BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from

BMJ Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available.

When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to.

The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript.

BMJ Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjopen.bmj.com).

If you have any questions on BMJ Open’s open peer review process please email [email protected] http://bmjopen.bmj.com/ on October 1, 2021 by guest. Protected copyright. BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from

Using patient-reported outcomes to manage postoperative symptoms in patients with lung cancer: protocol for a multicentre, randomised controlled trial ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2019-030041

Article Type: Protocol

Date Submitted by the 24-Feb-2019 Author:

Complete List of Authors: Dai, Wei; Cancer Hospital and Research Institute, Sichuan Cancer Center Zhang, Yuanqiang; First People's Hospital, Department of Cardiothoracic Surgery Feng, Wenhong ; People's Hospital, Department of Thoracic and Cardiovascular Surgery Liao, Xiaoqing ; Dazhu County People's Hospital, Department of Cardiothoracic Surgical Oncology Mu, Yunfei ; The Third People's Hospital of , Department of Thoracic Surgery Zhang, Rui ; The Seventh People's Hospital of Chengdu, Department of Thoracic Surgery Wei, Xing ; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery http://bmjopen.bmj.com/ Wu, Chuanmei ; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery Xie, Shaohua; Sichuan Cancer Hospital and Chengdu Medical College, Department of Thoracic Surgery Li, Qiang; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery Shi, Qiuling; University of Texas MD Anderson Cancer Center

lung cancer, patient reported outcomes, postoperative symptom on October 1, 2021 by guest. Protected copyright. Keywords: management, randomised controlled trial

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 27 BMJ Open

1 2 3 1 Title page 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 2 Title: Using patient-reported outcomes to manage postoperative symptoms in patients 7 8 3 with lung cancer: protocol for a multicentre, randomised controlled trial 9 10 11 4 Authors: Wei Dai,1# Yuanqiang Zhang,2# Wenhong Feng,3# Xiaoqing Liao,4# Yunfei 12 13 5 Mu,5# Rui Zhang,6 Xing Wei,1 Chuanmei Wu,1 Shaohua Xie,1,7 Qiang Li,1 Qiuling Shi8 14 15 6 Institutions: 16 17 1 18 7 Department of ForThoracic peerSurgery, Sichuan review Cancer Hospital only & Institute, Sichuan Cancer 19 8 Center, School of Medicine, University of Electronic Science and Technology of , 20 21 9 Chengdu, Sichuan, China. 22 23 10 2Department of Cardiothoracic Surgery, Zigong First People's Hospital, Zigong, 24 25 11 Sichuan, China. 26 12 3Department of Thoracic and Cardiovascular Surgery, Jiangyou People's Hospital, 27 28 13 Jiangyou, Sichuan, China. 29 4 30 14 Department of Cardiothoracic Surgical Oncology, Dazhu County People's Hospital, 31 15 Dazhu County, , Sichuan, China. 32 33 16 5Department of Thoracic Surgery, The Third People's Hospital of Chengdu, Chengdu, 34 35 17 Sichuan, China. 36 6 37 18 Department of Thoracic Surgery, The Seventh People's Hospital of Chengdu, Chengdu, http://bmjopen.bmj.com/ 38 19 Sichuan, China. 39 40 20 7Graduate School, Chengdu Medical College, Chengdu, Sichuan, China. 41 42 21 8Department of Symptom Research, The University of Texas MD Anderson Cancer 43 44 22 Center, Houston, Texas, USA. 45 23 on October 1, 2021 by guest. Protected copyright. 46 47 24 Correspondence to 48 49 25 Dr Qiuling Shi, Department of Symptom Research, The University of Texas MD 50 26 Anderson Cancer Center, 1515 Holcombe Blvd., Unit 1450 Houston, TX 77030, USA. 51 52 27 Telephone: 713/745-3504, e-mail: [email protected]. 53 54 28 Dr Qiang Li, Department of Thoracic Surgery, Sichuan Cancer Hospital & Institute, 55 56 29 Sichuan Cancer Center, School of Medicine, University of Electronic Science and 57 30 Technology of China, No. 55, Section 4, South Renmin Road, Chengdu 610041, 58 59 1 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 27

1 2 3 31 Sichuan, China. Telephone: +86-028-85420229, e-mail: [email protected].

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 32 6 # 7 33 These authors contributed equally to this work. 8 34 9 10 11 35 Word count: 2728 words 12 36 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 2 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 27 BMJ Open

1 2 3 37 ABSTRACT 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 38 Introduction Surgery is the primary treatment for lung cancer. The postoperative 7 8 39 symptom burden experienced by patients with lung cancer is substantial, seriously 9 10 40 delaying their recovery from surgery and impairing their quality of life significantly. 11 12 41 Patient-reported outcome (PRO)-based symptom management is increasingly regarded 13 14 42 as an optimal model for patient-centred care. Currently, clinical trial-based evidence 15 16 43 involving early-phase (immediately after surgery for up to one month) symptom 17 18 44 management of Forlung cancer peer is lacking. review We propose a randomisedonly trial to evaluate the 19 20 45 effect of a PRO-based symptom-monitoring program with over-threshold alerts and 21 22 46 responses for postoperative recovery in patients with lung cancer. 23 24 25 47 Methods and analysis The study will recruit 160 patients with lung cancer from six 26 hospitals. The patients will be randomly allocated to the intervention group or control 27 48 28 49 group in a ratio of 1:1. Patients in the intervention group will receive consultation from 29 30 50 clinicians to follow-up on symptoms when their reported target symptom (pain, 31 32 51 coughing, fatigue, disturbed sleep, and shortness of breath) scores reach the pre-set 33 34 52 threshold (score ≥ 4). Patients in the control group will not generate alerts and will 35 36 53 follow the standard procedure for symptom management. All patients will receive 37 http://bmjopen.bmj.com/ 38 54 symptom assessments via the MD Anderson Symptom Inventory – lung cancer module 39 40 55 on the day before surgery, daily after surgery, and twice a week after discharge for up 41 42 56 to 4 weeks or until the start of postoperative oncologic treatment. The primary 43 44 57 outcome—mean symptom threshold events—will be compared between the 45 on October 1, 2021 by guest. Protected copyright. 46 58 intervention and standard groups via independent sample Student’s t-test. 47 48 49 59 Ethics and dissemination The study was approved by the Ethics Committee of 50 51 60 Sichuan Cancer Hospital on November 22, 2018 (No. SCCHEC-02-2018-045). This 52 manuscript is based on Version 1.0, September 9, 2018 of the protocol. The study 53 61 54 62 results will be disseminated in publications in peer-reviewed journals and presentations 55 56 63 at academic conferences. 57 58 59 3 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 27

1 2 3 64 Trials registration number ChiCTR1900020846. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 65 7 8 9 66 Keywords: lung cancer, patient reported outcomes, postoperative symptom 10 11 67 management, randomised controlled trial. 12 13 14 68 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 4 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 27 BMJ Open

1 2 3 69 ARTICLE SUMMARY 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 70 Strengths and limitations of this study 7 8 9 71 1. This is an interventional study, comparing PRO-based postoperative symptom 10 11 72 management with standard postoperative symptom management in patients with lung 12 13 73 cancer. 14 15 74 2. It is a multicentre, randomised controlled trial, conducted in six tertiary hospitals in 16 17 75 China. 18 For peer review only 19 20 76 3. It focuses on the early postoperative period with a high frequency of data collection, 21 22 77 including daily in-hospital after surgery, and twice a week after discharge for up to 4 23 24 78 weeks or until the start of postoperative oncologic treatment. 25 26 27 79 4. We use a lung cancer-specific scale and the recall period is 24 hours, which is more 28 suitable to measure rapidly changing symptoms during the early recovery phase. 29 80 30 31 81 5. The lack of blinding for the participants and investigators delivering the intervention 32 33 82 may be a limitation. 34 35 36 83 37 http://bmjopen.bmj.com/ 38 84 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 5 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 27

1 2 3 85 INTRODUCTION 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 86 Lung cancer is the most common cancer and the leading cause of cancer death in China 7 8 87 and worldwide.1 2 With the application of low-dose computed tomography in screening, 9 10 88 more and more patients with early stage lung cancer are being diagnosed and treated 11 12 89 with surgery. However, the postoperative symptom burden of lung cancer patients is 13 14 90 very severe, and this detrimentally affects their quality of life (QOL). Patients have 15 16 91 various symptoms, such as pain, coughing, fatigue, and shortness of breath, in the early 17 18 92 stages after surgeryFor or even peer a long time review after surgery.3-5 onlyLowery et al followed 183 lung 19 20 93 cancer patients for 1-6 years and found that 79.8% of them had a variety of symptoms.3 21 22 94 Among these patients, 30.6% had one symptom, 27.9% had two symptoms, and 21.3% 23 24 95 had three or more symptoms. The most frequent symptoms were pain and shortness of 25 26 96 breath. If these symptoms are not effectively controlled, the postoperative recovery of 27 3 4 28 97 patients will be severely affected, resulting in a poor QOL. Therefore, effective 29 30 98 interventions are needed to alleviate post-surgery symptoms in patients with lung 31 3 4 32 99 cancer. 33 34 100 Symptom management is the foundation of clinical care, particularly for patients 35 36 101 with cancer. Patient-reported outcome (PRO)-based symptom management is 37 http://bmjopen.bmj.com/ 38 102 increasingly regarded as an optimal model for patient-centred care.6 7 A PRO is a 39 40 103 measurement of a patient's health status that comes directly from the patient's subjective 41 42 104 evaluation, with no interpretation by medical providers or anyone else.8 Studies have 43 44 105 shown that it may be more accurate for patients to evaluate their own health status 45 on October 1, 2021 by guest. Protected copyright. 46 106 themselves than evaluation by medical providers.9 The application of PRO-based 47 48 107 symptom monitoring and alerting followed by real-time symptom management from 49 50 108 health care professionals can improve the QOL of patients, prolong survival, increase 51 52 109 patient satisfaction, and allow evaluation of the treatment method.6 10-15 Basch et al 53 54 110 reported a randomised controlled trial (RCT) result, suggesting that PRO-based 55 56 111 proactive symptom monitoring could improve symptom management and thus bring 57 10 11 58 112 survival benefits in patients undergoing chemotherapy. When compared to a 59 6 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 27 BMJ Open

1 2 3 113 traditional reactive monitoring group, median survival was 5.2 months longer among 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 10 6 114 patients in the proactive monitoring group (31.2 vs. 26.0 months, p = 0.03). 7 8 115 Currently, PROs are mainly used in non-surgical treatment settings,10-13 and they are 9 10 116 still in the early stage of application in the surgical treatment setting.5 7 14-28 Studies on 11 12 117 postoperative symptom management of lung cancer, especially in the early 13 14 118 postoperative period are lacking. In addition, most of the published literature has a low 15 16 119 level of evidence, due to its design.5 18-28 The limitations of these studies include: (1) 17 18 120 most were observationalFor peerstudies; (2) reviewthey had small sample only sizes ranging from 30 to 19 20 121 200 subjects with few exceptions; (3) they did not focus on the early postoperative 21 22 122 period, typically including the in-hospital period immediately after surgery and 4 weeks 23 24 123 after discharge when patients frequently report multiple severe symptoms, leading to 25 26 124 later negative recovery events, i.e. higher symptom burden, delayed return to intended 27 28 125 oncologic therapy, and poorer QOL; (4) they used a variety of survey instruments and 29 30 126 some were not on a lung cancer-specific scale; (5) most of the scales used such as the 31 32 127 European Organization for Research and Treatment of Cancer Quality of Life 33 34 128 Questionnaire–Lung Cancer Module (EORTC QLQ-LC13), had a recall period of 1 35 36 129 week, which may not be able to identify the rapidly changing symptoms during early 37

5 http://bmjopen.bmj.com/ 38 130 postoperative phase; (6) only one study assessed in-hospital patients immediately after 39 131 surgery, but the MD Anderson Symptom Inventory (MDASI) scale used did not include 40 41 132 lung cancer-specific symptoms, i.e. coughing;5 (7) the symptom assessments were 42 43 133 inadequate, mostly at just two or three time points post-surgery; (8) most of the surgical 44 45 134 approaches were thoracotomies, not representing the current mainstream minimally on October 1, 2021 by guest. Protected copyright. 46 47 135 invasive thoracoscopic surgery for lung cancer; (9) there were very few studies on the 48 49 136 Chinese population. 50 51 52 137 We have already conducted an observational study of perioperative symptom 53 54 138 management in patients with lung cancer based on PRO (registration number 55 56 139 NCT03341377). Now, we propose an RCT, aiming to evaluate the efficacy of a PRO- 57 58 140 based symptom monitoring, alerting, and response system (SMARS) to improve 59 7 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 27

1 2 3 141 postoperative recovery of lung cancer patients. This study will provide evidence for 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 142 early postoperative phase symptom management for patients with lung cancer. We will 7 29 8 143 use the MDASI lung cancer-specific scale (MDASI-LC) to frequently monitor 9 10 144 symptoms and their impact on the functioning of lung cancer patients from pre- 11 operation to 4 weeks after discharge or until the beginning of postoperative oncologic 12 145 13 146 treatment. The recall period of MDASI-LC is 24 hours, which is more suitable to 14 15 147 measure rapidly changing, early postoperative period symptoms compared with other 16 17 148 QOL scales for 1 week or longer. Our research hypothesis is that patients with lung 18 For peer review only 19 149 cancer undergoing PRO-based symptom management have a lower postoperative 20 21 150 symptom burden than patients undergoing standard symptom management. 22 23 24 151 25 26 METHODS AND ANALYSIS 27 152 28 29 153 Study design 30 31 32 154 The trial is a multicentre, open, randomised, parallel group controlled, and superiority 33 34 155 design. This protocol will be consistent with the Standard Protocol Items: 35 30 36 156 Recommendations For Interventional Trials (SPIRIT). The results of this trial will be 37 http://bmjopen.bmj.com/ 38 157 reported according to the guidelines of Consolidated Standards of Reporting Trials 39 31 40 158 (CONSORT). A flow chart of this trial is shown in figure 1. 41 42 159 43 44 45 160 Participants on October 1, 2021 by guest. Protected copyright. 46 47 161 Participants will be recruited from six tertiary hospitals in different cities in China. The 48 49 162 six hospitals are Sichuan Cancer Hospital, Zigong First People's Hospital, Jiangyou 50 51 163 People's Hospital, Dazhu County People's Hospital, The Third People's Hospital of 52 53 164 Chengdu, and The Seventh People's Hospital of Chengdu. The anticipated dates of the 54 55 165 study are from December 1, 2018 to December 31, 2020. We haven't started recruiting 56 57 166 patients yet. The inclusion criteria for the participants are: (1) aged 18 to 75 years, (2) 58 59 8 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 27 BMJ Open

1 2 3 167 clinically diagnosed as primary lung cancer, (3) clinical stage I-IIIA (8th edition),32 and 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 168 (4) planning to receive surgery. The exclusion criteria are: (1) history of neoadjuvant 7 8 169 therapy, (2) having other malignant tumours, and (3) unable to understand the study 9 10 170 requirements. 11 12 171 13 14 15 172 Sample size calculation 16 17 173 The primary end point of this study will be the mean symptom threshold events within 18 For peer review only 19 174 4 weeks after discharge or before the start of postoperative oncologic treatment. To 20 21 175 meet the minimal clinically important difference (0.5 standard deviation)33 for the mean 22 23 176 symptom threshold events, the required sample size is 64 for each group, when rejecting 24 25 177 the null hypothesis (the difference between the two groups < 0.5 standard deviation). A 26 27 178 total of 128 cases with valid data are needed. Considering a 20% attrition rate, we will 28 29 179 need 80 patients for each group (64/0.8). The sample size calculation is based on the 30 31 180 independent sample Student’s t-test, using a two-tailed alpha level of 0.05 and a beta 32 33 181 error probability of 0.02 (80% power). 34 35 36 182 37 http://bmjopen.bmj.com/ 38 183 Randomisation, allocation concealment and blinding procedure 39 40 41 184 The process of randomisation will be carried out online using the central randomisation 42 43 185 module on the REDCap platform (http://125.71.214.100:888/redcap) of our hospital 44 45 186 after a participant has been recruited to the study and has signed an informed consent on October 1, 2021 by guest. Protected copyright. 46 47 187 form. The data analyst will upload the randomisation allocation table to the REDCap 48 49 188 platform and then save the randomisation allocation table independently. The 50 51 189 investigator will conduct randomisation by clicking on a randomisation button on the 52 53 190 REDCap platform. It will then be allocated to the intervention group or control group 54 with a 1:1 ratio. Each group will have 80 cases. The blinding of participants and 55 191 56 192 investigators is impossible due to the nature of the interventions. 57 58 59 9 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 27

1 2 3 193 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 194 Intervention procedure 7 8 9 195 Intervention group 10 11 12 196 The SMARS will be applied for the intervention group. Patients are required to fill out 13 the MDASI-LC (0-10 points) before surgery (baseline, typically 1 to 3 days before the 14 197 15 198 operation), daily after surgery (in-hospital, typically 1 to 7 days after the operation), 16 17 199 and twice a week after discharge for up to 4 weeks (±3 days) or until the start of 18 For peer review only 19 200 postoperative oncologic treatment (typically collecting PRO data six to eight times after 20 21 201 discharge). When there are one or more target symptoms (pain, coughing, fatigue, 22 23 202 disturbed sleep, and shortness of breath) and scores reach the pre-set intervention 24 25 203 threshold (score ≥ 4), a specialist will receive an alert and will contact the patient within 26 27 204 24 hours to implement symptom relief measures, e.g. consultation, education, 28 29 205 medication guidance, and clinic or hospital visit suggestions. 30 31 32 206 33 34 207 Control group 35 36 37 208 The control group will follow the standard symptom management procedure. Patients http://bmjopen.bmj.com/ 38 39 209 will fill out the MDASI-LC at the same schedule as those in the intervention group. 40 41 210 Investigators will inform the patients that the MDASI-LC data collected are only for 42 43 211 scientific research. And the patients' symptom management will follow the current 44 45 212 standard postoperative management model, that is, when the patients report discomfort, on October 1, 2021 by guest. Protected copyright. 46 47 213 the specialists will take clinical intervention measures according to their own 48 49 214 experience rather than based on the score of the MDASI-LC. Patients will be 50 51 215 encouraged to discuss their symptoms during clinical visits or to seek emergency help 52 53 216 if severe symptoms are reported. 54 55 217 56 57 58 218 Withdrawal criteria 59 10 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 27 BMJ Open

1 2 3 219 Participants will be withdrawn from the study and no further data will be collected if 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 220 they meet the following criteria: (1) unexpected cancellation of surgery, (2) severe 7 8 221 postoperative complications affecting symptom data collection, (3) postoperative 9 10 222 length of stay > 14 days, (4) postoperative pathology showed non-primary lung cancer, 11 (5) non-R0 resection, (6) pathological stage IV, (7) participant seriously violates the 12 223 13 224 study protocol, or (8) participant asks to withdraw from the study. 14 15 16 225 17 18 For peer review only 19 226 Outcomes and measurement 20 21 227 Primary outcomes 22 23 24 228 The primary end point of this study is the mean symptom threshold events, defined as 25 26 229 the average number of target symptom threshold events per patient, at each time point. 27 28 230 According to our pilot study, the five most common postoperative symptoms of lung 29 30 231 cancer patients are: pain, coughing, fatigue, disturbed sleep and shortness of breath. In 31 32 232 this study, these five symptoms assessed by the MDASI-LC are defined as target 33 34 233 symptoms. According to the recommendation of National Comprehensive Cancer 35 36 234 Network and published literature, when a patient's symptom score is ≥ 4, it is identified 37

