Efficacy
The key evidence supporting the use of this protocol comes from a large number of trials demonstrating local control and disease specific mortality improvement with pre/post-operative (chemo-) radiation therapy vs surgery alone as summarised in the CCCG meta-analysis.rThe most notable of these studies demonstrating the benefit of adding pre-operative short-course radiation therapy to surgery are the Swedish, MRC-CR07, and Dutch TME studies.rrr Superiority of pre-operative vs post-operative chemoradiation therapy is demonstrated in three phase III trials: the German CAO/ARO/AIO-94, NSABP R-03, and Korean randomised controlled trials (RCTs).rrr The equivalence of LC-CRT and SCRT for local control, disease-free survival (DFS), and overall survival (OS) is demonstrated in the Australian TROG 01.04 & two Polish RCTs, as well as a meta-analysis of the wider literature.rrr
Neoadjuvant short-course versus surgery alone
The Dutch Colorectal Cancer Group trial randomised 1,861 patients with resectable rectal cancer (tumours not further than 15 cm from the anal verge and below the level of S1-2) to receive either pre-operative radiation therapy (25 Gy in 5 fractions) followed by TME (n = 924) or TME alone (n = 937).r 85 patients received post-operative radiation therapy. Locoregional recurrence was lower at 10 years with the addition of pre-operative radiation therapy (11 vs 5%, p<0.0001), but this did not translate to a difference in OS. Survival at 10 years was improved by SCRT in a subgroup of patients with negative circumferential resection margin and stage III disease (50 vs 40%, p = 0.032).
Figure 1: Local recurrence in the 1,748 eligible patients who underwent a macroscopically complete local resection.
RT = radiation therapy, TME = total mesorectal excision.
© Lancet Oncol 2011r
The MRC CR-07 trial randomised 1350 patients to preoperative SCRT (25 Gy in 5 fractions) or selective-postoperative LC-CRT in patients with positive margins (45 Gy in 25 fractions with concurrent fluorouracil).r TME was performed in 92% of patients. 53 patients received postoperative chemoradiation therapy and 7 received postoperative radiation therapy only. Pre-operative SCRT reduced 3-year local relapse rate (4.4% vs 10.6%, p<0.0001), 3-year DFS but did not lead to an OS benefit.
Short-course radiation therapy (SCRT) vs long-course chemoradiation therapy (LC-CRT)
The Australian TROG 01.04 trial was a phase III multicentre randomised trial involving 326 patients comparing pre-operative long-course chemoradiation therapy (LC-CRT) and short-course radiation therapy (SCRT) in patients with T3 rectal cancers.r Between 2001 and 2006, 163 patients were randomised to LC-CRT (50.4 Gy in 28 fractions, 5 days per week, with continuous infusion fluorouracil, surgery was completed 4-6 weeks after CRT), and 163 patients were randomised to SCRT (25 Gy in 5 fractions, 5 days per week, with surgery 3-7 days later). Surgery was followed by six 28 day cycles of adjuvant fluorouracil and folinic acid given daily for 5 days, starting 4-6 weeks after surgery. There were no differences in the primary outcome of 3-year LR (4.4% vs 7.5% p = 0.24), or 5-year DR (27% SC vs 30% LC) or 5-year OS (74% SC vs 70% LC). There was a numerical increase in local recurrences for distal tumours that did not reach statistical significance; 6/48 SCRT patients and 1/31 LC-CRT patients (p = 0.21).
In a similar Polish RCT of LC-CRT (n = 157) vs SCRT (n = 155), Bujko et al reported no significant difference in sphincter preservation rates (58% vs 61%, p = 0.57), 4-year OS, DFS or local recurrence (14.2% vs 9.0%, p = 0.17).r
Total neoadjuvant therapy
There is emerging evidence that select patients with locally advanced rectal cancer may benefit from upfront total neoadjuvant therapy (TNT), rather than standard of care (SOC) pre-operative (C)RT followed by surgery and adjuvant chemotherapy. The rationale for this approach includes earlier treatment of micro-metastatic disease, maximising ‘down-staging’ and resectability (and perhaps the chance of non-operative management), administering chemotherapy in the pre-operative setting when it is better tolerated, and perhaps reducing the time with a stoma. Total neoadjuvant therapy with FOLFOX or CAPOX before or after pre-op LCCRT or SCRT has been included in the recently updated NCCN guidelines for management of rectal cancer.r
Earlier studies have demonstrated feasibility of the TNT approach, as well as increased pCR rates over SOC. The ‘RAPIDO’ phase III RCT demonstrated decreased 3-year disease-related treatment failure (24% vs 30%, p = 0.019), distant metastases (20% vs 27%), and higher pCR (14% vs 28%), with similar OS, for TNT consisting of SCRT followed by chemotherapy (CAPOX 6x, FOLFOX 9x) vs SOC.r There was no unexpected toxicity, and no differences in rates of R0 resection or close resection margin, postoperative complications or quality of life. The PRODIGE 23 phase 3 RCT assessed the efficacy of mFOLFIRINOX x6 followed by LC-CRT, TME, +/- adjuvant chemotherapy vs SOC. Despite a different TNT regimen to RAPIDO, it demonstrated similar 7% absolute improvements in 3-year DFS (75.7% vs 68.5%) and 3-year metastasis-free survival (MFS) (78.8% vs 71.7%) and doubling of pathological complete response (pCR) to roughly 28%.r The OPRA phase II RCT (presented in abstract form) did not demonstrate its primary endpoint of a 10% improvement in 3-year DFS of TNT over SOC, but did demonstrate rates of organ preservation (non-operative management) following TNT of 43-59%, far higher than historical controls.r
Given the recency of this evidence for TNT and the uncertainties in patient selection and specifics of TNT protocol (e.g. choice of chemotherapy agent, SCRT vs LC-CRT, and sequencing of modalities), it has not been incorporated into the current protocol. It is anticipated that this area will become more well-defined in the next update. Based on the current available evidence, it is already reasonable to employ this approach in select patients after appropriate multidisciplinary team discussion.
Timing of surgery
Recommended timing of surgery is within 10 days of starting short course radiation therapy, although a 4-8 week delay to surgery is considered a valid alternative.rr
The Stockholm III trial randomised patients to immediate surgery (within a week after SCRT) or delayed surgery (4-8 weeks after SCRT or LCCRT). Local control was non-inferior with the longer interval to surgery and there were also no significant differences in distant metastases or OS. Acute radiation induced toxicity was recorded in 1 (< 1%) patients in the immediate surgery group and 23 (7%) in the delayed surgery arm. In a pooled analysis of the two short-course radiation therapy regimens, the risk of post-operative complications was significantly lower after short-course radiation therapy with delay to surgery (53% vs 41%, OR 0.61, p = 0.001).
SCRT with 4-8 week delay to surgery is a valid alternative to immediate surgery, primarily for its benefits in reducing surgical morbidity or for logistical considerations. It must be balanced between the possible detrimental effect of delaying the time to adjuvant chemotherapy (and therefore treatment of micrometastatic disease in high-risk patients). A longer interval to surgery has been supported in retrospective studies, in particular for patients with poor performance status, or where maximal tumour downstaging is desired.rr