Rectal cancer radiotherapy: Towards European consensus

Abstract

Background and purpose. During the first decade of the 21st century several important European randomized studies in rectal cancer have been published. In order to help shape clinical practice based on best scientific evidence, the International Conference on ‘Multidisciplinary Rectal Cancer Treatment: Looking for an European Consensus’ (EURECA-CC2) was organized. This article summarizes the consensus about imaging and radiotherapy of rectal cancer and gives an update until May 2010. Methods. Consensus was achieved using the Delphi method. Eight chapters were identified: epidemiology, diagnostics, pathology, surgery, radiotherapy and chemotherapy, treatment toxicity and quality of life, follow-up, and research questions. Each chapter was subdivided by topic, and a series of statements were developed. Each committee member commented and voted, sentence by sentence three times. Sentences which did not reach agreement after voting round # 2 were openly debated during the Conference in Perugia (Italy) December 2008. The Executive Committee scored percentage consensus based on three categories: “large consensus”, “moderate consensus”, “minimum consensus”. Results. The total number of the voted sentences was 207. Of the 207, 86% achieved large consensus, 13% achieved moderate consensus, and only three (1%) resulted in minimum consensus. No statement was disagreed by more than 50% of members. All chapters were voted on by at least 75% of the members, and the majority was voted on by >85%. Considerable progress has been made in staging and treatment, including radiation treatment of rectal cancer. Conclusions. This Consensus Conference represents an expertise opinion process that may help shape future programs, investigational protocols, and guidelines for staging and treatment of rectal cancer throughout Europe. In spite of substantial progress, many research challenges remain.

Surgery remains the most important treatment of rectal cancer, nevertheless the management of this disease has evolved to become multidisciplinary [1]. During the past ten years a number of important European randomized studies have been published. They have examined a variety of preoperative approaches and most have required the use of total mesorectal excision (TME). In addition, advances in both pathology and imaging have further contributed to the multidisciplinary management [1].

In order to shape clinical practice based on best scientific evidence from literature, the International Conference on ‘Multidisciplinary Rectal Cancer Treatment: Looking for an European Consensus’ (EURECA-CC2) was organized in Italy under the endorsement of European Society of Medical Oncology (ESMO), European Society of Surgical Oncology (ESSO), and European Society of Therapeutic Radiation Oncology (ESTRO). The goal of this consensus conference was to help develop future programs, investigational protocols, and guidelines for staging and treatment of rectal cancer throughout Europe [2].

The aim of this manuscript is to focus on the main recommendations addressed in this Consensus to support the daily practice of radiation oncologists. An update of the literature published during 2009 to May 2010 was made.

Methods

The Departments of Radiotherapy of the Catholic University of Rome and the University of Perugia, who organized the first Consensus Conference [3] and are responsible for development of an ESTRO Multidisciplinary Teaching Course managed with ESSO and ESMO [1] organized the conference [2] with the endorsement of the three European Societies. An Executive Committee included two delegates from each of the three societies, two from the journal Radiotherapy & Oncology, and a radiologist, epidemiologist, and pathologist who participated in the multidisciplinary ESTRO teaching course. The Executive Committee then selected, based on majority vote, the participants to the Scientific Committee from experts who were involved in the major European published trials.

Consensus was achieved using the Delphi method [4]. Version 1 was created by the Executive Committee. The document was available to all committee members as a web-based document customized for the consensus process. Each member commented and voted, sentence by sentence. In addition, references to each sentence were presented and members were able to add additional ones. A consensus score was agreed between the experts. The outcome of each vote by web (% agree, % disagree, and new comments and references) was available to each member prior to the next vote. Another voting was performed during the conference in Perugia (Italy) from December 11–13, 2008. The Meeting was open to everyone interested in the topic. Sentences which did not reach agreement after the previous voting were openly debated by attendees at each session and the audience had the opportunity to ask further questions. Following the conclusion of the Perugia meeting the voting of the final document occurred. The final text was reviewed and collated by two experts (without changing the outcome of the votes).

