18 Applications in Rectal and Anal Cancer
Claus Rödel and Rolf Sauer
C. Rödel, MD
Klinik and Poliklinik für Strahlentherapie, Universitätsklini- kum Erlangen, Postfach 2306, Universitätsstrasse 27, 91054 Erlangen, Germany
R. Sauer, MD
Professor, Direktor der Strahlenklinik, Universitätsklinikum Erlangen, Postfach 2306, Universitätsstrasse 27, 91054 Erlan- gen, Germany
18.1
Rectal Cancer 18.1.1
Introduction
The rational for using combinations of radiation and systemic chemotherapy as a component of adjuvant treatment in stage T3/4 and/or node-positive rectal cancer (UICC stages II and III) is based on the risk of relapse after surgery alone and the evidence of radio- and drug-responsiveness derived from both laboratory studies and clinical trials. The past four decades have witnessed the development of a vari- ety of preoperative and postoperative radio- and radiochemotherapy schedules designed to optimize the sequence of treatment modalities and the most appropriate scheduling of irradiation and 5-fluorou- racil-based chemotherapy. Given that with optimized local treatment, including preoperative radiotherapy and total mesorectal excision (TME) surgery distant metastasis is by far the predominate pattern of tumor failure in rectal cancer presently, the future chal- lenge is to integrate more effective systemic therapy into the multimodal concepts for this disease.
18.1.2
5-Fluorouracil-Based Radiochemotherapy 5-fluorouracil (5-FU), an analog of the pyrimidine uracil with a fluorine atom substituted in place of hydrogen at the carbon 5 (C-5) position, has been the most commonly used single chemotherapeutic agent for colorectal cancer during the past four decades, and will certainly also continue to be the backbone of modern drug combinations in the near future. Since its synthesis in 1957 by Heidelberger et al. (1957), the metabolism and mechanism of action of 5-FU have been studied in detail. 5-FU enters a complex anabolic process that accounts for cytotoxicity at the cellular level by interfering with normal DNA and RNA func- tion. Heidelberger et al. (1958) also initially discov-
CONTENTS
18.1 Rectal Cancer 267 18.1.1 Introduction 267
18.1.2 5-Fluorouracil-Based Radiochemotherapy 267 18.1.2.1 Randomized Trials of
Postoperative Concurrent RCT 268 18.1.2.2 Randomized Trials to Optimize
5-FU-Based Postoperative RCT 269
18.1.2.3 Randomized Trials to Optimize the Sequence:
Preoperative RCT 270
18.1.2.4 Concomitant Chemotherapy with Preoperative Radiation Therapy? 271
18.1.2.5 Conclusions from Trials with 5-FU-Based RCT 272 18.1.3 Integrating Novel Chemotherapeutic Agents into
Preoperative Combined Modality Treatment 273 18.1.3.1 Oral Fluoropyrimidines 274
18.1.3.2 Oxaliplatin 276 18.1.3.3 Irinotecan 277 18.1.3.4 Targeted Therapy 277
18.1.4 Future Challenges for Combined-Modality Rectal Cancer Treatment 278
18.2 Anal Cancer 279 18.2.1 Introduction 279
18.2.2 Randomized Trials of RCT in Anal Cancer 279 18.2.3 Current Randomized Trails to
Optimize Concurrent RCT 280
18.2.4 Future Directions in Anal Cancer Treatment with New Agents 281
References 281
ered that the addition of 5-FU to radiation in rodent tumors markedly enhanced the effects of radiation therapy. Based on these early preclinical data, Moer- tel et al. (1969) administered 5-FU with radiation to patients with gastrointestinal cancers and noted significant activity. The pioneering contribution to the use of combined radiotherapy and FU was made by Byfield et al. (1982), who demonstrated that 5- FU radiosensitization resulted from specific time and concentration factors: The sensitizing effects of 5-FU in vitro are maximal when its exposure occurs for at least 24 h and up to 48 h after the radiation exposure, thus supporting a prolonged 5-FU exposure approach when given with fractionated irradiation.
18.1.2.1
Randomized Trials of Postoperative Concurrent RCT
Historically, the combination of postoperative radi- otherapy and 5-FU-based chemotherapy has been shown in several randomized trials to reduce local recurrence rates and to improve overall survival compared with (conventional) surgery alone or sur- gery plus postoperative radiotherapy (Table 18.1). In the early GITSG 7175 trial, the best local control was achieved with combined RCT (local relapse rate of 11 vs 20% with RT alone), whereas no impact on local control was noted with chemotherapy as single adju- vant treatment (local relapse rate of 27 vs 24% with surgery alone; Gastrointestinal Tumor Study
Group 1985). Although rates of distant metastases were slightly lower in the two arms that contained chemotherapy, no single arm had a significant impact on distant failure; thus, the survival advantage achieved with combined RCT appeared to relate pri- marily to the marked reduction in local relapse rates.
These result were later confirmed by a trial conducted by the Norwegian Adjuvant Rectal Cancer Project Group (Tveit et al. 1997). Again, the local relapse rate was significantly decreased from 30 to 12% by combined postoperative RCT compared with surgery alone, an effect which also translated into an improve- ment in 5-year survival, though no significant impact on distant metastases was achieved. The more recent NSABP R-02 also showed that combined RCT resulted in a significantly reduced local failure rate compared with chemotherapy alone (8 vs 13%); however, this small absolute reduction did no longer translate into a difference in overall survival (Wolmark et al. 2000).
Evidently, in all these trials, the effect of concomitant 5-FU chemotherapy was primarily mediated through its radiosensitization properties rather than through its own systemic efficacy. This conclusion is further strengthened by a recent Italian study that showed no significant effect on local control and survival when postoperative radiotherapy (RT) and chemotherapy was applied sequentially rather than concomitantly (Cafiero et al. 2003).
