An Official Journal of the European Society for Medical Oncology The Official Journal of ETOP IBCSG Partners Foundation The Official Journal of the British Thoracic Oncology Group LUNG CANCER AN INTERNATIONAL JOURNAL FOR LUNG CANCER AND OTHER THORACIC MALIGNANCIES Volume 178, Supplement 1, April 2023 ISSN 0169-5002 Poster abstracts of the 21st Annual British Thoracic Oncology Group Conference 2023 ICC Belfast, 26th to 28th April 2023
Editor-in-Chief R. Stahel Volume 178 (2023) AN INTERNATIONAL JOURNAL FOR LUNG CANCER AND OTHER THORACIC MALIGNANCIES Amsterdam—Boston—London—New York—Oxford—Paris—Philadelphia—San Diego—St. Louis Editor-in-Chief R. Stahel Volume 178, Supplement 1 (2023) Poster abstracts of the 21st Annual British Thoracic Oncology Group Conference 2023 ICC Belfast, 26th to 28th April 2023 Publication of this Abstract Book is supported by the British Thoracic Oncology Group (BTOG).
© 2023 Elsevier B.V. This journal and the individual contributions contained in it are protected under copyright, and the following terms and conditions apply to their use in addition to the terms of any Creative Commons or other user license that has been applied by the publisher to an individual article: Photocopying Single photocopies of single articles may be made for personal use as allowed by national copyright laws. Permission is not required for photocopying of articles published under the CC BY license nor for photocopying for non-commercial purposes in accordance with any other user license applied by the publisher. Permission of the publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-proÀ t educational classroom use. Derivative Works Users may reproduce tables of contents or prepare lists of articles including abstracts for internal circulation within their institutions or companies. Other than for articles published under the CC BY license, permission of the publisher is required for resale or distribution outside the subscribing institution or company. For any subscribed articles or articles published under a CC BY-NC-ND license, permission of the publisher is required for all other derivative works, including compilations and translations. Storage or Usage Except as outlined above or as set out in the relevant user license, no part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the publisher. Permissions For information on how to seek permission visit www.elsevier.com/permissions Author rights Author(s) may have additional rights in their articles as set out in their agreement with the publisher (more information at http://www. elsevier.com/authorsrights). Notice Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds or experiments described herein. Because of rapid advances in the medical sciences, in particular, independent veriÀ cation of diagnoses and drug dosages should be made. To the fullest extent of the law, no responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Although all advertising material is expected to conform to ethical (medical) standards, inclusion in this publication does not constitute a guarantee or endorsement of the quality or value of such product or of the claims made of it by its manufacturer. Funding body agreements and policies Elsevier has established agreements and developed policies to allow authors whose articles appear in journals published by Elsevier, to comply with potential manuscript archiving requirements as speciÀ ed as conditions of their grant awards. To learn more about existing agreements and policies please visit http://www.elsevier.com/fundingbodies Cited in: EMBASE/Excerpta Medica Oncology Information Service Elsevier BIOBASE/Current Awareness in Biological Sciences Current Contents/Clinical Medicine SciSearch Index Medicus/MEDLINE SCOPUS® ScienceDirect® The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper)
Vol. 177, March 2023 CONTENTS Cited in: EMBASE/Excertpa Medica, Oncology Information Service, Elsevier BIOBASE/Current Awareness in Biological Sciences; Current Contents/Clinical Medicine; SciSearch, Index Medicus/MEDLINE. Also covered in the abstract and citation database SCOPUS®. Full text available on ScienceDirect®. www.journals.elsevier.com/lung-cancer Poster abstracts of the 21st Annual British Thoracic Oncology Group Conference 2023 ICC Belfast, 26th to 28th April 2023 Basic Science S1 Clinical Networks and Pathways S2 COVID and Cancer S11 Diagnosis and Staging S11 Mesothelioma S19 NSCLC Systemic Treatment S25 Nursing and Allied Health Professionals S38 Other S42 Radiotherapy S53 Screening S66 Small Cell Lung Cancer S74 Supportive Care S76 Surgery S78 Trials in progress S85 Case Reports S90 Author Index S94 Volume 178, Supplement 1, April 2023
Aims and Scope Lung Cancer is an international publication covering the clinical, translational and basic science of malignancies of the lung and chest region. Original research articles, early reports, review articles, editorials and correspondence covering the prevention, epidemiology and etiology, basic biology, pathology, clinical assessment, surgery, chemotherapy, radiotherapy, combined treatment modalities, other treatment modalities and outcomes of lung cancer are welcome. Editor-in-Chief: R. Stahel (Switzerland) Deputy Editors: P. Baas (Netherlands) M. Edelman (USA) S. Peters (Switzerland) Y.