Neoadjuvant short-course radiotherapy with consolidation chemotherapy for locally advanced rectal cancer: A systematic review and meta-analysis.

Abstract

Background:

Short course radiotherapy with consolidation chemotherapy (SCRT-CCT) has emerged as a promising alternative to the long course chemoradiotherapy (LCRT) regimen in locally advanced rectal cancer management. The systematic review and meta-analysis is aimed at summarizing current evidence on SCRT-CCT and comparing it to LCRT.

Material and Methods:

Electronic databases of MEDLINE, Web of Science and Cochrane library were searched using a predefined search strategy returning 3314 articles. This review included 11 studies (6 randomized trials and 5 non-randomized studies) on SCRT-CCT regimen based on seven different cohorts. Weighted arithmetic means and forest plots were generated to determine summary estimates.

Results:

The probability of achieving pathological complete response (pCR) was higher with SCRT-CCT compared to LCRT (RR=1.75, 95% CI: 1.41-2.19). No statistically significant difference in 3-year OS was observed between the two groups (RR=1.06, 95% CI: 0.98-1.14). The weighted arithmetic mean of 3-year overall survival (OS) and pCR was 83.6% versus 80.9% and 24.5% versus 13.6% for SCRT-CCT and LCRT, respectively. R0 resection and T-downstaging rates ranged from 69.2-100% and 47-75% for SCRT-CCT, and 71%-92.3% and 41-75% for LCRT. The regimens had similar compliance, postoperative and late toxicity, however, acute toxicity rates varied primarily due to differences in treatment protocols.

Conclusions:

This review highlights the ability of SCRT-CCT to produce improved tumor response with comparable OS, R0 resection, and T-downstaging at the cost of increased acute toxicity. However, heterogeneity in treatment protocols across studies makes it difficult to provide definitive conclusions regarding the regimen. Several ongoing trials are expected to provide further confirm evidence provide by the RAPIDO trial and detail appropriate SCRT-CCT protocols to improve oncological outcomes, minimize toxicity and determine its effectiveness as the standard-of-care for LARC patients.

Background

Globally, colorectal cancer (CRC) is the third most commonly diagnosed malignancy with over 30% arising in the rectum [1,2] . It is also the second most common cause of cancer-related deaths [1]. The molecular features of cancerous lesions in the colon and rectum are known to be different; therefore, the treatment of CRC also varies depending on the primary location [2].

Primary treatment of rectal cancer is surgery; however, neoadjuvant radiation therapy (RT) is indicated in patients with locally advanced disease, with the aim of reducing local-recurrence rates [3]. Currently, there are three regimens to deliver radiation in locally advanced rectal cancer (LARC): (i) short-course radiotherapy (SCRT; 25 Gy in 5 fractions) with immediate surgery (within 1 week) or (ii) with delayed surgery (after 4-8 weeks), and (iii) long-course chemoradiotherapy (LCRT; 45-50.4 Gy in 25-28 fractions with concurrent chemotherapy) with surgery after 6-8 weeks [4]. Generally, SCRT with immediate surgery is used for patients with intermediate risk factors for local recurrence, while LCRT for large tumors or N2 nodal disease. These approaches have similar surgical, long-term oncologic, late toxicity and overall quality of life outcomes [5-9]. The LCRT approach is associated with greater tumor response, as measured by downstaging and higher pathologic complete response (pCR) rates [10]. On the other hand, the SCRT regimen is an appealing option considering treatment cost, patient compliance and complications [11].

Tumor response to radiation takes time. Considering that surgery after SCRT is undertaken much earlier compared to LCRT, the question was asked whether extending the interval between end of SCRT and surgery will produce an equivalent tumor response. Recently, studies such as the Stockholm III trial have demonstrated that delayed surgery after SCRT provides an improved tumor response and a better safety profile compared to immediate surgery [12,13]. Historically, SCRT regimens did not include systemic chemotherapy, however, there is a growing interest in combining chemotherapy before, during or after SCRT, with the goal of further increasing tumor response [14-18]. Of these new approaches, SCRT followed by consolidation chemotherapy (CCT) seems to be the most promising alternative to the traditional LCRT regimen. Several studies including recently published RAPIDO trial have sought to understand the short-term and long-term effects of this novel SCRT-CCT regimen, where courses of chemotherapy are administered in the interval between SCRT and surgery. This regimen is believed to improve local control and response of the primary tumor due to the additive effects of a long waiting period and chemotherapy, while curbing adverse events.

