Accepted Manuscript Short-course radiotherapy in the neoadjuvant treatment for rectal cancer: a systematic review and meta-analysis Bin Ma, Peng Gao, Yongxi Song, Xuanzhang Huang, Hongchi Wang, Qingzhou Xu, Shan Zhao, Zhenning Wang PII: S1533-0028(18)30295-0 DOI: 10.1016/j.clcc.2018.07.014 Reference: CLCC 494 To appear in: Clinical Colorectal Cancer Received Date: 13 June 2018 Revised Date: 23 July 2018 Accepted Date: 28 July 2018 Please cite this article as: Ma B, Gao P, Song Y, Huang X, Wang H, Xu Q, Zhao S, Wang Z, Shortcourse radiotherapy in the neoadjuvant treatment for rectal cancer: a systematic review and metaanalysis, Clinical Colorectal Cancer (2018), doi: 10.1016/j.clcc.2018.07.014. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. ACCEPTED MANUSCRIPT Title Page Short-course radiotherapy in the neoadjuvant treatment for rectal cancer: a RI PT systematic review and meta-analysis Bin Ma1, Peng Gao1, Yongxi Song1, Xuanzhang Huang1,2, Hongchi Wang1, Qingzhou Xu1, Shan Zhao1, Zhenning Wang1* Department of Surgical Oncology and General Surgery, the First Hospital of China SC 1 2 M AN U Medical University, Shenyang 110001, People’s Republic of China Department of Chemotherapy and Radiotherapy, the Second Affiliated Hospital and Yuying Children’s Hospital of Wenzhou Medical University, Wenzhou City 325027, TE D People’s Republic of China Bin Ma and Peng Gao contributed equally to this work EP *Correspondence to: Dr Zhen-ning Wang, Department of Surgical Oncology and AC C General Surgery, the First Hospital of China Medical University, Shenyang 110001, China. Tel: +86-24-83283556; Fax: +86-24-22703578; E-mail: email@example.com Funding The Special Prophase Program for National Key Basic Research Program of China (2014CB560712); Clinical Capability Construction Project for Liaoning Provincial Hospitals (LNCCC-A01-2014); the Doctoral Scientific Research Startup Foundation 1 ACCEPTED MANUSCRIPT of Liaoning Province (201601138) Acknowledgments RI PT We thank the department of Surgical Oncology of First Hospital of China Medical University and the College of China Medical University for technical assistance. SC Disclosure AC C EP TE D M AN U The authors declare that they have no competing interests. 2 ACCEPTED MANUSCRIPT MicroAbstract The effects of preoperative short-course radiotherapy (PSRT) remain controversial for rectal cancer patients. We conducted a systematic review and meta-analysis to explore RI PT the effects of PSRT and its optimized regimens when compared to preoperative long-course chemoradiotherapy (PLCRT). Our results indicated that PSRT could be the treatment of choice in comparison with PLCRT when complete respnse is not the SC primary aim. M AN U Abstract and Keywords Background: We aimed to show whether preoperative short-course radiotherapy (PSRT) could be the treatment of choice when compared to preoperative long-course chemoradiotherapy (PLCRT). TE D Methods: PubMed, Embase and Web of Science Databases were searched to conduct our systematic review and meta-analysis. Perioperative and survival outcomes between PSRT and PLCRT were selected as endpoints for our meta-analysis. In EP addition, Health related quality of life (HRQL) outcomes were also systematically AC C reviewed between PSRT and PLCRT. Finally, we also reviewed evidence of optimized regimens of PSRT (with delayed surgery or adding consolidation chemotherapy). Results: PLCRT showed a better pathological complete response (pCR) rate (OR = 0.05, 95% CI = 0.02-0.18, P < 0.01), but this benefit did not translate into a higher sphincter preservation rate (OR = 1.62, 95% CI = 0.72-3.67, P = 0.25) and other perioperative outcomes differences. In terms of survival outcomes, adding either PLCRT or PSRT both showed obvious advantage on local control when compared 3 ACCEPTED MANUSCRIPT with surgery alone, and PSRT and PLCRT had similar long-term outcomes irrespective of pairwise or network meta-analyses. Moreover, PSRT and PLCRT also had no overall differences based on HRQL scores. The insufficiency on pCR of PSRT RI PT might be improved by delayed surgery or adding consolidation chemotherapy based on systematic review of current evidence. Conclusions: Our results indicated that PSRT could be the treatment of choice in SC comparison with PLCRT when pCR is not the primary aim. PSRT with delayed evolution of neoadjuvant era. Keywords M AN U surgery or adding consolidation may provide further possibilities for advancing the Preoperative short-course radiotherapy, Preoperative long-course chemoradiotherapy, AC C EP TE D rectal cancer, neoadjuvant therapy, meta-analysis 4 ACCEPTED MANUSCRIPT Text Background Chemotherapy, radiotherapy, and surgery have been the primary methods for treating 1,2 . Based on the National Comprehensive Cancer RI PT rectal cancer over the past decades Network (NCCN) Clinical Practical Guidelines in Oncology 3, there are two primary neoadjuvant regimens accepted as standard for rectal cancer patients: preoperative SC long-course radiotherapy with concomitant chemotherapy (PLCRT) and preoperative M AN U short-course radiotherapy (PSRT). Both treatments are effective for local control and carry a low morbidity. PSRT is more accepted in northern Europe, whereas PLCRT enjoys more support in the United States and the rest of Europe 4. Whether PSRT or PLCRT is the preferred regimen has been a continuous topic of TE D debate 4-7, and two famous randomized controlled trials (RCTs) comparing PSRT and PLCRT have been performed 8,9 . Notably, a published investigation indicated that only 55% of rectal cancer patients received “standard” PLCRT in the United States EP according to the National Cancer Data Base 10,11. This result again put this regimen of AC C PLCRT under question. For patients, this situation could be partly explained by the higher cost and lower compliance seen for PLCRT 12. Although currently PLCRT may not require chemotherapy infusion devices, as oral xeloda is often being used, increased 5-fold radiotherapy schedule time and associated costs in managing additional acute toxicity were issues remained for patients received PLCRT5. For clinicians, does this indicate that we should reappraise our current label of this as the standard therapy? Robust evidence is needed to define the optimal, simple and 5 ACCEPTED MANUSCRIPT equivalent neoadjuvant therapy pattern. Accordingly, we aim to explore whether PSRT can be an alternative choice compared to conventional PLCRT for rectal cancer patients through a