The NCCTG 794751 trial was the first to integrate a course of full-dose chemotherapy before as well as
Table 18.1. Randomized trials of postoperative radiation (RT), chemotherapy (CT), or combined radiochemotherapy (RCT) for locally advanced rectal cancer (UICC II and III)
Reference Treatment Local failure (%) Distant failure (%) Five-year survival (%) GITSG 7175
(Gastointestinal Tumor Study Group 1985)
Surgery 24 34 45
Surgery + RT 20 (p=0.08) 30 52 (p<0.05)
Surgery + 5-FU/MeCCNU 27 27 56
Surgery + RT+5-FU/ MeCCNU 11 26 59
NCCTG/Mayo 794751 (Krook et al. 1991)
Surgery + RT 25 (p=0.04) 46 (p=0.01) 48 (p=0.025)
Surgery + RT+5-FU/MeCCNU 13.5 29 58
Norway trial (Tveit et al. 1997)
Surgery 30 (p=0.01) 39 50 (p=0.05)
Surgery + RT+5-FU 12 33 64
NSABP R-02
(Wolmark et al. 2000)
Surgery + CT a 13 (p=0.02) 29 64
Surgery + RCT 8 31 64
Italy trial
(Cafi ero et al. 2003)
Surgery + RT 20 38 59
Surgery + 5-FU/LEV + RT + 5-FU/LEV (RT and CT applied sequentially)
22 27 43
aMale patients received MOF (MeCCNU, Vincristin, 5-FU) or 5-FU/leucovorin; female patients only 5-FU/leucovorin
after combined RCT in an attempt to exploit both the radiosensitization properties of 5-FU and its potential to reduce the incidence of distant metas- tases (Krook et al. 1991). Indeed, this was also the first trial in which both local relapse and distant metastasis rates were significantly reduced in the experimental arm. The National Cancer Institute Consensus Conference concluded in 1990 that com- bined RCT was the standard adjuvant treatment for patients with TNM stage-II and stage-III rectal cancer (NIH Consensus Conference 1990).
18.1.2.2
Randomized Trials to Optimize 5-FU-Based Postoperative RCT
Further trials by the GITSG (7180) and NCCTG (864751) investigated the need for methyl-CCNU in the chemotherapy regimen and found that it added no benefit to the 5-FU regimen (Table 18.2; Gastro- intestinal Tumor Study Group 1992; O’Connell et al. 1994); thus, this compound is no longer used for
adjuvant RCT in rectal cancer. The NCCTG (864751) also tested the best method of administering 5-FU during radiotherapy: Bolus 5-FU (500 mg/m2 for 3 days during weeks 1 and 5 of radiation therapy) was compared with continuous infusion (225 mg/m2 during the whole course of radiotherapy): a 10%
disease-free and overall survival advantage was achieved with continuous infusion 5-FU during radiotherapy (Table 18.2). The INT 0144 trial tested the question whether additional continuous infusion 5-FU instead of bolus 5-FU before and after RCT (or modulation of 5-FU through addition of leucovorin and levamisol) may further increase tumor control.
Data were reported at the 2003 annual meeting of the American Society of Clinical Oncology (ASCO) and showed no difference in 3-year survival (Table 18.2;
Smalley et al. 2003).
Results of a four-arm intergroup trial, INT 0114, also showed no significant differences in local con- trol and survival among patients receiving either bolus 5-FU, bolus 5-FU + folinic acid, bolus 5-FU + levamisol, or bolus 5-FU + folinic acid + levami-
Table 18.2. Randomized trials of postoperative combined radiochemotherapy (RCT) in locally advanced rectal cancer
Reference Treatment DFS (%) OS (%)
GITSG 7180 (GITSG 1992)
RCT bolus 5-FU + bolus 5-FU (six cycles, escalating 5-FU)
68 (3 years) 75 (3 year)
RCT bolus 5-FU + bolus 5-FU/MeCCNU (12 months treatment)
54 (3 years;
p=0.20)
66 (3 years;
p=0.58) NCCTG 864751
(O’Connell et al. 1994)
Two cycles of bolus 5-FU (± MeCCNU)
+ RCT bolus 5-FU + two cycles of bolus 5-FU (± MeCCNU)
53 (4 years) 60 (4 years)
Two cycles of bolus 5-FU (± MeCCNU)
+ RCT PVI 5-FU + two cycles of bolus 5-FU (± MeCCNU)
63 (4 years;
p=0.01)
70 (4 years;
p=0.005) INT 0114
(Tepper et al. 2002)
Two cycles bolus 5-FU + RCT bolus 5-FU + two cycles bolus 5-FU
54% (all) 64% (all)
Two cycles bolus 5-FU/LV + RCT bolus 5-FU/LV + two cycles bolus 5-FU/LV
No significant difference
No significant difference Two cycles bolus 5-FU/LEV + RCT bolus 5-FU
+ 2 cycles bolus 5-FU/LEV
Two cycles bolus 5-FU/LV/LEV + RCT bolus 5-FU/LV + two cycles bolus 5-FU/LV/LEV
INT 0144
(Smalley et al. 2003)
Two cycles bolus 5-FU + RCT PVI 5-FU + two cycles bolus 5-FU
68–69 (3 years) 81–83 (3 years)
PVI 5-FU + RCT PVI 5-FU + PVI 5-FU No significant difference
No significant difference Two cycles bolus 5-FU/LV/LEV + RCT bolus 5-FU/LV
+ two cycles bolus 5-FU/LV/LEV Korean trial
(Lee et al. 2002)
RCT bolus FU/LV + six cycles bolus 5-FU/LV 81 (4 years) 84 (4 years) Two cycles bolus 5-FU/LV + RCT bolus FU/LV
+ 4 cycles bolus 5-FU/LV
70 (4 years;
p=0.04)
82 (4 years;
p=0.39) DFS disease-free survival, OS overall survival, LV leucovorin, LEV levamisole, PVI protracted venous infusion
sol (Tepper et al. 2002); however, gastrointestinal toxicity was higher in folinic acid containing regi- mens. The largest German adjuvant rectal cancer trial (FOGT 2) compared 5-FU + levamisol to 5-FU + levamisol + folinic acid and 5-FU + levamisol + interferone alpha as systemic treatment added to 45–50.4 Gy of radiotherapy. Toxicity was highest in the interferone containing arm, which was closed prematurely. Long-term results did not show any difference in failure rates and survival between the other groups (Staib et al. 2004).