34 35 http://bmjopen.bmj.com/ 38 235 as moderate severity. In this study, a score of 4 is set as the threshold value for 39 40 236 intervention, and a target symptom score of ≥ 4 is reported as a threshold event. 41 42 237 The primary PRO tool used in this study is the MDASI-LC.29 It is a measure that 43 44 238 contains sixteen items of lung cancer-related and treatment-related symptoms, and six 45 on October 1, 2021 by guest. Protected copyright. 46 239 items of interference to normal daily life caused by symptoms. All items are rated on 47 48 240 0-10 numerical scales, with 0 representing “symptom not present” or “symptom not 49 50 241 interfered with life” and 10 representing “symptom as bad as one can imagine” or 51 52 242 “symptom completely interfered with life”. The recall period of the MDASI-LC is 24 53 54 243 hours and it can usually be conducted in 5 minutes. It has been translated and validated 55 56 244 for application in a Chinese context. 57 58 59 11 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 27

1 2 3 245 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 246 Secondary outcomes 7 8 9 247 The secondary end points of this study include trajectories of PROs (symptom severity, 10 11 248 daily functioning, and QOL) and revisit rate after discharge. Trajectories of PROs are 12 13 249 defined as the longitudinal changing pattern of the mean score of the five target 14 15 250 symptoms for symptom severity, the mean score of the six MDASI-LC interference 16 17 251 items for daily functioning, and the mean score of the single-item QOL scale 18 For peer review only 36 19 252 (UNISCALE) for QOL, from the baseline to 4 weeks or until the start of postoperative 20 21 253 oncologic treatment. UNISCALE only has one question using a 0-10 scale, with 0 22 23 254 representing “worst QOL” and 10 representing “best QOL”. The revisit rate after 24 25 255 discharge is defined as the ratio of the number of patients who see the doctor again after 26 discharge including outpatient visits, emergency visits and hospitalisation divided by 27 256 28 257 the total number of patients. 29 30 31 258 32 33 34 259 Data collection, management, and quality control 35 36 260 REDCap,37 38 a worldwide popular research data collection and management platform 37 http://bmjopen.bmj.com/ 38 261 established in our hospital, will be used for data collection and management in this 39 40 262 study. PRO data will be collected using an e-questionnaire or a paper questionnaire and 41 42 263 recorded in REDCap. Participants should fill out the questionnaires by themselves. If 43 44 264 participants have difficulties in completing the questionnaires, investigators will help 45 on October 1, 2021 by guest. Protected copyright. 46 265 them by reading each item aloud and recording the participant’s responses. All data 47 48 266 including demographics, clinicopathological characteristics, follow-up information and 49 50 267 PRO data will be entered into the REDCap database. Data will be checked regularly by 51 52 268 the quality controller. Participant privacy information will not be recorded in REDCap. 53 54 269 A study number will be allocated to each participant and will be used on all study 55 56 270 documentation, which will only be available to the investigators. Before patients’ 57 58 271 enrolment, investigators from each research centre will receive standard operating 59 12 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 27 BMJ Open

1 2 3 272 procedure training. Each centre will receive on-site monitoring visits, telephone 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 273 monitoring, and online guidance during the course of the trial. 7 8 274 9 10 11 275 Data analysis 12 13 Data analyses will be performed on an intention-to-treat basis using the SAS 9.4 (SAS 14 276 15 277 Institute, Inc, Cary, NC). Per-protocol analyses may be conducted. To be included in 16 17 278 the analysis, a participant must provide MDASI-LC data from the baseline and at least 18 For peer review only 19 279 two additional time points. If a participant meets the withdrawal criteria, no data will 20 21 280 be included in the analysis. Two-sided P values of < 0.05 are considered to be 22 23 281 statistically significant. Continuous variables will be presented as mean ± standard 24 25 282 deviation or median and interquartile range. Comparisons between groups will be 26 27 283 conducted using the Student’s t-test or the Wilcoxon rank sum test. Categorical 28 29 284 variables will be presented as frequencies or proportions and compared between groups 30 31 285 using the chi-square test. Trajectories of PROs will be compared between the 32 33 286 intervention group and control group using generalised mixed effects models. Missing 34 35 287 data will be processed by the multiple imputation method. Results obtained from data 36 37 288 without missed observations will be compared with that from imputed data for sensitive http://bmjopen.bmj.com/ 38 39 289 analysis. 40 41 290 42 43 44 291 Data monitoring and interim analysis 45 on October 1, 2021 by guest. Protected copyright. 46 47 292 A data monitoring committee (DSM) consisting of one clinician, one statistician, and 48 49 293 the secretary of the Ethics Committee of Sichuan Cancer Hospital will be set up. Study 50 51 294 monitoring will be carried out regularly by DSM members and the process will be 52 53 295 independent from investigators. Due to the low-risk of the study content and short-term 54 study duration, interim analysis will not be performed. 55 296 56 57 297 58 59 13 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 27

1 2 3 298 Data availability statement 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 299 Deidentified data generated by this clinical trial to support future research articles will 7 8 300 be available from the corresponding author on reasonable request. 9 10 11 301 12 13 Patient and public involvement statement 14 302 15 16 303 Patients and the public will not be involved in the design, recruitment to, or conduct of 17 18 304 this study. We willFor provide peer all the participants review with free only long-term medical consultation 19 20 305 after this study. The burden of the intervention in this RCT will not be performed. We 21 22 306 will inform the applicants of the results. There are no plans to disseminate the results 23 24 307 to study participants. 25 26 27 308 28 29 309 ETHICS AND DISSEMINATION 30 31 32 310 This study was approved by the Ethics Committee of Sichuan Cancer Hospital on 33 34 311 November 22, 2018 (No. SCCHEC-02-2018-045). All recruited patients will be 35 36 312 required to give written informed consent. Any subsequent amendments to the protocol 37 http://bmjopen.bmj.com/ 38 313 will be submitted for further review and approval. Sub-centres will gain approval from 39 40 314 their hospital-specific ethics committees. The results of this study will be disseminated 41 42 315 through peer-reviewed publications and academic conferences. 43 44 316 45 on October 1, 2021 by guest. Protected copyright. 46 47 317 Acknowledgements The authors thank all the patients and patient advisers who are involved in this 48 49 318 study. 50 51 319 Author contributions WD and QLS conceived and designed the study. WD and QLS obtained the 52 53 320 funding. WD is the chief investigator of this study. YQZ, WHF, XQL, YFM, and RZ are sub-centre 54 55 321 principal investigators who contributed to the trial feasibility stage. WD, YQZ, WHF, XQL, YFM, 56 57 322 RZ, XW, CMW and SHX drafted the protocol. QLS participated in the statistical plan. QL and QLS 58 59 14 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 27 BMJ Open

1 2 3 323 revised the manuscript. All authors have read and approved the manuscript. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 324 Funding This work was supported by [Bethune charitable foundation], [Sichuan Science and 7 8 325 Technology Program] grant number [18PJ436] and [National Natural Science Foundation of China] 9 10 326 grant number [81872506]. 11 12 13 327 Competing interests None declared. 14 15 328 Patient consent Obtained. 16 17 18 329 Ethics approval EthicsFor Committee peer of Sichuan review Cancer Hospital only (No. SCCHEC-02-2018-045). 19 20 21 330 Provenance and peer review Not commissioned; externally peer reviewed. 22 23 331 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 15 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 27

1 2 3 332 REFERENCES 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 333 1. Chen W, Zheng R, Baade PD, et al. Cancer statistics in China, 2015. CA Cancer J Clin 7 8 334 2016;66:115-32. 9 10 335 2. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates 11 12 336 of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 13 14 337 2018;68:394-424. 15 16 338 3. Lowery AE, Krebs P, Coups EJ, et al. Impact of symptom burden in post-surgical non-small cell 17 18 339 lung cancer survivors.For Support peer Care Cancer review 2014;22:173-80. only 19 20 340 4. Yang P, Cheville AL, Wampfler JA, et al. Quality of life and symptom burden among long-term 21 22 341 lung cancer survivors. J Thorac Oncol 2012;7:64-70. 23 24 342 5. Fagundes CP, Shi Q, Vaporciyan AA, et al. Symptom recovery after thoracic surgery: Measuring 25 26 343 patient-reported outcomes with the MD Anderson Symptom Inventory. J Thorac Cardiovasc 27 28 344 Surg 2015;150:613-9 e2. 29 30 345 6. Basch E. Patient-Reported Outcomes - Harnessing Patients' Voices to Improve Clinical Care. N 31 32 346 Engl J Med 2017;376:105-8. 33 34 347 7. Khullar OV, Fernandez FG. Patient-Reported Outcomes in Thoracic Surgery. Thorac Surg Clin 35 2017;27:279-90. 36 348 37 349 8. U.S. Department of Health and Human Services FDA Center for Drug Evaluation and Research, http://bmjopen.bmj.com/ 38 39 350 U.S. Department of Health and Human Services FDA Center for Biologics Evaluation and 40 41 351 Research, U.S. Department of Health and Human Services FDA Center for Devices and 42 43 352 Radiological Health. Guidance for industry: patient-reported outcome measures: use in medical 44 45 353 product development to support labeling claims: draft guidance. Health Qual Life Outcomes on October 1, 2021 by guest. Protected copyright. 46 47 354 2006;4:79. 48 49 355 9. Basch E. The missing voice of patients in drug-safety reporting. N Engl J Med 2010;362:865-9. 50 51 356 10. Basch E, Deal AM, Dueck AC, et al. Overall Survival Results of a Trial Assessing Patient- 52 53 357 Reported Outcomes for Symptom Monitoring During Routine Cancer Treatment. JAMA 54 55 358 2017;318:197-8. 56 57 359 11. Basch E, Deal AM, Kris MG, et al. Symptom Monitoring With Patient-Reported Outcomes 58 59 16 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 27 BMJ Open

1 2 3 360 During Routine Cancer Treatment: A Randomized Controlled Trial. J Clin Oncol 2016;34:557- 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 361 65. 7 8 362 12. Novello S, Kaiser R, Mellemgaard A, et al. Analysis of patient-reported outcomes from the 9 LUME-Lung 1 trial: a randomised, double-blind, placebo-controlled, Phase III study of second- 10 363 11 364 line nintedanib in patients with advanced non-small cell lung cancer. Eur J Cancer 2015;51:317- 12 13 365 26. 14 15 366 13. Popat S. Patient reported outcomes from LUX-Lung 3: first-line afatinib is superior to 16 17 367 chemotherapy-would patients agree? Ann Palliat Med 2014;3:19-21. 18 For peer review only 19 368 14. Rolfson O, Donahue GS, Hallsten M, et al. Patient-reported outcomes in cemented and 20 21 369 uncemented total hip replacements. Hip Int 2016;26:451-7. 22 23 370 15. Day RW, Cleeland CS, Wang XS, et al. Patient-Reported Outcomes Accurately Measure the 24 25 371 Value of an Enhanced Recovery Program in Liver Surgery. J Am Coll Surg 2015;221:1023-30 26 27 372 e1-2. 28 29 373 16. Poghosyan H, Sheldon LK, Leveille SG, et al. Health-related quality of life after surgical 30 31 374 treatment in patients with non-small cell lung cancer: a systematic review. Lung Cancer 32 33 375 2013;81:11-26. 34 35 376 17. Cleeland CS, Wang XS, Shi Q, et al. Automated symptom alerts reduce postoperative symptom 36 37 377 severity after cancer surgery: a randomized controlled clinical trial. J Clin Oncol 2011;29:994- http://bmjopen.bmj.com/ 38 39 378 1000. 40 41 379 18. Khullar OV, Rajaei MH, Force SD, et al. Pilot Study to Integrate Patient Reported Outcomes 42 43 380 After Lung Cancer Operations Into The Society of Thoracic Surgeons Database. Ann Thorac 44 45 381 Surg 2017;104:245-53. on October 1, 2021 by guest. Protected copyright. 46 47 382 19. Shi Q, Wang XS, Vaporciyan AA, et al. Patient-Reported Symptom Interference as a Measure 48 of Postsurgery Functional Recovery in Lung Cancer. J Pain Symptom Manage 2016;52:822-31. 49 383 50 384 20. Li WW, Lee TW, Lam SS, et al. Quality of life following lung cancer resection: video-assisted 51 52 385 thoracic surgery vs thoracotomy. Chest 2002;122:584-9. 53 54 386 21. Kenny PM, King MT, Viney RC, et al. Quality of life and survival in the 2 years after surgery 55 56 387 for non small-cell lung cancer. J Clin Oncol 2008;26:233-41. 57 58 59 17 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 27

1 2 3 388 22. Balduyck B, Hendriks J, Lauwers P, et al. Quality of life after lung cancer surgery: a prospective 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 389 pilot study comparing bronchial sleeve lobectomy with pneumonectomy. J Thorac Oncol 7 8 390 2008;3:604-8. 9 23. Ferguson MK, Parma CM, Celauro AD, et al. Quality of life and mood in older patients after 10 391 11 392 major lung resection. Ann Thorac Surg 2009;87:1007-12; discussion 12-3. 12 13 393 24. Sartipy U. Prospective population-based study comparing quality of life after pneumonectomy 14 15 394 and lobectomy. Eur J Cardiothorac Surg 2009;36:1069-74. 16 17 395 25. Ostroff JS, Krebs P, Coups EJ, et al. Health-related quality of life among early-stage, non-small 18 For peer review only 19 396 cell, lung cancer survivors. Lung Cancer 2011;71:103-8. 20 21 397 26. Moller A, Sartipy U. Long-term health-related quality of life following surgery for lung cancer. 22 23 398 Eur J Cardiothorac Surg 2012;41:362-7. 24 25 399 27. Zhao J, Zhao Y, Qiu T, et al. Quality of life and survival after II stage nonsmall cell carcinoma 26 27 400 surgery: Video-assisted thoracic surgery versus thoracotomy lobectomy. Indian J Cancer 28 29 401 2015;52 Suppl 2:e130-3. 30 31 402 28. Yun YH, Kim YA, Sim JA, et al. Prognostic value of quality of life score in disease-free 32 33 403 survivors of surgically-treated lung cancer. BMC Cancer 2016;16:505. 34 35 404 29. Mendoza TR, Wang XS, Lu C, et al. Measuring the symptom burden of lung cancer: the validity 36 37 405 and utility of the lung cancer module of the M. D. Anderson Symptom Inventory. Oncologist http://bmjopen.bmj.com/ 38 39 406 2011;16:217-27. 40 41 407 30. Chan AW, Tetzlaff JM, Altman DG, et al. SPIRIT 2013 statement: defining standard protocol 42 43 408 items for clinical trials. Ann Intern Med 2013;158:200-7. 44 45 409 31. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting on October 1, 2021 by guest. Protected copyright. 46 47 410 parallel group randomized trials. Ann Intern Med 2010;152:726-32. 48 32. Goldstraw P, Chansky K, Crowley J, et al. The IASLC Lung Cancer Staging Project: Proposals 49 411 50 412 for Revision of the TNM Stage Groupings in the Forthcoming (Eighth) Edition of the TNM 51 52 413 Classification for Lung Cancer. J Thorac Oncol 2016;11:39-51. 53 54 414 33. Revicki DA, Cella D, Hays RD, et al. Responsiveness and minimal important differences for 55 56 415 patient reported outcomes. Health & Quality of Life Outcomes 2006;4:70. 57 58 59 18 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 27 BMJ Open

1 2 3 416 34. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 417 Adult Cancer Pain. Version 1.2019. Available at: 7 8 418 https://www.nccn.org/professionals/physician_ gls/pdf/pain.pdf. 9 35. Wang XS, Zhao F, Fisch MJ, et al. Prevalence and characteristics of moderate to severe fatigue: 10 419 11 420 a multicenter study in cancer patients and survivors. Cancer 2014;120:425-32. 12 13 421 36. Sloan JA, Loprinzi CL, Kuross SA, et al. Randomized comparison of four tools measuring 14 15 422 overall quality of life in patients with advanced cancer. J Clin Oncol 1998;16:3662-73. 16 17 423 37. Tomko RL, Gray KM, Oppenheimer SR, et al. Using REDCap for ambulatory assessment: 18 For peer review only 19 424 Implementation in a clinical trial for smoking cessation to augment in-person data collection. 20 21 425 Am J Drug Alcohol Abuse 2018:1-16. 22 23 426 38. Harvey LA. REDCap: web-based software for all types of data storage and collection. Spinal 24 25 427 Cord 2018;56:625. 26 27 428 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 19 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 27

1 2 3 429 Figure legends 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 430 Figure 1 Flow chart of this parallel group randomised trial. 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 20 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 27 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on October 1, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 27

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 7 8 SPIRIT 2013 Checklist: Recommended items to address in a clinical trial protocol and 9 related documents* 10 11 Section/item Item Description Page Number 12 No on which item 13 is reported 14 15 Administrative information 16 17 Title 1 Descriptive title identifying the study design, Page 1 18 Forpopulation, peer interventions, review and, if applicable, only trial 19 20 acronym 21 22 Trial 2a Trial identifier and registry name. If not yet registered, Page 4 23 registration name of intended registry 24 25 2b All items from the World Health Organization Trial Page 4 26 Registration Data Set 27 28 Protocol 3 Date and version identifier Page 3 29 version 30 31 Funding 4 Sources and types of financial, material, and other Page 14-15 32 33 support 34 35 Roles and 5a Names, affiliations, and roles of protocol contributors Page 1, 14 36 responsibilitie 37 5b Name and contact information for the trial sponsor Page 1, 14 s http://bmjopen.bmj.com/ 38 39 5c Role of study sponsor and funders, if any, in study Page 14 40 design; collection, management, analysis, and 41 42 interpretation of data; writing of the report; and the 43 decision to submit the report for publication, including 44 whether they will have ultimate authority over any of 45 these activities on October 1, 2021 by guest. Protected copyright. 46 47 5d Composition, roles, and responsibilities of the Page 12-13 48 49 coordinating centre, steering committee, endpoint 50 adjudication committee, data management team, and 51 other individuals or groups overseeing the trial, if 52 applicable (see Item 21a for data monitoring 53 54 committee) 55 56 Introduction 57 58 59 60

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 27 BMJ Open

1 2 Background 6a Description of research question and justification for Page 6-7 3 and rationale undertaking the trial, including summary of relevant