-L. Wu (China) Associate Editors: Epidemiology and Prevention: M. Janssen-Heijnen (Netherlands) Pathology: E. Brambilla (France) S. Finn (Ireland) Systemic Treatments: T. Berghmans (France) B.C. Cho (Korea) N. Girard (France) S. Lu (China) I. Okomoto (Japan) A. Passaro (Italy) C. Zhou (China) Radiation Oncology: M. Guckenberger (Switzerland) S. Senan (Netherlands) P. van Houtte (Belgium) Surgery: F. Yang (China) Pneumonology: D. Baldwin (UK) Radiology and Nuclear Medicine: I. Burger (Switzerland) C.A. Ridge (UK) Preclinical and Translational Research: D. Costa (USA) R. Soo (Singapore) Statistics Editor: U. Dafni (Greece) Social Media Editor: A. Passaro (Italy) Editorial Board: AN INTERNATIONAL JOURNAL FOR LUNG CANCER AND OTHER THORACIC MALIGNANCIES An Official Journal of the European Society for Medical Oncology. The Official Journal of ETOP IBCSG Partners Foundation. The Official Journal of the British Thoracic Oncology Group. L. Bubendorf (Switzerland) C. Faivre–Finn (UK) E. Felip (Spain) S. Lam (Canada) C. Mascaux (France) T. Mitsudomi (Japan) V. Ninane (Belgium) K.J. O’Byrne (Ireland) R. Pirker (Austria) M. Reck (Germany) G. Rocco (USA) C. Rolfo (Belgium) J.-P. Sculier (Belgium) T. Stinchcombe (USA) P. VanderLaan (USA) J.P. Van Meerbeeck (Belgium) A. Vergnenègre (France) G. Veronesi (Italy) H. Wakelee (USA) Y.-C. Wang (Taiwan) S. Yom (USA) Editorial Oɝ ce: e-mail: lungcancer@elsevier.com Processed at Thomson Digital, Gangtok (India) doi:10.1016/S0169-5002(23)00110-1
Lung Cancer 178, Suppl. 1 (2023) S1–S100 S1 Contents lists available at ScienceDirect Lung Cancer j o u r n a l h ome p a g e : www. e l s ev i e r . c om / l o c a t e / l u n g c a n An Official Journal of the European Society for Medical Oncology The Official Journal of ETOP IBCSG Partners Foundation The Official Journal of the British Thoracic Oncology Group LUNG CANCER AN INTERNATIONAL JOURNAL FOR LUNG CANCER AND OTHER THORACIC MALIGNANCIES Volume 178, Supplement 1, April 2023 ISSN 0169-5002 Poster abstracts of the 21st Annual British Thoracic Oncology Group Conference 2023 ICC Belfast, 26th to 28th April 2023 Basic Science 1 Prediction of next generation sequencing test failure in lung adenocarcinoma in a genomic laboratory hub setting Zhang, Yu Zhi1,2; Sherlock, Stuart2; MacMahon, Suzanne3; Brambilla, Cecilia2; Rice, Alexandra1,2; Robertus, Jan Lukas1,2; Wassilew, Katharina2; Lim, Eric4,1; Begum, Sofina4; Buderi, Silviu4; Jordan, Simon4; Anikin, Vladimir4,5; Finch, Jonathan4; Asadi, Nizar4; Beddow, Emma4; Garner, Justin L1,6; Morjaria, Jaymin6; Lee, Richard1,7,8; McDonald, Fiona7,8; Antoniou, Georgios9; Ridge, Carole10,11; Padley, Simon1,11; Dalal, Paras11; Morris-Rosendahl, Deborah10,12; Valganon-Petrizan, Mikel3; Shah, Pallav L1,6; Devaraj, Anand1,11; Popat, Sanjay1,7,8; Nicholson, Andrew G1,2 1National Heart & Lung Institute, Imperial College London, London, United Kingdom, 2Department of Histopathology, Royal Brompton and Harefield Hospitals, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom, 3Clinical Genomics Department, The Centre for Molecular Pathology, Royal Marsden Hospital, London, United Kingdom, 4Department of Thoracic Surgery, Royal Brompton and Harefield Hospitals, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom, 5Department of Oncology and Reconstructive Surgery, I.M. Sechenov First Moscow State Medical University, Moscow, Russia, 6Department of Respiratory Medicine, Royal Brompton and Harefield Hospitals, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom, 7Lung Unit, Royal Marsden NHS Foundation Trust, London, United Kingdom, 8Institute of Cancer Research, London, United Kingdom, 9Department of Oncology, Mount Vernon Cancer Centre, London, United Kingdom, 10National Heart & Lung Institute, Imperial College London, United Kingdom, 11Department of Radiology, Royal Brompton and Harefield Hospitals, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom, 12Clinical Genetics and Genomic Laboratory, Royal Brompton and Harefield Hospitals, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom Introduction: Next generation sequencing (NGS) has been established as the first line testing platform for detection of actionable driver mutations in lung adenocarcinoma. We aim to investigate the pre-analytical variables associated with test failure and develop a predictive tool to identify cases at higher risk of failure. Methods: This is a retrospective study including 903 cases of lung adenocarcinoma (Resection: 593, biopsy: 310) diagnosed over a 21-month period, that underwent reflex testing including hybrid DNA/RNA NGS at regional Genomic Laboratory Hub. Pre-analytical variables include procedure type, stage, tissue fixation time, hospital site, tumour cellularity, cancer cell fraction and tumour necrosis. NGS test failure is defined as either failure of pre-sequencing quality control or insufficient target coverage of any gene on designated lung cancer panels. Predictors of test failure were assessed using logistic regression, and performance of predictive models were evaluated using receiver operating characteristics (ROC). Results: Overall test failure rate was 29.2%, across all stages (0: 7.1%, I: 27.6%, II: 28.4%, III: 28.8%, IV: 39.1%). There were significant improvements within the study period (period 1: November Fig. 1 (abstract 1). A) Study population and breakdown of NGS test failure. B) Prediction of NGS test failure is derived from clinical and laboratory parameters, and microscopic assessment. The total score (0-9) is calculated from individual scores from each category. Higher total scores correlate with elevated risk of test failure. BAL, bronchoalveolar lavage; NGS, next generation sequencing. Poster abstracts of the 21st Annual British Thoracic Oncology Group Conference 2023 ICC Belfast, 26th to 28th April 2023 0169-5002/$ – see front matter © 2023 Elsevier BV. All rights reserved.