Given all these results, the standard design of postoperative RCT is to deliver six cycles of 5-FU chemotherapy with concurrent radiation therapy during cycles 3 and 4. During radiotherapy continu- ous infusion 5-FU regimens (e.g., 225 mg/m2 day –1 during the whole course of radiation, or 1000 mg/
m2 day –1 as 120-h continuous infusion during weeks 1 and 5 of radiation, like in the German CAO/
ARO/AIO-95-study; see below), are recommended.
A recent randomized Korean trial (see Table 18.2) suggests that radiation should start with cycle 1 rather than cycle 3, which supports the radiobiologi- cal paradigm that subclinical disease in the pelvis is best controlled if RT is applied as soon as possible after surgical resection to account for any regrowth of residual tumor cells (Lee et al. 2002).
18.1.2.3
Randomized Trials to Optimize the Sequence:
Preoperative RCT
The interest in preoperative radiochemotherapy for resectable tumors of the rectum is based not only on the success of the combined modality approach in the postoperative setting, but also on many radio- and tumorbiological advantages of the preopera- tive approach. Among those are downsizing effects that possibly enhance curative surgery in locally advanced disease, and sphincter preservation in low-lying tumors. The small bowel in an unviolated abdomen will be mobile and less likely to be within a pelvic radiation portal, the irradiated volume does not require coverage of the perineum, as in the cases after abdominoperineal resection (APR), and there is no irradiation of the anastomotic region; thus, preoperative irradiation should cause less acute and late toxicity and more patients are likely to receive full-dose therapy. In addition, a certain dose of irra- diation seems to be more effective if given preopera- tively compared with postoperatively, most prob- ably due to the fact that oxygen tension within the tumor may be higher prior to surgical compromise
of the regional blood flow. This may improve the radiosensitivity of the tumor by decreasing the more radioresistant hypoxic fraction.
Until recently, the only randomized trial that directly compared preoperative to postoperative radiation therapy (both without chemotherapy) in rectal cancer has been the Uppsala trial, which was carried out between 1980 and 1985 in Sweden ( Frykholm et al. 1993). In the preoperative arm, patients received intensive short-course radiation (five fractions of 5.1 Gy to a total dose of 25.5 Gy in 1 week), postoperatively conventional radiation therapy (2 Gy to a total of 60 Gy with a 2-week split after 40 Gy) was applied. Preoperative radiation sig- nificantly decreased local failure rate (13 vs 22%;
p=0.02), however, there was no significant differ- ence in 5-year survival rates (42 vs 38%). Prospec- tive randomized trials comparing the efficacy of preoperative with standard postoperative RCT in UICC stage-II and stage-III rectal cancer were initi- ated both in the United States through the Radia- tion Therapy Oncology Group (RTOG 94-01) and the NSABP (R-03) as well as in Germany (Protocol CAO/ARO/AIO-94). Unfortunately, both U.S. trials suffered from lack of accrual and were closed pre- maturely. A preliminary report of the NSABP R- 03 trial (with a median follow-up of only 1 year) revealed that the percentage of patients who under- went sphincter sparing surgery and were without evidence of disease was higher in the preoperative vs the postoperative arm (44 vs 34%; Roh et al. 2001).
The German study (CAO/ARO/AIO-94) has recently been completed with more than 820 patients included. The design of this trial and the treatment schedule is depicted in Fig. 18.1. Results were recently reported (Table 18.3): Compared with postoperative RCT, the preoperative com- bined modality approach was superior in terms of local control, downstaging, acute and chronic tox- icity, and sphincter preservation in those patients judged by the surgeon to require an APR (Sauer et al. 2004). Given these advantages preoperative RCT is now the preferred adjuvant treatment for patients with locally advanced rectal cancer in Germany as well as in most parts of Europe; however, it needs to be emphasized that, with a median follow-up of 46 months, there was no difference in 5-year dis- ease-free and overall survival rates between both treatment arms.
18.1.2.4
Concomitant Chemotherapy with Preoperative Radiation Therapy?
The concurrent use of chemotherapy as part of the preoperative regimen is another important point, as it is not clear by now whether data from postopera- tive radiochemotherapy in resectable rectal cancer can be translated to the preoperative setting.
For the treatment of primarily “unresectable,”
fixed T4-rectal cancer, several institutions have applied preoperative RT and RCT. The goal is to convert (“downsize”) a tumor, which is clinically not amenable to curative resection at presentation,
to a resectable status. Minsky et al. (1992) compared preoperative radiotherapy (50.4 Gy) with or with- out 5-FU/folinic acid and showed that 90% of the patients with initially “unresectable” tumors were converted to resectable lesions by preoperative com- bined therapy as compared with only 64% of those who received radiation therapy alone. Moreover, a complete pathological response was found in 20%
of patients receiving combined modality therapy as compared with 6% receiving radiotherapy alone, indicating an enhancement of radiation-induced
“downstaging” by concomitant 5-FU-based RCT.
