第一部分

局部晚期直肠癌全新辅助放化疗与长程同步放化疗的疗效比较：一项多中心II期随机研究的研究结果

目的：探究局部晚期直肠癌（locally advanced rectal cancer, LARC）患者接受不同模式全新辅助放化疗（total neoadjuvant therapy, TNT）的安全性及近期疗效。

方法：本研究设计为前瞻性、多中心、随机II期临床研究（ClinicalTrials.gov No.: NCT04543695）。研究纳入18-75岁初诊为II/III期直肠腺癌患者（至少包含一项高危因素：直肠系膜筋膜受累、T4、N2、侧方淋巴结阳性、壁外血管侵犯），按照1:1:1的比例随机进入到3组：术前同步放化疗(chemoradiotherapy, CRT)→根治性手术→辅助化疗（标准组）；术前同步放化疗(CRT)→新辅助化疗(neoadjuvant chemotherapy, NACT)→根治性手术（consolidation chemotherapy, CNCT组）；新辅助化疗(NACT)→术前同步放化疗(CRT)→根治性手术（induction chemotherapy, INCT组）。CRT为全盆腔照射，处方剂量剂50Gy/25次，放疗日同步口服卡培他滨1650 mg/m2/天。新辅助化疗和辅助化疗为6周期卡培他滨和奥沙利铂联合方案（XELOX）。主要研究终点为yp0-II期的比率，允许临床完全缓解（clinical complete response, cCR）患者采用观察等待策略。

结果：自2020年8月15日至2022年08月30日，全国7个中心共入组257例患者，其中255例患者完成治疗并纳入分析（标准组：巩固化疗组：诱导化疗组=84：86：85）。新辅助治疗期间，全新辅助放化疗组≥G3急性毒副反应概率更高（标准组：巩固化疗组：诱导化疗组=7.1%：23.3%：32.9%），其中骨髓抑制是最常见反应（标准组：巩固化疗组：诱导化疗组=1.2%：18.6%：23.5%）。全组未出现毒副反应相关死亡。从治疗完成情况来看，诱导化疗组的放疗完成率更低（标准组：巩固化疗组：诱导化疗组=98.8%：100% ：90.6%），巩固化疗组和诱导化疗组的6周期化疗完成率分别为57.0%和74.1%，4周期化疗完成率分别为76.7%和83.5%。巩固化疗组及诱导化疗组经过新辅助治疗后降期率（包括yp0-II期率以及维持cCR采用等待观察策略的概率）分别为75.6% 和74.1%，而标准组降期率为56.0%。巩固化疗组、诱导化疗组和标准组的完全缓解（complete response, CR）率（包括病理完全缓解率以及维持cCR采用等待观察策略的概率）分别为43.0%、38.8%和20.2%。全新辅助放化疗两组都没有增加手术相关并发症的发生率（标准组：巩固化疗组：诱导化疗组=6.0%：8.1%：10.9%），其中吻合口瘘及腹腔感染是最常见并发症。

结论：两组全新辅助放化疗组患者的增加了≥G3急性毒副反应概率，以血液学毒性为主，但同步放化疗耐受性好，大部分患者可接受4周期化疗。两组全新辅助放化疗组患者均获得较高的降期率和CR率。

第二部分

肿瘤负荷对局部晚期直肠癌疗效的影响：一项多中心III期随机临床研究的事后分析

目的：探究肿瘤体积和肿瘤长度对局部晚期直肠癌患者的近期疗效和远期疗效的影响。

方法：共纳入T3-T4和/或N+、无远处转移的289例局部晚期直肠癌患者，患者均来自STELLAR研究的中国医学科学院肿瘤医院单中心入组数据。根据疗前影像获得肿瘤体积 （gross tumor volume, GTV），根据结直肠镜获得肿瘤长度。观察终点主要为CR率、3年总生存率 (overall survival, OS)和3年无病生存率（disease-free survival, DFS）。应用单因素和多因素分析的方法分析影响研究终点的肿瘤负荷因素和临床病理因素，纳入分析的临床病理因素包括年龄、性别、ECOG评分、临床T分期、临床N分期、临床TNM分期、肿瘤下缘距肛缘距离、直肠系膜筋膜（mesorectal fascia involvement, MRF）、壁外血管侵犯(extramural vascular invasion, EMVI）、新辅助治疗方式、肿瘤体积、肿瘤长度、放疗结束至手术时间、手术切缘和新辅助直肠（Neoadjuvant Rectal， NAR）评分。应用接受者操作特性曲线（receiver operating characteristic curve, ROC）下面积和x-tile软件得到相应最佳界值。

