A comparison of the effectiveness of different treatment regimens for pancreatic cancer using English cancer registry data

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Objective for processing

The University of Sheffield request the continued retention and processing of require National Disease Registration (NDRS) Data from NHS England for the following research [6 words unchanged] of different treatment regimens for pancreatic cancer using English cancer registry data” [24 paragraphs unchanged] Yorkshire Cancer Research is the funder and provides funding via a fellowship which runs until 2024. Beyond this time the core funding is provided via the University of Sheffield. funder. The funders will have no ability to suppress or otherwise limit the publication of findings. [1 paragraph unchanged] In line with the national data opt-out policy, opt-outs are not applied because the data is not Confidential Patient Information as defined in section 251(10) and section 251(11) of the National Health Service Act 2006. Where individuals have opted out of disease registration by the National Disease Registration Service (NDRS), their data has been permanently removed from the registry and therefore will not be disseminated under this Data Sharing Agreement (DSA). https://digital.nhs.uk/ndrs/patients/opting-out.

Expected output

Update as per latest confirmation report submitted 22/12/2023 [9 paragraphs unchanged]

Benefits reported

The work The the principle investigator is leading on, of which this project is part, has [10 words unchanged] in health technology assessment. Several related projects have now been initiated, and publication of the principal investigators input to the NICE Real World Evidence Framework Framework, and published methods papers, is significant. Through seminars and presentations nationally and internationally the principle investigator has disseminated information on planned research and also on methodological developments that will inform subsequent analyses undertaken using NHS data. The analyses that have been conducted represent the first set of Target Trial replication studies undertaken using NHS cancer registry data, linked to SACT and HES data. All analyses have now been completed. Ongoing data access during the publication process is necessary, in case any amendments to the analyses are required in response to peer review.

Unchanged: Processing activities, Expected measurable benefits.

Text removed is struck through; text added is underlined. Unchanged paragraphs are summarised rather than repeated.