4 studies (published and unpublished) examining BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 benefits and harms for each intervention 7 8 6b Explanation for choice of comparators Page 8 9 10 Objectives 7 Specific objectives or hypotheses Page 3, 8 11 12 Trial design 8 Description of trial design including type of trial (eg, Page 8 13 parallel group, crossover, factorial, single group), 14 allocation ratio, and framework (eg, superiority, 15 equivalence, noninferiority, exploratory) 16 17 18 Methods: Participants,For interventions, peer andreview outcomes only 19 20 Study setting 9 Description of study settings (eg, community clinic, Page 8 21 academic hospital) and list of countries where data 22 will be collected. Reference to where list of study sites 23 24 can be obtained 25 26 Eligibility 10 Inclusion and exclusion criteria for participants. If Page 8-9 27 criteria applicable, eligibility criteria for study centres and 28 individuals who will perform the interventions (eg, 29 surgeons, psychotherapists) 30 31 Interventions 11a Interventions for each group with sufficient detail to Page 10 32 33 allow replication, including how and when they will be 34 administered 35 36 11b Criteria for discontinuing or modifying allocated Page 10-11 37 interventions for a given trial participant (eg, drug http://bmjopen.bmj.com/ 38 dose change in response to harms, participant 39 40 request, or improving/worsening disease) 41 42 11c Strategies to improve adherence to intervention Page 10 43 protocols, and any procedures for monitoring 44 adherence (eg, drug tablet return, laboratory tests) 45 on October 1, 2021 by guest. Protected copyright. 46 11d Relevant concomitant care and interventions that are Page 10 47 48 permitted or prohibited during the trial 49 50 Outcomes 12 Primary, secondary, and other outcomes, including Page 11 51 the specific measurement variable (eg, systolic blood 52 pressure), analysis metric (eg, change from baseline, 53 final value, time to event), method of aggregation (eg, 54 55 median, proportion), and time point for each outcome. 56 Explanation of the clinical relevance of chosen 57 efficacy and harm outcomes is strongly recommended 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 27

1 2 Participant 13 Time schedule of enrolment, interventions (including Page 10 3 timeline any run-ins and washouts), assessments, and visits

4 for participants. A schematic diagram is highly BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 recommended (see Figure) 7 8 Sample size 14 Estimated number of participants needed to achieve Page 9 9 study objectives and how it was determined, including 10 clinical and statistical assumptions supporting any 11 sample size calculations 12 13 Recruitment 15 Strategies for achieving adequate participant None 14 15 enrolment to reach target sample size 16 17 Methods: Assignment of interventions (for controlled trials) 18 For peer review only 19 Allocation: 20 21 Sequence 16a Method of generating the allocation sequence (eg, Page 9 22 generation computer-generated random numbers), and list of any 23 factors for stratification. To reduce predictability of a 24 random sequence, details of any planned restriction 25 26 (eg, blocking) should be provided in a separate 27 document that is unavailable to those who enrol 28 participants or assign interventions 29 30 Allocation 16b Mechanism of implementing the allocation sequence Page 9 31 concealme (eg, central telephone; sequentially numbered, 32 33 nt opaque, sealed envelopes), describing any steps to 34 mechanis conceal the sequence until interventions are assigned 35 m 36 37 Implement 16c Who will generate the allocation sequence, who will Page 9 http://bmjopen.bmj.com/ 38 39 ation enrol participants, and who will assign participants to 40 interventions 41 42 Blinding 17a Who will be blinded after assignment to interventions Page 9 43 (masking) (eg, trial participants, care providers, outcome 44 assessors, data analysts), and how 45 on October 1, 2021 by guest. Protected copyright. 46 17b If blinded, circumstances under which unblinding is Not 47 48 permissible, and procedure for revealing a Applicable 49 participant’s allocated intervention during the trial 50 51 Methods: Data collection, management, and analysis 52 53 54 55 56 57 58 59 60

3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 27 BMJ Open

1 2 Data 18a Plans for assessment and collection of outcome, Page 12 3 collection baseline, and other trial data, including any related

4 methods processes to promote data quality (eg, duplicate BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 measurements, training of assessors) and a 7 description of study instruments (eg, questionnaires, 8 laboratory tests) along with their reliability and validity, 9 if known. Reference to where data collection forms 10 11 can be found, if not in the protocol 12 13 18b Plans to promote participant retention and complete Page 11 14 follow-up, including list of any outcome data to be 15 collected for participants who discontinue or deviate 16 from intervention protocols 17 18 Data 19 ForPlans forpeer data entry, review coding, security, only and storage, Page 12 19 20 management including any related processes to promote data 21 quality (eg, double data entry; range checks for data 22 values). Reference to where details of data 23 management procedures can be found, if not in the 24 25 protocol 26 27 Statistical 20a Statistical methods for analysing primary and Page 13 28 methods secondary outcomes. Reference to where other 29 details of the statistical analysis plan can be found, if 30 not in the protocol 31 32 20b Methods for any additional analyses (eg, subgroup Page 13 33 34 and adjusted analyses) 35 36 20c Definition of analysis population relating to protocol Page 13 37 non-adherence (eg, as randomised analysis), and any http://bmjopen.bmj.com/ 38 statistical methods to handle missing data (eg, 39 multiple imputation) 40 41 Methods: Monitoring 42 43 44 Data 21a Composition of data monitoring committee (DMC); Page 13 45 monitoring summary of its role and reporting structure; statement on October 1, 2021 by guest. Protected copyright. 46 of whether it is independent from the sponsor and 47 competing interests; and reference to where further 48 49 details about its charter can be found, if not in the 50 protocol. Alternatively, an explanation of why a DMC 51 is not needed 52 53 21b Description of any interim analyses and stopping Page 13 54 guidelines, including who will have access to these 55 56 interim results and make the final decision to 57 terminate the trial 58 59 60

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 27

1 2 Harms 22 Plans for collecting, assessing, reporting, and Page 10 3 managing solicited and spontaneously reported

4 adverse events and other unintended effects of trial BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 interventions or trial conduct 7 8 Auditing 23 Frequency and procedures for auditing trial conduct, if Page 13 9 any, and whether the process will be independent 10 from investigators and the sponsor 11 12 13 Ethics and dissemination 14 Research 24 Plans for seeking research ethics Page 14 15 16 ethics committee/institutional review board (REC/IRB) 17 approval approval 18 For peer review only 19 Protocol 25 Plans for communicating important protocol Page 14 20 amendments modifications (eg, changes to eligibility criteria, 21 outcomes, analyses) to relevant parties (eg, 22 23 investigators, REC/IRBs, trial participants, trial 24 registries, journals, regulators) 25 26 Consent or 26a Who will obtain informed consent or assent from Page 14 27 assent potential trial participants or authorised surrogates, 28 29 and how (see Item 32) 30 31 26b Additional consent provisions for collection and use of Not 32 participant data and biological specimens in ancillary Applicable 33 studies, if applicable 34 35 Confidentiality 27 How personal information about potential and enrolled Page 12 36 participants will be collected, shared, and maintained 37 http://bmjopen.bmj.com/ 38 in order to protect confidentiality before, during, and 39 after the trial 40 41 Declaration of 28 Financial and other competing interests for principal Page 15 42 interests investigators for the overall trial and each study site 43 44 Access to 29 Statement of who will have access to the final trial Page 12-13 45 data dataset, and disclosure of contractual agreements on October 1, 2021 by guest. Protected copyright. 46 47 that limit such access for investigators 48 49 Ancillary and 30 Provisions, if any, for ancillary and post-trial care, and Page 14 50 post-trial care for compensation to those who suffer harm from trial 51 participation 52 53 Dissemination 31a Plans for investigators and sponsor to communicate Page 14 54 policy trial results to participants, healthcare professionals, 55 56 the public, and other relevant groups (eg, via 57 publication, reporting in results databases, or other 58 data sharing arrangements), including any publication 59 restrictions 60

5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 27 BMJ Open

1 2 31b Authorship eligibility guidelines and any intended use None 3 of professional writers

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 31c Plans, if any, for granting public access to the full Page 14 6 protocol, participant-level dataset, and statistical code 7 8 9 Appendices 10 11 Informed 32 Model consent form and other related documentation None 12 consent given to participants and authorised surrogates 13 materials 14 15 Biological 33 Plans for collection, laboratory evaluation, and Not 16 specimens storage of biological specimens for genetic or Applicable 17 18 Formolecular peer analysis reviewin the current trial only and for future 19 use in ancillary studies, if applicable 20 21 *It is strongly recommended that this checklist be read in conjunction with the SPIRIT 2013 22 Explanation & Elaboration for important clarification on the items. Amendments to the 23 protocol should be tracked and dated. The SPIRIT checklist is copyrighted by the SPIRIT 24 Group under the Creative Commons “Attribution-NonCommercial-NoDerivs 3.0 Unported” 25 26 license. 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from

Using patient-reported outcomes to manage postoperative symptoms in patients with lung cancer: protocol for a multicentre, randomised controlled trial ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2019-030041.R1

Article Type: Protocol

Date Submitted by the 15-May-2019 Author:

Complete List of Authors: Dai, Wei; Sichuan Cancer Hospital and Research Institute, Sichuan Cancer Center Zhang, Yuanqiang; Zigong First People's Hospital, Department of Cardiothoracic Surgery Feng, Wenhong ; Jiangyou People's Hospital, Department of Thoracic and Cardiovascular Surgery Liao, Xiaoqing ; Dazhu County People's Hospital, Department of Cardiothoracic Surgical Oncology Mu, Yunfei ; The Third People's Hospital of Chengdu, Department of Thoracic Surgery Zhang, Rui ; The Seventh People's Hospital of Chengdu, Department of Thoracic Surgery Wei, Xing ; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery http://bmjopen.bmj.com/ Wu, Chuanmei ; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery Xie, Shaohua; Sichuan Cancer Hospital and Chengdu Medical College, Department of Thoracic Surgery Li, Qiang; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery Shi, Qiuling; University of Texas MD Anderson Cancer Center

Primary Subject on October 1, 2021 by guest. Protected copyright. Patient-centred medicine Heading:

Secondary Subject Heading: Surgery, Oncology, Nursing

lung cancer, patient reported outcomes, postoperative symptom Keywords: management, randomised controlled trial

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 31 BMJ Open

1 2 3 1 Title page 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 2 Title: Using patient-reported outcomes to manage postoperative symptoms in patients 7 8 3 with lung cancer: protocol for a multicentre, randomised controlled trial 9 10 11 4 Authors: Wei Dai,1# Yuanqiang Zhang,2# Wenhong Feng,3 Xiaoqing Liao,4 Yunfei 12 13 5 Mu,5 Rui Zhang,6 Xing Wei,1 Chuanmei Wu,1 Shaohua Xie,1,7 Qiang Li,1 Qiuling Shi8 14 15 6 Institutions: 16 17 1 18 7 Department of ForThoracic peerSurgery, Sichuan review Cancer Hospital only & Institute, Sichuan Cancer 19 8 Center, School of Medicine, University of Electronic Science and Technology of China, 20 21 9 Chengdu, Sichuan, China. 22 23 10 2Department of Cardiothoracic Surgery, Zigong First People's Hospital, Zigong, 24 25 11 Sichuan, China. 26 12 3Department of Thoracic and Cardiovascular Surgery, Jiangyou People's Hospital, 27 28 13 Jiangyou, Sichuan, China. 29 4 30 14 Department of Cardiothoracic Surgical Oncology, Dazhu County People's Hospital, 31 15 Dazhu County, Dazhou, Sichuan, China. 32 33 16 5Department of Thoracic Surgery, The Third People's Hospital of Chengdu, Chengdu, 34 35 17 Sichuan, China. 36 6 37 18 Department of Thoracic Surgery, The Seventh People's Hospital of Chengdu, Chengdu, http://bmjopen.bmj.com/ 38 19 Sichuan, China. 39 40 20 7Graduate School, Chengdu Medical College, Chengdu, Sichuan, China. 41 42 21 8Department of Symptom Research, The University of Texas MD Anderson Cancer 43 44 22 Center, Houston, Texas, USA. 45 23 on October 1, 2021 by guest. Protected copyright. 46 47 24 Correspondence to 48 49 25 Dr Qiuling Shi, Department of Symptom Research, The University of Texas MD 50 26 Anderson Cancer Center, 1515 Holcombe Blvd., Unit 1450 Houston, TX 77030, USA. 51 52 27 Telephone: 713/745-3504, e-mail: [email protected]. 53 54 28 Dr Qiang Li, Department of Thoracic Surgery, Sichuan Cancer Hospital & Institute, 55 56 29 Sichuan Cancer Center, School of Medicine, University of Electronic Science and 57 30 Technology of China, No. 55, Section 4, South Renmin Road, Chengdu 610041, 58 59 1 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 31

1 2 3 31 Sichuan, China. Telephone: +86-028-85420229, e-mail: [email protected].

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 32 6 # 7 33 These authors contributed equally to this work. 8 34 9 10 11 35 Word count: 3508 words 12 36 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 2 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 31 BMJ Open

1 2 3 37 ABSTRACT 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 38 Introduction Surgery is one of the primary treatments for lung cancer. The 7 8 39 postoperative symptom burden experienced by patients with lung cancer is substantial, 9 10 40 seriously delaying their recovery from surgery and impairing their quality of life. 11 12 41 Patient-reported outcome (PRO)-based symptom management is increasingly regarded 13 14 42 as an optimal model for patient-centred care. Currently, clinical trial-based evidence 15 16 43 involving early-phase (immediately after surgery for up to one month) symptom 17 18 44 management of Forlung cancer peer is lacking. review We propose a randomisedonly trial to evaluate the 19 20 45 effect of a PRO-based symptom-monitoring program with over-threshold alerts and 21 22 46 responses for postoperative recovery in patients with lung cancer. 23 24 25 47 Methods and analysis The study will recruit 160 patients with lung cancer from six 26 hospitals. The patients will be randomly allocated to the intervention group or control 27 48 28 49 group in a ratio of 1:1. Patients in the intervention group will receive consultation from 29 30 50 clinicians to follow-up on symptoms when their reported target symptom (pain, 31 32 51 coughing, fatigue, disturbed sleep, and shortness of breath) scores reach the pre-set 33 34 52 threshold (score ≥ 4). Patients in the control group will not generate alerts and will 35 36 53 follow the standard procedures for symptom management. All patients will receive 37 http://bmjopen.bmj.com/ 38 54 symptom assessments via the MD Anderson Symptom Inventory – lung cancer module 39 40 55 on the day before surgery, daily after surgery, and twice a week after discharge until 4 41 42 56 weeks or the start of postoperative oncologic treatment. The primary outcome—mean 43 44 57 symptom threshold events—will be compared between the intervention and control 45 on October 1, 2021 by guest. Protected copyright. 46 58 group via independent sample Student’s t-test. 47 48 49 59 Ethics and dissemination The study was approved by the Ethics Committee of 50 51 60 Sichuan Cancer Hospital on November 22, 2018 (No. SCCHEC-02-2018-045). This 52 manuscript is based on Version 2.0, May 9, 2019 of the protocol. The study results will 53 61 54 62 be disseminated in publications in peer-reviewed journals and presentations at 55 56 63 academic conferences. 57 58 59 3 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 31

1 2 3 64 Trials registration number ChiCTR1900020846. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 65 7 8 9 66 Keywords: lung cancer, patient reported outcomes, postoperative symptom 10 11 67 management, randomised controlled trial. 12 13 14 68 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 4 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 31 BMJ Open

1 2 3 69 ARTICLE SUMMARY 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 70 Strengths and limitations of this study 7 8 9 71 1. This is an interventional study, comparing PRO-based postoperative symptom 10 11 72 management with standard postoperative symptom management in patients with lung 12 13 73 cancer. 14 15 74 2. It is a multicentre, randomised controlled trial, conducted in six tertiary hospitals in 16 17 75 China. 18 For peer review only 19 20 76 3. It focuses on the early postoperative period with a high frequency of data collection, 21 22 77 including baseline before surgery, daily in-hospital after surgery, and twice a week after 23 24 78 discharge until 4 weeks or the start of postoperative oncologic treatment. 25 26 27 79 4. We use a lung cancer-specific scale and the recall period is 24 hours, which is more 28 suitable to measure rapidly changing symptoms during the early recovery phase. 29 80 30 31 81 5. The lack of blinding for the participants and specialists delivering the intervention 32 33 82 may be a limitation. 34 35 36 83 37 http://bmjopen.bmj.com/ 38 84 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 5 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 31

1 2 3 85 INTRODUCTION 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 86 Lung cancer is the most common cancer and the leading cause of cancer death in China 7 8 87 and worldwide.1 2 With the application of low-dose computed tomography in screening, 9 10 88 more and more patients with early stage lung cancer are being diagnosed and treated 11 12 89 with surgery. However, the postoperative symptom burden of lung cancer patients is 13 14 90 very severe, and this detrimentally affects their quality of life (QOL). Patients have 15 16 91 various symptoms, such as pain, coughing, fatigue, and shortness of breath, in the early 17 18 92 stages after surgeryFor or even peer a long time review after surgery.3-5 onlyLowery et al followed 183 lung 19 20 93 cancer patients for 1-6 years and found that 79.8% of them had a variety of symptoms.3 21 22 94 Among these patients, 30.6% had one symptom, 27.9% had two symptoms, and 21.3% 23 24 95 had three or more symptoms. The most frequent symptoms were pain and shortness of 25 26 96 breath. If these symptoms are not effectively controlled, the postoperative recovery of 27 3 4 28 97 patients will be severely affected, resulting in a poor QOL. Therefore, effective 29 30 98 interventions are needed to alleviate post-surgery symptoms in patients with lung 31 3 4 32 99 cancer. 33 34 100 Symptom management is the foundation of clinical care, particularly for patients 35 36 101 with cancer. Patient-reported outcome (PRO)-based symptom management is 37 http://bmjopen.bmj.com/ 38 102 increasingly regarded as an optimal model for patient-centred care.6 7 A PRO is a 39 40 103 measurement of a patient's health status that comes directly from the patient's subjective 41 42 104 evaluation, with no interpretation by medical providers or anyone else.8 Studies have 43 44 105 shown that it may be more accurate for patients to evaluate their own health status 45 on October 1, 2021 by guest. Protected copyright. 46 106 themselves than evaluation by medical providers.9 The application of PRO-based 47 48 107 symptom monitoring and alerting followed by real-time symptom management from 49 50 108 health care professionals can improve the QOL of patients, prolong survival, increase 51 52 109 patient satisfaction, and allow evaluation of the treatment method.6 10-15 Basch et al 53 54 110 reported a randomised controlled trial (RCT) result, suggesting that PRO-based 55 56 111 proactive symptom monitoring could improve symptom management and thus bring 57 10 11 58 112 survival benefits in patients undergoing chemotherapy. When compared to a 59 6 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 31 BMJ Open

1 2 3 113 traditional reactive monitoring group, median survival was 5.2 months longer among 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 10 6 114 patients in the proactive monitoring group (31.2 vs. 26.0 months, p = 0.03). However, 7 8 115 it is still not clear if adequate symptom control and improved QOL in the surgical 9 10 116 population can ensure a potential better survival. 11 12 117 Currently, PROs are mainly used in non-surgical treatment settings,10-13 and they are 13 14 118 still in the early stage of application in the surgical treatment setting.5 7 14-28 Studies on 15 16 119 postoperative symptom management of lung cancer, especially in the early 17 18 120 postoperative periodFor are lacking.peer In addition, review most of the only published literature has a low 19 20 121 level of evidence, due to its design.5 18-28 The limitations of these studies include: (1) 21 22 122 most were observational studies; (2) they had small sample sizes ranging from 30 to 23 24 123 200 subjects with few exceptions; (3) they did not focus on the early postoperative 25 26 124 period, typically including the in-hospital period immediately after surgery and 4 weeks 27 28 125 after discharge when patients frequently report multiple severe symptoms, leading to 29 30 126 later negative recovery events, i.e. higher symptom burden, delayed return to intended 31 32 127 oncologic therapy, and poorer QOL; (4) they used a variety of survey instruments and 33 34 128 some were not a lung cancer-specific scale; (5) most of the scales used such as the 35 36 129 European Organization for Research and Treatment of Cancer Quality of Life 37 http://bmjopen.bmj.com/ 38 130 Questionnaire–Lung Cancer Module (EORTC QLQ-LC13), had a recall period of 1 39 131 week, which may not be able to identify the rapidly changing symptoms during early 40 41 132 postoperative phase;5 (6) only one study assessed in-hospital patients immediately after 42 43 133 surgery, but the MD Anderson Symptom Inventory (MDASI) scale used did not include 44 45 5 134 lung cancer-specific symptoms, i.e. coughing; (7) the symptom assessments were on October 1, 2021 by guest. Protected copyright. 46 47 135 inadequate, mostly at just two or three time points post-surgery; (8) most of the surgical 48 49 136 approaches were thoracotomies, not representing the current mainstream minimally 50 51 137 invasive thoracoscopic surgery for lung cancer; (9) there were very few studies on the 52 53 138 Chinese population. 54 55 56 139 We have been conducting an observational study of perioperative symptom in 57 58 140 patients with lung cancer based on PRO (registration number NCT03341377). Now, 59 7 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 31