S2 Poster abstracts of the 21st Annual BTOG / Lung Cancer 178S1 (2023) S1–S100 2020-January 2022, period 2: February 2022-September 2022) for both resections (31.5% vs 6.3%) and biopsies (54.8% vs 37.2%), associated with reduced fixation time and positive shift in tumour cellularity. Multivariate analysis showed image-guided biopsy, fineneedle aspiration cytology, prolonged fixation, lowtumour cellularity and low cancer cell fraction were independent predictors of test failure, but not stage or tumour necrosis. A four-factor predictive model using procedure type, fixation time, tumour cellularity and cancer cell fraction stratified cases into low, intermediate and high risk of test failure, with AUC of 0.782. Conclusions: We present a practical framework for predicting NGS failure using routinely available parameters. This allows preprocessing risk stratification, and may help refine diagnostic and testing pathways as part of an integrated molecular diagnostic service. Disclosure: No significant relationships. Clinical Networks and Pathways 2 Genomic Laboratory Hubs (GLHs) in the predictive profiling of non-small cell lung cancer: adapting to centralisation and improving the pathways Haragan, Alexander1; Gosney, John2 1Royal Liverpool University Hospital, Liverpool, United Kingdom, 2Royal Liverpool University Hospital Background: In April 2021, genomic profiling of non-small cell lung cancer (NSCLC) was centralised in England into seven Genomic Laboratory Hubs (GLHs) with the aim of providing equitable access for all patients to routine next-generation sequencing (NGS). In July 2022, because of long turnaround times (TATs) for NGS, our local profiling service was modified by adding regionally-funded EGFR testing to our existing, in-house reflex testing by immunochemistry for PD-L1 expression and ALK and ROS1 fusion. We thought it would be of value to examine the effects of centralisation and our subsequent adaptation to it. Methods: We reviewed the progress and results of analysis of 900 specimens of NSCLC submitted for genomic profiling between October 2020 and October 2022, divided into three groups; 300 precentralisation, 300 immediately post-centralisation into the North West GLH, and 300 following modification of our in-house testing service. Results: Three key metrics across the three groups, turnaround time (TAT), inadequacy rates and detection of actionable mutations, are shown in the Table. Overall, only 2% of specimens, diagnostic and resected tumours, failed in-house profiling due to inadequacy or for technical reasons. Discussion: The improvement since centralisation of genomics in the parameters considered here is attributable to obsessive attention to tissue stewardship, increased efficiency in the pathway generally, (including at the GLH) and, in particular, modest expansion of our in-house profiling to include testing for EGFR mutations, which means that crucial information on common actionable mutations applicable in the first-line setting is available to clinicians early in the pathway. We suggest that the formal devolution and funding of some of the testing involved in the entire analytical process would streamline the pathway, reduce pressure on GLHs and, ultimately, improve the service for patients. Disclosure: No significant relationships. 3 Patient outcomes following endovascular stent insertion for malignant superior vena-cava obstruction with a focus on anticoagulation practice Walker, Joshua1; Ravinthiranathan, Amsajini1; Diamantopoulos, Athanasios1; Danaee, Anicee1; Georgiou, Alexandros1,2 1Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom, 2School of Cancer and Pharmaceutical Sciences, King’s College London, Guy’s Campus, London, United Kingdom Introduction: Malignant superior vena cava obstruction (SVCO) results in significant morbidity and mortality. Lung cancer is the most common cause. Endovascular stent insertion for symptomatic relief is endorsed by NICE 2004 guidance. Following stent insertion, antithrombotic therapy is considered however there is a lack of consensus on the type, dose and duration of therapy. Methods: We retrospectively identified patients who underwent endovascular stenting for malignant SVCO between 2016-2021. We report the patient characteristics, stent-related complications and antithrombotic therapy details. Results: Forty-nine patients were identified. Median age was 61 and majority (n=39, 80%) had lung cancer (non-small cell, n=29 (60%)). Thirty-seven (76%) patients had metastatic disease. Median survival from stenting was 71 days (range 1-613). Twenty-three (47%) patients had a documented fast symptomatic benefit, 6 (12%) did not benefit and in 20 (40%) the response was not documented. Four (8%) patients experienced acute (within 48h) complications (3 stent migration, 1 dysphagia). Overall the commonest complication was thrombosis: 7 cases (14%) of stent thrombosis. There were 4 cases (8%) of stent narrowing. Ten cases (20%) of other thrombotic events (5-Pulmonary, 2-Subclavian, 1-Internal Jugular, 1-Saphenous and 1-Axillary) Twenty-one (43%) patients were on anticoagulants before stenting. In 88% of patients an anticoagulation plan was documented by the Interventional Radiologist. Treatment-dose LMWH was recommended in 25 (51%) cases. Figure 1 (overleaf) highlights the variability in anticoagulation plans. There were 2 cases of bleeding. Conclusion: Most patients with SVCO experience symptomatic benefit following stenting. These patients have a poor prognosis. Our findings highlight the significant burden of thrombotic complications. There is variability in anticoagulation practice and updated guidelines are required. Given the complexity of these cases with pro-thrombotic diathesis, planned anticancer therapies with possible drug interactions and the expanding anticoagulation options; specialist anticoagulation haematology input is required to optimise prevention of thrombotic complications. Disclosure: No significant relationships. Table 1 (abstract 2).
Poster abstracts of the 21st Annual BTOG / Lung Cancer 178S1 (2023) S1–S100 S3 4 Respiratory consultant triage of 2WW CT scans Hulston, George1; O’Donnell, Paul1; Naseer, Rehan1 1Fairfield General Hospital, Northern Care Alliance NHS Trust, Bury, United Kingdom Introduction: There is a shortfall of around 2,000 consultant radiologists in the UK, with UK having one of the lowest numbers of radiologists per head in Europe. Inevitably, such shortages have led to reporting delays for patients on 2WW pathways. Faster diagnostic standards aims for all patients with suspected cancer to have a clear diagnosis within 28 days of referral. Within our trust, a direct to CT pathway following an abnormal CXR was implemented in January 2022. Methods: Following an abnormal CXR alert, a CT is organised by the radiology department. Due to delay in reporting, the CT images are reviewed and triaged by a consultant respiratory physician. We aimed to analyse the accuracy and effectiveness of this triage. Results: Between the 17/01/2022 and 11/12/2022, 362 patients had a ‘straight to test’ CT scan, an average of 6 days from the CXR report. CT reporting took an average of a further 10 days. Recording of the date of respiratory physician triage commenced in July, with mean time to triage from CT scan 2 days (208 patients). In 10 cases the triage thought there was no evidence of malignancy, but the final report disagreed. Of these, only 2 were diagnosed with malignancy - Hodgkin’s lymphoma and an early stage radiological diagnosis of lung cancer. In 28 instances the triage thought the scan was possibly suggestive of malignancy, when the radiology report did not. In none of these cases was the ultimate diagnosis malignancy. Conclusions: Respiratory consultant triage of ‘straight to CT’ scans was accurate, missing only 2 cases of malignancy in 362 scans that were otherwise identified by the radiology report. This shows respiratory consultant triage can be used to appoint patients earlier, and highlights CT scans for expedited reporting. In our trust this saved 8 days. Disclosure: No significant relationships. 5 Impact of the New Optimal Lung Cancer Pathway (NOLCP) at Cambridge University Hospitals (CUH) NHS Trust Gordon, Laura1; Beckford, Jake1; Verissimo, Tiago1; Roberts, Hilary2; Babar, Judith1; McCaughan, Frank2 1Cambridge University Hospital, Cambridge, United Kingdom, 2Cambridge University Hospital, Cambridge, UK The NOLCP was proposed in 2017 with the aim of reducing the number of days between referral for initial assessment (typically a chest X-ray) and initiation of treatment for lung cancer. At CUH we introduced the NOLCP in stepwise fashion across the region starting late 2020. In 2021 16 GP practices referred via the NOLCP, others via the traditional 2WW pathway. We now report a retrospective audit comparing these pathways. A key aspect of the NOLCP was that GPrequested CXRs were reported by a trained radiographer whilst the patient was in the department and a CT offered within 72 hours as appropriate. Methods: All patients with a final diagnosis of lung cancer referred to the NOLCP/2WW pathways in 2021 were included. Patients who already had a reported CT suspicious for lung cancer were excluded. Results: 98 of 1400 NOLCP patients were offered a CT, 97% within 72h. 12 NOLCP (0.9%) patients were diagnosed with lung cancer. 45 2WW patients were included. The median time-to-CT from referral was 0 days for the NOLCP and 5 days for the 2WW cohort (p=<0.0001). Median interval between CXR and CT was 0 days for NOLCP and 8 days with 2WW (p=<0.001). Median time-to-clinic was Fig. 1 (abstract 3). Antithrombotic practice. Tx LMWH, treatment-dose low-molecularweight heparin; Pr LMWH, prophylactic-dose low-molecular-weight heparin; DOAC, direct oral anticoagulant. Fig. 1 (abstract 4). Fig. 1 (abstract 5). Comparison of times to a) CT, b) interval between CXR and CT, c) clinic, and d) diagnosis (pathology result or radiological diagnosis) between the NOLCP and 2WW pathway.