Objective for processing

The University of Sheffield request the continued retention and processing of National Disease Registration (NDRS) Data from NHS England for the following research project “A comparison of the effectiveness of different treatment regimens for pancreatic cancer using English cancer registry data” [1 paragraph unchanged] National Cancer Disease Registration and Analysis Service (NCRAS) data will be used to replicate clinical trials that have already been [19 words unchanged] from the trial-based analyses. If the results are similar, this suggests that registry data may be sufficient for comparing the effectiveness of different cancer treatments. registry data may be sufficient for comparing the effectiveness of different cancer treatments. This is an important first step in showing whether registry data can be relied upon to compare the effectiveness of different cancer treatments. If it can, researchers could more confidently use registry data to compare the effectiveness of different treatments in real world populations – going beyond the highly selected patient groups usually included in This is an important first step in showing whether registry data can be relied upon to compare the effectiveness of different cancer treatments. If it can, we can use registry data to compare the effectiveness of different treatments in real world populations – going beyond the highly selected patient groups usually included in [1 paragraph unchanged] If it is not possible to successfully replicate clinical trial results using [14 words unchanged] that effectiveness is different in the real world compared to in trials. We University of Sheffield will investigate this and if there are problems with the data we they will identify areas where data collection needs to be improved in order for registry data to be most useful. The University of Sheffield have identified pancreatic cancer as a suitable disease area for undertaking Target Trial analyses using NDRS data. In the following section University of Sheffield justify this choice, and provide background information on pancreatic cancer and treatment options in England. The University of Sheffield then specify 4 Target Trial analyses that they will undertake. Finally, University of Sheffield specify the NDRS data required to facilitate these analyses. Pancreatic Cancer In 2016, approximately 10,000 people were diagnosed with pancreatic cancer in the United Kingdom, and often pancreatic cancer is diagnosed at an advanced stage. The prognosis is poor even for people diagnosed at an early stage of pancreatic cancer, where surgical resection is possible, with 5-year survival rates estimated at between 7% and 25%. Survival rates are extremely poor for people with metastatic disease, with median survival of between 2 and 6 months if untreated. A NICE Guideline on the diagnosis and management of pancreatic cancer, published in 2018, recommends that gemcitabine plus capecitabine should be offered as adjuvant treatment for people who have had sufficient time to recover after pancreatic cancer resection. Gemcitabine monotherapy should be considered for people who are not well enough to tolerate combination chemotherapy. FOLFIRINOX, a combination regimen consisting of oxaliplatin, in rinotecan, leucovorin and fluorouracil, is not mentioned in the NICE guideline, but is beginning to be offered as adjuvant treatment in the NHS, due to trial results published in December 2018. For metastatic pancreatic cancer the NICE guideline recommends that FOLFIRINOX should be offered to people with an Eastern Cooperative Oncology Group (ECOG) performance status of 0-1. Gemcitabine combination therapy should be considered for people not well enough to tolerate FOLFIRINOX, with the first combination option being gemcitabine plus capecitabine. For people for whom FOLFIRINOX and gemcitabine plus capecitabine are unsuitable gemcitabine plus nab-paclitaxel is an option.[9]Gemcitabine monotherapy should be offered to people not well enough to tolerate combination chemotherapy. These guidelines seem to present a clear hierarchy of treatments for adjuvant and metastatic pancreatic cancer, and seem to suggest that there might be little overlap in prognostic characteristics of patients receiving different treatments. However, the NICE technology appraisal of gemcitabine plus nab-paclitaxel notes that some patients for whom FOLFIRINOX is otherwise suitable choose not to have this treatment because of its considerable toxicity. Further, it is noted that the current treatment options have a number of limitations, including serious adverse effects –in particular, the most effective treatment option (FOLFIRINOX) is associated with the most significant adverse events, whereas the least effective (gemcitabine monotherapy) is associated with the least significant adverse events. In addition, it is unfortunately the case that prognosis remains poor even with the most effective treatment. Therefore, it is likely that due to patient choice, there will be overlap in prognostic characteristics between patients who receive FOLFIRINOX and patients who receive gemcitabine for metastatic pancreatic cancer. Similarly, because gemcitabine combination therapies have lower effectiveness and toxicity than FOLFIRINOX, and higher effectiveness and toxicity than gemcitabine monotherapy, it is likely that there is some overlap in prognostic characteristics between patients who receive FOLFIRINOX, gemcitabine combination therapies, and gemcitabine monotherapy. The NICE technology appraisal guidance for gemcitabine plus nab-paclitaxel states that there is evidence of use of gemcitabine doublet chemotherapy for pancreatic cancer in the NHS. Similar is likely to be true for adjuvant treatment for pancreatic cancer, where gemcitabine plus capecitabine is more effective than gemcitabine monotherapy, but where toxicity is lower for the monotherapy option and prognosis is relatively poor with both treatment options. Hence, it is likely that there is variation in treatments received for adjuvant and metastatic pancreatic cancer in the NHS, with an overlap in characteristics of patients receiving different treatments. This echoes clinical expert opinion from Professor Jonathan Wadsley, who states that for both adjuvant and metastatic pancreatic cancer there is substantial overlap between patients receiving different treatments. For adjuvant treatment, Professor Wadsley believes that due to the additional side effects and limited increase in effectiveness associated with combination treatment, some patients choose gemcitabine monotherapy instead of gemcitabine plus capecitabine, and in fact some patients choose no treatment at all. For metastatic