The aim of this systematic review and meta-analysis of the literature is to summarize current data on the effectiveness of SCRT followed by CCT as neoadjuvant treatment for LARC, in terms of pCR, R0 resection rates and overall survival (OS). Additionally, evidence on the toxicity, compliance and other relevant information are also reviewed.

Methods and Material

Search strategy

The systematic review and meta-analysis was based on a PICO format (patients, interventions, comparisons, outcomes) question and performed in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement. A search of the electronic databases MEDLINE, Web of Science and Cochrane Library was performed to identify relevant articles, without using any filters. Additionally, backward citation chaining of reference lists from all suitable full-text articles was performed to identify pertinent articles.

Evidence acquisition

On the 4^th of January, 2021, two independent researchers (AP, PS) searched the abovementioned databases to identify eligible studies. The search strategy was as follows: (sequential OR consecutive OR concomitant OR consolidation) and (neoadjuvant OR preoperative) and (radiotherapy OR (radiation and therapy) OR (radiation and treatment) OR chemotherapy OR chemoradiotherapy OR radiochemotherapy) and (colon OR colorectal OR rectal OR rectum) and (cancer OR carcinoma OR adenocarcinoma). The initial search returned a total of 3314 studies. The screening protocol is shown as PRISMA flowchart. (Figure 1) Disagreements between researchers were resolved through discussion amongst them and consultation with another author (JK). The following data was retrieved from the included studies: first author, year of publication, enrollment period, center/country/clinical trial, study design, TNM stage of rectal cancer, number of patients, percentage of male patients, age, SCRT dose and modality, CCT regimen, treatment protocol, details on compliance with neoadjuvant treatment protocol, R0 resection, T-downstaging, pCR, OS and disease-free survival (DFS). Data on these parameters was also collected for the control groups if implemented. Additionally, information on acute toxicity, postoperative complication and late toxicities, as well as other relevant findings from each study was retrieved and described.

Figure 1:

Prisma Flowchart

PICO-based inclusion and exclusion criteria

The PICO-based questions utilized to identify relevant studies for inclusion in analysis was as follows: Patients – patients with LARC (defined as WHO II or III or IV (if T3-4 or n+)); Intervention – SCRT followed by CCT as neoadjuvant therapy; Comparisons – LCRT-treated control groups (if present); Outcomes – pCR, OS, DFS, toxicity rates and compliance rates. The inclusion criteria were any full text, original study articles on SCRT-CCT reporting on pCR, with or without LCRT control groups. Articles were excluded if they were not original studies (editorials, reviews and clinical trial protocols) and if they did not comply to PICO (studies on other SCRT schemes such as with induction, concurrent or no additional chemotherapy). Additionally, data from conference papers on studies focusing on SCRT-CCT was also retrieved and are presented as supplementary material.

Evidence synthesis and Risk of Bias assessment

The details regarding evidence synthesis and risk of bias assessment are provided in the supplementary material.

Statistical Analysis

Weighted arithmetic mean for pCR and 3-year OS was calculated to obtain a summary estimate. It was calculated by multiplying the weight of each study (number of patients in the study/total number of patients) with the outcome (pCR/3-year OS) value [21]. If several studies reported findings from the same cohort over different time points, data from the latest available study was used. A random-effects model and Mantel-Haenszel (M-H) test were utilized to pool data in terms of relative risk (RR) for pCR and 3-year OS reported in randomized clinical trials, which included an LCRT control group and were the latest reports for their respective cohorts. A forest plot was created for visual interpretation of the summary estimates. Statistical heterogeneity across studies was assessed using the I² coefficient. Excel version 19.11 and RevMan 5.4.1 were used for data abstraction and statistical analysis.

Results

After screening and full-text assessment (38 articles), 11 articles were included in this review. Of which, 4 were phase II trials, 4 phase III trials, 1 randomized controlled trials, 1 retrospective and 1 prospective cohort study. The characteristics of the included studies are presented in table 1.

Table 1:

Characteristics of included studies.