1 2 3 141 we propose an RCT, aiming to evaluate the efficacy of a PRO-based symptom 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 142 monitoring, alerting, and response system (SMARS) to improve postoperative recovery 7 8 143 of lung cancer patients. The SMARS (figure 1) includes a research electronic data 9 10 144 capture (REDCap) platform, an electronic PRO system (ePRO Hub), and a most 11 popular social software (WeChat)29 in China. This study will provide evidence for early 12 145 13 146 postoperative phase symptom management for patients with lung cancer. We will use 14 15 147 the MDASI lung cancer-specific scale (MDASI-LC)30 to frequently monitor symptoms 16 17 148 and their impact on the functioning of lung cancer patients from pre-operation to 4 18 For peer review only 19 149 weeks after discharge or until the beginning of postoperative oncologic treatment. The 20 21 150 recall period of MDASI-LC is 24 hours, which is more suitable to measure rapidly 22 23 151 changing, early postoperative period symptoms compared with other QOL scales for 1 24 25 152 week or longer. Our research hypothesis is that patients with lung cancer undergoing 26 27 153 PRO-based symptom management have a lower postoperative symptom burden than 28 29 154 patients undergoing standard symptom management. 30 31 32 155 33 34 156 METHODS AND ANALYSIS 35 36 37 157 Study design http://bmjopen.bmj.com/ 38 39 40 158 The trial is a multicentre, randomised, parallel group controlled, and superiority design. 41 159 This protocol will be consistent with the Standard Protocol Items: Recommendations 42 43 160 For Interventional Trials (SPIRIT).31 The results of this trial will be reported according 44 45 32 161 to the guidelines of Consolidated Standards of Reporting Trials (CONSORT). A flow on October 1, 2021 by guest. Protected copyright. 46 47 162 chart of this trial is shown in figure 2. 48 49 50 163 51 52 53 164 Setting 54 55 165 Participants will be recruited from six tertiary hospitals in different cities in China. The 56 57 166 six hospitals are Sichuan Cancer Hospital, Zigong First People's Hospital, Jiangyou 58 59 8 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 31 BMJ Open

1 2 3 167 People's Hospital, Dazhu County People's Hospital, The Third People's Hospital of 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 168 Chengdu, and The Seventh People's Hospital of Chengdu. The total number of lung 7 8 169 cancer operations in six hospitals is approximately 2,000 per year. 9 10 170 11 12 13 171 Participant recruitment 14 15 172 Participant recruitment will be carried out before the surgery by participating clinicians. 16 17 173 Eligible patients should meet all the inclusion criteria and not meet any of the exclusion 18 For peer review only 19 174 criteria. The inclusion criteria for the participants are: (1) aged 18 to 75 years, (2) 20 21 175 clinically diagnosed as primary lung cancer, (3) clinical stage I-IIIA (8th edition),33 and 22 23 176 (4) planning to receive surgery, and (5) able and willing to respond to a repeated 24 25 177 electronic questionnaire (e-questionnaire) on a smartphone or a tablet. The exclusion 26 27 178 criteria are: (1) history of neoadjuvant therapy, (2) having other malignant tumours, and 28 29 179 (3) unable to understand the study requirements. Strategies for achieving adequate 30 31 180 participant enrolment to reach a target sample size include inviting more doctors in each 32 33 181 centre to participate in the study and adding more research centres. Plans to promote 34 35 182 participant retention and complete follow-up include education, refill reminders, and 36 37 183 commitments to provide all the patients with free long-term medical consultations after http://bmjopen.bmj.com/ 38 39 184 the trial via WeChat. The anticipated dates of the study are from December 1, 2018 to 40 41 185 December 31, 2020. We haven't started recruiting patients yet. 42 43 186 44 45 on October 1, 2021 by guest. Protected copyright. 46 187 Sample size calculation 47 48 49 188 The primary end point of this study will be the mean symptom threshold events, defined 50 51 189 as the average number of target symptom threshold events per patient, at each time 52 34 53 190 point. To meet the minimal clinically important difference (0.5 standard deviation) 54 for the mean symptom threshold events, the required sample size is 64 for each group, 55 191 56 192 when rejecting the null hypothesis (the difference between the two groups < 0.5 57 58 59 9 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 31

1 2 3 193 standard deviation). A total of 128 cases with valid data are needed. Considering a 20% 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 194 attrition rate, we will need 80 patients for each group (64/0.8). The sample size 7 8 195 calculation is based on the independent sample Student’s t-test, using a two-tailed alpha 9 10 196 level of 0.05 and a beta error probability of 0.02 (80% power). 11 12 197 13 14 15 198 Randomisation and allocation concealment 16 17 199 The process of randomisation will be carried out online using the central randomisation 18 For peer review only 19 200 module on the REDCap platform (http://125.71.214.100:888/redcap) after a participant 20 21 201 has been recruited to the study and has signed an informed consent form. The data 22 23 202 analyst will upload the randomisation allocation table to the REDCap platform and then 24 25 203 save the randomisation allocation table independently. The investigator will conduct 26 27 204 randomisation by clicking on a randomisation button on the REDCap platform. It will 28 29 205 then be allocated to the intervention group or control group with a 1:1 ratio. Each group 30 31 206 will have 80 cases. 32 33 34 207 35 36 208 Blinding 37 http://bmjopen.bmj.com/ 38 39 209 The blinding of participants and specialists delivering the intervention is impossible 40 41 210 due to the nature of the interventions. But the data collectors who help administer PRO 42 43 211 collection will be blinded to group allocation to minimise measurement bias. The 44 45 212 statisticians analysing the results will also be blinded to group allocation. on October 1, 2021 by guest. Protected copyright. 46 47 213 48 49 50 214 Intervention 51 52 53 215 After enrolment, all the patients will use their WeChat app to connect with the 54 55 216 participating specialists’ WeChat app via a mini program (ePRO Cell). Then, they will 56 be taught how to use the program. The ePRO questionnaires will be set to send to the 57 217 58 59 10 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 31 BMJ Open

1 2 3 218 patients’ WeChat app automatically after randomisation. Patients are required to 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 219 complete the ePRO questionnaires on their smartphone or tablet before surgery 7 8 220 (baseline, typically 1 to 3 days before the operation), daily after surgery (in-hospital, 9 10 221 typically 1 to 7 days after the operation), and twice a week after discharge until 4 weeks 11 or the start of postoperative oncologic treatment (typically collecting PRO data six to 12 222 13 223 eight times after discharge). In a hospital setting, if the patients do not complete the 14 15 224 ePRO questionnaires within the scheduled time, an electronic reminder (e-reminder) 16 17 225 and up to two bedside reminders will be delivered at the same day. After discharge, if 18 For peer review only 19 226 the patients fail to complete the ePRO questionnaires within the scheduled time, an e- 20 21 227 reminder and up to two phone reminders will be delivered with 24 hours. 22 23 24 228 25 26 Comparison 27 229 28 29 230 Intervention group 30 31 32 231 Patients will not be informed about the threshold levels. When there are one or more 33 34 232 target symptoms (pain, coughing, fatigue, disturbed sleep, and shortness of breath) and 35 36 233 scores reach the pre-set intervention threshold (score ≥ 4), the participating specialist 37 http://bmjopen.bmj.com/ 38 234 (thoracic surgeon) will simultaneously receive an alert message on his or her WeChat. 39 40 235 Then the specialist will mainly use the WeChat or sometimes a telephone to contact the 41 236 patient within 24 hours to implement symptom relief measures, e.g. consultation, 42 43 237 education, medication guidance, and clinic or hospital visit suggestions. The symptom 44 45 238 relief measures of the intervention group patients will comply with the latest guidelines on October 1, 2021 by guest. Protected copyright. 46 47 239 and be standardised across all centres, in the form of a standard operating procedure 48 49 240 (SOP) handbook. Patients’ adherence to the interventions will be asked at each time 50 51 241 point. Those who do not follow the specialist’s advice will be monitored, and the 52 53 242 number of violations will be recorded. Those who refuse to follow the specialist’s 54 55 243 advice more than three times will be considered as seriously violating the study protocol 56 57 244 and will be withdrawn. 58 59 11 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 31

1 2 3 245 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 246 Control group 7 8 9 247 The control group patients will be informed that the ePRO data collected are only for 10 11 248 scientific research. They will not generate any alerts or get responses relating to their 12 13 249 symptoms. The patients' symptom management will follow the current standard 14 15 250 postoperative management model. During hospitalization, the doctors manage the 16 17 251 control group patients’ symptoms based on their own judgement rather than the score 18 For peer review only 19 252 of the PRO. After discharge, the patients will go home and the first clinic visit is 20 21 253 approximately 4 weeks later. Patients will be encouraged to seek medical help if severe 22 23 254 symptoms are reported. 24 25 255 26 27 28 256 Withdrawal criteria 29 30 257 Participants will be withdrawn from the study and no further data will be collected if 31 32 258 they meet the following criteria: (1) unexpected cancellation of surgery, (2) severe 33 34 259 postoperative complications affecting symptom data collection, (3) postoperative 35 36 260 length of stay > 14 days, (4) postoperative pathology shows non-primary lung cancer, 37 http://bmjopen.bmj.com/ 38 261 (5) non-R0 resection, (6) pathological stage IV, (7) participant seriously violates the 39 40 262 study protocol (continually not complying with the specialist’s advice, intentionally 41 42 263 letting a proxy to complete the PRO surveys, and deliberately providing false PROs), 43 44 264 or (8) participant asks to withdraw from the study. 45 on October 1, 2021 by guest. Protected copyright. 46 47 265 48 49 266 Outcomes and measurement 50 51 52 267 Primary outcome 53 54 55 268 The primary end point of this study is the mean symptom threshold events. According 56 to our pilot study, the five most common postoperative symptoms of lung cancer 57 269 58 59 12 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 31 BMJ Open

1 2 3 270 patients are: pain, coughing, fatigue, disturbed sleep and shortness of breath. In this 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 271 study, these five symptoms assessed by the MDASI-LC are defined as target symptoms. 7 8 272 According to the recommendation of National Comprehensive Cancer Network and 9 10 273 published literature, when a patient's symptom score is ≥ 4, it is identified as moderate 11 severity.35 36 In this study, a score of 4 is set as the threshold value for intervention, and 12 274 13 275 a target symptom score of ≥ 4 is reported as a threshold event. 14 15 16 276 The primary PRO tool used in this study is the MDASI-LC.30 It is a measure that 17 18 277 contains sixteenFor items of peer lung cancer-related review and treatment-related only symptoms, and six 19 20 278 items of interference to normal daily life caused by symptoms. All items are rated on 21 22 279 0-10 numerical scales, with 0 representing “symptom not present” or “symptom not 23 24 280 interfered with life” and 10 representing “symptom as bad as one can imagine” or 25 26 281 “symptom completely interfered with life”. The recall period of the MDASI-LC is 24 27 28 282 hours and it can usually be conducted in 5 minutes. It has been translated and validated 29 30 283 for application in a Chinese context. 31 32 284 33 34 35 285 Secondary outcomes 36 37 http://bmjopen.bmj.com/ 38 286 The secondary end points of this study include trajectories of PROs (symptom severity, 39 40 287 daily functioning, and QOL) and revisit rate after discharge. Trajectories of PROs are 41 288 defined as the longitudinal changing pattern of the mean score of the five target 42 43 289 symptoms for symptom severity, the mean score of the six MDASI-LC interference 44 45 290 items for daily functioning, and the mean score of the single-item QOL scale on October 1, 2021 by guest. Protected copyright. 46 47 291 (UNISCALE) for QOL,37 from the baseline to 4 weeks after discharge or until the start 48 49 292 of postoperative oncologic treatment. UNISCALE only has one question using a 0-10 50 51 293 scale, with 0 representing “worst QOL” and 10 representing “best QOL”. The revisit 52 53 294 rate after discharge is defined as the ratio of the number of patients who see the doctor 54 55 295 again after discharge including outpatient visits, emergency visits and hospitalisation 56 57 296 divided by the total number of patients. 58 59 13 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 31

1 2 3 297 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 298 Other data 7 8 9 299 The clinician workload, clinician system acceptability, and patient satisfaction of the 10 11 300 interventions will be assessed through surveys and interviews. Demographics, 12 13 301 clinicopathological characteristics, follow-up information, and adverse events of the 14 15 302 interventions will also be collected. All the adverse events will be assessed and 16 17 303 managed by a thoracic surgeon. 18 For peer review only 19 304 20 21 22 305 Data collection, management, and quality control 23 24 38 39 25 306 REDCap, a worldwide popular research data collection and management platform 26 27 307 established in Sichuan Cancer Hospital, will be used for data collection and 28 management in this study. PRO data will be collected using e-questionnaires and 29 308 30 309 recorded in REDCap. Participants should fill out the e-questionnaires by themselves. If 31 32 310 participants have difficulties in completing the e-questionnaires, data collectors or their 33 34 311 family members will help them by just reading each item aloud and recording the 35 36 312 participant’s responses. The control group patients’ PRO data will not be accessed by 37 http://bmjopen.bmj.com/ 38 313 the specialists. Specialists can only access the PRO data of the intervention group 39 40 314 patients. Other data including demographics, clinicopathological characteristics, and 41 42 315 follow-up information will also be entered into the REDCap database. 43 44 45 316 Data will be checked regularly by the quality controller. Participant privacy on October 1, 2021 by guest. Protected copyright. 46 47 317 information will not be recorded in REDCap. A study number will be allocated to each 48 49 318 participant and will be used on all study documentation, which will only be available 50 51 319 to the investigators. Before patients’ enrolment, investigators from each research centre 52 53 320 will receive SOP training. Each centre will receive on-site monitoring visits, telephone 54 monitoring, and online guidance during the course of the trial. 55 321 56 57 322 58 59 14 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 31 BMJ Open

1 2 3 323 Data analysis 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 324 Per-protocol analyses will be conducted. To be included in the analysis, a participant 7 8 325 must provide MDASI-LC data from the baseline and at least two additional time points. 9 10 326 If a participant meets the withdrawal criteria, no data will be included in the analysis. 11 12 327 Two-sided P values of < 0.05 are considered to be statistically significant. 13 14 328 Continuous variables will be presented as mean ± standard deviation or median and 15 16 329 interquartile range. Comparisons between groups will be conducted using the Student’s 17 18 330 t-test or the WilcoxonFor peer rank sum test.review Categorical variablesonly will be presented as 19 20 331 frequencies or proportions and compared between groups using the chi-square test. 21 22 332 Trajectories of PROs will be compared between the intervention group and control 23 24 333 group using generalised mixed effects models. Missing data will be processed by the 25 26 334 multiple imputation method. Results obtained from data without missed observations 27 28 335 will be compared with that from imputed data for sensitive analysis. 29 30 336 31 32 33 337 Data monitoring and interim analysis 34 35 36 338 A data monitoring committee (DSM) consisting of one clinician, one statistician, and 37 http://bmjopen.bmj.com/ 38 339 the secretary of the Ethics Committee of Sichuan Cancer Hospital will be set up. Study 39 40 340 monitoring will be carried out regularly by DSM members and the process will be 41 341 independent from investigators. Due to the low-risk of the study content and short-term 42 43 342 study duration, interim analysis will not be performed. 44 45 on October 1, 2021 by guest. Protected copyright. 46 343 47 48 49 344 Data availability statement 50 51 345 Deidentified data generated by this clinical trial to support future research articles will 52 53 346 be available from the corresponding author on reasonable request. 54 55 56 347 57 58 59 15 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 31

1 2 3 348 Patient and public involvement statement 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 349 Patients and the public will not be involved in the design, recruitment to, or conduct of 7 8 350 this study. The burden of the intervention in this RCT will not be performed. We will 9 10 351 inform the applicants of the results. There are no plans to disseminate the results to 11 12 352 study participants. Participants will be informed that they can obtain the final results of 13 14 353 this study through our future published articles. 15 16 17 354 18 For peer review only 19 355 ETHICS AND DISSEMINATION 20 21 22 356 This study was approved by the Ethics Committee of Sichuan Cancer Hospital on 23 24 357 November 22, 2018 (No. SCCHEC-02-2018-045). All recruited patients will be 25 26 358 required to give written informed consent. Any subsequent amendments to the protocol 27 28 359 will be submitted for further review and approval. Sub-centres will gain approval from 29 30 360 their hospital-specific ethics committees. The results of this study will be disseminated 31 32 361 through peer-reviewed publications and academic conferences. 33 34 362 35 36 37 363 DISCUSSION http://bmjopen.bmj.com/ 38 39 40 364 This trial focuses on the early-phase postoperative symptom management after lung 41 42 365 cancer surgery. The potential implications of the findings include: (1) identifying if 43 366 PRO-based symptom management is better than usual symptom management, (2) 44 45 367 identifying if proactive symptom management can reduce symptom burden and on October 1, 2021 by guest. Protected copyright. 46 47 368 improve QOL in the surgical population, (3) laying a foundation for future research on 48 49 369 whether postoperative symptom management improves survival, (4) investigating 50 51 370 whether SMARS is feasible and acceptable in real-world clinical practice in China, and 52 53 371 (5) identifying barriers which will be used to facilitate further revisions of the SMARS 54 55 372 and help extend its implementation in non-surgical settings. 56 57 58 373 There are many limitations in this trial. First, the trial will be carried out in well- 59 16 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 31 BMJ Open