In a recent randomized phase-III study comparing radiotherapy alone with combined radiochemo- I
I m r
A :
I m r
A :
P O
h t i w y p a r e h t o m e h c t n a v u j d a f o s e l c y c 4
m / g m 0 0 5 U F -
5 2/di.v.bolusfor5days k
a e r b s k e e w 3
t s o o B y G 4 . 5 + y G 4 . 0 5 : T R
U F - 5 U
F - 5
m / g m 0 0 0 1
5 2 5x1000mg/m2 n o i s u f n i - h 0 2 1 n o i s u f n i - h 0 2 1
P O
y G 4 . 0 5 : T R
U F - 5 U
F - 5
m / g m 0 0 0 1 x
5 2 5x1000mg/m2 n o i s u f n i - h 0 2 1 n o i s u f n i - h 0 2 1
h t i w y p a r e h t o m e h c t n a v u j d a f o s e l c y c 4
m / g m 0 0 5 U F -
5 2/di.v.bolusfor5days k
a e r b s k e e w 3 x
Fig. 18.1. Design of the German CAO/ARO/AIO-94 study comparing postoperative (arm I) with preoperative radiochemotherapy (arm II) in locally advanced rectal cancer
Table 18.3. German Rectal Cancer Study Group randomized trial of preoperative compared with postoperative radiochemotherapy for rectal cancer. (From Sauer et al. 2004)
Five-year outcome Preoperative RCT (%) Postoperative RCT (%) p-value
Locoregional recurrence rate 6 13 0.006
Distant recurrence rate 36 38 0.84
Disease-free survival 68 65 0.32
Overall survival 74 76 0.80
Any grade-3/4 acute toxicity 27 40 0.001
Any grade-3/4 late toxicity 14 24 0.01
Sphincter preservation ratea 39 19 0.004
RCT radiochemotherapy
aIn patients deemed to require abdominoperineal resection by the surgeon before randomization
therapy for primarily unresectable T4-rectal cancer, Frykholm et al. (2001) could demonstrate that the addition of chemotherapy to radiotherapy signifi- cantly improved local control rates, albeit again no significant difference in survival was found between the groups.
A Polish randomized trial compared preoperative short-course irradiation (5×5 Gy) and immediate surgery with conventionally fractionated RCT (1.8–
50.4 Gy) and delayed surgery in 316 patients with locally advanced (T3/T4) low rectal cancer (Bujko et al. 2004). The primary end point of the trial was the rate of sphincter preserving surgery. Despite a significant increase in tumor response in the RCT group (pathological complete remission, 16 vs 1%;
mean largest tumor diameter on the operative speci- men, 29 vs 48 mm), the rate of sphincter preserva- tion was 61% in the immediate surgery group and 58% in the delayed group, indicating a strong com- mitment of the surgeons in this trial not to change their choice whatever the tumor response was after neoadjuvant RCT.
In primarily resectable tumors (cT3/4 and/or cN+), the European Organization for Research and Treatment of Cancer (EORTC) has conducted a four- arm trial that treated all patients with preoperative radiation in conventional fractionation (45 Gy in
25 fractions) and tested whether preoperative con- current RCT with 5-FU/folinic acid, postoperative 5-FU/folinic acid, or both are superior to preop- erative radiation alone (Table 18.4; Bosset et al.
2005a; Bosset et al. 2005b). The FFCD 9203 was a 2-arm trial also randomizing patients to preopera- tive 45 Gy with or without bolus 5-FU/folinic acid (Gerard et al. 2005). All patients received postop- erative chemotherapy in this trial. First results of both trials were reported at the ASCO meeting 2005 and indicated that the addition of 5-FU/folinic acid to preoperative conventionally fractionated radia- tion therapy significantly increased the pathologi- cal complete response rates, reduced tumor size and lymph nodal invasion, increased sphincter preser- vation (in EORTC 22921 only), and long-term local control rates, yet, these advantages once again did not translate into a survival benefit for patients treated by combined radiation and chemotherapy (Table 18.4).
18.1.2.5
Conclusions from Trials with 5-FU-Based RCT
In summary, if the available data on post- and pre- operative RCT with 5-FU-based chemotherapy for rectal cancer are considered, the following conclu- sions can be drawn:
1. There is evidently a higher local effectiveness of concurrent radiation and 5-FU as compared with surgery alone or adjuvant radiotherapy or chemo- therapy alone. In the postoperative setting, this is manifested by a higher probability of the com- bined approach to eradicate subclinical disease left behind after potentially curative surgery and, consequentially, by improved long-term local con- trol rates (GITSG 7175, NCCTG 794751, Norway trial, NSABP R-02). In the preoperative setting, increased tumor downsizing and downstaging, increased pathological complete regression, and improved local control rates are seen with con- current radiation and 5-FU (EORTC 22921; FFCD 9203; Frykholm et al. 2001; Bujko et al. 2004).
2. It is likely that these phenomena are due not only to pure additive effects of both cytotoxic modali- ties but also to some forms of interaction between radiotherapy and 5-FU (as suggested by the nega- tive results of the sequential Italy trial as com- pared with the positive results of concurrent RCT trials; Table 18.1). These interactions may range from classical “radiosensitization,” as shown in in vitro and animal studies, to more com- plex processes, such as 5-FU-induced inhibition
Table 18.4. Preoperative conventionally fractionated radio- therapy with or without 5-FU/LV-based chemotherapy.
Results of EORTC 22921 and FFCD 9203 randomised trials.