comprehensive RI PT meta-analysis based on perioperative, long-term and health related quality of life (HRQL) outcomes. In addition, we explore whether optimized PSRT regimens could advance the future evolution of the neoadjuvant era. SC Methods M AN U Search strategy Searches were conducted using PubMed, Embase and Web of Science databases (for studies published up to August 2017) to identify all relevant studies (restricted to the English language). The following search algorithm was used in PubMed: “(rectal) TE D AND (neoadjuvant OR preoperative OR preoperatively) AND (chemotherapy OR radiotherapy OR irradiation OR radiation OR chemoradiotherapy OR chemoradiation OR radiochemotherapy)”. Furthermore, if multiple studies were published on the EP same patient population based on the same outcomes, only the most informative study AC C was included. If multiple studies reported different outcomes based on the same patient population, results were combined for a more comprehensive analysis. For comparative studies with multiple-arms, data was only extracted from the arms that matched the eligibility criteria. Selection criteria PICOS criteria (population, intervention, comparison, outcomes, and study design) was used to select eligible studies: (1) population: patients who were definitely 6 ACCEPTED MANUSCRIPT diagnosed with rectal cancer without metastases and received neoadjuvant treatment; (2) intervention: PSRT (25Gy, 5fractions) or PLCRT followed by surgical resections; (3) comparison: PSRT versus PLCRT; (4) outcomes: anastomotic leakage, bowel RI PT obstruction, total complications, total postoperative mortality, pathological complete response (pCR) rate, sphincter preservation rate, acute toxicities, late toxicities, survival or HRQL outcomes based on questionnaire; (5) studies: comparative studies SC including RCTs, prospective comparative studies (PCRS), or retrospective M AN U comparative studies (RCRS). Data extraction and study quality assessment Data extraction and study quality assessment were performed independently by two reviewers (Bin Ma and Peng Gao). Extraction data that was consistent between the TE D two reviewers were used directly for the final analyses, and disagreements between the two reviewers were discussed with a third reviewer (Hongchi Wang) to reach a final consensus. The Newcastle-Ottawa Scale (NOS) criterion recommended by the 13 . In addition, we appraised the risk of bias of each eligible individual study AC C studies EP Cochrane Library for included trials was used to evaluate the quality of the included by using the Cochrane risk of bias assessment tool (www. cochrane-handbook.org). Moreover, we used the grading of recommendations assessment, development and evaluation (GRADE) method to rate the level of outcomes (www. gradepro.org). The extracted data was used for the following aspects: (1) direct pairwise analysis of comparative studies based on perioperative and long-term outcomes; (2) network analysis of RCTs (PSRT, PLCRT and surgery alone) based on long-term outcomes; (3) 7 ACCEPTED MANUSCRIPT HRQL evaluation of cohort studies based on the QLQ-C30 questionnaire; (4) systematic review of the neoadjuvant era of comparative studies based on optimized PSRT regimens (delayed surgery or adding consolidation chemotherapy). RI PT Statistical methods For the direct pairwise meta-analysis, Stata software, version 12.0 (2011) (Stata Corp, College Station, TX, USA) was used to compare PSRT and PLCRT based on SC perioperative and long-term outcomes. Odds ratios (ORs) were used to evaluate the M AN U perioperative outcomes, and the hazard ratio (HR) was used to evaluate long-term outcomes. To further explore whether PSRT can be an alternative compared with conventional PLCRT in terms of patients’ long-term outcomes, network meta-analyses were conducted by the Bayesian model using the Markov chain Monte TE D Carlo method of WinBUGS 1.4.3 (MRC Biostatistics Unit, Cambridge, UK) based on RCTs of the network relationship between PLCRT, PSRT, and surgery alone calculation model proposed by Woods et al 15 14 . The was applied, because it possesses the EP superior feature of preserving the randomization of the RCTs and avoids potential AC C selection bias, misleading results, and loss of available treatment comparisons 15. All analyses in this study used a random-effects model, due to its conservative estimates and being more tailored to multicenter studies in which heterogeneity is usually present 16. Publication bias were checked by using begg’s test among studies of PSRT versus surgery alone, PLCRT versus surgery alone and PSRT versus PLCRT 17,18. All statistical values were reported with a 95% confidence interval (CI), and the two-tailed P value threshold for statistical significance was set at 0.05. 8 ACCEPTED MANUSCRIPT Results Characteristics of included studies A total of 18573 relevant studies (records from PubMed: n= 8846; records from other RI PT databases: n= 9727) were initially identified based on our systematic study search strategy. After removing duplicates between multiple databases, 9326 studies remained for screening. After screening the title and abstracts, 58 studies remained for 9,19-36 . Our study selection process is shown in Figure M AN U included for our meta-analysis SC further evaluation. Finally, data from 19 studies based on 14 different trials were 1. Data from eight studies based on 5 different trials were included for the direct pairwise analysis in terms of perioperative and long-term outcomes 9,19,20,24,26,29,30,35 . Long-term data from 7 RCTs were included for the network meta-analysis (PSRT TE D versus Surgery alone, n = 4; PLCRT versus Surgery alone, n = 1; PSRT versus PLCRT, n = 2) 9,24,28,31-34. Three trials were included for the review of HRQL outcomes 21,23,36. In addition, three trials were summarized for the optimized regimens of PSRT EP (delayed surgery, n = 1; adding consolidation chemotherapy, n = 2) in terms of 22,25,27 . Detailed information for the included AC C perioperative and long-term outcomes studies is summarized in Table 1. In addition, the risk of bias graph of individual studies was shown in Supplementary Figure 1. Pairwise meta-analysis of perioperative and long-term outcomes between PSRT and PLCRT Based on included comparative studies between PSRT and PLCRT, the pooled results showed both methods had similar total postoperative complications (OR = 1.19, 95% 9 ACCEPTED MANUSCRIPT CI = 0.86-1.65, P = 0.30). However, patients who