Minimally Invasive versus Open Resection

The application of minimally invasive approaches to rectal surgery represents a significant technical advance in the management of rectal cancer. The authors identified 10 studies that assessed minimally invasive approaches to surgery for rectal cancer. Of these, the only trial that fulfilled the authors’ criteria for level 1a evidence was the Conventional versus Laparoscopic-Assisted Surgery in Colorectal Cancer (CLASICC) trial from the UK Medical Research Council. In this multicenter trial, 794 patients with colorectal cancer, of whom 381 (48%) had rectal primaries, were randomized to laparoscopic-assisted or open resection. A TME technique was used in most patients in both groups. The short-term results demonstrated that the laparoscopic-assisted group had shorter hospital stays (by 2 days). In the subgroup of rectal cancer patients who underwent LAR, there was a trend toward a higher rate of positive circumferential margins in the laparoscopic-assisted compared with the open surgery group (12% versus 6%). On long-term follow-up, however, this trend did not translate into any significant differences between groups with regard to local recurrence, 3-year DFS, or overall survival (OS). Other findings from the CLASICC trial suggested a trend toward worse sexual function for men in the laparoscopic-assisted group.

Other smaller RCTs have shown results similar to the CLASICC trial with regard to shorter postoperative hospitalizations, but longer operating room times and higher costs for the laparoscopic groups. In addition, these smaller studies have not demonstrated any differences in DFS and OS between laparoscopic and open approaches, although they are underpowered for these types of analyses. Although the results of the CLASICC trial and these smaller studies are promising, the authors take the viewpoint that further data on the long-term oncologic outcomes of laparoscopic approaches are needed before minimally invasive surgery can be considered standard for mid- to low rectal cancers. Currently, the National Cancer Institutes-sponsored American College of Surgeons Oncology Group (ACOSOG) Z6051 multicenter RCT of laparoscopic-assisted rectal cancer resection is accruing patients to add further data to these issues.

Resection versus Local Excision

The authors identified one RCT comparing local excision with transanal endoscopic microsurgery (TEM) to laparoscopic resection (LAR or APR). This trial randomized 70 patients with clinically staged T2N0, well- to moderately differentiated rectal tumors to TEM or laparoscopic resection after neoadjuvant chemoradiation. With a minimum follow-up of 5 years, no differences in local recurrence (two in the TEM and one in the resection group) or DFS were seen between the TEM and laparoscopic resection groups. These results are consistent with an earlier RCT, described in the previous review, which compared TEM with anterior resection in 50 patients with clinically staged T1N0 rectal cancers. Even combined, however, these trials are severely underpowered. Furthermore, the length of follow-up needs to be taken into consideration, as data from Memorial Sloan-Kettering Cancer Center and the University of Minnesota suggest that greater than 5 years of follow-up are necessary to fully appreciate recurrence after local excision. The results from these small RCTs therefore must be considered in conjunction with multiple large, retrospective studies that demonstrate high rates of recurrence after local excision of rectal cancers on long-term follow-up.

Reconstructive Techniques after Anterior Resection

Reconstruction options to restore intestinal continuity following anterior resection were assessed in 17 RCTs. Seven RCTs comprising 506 patients compared functional outcomes of colonic J-pouches (CJPs) versus straight anastomoses (SA) after anterior resection. Of these seven trials, six reported better short-term functional outcomes with CJPs. Three of the seven trials also included quality-of-life assessments, and, of these, one showed a difference in quality of life favoring the CPJ group.

Two RCTs compared CJPs to side-to-end anastomoses and found similar functional outcomes between techniques. Five RCTs compared CJPs with transverse coloplasty pouches (TCPs); in two of these trials, better functional outcomes were seen with CJPs, while the other three RCTs showed comparable outcomes. Additionally, Fazio and colleagues compared TCPs with SA and found no differences in functional outcomes, although the analysis for this portion of the trial was underpowered. Considered together with previous RCTs, these data on reconstructive techniques support a functional benefit to reconstruction with CJPs over SA. In the event that CJP reconstruction is not technically feasible, however, the benefit of TCPs over SA is uncertain.

Defunctioning Stomas after Resection

Two large RCTs assessed whether defunctioning stomas reduced the rate of anastomotic leaks after rectal cancer surgery. In both of these studies, a lower incidence of anastomotic leaks was found when defunctioning loop stomas were performed after LAR. Matthiessen and colleagues showed a symptomatic leak rate of 10.3% in patients who underwent defunctioning stomas (compared with 28% for patients with no stomas). An interesting finding from this study was the low likelihood of stoma reversal in patients who developed anastomic leaks. Twenty-five patients in the nondefunctioned group developed a symptomatic leak, and of these patients, only eight were stoma-free after long-term follow-up. In comparison, 12 patients in the defunctioned group developed a leak, and seven of these patients were stoma-free on follow-up.

