted the greatest threat to the patients’ lives. Combination chemotherapy (with or without targeted drugs) is then the therapy having the greatest tumour cell kill. The chemotherapy that can be given concomitantly with radiation (1.8–2 Gy to 45–50 Gy) must be dose reduced by about 20% compared to when given alone [3], and thus result in less cell kill effect systemically. After 5–6 weeks of RTCT, a treatment-free interval of 4–6 weeks before surgery is also necessary. We hypothesized that after standard chemotherapy (e.g. fortnightly FLOX/FOLFOX ± bevacizumab), the systemic drug treatment can be given with higher dose intensity when integrating 5 · 5 Gy with delayed surgery than can be achieved using conventional RTCT (Fig. 4). Highest possible dose intensity is particularly important in individuals where down-sizing of distant metastases is required for subsequent R0 surgery. We had the intent to apply the alternative approach with curative intent in seven patients. One of them died during

348

Radiotherapy and surgery in rectal cancer RTCT week 1 Folfox C

week 2

week 3 Folfox C

week 4 week 5

week 6 week 7 week 8 week 9 week 10 week 11 week 12 week 13 week 14 week 15 Xeloda 825 mg/m² x 2 daily CCCCC CCCCC CCCCC CCCCC CCCCC CCC Surgery RRRRR RRRRR RRRRR RRRRR RRRRR RRR 28 x 1,8 Gy

Chemotherapy + short course RT week 1 Folfox C

week 2

week 3 Folfox C

week 4 week 5

week 6

week 7 Folfox C

week 8

week 9 Folfox C

week 10 week 11 week 12 week 13 week 14 week 15 Surgery

RRRRR 5 x 5 Gy C – chemotherapy, R - radiotherapy

Fig. 4. Schematic illustration of the conventional (above) and alternative (below) treatments in a patient with a locally advanced nonresectable (T4) rectal cancer with synchronous distant metastases in which up-front combination chemotherapy (such as the FOLFOXregimen ± an antibody) is decided upon. Note that only the two last chemotherapy cycles before the radiotherapy (RT) out of an individual number of cycles are indicated in the figure. Prior to surgery of the primary, when this is considered indicated, radiochemotherapy (RTCT) is conventionally indicated. In the alternative approach, the loco-regional treatment is 5 · 5 Gy with a planned delay before surgery. In this delay, further chemotherapy could be given to maximize treatment of the distant metastases.

therapy but six could fulfil the planned treatment without excessive toxicity. Surgery of the primary was possible in all, but metastatic surgery in only two due to insufficient response in the others. This supports the notion that the systemic disease poses the largest threat to the patients. The experience reported here is very limited, but based upon a theoretically interesting concept, and preliminarily favourable. Many details need more explorations, such as the minimal interval between chemotherapy and radiation. The toxic death, even if potentially caused by several mistakes, illustrates one of many uncertainties. Due to the retrospective nature of our study, toxicity evaluation is dependent upon information from the clinical records. Based upon prospective recordings, including quality-of-life questionnaires during and after the radiotherapy and prior to surgery, in the ongoing randomised study in resectable rectal cancers, the 5 · 5 Gy schedule has few adverse effects (Stockholm Rectal Cancer Group, unpublished information). Still, in a high-risk elderly population like the one treated here, special precautions must be taken. In the oldest population and in those with metastatic disease, the target volume could be reduced cranially. Further, prophylactic irradiation of lymph node stations is not logical if distant metastases are present. We did not pay sufficient attention to this when treating these patients. Furthermore, they are now informed about the risk of diarrhoea, prescribed loperamid and a nurse has telephone contact with the patient afterwards. In conclusion, a retrospective evaluation of a treatment policy utilizing the short-course 5 · 5 Gy schedule in selected patients with a locally advanced, non-resectable rectal cancer, introducing a delay prior to the surgery, shows that the approach is highly active and, although a few cases of serious toxicity were seen, has low toxicity in general. The approach is currently being evaluated in an ongoing randomised study in patients with resectable rectal cancer. More than 400 patients have been randomised up to October 2007. Formal evaluations in other patient groups are prompted.