	Author	Enrollment Years	Design	Stage (TNM)	Groups	N of Patients	Male%	Age2	RT dose (Gy)	CT regimen	Interval Period7	Treatment Compliance%8
Washington Cohort	Markovina et al. (2017)	2009-2012	P II	cT3-4 N0-2 M0	Study	69	71	57.2	25	mFOLFOX-6 (4 cycles)	11-17	96.7
					Control	69	67	56.6	45	5FU or capecitabine	6-8	100
	Myerson et al. (2014)	2009-2012	P II (single arm)	cT3-4 N0-2 M0-1	Study	76	71	56.43	25	mFOLFOX-6 (4 cycles)	11-17	56
					Control
Polish Cohort	Ciseł et al. (2019)	2008-2014	P III	cT3-4 N0-2	Study	261	70	60	25	FOLFOX-4 (3 cycles)	12	NR
					Control	254	67	60	50.4	5FU+Leucovorin plus Oxaliplatin (2 cycles)4	6	NR
	Bujko et al. (2016)	2008-2014	P III	cT3-4 N0-2	Study	261	70	60	25	FOLFOX-4 (3 cycles)	12	63
					Control	254	67	60	50.4	5FU+Leucovorin plus Oxaliplatin (2 cycles)4	6	66
	Bujko et al. (2013)	2008-2010	P III	cT3-4 N0-2	Study	49	67	60	25	FOLFOX-4 (3 cycles)	12	73.5
					Control	48	69	59	50.4	5FU+Leucovorin plus Oxaliplatin (2 cycles)	6	72.9
Iranian Cohort	Aghili et al. (2020)	2016-2020	RCT	cT3-4 N0-2 M0	Study	33	55	56	25	concurrent XELOX; Consolidative XELOX (3-4 cycles)	15-20	87.9
					Control	27	63	53	50-50.4	concurrent Capecitabine; Consolidative XELOX (3-4 cycles)	15-20	81.8
	Aghili et al. (2018)	2013-2015	P II (single arm)	cT3-4 N0-2 M0	Study	33	73	61	25	concurrent XELOX; Consolidative XELOX (1 cycle)	7-9	87.9
					Control
Baltimore Cohort	Jia et al. (2019)	2017-2019	R	cT2-4 N0-2	Study	26	77	52	25	mFOLFOX-6 (25 patients), CapeOX (1 patient)5	14.5	81
					Control
Danish Cohort	Van Dijk et al. (2013)	2006-2010	P II (single arm)	cT2-4 N0-2 M1	Study	50	54	59	25	CapeOX-Bevacizumab (6 cycles)	26	84
					Control
RAPIDO trial	Bahadoer et al. (2020)	2011-2016	P III	cT2-4 N0-2 M0	Study	462	65	62	25	CapeOX (6 cycles) or FOLFOX4 (9 cycles)6	24	85
					Control	450	69	62	50.4 or 50*	Capecitabine	6-10	90
Indian Cohort	Thakur et al. (2020) 1	2015-2016	Prospective	NR	Study	14	NR	NR	25	CapeOX (2 cycles)	11-13	100
					Control	13	NR	NR	45	Capecitabine	4-6	87

(RCT = randomized clinical trial; P II/III = phase II/III; R = retrospective cohort, Pr = prospective cohort, N = number, NR = not reported, RT = radiotherapy; CT = chemotherapy; FOLFOX = fluorouracil, leucovorin, and oxaliplatin; 5FU = fluorouracil; XELOX/CapeOX = capecitabine and oxaliplatin)

¹Specific data on CRC stages, Male% and Age were not reported (NR) by Thakur et al., however, the authors state that the study and control groups were comparable in these regards.

²Median Age reported in years.

³Mean Age reported in years.

⁴Oxaliplatin delivery left to local institution’s discretion after 2012.

⁵Number of chemotherapy cycles were left at the treating physician’s discretion

⁶Either chemotherapy regimen was used depending on the decision of the treating physician and hospital policy

⁷Approximate intervals between completion of radiotherapy (SCRT or LCRT) and Surgery (in weeks)

⁸Based on need for dose reduction and/or treatment delay due to toxicity or the proportion of patients completing the entire intended treatment

Two studies (Markovina et al., Myerson et al.) [22,23] reported findings from the Washington cohort, 3 on the Polish cohort (Cisel et al, Bujko et al. (2016), Bujko et al. (2013)) [24-26], 2 on the Iranian cohort (Aghili et al. (2020), Aghili et al. (2019)) [27,28] while each of remaining 4 reported on the Baltimore (Jia et al.) [29], Dutch (Van Dijk et al.) [30], RAPIDO trial (Bahadoer et al.) [31] and Indian (Thakur et al.) [32] cohorts. Eight studies included stage II or III rectal adenocarcinoma patients [22,24-29,31], while two studies additionally included stage IV patients [23,30]. Only one study (Thakur et al.) did not report tumor stage, but the authors mention that their groups were comparable in this regard [32].