Preoperative Radiation versus Surgery Alone

Three large, randomized studies examined preoperative RT versus surgery alone. Of these, only the RCT from the Dutch Colorectal Cancer Group presented new data, while the other two provided long-term follow-up data. The Dutch trial was a multicenter RCT that examined the added benefit of preoperative short-course RT to TME for rectal cancer. The authors included this trial as level 1a evidence because of its design and impact on practice. The distinguishing features of this trial were the rigorous efforts to standardize surgery with TME and the multiple measures instituted to ensure surgical and pathologic quality control. This level of standardization and quality control allowed the authors to address the key question of whether preoperative RT was beneficial when optimal surgical therapy with TME was performed. Patients (N = 1861) with resectable disease were randomized to short-course RT (5 × 5 Gy) followed by TME surgery or TME surgery alone. On short-term follow-up, the 2-year local recurrence rate was lower in the RT plus TME group compared with the TME alone group (2.4% versus 8.2%). On longer follow-up, the 5-year local recurrence risk remained lower in the RT plus TME group (5.6% versus 10.9%), but no difference in 5-year OS was observed between groups (64.2% versus 63.5%). Thus, these results showed that there was a benefit to preoperative RT even when optimal surgery was performed using a TME technique.

Two other RCTs reported long-term data on preoperative short-course RT for rectal cancer. The Stockholm II trial found that patients receiving preoperative RT had a lower incidence of pelvic recurrences compared with those who underwent surgery alone (12% versus 25%), but OS was not different between groups. However, the Swedish Rectal Cancer Trial, which randomized 1168 patients with resectable disease to preoperative RT or surgery alone, found improved disease-specific survival (DSS) (72% versus 62%) and OS favoring the preoperative RT group. These survival results differ from both the Dutch and the Stockholm II trials, which did not find a survival benefit with preoperative RT. Although subject to debate, some authors have pointed to the lack of standardization of TME surgery and high local recurrence rate in the Swedish trial as possible reasons for these contrasting results.

Multiple papers derived from the previously mentioned trials also have presented data related to the effects of preoperative RT on nononcologic outcomes. In terms of complications, these papers suggest a higher incidence of perioperative complications and wound problems, and more long-term complications and small bowel obstructions (SBO) for patients receiving preoperative RT. In terms of function and quality of life, patients who received preoperative RT report higher rates of sexual dysfunction and fecal incontinence, although these studies show few differences in overall health-related quality of life.

Preoperative versus Postoperative Chemoradiation

Three large RCTs comparing preoperative versus postoperative chemoradiation for clinically resectable rectal cancer were initiated, but two of these (NSABP R-03 and INT 0147) were closed early because of low accrual. Fortunately, the third trial, from the German Rectal Cancer Group (CAO/ARO/AIO 94), was completed, and the authors have included this trial as 1a level of evidence. The German trial randomized 823 patients with locally advanced rectal cancer to preoperative or postoperative chemoradiation, consisting of long-course RT (50.4 Gy in 25 fractions) and infusional fluorouracil (FU). The preoperative chemoradiation group was found to have fewer toxicities and an improved 5-year local recurrence rate compared with the postoperative chemoradiation group (6% versus 13%). There was no difference between groups with regard to 5-year OS, however.

Preoperative Short-Course Radiation versus Combined-Modality Therapy

A trial from the Polish Colorectal Group compared preoperative short-course RT with long course chemoradiation. In this trial, 312 patients with clinical T3 or T4 rectal cancers were randomized to either preoperative RT (25 Gy in five fractions) and TME surgery within 7 days or long course chemoradiation (RT 50.4 Gy in 28 fractions and infusional 5-FU and leucovorin [LV]) followed by TME surgery 4 to 6 weeks later. The results showed a higher rate of grade 3 or 4 toxicities for the chemoradiation group (18% versus 3%) but no differences in rates of sphincter preservation or postoperative complications. On long-term follow-up, there were no differences in the crude incidence of recurrence or 4-year DSS or OS between treatment groups. Unfortunately, the study was powered to detect differences of 15% or greater, and therefore was unlikely by design to detect smaller differences between groups.