(From Bosset et al. 2005; Gerard et al. 2005)
Five-year outcome Preoperative RT
Preoperative RCT
p-value
EORTC 22921 (n=1011)
pCR rate 5.3% 13.7% <0.001
ypN0 60.5% 71.9% <0.001
Tumor size (median) 30 mm 25 mm <0.0001 Sphincter preserved 52.4% 55.6% 0.05
Local failure 17% 8% 0.002
Overall survival 64.8% 65.6% 0.79
FFCD 9203 (n=762)
pCR rate 3.7% 11.7% <0.0001
Sphincter preserved 52.6% 51.7% n.s.
Grade-3+4 toxicity 2.7% 14.6% <0.0001
Local failure 8% 16.5% n.g.
Overall survival 66% 67% n.g.
RT radiotherapy, RCT radiochemotherapy, pCR pathologic complete remission, n.s. not signifi cant, n.g. not given
of (accelerated) repopulation during fraction- ated radiotherapy. As predicted by early labora- tory studies, prolonged administration of 5-FU during radiotherapy seems to be more effective in this respect compared with 5-FU bolus pro- grams (NCCTG 864751), although this has not consistently been shown (INT 0144). Conversely, biochemical modulation 5-FU (by folinic acid, levamisol, or both) has not been proven to be superior to 5-FU alone (INT 0114, INT 0144).
3. If in clinical trials the absolute differences in local control rates between the combined modal- ity approaches and unimodal adjuvant treatment or surgery alone are in the range of 10% and above, these differences potentially translate into survival benefi ts (GITSG 7175, NCCTG 794751, Norway trial). With the advent of modern surgi- cal techniques, including total mesorectal exci- sion, the absolute number of local events has been markedly reduced and, thus, the absolute differ- ences in local control rates, although still signifi - cant, are smaller and do not readily translate into any survival benefi t (as shown in NSABP R-02, CAO/ARO/AIO-94, EORTC 22921, FFCD 9203).
4. The impact of 5-FU-chemotherapy and the vari- ous ways of its administration and modulation on the control of distant disease is less clear. The NCCTG 794751 demonstrated both reduced local and distant relapse rates with 5-FU-based chemo- therapy during as well as before and after RCT.
Although the necessity of maintenance chemo- therapy has been questioned both in the adju- vant (Hellenic trial; Table 18.2) and neoadjuvant setting (EORTC 22921), most oncologists now use continuous infusion 5-FU concurrently with
preoperative radiation therapy and four courses of 5-FU with or without folinic acid as additional systemic treatment following surgery. As shown in the recent German CAO/ARO/AIO-94 trial, this approach, together with TME surgery, has led to a local failure rate of only 6%; however, it became also evident that the pattern of failure is now dominated by distant metastases (36% at 5 years in CAO/ARO/AIO-94). Improvement of results in the treatment of rectal cancer will obviously require a more effective systemic approach.
18.1.3
Integrating Novel Chemotherapeutic Agents into Preoperative Combined Modality Treatment
Novel chemotherapeutic agents, such as capecitab- ine, oral uracil and tegaful (UFT), tomudex, oxali- platin, and irinotecan, as well as targeted therapies, such as bevacizumab and cetuximab, which have improved results of patients treated in the adjuvant and metastatic setting for colorectal cancer, are currently incorporated into phase-I/II combined modality programs for rectal cancer as well. All sug- gest higher pathological complete response (pCR) rates compared with 5-FU-radiochemotherapy alone (Tables 18.5–18.10); however, for some agents, with this increased pCR rate is an associated increase in acute toxicity. Clearly, phase-III trials are needed to determine if these regimens offer an advantage compared with 5-FU-based combined modality regimen.
Table 18.5. Phase-I studies of radiochemotherapy for locally advanced rectal cancer using orally administered fl uoropyrimi- dines
Reference Number Concurrent radiochemotherapy Dose-limiting toxicity Recommended dose Dunst et al.
(2002)
36 Preop., postop., or palliative RT:
1.8–50.4 Gy
Capecitabine 250–1250 mg/m2 bid for duration of RT
Grade-3 hand-foot syndrome at a dose level of capecitabine 1000 mg/m2 day–1 bid
Capecitabine 825 mg/m2 bid for duration of RT
Souglakos et al.
(2003)
31 Postop. RT: 1.8–50.4 Gy
Capecitabine 500–850 mg/m2 bid for duration of RT
Grade-3 diarrhea at a dose level of capecitabine 850 mg/
m2 day–1
Capecitabine 800 mg/m2 bid for duration of RT
Ngan et al.
(2004)
28 Preop. RT: 1.8–50.4 Gy
Capecitabine 425–1000 mg/m2 bid Monday to Friday for duration of RT
Grade-3 diarrhea at a dose level of capecitabine 1000 mg/
m2 day–1 bid Monday to Friday for duration of RT
Capecitabine 900 mg/m2 bid Monday to Friday for duration of RT
RT radiotherapy
18.1.3.1
Oral Fluoropyrimidines
Capecitabine is an oral fluoropyrimidine that mimics the pharmacokinetics of continuous 5-FU infusion and is preferentially converted to the active 5-FU metabolite within tumor cells by exploiting the higher activity of the enzyme thymidine phos-
phorylase in tumor tissue compared with normal tissue (Schuller et al. 2000). This tumor-selec- tive activation of capecitabine might be improved further when combined with RT, which upregulates thymidine phosphorylase in tumor cells but not in healthy tissue (Sawada et al. 1999). Preclinical stud- ies have shown that the combination of capecitabine and radiotherapy has highly enhanced antitumor
Table 18.7. Phase-I studies of preoperative radiochemotherapy for locally advanced rectal cancer using oxaliplatin-based combined modality treatment
Reference Number Concurrent radiochemotherapy Dose-limiting toxicity Recommended dose of oxaliplatin
Freyer et al.