The total number of voted sentences was 207. Of the 207, 86% achieved large consensus, 13% achieved moderate consensus, and only three (1%) resulted in minimum consensus. The sentences with moderate or minimum consensus are identified in the text.

The literature update since the conference has not undergone the consensus process.

Results

During the past decades different treatment schedules have been examined such as postoperative chemoradiotherapy with different 5-fluorouracil (5FU)-based schedules, preoperative radiotherapy short course (5 × 5 Gy in 5 days), long course (alone or in combination with 5FU-based regimens or with new drugs), and intraoperative radiotherapy (IORT). These modalities are used differently in different parts of Europe and in North America, based upon the same evidence from studies performed in different parts of the world.

The evidence in the literature on the main advantages of a particular approach for different presentations of rectal cancer will be analyzed for early presentation, intermediate stage and locally advanced “unresectable” rectal cancers.

Imaging of rectal cancer

There are many different imaging modalities suitable for rectal cancer staging, tumour location and restaging but not all of them have the same accuracy for each indication.

Imaging tools to distinguish clinical stage (c)T1 vs cT2. With a moderate consensus, endorectal ultrasound (EUS) is considered the most accurate imaging modality for the assessment of tumor penetration into the rectal wall [5] even if, at times, it is difficult to determine the depth of invasion in large villous lesions.

For assessing the depth of tumour growth in the bowel wall EUS has an overall accuracy between 69% and 97% [5]. The highest accuracy is obtained in expert EUS centers. When EUS is performed in general clinical setting the accuracy significantly drops [6,7]. EUS cannot be reliably used in patients with high or stenosing tumors [8], but these latter tumors are rarely early.

Endorectal MRI can be considered as accurate as EUS for staging superficial tumors as shown by comparative studies between the two endoluminal techniques. Endorectal MRI allows an objective evaluation of the imaging also in high located or stenosing cancers, and is less observer dependent than EUS. Endorectal MRI is, however, more expensive, technically more demanding for the MR unit and less comfortable for patients. Therefore, investigation of superficial tumors is preferred with EUS in expert centers, even if the number of studies is limited [9].

Usually, the aim of these examinations is not to distinguish between cT1 vs cT2 lesions but is restricted to cT1 sm1/sm2 and sm3 evaluation. cT1 Sm3 and cT2 lesions require the same treatment. Most sm1 and sm2 lesions are removed at colonoscopy prior to staging, and subsequent management is often determined by histopathology and not by EUS or MRI.

Phased array MRI and multispiral CTs are not reliable in the differentiation between T1 versus T2 lesions [10].

Imaging tools to distinguish cT2 vs cT3. With a moderate consensus, endorectal MRI can be considered as accurate as EUS for differentiation of superficial (cT1 and/or cT2) rectal tumors from cT3.

Phased array MRI fails in the differentiation between T2 versus borderline T3 lesions and overstaging is the main cause of errors. It is difficult to distinguish by MRI between desmoplasia without tumor cells (stage pT2) and desmoplasia with tumor cells (stage pT3). EUS has the same limitations [9,11].

Phased array MRI and multidetector CT seem the have equal accuracy for staging advanced T3 tumors, although the number of available comparative studies is limited. CT cannot assess the depth of extramural spread as accurately as histology but MRI is equivalent to histology in measurement of extramural depth. Careful attention to technique and the use of high resolution sequences with scans planned perpendicular to the anal canal and coronal scans to show the sphincter complex make MRI superior for lower third rectal tumors [10–12].

With a moderate consensus, new generation 16 slice spiral CT with optimal bolus timing and reconstruction in multiple planes can be considered to achieve high sensitivity and specificity for prediction of tumor penetration in the bowel wall, even if it appears not to be as accurate as MRI in the low rectum locations. The CT accuracy can be superior to EUS when the latter is performed in less expert EUS centers [13].

Imaging tools to distinguish CRM+ vs CRM−. With a moderate consensus, EUS and endorectal MRI are not considered accurate for mesorectal fascia evaluation. Conventional CT is not helpful for predicting an involved resection margin [14].