1 2 3 374 resourced tertiary hospitals in China. This will limit the generalizability of this study. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 375 Second, the inclusion criteria and exclusion criteria are strict. For example, the program 7 8 376 is unsuitable for patients without internet access or with poor literacy. This will greatly 9 10 377 limit the population for which this study is applicable. Third, the lack of blinding for 11 the participants and specialists delivering the intervention will also be a limitation, 12 378 13 379 because it may increase the measurement bias. Fourth, the follow-up period is very 14 15 380 short. The results need confirmation in a study with a longer follow-up period. 16 17 18 381 In summary, For as a RCT, peer this study review will not only testonly the efficacy of SMARS in 19 20 382 postoperative care, but also it will provide data of feasibility for further unblinded 21 22 383 pragmatic study when implementing the SMARS in the real world, with the 23 24 384 involvement of community hospitals and patients with poor socioeconomic status, 25 26 385 while a wider internet access is available for the whole Chinese population. 27 28 386 29 30 31 387 Acknowledgements The authors thank all the patients and patient advisers who are involved in this 32 33 388 study. 34 35 36 389 Author contributions WD and QLS conceived and designed the study. WD and QLS obtained the 37 funding. WD is the chief investigator of this study. YQZ, WHF, XQL, YFM, and RZ are sub-centre http://bmjopen.bmj.com/ 38 390 39 391 principal investigators who contributed to the trial feasibility stage. WD, YQZ, WHF, XQL, YFM, 40 41 392 RZ, XW, CMW and SHX drafted the protocol. QLS participated in the statistical plan. QL and QLS 42 43 393 revised the manuscript. All authors have read and approved the manuscript. 44 45 on October 1, 2021 by guest. Protected copyright. 46 394 Funding This work was supported by [Bethune charitable foundation], [Sichuan Science and 47 48 395 Technology Program] grant number [18PJ436] and [National Natural Science Foundation of China] 49 50 396 grant number [81872506]. 51 52 Competing interests None declared. 53 397 54 55 398 Patient consent Obtained. 56 57 58 399 Ethics approval Ethics Committee of Sichuan Cancer Hospital (No. SCCHEC-02-2018-045). 59 17 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 31

1 2 3 400 Provenance and peer review Not commissioned; externally peer reviewed. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 401 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 18 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 31 BMJ Open

1 2 3 402 REFERENCES 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 403 1. Chen W, Zheng R, Baade PD, et al. Cancer statistics in China, 2015. CA Cancer J Clin 7 8 404 2016;66:115-32. 9 10 405 2. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates 11 12 406 of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 13 14 407 2018;68:394-424. 15 16 408 3. Lowery AE, Krebs P, Coups EJ, et al. Impact of symptom burden in post-surgical non-small cell 17 18 409 lung cancer survivors.For Support peer Care Cancer review 2014;22:173-80. only 19 20 410 4. Yang P, Cheville AL, Wampfler JA, et al. Quality of life and symptom burden among long-term 21 22 411 lung cancer survivors. J Thorac Oncol 2012;7:64-70. 23 24 412 5. Fagundes CP, Shi Q, Vaporciyan AA, et al. Symptom recovery after thoracic surgery: Measuring 25 26 413 patient-reported outcomes with the MD Anderson Symptom Inventory. J Thorac Cardiovasc 27 28 414 Surg 2015;150:613-9 e2. 29 30 415 6. Basch E. Patient-Reported Outcomes - Harnessing Patients' Voices to Improve Clinical Care. N 31 32 416 Engl J Med 2017;376:105-8. 33 34 417 7. Khullar OV, Fernandez FG. Patient-Reported Outcomes in Thoracic Surgery. Thorac Surg Clin 35 2017;27:279-90. 36 418 37 419 8. U.S. Department of Health and Human Services FDA Center for Drug Evaluation and Research, http://bmjopen.bmj.com/ 38 39 420 U.S. Department of Health and Human Services FDA Center for Biologics Evaluation and 40 41 421 Research, U.S. Department of Health and Human Services FDA Center for Devices and 42 43 422 Radiological Health. Guidance for industry: patient-reported outcome measures: use in medical 44 45 423 product development to support labeling claims: draft guidance. Health Qual Life Outcomes on October 1, 2021 by guest. Protected copyright. 46 47 424 2006;4:79. 48 49 425 9. Basch E. The missing voice of patients in drug-safety reporting. N Engl J Med 2010;362:865-9. 50 51 426 10. Basch E, Deal AM, Dueck AC, et al. Overall Survival Results of a Trial Assessing Patient- 52 53 427 Reported Outcomes for Symptom Monitoring During Routine Cancer Treatment. JAMA 54 55 428 2017;318:197-8. 56 57 429 11. Basch E, Deal AM, Kris MG, et al. Symptom Monitoring With Patient-Reported Outcomes 58 59 19 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 31

1 2 3 430 During Routine Cancer Treatment: A Randomized Controlled Trial. J Clin Oncol 2016;34:557- 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 431 65. 7 8 432 12. Novello S, Kaiser R, Mellemgaard A, et al. Analysis of patient-reported outcomes from the 9 LUME-Lung 1 trial: a randomised, double-blind, placebo-controlled, Phase III study of second- 10 433 11 434 line nintedanib in patients with advanced non-small cell lung cancer. Eur J Cancer 2015;51:317- 12 13 435 26. 14 15 436 13. Popat S. Patient reported outcomes from LUX-Lung 3: first-line afatinib is superior to 16 17 437 chemotherapy-would patients agree? Ann Palliat Med 2014;3:19-21. 18 For peer review only 19 438 14. Ng S, Pusic A, Parker E, et al. Patient-Reported Outcome Measures for Breast Implant Surgery: 20 21 439 A Pilot Study. Aesthet Surg J 2019;sjz023.doi:10.1093/asj/sjz023.[Epub ahead of print]. 22 23 440 15. Coronini-Cronberg S, Appleby J, Thompson J. Application of patient-reported outcome 24 25 441 measures (PROMs) data to estimate cost-effectiveness of hernia surgery in England. J R Soc 26 27 442 Med 2013;106:278-87. 28 29 443 16. Poghosyan H, Sheldon LK, Leveille SG, et al. Health-related quality of life after surgical 30 31 444 treatment in patients with non-small cell lung cancer: a systematic review. Lung Cancer 32 33 445 2013;81:11-26. 34 35 446 17. Cleeland CS, Wang XS, Shi Q, et al. Automated symptom alerts reduce postoperative symptom 36 37 447 severity after cancer surgery: a randomized controlled clinical trial. J Clin Oncol 2011;29:994- http://bmjopen.bmj.com/ 38 39 448 1000. 40 41 449 18. Khullar OV, Rajaei MH, Force SD, et al. Pilot Study to Integrate Patient Reported Outcomes 42 43 450 After Lung Cancer Operations Into The Society of Thoracic Surgeons Database. Ann Thorac 44 45 451 Surg 2017;104:245-53. on October 1, 2021 by guest. Protected copyright. 46 47 452 19. Shi Q, Wang XS, Vaporciyan AA, et al. Patient-Reported Symptom Interference as a Measure 48 of Postsurgery Functional Recovery in Lung Cancer. J Pain Symptom Manage 2016;52:822-31. 49 453 50 454 20. Li WW, Lee TW, Lam SS, et al. Quality of life following lung cancer resection: video-assisted 51 52 455 thoracic surgery vs thoracotomy. Chest 2002;122:584-9. 53 54 456 21. Kenny PM, King MT, Viney RC, et al. Quality of life and survival in the 2 years after surgery 55 56 457 for non small-cell lung cancer. J Clin Oncol 2008;26:233-41. 57 58 59 20 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 31 BMJ Open

1 2 3 458 22. Balduyck B, Hendriks J, Lauwers P, et al. Quality of life after lung cancer surgery: a prospective 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 459 pilot study comparing bronchial sleeve lobectomy with pneumonectomy. J Thorac Oncol 7 8 460 2008;3:604-8. 9 23. Ferguson MK, Parma CM, Celauro AD, et al. Quality of life and mood in older patients after 10 461 11 462 major lung resection. Ann Thorac Surg 2009;87:1007-12; discussion 12-3. 12 13 463 24. Sartipy U. Prospective population-based study comparing quality of life after pneumonectomy 14 15 464 and lobectomy. Eur J Cardiothorac Surg 2009;36:1069-74. 16 17 465 25. Ostroff JS, Krebs P, Coups EJ, et al. Health-related quality of life among early-stage, non-small 18 For peer review only 19 466 cell, lung cancer survivors. Lung Cancer 2011;71:103-8. 20 21 467 26. Moller A, Sartipy U. Long-term health-related quality of life following surgery for lung cancer. 22 23 468 Eur J Cardiothorac Surg 2012;41:362-7. 24 25 469 27. Zhao J, Zhao Y, Qiu T, et al. Quality of life and survival after II stage nonsmall cell carcinoma 26 27 470 surgery: Video-assisted thoracic surgery versus thoracotomy lobectomy. Indian J Cancer 28 29 471 2015;52 Suppl 2:e130-3. 30 31 472 28. Yun YH, Kim YA, Sim JA, et al. Prognostic value of quality of life score in disease-free 32 33 473 survivors of surgically-treated lung cancer. BMC Cancer 2016;16:505. 34 35 474 29. Montag C, Becker B, Gan C. The Multipurpose Application WeChat: A Review on Recent 36 37 475 Research. Front Psychol 2018;9:2247. http://bmjopen.bmj.com/ 38 39 476 30. Mendoza TR, Wang XS, Lu C, et al. Measuring the symptom burden of lung cancer: the validity 40 41 477 and utility of the lung cancer module of the M. D. Anderson Symptom Inventory. Oncologist 42 43 478 2011;16:217-27. 44 45 479 31. Chan AW, Tetzlaff JM, Altman DG, et al. SPIRIT 2013 statement: defining standard protocol on October 1, 2021 by guest. Protected copyright. 46 47 480 items for clinical trials. Ann Intern Med 2013;158:200-7. 48 49 481 32. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting 50 51 482 parallel group randomized trials. Ann Intern Med 2010;152:726-32. 52 53 483 33. Goldstraw P, Chansky K, Crowley J, et al. The IASLC Lung Cancer Staging Project: Proposals 54 55 484 for Revision of the TNM Stage Groupings in the Forthcoming (Eighth) Edition of the TNM 56 57 485 Classification for Lung Cancer. J Thorac Oncol 2016;11:39-51. 58 59 21 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 31

1 2 3 486 34. Revicki DA, Cella D, Hays RD, et al. Responsiveness and minimal important differences for 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 487 patient reported outcomes. Health & Quality of Life Outcomes 2006;4:70. 7 8 488 35. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: 9 Adult Cancer Pain. Version 1.2019. Available at: https://www.nccn.org/professionals/physician 10 489 11 490 _gls/pdf/pain.pdf. 12 13 491 36. Wang XS, Zhao F, Fisch MJ, et al. Prevalence and characteristics of moderate to severe fatigue: 14 15 492 a multicenter study in cancer patients and survivors. Cancer 2014;120:425-32. 16 17 493 37. Sloan JA, Loprinzi CL, Kuross SA, et al. Randomized comparison of four tools measuring 18 For peer review only 19 494 overall quality of life in patients with advanced cancer. J Clin Oncol 1998;16:3662-73. 20 21 495 38. Tomko RL, Gray KM, Oppenheimer SR, et al. Using REDCap for ambulatory assessment: 22 23 496 Implementation in a clinical trial for smoking cessation to augment in-person data collection. 24 25 497 Am J Drug Alcohol Abuse 2018:1-16. 26 27 498 39. Harvey LA. REDCap: web-based software for all types of data storage and collection. Spinal 28 29 499 Cord 2018;56:625. 30 31 32 500 33 34 501 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 22 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 31 BMJ Open

1 2 3 502 Figure legends 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 503 Figure 1 Schematic diagram of the SMARS. 7 8 9 504 Figure 2 Flow chart of this parallel group randomised trial. 10 11 12 505 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 23 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 31 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 Schematic diagram of the SMARS. 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on October 1, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 31 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 Flow chart of this parallel group randomised trial. 38 39 40

41 on October 1, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 31

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 7 8 SPIRIT 2013 Checklist: Recommended items to address in a clinical trial protocol and 9 related documents* 10 11 Section/item Item Description Page Number 12 No on which item 13 is reported 14 15 Administrative information 16 17 Title 1 Descriptive title identifying the study design, Page 1 18 Forpopulation, peer interventions, review and, if applicable, only trial 19 20 acronym 21 22 Trial 2a Trial identifier and registry name. If not yet registered, Page 4 23 registration name of intended registry 24 25 2b All items from the World Health Organization Trial Page 4 26 Registration Data Set 27 28 Protocol 3 Date and version identifier Page 3 29 version 30 31 Funding 4 Sources and types of financial, material, and other Page 17-18 32 33 support 34 35 Roles and 5a Names, affiliations, and roles of protocol contributors Page 1, 17 36 responsibilitie 37 5b Name and contact information for the trial sponsor Page 1, 17 s http://bmjopen.bmj.com/ 38 39 5c Role of study sponsor and funders, if any, in study Page 17-18 40 design; collection, management, analysis, and 41 42 interpretation of data; writing of the report; and the 43 decision to submit the report for publication, including 44 whether they will have ultimate authority over any of 45 these activities on October 1, 2021 by guest. Protected copyright. 46 47 5d Composition, roles, and responsibilities of the Page 14-15 48 49 coordinating centre, steering committee, endpoint 50 adjudication committee, data management team, and 51 other individuals or groups overseeing the trial, if 52 applicable (see Item 21a for data monitoring 53 54 committee) 55 56 Introduction 57 58 59 60

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 31 BMJ Open

1 2 Background 6a Description of research question and justification for Page 6-7 3 and rationale undertaking the trial, including summary of relevant

4 studies (published and unpublished) examining BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 benefits and harms for each intervention 7 8 6b Explanation for choice of comparators Page 11-12 9 10 Objectives 7 Specific objectives or hypotheses Page 8 11 12 Trial design 8 Description of trial design including type of trial (eg, Page 8 13 parallel group, crossover, factorial, single group), 14 allocation ratio, and framework (eg, superiority, 15 equivalence, noninferiority, exploratory) 16 17 18 Methods: Participants,For interventions, peer andreview outcomes only 19 20 Study setting 9 Description of study settings (eg, community clinic, Page 8-9 21 academic hospital) and list of countries where data 22 will be collected. Reference to where list of study sites 23 24 can be obtained 25 26 Eligibility 10 Inclusion and exclusion criteria for participants. If Page 9 27 criteria applicable, eligibility criteria for study centres and 28 individuals who will perform the interventions (eg, 29 surgeons, psychotherapists) 30 31 Interventions 11a Interventions for each group with sufficient detail to Page 10-11 32 33 allow replication, including how and when they will be 34 administered 35 36 11b Criteria for discontinuing or modifying allocated Page 12 37 interventions for a given trial participant (eg, drug http://bmjopen.bmj.com/ 38 dose change in response to harms, participant 39 40 request, or improving/worsening disease) 41 42 11c Strategies to improve adherence to intervention Page 9 43 protocols, and any procedures for monitoring 44 adherence (eg, drug tablet return, laboratory tests) 45 on October 1, 2021 by guest. Protected copyright. 46 11d Relevant concomitant care and interventions that are Page 11-12 47 48 permitted or prohibited during the trial 49 50 Outcomes 12 Primary, secondary, and other outcomes, including Page 13-14 51 the specific measurement variable (eg, systolic blood 52 pressure), analysis metric (eg, change from baseline, 53 final value, time to event), method of aggregation (eg, 54 55 median, proportion), and time point for each outcome. 56 Explanation of the clinical relevance of chosen 57 efficacy and harm outcomes is strongly recommended 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 31

1 2 Participant 13 Time schedule of enrolment, interventions (including Page 8-9 3 timeline any run-ins and washouts), assessments, and visits

4 for participants. A schematic diagram is highly BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 recommended (see Figure) 7 8 Sample size 14 Estimated number of participants needed to achieve Page 9-10 9 study objectives and how it was determined, including 10 clinical and statistical assumptions supporting any 11 sample size calculations 12 13 Recruitment 15 Strategies for achieving adequate participant Page 9 14 15 enrolment to reach target sample size 16 17 Methods: Assignment of interventions (for controlled trials) 18 For peer review only 19 Allocation: 20 21 Sequence 16a Method of generating the allocation sequence (eg, Page 10 22 generation computer-generated random numbers), and list of any 23 factors for stratification. To reduce predictability of a 24 random sequence, details of any planned restriction 25 26 (eg, blocking) should be provided in a separate 27 document that is unavailable to those who enrol 28 participants or assign interventions 29 30 Allocation 16b Mechanism of implementing the allocation sequence Page 10 31 concealme (eg, central telephone; sequentially numbered, 32 33 nt opaque, sealed envelopes), describing any steps to 34 mechanis conceal the sequence until interventions are assigned 35 m 36 37 Implement 16c Who will generate the allocation sequence, who will Page 10 http://bmjopen.bmj.com/ 38 39 ation enrol participants, and who will assign participants to 40 interventions 41 42 Blinding 17a Who will be blinded after assignment to interventions Page 10 43 (masking) (eg, trial participants, care providers, outcome 44 assessors, data analysts), and how 45 on October 1, 2021 by guest. Protected copyright. 46 17b If blinded, circumstances under which unblinding is Not 47 48 permissible, and procedure for revealing a Applicable 49 participant’s allocated intervention during the trial 50 51 Methods: Data collection, management, and analysis 52 53 54 55 56 57 58 59 60

3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 31 BMJ Open

1 2 Data 18a Plans for assessment and collection of outcome, Page 14-15 3 collection baseline, and other trial data, including any related

4 methods processes to promote data quality (eg, duplicate BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 measurements, training of assessors) and a 7 description of study instruments (eg, questionnaires, 8 laboratory tests) along with their reliability and validity, 9 if known. Reference to where data collection forms 10 11 can be found, if not in the protocol 12 13 18b Plans to promote participant retention and complete Page 9 14 follow-up, including list of any outcome data to be 15 collected for participants who discontinue or deviate 16 from intervention protocols 17 18 Data 19 ForPlans forpeer data entry, review coding, security, only and storage, Page 14-15 19 20 management including any related processes to promote data 21 quality (eg, double data entry; range checks for data 22 values). Reference to where details of data 23 management procedures can be found, if not in the 24 25 protocol 26 27 Statistical 20a Statistical methods for analysing primary and Page 15 28 methods secondary outcomes. Reference to where other 29 details of the statistical analysis plan can be found, if 30 not in the protocol 31 32 20b Methods for any additional analyses (eg, subgroup Page 15 33 34 and adjusted analyses) 35 36 20c Definition of analysis population relating to protocol Page 15 37 non-adherence (eg, as randomised analysis), and any http://bmjopen.bmj.com/ 38 statistical methods to handle missing data (eg, 39 multiple imputation) 40 41 Methods: Monitoring 42 43 44 Data 21a Composition of data monitoring committee (DMC); Page 15 45 monitoring summary of its role and reporting structure; statement on October 1, 2021 by guest. Protected copyright. 46 of whether it is independent from the sponsor and 47 competing interests; and reference to where further 48 49 details about its charter can be found, if not in the 50 protocol. Alternatively, an explanation of why a DMC 51 is not needed 52 53 21b Description of any interim analyses and stopping Page 15 54 guidelines, including who will have access to these 55 56 interim results and make the final decision to 57 terminate the trial 58 59 60

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 31

1 2 Harms 22 Plans for collecting, assessing, reporting, and Page 14 3 managing solicited and spontaneously reported