received PSRT had significant lower incidence rate of total acute toxicities in comparison with that of PLCRT (OR = 0.09, 95% CI = 0.02-0.35, P < 0.01). Meanwhile, PLCRT had significant better effects RI PT on pCR when compared to PSRT (OR = 0.05, 95% CI = 0.02-0.18, P < 0.01), however, this benefit of PLCRT did not translate into a higher sphincter preservation rate in comparison with PSRT (OR = 1.62, 95% CI = 0.72-3.67, P = 0.25) SC (Supplementary Figure 2). In terms of long-term outcomes, PSRT showed similar M AN U tendencies of overall survival (OS), disease-free survival (DFS), local recurrence (LR) and distant metastasis (DM) when compared with those treated with treatment of PLCRT (OS: HR = 0.88, 95% CI = 0.68-1.14, P = 0.34; DFS: HR = 0.72, 95% CI = 0.46-1.12, P = 0.14; LR: HR = 0.75, 95% CI = 0.48-1.17, P = 0.21; DM: HR = 0.97, TE D 95% CI = 0.75-1.27, P = 0.84) (Supplementary Figure 3). Notably, because of the evidence level is much higher for RCTs, hence, we conducted subgroup analysis based on the Polish trial and TROG 01.04 trial to further EP explore the difference between PSRT and PLCRT. We found that PSRT had similar AC C long-term outcomes in comparison with PLCRT (OS: HR = 0.95, 95% CI = 0.72-1.25, P = 0.72; LR: HR = 0.89, 95% CI = 0.44-1.78, P = 0.74; DM: HR = 0.92, 95% CI = 0.69-1.21, P = 0.55). Regarding to perioperative outcomes, PSRT also had similar incidence rate of postoperative complications irrespective of anastomosis leakage or bowel obstruction (total: OR = 1.20, 95% CI = 0.86-1.69, P = 0.29; anastomosis leakage: OR = 1.40, 95% CI = 0.64-3.06, P = 0.40; bowel obstruction: OR = 0.91, 95% CI = 0.46-1.81, P = 0.78). Meanwhile, both methods showed similar postoperative 10 ACCEPTED MANUSCRIPT mortality (OR = 1.26, 95% CI = 0.33-4.81, P = 0.74). Although PLCRT had significant better effects on pCR (OR = 0.06, 95% CI = 0.02-0.19, P < 0.01), this benefit did not translate into a higher sphincter preservation rate (OR = 0.87, 95% CI RI PT = 0.56-1.37, P = 0.56). However, PSRT had significant lower rate of grade 3-4 acute toxicities (OR = 0.10, 95% CI = 0.03-0.27, P < 0.01) and similar severe late toxicities (OR = 1.02, 95% CI = 0.48-2.16, P = 0.96) in comparison with PLRT. The results SC presented in forest plot were shown in Figure 2. The evidence level was recorded as RCTs (Supplementary Table 1). M AN U “high” using GRADE in terms of perioperative and long-term outcomes based on Network meta-analysis of long-term outcomes between PSRT, PLCRT, and surgery alone TE D To further explore the long-term efficacy between PRST and PLCRT, we conducted a network meta-analysis between PSRT, PLCRT, and surgery alone based on published RCTs. The network meta-analysis showed that PLCRT and PSRT have similar OS and EP DM compared to surgery alone. However, PLCRT and PSRT both showed obvious AC C lower LR compared with surgery alone (PSRT: HR = 0.47, 95% CI = 0.36-0.64; PLCRT: HR = 0.61, 95% CI = 0.33-1.15). Furthermore, the benefits of the long-term outcomes were similar between PLCRT and PSRT based on the network meta-analysis results (PSRT versus PLCRT, OS: HR = 0.95, 95% CI = 0.66-1.35; DM: HR = 0.93, 95% CI = 0.67-1.30; LR: HR = 0.78, 95% CI = 0.43-1.38; Figure 3). In addition, consistency could be observed between direct pairwise meta-analysis evidence and network meta-analysis evidence in terms of long-term outcomes 11 ACCEPTED MANUSCRIPT (Supplementary Table 2). We did not observe significant publication bias among studies of PSRT versus surgery alone, PLCRT versus surgery alone and PSRT versus PLCRT (Supplementary Figure 4). RI PT HRQL outcomes between PSRT and PLCRT Three trials reported the HRQL between PLCRT and PSRT 21,23,36. All three used the European Organization for Research and Treatment of Cancer (EORTC) QLQ-C30 SC questionnaire to evaluate quality of life in patients who received PSRT or PLCRT. The M AN U EORTC QLQ-C30 is a general cancer HRQL-questionnaire composed of a global health status, 3 functional scales, 3 symptom scales and 6-single-item scales shown in Table 2, except Wiltink et al 34 37 . As indicated that patients received PSRT had lower level of nausea/vomiting, other items showed no significant difference in TE D HRQL between patients who received PSRT or PLCRT based on the scores of QLQ-C30 from the original data of these three trials. Outcomes of optimized PSRT regimens in comparison with PLCRT EP There are two primary regimens for optimizing PSRT: use with delayed surgery and AC C adding consolidation chemotherapy. In this part, we qualitatively summarized comparative studies comparing optimized PSRT regimens with conventional PLCRT. Latkauskas et al 27 compared PSRT with delayed surgery with PLCRT. The results showed similar perioperative and long-term outcomes except for an improved DFS in patients who received PLCRT. Meanwhile, improved survival outcomes were observed in treatment of PSRT followed by consolidation chemotherapy from the results of Bujko et al 22 and Youssef et al 12 25 . Notably, all the three trials showed ACCEPTED MANUSCRIPT similar pCR rate between optimized PSRT regimens and PLCRT. The comparison data between optimized regimens of PSRT and PLCRT in terms of perioperative and long-term outcomes are shown in Table 3. RI PT Discussion Over the past decades, PLCRT and PSRT have been developed in parallel and regarded as the two main standards of care for patients with high-risk rectal cancer. In SC general, PSRT is preferred in Northern Europe and PLCRT in Southern Europe and M AN U America. The primary advantage of PSRT is its lower toxicities since it does not require concomitant chemotherapy in the radiotherapy regimen, as seen in with PLCRT. In addition, the delivery of PSRT is less expensive because it can be administered in 5 fractions compared to the 25-28 fractions seen for PLCRT 5,7 . TE D Theoretically, the main advantage of PLCRT is the downstaging effect caused by the concomitant chemotherapy and longer interval of use before surgery, and this may increase organ preservation chances for patients with low lying rectal cancer 4,38 . If EP PSRT could achieve similar benefits in perioperative, long-term, postoperative and AC C HRQL outcomes when compared with PLCRT, delivery of PLCRT may be seen as over-treating and PSRT would be considered the preferable alternative. In general, the standard schedule for PSRT includes immediate surgery (7 days), while PLCRT is more often paired with delayed surgery (6-8 weeks). Cancer cells sterilized