Multi-detector 4-16 slice CT appears promising for the prediction of a free circumferential resection margin, but not in low tumors, especially not those located in the low anterior rectal wall [12].

Phased array MRI is highly accurate for the prediction of CRM positivity in routine clinical practice [9,11,15,16]. PET-CT is not reliable in the evaluation of an involved CRM [17,18]. Substaging of the cT3 group with MRI according to the depth of tumor extension into the mesorectal fat penetration of the tumor in the mesorectum is recommended [11,19].

Imaging tools to distinguish cT3 vs cT4. With a moderate consensus, EUS is considered not to be accurate in the assessment of local tumor extent in bulky T3 and T4 rectal cancer [5].

Multidetector 4-16 slice CT is accurate for staging the advanced T3 tumors in the middle and high rectum with especially a high negative predictive value (NPV), at the expense of lower positive predictive value (PPV). However, the accuracy decreases for tumors located in the lower rectum [10–12].

Phased array MRI is highly accurate in staging advanced rectal cancer, in the assesment of mesorectal fascia infiltration, and to distinguish cT3 from cT4 [9,11].

PET-CT does not add to the accuracy in evaluating the cT stage in advanced rectal cancer [17].

Imaging tools to distinguish cN0 vs cN1-2. With a moderate consensus it was agreed that identifying nodal disease is a diagnostic problem for the radiologist. Nodes of >8 mm are defined as malignant nodes on CT, MRI and EUS. In contrast, despite the identification of lymph nodes as small as 2–3 mm on modern planar imaging, reliable detection of nodal metastases is presently not possible, because CT, MRI and EUS all rely on size criteria for predicting nodal metastases [16].

EUS is considered slightly superior to non-contrast enhanced MRI and CT for nodal staging but the entire mesorectum can not be explored. EUS guided fine needle aspiration has been reported to be a very reliable method with accuracy up to 100%, but it is a cumbersome technique that has not gained widespread acceptance [16].

New generation multislice spiral CT cannot accurately distinguish between malignant and benign lymph nodes measuring <8 mm [20].

Size is not a good predictor for malignancy and should not be used for defining whether lymph nodes are involved or not. The most reliable method of positively identifying nodal metastases is based on morphological features such as the presence of mixed signal intensity within the lymph node and/ or irregularity of the borders due to capsular penetration by malignancy. The overall accuracy is lower than other prognostic features that can be identified more reliably using high resolution MR scan techniques [21].

FDG PET-CT has shown disappointing results for N-staging in rectal cancer, especially in the mesorectum in the presence of a bulky tumor [17].

Imaging tools to distinguish responders vs non-responders. The detection of small clusters of residual tumor cells remains a problem and a complete remission after chemoradiation can not be reliably predicted with non-invasive imaging tools. Although EUS, CT and MRI can assess downsizing of the tumor, it is not accurate, especially when there is a fibrotic thickening of the rectal wall, in distinguishing between ypT0, ypT1, ypT2 or ypT3 tumors. The main source of error is overstaging.

With a moderate consensus it was agreed that reasonably high level of accuracy has been observed by phased array MRI when the endpoint is differentiating ypT0-2 vs ypT3. Although it could be useful for the surgeons to plan less extensive surgery, there is no solid evidence for this [22].

Many studies have reported a significant decrease of standardized uptake value (SUV) on postradiation FDG-PET in responders when compared to non-responders, but the clinical value of this information remains to be determined [23].

Early localized tumors

Early tumors are neoplasms limited to the rectal wall (c/pT1-2 N0 M0). They represent 3–5% of rectal cancers, and include small, exophytic, mobile tumors without adverse pathologic factors (i.e., high grade, blood or lymphatic vessel invasion, colloid histology, or the penetration of tumor into or through the bowel wall) and can be adequately treated with a variety of local therapies.

Radiotherapy approach to early rectal cancer. Patients who have received standard TME surgery for an early, localized tumor do not need further therapy.

