Aug 10, 2016
By Bertram Wiedenmann, MD, PhD
Charité—Universitätsmedizin Berlin, Germany
Recurrence of rectal cancer following R0 surgery can be observed in approximately one-third of all cases. Systemic treatment approaches are widely used in order to reduce this rather high recurrence rate. So far, little evidence exists for the benefit of adjuvant treatment for R0, tumor-free rectal carcinoma following neoadjuvant treatment. However, some argue that the question remains open. We are faced more and more frequently with patients with middle- and lower-third rectal cancer who respond completely to neoadjuvant chemoradiotherapy (CRT). The best treatment strategy for these patients is still evolving. The question is, based on recent evidence, how should these patients specifically be treated?
Treatment of rectal cancer has dramatically evolved during the last decades, shifting toward a tailored approach based on continuously improving preoperative staging and improved response to neoadjuvant therapy, and also possibly adjuvant therapy. However, for therapeutic decisions in this context to be sufficiently decisive across a wide spectrum of potential case scenarios, we still require that the following four benchmarks be met: 1) robust long-term prospective observational studies with 2) long duration of follow-up; 3) tightly uniform inclusion criteria; and 4) consistent application of the definition of clinical complete remissions with a demonstrable and close concordance with postoperative complete remissions. Only then can we be maximally confidant about the potential oncologic benefit of resection despite the attainment of postoperative complete remissions.
In this article, two leading experts in the treatment of rectal cancer, Dr. Bengt Glimelius and Dr. Bruce D. Minsky, debate the current role and benefits of the treatment options.
The Case for Adjuvant Treatment
By Bruce D. Minsky, MD
The University of Texas MD Anderson Cancer Center
For the past 3 decades, most randomized trials have confirmed a 10% to 15% survival benefit for patients with pTanyN+ colon or rectal cancer who receive 6 months of postoperative adjuvant 5-fluorouracil (5-FU)-based chemotherapy. It should be emphasized that these data are based on patients who did not receive preoperative treatment. For patients with rectal cancer who receive preoperative chemoradiation, the role for postoperative adjuvant chemotherapy, especially for those who have a favorable response (i.e., ypT0-1N0 disease), remains an open question. Further complicating the data is the fact that preoperative chemoradiation will decrease the number of nodes harvested from the postoperative specimen; therefore, the lymph node status (ypN) stage may underestimate the regional lymph node (pN) stage and the need for postoperative chemotherapy.
In the United States, extrapolating from the postoperative adjuvant colon and rectal trials, most patients with rectal cancer who are treated with preoperative chemoradiation receive eight cycles of postoperative mFOLFOX6 regardless of the primary tumor response. Even so, the question of postoperative adjuvant chemotherapy remains controversial; in fact, an international consensus conference also examined this issue and failed to reach a definitive recommendation.1
The question is: Why does this controversy persist, considering there are multiple randomized trials that have addressed this question?
EORTC 22921 was a four-arm randomized trial of 1,011 patients who received preoperative 45 Gy with or without concurrent bolus 5-FU/leucovorin, followed by surgery with or without four cycles of postoperative 5-FU/leucovorin. 2 Local recurrence was significantly decreased in those patients who received CRT compared with radiation alone (8% to 10% vs. 17%; p < 0.001). However, there was no difference in 5-year survival between the four arms (65% for all arms). An update with 10-year follow-up confirmed no significant differences in survival among the arms.3 However, in this updated trial, only 43% of patients received 95% or more of the planned postoperative chemotherapy, which may explain the negative results. Furthermore, a subset analysis revealed that patients who responded to preoperative chemoradiation had a survival benefit from postoperative chemotherapy.4
FFCD 9203 was a two-arm trial of 742 patients randomly assigned to preoperative 45 Gy with or without bolus 5-FU/leucovorin.5 All patients were scheduled to receive postoperative chemotherapy, and 73% did. It should be emphasized that this trial only tested the impact of concurrent chemotherapy with preoperative radiation and not the role of postoperative adjuvant chemotherapy. Overall, there was a decrease in the local failure rate (8% vs. 17%; p < 0.05) and a corresponding increase in pathologic complete response (11% vs. 4%; p < 0.05) among patients in the preoperative CRT arm compared with the patients in the radiation-alone arm. However, the preoperative CRT arm did not have an overall survival (OS) benefit compared to the radiation-alone arm (68% vs. 67%).
In the CHRONICLE trial, 113 patients received preoperative 5-FU-based CRT (> 45 Gy) and were randomly assigned to six cycles of postoperative XELOX versus observation.6 The trial was designed to accrue 780 patients but closed early due to low accrual, and was therefore underpowered. Similar to the experience with adjuvant chemotherapy in other trials, only 48% of trial participants were able to complete all six cycles of XELOX. The study found no improvement in overall 3-year survival with adjuvant chemotherapy (89% vs. 88%).