A total of 1728 patients were examined over the seven cohorts including 915 SCRT-CCT treated patients (study group) and 813 LCRT treated patients (control group). Across studies, the percentage of male patients ranged from 54-77%, and median age ranged from 52-62 years.

Treatment Regimen

The information relating to the treatment regimen and patient compliance is presented in table 1 and supplementary table 1.

The neoadjuvant regimen for SCRT-CCT group comprised on 25 Gy radiotherapy (5 Gy x 5 days) followed by CCT. The FOLFOX regimen (either modified FOLFOX-6 or FOLFOX-4) was given as consolidative treatment in the Washington (4 cycles), Polish (3 cycles), Baltimore, and RAPIDO (9 cycles) cohorts. The RAPIDO cohort, additionally, included patients treated with the CapeOX regimen (6 cycles) [31]. The CapeOX regimen was also used in the Dutch (6 cycles, in addition to Bevacizumab) and Indian (2 cycles) cohorts [30,32]. The Iranian cohort (Aghili et al. 2020) utilized XELOX regimen which was given concurrently with SCRT as well as after SCRT as consolidative therapy (3-4 cycles) [27].

In the LCRT group, the radiotherapy dose ranged from 45-50.4 Gy with concurrent chemotherapy. One study from the Washington cohort included a LCRT control group, which received either 5-flourouracil (5-FU) or capecitabine chemotherapy [22]. In the Polish cohort, 5FU, leucovorin and oxaliplatin were used as chemotherapy during LCRT. However, delivery of oxaliplatin was made non-mandatory and left to the treating center’s discretion starting from 2012 [24,25]. The RAPIDO, Indian and Iranian (Aghili et al. 2020) cohorts used only capecitabine therapy [27,31,32]. Aghili et al., in addition to capecitabine therapy, also utilized consolidative XELOX (3-4 cycle) therapy [27].

The time interval between the completion of radiotherapy and surgical resection was considerably longer with SCRT-CCT (range: 7-26 weeks) than LCRT (range: 4-20 weeks).

The information on treatment compliance was available in all but one included study [24]. It ranged from 56-100% for the SCRT-CCT regimen across cohorts [22,23,25,27-31]. SCRT-CCT group was found to have better compliance than LCRT group in the Indian cohort (dose reduction: 0 vs 8%, delay: 0 vs 5%, p<0.0001 for both) [32]. All other studies found the two regimens to be comparable in this regard.

Outcomes

The details on the outcome measures retrieved is abstracted in table 2 and supplementary table 1.

Table 2:

Outcomes abstracted from included studies.

	Author	Groups	R0 resection%	T downstaging%	pCR%	OS%	DFS/DRTF%
Washington Cohort	Markovina et al. (2017)	Study	NR	75	28	96 (3-year)	80 (3-year)
		Control	NR	41	16	88 (3-year)	70 (3-year)
	Myerson et al. (2014)	Study	NR	71	28	NR	NR
		Control
Polish Cohort	Ciseł et al. (2019)	Study	77	NR	16	49 (8-year)	43 (8-year)
		Control	71	NR	12	49 (8-year)	41 (8-year)
	Bujko et al. (2016)	Study	77	NR	16	73 (3-year)	53 (3-year)
		Control	71	NR	12	65 (3-year)	52 (3-year)
	Bujko et al. (2013)	Study	73	71	21	NR	NR
		Control	71	73	8	NR	NR
Iranian Cohort	Aghili et al. (2020)	Study	100	54.8	32.2	NR	NR
		Control	96.2	53.8	23.1	NR	NR
	Aghili et al. (2018)	Study	NR	NR	30.8	65 (3-year)	55 (3-year)
		Control
Baltimore Cohort	Jia et al. (2019)	Study	69.2	NR	35	NR	NR
		Control
Danish Cohort	Van Dijk et al. (2013)	Study	72	47	26	80 (2-year)	NR
		Control
RAPIDO trial	Bahadoer et al. (2020)	Study	90	NR	28	89.1 (3-year)	23.7 (3-year DRTF)
		Control	90	NR	14	88.8 (3-year)	30.4 (3-year DRTF)
Indian Cohort	Thakur et al. (2020)	Study	92.8	NR	0.071	NR	NR
		Control	92.3	NR	0	NR	NR