When patients with early, localized tumor have undergone a local surgical procedure, they are at risk for disease recurrence in the rectal wall or in the local nodes. Patients with pT1 tumors without adverse pathologic factors have a low rate of local failure (5–10%) and positive nodes (<10%) and usually do not need adjuvant therapy. However, there is a lack of evidence to demonstrate equivalent outcomes to radical surgery [24]. On the contrary, when adverse pathologic factors are present (involved margins, sm3, poor differentiation, lymph vessel invasion) or the tumor invades into or through the muscularis propria (pT2-3), the local failure rate increases to at least 17% and the incidence of positive nodes to above 10% and adjuvant treatments are recommended [25,26].

The role of postoperative radio(chemo)therapy. With a moderate consensus it was agreed that the main advantage of surgery prior to irradiation is that pathologic details such as margins, depth of bowel wall penetration, and histological features are known.

Patients with pT1 tumors (after local excision) with any of the adverse pathologic factors mentioned above or with any doubt about quality of the local excision procedure have to undergo a resection of the entire rectum. Postoperative radio(chemo)therapy could be considered for compromized general conditions or if the patient refuses surgery [24,27].

The optimal treatment of a pT2 tumor after a local excision is not clear, since large randomized trials are not available. Local excision alone is insufficient and radical surgery is therefore recommended. Postoperative radio(chemo)therapy is a reasonable alternative when adverse prognostic factors (involved margins, poorly differentiated tumor and lymphovascular invasion) are absent and the patient has co-morbidity or refuses surgery. However, in series with long-term follow-up, the pelvic failure rates are 18–25% [28], i.e. much higher than if radical surgery is done.

Salvage of local failures is possible after local excision. In half of the patients, with local failure after local excision +/− radio(chemo)therapy, local control can be achieved with salvage abdominoperineal resection (APR). Close follow up to detect early relapse and then perform curative resection is recommended. Local recurrence after local excision and postoperative radio(chemo)therapy tends to occur late (median about five years).

The series that have measured sphincter function after local excision and radiotherapy report favorable outcomes [28].

The role of radiotherapy alone. External radiotherapy alone in early rectal cancer might be a feasible alternative to local excision in patients with poor medical condition or who refuse any surgical treatment. However the evidence is limited and definitive recommendation requires further studies.

Contact therapy alone in early rectal cancer may be a feasible alternative to local excision, namely in patients with very poor medical condition and without adverse prognostic factors, or who refuses any surgical treatment. However, this should be done in specialized centers and the number of available studies is limited [29].

The role of preoperative radio(chemo)therapy. Preoperative short course radiotherapy in clinically operable cT2N0 rectal cancers <15 cm from anal verge results in an even lower risk of local failure, but is usually not indicated since the absolute risk of a local failure in these early tumors is very low, provided very high quality staging and surgery can be performed. Recommendation for its use depends on interdisciplinary decision making and institutional preferences [30,31].

With a moderate consensus it was agreed that patients, who are either medically inoperable or refuse radical surgery, can receive preoperative radiation followed by local excision. It is delivered usually with 5FU based concomitant chemotherapy, but a patient not fit for prolonged radio(chemo)therapy can receive short course radiotherapy alone and delayed surgery. This approach is reported in only a few series and its use must be limited to only this subset of patients [32–34].

Intermediate stage (Stage II–III resectable)

Intermediate tumors are defined as neoplasms extending beyond the rectal wall (c/p T3-4 or N1-2 M0) but without unresectable infiltration to surrounding organs (all cT4).

Radiotherapy approach to intermediate rectal cancer. There are two treatment approaches for patients with intermediate stage resectable rectal cancer. The first approach is preoperative radio(chemo)therapy followed by surgery and then postoperative chemotherapy can be considered. The second is initial surgery followed by postoperative combined modality therapy if the tumor is pT3-4 and/or N1-2 [1].