The combined PROCTOR-SCRIPT trial treated 437 patients with preoperative 5x5 Gy or 50 Gy plus 5-FU-based chemotherapy; after surgery, patients were randomly assigned to postoperative 5-FU/leucovorin or capecitabine versus observation. The trial was designed to accrue 840 patients but, similar to the CHRONICLE trial, closed early due to low accrual and therefore was underpowered. Postoperative chemotherapy did not decrease the incidence of local failure (8%) or distant failure (39% vs. 35%) and had no impact on 5-year survival (79% vs. 80%).7
Another trial limited random assignment to patients with stage yp II and III disease. The goal was to determine which type of chemotherapy, if any, was beneficial among the subset of patients with a less favorable response to preoperative chemoradiation. In a subset of 321 patients with yp stage II/III disease following preoperative chemoradiation, the ADORE phase II randomized trial reported that postoperative adjuvant FOLFOX improved 3-year disease-free survival (DFS) compared to the MAYO regimen of postoperative adjuvant 5-FU/leucovorin (72% vs. 63%, p = 0.047).8
A recent meta-analysis looked at 1,196 patients with rectal cancer and compared those who received adjuvant chemotherapy after preoperative (chemo)radiotherapy and surgery to those who underwent observation alone. The study revealed that in a subset of patients with tumors 10 cm to 15 cm from the anal verge, there was an improvement in DFS and fewer distant recurrences among those treated with adjuvant chemotherapy, compared with patients who underwent observation alone.9 However, there were no significant differences in OS between patients who received adjuvant chemotherapy and those who underwent observation. In summary, although the majority of the data do not strongly support the addition of postoperative chemotherapy following preoperative chemoradiation, the practice is still open to interpretation because of the issues discussed above. As with any medical decision, the treatment chosen is a balance between benefit and risk. For patients who want to do everything possible to improve outcomes, with the understanding that they might not experience significant benefit, it is reasonable to offer eight cycles of postoperative FOLFOX.
- Valentini V, Aristei C, Glimelius B, et al. Radiother Oncol. 2009;92:148-63.
- Bosset JF, Collette L, Calias G, et al. New Engl J Med. 2006;355:1114-23.
- Bosset JF, Calais G, Mineur L, et al. Lancet Oncol. 2014;15:184-90.
- Collette L, Bosset JF, den Dulk M, et al. J Clin Oncol. 2007;25:4379-86.
- Gerard JP, Conroy T, Bonnetain F. J Clin Oncol. 2006;28:4620-5.
- Glynne-Jones R, Counsell N, Quirke P, et al. Ann Oncol. 2014;25:1356-62.
- Breugom AJ, van Gijn W, Muller EW, et al. Ann Oncol. 2015;26:696-701.
- Hong YS, Nam BH, Kim KP, et al. Lancet Oncol. 2014;15:1245-53.
- Breugom AJ, Swets M, Bosset JF, et al. Lancet Oncol. 2015;16:200-7.
The Case Against Adjuvant Treatment
By Bengt Glimelius, MD, PhD
Uppsala University, Uppsala, Sweden
Since recurrences occur after rectal carcinoma surgery, there is a need for systemic treatment aimed at eliminating the disseminated tumor deposits present in some patients. This treatment goal is not controversial. What is controversial is the question of whether adjuvant chemotherapy prevents recurrences. In theory, the need for adjuvant chemotherapy is apparent. However, benefit has not been proven; no trial has shown that preoperative treatment sufficiently improves outcomes for patients with rectal carcinoma.
The trials that have looked at the benefit of adjuvant chemotherapy have been summarized in two recent meta-analyses, one using individual data and one using aggregated data.1,2 These meta-analyses did not include the results of the German CAO/ARO/ AIO-04 study, which found that among patients randomly assigned to preoperative CRT with 5-FU/oxaliplatin and postoperative 5-FU/oxaliplatin, DFS at 3 years was superior to preoperative CRT with 5-FU alone and adjuvant bolus 5-FU/leucovorin (76% vs. 71%, respectively; HR 0.79, 95% CI [0.64, 0.98]; p = 0.03).3 However, this benefit is too small to warrant the addition of oxaliplatin to the “intermediate-risk” rectal cancers included in the trial.4 In addition, the study used two different 5-FU schedules, so it is impossible to ascribe the gain solely to the addition of oxaliplatin pre- or postoperatively. Adding oxaliplatin could be more warranted in the setting of truly locally advanced rectal cancers (cT3 mrf+, cT4, often designated “ugly”), which have a higher risk of dissemination.
