Total neoadjuvant therapy in locally advanced rectal cancer: Role of systemic chemotherapy

Abstract For the past several decades, disease‐related outcomes, particularly local recurrence rate, in patients with locally advanced rectal cancer have significantly improved as a result of advancement of surgical technique and implementation of neoadjuvant chemoradiation. However, distant metastasis remains unresolved, being a significant cause of cancer death. To focus on micrometastases early in the course of multimodal treatment, delivering systemic chemotherapy in the neoadjuvant setting is emerging. Also, driven by patient demand and interest in preserving quality of life, upfront chemotherapy prior to surgery serves as a strategy for organ preservation in the management of rectal cancer. Herein, currently available literature on different methods and strategies of the multimodal approach is critically appraised.


| INTRODUC TI ON
The multidisciplinary approach of neoadjuvant chemoradiotherapy (nCRT) followed by total mesorectal excision (TME) and postoperative adjuvant chemotherapy has been accepted worldwide as the standard treatment for patients with locally advanced rectal cancer. This therapeutic approach markedly reduced the local recurrence rate from 35% to 5%-10% and significantly improved overall survival. [1][2][3][4] Local recurrence now seems less of a concern, whereas distant metastasis is now the leading cause of cancer death in rectal cancer. 5 Although adjuvant chemotherapy has been recommended in patients with locally advanced rectal cancer treated with nCRT and TME, it does not show clear benefit in improving overall survival or cancer-specific survival. 6 Poor patient compliance with adjuvant chemotherapy is also a serious concern in clinical practice. 6,7 Therefore, the efficacy of adjuvant chemotherapy in rectal cancer treatment remains controversial.
With the purpose of improving patient survival, delivery of chemotherapy before surgery had been proposed to treat occult micrometastases early and increase treatment compliance. 8 Multiple trials evaluating various modes of incorporating both chemotherapy and CRT in the neoadjuvant setting, referred to as "total neoadjuvant therapy (TNT)," have reported optimistic results. This review discusses the rationale for TNT and available evidence from clinical trials, appraising the emerging literature on the conception of organ preservation to the current therapeutic paradigm.

| CHRONOLOGY OF THE MULTIMODAL APPROACH IN REC TAL C AN CER TRE ATMENT
In the 1980s, reports by Heald et al, including a report on the "Holy Plane" by Heald alone, were acclaimed in the history of rectal cancer surgery. 9,10 Dissecting the entire rectum through the "Holy Plane," meaning the surgical plane based on the embryological development of the hindgut, resulted in removal of the entire circumferential perirectal tissue without leaving mesorectal tissue. 10 This artful surgical technique reduced the rate of positive circumferential resection margin; consequently, the local recurrence rate dropped dramatically from 30%-50% to a single digit, and the survival outcome ultimately improved up to 80%. 1 TME also allowed patients to undergo a sphincter-saving procedure. 11 Now, the prime role of TME in rectal cancer management is indisputable.
At a similar time, the effectiveness of chemotherapy and radiation therapy (RT) was eagerly investigated for the treatment of locally advanced rectal cancer as shown in Figure 1. Before TME had gained popularity and general application in rectal cancer surgery, early clinical trials, such as NSABP R-01, NCCTG or GITSG G-7175, were conducted to test whether postoperative RT, chemotherapy, or combined chemoradiotherapy could improve survival benefit with effective local control. [12][13][14] These trials indicated that adjuvant radiotherapy could improve local control and, in a combined setting with chemotherapy, possible improvement in survival was suggested. Based on these findings, adjuvant chemotherapy with radiotherapy was recommended in the management of rectal cancer. 15 Although the trials of adjuvant therapy in a postoperative setting were mainly conducted in the USA, preoperative radiation therapy was common in European countries. The Swedish Trial evaluating preoperative short-course RT combined with conventional rectal resection demonstrated that both the 5-year local recurrence rate and the 5-year survival rate significantly improved with preoperative short-course RT, 9% vs 23% and 58% vs 48% respectively, among the curatively treated patients. 16 Along with implementation of TME, the Dutch TME trial showed a significant benefit of preoperative RT in reducing the local recurrence rate for patients with TNM stage III. 5 However, the overall survival or cancer-specific survival was similar when compared with the patients who received TME surgery only.
In terms of sequencing multimodal therapy, the German trial clearly showed the superiority of preoperative RT over postoperative therapy. 17 Similar trials indicated that preoperative radiotherapy combined with TME effectively lowered the risk of pelvic local recurrence, while high patient compliance with and low toxicity of the treatment were obtained. [18][19][20]

