Short-course radiotherapy followed by neo-adjuvant chemotherapy in locally advanced rectal cancer – the RAPIDO trial

Over the past decades, management of rectal cancer has evolved immensely leading to improved patient outcomes. Although increased awareness, introduction of screening programmes [1] and enhanced perioperative care [2] have played roles in reducing mortality and morbidity rates, two distinct therapeutic developments have been of key importance. Firstly, surgical techniques have been refined [3, 4] and disseminated to the colorectal surgical community [5]. Randomized controlled studies to prove the effects of new surgical techniques are difficult to design and run, but observational data clearly indicate benefits [6‐8]. Secondly, (neo)adjuvant therapies including radiotherapy and chemotherapy have become integrated parts of rectal cancer management.

The Uppsala trial on pre- or postoperative radiotherapy for rectal cancer was the starting point for a line of development towards full acceptance for preoperative short-course radiotherapy [9]. The Stockholm I trial run in parallel [10], and the subsequent Swedish Rectal Cancer Trial with randomization between surgery alone and preoperative radiotherapy with 5 Gy x 5 in resectable rectal cancer showed not only improved local control but also a survival benefit [11]. Criticism against these trials regarding the quality of surgery performed was met in the Dutch TME trial where only surgeons trained in total mesorectal excision (TME) were operating. This study confirmed improved local control with the use of preoperative short-course radiotherapy although no significant effects on survival could be seen, apart from certain subgroups [12, 13]. Randomized data indicating decreased short-term toxicity and no differences in long-term oncologic outcomes and late toxicity compared to preoperative chemoradiation in resectable rectal cancer suggest that short-course radiation should be preferred in these patients [14].

There is an overwhelming body of evidence supporting adjuvant chemotherapy in colon cancer with indisputable effects on survival [15]. However, in rectal cancer the corresponding evidence is weaker although a recent metaanalysis [16] showed favourable outcomes with the use of fluoropyrimidine-based chemotherapy. Another systematic overview questioned its use, particularly after preoperative chemoradiotherapy [17]. The risk of developing metachronous metastases in intermediate and locally advanced rectal cancer is 25–65% [18‐20] and systemic chemotherapy aim to treat occult or micrometastatic disease that may later appear as distant metastases. At least two arguments against postoperative delivery of chemotherapy can be brought up. Firstly, rectal cancer surgery is connected with a substantial risk of postoperative complications that may lead to patient inability to tolerate postoperative chemotherapy [21, 22]. Secondly, in an early disease phase when microscopic dissemination is limited the efficacy of systemic chemotherapy would theoretically be expected to be the greatest. Surgery, particularly if extensive, may also accelerate tumor growth [23, 24]. Neoadjuvant chemotherapy has been proven favourable in the management of colorectal liver metastases [25], in gastric cancer [26, 27] and is currently being studied in colon cancer in the ongoing FOxTROT trial [28].

Magnetic resonance imaging (MRI) is currently the most accepted modality for preoperative local staging of rectal cancer. Following the MERCURY-trial and further studies there are criteria, although not yet universally accepted or used, for identification of patients with a high risk of local and/or systemic relapse [29]. For patients with early, clearly resectable tumours (also designated “good”) surgery alone may provide excellent results with respect to both local control and survival. In slightly more advanced tumours, or “intermediate” following the terminology of the European Rectal Cancer Consensus Conference (EURECA-CC2) [30] or the European Society of Medical Oncology (ESMO) Consensus Guidelines [31], the increased risk of local recurrence justifies preoperative radiotherapy. Short-course radiotherapy with immediate surgery is a valid option in this situation as it reduces the risk of recurrence by 60-70% [32, 33]. However, in this “intermediate” group (also designated “bad”), frequently termed “locally advanced”, chemoradiation is the preferred option by many [22, 30, 34]. In patients showing high-risk features on MRI, tumours best termed “locally advanced”, or “ugly”, there is a substantial risk of treatment failure either locally or systemically and chemoradiation is the reference regimen, since the addition of chemotherapy to conventionally fractionated radiotherapy improves local control and cancer-specific survival [35]. However, some patients may be considered too frail for chemoradiation and three recent reports have shown promising results with a strategy of delivering 5 Gy x 5 with delayed surgery [36‐38]. These studies, although non-randomized, support the notion that also short-course preoperative radiation results in down-staging if surgery is postponed. In addition, a Dutch phase II trial [39], in which patients with resectable metastatic rectal cancer were given short-course radiation followed by preoperative chemotherapy including bevacizumab reported high response rates and radical (R0) resection was achieved in 80% of the patients.

Following evidence of tumour down-staging or down-sizing with short-course radiation and arguments for neo-adjuvant rather than adjuvant chemotherapy there is a rationale for applying this concept on patients with rectal cancer at high risk of local or systemic failure. The present RAPIDO (Radiotherapy And Preoperative Induction therapy followed by Dedicated Operation) trial has been designed to assess whether short-course radiation followed by up-front chemotherapy before surgery improves 3-year disease-free survival (DFS) in patients with locally advanced rectal cancer compared to conventional chemoradiation with optional (according to institution policy) postoperative adjuvant chemotherapy.