S4 Poster abstracts of the 21st Annual BTOG / Lung Cancer 178S1 (2023) S1–S100 3 days (NOLCP) and 9 days with 2WW (p=<0.0006). Median time-todiagnosis was 24 days for NOLCP and 28 days with 2WW (p=0.8259). Discussion: Introduction of NOLCP at CUH significantly reduced the interval between CXR and CT scan for lung cancer patients. It also improved the time-to-CT and clinic compared to the 2WW pathway. The time to diagnosis was not significantly changed reflecting the prior excellent access to CT scan for lung 2WW patients. The NOLCP has now been expanded to include all GP practices in the region. Disclosure: No significant relationships. 6 BTOG Thymic Special Interest Group: Establishment of a national clinical network for thymic malignancies in the UK Lynskey, Deirdre1; Gault, Abigail2; Robinson, Stephen3; on behalf of the BTOG Thymic Special Interest Group 1Cambridge University Hospitals NHS Trust, Cambridge, United Kingdom, 2Northern Centre for Cancer Care, Newcastle upon Tyne, United Kingdom, 3The Royal Marsden Hospital NHS Foundation Trust, London, United Kingdom Introduction: Specialist thymic tumour provision has recently been under scrutiny, given international efforts to improve care for these patients. A recent multi-centre UK service review demonstrated differing UK practice [1], whilst the French nationwide RHYTHMIC network [2] is set to prospectively evaluate outcomes following centralisation of multidisciplinary discussions and improved specialist input into standard care. In 2022 the British Thoracic Oncology Group (BTOG) formed a Thymic Special interest Group (SiG) to develop and improve consistency in UK practice. Methods: A multidisciplinary SiG was convened involving a network of specialists from across the UK (thoracic oncologists, surgeons, respiratory physicians, radiologists, neurologists, nurse specialists and Thymic UK charity stakeholders) invested in improving opportunities for sharing of best practice. Results: Over 2022, we have successfully established routine meetings, terms of reference and raising awareness of the SiG. We collected data on how UK multi-disciplinary meetings cater for this rare tumour group to understand how to standardise UK practice. A key output was delivery of a Thymic Malignancies Essential Update educational day; further educational events are planned. We are building a webpage with free relevant integrated resources/ publications accessible to healthcare professionals. We have begun drafting early-stage and late-stage clinical diagnostic pathways to share nationally. Conclusion: The creation of a BTOG Thymic SiG demonstrates significant UK engagement and multidisciplinary collaboration for this rare malignancy and the collective outputs of the group will lead to improved and consistent care. We hope that this will be helpful to UK clinicians and other interested groups internationally. References: [1] National Oncology Trainees Collaborative for Healthcare Research (NOTCH). Clinical features, management and outcomes of patients with thymoma and thymic carcinoma in a multicentre UK series. Lung Cancer 2020;139(1):S73 [2] Merveilleux du Vignaux C et al. Systemic Therapy in Advanced Thymic Epithelial Tumors: Insights from the RYTHMIC Prospective Cohort. J Thorac Oncol. 2018;13(11):1762-1770 Disclosure: No significant relationships. 7 The utility of high-fidelity simulation in improving trainee confidence within the lung cancer multidisciplinary team setting Patel, Kishen1,2; Mackay, Ewan3; Tipples, Karen4; Januszewski, Adam5; Ricketts, William6 1Clinical Oncology Department Mount Vernon Cancer Centre, Northwood, United Kingdom, 2NIHR Academic Clinical Fellow, University College London, London, United Kingdom, 3Respiratory Medicine, King’s College Hospital NHS Foundation Trust, London, United Kingdom, 4Clinical Oncology, St Bartholomew’s Hospital, London, United Kingdom, 5Medical Oncology, St Bartholomew’s Hospital, London, United Kingdom, 6Respiratory Medicine, St Bartholomew’s Hospital, London, United Kingdom Introduction: Treatment of lung cancer is dependent on many factors including staging, performance status and genomics. Due to this complexity, management decisions nearly always occur at multidisciplinary team (MDT) meetings to optimise care outcomes [1]. MDT meetings are an important component of respiratory and oncology consultant workload, however 50% of trainees report that they are not satisfied with their current training for MDT working [2]. Our hypothesis was that simulated MDT sessions could be an effective means to improve confidence in these meetings and promote inter-disciplinary working. Methods: The aim of this study was to assess whether multispeciality simulated scenarios prepared specialist registrars for interacting within an MDT. Our team created multiple lung oncology MDT scenarios with focused learning objectives and pre-determined end points. For each scenario, a respiratory and oncology trainee played the role of a new consultant, leading the presentation and discussion of simulated patients, interacting with other participants and actors playing the role of additional MDTmembers. Feedbackwas provided by two lung oncology specialists. Participants completed a pre- and post-questionnaire utilising visual analogue scales which assessed their current training, confidence and whether the course would alter future practice. Results: Our results demonstrated an increase in trainee confidence from a mean score of 4.92 to 7.33. Trainees rated the session highly in terms of utility and altering their future practice (mean scores of 9.25 and 8.67, respectively, out of 10). Conclusion: Simulation has shown success in othermultidisciplinary teaching; however there are no lung cancer specific training programmes. Our results highlight a potential gap in UK specialist training curriculums, and suggest that simulation may be beneficial in preparing trainees to present in MDT meetings. Reference: [1] Eberhardt. 2nd ESMO Consensus Conference in Lung Cancer. 2015. [2] Nally. Improving the educational value of the MDT meeting. 2021. Disclosure: No significant relationships. 8 The impact of military and civilian health partnership in raising awareness of asbestos exposure in military personnel and veterans Moylan, Anne1; Darlison, Liz1; Hutchinson, George2 1Mesothelioma UK, Leicester, United Kingdom, 2Ministry of Defense, London, United Kingdom Introduction: Understanding more about the impact of asbestos exposure and the incidence of mesothelioma in British Armed Forces personnel from a public health perspective has been a principle aim of Mesothelioma UK Supporting our Armed Forces (MUK-SOAF) project. With support of a grant from the Armed Forces Covenant