(pCR = pathological complete response; OS = overall survival; DFS = disease free survival, DRTF = disease-related treatment failure, NR = not reported)

Data on R0 resection rates was provided by all included studies except for those originating from Washington [24-27,30-33]. Across studies, the R0 resection rates ranged from 69.2-100% for SCRT-CCT group and 71-92.3% for LCRT group. There were no statistically significant differences in R0 resection rates between the regimens in any of the comparative studies. The information on T-downstaging was available for the Washington, Polish, Iranian and Dutch cohorts [22,23,26,27,30]. It ranged from 47-75% and 41-75% for the SCRT-CCT and LCRT, respectively. Only Markovina et al. (Washington) reported a statistically significant difference in T-downstaging between the two regimens (75% vs 41%, p<0.001) [22].

The data on pCR was published for all included cohorts, ranging from 7.1% to 35% and 0% to 23.1% for SCRT-CCT and LCRT regimens, respectively. In the RAPIDO trial, SCRT-CCT was found to result in higher pCR than LCRT (28% vs 14%, odds ratio=2.37, 95% CI:1.67-3.37) [31]. A prospective cohort study (Thakur et al.) found that 6.7% of SCRT-CCT- and 0% of the LCRT-treated patients achieved pCR. The weighted arithmetic mean of pCR were 24.5% and 13.6% for SCRT-CCT and LCRT regimen, respectively. On analysis for the 4 comparative trials, the probability of achieving pCR was found to be higher in the SCRT-CCT-treated patients compared to LCRT-treated patients (RR=1.75, 95% CI: 1.41-2.19, I²=0%). (Figure 2A)

Figure 2:

Forest Plot using the Mantel-Haenszel (M-H) analysis for A) pathological complete response (pCR) and B) 3-year overall survival (OS) based on parallel-arm clinical trials comparing SCRT-CCT to LCRT and reporting latest results for their respective cohorts. (SCRT-CCT = short course radiotherapy followed by consolidation chemotherapy, LCRT = long course radiochemotherapy)

The results of OS were available for the Washington, Polish, Iranian, Dutch and RAPIDO trial cohorts [24,25,28,30,31]. Van Dijk et al. (Dutch) reported an OS of 80% at 2-years after SCRT-CCT regimen [30]. Markovina et al. reported similar 3-year OS between SCRT-CCT and LCRT groups of the Washington cohort (96% vs 88%, p=0.67) [22]. Similarly, Bujko et al. (2016) (73% vs 65%, hazard ratio (HR)=0.73, 95% confidence interval (CI): 0.53-1.01) and Bahadoer et al. (89.1% vs 88.8%, HR=0.92, 95% CI: 0.67-1.25) reported comparable OS at 3-years between the two regimens in the Polish and RAPIDO cohorts [25,31]. Aghili et al. (2018) reported a 3-year OS of 65% with SCRT-CCT regimen in the Iranian cohort [28]. The weighted arithmetic mean of 3-year OS was 83.6% and 80.9% for SCRT-CCT and LCRT regimens, respectively. No statistically significant difference in 3-year OS was observed between the two groups (Risk Ratio (RR)=1.06, 95% CI: 0.98-1.14, I²=60%). (Figure 2B) Cisel et al. also reported OS at 8-years for the Polish cohort of 49% for both regimens [24]. Additionally, data on DFS and disease-related treatment failure (RAPIDO) were reported for 4 included cohorts [22,24,25,27,31]. (Table 2)

Toxicity

The incidence of acute toxicity in SCRT-CCT and LCRT groups varied across studies. However, SCRT-CCT produced comparable postoperative complications and late toxicities in all included studies.