4 adverse events and other unintended effects of trial BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 interventions or trial conduct 7 8 Auditing 23 Frequency and procedures for auditing trial conduct, if Page 15 9 any, and whether the process will be independent 10 from investigators and the sponsor 11 12 13 Ethics and dissemination 14 Research 24 Plans for seeking research ethics Page 16 15 16 ethics committee/institutional review board (REC/IRB) 17 approval approval 18 For peer review only 19 Protocol 25 Plans for communicating important protocol Page 16 20 amendments modifications (eg, changes to eligibility criteria, 21 outcomes, analyses) to relevant parties (eg, 22 23 investigators, REC/IRBs, trial participants, trial 24 registries, journals, regulators) 25 26 Consent or 26a Who will obtain informed consent or assent from Page 16 27 assent potential trial participants or authorised surrogates, 28 29 and how (see Item 32) 30 31 26b Additional consent provisions for collection and use of Not 32 participant data and biological specimens in ancillary Applicable 33 studies, if applicable 34 35 Confidentiality 27 How personal information about potential and enrolled Page 14--15 36 participants will be collected, shared, and maintained 37 http://bmjopen.bmj.com/ 38 in order to protect confidentiality before, during, and 39 after the trial 40 41 Declaration of 28 Financial and other competing interests for principal Page 18 42 interests investigators for the overall trial and each study site 43 44 Access to 29 Statement of who will have access to the final trial Page 16 45 data dataset, and disclosure of contractual agreements on October 1, 2021 by guest. Protected copyright. 46 47 that limit such access for investigators 48 49 Ancillary and 30 Provisions, if any, for ancillary and post-trial care, and Page 9 50 post-trial care for compensation to those who suffer harm from trial 51 participation 52 53 Dissemination 31a Plans for investigators and sponsor to communicate Page 16 54 policy trial results to participants, healthcare professionals, 55 56 the public, and other relevant groups (eg, via 57 publication, reporting in results databases, or other 58 data sharing arrangements), including any publication 59 restrictions 60

5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 31 BMJ Open

1 2 31b Authorship eligibility guidelines and any intended use None 3 of professional writers

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 31c Plans, if any, for granting public access to the full Page 16 6 protocol, participant-level dataset, and statistical code 7 8 9 Appendices 10 11 Informed 32 Model consent form and other related documentation None 12 consent given to participants and authorised surrogates 13 materials 14 15 Biological 33 Plans for collection, laboratory evaluation, and Not 16 specimens storage of biological specimens for genetic or Applicable 17 18 Formolecular peer analysis reviewin the current trial only and for future 19 use in ancillary studies, if applicable 20 21 *It is strongly recommended that this checklist be read in conjunction with the SPIRIT 2013 22 Explanation & Elaboration for important clarification on the items. Amendments to the 23 protocol should be tracked and dated. The SPIRIT checklist is copyrighted by the SPIRIT 24 Group under the Creative Commons “Attribution-NonCommercial-NoDerivs 3.0 Unported” 25 26 license. 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from

Using patient-reported outcomes to manage postoperative symptoms in patients with lung cancer: protocol for a multicentre, randomised controlled trial ForJournal: peerBMJ Open review only Manuscript ID bmjopen-2019-030041.R2

Article Type: Protocol

Date Submitted by the 18-Jul-2019 Author:

Complete List of Authors: Dai, Wei; Sichuan Cancer Hospital and Research Institute, Sichuan Cancer Center Zhang, Yuanqiang; Zigong First People's Hospital, Department of Cardiothoracic Surgery Feng, Wenhong ; Jiangyou People's Hospital, Department of Thoracic and Cardiovascular Surgery Liao, Xiaoqing ; Dazhu County People's Hospital, Department of Cardiothoracic Surgical Oncology Mu, Yunfei ; The Third People's Hospital of Chengdu, Department of Thoracic Surgery Zhang, Rui ; The Seventh People's Hospital of Chengdu, Department of Thoracic Surgery Wei, Xing ; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery http://bmjopen.bmj.com/ Wu, Chuanmei ; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery Xie, Shaohua; Sichuan Cancer Hospital and Chengdu Medical College, Department of Thoracic Surgery Li, Qiang; Sichuan Cancer Hospital and Research Institute, Department of Thoracic Surgery Shi, Qiuling; University of Texas MD Anderson Cancer Center

Primary Subject on October 1, 2021 by guest. Protected copyright. Patient-centred medicine Heading:

Secondary Subject Heading: Surgery, Oncology, Nursing

lung cancer, patient reported outcomes, postoperative symptom Keywords: management, randomised controlled trial

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 1 of 32 BMJ Open

1 2 3 1 Title page 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 2 Title: Using patient-reported outcomes to manage postoperative symptoms in patients 7 8 3 with lung cancer: protocol for a multicentre, randomised controlled trial 9 10 11 4 Authors: Wei Dai,1# Yuanqiang Zhang,2# Wenhong Feng,3 Xiaoqing Liao,4 Yunfei 12 13 5 Mu,5 Rui Zhang,6 Xing Wei,1 Chuanmei Wu,1 Shaohua Xie,1,7 Qiang Li,1 Qiuling Shi8 14 15 6 Institutions: 16 17 1 18 7 Department of ForThoracic peerSurgery, Sichuan review Cancer Hospital only & Institute, Sichuan Cancer 19 8 Center, School of Medicine, University of Electronic Science and Technology of China, 20 21 9 Chengdu, Sichuan, China. 22 23 10 2Department of Cardiothoracic Surgery, Zigong First People's Hospital, Zigong, 24 25 11 Sichuan, China. 26 12 3Department of Thoracic and Cardiovascular Surgery, Jiangyou People's Hospital, 27 28 13 Jiangyou, Sichuan, China. 29 4 30 14 Department of Cardiothoracic Surgical Oncology, Dazhu County People's Hospital, 31 15 Dazhu County, Dazhou, Sichuan, China. 32 33 16 5Department of Thoracic Surgery, The Third People's Hospital of Chengdu, Chengdu, 34 35 17 Sichuan, China. 36 6 37 18 Department of Thoracic Surgery, The Seventh People's Hospital of Chengdu, Chengdu, http://bmjopen.bmj.com/ 38 19 Sichuan, China. 39 40 20 7Graduate School, Chengdu Medical College, Chengdu, Sichuan, China. 41 42 21 8Department of Symptom Research, The University of Texas MD Anderson Cancer 43 44 22 Center, Houston, Texas, USA. 45 23 on October 1, 2021 by guest. Protected copyright. 46 47 24 Correspondence to 48 49 25 Dr Qiuling Shi, Department of Symptom Research, The University of Texas MD 50 26 Anderson Cancer Center, 1515 Holcombe Blvd., Unit 1450 Houston, TX 77030, USA. 51 52 27 Telephone: 713/745-3504, e-mail: [email protected]. 53 54 28 Dr Qiang Li, Department of Thoracic Surgery, Sichuan Cancer Hospital & Institute, 55 56 29 Sichuan Cancer Center, School of Medicine, University of Electronic Science and 57 30 Technology of China, No. 55, Section 4, South Renmin Road, Chengdu 610041, 58 59 1 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 2 of 32

1 2 3 31 Sichuan, China. Telephone: +86-028-85420229, e-mail: [email protected].

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 32 6 # 7 33 These authors contributed equally to this work. 8 34 9 10 11 35 Word count: 3699 words 12 36 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 2 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 3 of 32 BMJ Open

1 2 3 37 ABSTRACT 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 38 Introduction Surgery is one of the primary treatments for lung cancer. The 7 8 39 postoperative symptom burden experienced by patients with lung cancer is substantial, 9 10 40 seriously delaying their recovery from surgery and impairing their quality of life. 11 12 41 Patient-reported outcome (PRO)-based symptom management is increasingly regarded 13 14 42 as an optimal model for patient-centred care. Currently, clinical trial-based evidence 15 16 43 involving early-phase (immediately after surgery for up to one month) symptom 17 18 44 management of Forlung cancer peer is lacking. review We propose a randomisedonly trial to evaluate the 19 20 45 effect of a PRO-based symptom-monitoring program with over-threshold alerts and 21 22 46 responses for postoperative recovery in patients with lung cancer. 23 24 25 47 Methods and analysis The study will recruit 160 patients with lung cancer from six 26 hospitals. The patients will be randomly allocated to the intervention group or control 27 48 28 49 group in a ratio of 1:1. Patients in the intervention group will receive PRO-based 29 30 50 symptom management from the specialists when their reported target symptom (pain, 31 32 51 coughing, fatigue, disturbed sleep, and shortness of breath) scores reach the pre-set 33 34 52 threshold (score ≥ 4). Patients in the control group will not generate alerts and will 35 36 53 follow the standard procedures for symptom management. All patients will receive 37 http://bmjopen.bmj.com/ 38 54 symptom assessments via the MD Anderson Symptom Inventory – lung cancer module 39 40 55 on the day before surgery, daily after surgery, and twice a week after discharge until 4 41 42 56 weeks or the start of postoperative oncologic treatment. The primary outcome—mean 43 44 57 symptom threshold events—will be compared between the intervention and control 45 on October 1, 2021 by guest. Protected copyright. 46 58 group via independent sample Student’s t-test. 47 48 49 59 Ethics and dissemination The study was approved by the Ethics Committee of 50 51 60 Sichuan Cancer Hospital on November 22, 2018 (No. SCCHEC-02-2018-045). This 52 manuscript is based on Version 2.0, May 9, 2019 of the protocol. The study results will 53 61 54 62 be disseminated in publications in peer-reviewed journals and presentations at 55 56 63 academic conferences. 57 58 59 3 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 4 of 32

1 2 3 64 Trials registration number ChiCTR1900020846. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 65 7 8 9 66 Keywords: lung cancer, patient reported outcomes, postoperative symptom 10 11 67 management, randomised controlled trial. 12 13 14 68 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 4 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 5 of 32 BMJ Open

1 2 3 69 ARTICLE SUMMARY 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 70 Strengths and limitations of this study 7 8 9 71 1. This is an interventional study, comparing PRO-based postoperative symptom 10 11 72 management with standard postoperative symptom management in patients with lung 12 13 73 cancer. 14 15 74 2. It is a multicentre, randomised controlled trial, conducted in six tertiary hospitals in 16 17 75 China. 18 For peer review only 19 20 76 3. It focuses on the early postoperative period with a high frequency of data collection, 21 22 77 including baseline before surgery, daily in-hospital after surgery, and twice a week after 23 24 78 discharge until 4 weeks or the start of postoperative oncologic treatment. 25 26 27 79 4. We use a lung cancer-specific scale and the recall period is 24 hours, which is more 28 suitable to measure rapidly changing symptoms during the early recovery phase. 29 80 30 31 81 5. The lack of blinding for the participants and specialists delivering the intervention 32 33 82 may be a limitation. 34 35 36 83 37 http://bmjopen.bmj.com/ 38 84 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 5 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 6 of 32

1 2 3 85 INTRODUCTION 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 86 Lung cancer is the most common cancer and the leading cause of cancer death in China 7 8 87 and worldwide.1 2 With the application of low-dose computed tomography in screening, 9 10 88 more and more patients with early stage lung cancer are being diagnosed and treated 11 12 89 with surgery. However, the postoperative symptom burden of lung cancer patients is 13 14 90 very severe, and this detrimentally affects their quality of life (QOL). Patients have 15 16 91 various symptoms, such as pain, coughing, fatigue, and shortness of breath, in the early 17 18 92 stages after surgeryFor or even peer a long time review after surgery.3-5 onlyLowery et al followed 183 lung 19 20 93 cancer patients for 1-6 years and found that 79.8% of them had a variety of symptoms.3 21 22 94 Among these patients, 30.6% had one symptom, 27.9% had two symptoms, and 21.3% 23 24 95 had three or more symptoms. The most frequent symptoms were pain and shortness of 25 26 96 breath. If these symptoms are not effectively controlled, the postoperative recovery of 27 3 4 28 97 patients will be severely affected, resulting in a poor QOL. Therefore, effective 29 30 98 interventions are needed to alleviate post-surgery symptoms in patients with lung 31 3 4 32 99 cancer. 33 34 100 Symptom management is the foundation of clinical care, particularly for patients 35 36 101 with cancer. Patient-reported outcome (PRO)-based symptom management is 37 http://bmjopen.bmj.com/ 38 102 increasingly regarded as an optimal model for patient-centred care.6 7 A PRO is a 39 40 103 measurement of a patient's health status that comes directly from the patient's subjective 41 42 104 evaluation, with no interpretation by medical providers or anyone else.8 Studies have 43 44 105 shown that it may be more accurate for patients to evaluate their own health status 45 on October 1, 2021 by guest. Protected copyright. 46 106 themselves than evaluation by medical providers.9 The application of PRO-based 47 48 107 symptom monitoring and alerting followed by real-time symptom management from 49 50 108 health care professionals can improve the QOL of patients, prolong survival, increase 51 52 109 patient satisfaction, and allow evaluation of the treatment method.6 10-15 Basch et al 53 54 110 reported a randomised controlled trial (RCT) result, suggesting that PRO-based 55 56 111 proactive symptom monitoring could improve symptom management and thus bring 57 10 11 58 112 survival benefits in patients undergoing chemotherapy. When compared to a 59 6 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 7 of 32 BMJ Open

1 2 3 113 traditional reactive monitoring group, median survival was 5.2 months longer among 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 10 6 114 patients in the proactive monitoring group (31.2 vs. 26.0 months, p = 0.03). However, 7 8 115 it is still not clear if adequate symptom control and improved QOL in the surgical 9 10 116 population can ensure a potential better survival. 11 12 117 Currently, PROs are mainly used in non-surgical treatment settings,10-13 and they are 13 14 118 still in the early stage of application in the surgical treatment setting.5 7 14-28 Studies on 15 16 119 postoperative symptom management of lung cancer, especially in the early 17 18 120 postoperative periodFor are lacking.peer In addition, review most of the only published literature has a low 19 20 121 level of evidence, due to its design.5 18-28 The limitations of these studies include: (1) 21 22 122 most were observational studies; (2) they had small sample sizes ranging from 30 to 23 24 123 200 subjects with few exceptions; (3) they did not focus on the early postoperative 25 26 124 period, typically including the in-hospital period immediately after surgery and 4 weeks 27 28 125 after discharge when patients frequently report multiple severe symptoms, leading to 29 30 126 later negative recovery events, i.e. higher symptom burden, delayed return to intended 31 32 127 oncologic therapy, and poorer QOL; (4) they used a variety of survey instruments and 33 34 128 some were not a lung cancer-specific scale; (5) most of the scales used such as the 35 36 129 European Organization for Research and Treatment of Cancer Quality of Life 37 http://bmjopen.bmj.com/ 38 130 Questionnaire–Lung Cancer Module (EORTC QLQ-LC13), had a recall period of 1 39 131 week, which may not be able to identify the rapidly changing symptoms during early 40 41 132 postoperative phase;5 (6) only one study assessed in-hospital patients immediately after 42 43 133 surgery, but the MD Anderson Symptom Inventory (MDASI) scale used did not include 44 45 5 134 lung cancer-specific symptoms, i.e. coughing; (7) the symptom assessments were on October 1, 2021 by guest. Protected copyright. 46 47 135 inadequate, mostly at just two or three time points post-surgery; (8) most of the surgical 48 49 136 approaches were thoracotomies, not representing the current mainstream minimally 50 51 137 invasive thoracoscopic surgery for lung cancer; (9) there were very few studies on the 52 53 138 Chinese population. 54 55 56 139 We have been conducting an observational study of perioperative symptom in 57 58 140 patients with lung cancer based on PRO (registration number NCT03341377). Now, 59 7 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 8 of 32

1 2 3 141 we propose an RCT, aiming to evaluate the efficacy of a PRO-based symptom 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 142 monitoring, alerting, and response system (SMARS) to improve postoperative recovery 7 8 143 of lung cancer patients. The SMARS (figure 1) includes a research electronic data 9 10 144 capture (REDCap) platform, an electronic PRO system (ePRO Hub), and a most 11 popular social software (WeChat)29 in China. This study will provide evidence for early 12 145 13 146 postoperative phase symptom management for patients with lung cancer. We will use 14 15 147 the MDASI lung cancer-specific scale (MDASI-LC)30 to frequently monitor symptoms 16 17 148 and their impact on the functioning of lung cancer patients from pre-operation to 4 18 For peer review only 19 149 weeks after discharge or until the beginning of postoperative oncologic treatment. The 20 21 150 recall period of MDASI-LC is 24 hours, which is more suitable to measure rapidly 22 23 151 changing, early postoperative period symptoms compared with other QOL scales for 1 24 25 152 week or longer. Our research hypothesis is that patients with lung cancer undergoing 26 27 153 PRO-based symptom management have a lower postoperative symptom burden than 28 29 154 patients undergoing standard symptom management. 30 31 32 155 33 34 156 METHODS AND ANALYSIS 35 36 37 157 Study design http://bmjopen.bmj.com/ 38 39 40 158 The trial is a multicentre, randomised, parallel group controlled, and superiority design. 41 159 This protocol will be consistent with the Standard Protocol Items: Recommendations 42 43 160 For Interventional Trials (SPIRIT).31 The results of this trial will be reported according 44 45 32 161 to the guidelines of Consolidated Standards of Reporting Trials (CONSORT). A flow on October 1, 2021 by guest. Protected copyright. 46 47 162 chart of this trial is shown in figure 2. 48 49 50 163 51 52 53 164 Setting 54 55 165 Participants will be recruited from six tertiary hospitals in different cities in China. The 56 57 166 six hospitals are Sichuan Cancer Hospital, Zigong First People’s Hospital, Jiangyou 58 59 8 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 9 of 32 BMJ Open

1 2 3 167 People’s Hospital, Dazhu County People’s Hospital, The Third People’s Hospital of 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 168 Chengdu, and The Seventh People’s Hospital of Chengdu. The total number of lung 7 8 169 cancer operations in six hospitals is approximately 2,000 per year. 9 10 170 11 12 13 171 Participant recruitment 14 15 172 Participant recruitment will be carried out before the surgery by participating clinicians. 16 17 173 Eligible patients should meet all the inclusion criteria and not meet any of the exclusion 18 For peer review only 19 174 criteria. The inclusion criteria for the participants are: (1) aged 18 to 75 years, (2) 20 21 175 clinically diagnosed as primary lung cancer, (3) clinical stage I-IIIA (8th edition),33 and 22 23 176 (4) planning to receive surgery, and (5) able and willing to respond to a repeated 24 25 177 electronic questionnaire (e-questionnaire) on a smartphone or a tablet. The exclusion 26 27 178 criteria are: (1) history of neoadjuvant therapy, (2) having other malignant tumours, and 28 29 179 (3) unable to understand the study requirements. 30 31 32 180 Strategies for achieving adequate participant enrolment to reach a target sample size 33 34 181 include inviting more doctors in each centre to participate in the study and adding more 35 36 182 research centres. Plans to promote participant retention and complete follow-up include 37 http://bmjopen.bmj.com/ 38 183 education, refill reminders, and commitments to provide all the patients with free long- 39 40 184 term medical consultations after the trial via WeChat. In China, the first follow-up clinic 41 185 visit of surgical lung cancer patient is approximately 4 weeks after discharge. There is 42 43 186 no usual follow-up within these 4 weeks. In addition, usual care does not include free 44 45 187 medical consultations after discharge. Patients usually have to pay for follow-up care. on October 1, 2021 by guest. Protected copyright. 46 47 188 Free long-term medical consultation is an incentive for patients who participate in the 48 49 189 study, which may improve compliance. This incentive will do more good than harm to 50 51 190 the patients, so it is approved and recommended by the Ethics Committee of Sichuan 52 53 191 Cancer Hospital. The anticipated dates of the study are from December 1, 2018 to 54 55 192 December 31, 2020. We haven't started recruiting patients yet. 56 57 58 193 59 9 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 10 of 32