结果：289例局部晚期直肠癌患者年龄20-70岁（中位数55岁），男性为主（70.2%）。肿瘤临床T分期达cT3者占87.9%，达cT4者占9.3%。肿瘤临床N分期达cN1者占40.1%，达cN2者占39.4%。TNT组148人（51.2%），CRT组141人（48.8%）。中位随访时间为46.2个月（16.1-63.9个月），3年总生存率为82.9%，3年无病生存率为62.9%。肿瘤中位体积（四分位距）为53.90（36.71-71.23）ml，肿瘤中位长度（四分位距）为5.00（4.00-6.00）cm。单因素分析中，肿瘤长度与CR率相关（P=0.014），肿瘤体积与CR率无关（P=0.127），而肿瘤体积、肿瘤长度均与OS和DFS相关（均P＜0.05）。多因素分析中，肿瘤长度与CR率（OR=0.581，95%CI=0.379-0.889，P=0.012）、OS（HR=1.500，95%CI=1.127-1.995，P=0.005）和DFS（HR=1.216，95%CI=1.013-1.460，P=0.036）相关，肿瘤体积与CR率（P=0.722）、OS（P=0.540）和DFS（P=0.197）无关。肿瘤长度预测CR率的最佳界值为4.5cm（曲线下面积为0.624，P=0.005）。在接受TNT治疗的患者中，肿瘤长度＜4.5cm者在新辅助治疗后CR率更高（P=0.002）。在接受CRT治疗的患者中，肿瘤长度＜4.5cm者在新辅助治疗后CR率同样更高（P=0.027）。预测OS(P=0.015)和DFS（P=0.016）的最佳界值为7cm。接受TNT治疗的患者中，肿瘤长度＜7cm者和≥7cm者OS（中位OS：36个月 vs. 31个月，P=0.549）和DFS（中位DFS：36个月 vs.31个月，P=0.151）均无统计学差异。在接受CRT治疗的患者中，肿瘤长度＜7cm者的OS（中位OS：35个月 vs. 29个月，P=0.005）和DFS（中位DFS：35个月 vs. 19个月，P=0.047）均明显高于肿瘤长度≥7cm者。  

结论：肿瘤长度较长的患者更难以在新辅助治疗后达到CR，OS和DFS更差，更倾向于加强治疗强度。

第三部分

多学科团队质量对局部晚期直肠癌疗效的影响：一项多中心Ⅲ期随机临床研究的事后分析

目的：探究多学科团队（multidisciplinary team, MDT）质量对局部晚期直肠癌患者的生存的影响。

方法：在随机III期STELLAR研究的事后分析中，共纳入464例接受根治性手术的局部晚期直肠癌患者（临床肿瘤分期cT3-T4和/或N+、无远处转移）。研究将无病生存期（DFS）和总生存期（OS）按多学科团队（MDT）质量分层，并将MDT质量纳入DFS和OS的单变量及多变量分析中。

结果：单变量分析显示，参与MDT的科室数目较少（<5 vs ≥5；P=0.049）、MDT会议频率较低（<每周一次 vs ≥每周一次；P=0.021）以及MDT年度直肠癌讨论病例数较少（≤200 vs >200；P=0.039）与较差的DFS显著相关。此外，参与MDT的科室数目较少（<5 vs ≥5；P<0.001）、MDT会议频率较低（<每周一次 vs ≥每周一次；P<0.001）以及MDT年度直肠癌讨论病例数较少（≤200 vs >200；P=0.001）也与较差的OS显著相关。综合以上3个因素评估MDT质量，将MDT质量分为两类：高质量组与普通质量组。与在MDT质量普通的医院接受治疗的患者相比，在MDT质量高的医院接受治疗的患者3年OS更长（90.5% vs 78.1%；P=0.001），而3年DFS无显著差异（70.3% vs 61.3%；P=0.109）。多变量分析进一步表明，MDT质量对DFS（HR=1.648；95% CI 1.143–2.375；P=0.007）和OS（HR=2.771；95% CI 1.575–4.877；P<0.001）均具有显著影响。