Four meta-analyses about the value of radio(chemo)therapy (pre-or postoperatively) report partly conflicting results [35–38]. All of them reveal a decrease in local recurrence rates. The analysis by Camma et al. [35] and the Collaborative Colorectal Cancer Group [36] reported a survival advantage, whereas the analysis by Munro and Bentley [37] did not. The Swedish Council of Technology Assessment in Health Care (SBU) performed a systematic review of radiation therapy trials [38] and reported that survival is improved by about 10% using preoperative radiotherapy at adequate doses. The partly conflicting results are at least partly due to inabilities to recognize overlaps between some of the trials.

The role of short course preoperative radiotherapy. With a moderate consensus it was agreed that short course radiotherapy definitively reduces local recurrence risk for patients with most rectal cancers. The relative risk reduction may actually be higher the lower the absolute risk of a local failure is. The largest absolute gains have in the trials been seen in patients with extramural spread and node positive disease [30,31].

For patients with positive CRM, there is a reduction in local failure rates after short-course radiation although the magnitude of benefit is not considered sufficient for routine use [31].

After standardization of TME there is no evidence of overall survival benefit in the single short course randomized trial, the TME-trial [31]. The reduction in local failure rates in most intermediate cancers after TME standardization are too small to translate into an overall survival benefit irrespective of which radiotherapy modality is used. However population-based studies have demonstrated that since standardization of rectal cancer surgery with TME and the implementation of preoperative radiotherapy there has been a survival benefit [30,31,39].

The role of long-course preoperative radiotherapy. The analysis of the randomized trials which compare preoperative radiotherapy vs surgery alone showed that the combined treatment at biologically effective doses above 30 Gy decreases the relative risk of local failure. However, these analyses did not include only patients receiving long-course RT and most of the patients in the trials received short course [35–38]. Since the standardization of TME there has been no randomized trial comparing long-course preoperative radiotherapy versus surgery alone.

The role of long-course preoperative radiochemotherapy. Two recent randomized trials have showed an improvement in the results of preoperative radiation in patients with intermediate stage rectal cancer when 5FU based chemotherapy is added to radiotherapy. A statistically significant decrease in local recurrence was observed in those receiving chemotherapy as well as an increased rate of pCR. Five year overall survival was not changed by chemotherapy, but the trials were underpowered to detect a 5% difference in overall survival [40,41].

After standardization of TME, there has been no randomized trial comparing long-course preoperative radiochemotherapy with surgery alone. The relative efficacy of short-course RT and long-course chemotherapy has been evaluated in two randomized trials [42,43]. None of the trials, both including more than 300 patients could detect any difference in terms of local recurrence rate, disease-free and overall survival and late toxicity. Acute toxicity was less using short-course radiotherapy.

After preoperative radiochemotherapy a variable percentage of pathological complete response (pCR) specimens has been reported. Although some series show no correlation [44], many series report that patients who achieve a pCR following preoperative radiochemotherapy have improved long-term outcomes in terms of excellent local control rates and this is independent of their initial clinical T and N stage [45,46]. The increased incidence of pCR in the radiochemotherapy arms did not improve the final outcome of the randomized studies [40–43].

To increase the efficacy of bolus or infused 5-FU or capecitabine these agents have been combined, in several phase II studies, with oxaliplatin or irinotecan plus radiation. The apparently positive results of these studies have supported many ongoing phase III studies. At the present, infused 5-FU as well as oral fluoropyrimidines remain the standard agents to combine with preoperative radiotherapy [1]. Two studies have recently reported early results from randomized phase III trials evaluating the role of adding oxaliplatin to chemoradiotherapy with 5-FU. Neither the Prodige 2-ACCORD 12/0405 trial [47], nor the STAR-01 trial [48] could report any significant benefit in pCR rates.

About 20% of cT3N0 patients are overstaged and have cT1-2N0 disease and therefore overtreated with preoperative radio-chemotherapy. However, an even larger number would be understaged since following preoperative radiochemotherapy, 22% will have ypN+ disease. These data illustrate the weaknesses of nodal staging by imaging [49].

Sphincter saving after preoperative radio(chemo)therapy. Sphincter preservation is usually considered when tumor is found in the lower third of the rectum. Since the mesorectum decreases in size close to the top of the anal canal, tumors arising in this area can easily invade surrounding structures, such as the internal and external sphincters and the levator muscles. This is common if the depth of invasion is beyond T2. Consequently, it is crucial to ensure that the pelvic floor is free from tumor if a loco-regional curative procedure, with the sphincters intact, is to be performed in very low rectal cancers.