1 2 3 194 Sample size calculation 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 195 The primary end point of this study is the mean symptom threshold events, defined as 7 8 196 the average number of target symptom threshold events per patient, at each time point. 9 10 197 To meet the minimal clinically important difference (0.5 standard deviation)34 for the 11 12 198 mean symptom threshold events, the required sample size is 64 for each group, when 13 14 199 rejecting the null hypothesis (the difference between the two groups < 0.5 standard 15 16 200 deviation). A total of 128 cases with valid data are needed. Considering a 20% attrition 17 18 201 rate, we will needFor 80 patients peer for each review group (64/0.8). onlyThe 20% attrition rate is based 19 20 202 on our ongoing observational research (NCT03341377). The current withdrawal rate is 21 22 203 about 17% in the observational research. The rate of loss to follow-up in this trial is 23 24 204 estimated to be less than 3%, because this trial is an interventional study and the follow- 25 26 205 up time is very short (less than 4 weeks). In this trial, the 20% attrition include both 27 28 206 withdrawal and loss to follow-up. The sample size calculation is based on the 29 30 207 independent sample Student’s t-test, using a two-tailed alpha level of 0.05 and a beta 31 32 208 error probability of 0.02 (80% power). 33 34 209 35 36 37 210 Randomisation and allocation concealment http://bmjopen.bmj.com/ 38 39 40 211 The process of randomisation will be carried out online using the central randomisation 41 212 module on the REDCap platform (http://125.71.214.100:888/redcap) after a participant 42 43 213 has been recruited to the study and has signed an informed consent form. The data 44 45 214 manager will upload the randomisation allocation table to the REDCap platform and on October 1, 2021 by guest. Protected copyright. 46 47 215 then save the randomisation allocation table independently. The investigator will 48 49 216 conduct randomisation by clicking on a randomisation button on the REDCap platform. 50 51 217 It will then be allocated to the intervention group or control group with a 1:1 ratio. Each 52 53 218 group will have 80 cases. 54 55 56 219 57 58 59 10 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 11 of 32 BMJ Open

1 2 3 220 Blinding 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 221 The blinding of participants and specialists delivering the intervention is impossible 7 8 222 due to the nature of the interventions. But the data collectors who help administer PRO 9 10 223 collection will be blinded to group allocation to minimise measurement bias. The 11 12 224 statisticians analysing the results will also be blinded to group allocation. 13 14 15 225 16 17 226 Intervention 18 For peer review only 19 20 227 After enrolment, all the patients will use their WeChat app to connect with the 21 22 228 participating specialists’ WeChat app via a mini program (ePRO Cell). Then, they will 23 24 229 be taught how to use the program. The ePRO questionnaires will be set to send to the 25 26 230 patients’ WeChat app automatically after randomisation. Patients are required to 27 28 231 complete the ePRO questionnaires on their smartphones or tablets before surgery 29 30 232 (baseline, typically 1 to 3 days before the operation), daily after surgery (in-hospital, 31 32 233 typically 1 to 7 days after the operation), and twice a week after discharge until 4 weeks 33 34 234 or the start of postoperative oncologic treatment (typically collecting PRO data six to 35 36 235 eight times after discharge). In a hospital setting, if the patients do not complete the 37 http://bmjopen.bmj.com/ 38 236 ePRO questionnaires within the scheduled time, an electronic reminder (e-reminder) 39 40 237 and up to two bedside reminders will be delivered at the same day. After discharge, if 41 238 the patients fail to complete the ePRO questionnaires within the scheduled time, an e- 42 43 239 reminder and up to two phone reminders will be delivered with 24 hours. 44 45 on October 1, 2021 by guest. Protected copyright. 46 240 47 48 49 241 Comparison 50 51 242 Intervention group 52 53 54 243 Patients will not be informed about the threshold levels. When there are one or more 55 56 244 target symptoms (pain, coughing, fatigue, disturbed sleep, and shortness of breath) and 57 58 245 scores reach the pre-set intervention threshold (score ≥ 4), the participating specialist 59 11 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 12 of 32

1 2 3 246 (thoracic surgeon) will simultaneously receive an alert message on his or her WeChat. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 247 The specialist will manage the patients’ symptoms based on the scores of the PRO. 7 8 248 After discharge, the specialist will mainly use the WeChat or sometimes a telephone to 9 10 249 contact the patient within 24 hours to implement symptom relief measures, e.g. 11 consultation, education, medication guidance, and clinic or hospital visit suggestions. 12 250 13 251 The symptom relief measures of the intervention group patients will comply with the 14 15 252 latest guidelines and be standardised across all centres, in the form of a standard 16 17 253 operating procedure (SOP) handbook. Patients’ adherence to the interventions will be 18 For peer review only 19 254 asked at each time point. Those who do not follow the specialist’s advice will be 20 21 255 monitored, and the number of violations will be recorded. Those who refuse to follow 22 23 256 the specialist’s advice more than three times will be considered as seriously violating 24 25 257 the study protocol and will be withdrawn. Patients will be educated and allowed to seek 26 27 258 medical help through usual channels for severe symptoms. 28 29 30 259 31 32 260 Control group 33 34 35 261 The control group patients will be informed that the ePRO data collected are only for 36 37 262 scientific research. They will not generate any alerts or get responses relating to their http://bmjopen.bmj.com/ 38 39 263 symptoms. The patients’ symptom management will follow the current standard 40 41 264 postoperative management model. During hospitalization, the doctors manage the 42 43 265 control group patients’ symptoms based on their own judgement rather than the scores 44 45 266 of the PRO. After discharge, the patients will go home and the first clinic visit is on October 1, 2021 by guest. Protected copyright. 46 47 267 approximately 4 weeks later. Patients will be encouraged to seek medical help if severe 48 49 268 symptoms are reported. 50 51 269 52 53 54 270 Withdrawal criteria 55 56 271 Participants will be withdrawn from the study and no further data will be collected if 57 58 59 12 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 13 of 32 BMJ Open

1 2 3 272 they meet the following criteria: (1) unexpected cancellation of surgery, (2) severe 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 273 postoperative complications (≥grade Шb according to the Clavien‑Dindo classification 7 8 274 of surgical complications) affecting symptom data collection, (3) postoperative length 9 10 275 of stay > 14 days (because patient with a postoperative hospital stay > 14 days usually 11 12 276 has a severe complication, and the patient compliance will gradually decrease, affecting 13 14 277 the accuracy of PRO data), (4) postoperative pathology shows non-primary lung cancer, 15 16 278 (5) non-R0 resection, (6) pathological stage IV, (7) participant seriously violates the 17 18 279 study protocol For(continually peer not complying review with the specialist’sonly advice, intentionally 19 20 280 letting a proxy to complete the PRO surveys, and deliberately providing false PROs), 21 22 281 or (8) participant asks to withdraw from the study. 23 24 25 282 26 27 283 Outcomes and measurement 28 29 30 284 Primary outcome 31 32 285 The primary end point of this study is the mean symptom threshold events. According 33 34 286 to our pilot study, the five most common postoperative symptoms of lung cancer 35 36 287 patients are: pain, coughing, fatigue, disturbed sleep and shortness of breath. In this 37 http://bmjopen.bmj.com/ 38 288 study, these five symptoms assessed by the MDASI-LC are defined as target symptoms. 39 40 289 According to the recommendation of National Comprehensive Cancer Network and 41 42 290 published literature, when a patient’s symptom score is ≥ 4, it is identified as moderate 43 44 291 severity.35 36 In this study, a score of 4 is set as the threshold value for intervention, and 45 on October 1, 2021 by guest. Protected copyright. 46 292 a target symptom score of ≥ 4 is reported as a threshold event. 47 48 30 49 293 The primary PRO tool used in this study is the MDASI-LC. It is a measure that 50 51 294 contains sixteen items of lung cancer-related and treatment-related symptoms, and six 52 53 295 items of interference to normal daily life caused by symptoms. All items are rated on 54 0-10 numerical scales, with 0 representing “symptom not present” or “symptom not 55 296 56 297 interfered with life” and 10 representing “symptom as bad as one can imagine” or 57 58 59 13 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 14 of 32

1 2 3 298 “symptom completely interfered with life”. The recall period of the MDASI-LC is 24 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 299 hours and it can usually be conducted in 5 minutes. It has been translated and validated 7 8 300 for application in a Chinese context. 9 10 301 11 12 13 302 Secondary outcomes 14 15 303 The secondary end points of this study include trajectories of PROs (symptom severity, 16 17 304 daily functioning, and QOL) and revisit rate after discharge. Trajectories of PROs are 18 For peer review only 19 305 defined as the longitudinal changing pattern of the mean score of the five target 20 21 306 symptoms for symptom severity, the mean score of the six MDASI-LC interference 22 23 307 items for daily functioning, and the mean score of the single-item QOL scale 24 25 308 (UNISCALE) for QOL,37 from the baseline to 4 weeks after discharge or until the start 26 27 309 of postoperative oncologic treatment. UNISCALE has only one question using a 0-10 28 29 310 scale, with 0 representing “worst QOL” and 10 representing “best QOL”. The revisit 30 31 311 rate after discharge is defined as the ratio of the number of patients who see the doctor 32 33 312 again after discharge including outpatient visits, emergency visits and hospitalisation 34 35 313 divided by the total number of patients. 36 37 http://bmjopen.bmj.com/ 38 314 39 40 315 Other data 41 42 43 316 The clinician workload, clinician system acceptability, and patient satisfaction of the 44 45 317 interventions will be assessed through surveys and interviews. Demographics, on October 1, 2021 by guest. Protected copyright. 46 47 318 clinicopathological characteristics, follow-up information, and adverse events of the 48 49 319 interventions will also be collected. All the adverse events will be assessed and 50 51 320 managed by a thoracic surgeon. 52 53 321 54 55 56 322 Data collection, management, and quality control 57 58 59 14 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 15 of 32 BMJ Open

1 2 3 323 REDCap,38 39 a worldwide popular research data collection and management platform 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 324 established in Sichuan Cancer Hospital, will be used for data collection and 7 8 325 management in this study. PRO data will be collected using e-questionnaires and 9 10 326 recorded in REDCap. Participants should fill out the e-questionnaires by themselves. If 11 participants have difficulties in completing the e-questionnaires, data collectors or their 12 327 13 328 family members will help them by just reading each item aloud and recording the 14 15 329 participant’s responses. The control group patients’ PRO data will not be accessed by 16 17 330 the specialists. Specialists can only access the PRO data of the intervention group 18 For peer review only 19 331 patients. Other data including demographics, clinicopathological characteristics, and 20 21 332 follow-up information will also be entered into the REDCap database. 22 23 24 333 Data will be checked regularly by the quality controller. Participant privacy 25 26 334 information will not be recorded in REDCap. A study number will be allocated to each 27 28 335 participant and will be used on all study documentation, which will only be available 29 30 336 to the investigators. Before patients’ enrolment, investigators from each research centre 31 32 337 will receive SOP training. Each centre will receive on-site monitoring visits, telephone 33 34 338 monitoring, and online guidance during the course of the trial. 35 36 339 37 http://bmjopen.bmj.com/ 38 39 340 Data analysis 40 41 341 Per-protocol analyses will be conducted. To be included in the analysis, a participant 42 43 342 must provide MDASI-LC data from the baseline and at least two additional time points. 44 45 343 If a participant meets the withdrawal criteria, no data will be included in the analysis. on October 1, 2021 by guest. Protected copyright. 46 47 344 Two-sided P values of < 0.05 are considered to be statistically significant. 48 49 345 Continuous variables will be presented as mean ± standard deviation or median and 50 51 346 interquartile range. Comparisons between groups will be conducted using the Student’s 52 53 347 t-test or the Wilcoxon rank sum test. Categorical variables will be presented as 54 55 348 frequencies or proportions and compared between groups using the chi-square test. 56 57 349 Trajectories of PROs will be compared between the intervention group and control 58 59 15 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 16 of 32

1 2 3 350 group using generalised mixed effects models. Missing data will be processed by the 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 351 multiple imputation method. Results obtained from data without missed observations 7 8 352 will be compared with that from imputed data for sensitive analysis. 9 10 353 11 12 13 354 Data monitoring and interim analysis 14 15 355 A data monitoring committee (DSM) consisting of one clinician, one statistician, and 16 17 356 the secretary of the Ethics Committee of Sichuan Cancer Hospital will be set up. Study 18 For peer review only 19 357 monitoring will be carried out regularly by DSM members and the process will be 20 21 358 independent from investigators. Due to the low-risk of the study content and short-term 22 23 359 study duration, interim analysis will not be performed. 24 25 26 360 27 28 Data availability statement 29 361 30 31 362 Deidentified data generated by this clinical trial to support future research articles will 32 33 363 be available from the corresponding author on reasonable request. 34 35 36 364 37 http://bmjopen.bmj.com/ 38 365 Patient and public involvement statement 39 40 41 366 Patients and the public will not be involved in the design, recruitment to, or conduct of 42 43 367 this study. We will inform the applicants of the results. There are no plans to 44 45 368 disseminate the results to study participants, because it is not a routine practice to feed on October 1, 2021 by guest. Protected copyright. 46 47 369 back research results to participants in China. Participants will be informed that they 48 49 370 can obtain the final results of this study through our future published articles. 50 51 371 52 53 54 372 ETHICS AND DISSEMINATION 55 56 57 373 This study was approved by the Ethics Committee of Sichuan Cancer Hospital on 58 59 16 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 17 of 32 BMJ Open

1 2 3 374 November 22, 2018 (No. SCCHEC-02-2018-045). All recruited patients will be 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 375 required to give written informed consent. Any subsequent amendments to the protocol 7 8 376 will be submitted for further review and approval. Sub-centres will gain approval from 9 10 377 their hospital-specific ethics committees. The results of this study will be disseminated 11 through peer-reviewed publications and academic conferences. 12 378 13 14 379 15 16 17 380 DISCUSSION 18 For peer review only 19 381 This trial focuses on the early-phase postoperative symptom management after lung 20 21 382 cancer surgery. The potential implications of the findings include: (1) identifying if 22 23 383 PRO-based symptom management is better than usual symptom management, (2) 24 25 384 identifying if proactive symptom management can reduce symptom burden and 26 27 385 improve QOL in the surgical population, (3) laying a foundation for future research on 28 29 386 whether postoperative symptom management improves survival, (4) investigating 30 31 387 whether SMARS is feasible and acceptable in real-world clinical practice in China, and 32 33 388 (5) identifying barriers which will be used to facilitate further revisions of the SMARS 34 35 389 and help extend its implementation in non-surgical settings. 36 37 http://bmjopen.bmj.com/ 38 390 There are many limitations in this trial. First, the trial will be carried out in well- 39 40 391 resourced tertiary hospitals in China. This will limit the generalisability to non-tertiary 41 392 hospitals. Second, the inclusion criteria and exclusion criteria are strict. For example, 42 43 393 the program is unsuitable for patients without internet access or with poor literacy. This 44 45 394 will greatly limit the population for which this study is applicable. Third, the withdrawal on October 1, 2021 by guest. Protected copyright. 46 47 395 criteria will create selection bias and limit the external validity, although the strict 48 49 396 criteria will ensure the compliance of this study. In the future, we will conduct 50 51 397 pragmatic clinical trials (PCT) to evaluate the effectiveness of the monitoring system 52 53 398 in a more heterogeneous population, to improve the generalisability. Fourth, the lack of 54 55 399 blinding for the participants and specialists delivering the intervention will also be a 56 57 400 limitation, because it may increase the measurement bias. Fifth, it may affect the 58 59 17 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 18 of 32

1 2 3 401 establishment of feasibility if patients are not involved in the design and development 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 402 of this trial, although previous studies and our ongoing observational study have 7 8 403 provided pilot data for the design and development of this trial in terms of feasibility 9 10 404 and acceptability. This RCT is designed to test the efficacy of the PRO monitoring 11 system. We will evaluate the effectiveness in a future PCT, with patients’ involvement 12 405 13 406 in study design, conduct and interpretation. Sixth, the follow-up period is very short. 14 15 407 The results need confirmation in a study with a longer follow-up period. 16 17 18 408 In summary, For as a RCT, peer this study review will not only testonly the efficacy of SMARS in 19 20 409 postoperative care, but also it will provide data of feasibility for further unblinded 21 22 410 pragmatic study when implementing the SMARS in the real world, with the 23 24 411 involvement of community hospitals and patients with poor socioeconomic status, 25 26 412 while a wider internet access is available for the whole Chinese population. 27 28 413 29 30 31 414 Acknowledgements The authors thank all the patients and patient advisers who are involved in this 32 33 415 study. 34 35 36 416 Author contributions WD and QLS conceived and designed the study. WD and QLS obtained the 37 funding. WD is the chief investigator of this study. YQZ, WHF, XQL, YFM, and RZ are sub-centre http://bmjopen.bmj.com/ 38 417 39 418 principal investigators who contributed to the trial feasibility stage. WD, YQZ, WHF, XQL, YFM, 40 41 419 RZ, XW, CMW and SHX drafted the protocol. QLS participated in the statistical plan. QL and QLS 42 43 420 revised the manuscript. All authors have read and approved the manuscript. 44 45 on October 1, 2021 by guest. Protected copyright. 46 421 Funding This work was supported by [Bethune charitable foundation], [Sichuan Science and 47 48 422 Technology Program] grant number [18PJ436 and 2019YFH0070] and [National Natural Science 49 50 423 Foundation of China] grant number [81872506]. 51 52 Competing interests None declared. 53 424 54 55 425 Patient consent Obtained. 56 57 58 426 Ethics approval Ethics Committee of Sichuan Cancer Hospital (No. SCCHEC-02-2018-045). 59 18 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 19 of 32 BMJ Open

1 2 3 427 Provenance and peer review Not commissioned; externally peer reviewed. 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 428 7 8 9 10 11 12 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 19 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 20 of 32

1 2 3 429 REFERENCES 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 430 1. Chen W, Zheng R, Baade PD, et al. Cancer statistics in China, 2015. CA Cancer J Clin 7 8 431 2016;66:115-32. 9 10 432 2. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates 11 12 433 of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 13 14 434 2018;68:394-424. 15 16 435 3. Lowery AE, Krebs P, Coups EJ, et al. Impact of symptom burden in post-surgical non-small cell 17 18 436 lung cancer survivors.For Support peer Care Cancer review 2014;22:173-80. only 19 20 437 4. Yang P, Cheville AL, Wampfler JA, et al. Quality of life and symptom burden among long-term 21 22 438 lung cancer survivors. J Thorac Oncol 2012;7:64-70. 23 24 439 5. Fagundes CP, Shi Q, Vaporciyan AA, et al. Symptom recovery after thoracic surgery: Measuring 25 26 440 patient-reported outcomes with the MD Anderson Symptom Inventory. J Thorac Cardiovasc 27 28 441 Surg 2015;150:613-9 e2. 29 30 442 6. Basch E. Patient-Reported Outcomes - Harnessing Patients' Voices to Improve Clinical Care. N 31 32 443 Engl J Med 2017;376:105-8. 33 34 444 7. Khullar OV, Fernandez FG. Patient-Reported Outcomes in Thoracic Surgery. Thorac Surg Clin 35 2017;27:279-90. 36 445 37 446 8. U.S. Department of Health and Human Services FDA Center for Drug Evaluation and Research, http://bmjopen.bmj.com/ 38 39 447 U.S. Department of Health and Human Services FDA Center for Biologics Evaluation and 40 41 448 Research, U.S. Department of Health and Human Services FDA Center for Devices and 42 43 449 Radiological Health. Guidance for industry: patient-reported outcome measures: use in medical 44 45 450 product development to support labeling claims: draft guidance. Health Qual Life Outcomes on October 1, 2021 by guest. Protected copyright. 46 47 451 2006;4:79. 48 49 452 9. Basch E. The missing voice of patients in drug-safety reporting. N Engl J Med 2010;362:865-9. 50 51 453 10. Basch E, Deal AM, Dueck AC, et al. Overall Survival Results of a Trial Assessing Patient- 52 53 454 Reported Outcomes for Symptom Monitoring During Routine Cancer Treatment. JAMA 54 55 455 2017;318:197-8. 56 57 456 11. Basch E, Deal AM, Kris MG, et al. Symptom Monitoring With Patient-Reported Outcomes 58 59 20 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 21 of 32 BMJ Open