by radiotherapy usually require time to undergo necrosis 39, and, accordingly, our results showed that PLCRT with delayed surgery had a significant advantage on pCR rate compared with PSRT with immediate surgery. However, this advantage did 13 ACCEPTED MANUSCRIPT not translate into a higher sphincter preservation rate for PLCRT. This builds upon other studies showing that higher downstaging was not correlated to a higher sphincter preservation rate 6,40, because sphincter preservation rate can be affected by RI PT multiple factors such as tumor stage and location, patient desire, and the surgeon’s skill and conceptual approach of such a “conservative” treatment 41 . Hence, PSRT with delayed surgery might improve the insufficient results seen on tumor SC downstaging in standard PSRT. Notably, the Stockholm III trial compared three arms: M AN U PSRT with immediate surgery, PSRT with delayed surgery and preoperative long-course radiotherapy (PLRT) with delayed surgery 42-45 . The Stockholm III trial published its latest results with long-term outcomes indicating PSRT with delay surgery had similar long-term outcomes in comparison with other two methods 42 . However, PSRT with delayed surgery TE D irrespective of OS, DFS, LR or DM rates showed higher acute toxicities when compared with other two methods (PSRT: < 1%; PSRT-delay: 7%; PLRT: 5%). In addition, PSRT showed the highest effects on tumor EP downstaging of ypStage 0-I (PSRT: 26.8%; PSRT-delay: 38.9%; PLRT: 28.9%). AC C Moreover, PSRT with delayed surgery showed significant lower incidence of postoperative complications when compared to PSRT with immediate surgery (41% versus 53%). Since only two RCTs to date compare PSRT and PLCRT directly, we conducted a network meta-analysis to further explore whether both treatments could provide similar long-term outcomes. In comparison with surgery alone, adding radio(chemo)therapy did not improve OS and DM. The significant benefit of 14 ACCEPTED MANUSCRIPT neoadjuvant therapy is still the local control it grants, irrespective of treatment pattern changes. However, PSRT and PLCRT showed similar long-term times based on our results. Hence, we further studied preoperative regimens which could improve RI PT survival time for rectal cancer patients. Notably, Myerson et al indicated that PSRT followed by 4 cycles of FOLFOX chemotherapy achieved a favorable response and morbidity in 76 patients 46 . A recent RCT conducted by Bujko et al indicated that SC PSRT with consolidation chemotherapy (3 cycles FOLFOX) showed an improved OS M AN U (73% versus 65%, P < 0.05) and similar local recurrence incidences compared with PLCRT (22% versus 21%, P = 0.82) after a follow-up of 3-year 22. In addition, using match pair analysis, one proceeding from the American Society for Radiation Oncology 57th Annual Meeting demonstrated improved DFS and metastasis-free 25 . TE D survival in PSRT followed by consolidation chemotherapy compared to PLCRT Furthermore, an ongoing RCT-RAPIDO trial also exploring the PSRT followed by consolidation chemotherapy in comparison with PLCRT will further characterize the EP efficiency of this optimized regimen of PSRT 47. We believe that PSRT followed by AC C consolidation chemotherapy may be a worthy therapy needed for further exploration. Surgery had the most negatively affects quality of life most in the early months, and its effect was similar in both PSRT and PLCRT based on our summarized results. This is in contrast with the idea that the higher dose per fraction and short overall treatment time of PSRT would increase the risk for the long-term functional effects compared with the lower doses per fraction within PLCRT. With this data, doctors can inform patients contemplating such treatments that there is no overall difference in 15 ACCEPTED MANUSCRIPT HRQL between PSRT and PLCRT strategies. In addition, this HRQL information is critical so that patients can learn about post-treatment function and adequate support can provide in time. RI PT Based on the latest guideline of NCCN 3, PSRT was a suboptimal option for cT3 and not recommended for cT4 rectal cancer patients. In addition, European Society for Medical Oncology (ESMO) guidelines recommended that rectal cancer patients SC should be divided into early group (the “good” cancers: cT1-2; cT3 if middle or high, M AN U N0 (or cN1 if high), no mesorectal fasica (MRF), no extramural venous invasion (EMVI)), intermediate group (the “bad” cancers: cT2 very low, cT3 MRF- (unless cT3 and mid- or high rectum, N1-2, EMVI+, limited cT4aN0)) and advanced group (the “ugly” cancers: cT3 MRF+, cT4, lateral node+) according to the recurrence risk. TE D In general, the ugly cancers may require PLCRT, the bad cancers require PSRT and the good cancers may follow radical surgery alone. Our results indicate that PSRT could be a preoperative treatment option for mixed patients with cT3 or cT4 tumors. EP However, we still cannot draw a definite conclusion on whether PSRT and PLCRT is AC C the preferred regimen for cT4 tumors from the current evidence, since we could not obtain individualized data to further stratify patients into cT3 or cT4 subgroups in the present study. Therefore, we recommend that future trials provide detailed results for cT3 and cT4 patients separately. In addition, multiple survival analyses and longer follow-up should be undertaken for cT4 patients, since the primary advantage of radiation is local control, but chemotherapy could also potentially control distant metastases. Moreover, the interval between preoperative treatments and surgery was 16 ACCEPTED MANUSCRIPT an important factor for tumor response 48 . This deficiency of PSRT with immediate surgery restricted non-resectable cT4 tumor downstaging and shrinkage. Hence, optimized PSRT regimens (with delayed surgery or adding consolidation RI PT chemotherapy) which could improve these insufficiencies necessitate further exploration. A recent published meta-analysis has provided a general summary of advantages SC and adverse effects of neoadjuvant therapies based on 41,121 rectal cancer patients M AN U over the past decades 49. The main advantage of our present study is that we provide a more comprehensive examination of PSRT including perioperative, long-term, and HRQL outcomes compared with PLCRT. In addition, we conducted network analyses to provide updated evidence between PSRT, PLCRT and surgery alone in terms of TE D long-term outcomes. Furthermore, optimized regimens for PSRT were