Poster abstracts of the 21st Annual BTOG / Lung Cancer 178S1 (2023) S1–S100 S5 Fund Mesothelioma UK (MUK) have continued to work with the Ministry of Defence (MOD) to raise awareness of risks of asbestos exposure to armed forces and civilian personnel. Methods: Contact was established with the Health Safety and Environmental Protection (HS and EP) Directorate at the MOD and potential asbestos exposure from both military equipment and estates as well as themanagement of asbestos in situwas explored. In response to a Parliamentary question posed in January 2021 the MOD pledged to continue their engagement with MUK and committed to the management (removal or encapsulation) of asbestos wherever practicable. MUK has proposed adding mesothelioma and the risk of asbestos exposure to both the armed forces medical community education programme across the board and to the health and safety training for all recruits. Engagement with other agencies within the MOD has commenced to facilitate these objectives. Results: As a result of this project the MOD have committed to the management of asbestos within estates and equipment wherever practicable and have established a programme of work to achieve this. We will continue to focus developing an awareness and education module for health professionals within the military as well as introducing asbestos into the health and safety training of all recruits. Conclusion: Achieving the objectives has been protracted due to the global pandemic but both agencies are focused on building on the progress to date. This project illustrates what can be achieved by military and civilian health agencies working in partnership. Disclosure: No significant relationships. 9 A Quality Improvement Project to Improve the Efficiency of Pulmonary Nodule Service at a Tertiary Centre Khan, Muhammad Ziaullah1; Park, John1; Moore, Alastair1; Benamore, Rachel1; Yen, Duong1; Falolu, Adedamola1; Yates, Joanne1; Ridgeon, Claire1; O’Mahony, Anne1; Ghidoni, Giulia1; Talwar, Ambika1 1Oxford University Hospital, Oxfordshire, United Kingdom Introduction: With the increasing detection of incidental pulmonary nodules (IPNs), there is a clinical need for a dedicated IPN service to ensure that growing PNs are managed in a timely manner. Pre COVID-19, our centre ran a virtual nodule service, delivered on an ad-hoc basis by the lung cancer physicians. We hypothesised that efficiency of the service would improve with a dedicated nodule team. We were awarded a pump priming grant by the Thames Valley Cancer Alliance to implement a nodule navigator run service. We report the initial outcomes of this project here. Objective: To evaluate the PN navigator service. Methods: Retrospective data pre-service development was collected from patients presenting to the PN service between April and June 2022. The service was established in October 2022 and data from October and November 2022 collected. Student t-test was used to compare means. Results: 107 patients were included pre-service and 92 patients in the post-service development cohorts. Data for time to CT report and patient contact are summarised in Table 1. There was no reduction in mean time from CT scan date to CT report (Table 1; 31vs 27; p=0.143) but a reduction was seen between CT report and patient contact (Table 1, 45 vs 20; p<0.001). Conclusion: This small cohort study shows an improvement in the time between CT scan and patient contact following the introduction of a dedicated PN service. This may reduce delays in the diagnosis of early-stage lung cancer. Whilst there was no significant difference between the CT scan date and CT report, these data highlight an area in the pathway that can improve. Further aims of the project are to collect patient satisfaction and IPN discharge. Disclosure: No significant relationships. 10 The routine clinical use of electronic patient-reported outcome measures (ePROMs) for patients with lung cancer Crockett, Cathryn1; Price, James2; Pham, Mai2; Abdulwahid, Danya2; Bayman, Neil2; Blackhall, Fiona3,2; Bostock, Layla2; Califano, Raffaele2,3; Chan, Clara2; Coote, Joanna2; Cove-Smith, Laura2; Eaton, Marie2; Fenemore, Jacqueline2; Gomes, Fabio2; Harris, Margaret2; Halkyard, Emma2; Hughes, Sarah2; Lindsay, Colin2; Neal, Hilary2; McEntee, Delyth2; Pemberton, Laura2; Sheikh, Hamid2; Summers, Yvonne2; Taylor, Paul2; Woolf, David2; Yorke, Janelle2,4; Faivre-Finn, Corinne2,3 1Northern Ireland Cancer Centre, Belfast, United Kingdom, 2The Christie NHS Foundation Trust, Manchester, United Kingdom, 3Division of Cancer Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Manchester, United Kingdom, 4Christie Patient-Centred Research, Division of Nursing, Midwifery & Social Work, The University of Manchester, Manchester, United Kingdom Background: The Christie NHS Foundation Trust launched their routine clinical ePROMs service in January 2019. The lung cancerspecific questionnaire consists of 14 symptom items, adapted from the CTCAE (v5.0) and the EuroQol EQ-5D-5L QoL tool. Both new and follow-up lung cancer patients are invited to complete questionnaires assessing their symptoms and QoL using an online platform called MyChristie-MyHealth. Methods: For patients who completed questionnaires between January 2019 and December 2020, the ePROM responses and clinical, pathological and treatment data were extracted from electronic medical records. The symptom and QoL scores of patients who completed baseline pre-treatment ePROMs and also those who completed ePROMs pre- and post- palliative lung systemic anticancer therapy (SACT) or radical thoracic radiotherapy were evaluated. Pre-treatment questionnaires were analysed based on age group, ECOG PS and ACE-27 comorbidity score. Results: The analysis included 1480 lung cancer patients. There were no statistically significant differences in symptom and QoL scores between age groups. Cough (p=0.006) and EQ-5D-5L mobility scores (p=0.006) were significantly worse for patients with an ECOG PS of 0-1. Dyspnoea (p=0.035), haemoptysis (p=0.023), nausea (p=0.041), mobility (p=0.004) and self-care (p=0.0420) were significantly worse for those with higher ACE-27 scores (2-3 vs. 0-1). Palliative SACT was associated with a significant improvement in cough (p=<0.001) and haemoptysis (p=0.025), but significantly negatively impacted mobility (p=0.013). Patients receiving radical thoracic radiotherapy reported a significant improvement in haemoptysis (p=0.042) but worse pain (p=0.002) and fatigue (p=0.01). Other changes in symptom and QoL scores were not significant. Table 1 (abstract 9). Pre- and post-service timings of CT date to CT report, and CT report to patient letter.