disease, Professor Wadsley believes that treatment with gemcitabine monotherapy remains common, with people choosing it instead of the highly toxic FOLFIRINOX regimen, whilst some patients receive gemcitabine combination therapy. To be able to compare the effectiveness of different treatment options in registry data there needs to be some overlap in prognostic characteristics between patients receiving the different treatments. Based on statements made by clinical and patient experts in NICE technology appraisal documents and information from a practicing clinician who treats people with pancreatic cancer, University of Sheffield are confident that such overlap exists for the treatment of both adjuvant and metastatic pancreatic cancer in the NHS. Target Trial Analyses University of Sheffield have identified four pancreatic cancer trials that the study team will try to replicate using NDRS data, using Hernan and Robins’ Target Trial framework. The details of these analyses, under the headings used by Hernan and Robins, are presented for each Target Trial. For each Target Trial, two sets of analyses will be completed. One set of analyses will be undertaken whereby the population analysed will match that included in the RCT being emulated as closely as possible, based on the eligibility criteria of the RCT. These analyses will be compared to the RCT results, allowing us to determine whether or not it has been possible to successfully emulate the RCT. A second set of analyses will be undertaken for a broader real world population, without applying the same eligibility criteria used in the associated RCT. For example, in Target Trial 1, the ESPAC-4 RCT included strict eligibility criteria (shown in the Table below). In the studys first set of analyses University of Sheffield will attempt to replicate the trial population using these eligibility criteria. In the second set of analyses the study will not use these eligibility criteria, instead analysing the effectiveness of gemcitabine monotherapy compared to gemcitabine plus capecitabine in any patient aged 18 or older who had adjuvant pancreatic cancer and received either of these regimens. The second set of analyses will allow the study to estimate the effectiveness of treatment in a more general real world population. Note that for all main analyses the minimum follow-up time used in the Target Trials will match that used in the trials being emulated. However, to make full use of available data, supplementary analyses will be conducted that do not include a minimum follow-up time. In order to complete the above objectives, University of Sheffield request the retention and continued processing of the following National Disease Registration (NDRS) Datasets for a further 2 years. -NDRS Cancer Registrations (diagnoses from 2nd April 2012 to 31st Dec 2018, with the latest follow-up (for death) date being 16th April 2021) - NDRS Linked Linked Hospital Episode Statistics A&E (data spans spans 4th April 2012 to 31st March 2019) -NDRS Linked Linked Hospital Episode Statistics Admitted Patient Care (data spans 2nd April 2012 to 31st March 2019) -NDRS Linked Hospital Episode Statistics Outpatients (spans 5th April 2012 to 31st March 2019) -NDRS Radiotherapy (RTDS) (data spans 5th April 2012 to 31st March 2019) -NDRS Systemic Anti Cancer (SACT) (data spans 4th April 2012 to 31st October 2020) The level of data is pseudonymised and minimised according to the following criteria: Diagnosis code C25x and limited to patients who received some kind of systemic anti-cancer therapy for their pancreatic cancer. Patients who were diagnosed with pancreatic cancer but did not receive systemic anti-cancer therapy were excluded from the data extract. Previous data disseminated was between the years 2012-2019 (with the latest follow-up for death date being 16th April 2021). University of Sheffield is the research sponsor and the controller as the organisation responsible for ensuring that the data will only be processed for the purpose described above. The lawful basis for processing personal data under the UK GDPR is: Article 6(1)(e) - processing is necessary for the performance of a task carried out in the public interest or in the exercise of official authority vested in the controller. The lawful basis for processing special category data under the UK GDPR is: Article 9(2)(j) - processing is necessary for archiving purposes in the public interest, scientific or historical research purposes or statistical purposes in accordance with Article 89(1) based on Union or Member State law which shall be proportionate to the aim pursued, respect the essence of the right to data protection and provide for suitable and specific measures to safeguard the fundamental rights and the interests of the data subject. All data will be processed by substantive employees of the University of Sheffield and a registered PhD student affiliated with the University of Sheffield, all individuals processing the data will receive appropriate training in data protection and confidentiality. Yorkshire Cancer Research is the funder and provides funding via a fellowship which runs until 2024. Beyond this time the core funding is provided via the University of Sheffield. The funders will have no ability to suppress or otherwise limit the publication of findings. The research described in this document is part of a broader programme of research being conducted at the University of Sheffield, investigating the use of cancer registry data to inform healthcare decision making. As part of this work, researchers consult with patient and public involvement and engagement (PPIE) representatives on an ongoing basis. This consultation has informed the planning of the research described here and related projects being undertaken by the research team in Sheffield, and PPIE input will also be crucial when results are disseminated. PPIE input to these projects is particularly relevant and important, because the research is about data routinely collected from cancer patients, its reliability, whether improved or adapted data collection is required, and the findings of this research will have important implications for patients and the public. In line with the national data opt-out policy, opt-outs are not applied because the data is not Confidential Patient Information as defined in section 251(10) and section 251(11) of the National Health Service Act 2006. Where individuals have opted out of disease registration by the National Disease Registration Service (NDRS), their data has been permanently removed from the registry and therefore will not be disseminated under this Data Sharing Agreement (DSA). https://digital.nhs.uk/ndrs/patients/opting-out.