(2001)
17 Preop. RT: 1.8–45 Gy Days 1–5 and 29–33:
5-FU 350 mg/m2 day-1,
LV 100 mg/m2 day-1; days 1 and 29:
oxaliplatin 80, 100, 130 mg/m2 day-1
MTD not reached Days 1, 29: oxaliplatin 130 mg/m2 day-1
Rödel et al.
(2003)
6 Preop. RT: 1.8–50.4 Gy
Days 1–14 and 22–35: capecitabine 825 mg/m2 bid; days 1, 8 and 22, 29:
oxaliplatin 50–60 mg/m2 day-1
Grade-3 diarrhea at a dose level of oxaliplatin 60 mg/m2 day-1
Days 1, 8, 22, 29: oxaliplatin 50 mg/m2 day-1
Gambacorta et al. (2004)
18 Preop. RT: 1.8–50.4 Gy
Days 1, 19, 38: Raltitrexed 3 mg/m2 day-1, oxaliplatin 65, 85, 110, 130 mg/m2 day-1
MTD not reached Days 1, 19, 38: oxaliplatin 130 mg/m2 day-1
Aschele et al.
(2005)
25 Preop. RT: 1.8–50.4 Gy Duration of RT: 5-FU 200,
225 mg/m2 day-1, oxaliplatin 25, 35, 45, 60 mg/m2 day-1 once weekly for a total of six courses
MTD not reached Weekly oxaliplatin 60 mg/
m2 day-1;
5-FU 225 mg/m2 day-1 during RT
Loi et al.
(2005)
16 RT: 1.8–50.4 Gy
5-FU 200 mg/m2 day-1 5 days, 7 days per week during RT; Days 1, 15, 29:
oxaliplatin 70, 85 mg/m2 day-1
Grade-3 diarrhea at a dose level of 5-FU 200 mg/m2 day-1, 7 days per week, and oxalipla- tin 85 mg/m2 day-1
5-FU 200 mg/m2 day-1, 5 days per week during RT;
days 1, 15, 29: oxaliplatin 85 mg/m2 day-1
RT radiotherapy, LV leucovorin, MTD maximum tolerated dose
Table 18.6. Phase-II studies of preoperative radiochemotherapy for locally advanced rectal cancer with orally administered fl uoropyrimidines
Reference Number Concurrent radiochemotherapy Toxicity pCR (%)
Kim et al. (2002) 45 Preop. RT: 1.8–50.4 Gy
Days 1–14 and 22–35: Capecitabine 825 mg/m2 bid
Grade-3 hand–foot syn- drome 7%, diarrhea 4%, fatigue 4%
31
Fernandez-Martos et al. (2004)
94 Preop. RT: 1.8–45 Gy
UFT 400 mg/m2 day–1, 5 days a week for 5 weeks
Grade-3 diarrhea 14%, leukopenia 14%
9
Kim et al. (2005) 95 Preop. RT: 2.0–50 Gy
Capecitabine 825 mg/m2 bid during RT
Grade-3 diarrhea 3%, neutropenia 1%
12
RT radiotherapy, UFT oral uracil and tegaful, pCR pathological complete response
Table 18.8. Phase-II studies of preoperative radiochemotherapy for locally advanced rectal cancer using oxaliplatin-based com- bined-modality treatment
Reference Number Concurrent radiochemotherapy Toxicity pCR (%)
Carraro et al.
(2002)
22 Preop. RT: 1.8–50.4 Gy
Days 1–4 and 29–32: 5-FU 375 mg/m2 day-1, LV 20 mg/m2 day-1, oxaliplatin 25 mg/m2 day-1; day 15:
oxaliplatin 50 mg/m2; four weeks after RT, one addi- tional cycle of oxaliplatin, 5-FU/LV (same dose as during RT)
Grade-4: leukopenia 4.5%
Grade-3: diarrhea 27%;
leukopenia 4.5%
25
Gérard et al.
(2003)
40 Preop. RT: 1.8–45 Gy (concomitant boost to 50 Gy) Days 1–5 and 29–33: 5-FU 350 mg/m2 day-1, LV 100 mg/m2 day-1; days 1 and 29: oxaliplatin 130 mg/
m2 day-1
Grade-4: diarrhea 2.5%;
mucositis 2.5%
Grade-3: fatigue 7.5%;
diarrhea 5%; proctitis 5%;
neutropenia 2.5%
15
Rödel et al.
(2003)
26 Preop. RT: 1.8 Gy to 50.4 Gy
Days 1–14 and 22–35: capecitabine 825 mg/m2 bid;
days 1, 8, 22, 29: oxaliplatin 50 mg/m2 day-1
Grade-3: diarrhea 8%; skin (local) 8%
19
Gambacorta et al.
(2004)
30 Preop. RT: 1.8–50.4 Gy
Days 1, 19, 38: raltitrexed 3 mg/m2 day-1; oxaliplatin 130 mg/m2 day-1
Grade-3: leukopenia 10%;
vomiting 3%; proctitis 3%
30
Aschele et al.