The incidence of acute toxicity was higher in the SCRT-CCT groups of the Washington and RADIPO trial cohorts than LCRT groups. In the Washington cohort, acute grade 3-4 hematological toxicities were higher with SCRT-CCT versus LCRT (22% vs 0%, p<0.001) likely due to use of mFOLFOX6 regimen (6 cycles), which is known to have higher rates of neutropenia [22,34]. Both groups were comparable in terms of any other form of acute toxicities. In the RAPIDO cohort, any acute grade 3+ toxicities occurred in 48% of SCRT-CCT and 25% of LCRT group, possibly due to greater number of CCT cycles (CapeOX (6 cycles) or FOLFOX4 (9 cycles)) given in the former group [31].

On the other hand, acute toxicity rates were lower after SCRT-CCT in the Polish (FOLFOX-4, 3 cycles) and Indian cohorts (CAPEOX, 2 cycles), with both using considerably lower number of CCT cycles than the abovementioned studies. Bujko et al. (2016) reported lower acute toxicities with SCRT-CCT versus LCRT (75% vs 83%, p=0.006) [25]. However, neutropenia was more common after SCRT-CCT regimen (p=0.032) likely due to treatment with FOLFOX-4 regimen [34]. Thakur et al. reported that incidence of any grade 3+ acute toxicities was lower in their SCRT-CCT group than in LCRT group (14.2% vs 61.5%, p=0.011) [32].

Lastly, Aghili et al. (2020) found grade 3+ acute toxicities rates to be similar in their SCRT-CCT and LCRT groups (24.2% vs 22.2%, p=0.55) [27]. These findings are likely related to the use of similar number of consolidation XELOX cycles (3-4 cycles) in both groups.

Risk of Bias assessment and Study quality

Overall, the quality of included studies was classified as low risk of bias. (Supplementary Figure 2) The details related to risk of bias assessment is presented in supplementary material.

Discussion

To the best of our knowledge, this is the first systematic review and meta-analysis focusing on the sequential SCRT followed by CCT regimen for neoadjuvant treatment of LARC. The reviews performed to date have mainly focused on the standard SCRT regimen and its comparison with LCRT including only brief descriptions of different SCRT schemes such as with induction, consolidation, or concurrent chemotherapy [10,35-38]. This review summarizes the largest number of articles (n = 11) pertaining to the SCRT-CCT scheme. Our review highlights the ability of the SCRT-CCT regimen to achieve improved tumor response (pCR) with comparable R0 resection rates, T-downstaging and OS. Acute toxicity rates between SCRT-CCT and LCRT varied across studies, likely due to differences in CCT regimen and number of cycles given. However, SCRT-CCT appears to result in comparable postoperative complications and late toxicity. Furthermore, patient compliance with SCRT-CCT was also found to be similar to LCRT, with more than 80% of the patients completing the entire planned treatment course across studies. Recent data from the RAPIDO trial demonstrated extremely high compliance rates with the SCRT-CCT regimen [39].

pCR defined as absence of residual cancer in histopathological examination has been long investigated as a surrogate for long-term oncological outcomes. Maas et al. hypothesized that tumors responsive to neoadjuvant therapy possess a prognostically favorable biological profile, which in turn lowers the tendency of recurrence and metastasis, and improves survival [40]. Studies examining this concept have provided evidence supporting prognostication through pCR [41]. A systematic review on LCRT regimen concluded that achieving pCR was associated with statistically fewer local recurrences and distant failures, and better 5-year OS and DFS [42]. However, the rate of distant failure (8.7%) still remains considerably higher than local recurrence rate (0.7%). This presents a rationale for additional systemic therapy in either neoadjuvant or adjuvant setting. Furthermore, utilizing pCR as a prognostic indicator may in the future allow responsive patients to undergo a less radical surgery or a watch-and-wait approach. These less intensive pathways would allow for better organ preservation, an aspect becoming increasingly important in rectal cancer management [43]. Although most studies on non-operative management are related to LCRT, some reports suggest that it may be plausible with SCRT or SCRT-CCT regimens.[29,44].

