- Although colon and rectal cancers are frequently grouped as a single disease entity, there are important differences in treatment approaches and patterns of recurrence between these malignancies.
- The role of adjuvant therapy in colon cancers is well established, but available data do not support the routine use of adjuvant chemotherapy in patients with locally advanced rectal cancer receiving neoadjuvant therapy.
- The landscape of treatment for rectal cancer continues to evolve, making the role of adjuvant chemotherapy increasingly difficult to define.
Approximately 39,000 people are diagnosed with rectal adenocarcinoma in the United States annually.1 Adjuvant chemotherapy following chemoradiation and total mesorectal excision (TME) for locoregionally advanced rectal cancer has become a standard of care in the United States despite limited data.
The role of adjuvant chemotherapy in rectal cancer was first evaluated in the NSABP R-01 trial in 555 patients with Dukes’ B and C disease. Patients were randomly assigned to adjuvant chemotherapy with 5-fluorouracil (5-FU), semustine, and vincristine; adjuvant radiotherapy; or no further treatment. Both overall survival (OS) and disease-specific survival benefits were seen with the addition of chemotherapy compared to surgery alone (p = 0.05 and p = 0.006, respectively) and established adjuvant chemotherapy following rectal cancer resection as a standard of care.2 Despite wide adoption of adjuvant chemotherapy following publication, the analysis did not account for meaningful differences in baseline patient characteristics, as healthier patients were more likely to receive adjuvant chemotherapy and expected to live longer regardless of additional therapy. Furthermore, although this outcome was meaningful during this era, this study was conducted prior to routine use of neoadjuvant therapy and the TME technique.
Modern Era of Treatment
Since the publication of early trials, the treatment of rectal cancer has evolved. The practice-changing trial from Sauer et al in 2004 established neoadjuvant, as opposed to adjuvant, chemoradiation as a standard of care for rectal cancer.3 In the 11-year update, chemoradiation resulted in 7.1% cumulative incidence of local recurrence versus 10.1% with adjuvant chemoradiation.4 Additionally, changes in surgical techniques have dramatically reduced local recurrence rates for patients with rectal cancer. Prior to routine use of TME, local recurrence rates were as high as 26% in patients treated with surgery alone.5,6 TME has since become standard of care, and more recent clinical trials examining the role of neoadjuvant radiation therapy (RT) found local recurrence rates of 10.9% with TME alone.7
The EORTC 22921 study examined the role of adjuvant chemotherapy for locally advanced rectal cancer. Patients with stage T3-4 resectable rectal cancer (1,011 patients) were randomly assigned to preoperative RT, preoperative chemoradiation, preoperative RT and postoperative chemotherapy, or preoperative chemoradiation and postoperative chemotherapy. The 5-year cumulative incidence rates for local recurrence were 8.7%, 9.6%, and 7.6% for groups that received chemotherapy preoperatively, postoperatively, or both, and 17.1% in those that did not receive any chemotherapy (p = 0.002). These data established that the addition of chemotherapy in both the neoadjuvant and adjuvant settings decreases local recurrence.
Despite long-term follow-up, there was no effect on OS or disease-free survival (DFS) with the addition of postoperative chemotherapy compared to no adjuvant chemotherapy (p = 0.32 and 0.29, respectively). In an unplanned subgroup analysis, there was a DFS benefit for adjuvant chemotherapy seen in those patients with tumor down-staging following neoadjuvant therapy (p = 0.013). However, the 10-year rate of distant metastases was nearly 30% in all study arms.8-11 It is important to recognize that this study did not directly compare adjuvant chemotherapy versus observation in patients treated with neoadjuvant chemoradiation. Additionally, the trial design has been criticized for being underpowered and for using a bolus 5-FU regimen that is not standard in the United States.