(2005)
25 Preop. RT: 1.8–50.4 Gy
Duration of RT: 5-FU 225 mg/m2 day-1, oxaliplatin 60 mg/m2 day-1 once weekly for a total of six courses
Grade-3: diarrhea 16%; skin (local) 12%; anemia 4%
28
pCR pathological complete response, RT radiotherapy, LV leucovorin
Table 18.9. Phase-I studies of preoperative radiochemotherapy for locally advanced rectal cancer using irinotecan-based com- bined-modality treatment
Reference Number Concurrent radiochemotherapy Dose limiting toxicity Recommended dose of CPT-11 Voelter
et al. (2003)
28 Preop. RT: 1.6 Gy bid to 41.6 Gy (started on day 8)
Days 1, 8, 15: CPT-11 30–105 mg/m2 day-1
Grade 3 diarrhea at dose level CPT-11 105 mg/m²/d
Days 1, 8, 15:
CPT-11 90 mg/m²/d
Hofheinz et al. (2005)
19 Preop. RT: 1.8–50.4 Gy
Days 1–38: capecitabine 500, 625 mg/m2 day-1 Days 1, 8, 15, 22, 29: CPT-11 50 mg/m2 day-1
Grade 3 diarrhea at dose level Capecitab- ine 625 mg/m²/d
Days 1–38:
Capecitabine 500 mg/m²/d;
days 1, 8, 15, 22, 29:
CPT-11 50 mg/m²/d
Table 18.10. Phase-II studies of preoperative radiochemotherapy for locally advanced rectal cancer using irinotecan-based combined-modality treatment
Reference Number Concurrent radiochemotherapy Toxicity pCR (%)
Mehta et al.
(2003)
32 Preop. RT: 1.8–50.4 Gy
Days 1–33: 5-FU 200 mg/m2 day-1; days 1, 8, 15, 22: CPT-11 50 mg/m2 day-1
Grade 3: diarrhea 28%; mucositis 21%;
proctitis 21%; abdominal cramping 9%
37
Klautke et al.
(2005)
37 Preop. RT: 1.8–50.4 Gy
Duration of RT: 5-FU 250 mg/m2 day-1; CPT-11 40 mg/m2 day-1 once weekly for a total of six courses
Grade 4: diarrhea 5%; leukopenia 2%;
grade 3: diarrhea 27%; leukopenia 8%
22
pCR pathological complete response, RT radiotherapy
activity compared with radiation or capecitabine alone (Sawada et al. 1999).
Three phase-I studies have been conducted to determine the maximum tolerated dose (MTD) of capecitabine in combination with standard RT in patients with rectal cancer (Table 18.5). Capecit- abine plus RT demonstrated promising activity in the study by Dunst et al. (2002), including one pathological complete remission (pCR) and nine partial responses (PR) in the 10 patients treated in the neo-adjuvant setting. Furthermore, no grade-3 or grade-4 toxicities occurred in patients treated at the recommended dose (continuous capecitabine 825 mg/m2 twice daily in combination with stand- ard RT). In the two other phase-I studies, the MTD of capecitabine was reached at a dose level of 1000 mg/
m2 twice daily on Monday to Friday throughout the course of preoperative pelvic RT, and at a dose of 800 mg/m2 twice daily during postoperative radio- therapy, respectively (Ngan et al. 2004; Souglakos et al. 2003).
Phase-II studies of preoperative radiotherapy with capecitabine and UFT indicate pCR rates in the range of 9 and 31%, with diarrhea and hand–
foot syndrome as the main grade-3 acute toxicity (Table 18.6; Fernandez-Martos et al. 2004; Kim et al. 2005; Kim et al. 2002). An ongoing German phase-III trial currently compared capecitabine with standard infusional 5-FU in patients undergo- ing pre- or postoperative radiotherapy for locally advanced rectal cancer. The NSABP R-04 trial is a prospective randomized trial of neoadjuvant radi- otherapy with capecitabine vs infusional 5-FU (r oxaliplatin in both arms). The use of postoperative adjuvant chemotherapy is left open to investigators or they may enter intergroup study E5202 (postoper- ative FOLFOX r avastin vs 5-FU/LV r avastin). Based on the available data in adjuvant and metastatic colon cancer trials, that compared capecitabine to intravenous 5-FU plus folinic acid (Twelves et al.
2005; Van Cutsem et al. 2004), it is likely that oral fluoropyrimidines will finally replace infusional 5- FU with radiotherapy in rectal cancer as well.
18.1.3.2 Oxaliplatin
Oxaliplatin is a very reasonable candidate for inclu- sion into neoadjuvant downsizing regimens because of its rapid cytoreductive capacity and its relative lack of acute dose-limiting side effects when added to 5-FU or capecitabine. As a preoperative regimen for initially unresectable liver metastases, the com-
bination of oxaliplatin and 5-FU/folinc acid resulted in tumor downsizing in 59% of patients and in a complete resection rate of 38% (Giacchetti et al.
1999). Randomized phase-III trials have demon- strated the superiority of combined oxaliplatin and 5-FU/folinic acid compared with 5-FU/folinic acid in metastatic colorectal disease in terms of response and progression-free survival and, more recently, also in the adjuvant treatment of colon cancer in terms of improved disease-free survival (Andre et al. 2004; Giacchetti et al. 2000).
In vitro and in vivo preclinical studies have dem- onstrated oxaliplatin to be a potent radiosensitizing agent (Cividalli et al. 2002; Magne et al. 2003). In the in vivo xenograft model of HT-29 colon carci- noma, oxaliplatin showed increased cytotoxicity when combined with irradiation (Blackstock et al. 1999). Although it remains controversial whether oxaliplatin should be delivered before or after radia- tion to maximize its radiosensitizing activity, more recent investigations suggest that oxaliplatin should likely be given a number of hours before RT to maxi- mize benefit (A.W. Blackstock et al., unpublished data).