A non-significant improvement in sphincter saving surgery was reported in a French study which randomized patients to surgery within two weeks after completion of radiation therapy, compared with six to eight weeks. The long-interval between preoperative irradiation and surgery provided increased tumor downstaging with no detrimental effect on toxicity, but did not result in significant differences in long-term local control or survival [50]. None of the other randomized trials nor meta-analyses of the trials support the idea of increased possibilities for sphincter preservation after radio(chemo)therapy [38,51,52]. Cultural differences are significant. For example a stoma may be more or less disastrous for the patient than a local failure in southern parts of Europe and the Arabic world. Therefore, many patients from the Mediterranean areas will accept poor bowel function in preference to a stoma, and will also accept using diapers [53].

The role of adjuvant postoperative chemotherapy after preoperative radio(chemo)therapy. There is insufficient evidence on the benefit of adjuvant postoperative chemotherapy after preoperative chemoradiation to come to a consensus about its use [1].

Exploratory posthoc subgroup analyses suggest that only patients who respond and are downstaged from cT3-4 to ypT0-2 benefit from 5-FU based adjuvant chemotherapy [54]. These data supports that – as shown in other trials such as the QUASAR trial [55] or the Japanese trial investigating 5FU/FA or UFT respectively – a significant survival benefit of 3–4% with 5-FU based chemotherapy. The role of adjuvant treatment strategy after preoperative chemoradiation is still being investigated (see also below).

Both bolus and continuous infusion 5-FU as well as capecitabine have been combined in several phase II studies with oxaliplatin or irinotecan plus preoperative radiation as well delivered alone as adjuvant treatment after surgery. However, the compliance of adjuvant 5-FU/LV in the adjuvant setting in the two large randomized phase III trials was suboptimal [40,41]. Ongoing studies will help clarify the role of the combination of these new drugs.

The role of postoperative radio(chemo)therapy. The main advantage of postoperative radio(chemo)therapy is better selection of patients since it can be based on pathologic staging. Postoperative therapy was a common approach in North-America, however since 2004 this is no longer the case. The primary disadvantages include increased toxicity related to the amount of small bowel in the radiation field, a potentially more radio-resistant hypoxic postsurgical bed and, if the patient has undergone an APR, the radiation beams have to be extended to include the perineal scar.

Four randomized trials have reported data on the use of adjuvant postoperative RT alone in stages pT3 and/or N1–2 rectal cancer [38]. No single study showed an improvement in overall survival seen in the meta-analyses [35–37]. No survival advantage was observed from pelvic radiation plus elective para-aortic and liver radiation versus pelvic radiation alone [56].

In 1990, the NCI Consensus Conference, analyzing the postoperative North American chemoradiotherapy studies, stated that combined modality therapy was the standard postoperative treatment for patients with pT3 and/or N1-2 disease [57]. However, based on the German trial [58], most patients today receive preoperative chemoradiotherapy in the US.

Although the 1990 NCI Consensus Conference recommended postoperative combined modality therapy for patients with pT3 and/or N1-2 disease [57], retrospective data has suggested that a subset of patients with pT3N0 disease may not require adjuvant therapy. Reports from a US pooled analysis have identified favorable subsets of patients with pT3N0 disease who, following surgery alone, have a ten-year actuarial local recurrence rate of <10% [59]. Their data suggest that patients with upper rectal cancers who undergo a TME, have at least 12 nodes examined and have stage pT3N0 disease with an adequate radial resection margin likely do not need radiation therapy. The 4–5% benefit in local control with radiation may not be worth the risks.

With a moderate consensus it was agreed that a randomized trial suggested that radiation should start during cycle 1 rather than during cycle 3 [60]. However, more data are needed before recommending a change in sequence.