| CLINI C AL IMP ORTAN CE OF PATHOLOG IC AL COMPLE TE RE S P ONS E AF TER N CRT
It is well known that nCRT not only results in significant downsizing or downstaging of tumor, but also in pathological complete response (pCR) defined as the complete absence of cancer cells in the resected specimen. In trials evaluating nCRT in locally advanced rectal cancer, pCR was reported in 10%-32% of patients. [31][32][33] In general, patients who were treated with nCRT and TME did not show significant improvement in overall survival despite the excellent local control rate. 12,34 However, the prognosis in patients who obtain pCR is usually excellent, so that it is often used as a surrogate marker for good oncological outcome. 35 Analysis of the factors that might increase the rate of pCR showed that the use of continuous infusion of 5-fluorouracil, the addition of a second cytotoxic drug and high radiation dose were associated with a higher rate of pCR. 31,32 Noninferiority, phase III randomized trials comparing capecitabine to 5-fluorouracil in CRT showed that capecitabine could be an alternative in neoadjuvant or adjuvant CRT regimens. 36 Given that capecitabine functions equally as a radiosensitizer, addition of oxaliplatin to either continuous 5-fluorouracil or capecitabine and radiotherapy during chemoradiation was attempted. [37][38][39] Unfortunately, these trials failed to show improvement of tumor response but only increased toxicity.
In terms of the long-term oncological outcomes, conflicting results were reported. In the ACCORD 12 trial, the addition of oxaliplatin did not show significant improvement in 5-year disease free survival or overall survival. 40 However, in the German CAO/ARO/AIO-04 trial, 3-year disease-free survival was significantly enhanced. 41 Nevertheless, at this time, the addition of oxaliplatin to standard nCRT is not supported. Finally, although the rate of pCR was higher with increasing radiation dose, a dose-response effect beyond 45 Gy was not recommended due to a lack of data. 31,32 Therefore, 5-fluoropyrimidine-based nCRT remains the current standard treatment, and a way of intensifying the effect of nCRT is the primary focus of ongoing clinical trials.
One problem related to the identification of patients who have achieved pCR is the optimal interval from the end of chemoradiation to surgery. It is known that the effect of chemoradiation takes time for a result to be seen, and complete clinical regression of tumor could be achieved when surgery occurred after a longer time interval. 42 The Lyon R90-01 trial showed that patients undergoing surgery at 6-8 weeks after completion of nCRT had significantly increased rates of pCR compared to those operated at 2 weeks after the completion of nCRT, 26% vs 10% respectively, supporting the currently accepted timing of surgery. 43 The 17-year follow-up report indicated that the long-term oncological outcome was similar in the two groups: local recurrence was 14% in the 6-to 8-week group and 12% in the 2-week group, and the overall survival was 42% and 40%, respectively. 33 With a longer interval after therapy, tumor shrinkage and downstaging seemed to be more apparent. Other retrospective studies and meta-analyses also support that tumor response to nCRT is time-dependent and suggest that an extended interval between completion of nCRT and surgery is an important determinant in achieving a pCR. [44][45][46][47] However, the GRECCAR-6 trial, a phase III randomized, multicenter trial evaluating the effect of interval (7 or 11 weeks) between nCRT and surgery on pCR, failed to show a significant difference in the pCR rate between the two groups. 48 The authors concluded that the longer interval resulted in

| MODE OF DELIVERING SYS TEMIC CHEMOTHER APY: CONSOLIDATION CHEMOTHER APY
In order to address the risk of disease progression and distant metastasis during a longer interval between nCRT and surgery and to improve tumor regression, delivering systemic chemotherapy after nCRT, called consolidation chemotherapy, was suggested. Currently available reports on consolidation chemotherapy are listed in intervals between the start of irradiation and surgery, approximately 12 weeks. The R0 resection rate and postoperative surgical complication rate were similar in both groups. Also, there was no difference in disease-free survival, local failure rate, and distant metastases rate in the two groups. However, the patients given short-course radiotherapy followed by consolidation chemotherapy showed better treatment compliance and lower acute toxicity than the long-course CRT group. Interestingly, the 3-year overall survival was better in the patients with short-course radiotherapy followed by consolidation chemotherapy, despite the rate of distant metastasis or local recurrence being similar in the two groups. The authors presumed that large irradiation doses result in activation of antitumor immune responses during the waiting period before surgery. Observation on longer follow ups will probably clarify this phenomenon. So far, these studies suggest that consolidation chemotherapy can be a safe and feasible option in the multimodal approach, encouraging implementation in future trials.