S6 Poster abstracts of the 21st Annual BTOG / Lung Cancer 178S1 (2023) S1–S100 Conclusion: The symptom and QoL scores reported at baseline and pre- and post- both palliative SACT and radical thoracic radiotherapy are clinically relevant and meaningful. We have demonstrated that it is feasible to implement ePROMs into routine clinical practice and they have the potential to inform and transform clinical practice and future research. Disclosure: No significant relationships. 11 Process mining as a research tool to evaluate care pathways for patients with an indeterminate pulmonary nodule demonstrates the high variability between patient pathways across a region Ratnakumar, Ratnaprashanthika1; Derbyshire, Jane2; Morris, Katie2; Burchill, John2; Mayer, Erik1; Quint, Jennifer3; Bloch, Susannah1 1Imperial College Healthcare NHS Trust, London, United Kingdom, 2Royal Marsden Partners; West London Cancer Alliance, London, United Kingdom, 3Royal Brompton Hospital & Imperial College Healthcare NHS Trust, London, United Kingdom Introduction: Earlier diagnosis of lung cancer remains imperative within UK cancer services [1].As of 2019 national data, just 26.6% of lung cancers were diagnosed at Stage 1 or 2 [2].The efficient management of indeterminate pulmonary nodules (IPNs) can improve this. Whilst national guidance dictates surveillance timelines, understanding models of care which exist and evaluating variation is key to support equitable and timely management as well as efficient resourceuse. Processmining is a set of research techniques which have evolved from business process modelling. Extracting information from routinely-collected event data in clinical systems supports objective analysis using ‘miners’(algorithms) to model pathways. This allows characterisation of services at a regional level, assessing for deviation, as well as modelling improvements. The aim of this project was to demonstrate use of process mining to evaluate IPN patient pathways across the West London region using patientlevel data and measure the degree of variability seen. Methods: Patient-level data was collected from 9 NHS Trusts across West London. Event logs were created from primary data, and miners then tested to describe pathway patterns, identify similarities, and create models for evaluation. Results: A total of 80 patients have been analysed to date(pathway duration ranging from 3-111.3 months). Event logs demonstrate a total of 866 individual events(10.8 events per patient),with 56 variants of pathway observed within the data. Analysis using the Alpha, Heuristic and Inductive miners demonstrates that the Inductive miner produces the best-fit pathway model (Figure 1).This has a trace fitness of 0.47, indicating high variability between patient pathways in clinical practice. Conclusions: Understanding care pathways for patients with IPNs supports early cancer diagnosis. To date, most pathways are evaluated solely through service evaluation. Process mining provides a valuable research tool which can be applied to healthcare to objectively identify and analyse variance, as well as model improvement. Further work will involve larger models using additional data, and analysing variance between patient subgroups. References: [1] NHS. The NHS long term plan. https://www.longtermplan.nhs. uk. 2019. [2] Cancer Research UK. https://www.cancerresearchuk.org/healthprofessional/cancer-statistics/statistics-by-cancer-type/lungcancer#heading-Zero. Accessed April 2022. Disclosure: No significant relationships. 12 Developing a regional hub-and-spoke model for volumetric assessment of indeterminate pulmonary nodules supports efficient and effective management of pulmonary nodule pathways Ratnakumar, Ratnaprashanthika1; Sabarwal, Kivraj1; Derbyshire, Jane2; Stephenson, Candice3; Rhodes, Anita4; Morris, Katie2; Burchill, John2; Devaraj, Anand5; Orhan, Orhan6; Sheard, Sarah1; Mayer, Erik1; Quint, Jennifer7; Bloch, Susannah1 1Imperial College Healthcare NHS Trust, London, United Kingdom, 2Royal Marsden Partners; West London Cancer Alliance, London, United Kingdom, 3Kingston Hospitals NHS Foundation Trust, London, United Kingdom, 4Kingston Hospitals NHS Trust, London, United Kingdom, 5Royal Brompton Hospital, London, United Kingdom, 6Chelsea & Westminster Hospitals NHS Trust, London, United Kingdom, 7Royal Brompton Hospital & Imperial College Healthcare NHS Trust, London, United Kingdom Introduction: Indeterminate pulmonary nodules (IPNs) are a growing cohort, with a reported prevalence of up to 53% within screening populations, or 13% in non-screening populations [1]. Effective management is key for efficient discharge and resource management, reduction of anxiety for patients, and early detection and treatment of cancerous nodules. Whilst national guidelines recommend volume measurements for more efficient management, technical and resource constraints mean that this is not always available. Within the West London Cancer Alliance group, a 2019 survey demonstrated that only 2 of 10 NHS Trusts regularly had access to volumetry software. A regional pilot was set up to: • develop a protocol for a hub-and-spoke. • evaluate usage(patient numbers, outcomes, qualitative feedback). • measure pathway costs Two ‘hub-and-spoke’ models were set up, with 5 participating hospitals. The pilot ran for 12 months, with a cap of two referrals per week per hub. Modelling was used to evaluate cost and capacity impact. Results: In total,47 patients were referred (55 nodules evaluated). Mean nodule diameter was 8.1mm (IQR 6-9.75, range 5-14mm). Mean volume was 300mm3. Table 1 summarises overall outcomes. Cost data was derived from National Tariff documents (2021/22 financial year).Where surveillance was shortened to 1 year through volumetry, total cost per patient was £581: a cost saving of £316 compared to a two-year pathway. This results in actual savings of £11,481 per 50 patients, and yearly savings of £31,200-£43,200, to a maximum of £82,000 in full service. A return capacity of 195-270 additional clinic and scan slots (up to 810 if discharged at baseline) could be generated. Fig. 1 (abstract 11). Process model (petri net formation) constructed using Inductive Miner. The miner is better able to handle loops of activity and allows filtration of logs. There is a sequential pattern (3 months, 12 months, 24 months) which can be seen. However, this is still not a completely sound model; the pathways grouped at the top (circled in blue) do not appear to reflect log activity accurately, and there are dead-end transitions (circled to the right of the model) which need to be removed.