1 2 3 457 During Routine Cancer Treatment: A Randomized Controlled Trial. J Clin Oncol 2016;34:557- 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 458 65. 7 8 459 12. Novello S, Kaiser R, Mellemgaard A, et al. Analysis of patient-reported outcomes from the 9 LUME-Lung 1 trial: a randomised, double-blind, placebo-controlled, Phase III study of second- 10 460 11 461 line nintedanib in patients with advanced non-small cell lung cancer. Eur J Cancer 2015;51:317- 12 13 462 26. 14 15 463 13. Popat S. Patient reported outcomes from LUX-Lung 3: first-line afatinib is superior to 16 17 464 chemotherapy-would patients agree? Ann Palliat Med 2014;3:19-21. 18 For peer review only 19 465 14. Ng S, Pusic A, Parker E, et al. Patient-Reported Outcome Measures for Breast Implant Surgery: 20 21 466 A Pilot Study. Aesthet Surg J 2019;sjz023.doi:10.1093/asj/sjz023.[Epub ahead of print]. 22 23 467 15. Coronini-Cronberg S, Appleby J, Thompson J. Application of patient-reported outcome 24 25 468 measures (PROMs) data to estimate cost-effectiveness of hernia surgery in England. J R Soc 26 27 469 Med 2013;106:278-87. 28 29 470 16. Poghosyan H, Sheldon LK, Leveille SG, et al. Health-related quality of life after surgical 30 31 471 treatment in patients with non-small cell lung cancer: a systematic review. Lung Cancer 32 33 472 2013;81:11-26. 34 35 473 17. Cleeland CS, Wang XS, Shi Q, et al. Automated symptom alerts reduce postoperative symptom 36 37 474 severity after cancer surgery: a randomized controlled clinical trial. J Clin Oncol 2011;29:994- http://bmjopen.bmj.com/ 38 39 475 1000. 40 41 476 18. Khullar OV, Rajaei MH, Force SD, et al. Pilot Study to Integrate Patient Reported Outcomes 42 43 477 After Lung Cancer Operations Into The Society of Thoracic Surgeons Database. Ann Thorac 44 45 478 Surg 2017;104:245-53. on October 1, 2021 by guest. Protected copyright. 46 47 479 19. Shi Q, Wang XS, Vaporciyan AA, et al. Patient-Reported Symptom Interference as a Measure 48 of Postsurgery Functional Recovery in Lung Cancer. J Pain Symptom Manage 2016;52:822-31. 49 480 50 481 20. Li WW, Lee TW, Lam SS, et al. Quality of life following lung cancer resection: video-assisted 51 52 482 thoracic surgery vs thoracotomy. Chest 2002;122:584-9. 53 54 483 21. Kenny PM, King MT, Viney RC, et al. Quality of life and survival in the 2 years after surgery 55 56 484 for non small-cell lung cancer. J Clin Oncol 2008;26:233-41. 57 58 59 21 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 22 of 32

1 2 3 485 22. Balduyck B, Hendriks J, Lauwers P, et al. Quality of life after lung cancer surgery: a prospective 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 486 pilot study comparing bronchial sleeve lobectomy with pneumonectomy. J Thorac Oncol 7 8 487 2008;3:604-8. 9 23. Ferguson MK, Parma CM, Celauro AD, et al. Quality of life and mood in older patients after 10 488 11 489 major lung resection. Ann Thorac Surg 2009;87:1007-12; discussion 12-3. 12 13 490 24. Sartipy U. Prospective population-based study comparing quality of life after pneumonectomy 14 15 491 and lobectomy. Eur J Cardiothorac Surg 2009;36:1069-74. 16 17 492 25. Ostroff JS, Krebs P, Coups EJ, et al. Health-related quality of life among early-stage, non-small 18 For peer review only 19 493 cell, lung cancer survivors. Lung Cancer 2011;71:103-8. 20 21 494 26. Moller A, Sartipy U. Long-term health-related quality of life following surgery for lung cancer. 22 23 495 Eur J Cardiothorac Surg 2012;41:362-7. 24 25 496 27. Zhao J, Zhao Y, Qiu T, et al. Quality of life and survival after II stage nonsmall cell carcinoma 26 27 497 surgery: Video-assisted thoracic surgery versus thoracotomy lobectomy. Indian J Cancer 28 29 498 2015;52 Suppl 2:e130-3. 30 31 499 28. Yun YH, Kim YA, Sim JA, et al. Prognostic value of quality of life score in disease-free 32 33 500 survivors of surgically-treated lung cancer. BMC Cancer 2016;16:505. 34 35 501 29. Montag C, Becker B, Gan C. The Multipurpose Application WeChat: A Review on Recent 36 37 502 Research. Front Psychol 2018;9:2247. http://bmjopen.bmj.com/ 38 39 503 30. Mendoza TR, Wang XS, Lu C, et al. Measuring the symptom burden of lung cancer: the validity 40 41 504 and utility of the lung cancer module of the M. D. Anderson Symptom Inventory. Oncologist 42 43 505 2011;16:217-27. 44 45 506 31. Chan AW, Tetzlaff JM, Altman DG, et al. SPIRIT 2013 statement: defining standard protocol on October 1, 2021 by guest. Protected copyright. 46 47 507 items for clinical trials. Ann Intern Med 2013;158:200-7. 48 49 508 32. Schulz KF, Altman DG, Moher D. CONSORT 2010 statement: updated guidelines for reporting 50 51 509 parallel group randomized trials. Ann Intern Med 2010;152:726-32. 52 53 510 33. Goldstraw P, Chansky K, Crowley J, et al. The IASLC Lung Cancer Staging Project: Proposals 54 55 511 for Revision of the TNM Stage Groupings in the Forthcoming (Eighth) Edition of the TNM 56 57 512 Classification for Lung Cancer. J Thorac Oncol 2016;11:39-51. 58 59 22 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 23 of 32 BMJ Open

1 2 3 513 34. Revicki DA, Cella D, Hays RD, et al. Responsiveness and minimal important differences for 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 514 patient reported outcomes. Health & Quality of Life Outcomes 2006;4:70. 7 8 515 35. National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology: 9 Adult Cancer Pain. Version 1.2019. Available at: https://www.nccn.org/professionals/physician 10 516 11 517 _gls/pdf/pain.pdf. 12 13 518 36. Wang XS, Zhao F, Fisch MJ, et al. Prevalence and characteristics of moderate to severe fatigue: 14 15 519 a multicenter study in cancer patients and survivors. Cancer 2014;120:425-32. 16 17 520 37. Sloan JA, Loprinzi CL, Kuross SA, et al. Randomized comparison of four tools measuring 18 For peer review only 19 521 overall quality of life in patients with advanced cancer. J Clin Oncol 1998;16:3662-73. 20 21 522 38. Tomko RL, Gray KM, Oppenheimer SR, et al. Using REDCap for ambulatory assessment: 22 23 523 Implementation in a clinical trial for smoking cessation to augment in-person data collection. 24 25 524 Am J Drug Alcohol Abuse 2018:1-16. 26 27 525 39. Harvey LA. REDCap: web-based software for all types of data storage and collection. Spinal 28 29 526 Cord 2018;56:625. 30 31 32 527 33 34 528 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 23 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 24 of 32

1 2 3 529 Figure legends 4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 530 Figure 1 Schematic diagram of the SMARS. 7 8 9 531 Figure 2 Flow chart of this parallel group randomised trial. 10 11 12 532 13 14 15 16 17 18 For peer review only 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 24 60

For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 25 of 32 BMJ Open BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 Schematic diagram of the SMARS. 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 38 39 40

41 on October 1, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 26 of 32 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 For peer review only 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

33 http://bmjopen.bmj.com/ 34 35 36 37 Flow chart of this parallel group randomised trial. 38 39 40

41 on October 1, 2021 by guest. Protected copyright. 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 27 of 32 BMJ Open

1 2 3

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 7 8 SPIRIT 2013 Checklist: Recommended items to address in a clinical trial protocol and 9 related documents* 10 11 Section/item Item Description Page Number 12 No on which item 13 is reported 14 15 Administrative information 16 17 Title 1 Descriptive title identifying the study design, Page 1 18 Forpopulation, peer interventions, review and, if applicable, only trial 19 20 acronym 21 22 Trial 2a Trial identifier and registry name. If not yet registered, Page 4 23 registration name of intended registry 24 25 2b All items from the World Health Organization Trial Page 4 26 Registration Data Set 27 28 Protocol 3 Date and version identifier Page 3 29 version 30 31 Funding 4 Sources and types of financial, material, and other Page 17-18 32 33 support 34 35 Roles and 5a Names, affiliations, and roles of protocol contributors Page 1, 17 36 responsibilitie 37 5b Name and contact information for the trial sponsor Page 1, 17 s http://bmjopen.bmj.com/ 38 39 5c Role of study sponsor and funders, if any, in study Page 17-18 40 design; collection, management, analysis, and 41 42 interpretation of data; writing of the report; and the 43 decision to submit the report for publication, including 44 whether they will have ultimate authority over any of 45 these activities on October 1, 2021 by guest. Protected copyright. 46 47 5d Composition, roles, and responsibilities of the Page 15-16 48 49 coordinating centre, steering committee, endpoint 50 adjudication committee, data management team, and 51 other individuals or groups overseeing the trial, if 52 applicable (see Item 21a for data monitoring 53 54 committee) 55 56 Introduction 57 58 59 60

1 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 28 of 32

1 2 Background 6a Description of research question and justification for Page 6-7 3 and rationale undertaking the trial, including summary of relevant

4 studies (published and unpublished) examining BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 benefits and harms for each intervention 7 8 6b Explanation for choice of comparators Page 11-12 9 10 Objectives 7 Specific objectives or hypotheses Page 8 11 12 Trial design 8 Description of trial design including type of trial (eg, Page 8 13 parallel group, crossover, factorial, single group), 14 allocation ratio, and framework (eg, superiority, 15 equivalence, noninferiority, exploratory) 16 17 18 Methods: Participants,For interventions, peer andreview outcomes only 19 20 Study setting 9 Description of study settings (eg, community clinic, Page 8-9 21 academic hospital) and list of countries where data 22 will be collected. Reference to where list of study sites 23 24 can be obtained 25 26 Eligibility 10 Inclusion and exclusion criteria for participants. If Page 9 27 criteria applicable, eligibility criteria for study centres and 28 individuals who will perform the interventions (eg, 29 surgeons, psychotherapists) 30 31 Interventions 11a Interventions for each group with sufficient detail to Page 11-12 32 33 allow replication, including how and when they will be 34 administered 35 36 11b Criteria for discontinuing or modifying allocated Page 12-13 37 interventions for a given trial participant (eg, drug http://bmjopen.bmj.com/ 38 dose change in response to harms, participant 39 40 request, or improving/worsening disease) 41 42 11c Strategies to improve adherence to intervention Page 9 43 protocols, and any procedures for monitoring 44 adherence (eg, drug tablet return, laboratory tests) 45 on October 1, 2021 by guest. Protected copyright. 46 11d Relevant concomitant care and interventions that are Page 12-13 47 48 permitted or prohibited during the trial 49 50 Outcomes 12 Primary, secondary, and other outcomes, including Page 13-14 51 the specific measurement variable (eg, systolic blood 52 pressure), analysis metric (eg, change from baseline, 53 final value, time to event), method of aggregation (eg, 54 55 median, proportion), and time point for each outcome. 56 Explanation of the clinical relevance of chosen 57 efficacy and harm outcomes is strongly recommended 58 59 60

2 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 29 of 32 BMJ Open

1 2 Participant 13 Time schedule of enrolment, interventions (including Page 9-11 3 timeline any run-ins and washouts), assessments, and visits

4 for participants. A schematic diagram is highly BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 recommended (see Figure) 7 8 Sample size 14 Estimated number of participants needed to achieve Page 10 9 study objectives and how it was determined, including 10 clinical and statistical assumptions supporting any 11 sample size calculations 12 13 Recruitment 15 Strategies for achieving adequate participant Page 10 14 15 enrolment to reach target sample size 16 17 Methods: Assignment of interventions (for controlled trials) 18 For peer review only 19 Allocation: 20 21 Sequence 16a Method of generating the allocation sequence (eg, Page 10 22 generation computer-generated random numbers), and list of any 23 factors for stratification. To reduce predictability of a 24 random sequence, details of any planned restriction 25 26 (eg, blocking) should be provided in a separate 27 document that is unavailable to those who enrol 28 participants or assign interventions 29 30 Allocation 16b Mechanism of implementing the allocation sequence Page 10 31 concealme (eg, central telephone; sequentially numbered, 32 33 nt opaque, sealed envelopes), describing any steps to 34 mechanis conceal the sequence until interventions are assigned 35 m 36 37 Implement 16c Who will generate the allocation sequence, who will Page 10 http://bmjopen.bmj.com/ 38 39 ation enrol participants, and who will assign participants to 40 interventions 41 42 Blinding 17a Who will be blinded after assignment to interventions Page 10-11 43 (masking) (eg, trial participants, care providers, outcome 44 assessors, data analysts), and how 45 on October 1, 2021 by guest. Protected copyright. 46 17b If blinded, circumstances under which unblinding is Not 47 48 permissible, and procedure for revealing a Applicable 49 participant’s allocated intervention during the trial 50 51 Methods: Data collection, management, and analysis 52 53 54 55 56 57 58 59 60

3 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 30 of 32

1 2 Data 18a Plans for assessment and collection of outcome, Page 14-15 3 collection baseline, and other trial data, including any related

4 methods processes to promote data quality (eg, duplicate BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 measurements, training of assessors) and a 7 description of study instruments (eg, questionnaires, 8 laboratory tests) along with their reliability and validity, 9 if known. Reference to where data collection forms 10 11 can be found, if not in the protocol 12 13 18b Plans to promote participant retention and complete Page 9 14 follow-up, including list of any outcome data to be 15 collected for participants who discontinue or deviate 16 from intervention protocols 17 18 Data 19 ForPlans forpeer data entry, review coding, security, only and storage, Page 15-16 19 20 management including any related processes to promote data 21 quality (eg, double data entry; range checks for data 22 values). Reference to where details of data 23 management procedures can be found, if not in the 24 25 protocol 26 27 Statistical 20a Statistical methods for analysing primary and Page 15 28 methods secondary outcomes. Reference to where other 29 details of the statistical analysis plan can be found, if 30 not in the protocol 31 32 20b Methods for any additional analyses (eg, subgroup Page 15 33 34 and adjusted analyses) 35 36 20c Definition of analysis population relating to protocol Page 15 37 non-adherence (eg, as randomised analysis), and any http://bmjopen.bmj.com/ 38 statistical methods to handle missing data (eg, 39 multiple imputation) 40 41 Methods: Monitoring 42 43 44 Data 21a Composition of data monitoring committee (DMC); Page 16 45 monitoring summary of its role and reporting structure; statement on October 1, 2021 by guest. Protected copyright. 46 of whether it is independent from the sponsor and 47 competing interests; and reference to where further 48 49 details about its charter can be found, if not in the 50 protocol. Alternatively, an explanation of why a DMC 51 is not needed 52 53 21b Description of any interim analyses and stopping Page 16 54 guidelines, including who will have access to these 55 56 interim results and make the final decision to 57 terminate the trial 58 59 60

4 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml Page 31 of 32 BMJ Open

1 2 Harms 22 Plans for collecting, assessing, reporting, and Page 14 3 managing solicited and spontaneously reported

4 adverse events and other unintended effects of trial BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 6 interventions or trial conduct 7 8 Auditing 23 Frequency and procedures for auditing trial conduct, if Page 16 9 any, and whether the process will be independent 10 from investigators and the sponsor 11 12 13 Ethics and dissemination 14 Research 24 Plans for seeking research ethics Page 16 15 16 ethics committee/institutional review board (REC/IRB) 17 approval approval 18 For peer review only 19 Protocol 25 Plans for communicating important protocol Page 16 20 amendments modifications (eg, changes to eligibility criteria, 21 outcomes, analyses) to relevant parties (eg, 22 23 investigators, REC/IRBs, trial participants, trial 24 registries, journals, regulators) 25 26 Consent or 26a Who will obtain informed consent or assent from Page 16 27 assent potential trial participants or authorised surrogates, 28 29 and how (see Item 32) 30 31 26b Additional consent provisions for collection and use of Not 32 participant data and biological specimens in ancillary Applicable 33 studies, if applicable 34 35 Confidentiality 27 How personal information about potential and enrolled Page 14--15 36 participants will be collected, shared, and maintained 37 http://bmjopen.bmj.com/ 38 in order to protect confidentiality before, during, and 39 after the trial 40 41 Declaration of 28 Financial and other competing interests for principal Page 18 42 interests investigators for the overall trial and each study site 43 44 Access to 29 Statement of who will have access to the final trial Page 16 45 data dataset, and disclosure of contractual agreements on October 1, 2021 by guest. Protected copyright. 46 47 that limit such access for investigators 48 49 Ancillary and 30 Provisions, if any, for ancillary and post-trial care, and Page 9 50 post-trial care for compensation to those who suffer harm from trial 51 participation 52 53 Dissemination 31a Plans for investigators and sponsor to communicate Page 16 54 policy trial results to participants, healthcare professionals, 55 56 the public, and other relevant groups (eg, via 57 publication, reporting in results databases, or other 58 data sharing arrangements), including any publication 59 restrictions 60

5 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml BMJ Open Page 32 of 32

1 2 31b Authorship eligibility guidelines and any intended use None 3 of professional writers

4 BMJ Open: first published as 10.1136/bmjopen-2019-030041 on 26 August 2019. Downloaded from 5 31c Plans, if any, for granting public access to the full Page 16 6 protocol, participant-level dataset, and statistical code 7 8 9 Appendices 10 11 Informed 32 Model consent form and other related documentation None 12 consent given to participants and authorised surrogates 13 materials 14 15 Biological 33 Plans for collection, laboratory evaluation, and Not 16 specimens storage of biological specimens for genetic or Applicable 17 18 Formolecular peer analysis reviewin the current trial only and for future 19 use in ancillary studies, if applicable 20 21 *It is strongly recommended that this checklist be read in conjunction with the SPIRIT 2013 22 Explanation & Elaboration for important clarification on the items. Amendments to the 23 protocol should be tracked and dated. The SPIRIT checklist is copyrighted by the SPIRIT 24 Group under the Creative Commons “Attribution-NonCommercial-NoDerivs 3.0 Unported” 25 26 license. 27 28 29 30 31 32 33 34 35 36 37 http://bmjopen.bmj.com/ 38 39 40 41 42 43 44 45 on October 1, 2021 by guest. Protected copyright. 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

6 For peer review only - http://bmjopen.bmj.com/site/about/guidelines.xhtml