The contribution of adjuvant chemotherapy in the postoperative combined treatment has been questioned. Two European randomized trials support the argument. In a Hellenic trial [61] the chemoradiation alone arm was less toxic than the arm with four additional cycles of chemotherapy while maintaining the same efficacy. In a small Norwegian trial [62] chemoradiation had better survival than surgery alone.

Pre- vs postradio(chemo)therapy. Preoperative and postoperative therapy have been compared in randomized trials [30,58,63]. Two trials (Intergroup 0147 and NSABP R-03) closed early due to lack of accrual. The completed trial, the German Rectal Cancer Trial [58] showed fewer local recurrences and less acute and late toxicity, but no survival benefit with preoperative therapy. In one trial where short-course preoperative radiation was compared with long-course RT alone [63] and in another trial where it was compared with long-course chemoradiation [30] for the subsets with a high-risk of recurrence, more favorable results were seen in the preoperative groups.

At the present time, given the improved local control, and acute and long-term toxicity profile, patients with cT3 rectal cancer who require additional therapy to surgery (chemoradiation or short course radiotherapy) should receive it preoperatively [30,38,58,63].

T4 “unresectable”

Locally advanced tumors are defined as neoplasms extending beyond the rectal wall with infiltration to surrounding organs or structures, and/or perforation of the visceral peritoneum (c/p T4 N0-2 M0). These tumors have traditionally been look upon as “unresectable”, although previous staging using the finger of the surgeon was far from accurate in the trials.

The role of long-course preoperative radio(chemo)therapy. All patients with primarily unresectable disease should receive preoperative chemoradiation. This includes radiation in the range of 50–54 Gy plus 5FU-based chemotherapy with the goal of increasing the chances of R0 resectability [38,64]. Compared to the same radiotherapy alone, chemoradiation improved in a randomized trial local and systemic tumor control [64].

Given the limitation of the total radiation dose which can be delivered to the bulky tumor in the pelvis and the frequent problem of local recurrence, the surgeon should be “aggressive” and not risk leaving microscopic residual tumor. Extended surgery to the infiltrated organ(s) should be considered even if there is a favorable response after preoperative therapy [64].

An alternative strategy under clinical evaluation for patients who are not medically able to receive long-course chemoradiation is short-course RT followed by delayed surgery [33,34].

The role of radiotherapy intensification (altered fractionation, IORT). Although 50–90% of patients will be able to undergo a resection with negative margins, depending on the degree of tumor fixation, many still develop a local recurrence. To reduce this concomitant or sequential boosts can be delivered in the preoperative setting with the goal of increasing the dose. However, doses above about 50 Gy may be associated with a higher complication rates. Positive evidence of the role of higher doses is still to be confirmed in randomized studies [65–68].

To increase local control of unresectable rectal cancer a large single dose (10–20 Gy) of radiation by electron beam or brachytherapy (Intraoperative RT or IORT) can be delivered to the tumor bed. Many North American and European single institution studies suggest a favorable local control rate in patients who also have positive margins or microscopic residual disease [69,70]. However, not all series show a benefit [71].

The results (and recommended dose) of IORT depend on whether the margins of resection are negative or whether there is microscopic or gross residual disease. IORT does not compensate a suboptimal surgery. IORT-related toxicity increases with IORT doses >18–20 Gy.

The role of adjuvant chemotherapy. The high incidence of metastases in these patients is the rationale for the use of adjuvant chemotherapy after chemoradiation and surgery. However the definitive study in patients with rectal cancer is not available [64,72,73].

Research scenario

In this time of changing therapeutic approaches, a common standard for large heterogeneous patient groups will likely be substituted by more individualized therapies. It will depend on new evidence of more tailored diagnosis, surgery, radiotherapy and chemotherapy. The main questions addressed by ongoing research in these different fields are outlined.

Radiotherapy schedule. There is a study underway which compares short-course versus long-course preoperative radiotherapy [74,75]. Hopefully this study will define the relative tumor cell kill effect of these two schedules, and the long-term toxicities. This can create an opportunity for a more tailored approach based on stage, location of the tumor, and the prediction of the CRM.