Poster abstracts of the 21st Annual BTOG / Lung Cancer 178S1 (2023) S1–S100 S7 Conclusions: Our pilot demonstrates viability and acceptability of a hub-and-spoke model for volumetry. Feedback highlighted benefits of widening access and earlier discharge, with key limitations being administrative capacity to ensure streamlining. Overall, this proofof-concept study improves early discharge of patients with IPNs regionally, with associated cost and capacity benefits to Respiratory services. Reference: [1] Larici AR, Farchione A, Franchi P, et al. Lung nodules: size still matters. Eur Respir Rev 2017;26(146):170025. Disclosure: No significant relationships. 13 Long term follow up of patients referred to the lung two week wait pathway with haemoptysis and a normal CT: do we miss cancer? Brackenborough, Kate Brackenborough1; Trevelyan, Gareth1; Park, John1; Benamore, Rachel1; Hameed, Fawad1; Joshi, Anuja1; Mayhew, Jonathan1; Moore, Alastair1 1Oxford University Trust, Oxford, United Kingdom In 2019 we conducted a retrospective 5-year study (Jan 2014 – Dec 2018) of haemoptysis referral data. 39 cases had a final diagnosis of cancer, all with abnormal CT. In the patients with a normal CT there were no additional cancers identified at bronchoscopy. Consequently, patients with haemoptysis and a normal CT scan do not have bronchoscopy. To bring further reassurance we reviewed the same patients for the development of malignancy. 139 (34.6%) of the 402 patients included in the 2019 study had a normal CT and flexible bronchoscopy. The 139 patients’ electronic notes were interrogated (2022) looking for evidence of subsequent cancer diagnoses. Where malignancy had been identified an experienced thoracic radiologist reviewed the original images to look for retrospective evidence of lesions present on the initial imaging. The timeframe of review ranged from 4 years to 9 years (median 6 years 7 months) from original referral. When reviewing the 139 patients from the original data with an initially normal CT and normal flexible bronchoscopy; seven (5%) have subsequently developed a malignancy. Three (2.16%) would not have been identifiable on the original limited CT scans (colon, testicular, myeloma). Four (0.4%) developed malignancies that might have been seen on initial imaging (neuroendocrine carcinoma of pancreas, breast, vocal cord/ lung, larynx) however of the four patients, retrospective analysis of the original CT scans by experienced thoracic radiologists failed to find any evidence of these tumours. This longer-term retrospective analysis of previously normal CT and bronchoscopy at referral confirms that clinical and radiological assessment in patients who have been referred with a discrete episode of haemoptysis is sufficient to exclude lung malignancy. The development of two subsequent laryngeal malignancies suggests ENT review remains important if haemoptysis is recurrent. Disclosure: No significant relationships. 14 Community Access To Chest CT (CATCH): experiences from a tertiary centre of the CATCH CT protocol for equivocal chest radiographs requested in primary care Broadhurst, Philip1; Lamb, Rochelle1; Barclay, Elizabeth1; Sharman, Anna1 1Manchester University NHS Foundation Trust, Manchester, United Kingdom Introduction: CATCH protocol was established in 2016 at the cardiothoracic tertiary centre in Manchester. The protocol allows patients with equivocal chest radiograph findings requested in primary care to be referred directly for a low dose CT thorax; avoiding potentially unnecessary 2WW referrals. The target time taken for the CT to be performed following the chest radiograph report was set at three weeks. We present our experiences of this protocol over a six year period. Methods: Patients who underwent a low dose unenhanced CT Thorax following an equivocal chest radiograph requested by primary care were identified through the local radiology information system (RIS). Patients were identified across three discrete periods between March 2016 and March 2022. Data were extracted to assess the time taken for the chest radiograph to be reported, time taken for the CT scan to be reported and the total time taken for the diagnostic journey. The final abnormality described on the CT scan was recorded, in order to assess the prevalence of benign versus malignant pathology. Results: Across the time periods interrogated, a total of 288 CTs were performed. Following descriptive analysis, there has been a reduction in the percentage performance across all three metrics (Table 1). In each time period, the percentage of malignant pathology detected on CTs were 16, 7 and 6, respectively. Conclusion: CATCH protocol plays a useful role in the radiology department. The reasons for the decline in performance at our centre are multifactorial; but may relate to the utility of outsourcing primary care chest radiographs, limited capacity for cross-sectional imaging and the need to prioritise diagnostic waiting list recovery following the pandemic. Disclosure: No significant relationships. 15 Is it time to ring an alarm among young adults with NSCLC? A decade of follow up from an Indian cohort Saikia, Jyoutishman1; Malik, Prabhat2; Kumar, Sachin2; Jain, Deepali2; Madan, Karan2; Bharati, Sachidanand2; Deo, Suryanarayana2; Kumar, Sunil2 1Department of Surgical Oncology, All India Institute of Medical Sciences, New Delhi, India, 2All India institute of Medical Sciences, New Delhi, India Introduction: The incidence of young lung cancer patients is reported to be increasing globally. Although they are considered to be aggressive, outcomes after multimodality treatment were encouraging. However, behavior of young NSCLCs in developing countries were less clear. Methods: From 2012 to 2021 patients with non-small cell lung cancer belonging to Stage IA to IIIB undergoing operative resection were included in the study. Data was generated prospectively from a dedicated database team who would look after the postop and follow-up details. The follow-up was prescribed for a minimum of 6months. Results: A total of 226 patients underwent resection. A third of the patients were younger, less than 50 years (n=66, 29.2%). In younger age group adenocarcinoma were around 55%. Many of these patients underwent some form of neoadjuvant treatment before surgery. Smokers in the older subgroup were nearly two times that of younger patients. The median survival was 52 months. Figure Survival plots as shown in figure 1. Cox proportionate hazards model suggested significantly better survival than his younger counterparts (HR=1.944, 95% CI-1.224-3.087, p=0.005). Table 1 (abstract 14). Finding for performance metrics of the CATCH protocol over time.