* Corresponding author. Calin Radu, Department of Oncology, Radiology and Clinical Immunology, Uppsala University Hospital, SE751 85 Uppsala, Sweden. E-mail address: calin.radu@onkologi. uu.se

Received 19 October 2007; received in revised form 23 November 2007; accepted 25 November 2007; Available online 21 February 2008

References [1] Frykholm G, Glimelius B, Pa ˚hlman L. Preoperative irradiation with and without chemotherapy (MFL) in the treatment of primary non-resectable adenocarcinoma of the rectum. Results from two consecutive studies. Eur J Cancer Clin Oncol 1989;25:1535–41. [2] Sauer R, Becker H, Hohenberger W, et al. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med 2004;351:1731–40. [3] Sebag-Montefiore D. Developments in the use of chemoradiotherapy in rectal cancer. Colorectal Dis 2006;8:14–7. [4] Swedish Rectal Cancer Trial. Improved survival with preoperative radiotherapy in resectable rectal cancer. N Engl J Med 1997;336:980–7. [5] Kapiteijn E, Marijnen CAM, Nagtegaal ID, et al. Preoperative radiotherapy in combination with total mesorectal excision improves local control in resectable rectal cancer. Report from a multicenter randomized trial. New Engl J Med 2001;345:638–46. [6] Peeters KCMJ, Marijnen CAM, Nagtegaal ID, et alFor the Dutch Colorectal Cancer Group. The TME trial after a median followup of 5 years: increased local control but no survival benefit in irradiated patients with resectable rectal carcinoma. Ann Surg 2007;246:693–701. [7] Glimelius B, Gro ¨nberg H, Ja ¨rhult J, Wallgren A, Cavallin-sta ˚hl E. A systematic overview of radiation therapy effects in rectal cancer. Acta Oncol 2003;42:476–92. [8] Glimelius B. Rectal cancer irradiation. Long course, short course or something else? Acta Oncol 2006;45:1013–7. [9] Gerard JP, Conroy T, Bonnetain F, et al. Preoperative radiotherapy with or without concurrent fluorouracil and leucovorin in T3-4 rectal cancers: results of FFCD 9203. J Clin Oncol 2006;24:4620–5.

C. Radu et al. / Radiotherapy and Oncology 87 (2008) 343–349 [10] Bosset JF, Collette L, Calais G, et al. Chemotherapy with preoperative radiotherapy in rectal cancer. N Engl J Med 2006;355:1114–23. [11] Widder J, Herbst F, Scheithauer W. Preoperative sequential short-term radiotherapy plus chemotherapy can induce complete remission in T3N2 rectal cancer. Acta Oncol 2005;44:921–3. ˚, et al. Multicentre phase II [12] Sørbye H, Glimelius B, Berglund A study of Nordic 5-fluorouracil/leucovorin bolus schedule combined with oxaliplatin (FLOX) as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2004;22:31–8. [13] 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;18:2938–47. [14] Bujko K, Nowacki MP, Nasierowska-Guttmejer A, et al. Longterm results of a randomised trial comparing preoperative short-course radiotherapy vs preoperative conventionally frac-

[15]

[16]

[17]

[18] [19]

349

tionated chemoradiation for rectal cancer. Br J Surg 2006;93:1215–23. Desch CE, Benson III AB, Smith TJ, et al. Recommended colorectal cancer surveillance guidelines by the American Society of Clinical Oncology. J Clin Oncol 1999;17:1312–21. ˚, et al. A randomized Braendengen M, Tveit KM, Berglund A phase III study (LARCS) comparing preoperative radiotherapy alone versus chemoradiotherapy in non-resectable rectal cancer. Eur J Cancer 2005; ECCO 13. Abstr 612. Courdi A. Fractionation sensitivity and equivalent doses. Commenting the Editorial by Glimelius. Acta Oncol 2007;46:395–6 [author reply 396]. Glimelius B, Isacsson U. In reply to Dr. Courdi. Acta Oncol 2007;46:396. Smith N, Brown G. Preoperative staging of rectal cancer. Acta Oncol 2008;47:1–12.