also summarized to explore the future directions of the neoadjuvant era. Our comprehensive analyses and summary further verified the efficiency of PSRT. There EP were some limitations in our present study which should be pointed out. In our overall AC C analyses, the differences in study design, clinical stage, age, gender rate and tumor location could not be balanced for all potential methods between studies. We observed that the tumor location of patients in PSRT group is higher than that in PLCRT group in studies of Yeh et al 20 and Vironen et al 35 . As high rectal cancers have a better outcome, the overall pairwise meta-analysis might be affected by this inherent bias. In addition, from some of the trials included in our study, the tumour stage of the trial patients ranged from T1 to T3, and some resectable T4, which may cause bias to the 17 ACCEPTED MANUSCRIPT results of our overall analysis. Moreover, the main limitation was the heterogeneities (study type, published year, enrolled patients) of our included studies. Although we did not pool all the included studies for our overall analyses and data from included RI PT studies were used into different section analyses or review, we may rely on more large-scale prospective trials to further update our findings, especially focus on one specific topic (such as effect of PSRT with delayed surgery). Finally, the eligibility SC criteria in terms of TN stage, tumour height, staging methodology (from digital M AN U examination to magnetic resonance imaging) were different in these randomized trials. As these patients were treated over a long period of time, radiotherapy techniques (from 2-field chimney technique to intensity modulated radiation therapy), surgical techniques (blunt dissection to total mesorectal exicision) and postoperative care were TE D different. There were variations in chemotherapy (from no chemotherapy to adjuvant chemotherapy of different types) as well. They all have impact on survival, toxicity, and quality of life. Hence, we may rely on individual data of prospective studies to AC C Conclusions EP further explore the impact of these factors. PSRT showed comparable perioperative, long-term, and HRQL outcomes in comparison with PLCRT. Although PLCRT has a better pCR effect, this benefit does not translate into a higher sphincter preservation rate. When pCR is not the primary aim for surgeons and patients, PSRT seems to be the better option for preoperative treatment for rectal cancer patients based on its main advantages of simplicity and lower cost. Furthermore, optimized PSRT regimens could improve the current 18 ACCEPTED MANUSCRIPT deficiencies on tumor response with PSRT with immediate surgery. Hence, optimized PSRT-based regimens provide possibilities for the future evolution of the neoadjuvant era. RI PT Clinical Practice Points 1. Based on the NCCN Clinical Practical Guidelines in Oncology, PLCRT accepted as standard for locally advanced rectal cancer patients. SC 2. A published investigation indicated that only 55% of rectal cancer patients Data Base. M AN U received “standard” PLCRT in the United States according to the National Cancer 3. Whether PSRT and its optimized regimens can be alternative choices compared to PLCRT for rectal cancer patients based on perioperative, long-term and HRQL TE D outcomes remain controversial. 4. PLCRT showed a better pCR rate, but this benefit did not translate into a higher sphincter preservation rate and other perioperative outcomes differences when EP compared to PSRT. AC C 5. In terms of survival outcomes, adding either PLCRT or PSRT both showed obvious advantage on local control when compared with surgery alone, and PSRT and PLCRT had similar long-term outcomes irrespective of pairwise or network meta-analyses. 6. PSRT and PLCRT had no overall differences based on HRQL scores. 7. The insufficiency on pCR of PSRT might be improved by delayed surgery or adding consolidation chemotherapy based on systematic review of current 19 ACCEPTED MANUSCRIPT evidence. 8. Our results indicated that PSRT could be the treatment of choice in comparison with PLCRT when pCR is not the primary aim. PSRT with delayed surgery or RI PT adding consolidation may provide further possibilities for advancing the evolution of neoadjuvant era. SC Disclosure AC C EP TE D M AN U The authors declare that they have no competing interests. 20 ACCEPTED MANUSCRIPT References: 1. Murugappan S, Harris WP, Willett CG, et al. Multidisciplinary management of locally advanced rectal cancer: neoadjuvant approaches. J Natl Compr Canc Netw. May 1 2013;11(5):548-57. 2. Benson AB, 3rd, Bekaii-Saab T, Chan E, et al. Rectal cancer. J Natl Compr Canc Netw. Dec 1 2012;10(12):1528-64. Benson AB, 3rd, Venook AP, Al-Hawary MM, et al. Rectal Cancer, Version 2.2018, NCCN RI PT 3. Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. Jul 2018;16(7):874-901. 4. Rodel C, Trojan J, Bechstein WO, et al. Neoadjuvant short- or long-term radio(chemo)therapy for rectal cancer: how and who should be treated? Dig Dis. 2012;30 Suppl 2:102-8. 5. Bujko K, Bujko M. Point: short-course radiation therapy is preferable in the neoadjuvant treatment of rectal cancer. Semin Radiat Oncol. Jul 2011;21(3):220-7. Zhou ZR, Liu SX, Zhang TS, et al. Short-course preoperative radiotherapy with immediate SC 6. surgery versus long-course chemoradiation with delayed surgery in the treatment of rectal cancer: a systematic review and meta-analysis. Surg Oncol. Dec 2014;23(4):211-21. Palta M, Willett CG, Czito BG. Short-course versus long-course chemoradiation in rectal 8. Bosset JF, Calais G, Mineur L, et al. Fluorouracil-based adjuvant chemotherapy after M AN U 7. cancer--time to change strategies? Curr Treat Options Oncol. Sep 2014;15(3):421-8. preoperative chemoradiotherapy in rectal cancer: long-term results of the EORTC 22921 randomised study. Lancet Oncol. Feb 2014;15(2):184-90. 9. Ngan SY, Burmeister B, Fisher RJ, et al. Randomized trial of short-course radiotherapy versus long-course chemoradiation comparing rates of local recurrence in patients with T3 rectal TE D cancer: Trans-Tasman Radiation Oncology Group trial 01.04. J Clin Oncol. Nov 1 2012;30(31):3827-33. 10. Barton MK. Delivery of neoadjuvant chemoradiation for patients with stage II and III rectal cancer is suboptimal. CA Cancer J Clin. Jul 1 2016. 11. Sineshaw HM, Jemal A, Thomas CR, Jr., et al. Changes in treatment patterns for patients with EP locally advanced rectal cancer in the United States over the past decade: An analysis from the National Cancer Data Base. Cancer. Jul 1 2016;122(13):1996-2003. 