Processing activities

The University of Sheffield have identified pancreatic cancer as a suitable disease area for undertaking Target Trial analyses using NCRAS data. In the following section we justify this choice, and provide background information on pancreatic cancer and treatment options in England. We then specify 4 Target Trial analyses that we will undertake. Finally we specify the NCRAS data required to facilitate these analyses. Pancreatic Cancer In 2016, approximately 10,000 people were diagnosed with pancreatic cancer in the United Kingdom, and often pancreatic cancer is diagnosed at an advanced stage. Data historically flowed from NDRS to the University of Sheffield. There will be no further flows of data under this agreement. The prognosis is poor even for people diagnosed at an early stage of pancreatic cancer, where surgical resection is possible, with 5-year survival rates estimated at between 7% and 25%. Survival rates are extremely poor for people with metastatic disease, with median survival of between 2 and 6 months if untreated. A NICE Guideline on the diagnosis and management of pancreatic cancer, published in 2018,recommends that gemcitabine plus capecitabine should be offered as adjuvant treatment for people who have had sufficient time to recover after pancreatic cancer resection. Gemcitabine monotherapy should be considered for people who are not well enough to tolerate combination chemotherapy. FOLFIRINOX, a combination regimen consisting of oxaliplatin, in rinotecan, leucovorin and fluorouracil, is not mentioned in the NICE guideline, but is beginning to be offered as adjuvant treatment in the NHS, due to trial results published in December 2018. The data will be stored securely on centrally provisioned University of Sheffield virtual servers and research data storage infrastructure as Stata datasets for the term of the agreement. Access control is by authorised University computer account username and password. Off-site access is facilitated by secure VPN connection authenticated by university username and remote password. Copies of data are kept in two physical data centres on-campus with a retention period of 7 days of snapshots. Both of these data centres are secured requiring access through multiple doors and shutters, there is also swipe access. Both data centres also have CCTV coverage. The third site where backups are located for 28 days is in a tier 3 data centre in Manchester, the data is encrypted while in transit between the local and remote data centre and the data is encrypted at rest at the remote data centre. Only IT Services Server & Storage admins have administrative access to these systems, no third party has any logical access to the backup storage systems. This system is maintained by the University’s IT Services, and is updated and amended over time. Further information regarding university server storage security, backup and recovery is available from IT Services. University of Sheffield will comply with the Data Protection Act and the University's own Information Security and Data Protection Policies as well as the Division of Population Health Information Governance Policy. Because the data will be de-personalised rather than completely anonymous data will not be placed in a repository or made publicly available. On or before the effective date of termination or End Date of the data sharing agreement, the data provided will be securely and permanently destroyed or erased such that it cannot be recovered or reconstructed, together with all hard or soft copies of the manipulated or derived data generated from the data. For metastatic pancreatic cancer the NICE guideline recommends that FOLFIRINOX should be offered to people with an Eastern Cooperative Oncology Group (ECOG) performance status of 0-1. Gemcitabine combination therapy should be considered for people not well enough to tolerate FOLFIRINOX, with the first combination option being gemcitabine plus capecitabine. For people for whom FOLFIRINOX and gemcitabine plus capecitabine are unsuitable gemcitabine plus nab-paclitaxel is an option.[9]Gemcitabine monotherapy should be offered to people not well enough to tolerate combination chemotherapy. These guidelines seem to present a clear hierarchy of treatments for adjuvant and metastatic pancreatic cancer, and seem to suggest that there might be little overlap in prognostic characteristics of patients receiving different treatments. However, the NICE technology appraisal of gemcitabine plus nab-paclitaxel notes that some patients for whom FOLFIRINOX is otherwise suitable choose not to have this treatment because of its considerable toxicity. Further, it is noted that the current treatment options have a number of limitations, including serious adverse effects –in particular, the most effective treatment option(FOLFIRINOX) is associated with the most significant adverse events, whereas the least effective (gemcitabine