结论：具备更完善多学科团队的医院，对局部晚期直肠癌患者的生存有显著改善作用。

第四部分

局部晚期直肠癌全新辅助放化疗和长程同步放化疗的对比：一项多中心Ⅲ期随机临床研究的生活质量结果

目的：探究全新辅助放化疗和术前同步放化疗对局部晚期直肠癌患者的生活质量和肛门功能的影响。

方法：STELLAR研究中，591例中低位、T3-T4和/或N+、无远处转移的直肠腺癌患者被随机（1：1）分入短程全新辅助放化疗（TNT，n=297）组或术前长程同步放化疗（CRT, n=294）组。随诊6年后，向仍存活的患者发送EORTC QLQ-C30、EORTC QLQ-CR29和Wexner便失禁评分调查问卷。EORTC QLQ-C30和EORTC QLQ-CR29用于评估患者的生活质量，项目包括功能、症状和总体生活质量评估。Wexner便失禁评分调查问卷用于评估患者的肛门功能。对比两组间数据差异，并与完成EORTC QLQ-C30评估的一般人群数据进行对比。

结果：中位随访77.38（59.07-103.20）月，414名存活患者中有196人（47.3%）完成问卷调查，TNT组（49.5%）和CRT组（44.9%）应答率无明显差异（p=0.350）。TNT组患者情绪功能更好（94.16±10.19分vs.90.17±14.63分，p=0.031），腹泻更严重（12.46±21.73分 vs. 6.74±16.03分，p=0.036），但两组间均值差异均小于5分，不具有临床意义，其余生活质量项目均无明显统计学差异。接受保肛手术的两组Wexner便失禁评分分别为5（0-6）分和3（0-6）分，组间比较差异无统计学意义（p=0.357）。与完成EORTC QLQ-C30评估的一般人群数据相比，TNT组患者未出现功能更差或症状更严重且均值差异＞5分的项目。

结论：接受全新辅助放化疗和术前同步放化疗的患者生活质量和肛门功能无明显差异，随机后6年可恢复到治疗前水平并维持稳定。

Part I: Results of a phase II study of total neoadjuvant therapy for locally advanced rectal cancer

Purpose: To investigate the safety and short-term efficacy of different modalities of total neoadjuvant therapy (TNT) in patients with locally advanced rectal cancer.

Methods: We conducted a multicenter, randomized, phase II trial (ClinicalTrials.gov No.: NCT04543695). Patients aged 18-75 years with initial stage II / III rectal adenocarcinoma ( At least one of high-risk factors should be included: mesorectal fascia involvement, T4, N2, positive lateral lymph node, or extramural vessel invasion) were assigned into three groups, 1) Chemoradiotherapy(50 Gy of radiation combined with oral capecitabine)  followed by consolidation chemotherapy using six cycles of XELOX (CNCT group), 2) Induction chemotherapy before chemoradiotherapy (INCT group), or 3) Chemoradiotherapy alone (Control group). Then enrolled patients were required to undergo radical surgery after neoadjuvant therapy. Patients in Control group need six cycles of XELOX after surgery. The primary end point was proportion of yp0-II, and watch-and-wait strategy after complete clinical response (cCR) was allowed.