12. Ohri N, Rapkin BD, Guha D, et al. Predictors of radiation therapy noncompliance in an urban AC C academic cancer center. Int J Radiat Oncol Biol Phys. Jan 1 2015;91(1):232-8. 13. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. Sep 2010;25(9):603-5. 14. Fryback DG, Stout NK, Rosenberg MA. An elementary introduction to Bayesian computing using WinBUGS. Int J Technol Assess Health Care. Winter 2001;17(1):98-113. 15. Woods BS, Hawkins N, Scott DA. Network meta-analysis on the log-hazard scale, combining count and hazard ratio statistics accounting for multi-arm trials: a tutorial. BMC Med Res Methodol. 2010;10:54. 16. Schmidt FL, Oh IS, Hayes TL. Fixed- versus random-effects models in meta-analysis: model properties and an empirical comparison of differences in results. Br J Math Stat Psychol. Feb 2009;62(Pt 1):97-128. 17. Mavridis D, Welton NJ, Sutton A, et al. A selection model for accounting for publication bias in a full network meta-analysis. Stat Med. Dec 30 2014;33(30):5399-412. 21 ACCEPTED MANUSCRIPT 18. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for publication bias. Biometrics. Dec 1994;50(4):1088-101. 19. Ansari N, Solomon MJ, Fisher RJ, et al. Acute Adverse Events and Postoperative Complications in a Randomized Trial of Preoperative Short-course Radiotherapy Versus Long-course Chemoradiotherapy for T3 Adenocarcinoma of the Rectum: Trans-Tasman Radiation Oncology Group Trial (TROG 01.04). Ann Surg. Sep 14 2016. 20. Yeh CH, Chen MF, Lai CH, et al. Comparison of treatment results between surgery alone, RI PT preoperative short-course radiotherapy, or long-course concurrent chemoradiotherapy in locally advanced rectal cancer. Int J Clin Oncol. Oct 2012;17(5):482-90. 21. McLachlan SA, Fisher RJ, Zalcberg J, et al. The impact on health-related quality of life in the first 12 months: A randomised comparison of preoperative short-course radiation versus long-course chemoradiation for T3 rectal cancer (Trans-Tasman Radiation Oncology Group Trial 01.04). Eur J Cancer. Mar 2016;55:15-26. Bujko K, Wyrwicz L, Rutkowski A, et al. Long-course oxaliplatin-based preoperative SC 22. chemoradiation versus 5 x 5 Gy and consolidation chemotherapy for cT4 or fixed cT3 rectal cancer: results of a randomized phase III study. Ann Oncol. May 2016;27(5):834-42. Wiltink LM, Nout RA, van der Voort van Zyp JR, et al. Long-Term Health-Related Quality of Life M AN U 23. in Patients With Rectal Cancer After Preoperative Short-Course and Long-Course (Chemo) Radiotherapy. Clin Colorectal Cancer. Feb 13 2016. 24. Bujko K, Nowacki MP, Nasierowska-Guttmejer A, et al. Long-term results of a randomized trial comparing preoperative short-course radiotherapy with preoperative conventionally fractionated chemoradiation for rectal cancer. Br J Surg. Oct 2006;93(10):1215-23. 25. Youssef FF, Markovina S, Khwaja SS, et al. Matched pair analysis of sequential short course TE D radiation therapy and FOLFOX chemotherapy as preoperative therapy for rectal cancer: improved DFS compared to institutional controls favors near-total neoadjuvant therapy. Int J Radiat Oncol Biol Phys. 2015;93(3):S125. 26. Bujko K, Nowacki MP, Kepka L, et al. Postoperative complications in patients irradiated pre-operatively for rectal cancer: report of a randomised trial comparing short-term 27. EP radiotherapy vs chemoradiation. Colorectal Dis. Jul 2005;7(4):410-6. Latkauskas T, Pauzas H, Kairevice L, et al. Preoperative conventional chemoradiotherapy versus short-course radiotherapy with delayed surgery for rectal cancer: results of a AC C randomized controlled trial. BMC Cancer. Dec 01 2016;16(1):927. 28. van Gijn W, Marijnen CA, Nagtegaal ID, et al. Preoperative radiotherapy combined with total mesorectal excision for resectable rectal cancer: 12-year follow-up of the multicentre, randomised controlled TME trial. Lancet Oncol. Jun 2011;12(6):575-82. 29. Krajcovicova I, Boljesikova E, Sandorova M, et al. Preoperative radiotherapy of locally advanced rectal cancer: clinical outcome of short-course and long-course treatment with or without concomitant chemotherapy. Klin Onkol. 2012;25(5):364-9. 30. Bujko K, Nowacki MP, Nasierowska-Guttmejer A, et al. Sphincter preservation following preoperative radiotherapy for rectal cancer: report of a randomised trial comparing short-term radiotherapy vs. conventionally fractionated radiochemotherapy. Radiother Oncol. Jul 2004;72(1):15-24. 31. Cedermark B, Johansson H, Rutqvist LE, et al. The Stockholm I trial of preoperative short term radiotherapy in operable rectal carcinoma. A prospective randomized trial. Stockholm 22 ACCEPTED MANUSCRIPT Colorectal Cancer Study Group. Cancer. May 1 1995;75(9):2269-75. 32. Martling A, Holm T, Johansson H, et al. The Stockholm II trial on preoperative radiotherapy in rectal carcinoma: long-term follow-up of a population-based study. Cancer. Aug 15 2001;92(4):896-902. 33. Fan WH, Wang FL, Lu ZH, et al. Surgery with versus without preoperative concurrent chemoradiotherapy for mid/low rectal cancer: an interim analysis of a prospective, randomized trial. Chin J Cancer. Sep 2015;34(9):394-403. Folkesson J, Birgisson H, Pahlman L, et al. Swedish Rectal Cancer Trial: long lasting benefits RI PT 34. from radiotherapy on survival and local recurrence rate. J Clin Oncol. Aug 20 2005;23(24):5644-50. 35. Vironen J, Juhola M, Kairaluoma M, et al. Tumour regression grading in the evaluation of tumour response after different preoperative radiotherapy treatments for rectal carcinoma. Int J Colorectal Dis. Sep 2005;20(5):440-5. Pietrzak L, Bujko K, Nowacki MP, et al. Quality of life, anorectal and sexual functions after SC 36. preoperative radiotherapy for rectal cancer: report of a randomised trial. Radiother Oncol. Sep 2007;84(3):217-25. Aaronson NK, Ahmedzai S, Bergman B, et al. The European Organization for Research and M AN U 37. Treatment of Cancer QLQ-C30: a quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. Mar 3 1993;85(5):365-76. 38. Ngan SY. Preoperative Treatment of Locally Advanced Rectal Cancer: Assets and Drawbacks of Short Course and Long Course in Clinical Practice. Semin Radiat Oncol. Jul 2016;26(3):186-92. 39. Francois Y, Nemoz CJ, Baulieux J, et al. Influence of the interval between preoperative radiation therapy and surgery on downstaging and on the rate of sphincter-sparing surgery 40. TE D for rectal cancer: the Lyon R90-01 randomized trial. J Clin Oncol. Aug 1999;17(8):2396. De Caluwe L, Van Nieuwenhove Y, Ceelen WP. Preoperative chemoradiation versus radiation alone for stage II and III resectable rectal cancer. Cochrane Database Syst Rev. 2013(2):CD006041. 