| MODE OF DELIVERING SYS TEMIC CHEMOTHER APY: INDUC TION CHEMOTHER APY
Another way of delivering chemotherapy before surgery is dividing adjuvant chemotherapy and delivering a limited number of cycles before nCRT, and then delivering the remaining cycles postoperatively, named induction therapy. 54 Studies assessing the effect of induction chemotherapy are listed in Table 2 Regarding long-term oncological outcome, the Spanish GCR-3 phase II trial reported that local recurrence rate and survival outcomes at 5 years of follow up were not different from the control group, although the experimental group showed lower toxicity and better compliance. 55 The CONTRE trial notably demonstrated a similar pCR rate to the consolidation chemotherapy trial conducted by Garcia-Aguilar et al. 57 This prospective single-arm study reported the outcome of patients who received eight cycles of mFOLFOX6 before CRT. Again, the compliance rate was over 90%, and the R0 resection rate was 100%. It appears that approximately 6-8 cycles of induction chemotherapy are needed to achieve a similar rate of pCR. In a future trial, the optimal duration and number of cycles of induction chemotherapy would be an important topic to address.
Two clinical studies included target agents in the induction therapy. EXPERT-C, a branching trial of EXPERT, investigated the effect of adding cetuximab to capecitabine-based induction chemotherapy followed by chemoradiotherapy, then again adding cetuximab in adjuvant chemotherapy. 61 Similar to other studies, this randomized trial indicated that compliance was over 90% in both groups, but either pCR rate or R0 resection in the two groups was not different.
AVACROSS was a phase II single-arm study evaluating the effect of bevacizumab to induction chemotherapy with capecitabine and oxaliplatin followed by chemoradiotherapy. 62 Along with high compliance with the treatment, the pCR rate was as high as 34%, and the R0 resection rate was 98%. However, postoperative morbidity occurred in 58% of patients, and 24% required surgical reintervention. The safety of adding bevacizumab to induction chemotherapy should be addressed. Nevertheless, the role of target agent in induction chemotherapy is not yet fully investigated.

| RIS K-ADAP TED APPROACH AND ORG AN PRE S ERVATION
Based on previous reports, it is quite certain that TNT, consolidation or induction chemotherapy with nCRT, provides several benefits in the treatment of locally advanced rectal cancer. It not only improves pCR rate and treatment compliance but also delivers early systemic control for possible micrometastasis. Beyond these benefits, TNT gives the opportunity to assess chemosensitivity and tumor response prior to surgery. This can lead to risk stratification and identification of patients who may not require surgery or radiation therapy. In fact, several studies reporting a nonoperative approach have suggested that with robust follow ups, a certain subgroup of patients who achieved a complete clinical response could be safely left with the rectum and have good long-term oncological outcomes. [63][64][65] However, an important concern arising from this "watch and wait" approach is how to accurately access tumor regression and identify patients with no residual disease.
Reflecting back to the past, pelvic magnetic resonance imaging (MRI) enormously contributed to the development of a multidisciplinary therapeutic approach, providing accurate clinical staging. Pelvic MRI has been used as a tool to guide preoperative decision on treatment modality, essentially determining which patients undergo nCRT before surgery. 66 whereas the poor response group is planned for upfront systemic chemotherapy before surgery. Results from this study will provide much vital information regarding the role of MRI in risk-adapted approach.
A large phase II multicentered randomized trial is also ongoing investigating the efficacy of TNT and a selective nonoperative approach in locally advanced rectal cancer. 71 In this trial, patients are randomly assigned to either an induction chemotherapy or a con- Several clinical trials, listed in Table 3, are ongoing, investigating the effectiveness and efficacy of the TNT approach. 53,70,71,[75][76][77][78] All these trials uniformly show the immense effort that is being made to intensify neoadjuvant therapy and to improve survival outcome. Outcomes of the trials will help understand tumor characteristics and provide essential information on optimization of the TNT approach. The results of these studies are earnestly anticipated.

| CON CLUS ION
Total neoadjuvant therapy offers a chance to deliver aggressive treatment against the development and progression of micrometastases, potentially increasing survival rates in locally advanced rectal cancer. Furthermore, there is tremendous interest and desire for organ preservation in rectal cancer partly driven by patients who want to preserve a decent quality of life in the modern era. The total neoadjuvant therapy approach may facilitate a greater number of patients having the potential for organ preservation. Upcoming results from multiple ongoing and future trials will assist the clinical decisions that fulfil optimal oncological outcomes as well as quality of life.