Data from the Uppsala group, Sweden, have shown that short-course radiotherapy and delayed surgery in T4 tumors based upon MRI-staging also results in a chance of R0 resection, indicating that down-sizing will occur after this treatment regimen [33]. Similar experience has been reported from a group in Leeds, UK [34]. Ongoing studies are evaluating the role of short-course radiotherapy and delayed surgery in resectable patients.

Radiotherapy and Chemotherapy. Open questions of intensification of preoperative chemoradiation and postoperative adjuvant treatment are currently addressed by three large trials (CAO/ARO/AIO-04 in Germany, PETACC 6 in Europe, and NSABP R-04 in the US). They investigate the value of oxaliplatin in addition to preoperative chemoradiation with 5-FU (CAO/ARO/AIO-04) or capecitabine (PETACC 6) as well as in the postoperative phase for the prolonged period of four to five months. The NSABP R-04 trial compares capecitabine with 5-FU in a 2 × 2 factorial design with or without oxaliplatin.

In patients treated with 5 ×5 Gy preoperatively, postoperative chemotherapy has not been evaluated so far but is currently being tested in a randomized trial (SCRIPT trial, “Simply Capecitabine in Rectal cancer after Irradiation Plus TME”).

An Italian trial (INTERACT-LEADER) is testing a combination of preoperative radiotherapy with capacitabine and oxaliplatin versus accelerated radiotherapy by concomitant boost and only capecitabine. The cT3N0-1 MRI responding patients receive local excision, and if pCR is confirmed no further surgery is performed.

The early delivery of highly active systemic combination treatment before chemoradiation and TME is currently being investigated in phase II trials [76,77]. Both approaches indicate that treatment of advanced rectal cancer has become truely ‘multidisciplinary’, requiring improvement in all fields of surgery, radiation and chemotherapy for optimal local control and reduction of distant metastases in order to improve overall prognosis.

(Chemo)radiotherapy + Target Therapy. The next generation of clinical trials will integrate novel ‘targeted’ drugs like bevacizumab and cetuximab in both the preoperative and postoperative settings. The Epidermal Growth Factor Receptor (EGFR) is a promising target of antitumor treatment because it is involved in cell division, inhibition of apoptosis, and angiogenesis. Current trials with a traditional sequence/timing did not show improved results indicating that more intense preclinical investigations are needed to identify the relationship with molecular constraints (like K-ras) and to establish the best sequence of triple combinations [78–81].

Inhibition of Vascular endothelial growth factor (VEGF) via an anti-VEGF antibody (bevacizumab) has been shown to block the growth of a number of human cancer cell lines, including colorectal, in nude mice. Preliminary clinical data indicate significant activity, however data on safety are limited [82,83]. Several trials are ongoing regarding this issue.

In the face of current and future schedules and the increasing number of therapeutic options and intensities, translational research is urgently required for the identification of patient groups, by both clinical-pathological features and molecular and genetic markers, that will gain maximum benefit from each treatment option.

Need for quality assurance. There has been an increasing emphasis on quality assurance within clinical trials in recent European trials, particularly relating to surgical technique and histopathological assessment of the rectal cancer specimen. This needs to encompass imaging, radiotherapy and assessment of late toxicity. From the perspective of radiotherapy, future studies should include transparency of target volume definition, the use of techniques to minimize normal tissue toxicity and evaluation of approaches such as image modulated and imaged-guided radiotherapy [84]. It will also be important to identify where it is possible to use tighter margins with a reproducible patient set-up to miminize late toxicity (e.g. anal sphincter and perineum) and where greater than conventional margins are required, e.g. due to internal target volume movement.

Many of these approaches are underway or planned but it is of paramount importance that future research studies are well designed, the study group well defined and the treatments and assessment of long-term outcome (including toxicity) are standardized. Particular attention to the aspects of target volume definition, quality assurance of its application are critical components in achieving future high quality tailored studies.

Acknowledgements

Presented at the Swedish Oncology Association (SOF) meeting in Kalmar, Sweden, March 24–26, 2010.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.