41. Gerard JP, Rostom Y, Gal J, et al. Can we increase the chance of sphincter saving surgery in EP rectal cancer with neoadjuvant treatments: lessons from a systematic review of recent randomized trials. Crit Rev Oncol Hematol. Jan 2012;81(1):21-8. 42. Erlandsson J, Holm T, Pettersson D, et al. Optimal fractionation of preoperative radiotherapy AC C and timing to surgery for rectal cancer (Stockholm III): a multicentre, randomised, non-blinded, phase 3, non-inferiority trial. Lancet Oncol. Mar 2017;18(3):336-46. 43. Pettersson D, Cedermark B, Holm T, et al. Interim analysis of the Stockholm III trial of preoperative radiotherapy regimens for rectal cancer. Br J Surg. Apr 2010;97(4):580-7. 44. Pettersson D, Glimelius B, Iversen H, et al. Impaired postoperative leucocyte counts after preoperative radiotherapy for rectal cancer in the Stockholm III Trial. Br J Surg. Jun 2013;100(7):969-75. 45. Pettersson D, Lorinc E, Holm T, et al. Tumour regression in the randomized Stockholm III Trial of radiotherapy regimens for rectal cancer. Br J Surg. Jul 2015;102(8):972-8; discussion 78. 46. Myerson RJ, Tan B, Hunt S, et al. Five fractions of radiation therapy followed by 4 cycles of FOLFOX chemotherapy as preoperative treatment for rectal cancer. Int J Radiat Oncol Biol Phys. Mar 15 2014;88(4):829-36. 47. Nilsson PJ, van Etten B, Hospers GA, et al. Short-course radiotherapy followed by 23 ACCEPTED MANUSCRIPT neo-adjuvant chemotherapy in locally advanced rectal cancer--the RAPIDO trial. BMC Cancer. 2013;13:279. 48. Bujko K, Partycki M, Pietrzak L. Neoadjuvant radiotherapy (5 x 5 Gy): immediate versus delayed surgery. Recent Results Cancer Res. 2014;203:171-87. 49. Ma B, Gao P, Wang H, et al. What has preoperative radio(chemo)therapy brought to localized rectal cancer patients in terms of perioperative and long-term outcomes over the past decades? A systematic review and meta-analysis based on 41,121 patients. Int J Cancer. May RI PT 30 2017. Figure Legends SC Figure 1: Flow chart of study selection M AN U Figure 2: Forest plot based on RCTs between PSRT and PLCRT. A: Overall survival; B: Local recurrence; C: Postoperative morbidity; D: Postoperative mortality; E: Pathological complete response; F: Sphincter preservation rate; G: Grade 3-4 acute toxicities; H: Severe late toxicities. TE D Figure 3: Network analyses between PSRT, PLCRT and surgery alone in terms of long-term outcomes. Each link represents at least 1 study and the widths of each link are proportional to the number of studies comparing the particular arms. The size of EP each node is proportional to the total sample size. Results were presented as hazard AC C ratio and its 95% confidence interval (row-defining treatment compared with column-defining treatment). Supplementary Figure 1： Risk of bias graph of individual studies Supplementary Figure 2: Forest plot based on comparative studies between PSRT and PLCRT. A: Postoperative complications; B: Acute toxicities; C: Pathological complete response; D: Sphincter preservation rate. Supplementary Figure 3: Forest plot based on comparative studies between PSRT and 24 ACCEPTED MANUSCRIPT PLCRT. A: Overall survival; B: Disease-free survival; C: Local recurrence; D: Distant metastasis. Supplementary Figure 4: Publication bias among studies of PSRT versus surgery AC C EP TE D M AN U SC RI PT alone, PLCRT versus surgery alone and PSRT versus PLCRT. 25 ACCEPTED MANUSCRIPT Table 1 Baseline characteristics and quality assessment of included studies Armc Designb Krajcovicova* Taiwan Slovak RCT RCRS RCRS Republic Vironen Stockholm I trial Stockholm II trial Finland Sweden Sweden PCRS RCT RCT SRCT trial Sweden RCT TME trial Netherlands RCT & Sweden Fan TROG 01.04 trial China Australia RCT RCT (QoF outcomes) Wiltink Netherlands RCRS Poland RCT (QoF outcomes) Latkauskas Lithuania RCT Tumor Interval (M/F, %) Stage location T3-4N0 5.7 cm 50.4Gy, 28 fractions (5-FU + LV) 4-6 weeks 5.8 cm 25Gy, 5 fractions 7 days 7.0 cm 50.4Gy, 28 fractions (FU) 5-6 weeks 6.2 cm 25Gy, 5 fractions 3-7 days PLCRT 157 59 66/34 PSRT 155 60 65/35 PLCRT 161 64 75/25 PSRT 162 63 72/28 PLCRT 37 60 65/35 T3-4 4.7 cm 50.4Gy, 28 fractions (5-FU + LV) 4-6 weeks PSRT 28 67 61/39 or N1-2 5.9 cm 25Gy, 5 fractions 7 days PLCRT 55 64 73/27 II/III NA 45-46Gy, 23-25 fractions (5-FU + LV or Ca) 6 weeks PSRT 96 62 66/34 25Gy, 5 fractions 5 days PLCRT 44 65 80/20 50Gy, 25 fractions (5-FU) 4-5 weeks PSRT 42 68 64/36 PSRT 424 69 60/40 S 425 67 57/43 PSRT 272 66 61/39 S 285 66 61/39 PSRT 573 69 59/41 S 574 69 61/39 PSRT 897 65 64/36 S 908 66 64/36 PLCRT 90 56 62/38 S 94 58 PLCRT 154 64 PSRT 143 PLCRT 85 (QoF outcomes) Polish trial Clinical PSRT 306 PLCRT PSRT T3Nx T3-4Nx 8.5 cm 25Gy, 5 fractions 7 days Promon- 25Gy, 5 fractions 1-7 days tory - - Promon- 25Gy, 5 fractions 7 days tory - - Resectable Promon- 25Gy, 5 fractions 7 days tory - - Resectable ≤15cm 25Gy, 5 fractions ≤ 10 days ≤15cm - - 10cm 46-50Gy, 23-25 fractions (XELOX) 4-6 weeks - - 7.0 cm 50.4Gy, 28 fractions (FU) 4-6 weeks 6.2 cm 25Gy, 5 fractions 3-7 days NA 50-50.4Gy, 25-28 fractions (5-FU + LV or Ca or 5-8 weeks Resectable Resectable T3-4 54/36 or N1-2 73/27 T3Nx 63 71/29 69 65/35 68 65/35 110 60 65/35 110 62 61/39 PLCRT 72 63 69/31 PSRT 68 66 63/37 5 cm RI PT Yeh Australia RCT Number Regimens of preoperative treatmentd Gender SC TROG 01.04 trial Poland Age M AN U Polish trial Patient TE D Study EP Country AC C Studiesa LARC Q-Af Out- Follow-up comese (months) P, N 48 H P, N 70.8 H P 36 M P 48 M P NA M N 60 M N 106 M N 156 H N 144 H N 38 M Q ≥ 12 H Q 58 M Q ≥ 12 H O 40 M XELOX or Ca + bevacizumab) 25Gy, 5 fractions ≤ 10 days 5.7 cm 50.4Gy, 28 fractions (5-FU + LV) 4-6 weeks 5.8 cm 25Gy, 5 fractions 7 days T3-4 ≤15cm 50Gy, 25 fractions (5-FU + LV) 6 weeks or N1-2 ≤15cm 25Gy, 5 fractions 6 weeks T3-4 ACCEPTED MANUSCRIPT Bujko et al Youssef# Poland US RCT RCRS PLCRT 254 60 67/33 PSRT 261 60 70/30 PLCRT 80 NA NA PSRT 69 NA NA T3-4 T3-4Nx ≤15cm 50.4Gy, 28 fractions (FOLFOX or 5-FU + LV) 12.4 weeks ≤15cm 25Gy, 5 fractions followed by FOLFOX 12.4 weeks NA Long-course chemoradiotherapy (without details) NA 25Gy, 5 fractions followed by FOLFOX NA O 35 H O 49 NA AC C EP TE D M AN U SC RI PT a: Data of 19 studies based on 14 different trials were included in our systematic review and meta-analysis; b: Study design was