monotherapy) is associated with the least significant adverse events. In addition, it is unfortunately the case that prognosis remains poor even with the most effective treatment. Therefore, it is likely that due to patient choice, there will be overlap in prognostic characteristics between patients who receive FOLFIRINOX and patients who receive gemcitabine for metastatic pancreatic cancer. Similarly, because gemcitabine combination therapies have lower effectiveness and toxicity than FOLFIRINOX, and higher effectiveness and toxicity than gemcitabine monotherapy, it is likely that there is some overlap in prognostic characteristics between patients who receive FOLFIRINOX, gemcitabine combination therapies, and gemcitabine monotherapy. The NICE technology appraisal guidance for gemcitabine plus nab-paclitaxel states that there is evidence of use of gemcitabine doublet chemotherapy for pancreatic cancer in the NHS. In order to allow the analyses conducted during this study to be re-produced, detailed information regarding the exact data received and the programming code used to analyse will be recorded and made publicly available. This would allow an interested party to request the same extract of data from NDRS, and to re-produce the analyses. The data will be analysed in Stata by the study team at the University of Sheffield to estimate the comparative effectiveness of treatments for pancreatic cancer, as described above. All analyses will be documented in Stata .do files. The principle investigator will be responsible for implementing the data management plan, and ensuring it is reviewed and revised if required. Similar is likely to be true for adjuvant treatment for pancreatic cancer, where gemcitabine plus capecitabine is more effective than gemcitabine monotherapy, but where toxicity is lower for the monotherapy option and prognosis is relatively poor with both treatment options. Hence, it is likely that there is variation in treatments received for adjuvant and metastatic pancreatic cancer in the NHS, with an overlap in characteristics of patients receiving different treatments. This echoes clinical expert opinion from Professor Jonathan Wadsley, who states that for both adjuvant and metastatic pancreatic cancer there is substantial overlap between patients receiving different treatments. For adjuvant treatment, Professor Wadsley believes that due to the additional side effects and limited increase in effectiveness associated with combination treatment, some patients choose gemcitabine monotherapy instead of gemcitabine plus capecitabine, and in fact some patients choose no treatment at all. For metastatic disease, Professor Wadsley believes that treatment with gemcitabine monotherapy remains common, with people choosing it instead of the highly toxic FOLFIRINOX regimen, whilst some patients receive gemcitabine combination therapy. To be able to compare the effectiveness of different treatment options in registry data there needs to be some overlap in prognostic characteristics between patients receiving the different treatments. Based on statements made by clinical and patient experts in NICE technology appraisal documents and information from a practicing clinician who treats people with pancreatic cancer, we are confident that such overlap exists for the treatment of both adjuvant and metastatic pancreatic cancer in the NHS. Target Trial Analyses We have identified four pancreatic cancer trials that we will try to replicate using NCRAS data, using Hernan and Robins’ Target Trial framework. The details of these analyses, under the headings used by Hernan and Robins, are presented for each Target Trial in the following four tables. For each Target Trial, two sets of analyses will be completed. One set of analyses will be undertaken whereby the population analysed will match that included in the RCT being emulated as closely as possible, based on the eligibility criteria of the RCT. These analyses will be compared to the RCT results, allowing us to determine whether or not it has been possible to successfully emulate the RCT. A second set of analyses will be undertaken for a broader real world population, without applying the same eligibility criteria used in the associated RCT. For example, in Target Trial 1, the ESPAC-4 RCT included strict eligibility criteria (shown in the Table below). In our first set of analyses we will attempt to replicate the trial population using these eligibility criteria. In our second set of analyses we will not use these eligibility criteria, instead analysing the effectiveness of gemcitabine monotherapy compared to gemcitabine plus capecitabine in any patient aged 18 or older who had adjuvant pancreatic cancer and received either of these regimens. The second set of analyses will allow us to estimate the effectiveness of treatment in a more general real world population. Note that for all main analyses the minimum follow-up time used in