In regards to the trials included in the meta-analyses, several issues must be discussed, including the quality of the trials and the reasons they do not reveal sufficient benefit. (Prior to even knowing the negative results of the recent trials, researchers claimed that adjuvant therapy in rectal cancer is a “non-issue.”5 It is indeed an issue, and many patients continue to be treated with little benefit and much harm.) First, whenever adjuvant therapy after rectal cancer surgery is discussed, clinicians and researchers extrapolate from one clinical situation to another and state that the same therapy works in colon cancer. The rectal cancer trials were not large, and some were prematurely interrupted, which is why meaningful gains may have gone undetected. Pooling the data in the meta-analyses increased power, but did not help in uncovering any benefit to adjuvant chemotherapy.
The second issue in regards to the five trials included in one of the metaanalyses is that some of the larger trials randomly assigned patients prior to preoperative treatment, thus lowering the possibility of detecting a difference in patients who were radically operated upon. In a subgroup analysis of the three trials that randomly assigned 753 patients postoperatively, there was a benefit in DFS (HR 0.79 CI [0.62, 1.00]; p = 0.047), but not in OS (p = 0.39) among patients who received adjuvant chemotherapy. In all five trials included in the meta-analysis (2,398 patients), the hazard ratio for DFS was 0.92, with a confidence interval of 0.80 to 1.04.2 In the other meta-analysis, a possible gain in high tumors (10% to 15% from the anal verge) appeared among patients who received adjuvant chemotherapy.1 Also in this meta-analysis, the hazard ratio for the DFS for all rectal cancers was 0.91 (CI [0.77, 1.07]; 1,196 patients). For high tumors, DFS was 0.59 (CI [0.40, 0.85]; p = 0.005).
Extrapolating From Rectal to Colon Cancer
Another potential issue regarding the apparent discrepancy between the colon and rectal cancer trials is that the gains have been overestimated in colon cancer. The colon cancer trials that had a surgery-alone group are old, and in the years since they were carried out, there have been many improvements, such as better staging, improved surgical and pathologic techniques, and more rigid trial methodology. These developments challenge the benefits reported in the earlier studies, at least in absolute terms.6
Sensitivity to chemotherapy in metastatic colorectal cancer has traditionally been considered the same irrespective of primary tumor location. More recently, differences have been noticed according to site,7 but this is an unlikely argument for the benefit of adjuvant therapy in colon but not in rectal cancer.
The gains in colon and rectal cancer may not be fundamentally different, although the gains in colon cancer may potentially be slightly larger since timing of the chemotherapy may be important—after diagnosis, colon cancer surgery usually follows within a few weeks and most patients are fit for adjuvant treatment within the next 6 weeks; in the trials showing adjuvant treatment in colon cancer to be effective, treatment started within 5 to 6 weeks of surgery. In rectal cancer, however, particularly after preoperative CRT, adjuvant treatment cannot start earlier than 16 to 18 weeks after surgery. There is also a tendency in rectal cancer to prolong the interval between the end of CRT (or shortcourse radiotherapy) and surgery to beyond about 6 weeks.8 This prolonged wait time may result in better downsizing and down-staging, but can result in a lower possibility for adjuvant chemotherapy to eradicate the disseminated tumor cells.9 Eradication would call for neoadjuvant systemic treatment, and trials assessing the benefit of such treatment are ongoing. In a Polish trial, 541 patients were randomly assigned to receive 5x5 Gy and three cycles of FOLFOX4 or conventional CRT preoperatively. The study found improved OS in the experimental group, however, without any gains in local or systemic control.10 The Dutch-Swedish RAPIDO trial, comparing 5x5 Gy and six cycles of CAPOX and surgery with CRT, surgery, and an optional eight cycles of CAPOX, closed patient recruitment in early June 2016 after randomly assigning 920 patients.11 Hopefully, after a lack of evidence for substantial benefit of adjuvant chemotherapy in rectal cancer, these trials will reveal sufficient gains when systemic treatment is given prior to surgery.
- Breugom AJ, Swets M, Bosset JF, et al. Lancet Oncol. 2015;16:200-7.
- Bujko K, Glimelius B, Valentini V, et al. Eur J Surg Oncol. 2015;41:713-23.
- Rödel C, Graeven U, Fietkau R, et al. Lancet Oncol. 2015;16:979-89.
- Glimelius B. Lancet Oncol. 2015;16:883-5.
- Glimelius B. Ann Oncol. 2010;21:1739- 41.
- Pahlman LA, Hohenberger WM, Matzel K, et al. J Clin Oncol. 2016;34:1297-9.
- Price TJ, Beeke C, Ullah S, et al. Cancer. 2015;121:830-5.
- Glimelius B. Front Oncol. 2014;4:50.
- Pettersson D, Lorinc E, Holm T, et al. Br J Surg. 2015;102:972-8.
- Bujko K, Wyrwicz L, Rutkowski A, et al. Ann Oncol. 2016;27:834-42.
- Nilsson PJ, van Etten B, Hospers GA, et al. BMC Cancer. 2013;13:279.
The views and opinions expressed in Current Controversies in Oncology are those of the authors alone. They do not necessarily reflect the views or positions of the Editor or of the American Society of Clinical Oncology.