A second randomized trial, by Cionini et al, directly questioned the utility of adjuvant chemotherapy for patients with cT3 or T4 disease who underwent neoadjuvant chemoradiation followed by surgery. The 655 patients in this study were randomly assigned to receive six cycles of 5-FU versus observation. At 5 years of follow-up, there was no difference in OS, local recurrence, or distant metastases.12 Notably, this trial also used a bolus 5-FU regimen.
The PROCTOR-SCRIPT trial randomly assigned patients with histologically proven stage II or III rectal cancer who received neoadjuvant RT or chemoradiation and TME to adjuvant chemotherapy versus observation. The trial accrued 437 eligible patients but closed because of poor accrual. After median follow-up at 5 years, there was no difference in OS, DFS, local recurrence, or distant metastases.13 An additional study from the United Kingdom closed prematurely after accruing 113 of an intended 800 patients. After neoadjuvant chemoradiation and resection, patients were randomly assigned to adjuvant capecitabine and oxaliplatin for six cycles or observation; compliance was poor, with only 48% of those assigned to chemotherapy completing six cycles. No differences were observed in OS or DFS.14
In an effort to improve statistical power through patient numbers, a meta-analysis of individual patient data from four European randomized trials included 1,196 patients with stage II or III rectal cancer receiving neoadjuvant therapy and surgical resection. No differences were seen in OS, DFS, or distant metastases. In these studies, compliance with adjuvant chemotherapy ranged from 43% to 73%.15 Collectively, these data do not support the routine use of adjuvant chemotherapy in local advanced rectal cancer patients treated with neoadjuvant chemoradiation and surgery.
Optimal Chemotherapy Regimen
Although colon and rectal cancers are frequently grouped as a single disease entity, there are important differences in treatment approaches and patterns of recurrence between these malignancies. Where data are lacking in the adjuvant management of rectal cancer, extrapolation from colon cancer studies is commonly used. Early trials, including the NSABP-C01 and a pooled analysis, demonstrated a 25% to 30% relative risk reduction in mortality and distant metastases with the use of adjuvant chemotherapy compared to surgery alone.2,6,16-18
The MOSAIC trial, which randomly assigned 2,246 patients with stage II-III colon cancer to either adjuvant bolus plus continuous-infusion 5-FU and leucovorin (LV5FU2) versus the same regimen with the addition of oxaliplatin (FOLFOX4), found the addition of oxaliplatin significantly improved both 5-year DFS and 6-year OS (p = 0.003 and p = 0.023, respectively).19,20
Although multiple studies have evaluated the optimal systemic therapy regimen in colon cancer, the data for selecting a specific chemotherapy regimen for adjuvant therapy in rectal cancer are sparse. The ADORE trial, a multicenter, phase II, randomized controlled study, examined the addition of oxaliplatin to a 5-FU–based adjuvant chemotherapy regimen in the setting of patients with rectal cancer who had previously been treated with neoadjuvant chemoradiation and surgery. A total of 321 patients with pathologic stage II or III (ypT3-4N0 or ypTanyN1-2) rectal cancer and were randomly assigned to receive either four cycles of 5-FU and leucovorin versus eight cycles of FOLFOX. The primary study endpoint was 3-year DFS. With a median follow-up of 38.2 months, there was a statistically significant improvement favoring the FOLFOX regimen (71.6% vs. 62.9%, HR 0.657, p = 0.047).21 General treatment-related toxicity, including fatigue, neuropathy, thrombocytopenia, and nausea, was greater with the FOLFOX regimen, although there was no difference in grade 3 or 4 events.