the Target Trials will match that used in the trials being emulated. However, to make full use of available data, supplementary analyses will be conducted that do not include a minimum follow-up time. The Data will not be transferred to any other location, not be linked with any other data and will not leave England/Wales at any time. There will be no requirement and no attempt to reidentify individuals when using the Data. The data will be stored securely on centrally provisioned University of Sheffield virtual servers and research data storage infrastructure as Stata datasets for a period of two years. Access control is by authorised University computer account username and password. Off-site access is facilitated by secure VPN connection authenticated by University username and remote password. By default, two copies of data are kept across two physical plant rooms, with a 28 day snapshot made of data and backed up securely offsite at least daily. This service is maintained by the University’s Corporate Information and Computing Services. We will comply with the Data Protection Act and the University's own Information Security and Data Protection Policies as well as the School of Health and Related Research (ScHARR) Information Governance Policy. Because the data will be de-personalised rather than completely anonymous data will not be placed in a repository or made publicly available. On or before the effective date of termination or End Date of the data sharing agreement (expected to be 2 years after data receipt), the data provided will be securely and permanently destroyed or erased such that it cannot be recovered or reconstructed, together with all hard or soft copies of the manipulated or derived data generated from the data. The Data will be accessed onsite at the premises of the University of Sheffield and via remote access. In order to allow the analyses conducted during this study to be re-produced detailed information regarding the exact data extract received and the programming code used to analyse it will be recorded and made publicly available. This would allow an interested party to request the same extract of data from ODR, and to re-produce the analyses. The data will be analysed in Stata by Dr Nicholas Latimer to estimate the comparative effectiveness of treatments for pancreatic cancer, as described above. All analyses will be documented in Stata .do files. Dr Nicholas Latimer will be responsible for implementing the data management plan, and ensuring it is reviewed and revised if required. ODR operate a cost recovery framework, and charge for the time taken to provide the data extract. For remote access: All data will be processed by substantive employees and a registered PhD student of the University Of Sheffield, all individuals processing the data will receive appropriate training in data protection and confidentiality. The research team and student will comply with the Data Protection Act and University Of Sheffield’s own Research Degree Students Code of Practice, University Of Sheffield’s Misconduct Regulations, University Of Sheffield Information Security Policy and Data protection policy. The listed data controller accepts liability for any action of the PhD student in relation to NHSD data. - Remote access will only be from secure locations situated within the territory of use (as further restricted elsewhere within the DSA if so done) stated within this DSA; - Access controls granting users the minimum level of access required are in place; - Remote access is only via secure connections (e.g., VPNs or secure protocols) to protect data; - Multifactor authentication (MFA) is required for remote access; - Device security, including up-to-date software and operating systems, antivirus software, and enabled firewalls are utilised for the remote access; - All remote access is undertaken within the scope of the organisation’s DSPT (or other security arrangements as per this agreement) and complies with the organisation’s Ensuring Data Security When Working Remotely policy. The above applies in addition to any condition set out elsewhere within the DSA (e.g. who may carry out processing, and for what purpose). All data access is restricted too and will be processed by substantive employees and a registered PhD student of the University Of Sheffield, all individuals processing the data will receive appropriate training in data protection and confidentiality. The research team and student will comply with the Data Protection Act and University Of Sheffield’s own Research Degree Students Code of Practice, University Of Sheffield’s Misconduct Regulations, University Of Sheffield Information Security Policy and Data protection policy. The listed data controller accepts liability for any action of the PhD student in relation to NHS England data. Analysts and researchers from the University of Sheffield will process the Data for the purposes described above