The pCR rates for SCRT with immediate surgery, with delayed surgery, and LCRT have been reported to be approximately 1.7-7%, 2.7-14%, and 5-25%, respectively [42,45,46]. In this review, the pCR rates for SCRT-CCT regimen ranged from 7.1% to 35% (summary=24.5%) and LCRT regiment from 0% to 23.1% (summary=13.62%). The lack of pCR achievement in LCRT group of Thakur et al. should be interpreted with caution considering the very small sample size (28 patients) and observational design [32]. There are two aspects contributing to improved pCR with SCRT-CCT: 1) longer interval between neoadjuvant treatment and surgery, and 2) additional systemic CCT [47]. Delayed surgery after neoadjuvant treatment has previously been proven to result in superior pCR rates, with similar complications and outcomes, in comparison to immediate surgery [12]. This can be explained by the phenomenon wherein DNA damage of tumor cells occurs during radiotherapy but cell lysis occurs with delay after few weeks [48]. Preclinical studies also suggest that radiotherapy elicits T-cell immune response, which peaks within 8-10 days after irradiation [49]. A recent review suggested that optimal interval between radiotherapy and surgery may be around 11-16 days after SCRT [50]. The interval between completion of SCRT and surgery in the included studies ranged from 7-26 weeks. Furthermore, the addition of neoadjuvant CCT is known to induce local tumor response while also serving as systemic treatment of disseminated disease [51]. A phase II, multicenter trial has shown that pCR rates increased proportionally with longer interval to surgery and increasing number of CCT cycles [52].

Data regarding another important endpoint assessed by major LARC trials, DFS, was reported for three included cohorts [22,24,25,27,31]. In the Washington trials, SCRT-CCT resulted in superior 3-year DFS than LCRT regimen, which was possibly a result of two factors: 1) greater number and earlier (neoadjuvant) delivery of systemic CCT cycles compared to LCRT and 2) better patient compliance with neoadjuvant chemotherapy rather than with adjuvant chemotherapy after a major surgical resection [22]. In this cohort, SCRT-CCT also resulted in greater T-downstaging and pCR, and lower recurrence rates than LCRT. Similar explanation can be given for the lower 3-year disease-related treatment failure with SCRT-CCT noticed in RAPIDO trial [31]. However, this benefit was not noticed in the Polish trials, which reported similar 3-year DFS and other outcomes between the two regimens [24,25]. The relatively fewer CCT cycles (compared to RAPIDO and Washington trials) confounded by a lack of planned adjuvant chemotherapy (left at the local institution’s discretion) might explain these findings [22,25,31]. Additionally, it must be highlighted that, although greater CCT cycles may improve oncological outcomes, they also lead to greater acute toxicities as noticed in RAPIDO and Washington trials and vice-versa in Polish trials.

This review has several limitations. Firstly, the number of patients studied in the included cohorts was small, with only two cohorts comprising of more than 100 patients. Secondly, the treatment protocols for SCRT-CCT and LCRT regimens varied across studies. The effect of various chemotherapy regimens was not taken into consideration while calculating summary estimates. Thirdly, the tumor characteristics of included patients differed slightly across studies, with two studies additionally including patients with metastatic disease. However, the rate of pCR achieved in these studies was comparable with the rest of the included studies. Fourthly, the period of follow-up also varied across studies (range: 1.5 to 7 years). These considerations limits our ability to make precise conclusions in favor or against the SCRT-CCT regimen. Regardless, this review is the first to provide an overview of the current scientific evidence on this topic.

Several clinical trials on this regimen are either ongoing or planned such as STELLAR, LARCT-US, ESCORT, SUNRISE and seven others [16,53-57]. These are likely to shed light on several questions regarding SCRT-CCT such as the optimal CCT regimen and dose, role of adjuvant chemotherapy, its role in replacing traditional LCRT, impact on quality of life, utility in metastatic rectal cancer, and concomitant immunotherapeutic agent use. Preliminary data from the STELLAR trial have been published reporting greater acute toxicities and pCR rates with SCRT-CCT than LCRT, with comparable postoperative complications, compliance rates and R0 resection rates [58-60]. The conference abstracts of included full-text studies covered in the review and from STELLAR trial are summarized in supplementary table.

In conclusion, this systematic review presents encouraging evidence regarding the role of SCRT-CCT in the neoadjuvant treatment of LARC. Current evidence suggests that SCRT-CCT regimen improves pCR and possibly DFS, while having comparable other oncological outcomes, at the cost of increased, but acceptable, acute toxicities. Our findings are in line with the RAPIDO trial, which concluded that SCRT-CCT may now be regarded as a standard of care for LARC patients. This regimen is already gaining acceptance, or is being currently introduced in centers globally. Results from currently ongoing clinical trials are expected to provide confirm its effectiveness and inform regarding its efficient use to improve oncological outcomes and minimize related toxicities, consolidating its position as the standard of care for LARC patients.