Freyer et al. (2001) have recently published results from a phase-I study of RT (45 Gy over 5 weeks) plus oxaliplatin and 5-FU/folinic acid demonstrating the feasibility of such an intensified chemotherapy regi- men when given concomitantly with preoperative RT (Table 18.7). For historical reasons, chemother- apy was administered only in the first and fifth week of RT in this study: using escalating doses of oxali- platin (80, 100, or 130 mg/m2 on days 1 and 29), 5-FU (350 mg/m2 on days 1–5 and 29–33) and leucovorin (100 mg/m2 on days 1–5 and 29–33) the MTD was not reached. In a subsequent phase-II trial, Gerard et al. (2003) demonstrated that such a preoperative combined RCT regimen (with 130 mg/m2 oxaliplatin on days 1 and 29) was well tolerated with no increase in surgical toxicity and a promising complete patholological response rate of 15%. If the goal of preoperative chemoradiation is to maximize tumor shrinkage prior to surgery as well as to improve sys- temic control, concomitant chemotherapy should be as dense as possible (i.e., applied as often as possi- ble during RT) to maximize local effectiveness by radiation sensitization, and as intense as possible to effectively eradicate microscopic distant disease.
In addition, acute toxicity may be substantially reduced through chemotherapy dose fractionation, that may also allow a higher cumulative dose to be administered during radiotherapy; thus, subsequent trials developed and tested novel combined modal-
ity regimen that incorporated oxaliplatin at a weekly rather than a monthly basis. Aschele et al. (2005) reported the feasibility and promising activity (pCR rate 28%) of an intensified preoperative RCT regi- men where oxaliplatin was administered weekly at a dose of 60 mg/m2 for six cycles together with 5-FU 225 mg/m2 day-1 as continuous infusion during the whole course of RT (50.4 Gy/28 fractions). Rödel et al. (2003) used a combination of capecitabine and oxaliplatin given weekly with preoperative RT, apart from a 7-day break during the third week of RT. This group also found this preoperative regimen to be well tolerated and effective (pCR rate 19%).
It is noteworthy that, for all the studies sum- marized in Table 18.7 and 18.8, either no postop- erative chemotherapy at all was applied or adjuvant chemotherapy was left to the discretion of the treat- ing physician. Given the fact that (a) in previous 5-FU-based phase-III trials concomitant as well as sequential (adjuvant) cycles of chemotherapy were applied, and (b) the overall cumulative doses of the new drugs reached during radiotherapy is substan- tially lower than in adjuvant colon cancer trials, two multicenter phase-II trials (the French CORE study, and a multicenter phase-II study from the German Rectal Cancer Study Group) currently are investigat- ing the role of preoperative concomitant RCT with capecitabine and oxaliplatin plus four to six cycles of adjuvant XELOX chemotherapy. The double-arm randomized CHRONICLE trial in the United King- dom tests the question of whether six cycles of adju- vant XELOX after 5-FU based preoperative RCT and curative resection improves disease-free survival compared with no adjuvant chemotherapy at all.
18.1.3.3 Irinotecan
Irinotecan, a semisynthetic derivate of camp- tothecin, is a potent inhibitor of topoisomerase I, a nuclear enzyme that plays a critical role in DNA replication and transcription. This drug has now become standard therapy in first- and second-line treatment of metastatic colorectal cancer in combi- nation with bolus or infusional 5-FU. With a mecha- nism of action interfering with DNA replication, iri- notecan also showed radiosensitizing properties in preclinical studies, which showed maximum syner- gistic effects when irinotecan was administered 1 h before irradiation (Rich et al. 2001). A valid concern in using topoisomerase inhibitors with 5-FU and pelvic irradiation, however, is overlapping toxicity, particularly the development of severe diarrhea.
Reminiscent of the combined modality approaches with oxaliplatin and for the same reasons, weekly schedules of irinotecan administration (in the dose range of 40–50 mg/m2 day-1 given once weekly) have been developed in combination with continuous infusion 5-FU or capecitabine (Tables 18.9, 18.10).
The published phase-I/II trials with this novel com- bination suggest that gastrointestinal acute toxicity is manageable and pCR rates are highly promising (Hofheinz et al. 2005; Klautke et al. 2005; Mehta et al. 2003; Voelter et al. 2003).
18.1.3.4
Targeted Therapy
The epidermal growth factor receptor (EGFR) is expressed in 50–70% of colorectal cancer. Enhanced EGFR signaling may affect cell cycle control, apop- tosis, and angiogenesis, leading to tumor growth and progression. In a cohort of patients with rectal cancer treated by surgery alone, increased EGFR levels were significantly associated with unfavorable histopatho- logical factors, such as lymph node involvement, more advanced UICC-TNM stages, and predicted for a reduced overall survival (Kopp et al. 2003). In a group of rectal cancer patients treated with preop- erative radiotherapy and surgery, Giralt et al. (2005) identified EGFR expression as an indicator for poor pathological response to radiotherapy, and reduced disease-free survival following curative surgery.
Several preclinical studies have shown that block- ing EGFR ligand binding via monoclonal antibody (e.g., cetuximab) can enhance tumor radiosensitiv- ity (Baumann and Krause 2004). Clinical studies of preoperative RCT have now been initiated to evalu- ate EGFR inhibitors as radiosensitizers in rectal cancer as well. Figure 18.2 depicts the design of our own study incorporating capecitabine, oxaliplatin, and cetuximab plus preoperative radiotherapy in locally advanced rectal cancer. First experiences with this schedule do not indicate increased toxicity apart from acne-like rash commonly associated with the EGFR antibody. The ongoing EXPERT-C trial is a multicenter randomized phase-II trial comparing oxaliplatin, capecitabine, and preoperative radio- therapy with or without cetuximab for magnetic resonance imaging (MRI)-defined poor risk rectal cancer. Whether or not these combinations will fur- ther improve the pathological complete response rates, and whether or not this possible improvement will finally translate into an improved disease-free survival, needs to be awaited.