Preoperative Radiation versus Surgery Alone

Three large, randomized studies examined preoperative RT versus surgery alone. Of these, only the RCT from the Dutch Colorectal Cancer Group presented new data, while the other two provided long-term follow-up data. The Dutch trial was a multicenter RCT that examined the added benefit of preoperative short-course RT to TME for rectal cancer. The authors included this trial as level 1a evidence because of its design and impact on practice. The distinguishing features of this trial were the rigorous efforts to standardize surgery with TME and the multiple measures instituted to ensure surgical and pathologic quality control. This level of standardization and quality control allowed the authors to address the key question of whether preoperative RT was beneficial when optimal surgical therapy with TME was performed. Patients (N = 1861) with resectable disease were randomized to short-course RT (5 × 5 Gy) followed by TME surgery or TME surgery alone. On short-term follow-up, the 2-year local recurrence rate was lower in the RT plus TME group compared with the TME alone group (2.4% versus 8.2%). On longer follow-up, the 5-year local recurrence risk remained lower in the RT plus TME group (5.6% versus 10.9%), but no difference in 5-year OS was observed between groups (64.2% versus 63.5%). Thus, these results showed that there was a benefit to preoperative RT even when optimal surgery was performed using a TME technique.

Two other RCTs reported long-term data on preoperative short-course RT for rectal cancer. The Stockholm II trial found that patients receiving preoperative RT had a lower incidence of pelvic recurrences compared with those who underwent surgery alone (12% versus 25%), but OS was not different between groups. However, the Swedish Rectal Cancer Trial, which randomized 1168 patients with resectable disease to preoperative RT or surgery alone, found improved disease-specific survival (DSS) (72% versus 62%) and OS favoring the preoperative RT group. These survival results differ from both the Dutch and the Stockholm II trials, which did not find a survival benefit with preoperative RT. Although subject to debate, some authors have pointed to the lack of standardization of TME surgery and high local recurrence rate in the Swedish trial as possible reasons for these contrasting results.

Multiple papers derived from the previously mentioned trials also have presented data related to the effects of preoperative RT on nononcologic outcomes. In terms of complications, these papers suggest a higher incidence of perioperative complications and wound problems, and more long-term complications and small bowel obstructions (SBO) for patients receiving preoperative RT. In terms of function and quality of life, patients who received preoperative RT report higher rates of sexual dysfunction and fecal incontinence, although these studies show few differences in overall health-related quality of life.

Preoperative versus Postoperative Chemoradiation

Three large RCTs comparing preoperative versus postoperative chemoradiation for clinically resectable rectal cancer were initiated, but two of these (NSABP R-03 and INT 0147) were closed early because of low accrual. Fortunately, the third trial, from the German Rectal Cancer Group (CAO/ARO/AIO 94), was completed, and the authors have included this trial as 1a level of evidence. The German trial randomized 823 patients with locally advanced rectal cancer to preoperative or postoperative chemoradiation, consisting of long-course RT (50.4 Gy in 25 fractions) and infusional fluorouracil (FU). The preoperative chemoradiation group was found to have fewer toxicities and an improved 5-year local recurrence rate compared with the postoperative chemoradiation group (6% versus 13%). There was no difference between groups with regard to 5-year OS, however.

Preoperative Short-Course Radiation versus Combined-Modality Therapy

A trial from the Polish Colorectal Group compared preoperative short-course RT with long course chemoradiation. In this trial, 312 patients with clinical T3 or T4 rectal cancers were randomized to either preoperative RT (25 Gy in five fractions) and TME surgery within 7 days or long course chemoradiation (RT 50.4 Gy in 28 fractions and infusional 5-FU and leucovorin [LV]) followed by TME surgery 4 to 6 weeks later. The results showed a higher rate of grade 3 or 4 toxicities for the chemoradiation group (18% versus 3%) but no differences in rates of sphincter preservation or postoperative complications. On long-term follow-up, there were no differences in the crude incidence of recurrence or 4-year DSS or OS between treatment groups. Unfortunately, the study was powered to detect differences of 15% or greater, and therefore was unlikely by design to detect smaller differences between groups.

Combined Chemoradiation versus Radiation

The European Organization for Research and Treatment of Cancer (EORTC) developed protocol 22921 to assess the effect of adding chemotherapy to preoperative RT and the value of postoperative chemotherapy in patients with rectal cancers. In this trial, patients (N=1011) with clinical T3 or T4 rectal cancer were randomized to one of four arms:

• 1.
  

  Preoperative RT (45 Gy in 25 fractions)

  
  
• 2.
  

  Preoperative combined chemoradiation (RT plus 5-FU and LV)

  
  
• 3.
  

  Preoperative RT followed by postoperative chemotherapy

  
  
• 4.
  