S8 Poster abstracts of the 21st Annual BTOG / Lung Cancer 178S1 (2023) S1–S100 Conclusions: Among Indian lung cancer patients’ young age lung cancer patients did better than its counterparts. Lung cancer in the young population is a distinct entity that predominantly presents as adenocarcinoma, with lower smoking behavior relative to older patients. Disclosure: No significant relationships. 16 Preliminary review of the Lung One Stop Clinic in Greater Manchester Hiu Nian Chung, Jonathan1; Sivabalah, Kavita1; Banfill, Kathryn2; Hayes, Tim1; Joshi, Vijay1; Granato, Felice1; Ng, Cassandra1; Huddart, Helen1; Rowlinson-Groves, Kirsty1; Peplow, Karen1; Watts, Julie1; Keeghan, Siobhan1; Galligan-Dawson, Lisa1; Lyons, Sarah1; Hewitt, Kath1; Evison, Matthew1 1Wythenshawe Hospital, Manchester University NHS Foundation Trust, Manchester, United Kingdom, 2The Christie NHS Foundation Trust, Manchester, United Kingdom Introduction: Lung cancer is the leading cause of cancer-related death in the UK, with delays in diagnosing and treating lung cancer leading to reduction in life expectancy. The Lung Cancer One Stop Clinic was developed as a clinical service model to patients across Greater Manchester with a multi-disciplinary approach by meeting different treating specialists in a single morning (surgeons, oncologists, thoracic anaesthetists, onco-geriatrics, Prehab4Cancer and smoking cessation). This new clinic model aims to eliminate multiple appointments across a range of providers and accelerate lung cancer treatment for this highly complex cohort of patients with uncertain treatment modalities, typically driven by a higher level of risk from treatment-related complications due to comorbidity and frailty. Methods: One Stop Clinic patients were reviewed over a 3-month period from its launch. Primary metrics measured include median time from Referral Acceptance to decision to treat(DTT) and proportion of patients in which a decision to treat is made on the day. Adherence to optimisation protocols and patient satisfaction surveys were also reviewed. Results: 59 patients were seen in fully staffed clinics, with no patient cancellations or DNAs. The median wait from Referral Acceptance to DTT was 7 days compared to 35 days in the pre-clinic launch cohort accruing to a median saving of 28 days. 51 out of 59 patients(86%) had a decision to treat on the day. There were 100% screening rates for tobacco dependency, frailty(oncogeriatrics), prehab, nutrition, and alcohol dependency. 100% of patients said they felt ‘very well supported’ with a 9.5/10 overall rating of the clinic experience. Conclusion: The One Stop Clinic demonstrates successful implementation of a new model of cancer treatment care with immediate and substantial pathway improvements. The benefits of holistic care for lung cancer patients are unquestionable. We aim to open a second clinic per week to further improve pathways. Disclosure: No significant relationships. 17 Molecular profiling of lung adenocarcinoma using a genomic laboratory hub approach: a single centre realworld experience Yang, Yu-Hsuen1; Reeves, Sarah1; Whittle, Bethany1; Gilliam, Harry1; Chan, Cheng Pou1; Chu, James1; Britten, Anna1; Guy, Catherine2; Zaki, Kamarul1 1Sussex Cancer Centre, Royal Sussex County Hospital, Brighton, United Kingdom, 2Department of Cellular Pathology, Royal Sussex County Hospital, Brighton, United Kingdom NHS England have implemented a national network of regional Genomic Laboratory Hubs (GLHs) to integrate routine molecular profiling into oncological practice. Since April 2022, the Royal Sussex County Hospital (RSCH) has utilised the South East GLH to identify molecular targets in lung adenocarcinoma (LUAD) samples. We conducted an audit of this novel pathway. Data on patients diagnosed with LUAD between April to October 2022 at RSCH was reviewed retrospectively. Immunohistochemical PD-L1 expression was obtained locally. Within the GLH, next generation DNA sequencing was used to detect mutations in EGFR/KRAS/BRAF, and RNA sequencing was used to detect structural variants of ALK/ROS1/ NTRK/MET/RET. Of the 30 patients included, 14 presented with early-stage, 14 with metastatic and 2 with relapsed disease. Median times to diagnosis and treatment were 31 and 86.5 days respectively. The median turnaround time for GLH molecular profiling was 28 days (range: 15-42 days). Failure rates for DNA and RNA testing were 3% (n=1/30) and 59% (n=16/27) respectively. Salvage testing was performed on 38% (n=6/16) of initial failures, and all attempts passed. Activating EGFR, KRAS and BRAF mutations were detected in 24% (n=7), 31% (n=9) and 0% (n=0) respectively. The actionable KRAS G12C mutation was present in 17% (n=5). No structural variants were identified apart from for NTRK in 1 sample. 3 patients with metastatic disease and targetable EGFR mutations received first line Osimertinib. Median PD-L1 expression across 29 samples was 7.5%. 2 patients with advanced disease and PD-L1 expression of >50% received first-line Pembrolizumab. This single centre audit demonstrates that combining local PD-L1 testing with regional GLH sequencing can effectively identify molecular targets in LUAD, although the efficiency of this pathway requires improvement. Fig. 1 (abstract 15). Showing overall survival comparing the two age groups. Fig. 1 (abstract 16).
RkJQdWJsaXNoZXIy MTk3NTQxMg==