Results: Of the 257 patients enrolled, 255 patients were evaluable (Control: CNCT: INCT= 84: 86: 85). Neoadjuvant treatment related grade 3 or 4 toxicity was higher in TNT approach (7.1%, 23.3% and 32.9% in Control, CNCT and INCT, respectively), of with myelosuppression being most common. Compliance of chemoradiotherapy was lower in INCT group (98.8%, 100% and 90.6% in Control, CNCT and INCT, respectively). As to CNCT and INCT, 57.0% and 74.1% of patients completed six cycles of chemotherapy, with 76.7% and 83.5% completing four cycles of chemotherapy, respectively. Proportions of downstage (yp0-II + sustained cCR) were achieved 75.6% in CNCT group and 74.1% in INCT group, respectively, compared to 56.0% in Control group. CR rates (pCR + sustained cCR) were achieved 43.0% in CNCT group and 38.8% in INCT group, respectively, compared to 20.2% in Control group. Both TNT approaches did not increase the incidence of surgical complications (6.0%, 8.1% and 10.9% in Control, CNCT and INCT, respectively), with anastomotic fistula and abdominal infection being of most common.

Conclusion: Both TNT approaches increased the probability of ≥ G3 acute toxicities, but most patients could tolerate 4 cycles of neoadjuvant chemotherapy. Both TNT approaches could achieve a higher proportion of downstage and CR rate.

Part II: Value of tumor burden in predicting short-term and long-term efficacy in locally advanced rectal cancer patients: A post hoc analysis of a phase III randomized trial

Purpose: To explore the influence of tumor volume and tumor length on short-term and long-term efficacy in patients with locally advanced rectal cancer.

Methods: A total of 289 patients with locally advanced rectal cancer, including T3-T4 and/or N+ disease without distant metastasis, who were registered in the Cancer Hospital of the Chinese Academy of Medical Sciences as part of the STELLAR study were included in this study. The tumor volume (GTV, gross tumor volume) was measured on pretreatment computed tomography (CT) images obtained during radiotherapy planning, and the tumor length was measured during colonoscopy. The primary endpoints observed were the complete response (CR) rate, 3-year overall survival (OS) rate, and 3-year disease-free survival (DFS) rate. Clinicopathological factors, including age, sex, ECOG score, clinical T stage, clinical N stage, clinical stage, distance to anal verge, mesorectal fascia (MRF), extramural vascular invasion (EMVI), treatment category, tumor volume, tumor length, time from the end of radiotherapy to surgery, surgical margin status and neoadjuvant rectal (NAR) score, were analyzed via univariable and multivariable analyses. Optimal cutoff values were obtained using receiver operating characteristic (ROC) curves and X-tile software.

Results: A total of 289 locally advanced rectal cancer patients aged 20–70 years (median age 55 years), predominantly male (70.2%), were included in the study. The median follow-up time was 46.2 months (range 16.1–63.9 months), with a 3-year overall survival rate of 82.9% and a 3-year disease-free survival rate of 62.9%. The median tumor volume was 53.90 ml (IQR 36.71–71.23 ml), and the median tumor length was 5.00 cm (IQR 4.00–6.00 cm). According to the univariable analysis, tumor length was associated with the CR rate (P=0.014), whereas tumor volume was not (P=0.127). Both tumor volume and tumor length were associated with OS and DFS (both P<0.05). According to the multivariable analysis, tumor length was associated with the CR rate (OR=0.581, 95% CI=0.379-0.889, P=0.012), OS (HR=1.500, 95% CI=1.127-1.995, P=0.005) and DFS (HR=1.216, 95% CI=1.013-1.460, P=0.036), whereas tumor volume was not associated with the CR rate (P=0.722), OS (P=0.540) or DFS (P=0.197). The optimal threshold for predicting the CR rate on the basis of tumor length was 4.5 cm (area under the curve (AUC)=0.624, P=0.005). Patients with tumor length <4.5 cm had a higher CR rate after neoadjuvant treatment in both the TNT group (P=0.002) and CRT group (P=0.027). The optimal threshold for predicting OS (P=0.015) and DFS (P=0.016) was 7 cm. In the TNT group, there was no significant difference in OS (median OS: 36 months vs. 31 months, P=0.549) or DFS (median DFS: 36 months vs. 31 months, P=0.151) between patients with tumor length <7 cm and those with tumor length ≥7 cm. However, in the CRT group, patients with tumor length <7 cm had significantly longer OS (median OS: 35 months vs. 29 months, P=0.005) and DFS (median DFS: 35 months vs. 19 months, P=0.047) than those with tumor length ≥7 cm.