There are currently no trials addressing the appropriate duration of adjuvant chemotherapy for rectal cancer. The optimal duration of therapy is largely based on the experience in colon cancer where 6 months of adjuvant chemotherapy is standard. Recent results from the IDEA collaboration (six randomized trials of 6 vs. 3 months of oxaliplatin-based adjuvant therapy) reported during the 2017 ASCO Annual Meeting demonstrated a significantly higher incidence of grade 3 or 4 neurotoxicity with 6 compared to 3 months of treatment (16% vs. 3% with FOLFOX, 9% vs. 3% with CAPOX). At a median follow-up of 39 months, the 3-year DFS for 3 versus 6 months of treatment was 74.6% versus 75.5% (HR = 1.07, 95% CI 1.00-1.15).22
Extrapolating these data to rectal cancer, however, is challenging given many patients with rectal cancer are treated with neoadjuvant chemoradiation prior to surgery, whereas patients with colon cancer typically proceed directly to surgery followed by adjuvant therapy. Consensus-based guidelines, such as those from the National Comprehensive Cancer Network, currently recommend a total of 6 months of perioperative therapy for treatment of resected rectal cancer, which generally includes a 5- to 6-week course of chemoradiotherapy and 4 additional months of adjuvant chemotherapy.
Despite the paucity of data supporting adjuvant chemotherapy in patients with locally advanced rectal cancer that have undergone neoadjuvant chemoradiation followed by TME, it has been incorporated into standard treatment. The landscape of treatment for rectal cancer continues to evolve, making the question of adjuvant chemotherapy increasingly difficult to answer.
As cost of care and resource utilization becomes ever more important, some centers are beginning to incorporate a hypofractionated, neoadjuvant radiation approach (5 Gy x 5 fractions) both with and without consolidation chemotherapy.23-25 The ongoing PROSPECT trial is randomly assigning 1,000 patients to a standard arm of 5-FU or capecitabine-based chemoradiation, followed by TME and adjuvant FOLFOX, versus six cycles of neoadjuvant FOLFOX (and no chemoradiation if reassessment shows a greater than 20% response at the primary tumor), TME, and six cycles of adjuvant FOLFOX.26 Additionally, a nonoperative management approach pioneered by Habr-Gama et al suggests that patients with a clinical complete response to neoadjuvant chemoradiation many have preserved oncologic outcomes and may not derive additional benefit from surgery.27-30
In these contexts, the need for adjuvant chemotherapy has yet to be explored. The ongoing phase II trial from Memorial Sloan Kettering Cancer Center is accruing patients to further explore a nonoperative treatment approach for patients with near or complete clinical response to the neoadjuvant treatment. Additionally, this study is examining the timing of chemotherapy and whether chemotherapy given prior to chemoradiation can improve DFS.31
Sensitivity to systemic agents appears to be dependent on microsatellite status as patients with colon cancer and high microsatellite instability (MSI-H) or deficient mismatch repair enzymes (dMMR) may not benefit from adjuvant therapy.32,33 Le et al found that in patients with metastatic colorectal cancer and dMMR, delivery of anti–PD-1 antibody monotherapy resulted in high tumor response rates and increased progression-free survival compared to proficient MMR tumors.34 An ongoing phase III randomized trial is examining whether the addition of the anti–PD-L1 antibody atezolizumab to FOLFOX in patients with stage III colon cancer with dMMR or MSI-H can improve DFS compared to FOLFOX alone35; it is plausible that the use of immunotherapy could be extrapolated to rectal cancer.
Given the nearly 30% rate of distant metastases with locally advanced rectal cancer, better systemic agents are clearly needed. At present, oncologists will need to consider patients’ comorbidities, tumor stage (both pre- and post-neoadjuvant treatment), tumor response, pathologic characteristics, and postoperative performance status resulting in a nuanced discussion regarding the potential advantages and drawbacks of adjuvant chemotherapy.
About the Authors: Dr. Spiegel is chief resident of the Duke Radiation oncology residency program at the Duke University School of Medicine. Dr. Uronis is an associate professor of medicine at the Duke University School of Medicine and a member of the Duke Cancer Institute. Dr. Czito is an associate professor of radiation oncology at the Duke University School of Medicine and a member of the Duke Cancer Institute. Dr. Willett is professor and chair of radiation oncology at the Duke University School of Medicine and a member of the Duke Cancer Institute. Dr. Palta is an associate professor of radiation oncology at the Duke University School of Medicine and a member of the Duke Cancer Institute.