divided into randomized controlled trial (RCT), retrospective comparative study (RCRS) and prospective comparative study (PCRS); c: Arm of included studies contained preoperative long-course chemoradiotherapy (PLCRT); preoperative short-course radiotherapy (PSRT); surgery alone (S); d: 5-FU: 5-fluorouracil; LV: leucovorin; Ca: capecitabine; e: We used the data of included studies to conduct pairwise meta-analysis (P), network meta-analysis (N), quality of life evaluation (Q) and optimized PSRT regimens summary (O); f: Quality assessment (Q-A) was divided into low, moderate and high according to Newcastle–Ottawa Scale (NOS) criterion; * PLCRT group enrolled some patients only received long-course radiotherapy without concomitant chemotherapy; # Proceedings from the American Society for Radiation Oncology 57th Annual Meeting; M: male; F: female; LARC: locally advanced rectal cancer; NA: not applicable. ACCEPTED MANUSCRIPT Table 2 Health-related quality of life outcomes after PSRT or PLCRT based on EORTC QLQ-C30 questionnaire EORTC QLQ-C30 subscales Polish trial Wiltink et al PLCRT P value PSRT PLCRT P value PSRT PLCRT P value 143 154 - 111 110 - 306 85 - 78.0 63.9 NR NR 68.5 61.1 79.9 68.2 NR NR 68.9 61.8 0.26 0.26 NR NR 0.31 0.44 76 74 66 77 75 57 75 73 67 76 73 61 0.78 0.76 0.69 0.72 0.58 0.22 82.6 83.3 86.3 84.1 87.7 78.9 84.5 82.5 86.9 84.0 84.6 79.6 0.56 0.73 0.85 0.90 0.27 0.90 36.4 NR 23.0 NR NR 20.6 NR NR NR 35.5 NR 22.0 NR NR 17.3 NR NR NR 0.62 NR 0.98 NR NR 0.37 NR NR NR 34 8 28 18 36 13 23 23 33 0.67 0.03 0.73 0.64 0.62 0.88 0.34 0.19 0.20 22.5 1.3 11.1 11.6 18.5 4.6 10.8 10.6 6.8 23.8 5.9 11.2 11.8 15.4 8.5 8.6 5.8 9.5 0.59 < 0.01 0.92 0.89 0.42 0.12 0.51 0.09 0.27 TE D M AN U SC RI PT PSRT EP Patients Function Physical Role Emotional Cognitive Social Global Symptoms Fatigue Nausea/vomiting Pain Dyspnoea Insomnia Appetite Constipation Diarrhoea Financial TROG 01.04 trial 36 5 31 18 34 14 19 18 38 A higher score for functioning reflects better functioning, whereas a higher score for symptoms represents a higher level of symptoms and decreased health-related quality of life. NR: not reported. AC C EORTC: European Organization for Research and Treatment of Cancer; PSRT: preoperative short-course radiotherapy; PLCRT: preoperative long-course chemoradiotherapy; ACCEPTED MANUSCRIPT Table 3 Systematic review of optimized PSRT (delayed surgery or adding consolidation chemotherapy) versus PLCRT based on perioperative and long-term outcomes Outcomes Latkauskas et al Youssef et al * Bujko et al PSRT+delay PLCRT PSRT+CCT PLCRT PSRT+CCT PLCRT Would complications 35.3% 26.8% NA Anastomosis leakage 5.8% 6.9% 5.9% 3.9% NA Total complications 35.3% 26.8% 28.9% 24.7% NA Pathological complete response 4.4% 11.1% 16.8% 11.7% 28% Sphincter preservatipn 69.1% 72.2% 50.0% 48.8% NA Acute toxicities NA 74.9% 82.6% NA Late toxicities NA 19.1% 20.3% NA 65% 96% 89% 52% 83% 66% 21% a 93% 93%a 87%b 69%b Perioperative outcomes 78% 82.4% 73% Disease-free survival 59% 75.1% 53% Local recurrence 0% 4.2% 22% Distant metastasis 18.8% 11.3% NA RI PT Overall survival SC Survival outcomes NA 32% M AN U PSRT: preoperative short-course radiotherapy (25 Gy, 5 fractions); PLCRT: preoperative long-course chemoradiotherapy; CCT: consolidation chemotherapy; NA: not applicable;*Proceedings from the American Society for Radiation Oncology 57th Annual Meeting; a: local control rate; b: distant metastasis-free survival; AC C EP TE D Significant difference was presented by bold style. AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT ACCEPTED MANUSCRIPT ACCEPTED MANUSCRIPT Supplementary Table 1: Evidence level assessment in terms of perioperative and long-term outcomes between PSRT and PLCRT based on RCTs Quality assessment No. of Outcomes studies Study Limitations Inconsistency Indirectness Imprecision design Other No. of patients Effect PSRT Relative Absolute (95% CI) (95% CI) HR (0.95) 19 fewer per 1000 PLCRT consider Quality Importance -ations OS RCT Not serious Not serious Not serious Not serious None 218/ 320 2 2 2 Mortality pCR rate SP rate Acute RCT RCT RCT RCT RCT Not serious Not serious Not serious Not serious Not serious toxities 2 Late RCT Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious Not serious toxicities (from 120 fewer to 288/ 288/ HR (0.89) 28 fewer per 1000 320 318 0.44-1.78 (from 260 more to 215/ 224/ HR (0.92) 30 fewer per 1000 320 0.69-1.21 (from 136 more to OR (1.26) 3 more per 1000 311 313 0.33-4.81 (from 9 fewer to 46 Not serious Not serious None HIGH IMPORTANT HIGH 79 fewer) 4/ None IMPORTANT 78 more) 5/ Not serious Not serious None 0.72-1.25 318 Not serious Not serious None SC Not serious Not serious 318 M AN U 2 Morbidity RCT Not serious TE D 2 DM RCT EP 2 LR AC C 2 224/ RI PT 2 IMPORTANT HIGH 67 fewer) 123/ 112/ OR (1.20) 43 more per 1000 311 313 0.86-1.69 (from 34 more to IMPORTANT HIGH 127 fewer) IMPORTANT HIGH more) None 3/ 45/ OR (0.06) 140 fewer per 1000 310 299 0.02-0.19 (from 118 fewer to 193/ 204/ OR (0.87) 32 fewer per 1000 311 313 0.56-1.37 (from 140 fewer to CRITICAL HIGH 147 more) None IMPORTANT HIGH 68 more) Not serious None 8/ 73/ OR (0.10) 200 fewer per 1000 310 319 0.03-0.27 (from 155 fewer to CRITICAL HIGH 220 more) Not serious None 23/ 23/ OR (1.02) 1 more per 1000 293 299 0.48-2.16 (from 38 fewer to IMPORTANT HIGH 76 more) PSRT: preoperative short-course radiotherapy; PLCRT: preoperative long-course chemoradiotherapy; OS: overall survival; LR: local recurrence; DM: Distant metastasis; RCT: randomized controlled trial; pCR: pathological complete response; SP: sphincter preservation; ACCEPTED MANUSCRIPT Supplementary Table 2. Consistency of long-term results between direct pairwise meta-analysis and network meta-analysis A. PLCRT versus S OS LR DM B. PSRT versus S OS LR DM C. PSRT versus PLCRT OS LR DM 1.09 (0.50, 2.63) 1.22 (0.30, 4.92) 0.91 (0.38, 2.17) M AN U 0.94 (0.83, 1.08) 0.46 (0.39, 0.55) 0.90 (0.79, 1.02) TE D 0.95 (0.72, 1.25) 0.89 (0.44, 1.78) 0.92 (0.69, 1.21) Network meta-analysis RI PT Direct pairwise meta-analysis 1.00 (0.67, 1.47) 0.61 (0.33, 1.15) 0.96 (0.66, 1.40) SC Treatments 0.95 (0.77, 1.14) 0.47 (0.36,0.64) 0.90 (0.73, 1.13) 0.95 (0.66, 1.35) 0.78 (0.43, 1.38) 0.93 (0.67, 1.30) AC C EP PLCRT: preoperative long-course chemoradiotherapy; PSRT: preoperative short-course radiotherapy; S: surgery alone; OS: overall survival; LR: local recurrence; DM: Distant metastasis. Results were presented by hazard ratio; Results with significance were presented by bold type.