Conclusions: Patients with longer tumors were less likely to exhibit a CR after neoadjuvant treatment, had shorter OS and DFS times and might benefit from intensified treatment regimens.

Part III: Multidisciplinary team quality improves the survival outcomes of locally advanced rectal cancer patients: A post hoc analysis of a phase III randomized trial

Purpose: We sought to determine the association between multidisciplinary team (MDT) quality and survival of patients with locally advanced rectal cancer.

Methods: In a post hoc analysis of the randomized phase III STELLAR trial, 464 patients with distal or middle-third, clinical tumor category cT3-4 and/or regional lymph node-positive rectal cancer who completed surgery were evaluated. Disease-free survival (DFS) and Overall survival (OS) were stratified by Multidisciplinary team (MDT) quality, which was also included in the univariable and multivariable analyses of DFS and OS.

Results: According to the univariable analyses, a significantly worse DFS was associated with a fewer specialized medical discipline participating in MDT (<5 vs ≥5; P=0.049), a lower frequency of MDT meetings ( 200; P=0.039). In addition, a lower number of specialized medical disciplines participating in MDT (<5 vs ≥5; P<0.001), a lower frequency of MDT meetings ( 200; P=0.001) were the variables associated with OS. These 3 factors were considered when assessing MDT quality, which was classified into 2 categories: high quality or general quality. Patients treated in hospitals with high MDT quality had longer 3-year OS (90.5% vs 78.1%; P=0.001) and similar 3-year DFS (70.3% vs 61.3%; P=0.109) compared to those treated in hospitals of the general MDT quality group. Furthermore, multivariable analyses revealed a significance for DFS (HR, 1.648; 95% CI, 1.143–2.375; P=0.007) and OS (HR, 2.771; 95% CI, 1.575–4.877; P<0.001) in MDT quality.

Conclusions: The use of hospitals with optimized multidisciplinary infrastructure had a significant influence on survival of patients with locally advanced rectal cancer.

Part IV: Quality of life and functional outcomes in patients with locally advanced rectal cancer receiving total neoadjuvant therapy versus concurrent chemoradiotherapy: An analysis of a phase III randomized trial

Purpose: To explore differences in the effects of total neoadjuvant therapy (TNT) and preoperative concurrent chemoradiotherapy (CRT) on quality of life and functional outcomes in patients with locally advanced rectal cancer.

Methods: In the study, 591 patients with distal or middle-third, clinical primary tumor stage cT3-4 and/or regional lymph node–positive rectal cancer were randomly assigned (1:1) to receive short-term radiotherapy (25 Gy in five fractions) followed by 4 cycles of CAPOX (TNT group, n=297) or standard concurrent chemoradiotherapy (50 Gy in 25 fractions concurrently with oral capecitabine) (CRT group, n=294) before surgery. After a 6-year follow-up, the surviving patients were sent surveys, including the EORTC QLQ-C30, EORTC QLQ-CR29, and Wexner incontinence score questionnaires. Differences between the two groups were compared, and baseline data from the general population who completed the EORTC QLQ-C30 were also compared.

Results: The median follow-up was 77.38 (59.07-103.20) months, with 196 out of 414 surviving patients (47.3%) completing the questionnaire. Patients in the TNT group had better emotional function (94.16±10.19 vs. 90.17±14.63, P=0.031) but more severe diarrhea (12.46±21.73 vs. 6.74±16.03, P=0.036) than did those in the CRT group. However, the mean differences between the two groups were less than 5 points, which is not clinically significant, and there were no significant differences in other quality of life items. The Wexner incontinence scores were 5 (0-6) and 3 (0-6) for the TNT and CRT groups, respectively, with no significant difference between the groups (P=0.357). Compared to the general population data from the completed EORTC QLQ-C30 assessment, the patients did not exhibit differences greater than 5 points in terms of worse function or more severe symptoms.

Conclusions: There were no significant differences in quality of life or anal function between patients receiving TNT and those receiving CRT. After 6 years, patients were able to maintain stability.