  Preoperative combined chemoradiation followed by postoperative chemotherapy

The 5-year local recurrence rate was found to be significantly lower in all three arms that received any chemotherapy (preoperative and/or postoperative) compared with the preoperative RT-alone group (8% to 10% versus 17%), which suggests a benefit for chemotherapy regardless of when it is administered. The addition of chemotherapy to RT (preoperatively or postoperatively), however, did not impact survival.

The authors identified one additional large RCT comparing preoperative combined chemoradiation to RT. Federation Francophone de la Cancerologie Digestive (FFCD) protocol 9203 randomized 733 patients with T3-4 Nx rectal cancer to preoperative long-course RT with or without concurrent 5-FU and LV. Consistent with EORTC 22921, this study showed a decrease in the 5-year local recurrence rate with combined chemoradiation over RT alone (8% versus 17%) but no difference in OS. Thus, these most recent data from two large RCTs both demonstrate that the addition of 5-FU to RT improves local control but not survival. These results challenge previous findings of a survival benefit to postoperative RT with concurrent chemotherapy from the older and much smaller North Central Cancer Treatment Group (NCCTG) trial.

One additional trial compared the effects of preoperative combined chemoradiation (RT plus 5-FU and LV) to RT alone in 207 patients with unresectable primary or recurrent rectal cancers. The authors defined unresectable tumors as those that were fixed on digital rectal examination or had computed tomography (CT) or magnetic resonance imaging evidence of sacral, pelvic sidewall/floor, base of bladder, or prostate invasion. The chemoradiation arm in this study had higher rates of R0 resection, local control, and DSS, although there was more toxicity in chemoradiation (CRT) group as well.

Adjuvant Chemotherapy versus Surgery

The Quick and Simple and Reliable (QUASAR) trial was a multicenter trial that randomized 3239 patients with colorectal cancer (including 948 with rectal cancer), most of whom had stage 2 disease, to chemotherapy with intravenous FU or observation following curative surgical resection. Approximately half of the rectal cancer patients in this trial received RT (either pre- or postoperatively). This study found that adjuvant chemotherapy was associated with a small reduction in the relative risk of recurrence and death when compared with observation alone, irrespective of site (colon versus rectal) or chemotherapy schedule.

However, a survival benefit with adjuvant chemotherapy in patients with locally advanced rectal cancer has not been a consistent finding in modern randomized trials when pre- or postoperative RT is given. EORTC 22921, described previously, found that the 5-year OS rate was not different for rectal cancer patients who received postoperative chemotherapy compared with those who did not receive postoperative chemotherapy (hazard ratio [HR] for death in the adjuvant chemotherapy group 0.85, 95% confidence interval [CI] 0.68 to 1.04). Another RCT of 1029 stage 2 or 3 colorectal patients (including 299 rectal cancer patients, of whom more than 50% received RT) failed to demonstrate a survival benefit with FU-based chemotherapy following surgery in the subset of rectal cancer patients. Despite these data, postoperative intravenous chemotherapy still is considered standard treatment for patients who receive preoperative chemoradiation, while further trials and meta-analyses on adjuvant chemotherapy are being conducted.

Several RCTs from Japan have investigated oral chemotherapy in the adjuvant setting. Consistent with previous results, these RCTs have demonstrated higher DFS in patients with resected rectal cancers receiving oral uracil–tegafur over no chemotherapy, although the data on OS were more inconsistent.

In addition to the previously mentioned RCTs on systemic (intravenous or oral) chemotherapy, the authors identified one large RCT that examined the role of regional chemotherapy (given by intraperitoneal or intraportal methods) combined with systemic chemotherapy for patients with stage 2 or 3 colorectal cancers. In this study, FU-based regional chemotherapy did not add any benefit with regards to recurrence or survival over systemic chemotherapy alone.

Comparisons of Chemotherapeutic Regimens

One large RCT (N = 1917) compared infusional to bolus regimens of intravenous FU in patients with resected T3 to T4 or N1 rectal cancers who also were receiving concurrent RT. The regimens were found to be comparable in terms of local recurrence, DSS, and OS. There were differences in the toxicity profiles, however, with patients receiving infusional FU experiencing fewer hematologic toxicities.

Another RCT assessed the addition of LV or levamisole, or both LV and levamisole with intravenous 5-FU alone in patients with stage 2 or 3 colorectal cancer. This trial failed to demonstrate a benefit in DFS or OS with the addition of any of these agents alone or in combination over 5-FU alone.

Finally, in a comparison of oral versus intravenous FU, an RCT of patients with clinically staged T3 to T4 and/or N-positive rectal cancers found that those treated with preoperative oral uracil, tegafur, and LV concomitant with RT experienced fewer toxicities than those treated with intravenous 5-FU and LV plus RT, with no differences in local control or suvival.