Improved staging, introduction of multidisciplinary decision-making, refined surgery and appropriate use of preoperative radiotherapy, together with quality assessment by pathology and registries have all contributed to substantially lowering rates of local recurrence in rectal cancer from historical figures of above 30% to below 10% in many cohorts. Although there are certain subgroups who still suffer a high risk of not having R0 surgery or a local failure, the problem of local control can be seen as solved for a majority of rectal cancer patients. However, the improvements regarding local control achieved over the past decades are not matched by same-size improvements with respect to survival. Without compromising local therapy, it may therefore be justified to shift focus from local control to systemic control and survival when designing trials aiming at further development of rectal cancer treatment.

The primary aim of adjuvant systemic chemotherapy is to treat occult disease dissemination that may later occur as distant metastases. Current standard for locally advanced rectal cancer includes preoperative chemoradiation but, because of the risk of toxicity, dosage of chemotherapeutic agents must be reduced which may negatively affect the systemic efficacy. In many centres additional chemotherapy is administered postoperatively. However, since rectal cancer surgery is afflicted with high rates of postoperative complications, a substantial number of eligible patients are not fit to receive chemotherapy postoperatively [21, 22, 44]. In addition, when preoperative chemoradiation is administered (5 weeks) followed by surgery after 6–8 weeks and patients having to recover after surgery for 5–6 weeks, postoperative chemotherapy for suspected occult metastases cannot be delivered until after 4–5 months. On the other hand, neo-adjuvant chemotherapy yields favourable outcomes in oesophageal and gastric cancer [26, 27], in colorectal hepatic metastases [25] and is currently investigated in primary colon cancer [28]. In contrast to these trials, in which chemotherapy is delivered pre- and postoperatively, all chemotherapy is given preoperatively in the experimental arm of the RAPIDO-trial. This was also the regimen in the Dutch “M1”-trial [39]. In this trial 50 patients with metastatic rectal cancer received short-course radiation followed by 6 cycles of full-dose capecitabine/oxaliplatin and bevacizumab preoperatively. The completion rate for all 6 cycles of chemotherapy was 85% and more than 90% received 4 cycles or more, and toxicity reported was mostly mild. Thus, the assumption that preoperatively administered chemotherapy is more likely to be accepted in full-dose than concomitant or postoperative appears to be valid.

Two randomized trials have compared short-course preoperative radiotherapy to preoperative chemoradiation in patients with resectable rectal cancer [45, 46]. No evidence that chemoradiation was superior concerning local control or survival was provided in these studies. Furthermore, there were no indications of differences in late toxicity between the two regimens and short-course radiation resulted in less acute toxicity [14]. In these studies, in which patients underwent surgery the immediate week following short-course radiotherapy down-staging occurred to a much greater extent in patients who had received chemoradiation with delayed surgery. However, in an interim analysis of the on-going Stockholm III trial, also in resectable (intermediate stage) rectal cancer, the rate of pCR after short-course radiotherapy with delayed surgery was 12.5% [47]. Additionally, there are also other randomized data [39] and observational data [36‐38] indicating a down-staging effect of short-course radiotherapy, given that surgery is after a delay. Among patients who underwent resection of the primary tumour in the “M1”-trial (approximately 75% with T3/T4N+ tumours), 91% had a R0 resection and pCR was found in 27% [39]. None of the included patients was considered inoperable because of primary tumour progression and only one patient suffered local symptoms which occurred due to massive tumour response with tumour necrosis and an abscess. The observational data from Stockholm also indicate that the risk of tumour progression during the “waiting time” is low even if chemotherapy is not administered in this period [38, 47]. With this background, a rationale to test short-course radiation also for the locally advanced tumours appears to exist.

The RAPIDO-trial is designed with the aim of improving survival without compromising local control in patients with locally advanced rectal cancer. The aforementioned support in the literature for down-staging following short-course radiation with delayed surgery and the rationale for neo-adjuvant chemotherapy opens a window to test the combination of these two concepts, and to compare results with current standard being preoperative chemoradiotherapy with or without adjuvant chemotherapy. Although it is reasonable to assume that neo-adjuvant chemotherapy not only has systemic effects but also acts on the primary tumour it would be difficult to gain acceptance for a trial in which surgery is postponed after a prolonged chemotherapy period. Even if chemotherapy with the addition of biologics has improved substantially during the past decade [48], the medical therapy has limited cell kill effect and is the weakest component in the treatment armamentarium. The initial 5 Gy x 5 will prevent local progression during the prolonged chemotherapy aimed at killing all potential subclincal cancer cells, without postponing the start of the systemic therapy more than marginally. Most adjuvant schedules in colorectal cancer consist of 8 cycles (24 weeks) of chemotherapy but in the “M1”-trial only 6 cycles were given preoperatively [39], an approach that reduces the “waiting time” between short-course radiation and surgery. Bevacizumab was included in that trial but there is no evidence supporting an effect of either bevacizumab or cetuximab against sub-clinical disease [49, 50] which renders it superfluous in the RAPIDO setting.

In the RAPIDO-trial, the logistically simple approach with initial local therapy with short-course radiotherapy followed by systemic adjuvant full-dose up-front chemotherapy in 6 cycles, before local treatment is finalized with surgery according to TME-principles, is being explored. If this concept in the RAPIDO trial yields improved survival with maintained results regarding local control, the established management of locally advanced rectal cancer with preoperative chemoradiation is challenged.