Expected output

I have co-authored the following papers in this area since applying for these data: Update as per latest confirmation report submitted 22/12/2023 - Gomes M, Latimer N, Soares M, Dias S, Baio G, Freemantle N, Dawoud D, Wailoo A, Grieve R. Target trial emulation for transparent and robust estimation of treatment effects for health technology assessment using real-world data: opportunities and challenges. PharmacoEconomics. Published online 25 Mar 2022. Since gaining access to the datasets supplied by NHS England, several other projects have combined to develop a research programme, led by Prof Nicholas Latimer, the principle investigator responsible for the research described in this report. In particular, two PhD students began their studies at the University of Sheffield under the supervision of the principle investigator responsible, with the shared goal of investigating statistical methods for estimating the comparative effectiveness of cancer treatments using real world data sources. These students have each made their own successful applications for data from NHS England. The principle investigator has also been given the role of theme lead for the University of Sheffield's cancer research strategy (Theme 1, focusing on epidemiology, screening and early diagnosis, utilising big data sources), and has led a proposal for a 5-year research programme further investigating the use of cancer registry datasets to inform the health technology assessment process and the development of targeted interventions. The principle investigator has presented on the research being conducted by this research team at numerous seminars nationally and internationally. - Gray J, Sullivan T, Latimer NR, Salter A, Sorich MJ, Ward RL, Karnon J. Extrapolation of survival curves using standard parametric models and flexible parametric spline models: Comparisons in Large Registry Cohorts with Advanced Cancer. Medical Decision Making 2021 Feb;41(2):179-193. The principle investigator also recently contributed to the development of the National Institute for Health and Care Excellence (NICE) Real World Evidence Framework, published in 2022. This framework solidifies the role of real world data in healthcare decision making, and makes the findings of the research described here even more important: we need to know whether comparative effectiveness analyses that use English cancer registry data can be reliably used to inform healthcare decision making in the UK. The analyses of the pancreatic cancer registry data that is the subject of this report remains a key part of this research programme: these analyses will make an important and valuable contribution to understanding in this area, which will directly influence the use of these data sources by healthcare decision makers. I have also taken on two PhD students to investigate methods for estimating the comparative effectiveness of cancer treatments using real world data sources, who have each made their own applications for data from NHS England. I have also been given the role of theme lead for the University of Sheffield's cancer research strategy, focusing on epidemiology, screening and early diagnosis, utilising big data sources. I have led a proposal for a 5-year research programme (currently under review) further investigating the use of cancer registry datasets to inform the health technology assessment process and the development of targeted interventions. I have presented a plan of the research that my PhD students and I seek to deliver in numerous seminars nationally and internationally. I also contributed to the development of the National Institute for Health and Care Excellence (NICE) Real World Evidence Framework, published earlier this year. Since gaining access to these data, the principle investigator has co-authored the following papers in the area of target trial emulation and survival analysis (though none of these refer directly to the pancreatic cancer analyses being undertaken because those analyses are still in process): The analysis of the pancreatic cancer registry data that is the subject of this data access extension request remains a key part of the research programme, but conducting the analyses has been delayed due to the time commitments associated with the work described above. Hence my request for an extension. The work described above helps demonstrate the growing importance of comparative effectiveness analysis using real world data sources, and hence my planned analyses will make an important and valuable contribution to understanding in this area, which will directly influence the use of these data sources by health care decision makers. - Gomes M, Latimer N, Soares M, Dias S, Baio G, Freemantle N, Dawoud D, Wailoo A, Grieve R. Target trial emulation for transparent and robust estimation of treatment effects for health technology assessment using real world data: opportunities and challenges. PharmacoEconomics. Published online 25 Mar 2022. - Lee S, Lambert PC, Sweeting MJ, Latimer NR, Rurtherford MJ. Evaluation of Flexible Parametric Relative Survival Approaches for Enforcing Long-Term Constraints When Extrapolating All-Cause Survival. Value in Health, 2023 Oct 17. - Sweeting MJ, Rutherford MJ, Jackson D, Lee S, Latimer NR, Hettle R, Lambert PC. Survival extrapolation incorporating general population mortality using excess hazard and cure models: a tutorial. Medical Decision Making, 2023; 43(6):737-748. - Gray J, Sullivan T, Latimer NR, Salter A, Sorich MJ, Ward RL, Karnon J. Extrapolation of survival curves using standard parametric models and flexible parametric spline models: Comparisons in Large Registry Cohorts with Advanced Cancer. Medical Decision Making 2021 Feb;41(2):179-193 The analysis of the pancreatic cancer registry data that is the subject of this data access extension request remains a key part of the research programme, but conducting the analyses has been delayed due to the time commitments associated with the work described above. Hence the request for an extension. The work described above helps demonstrate the growing importance of comparative effectiveness analysis using real world data sources, and hence the planned analyses will make an important and valuable contribution to understanding in this area, which will directly influence the use of these data sources by health care decision makers. The outputs will not contain NHS England data and will only contain aggregated information with small numbers suppressed as appropriate in line with the relevant disclosure rules for the dataset(s) from which the information was derived.

Benefits reported

The work I am leading, The principle investigator is leading on, of which this project is part, has helped drive progress in the [22 words unchanged] World Evidence Framework is significant. Through seminars and presentations nationally and internationally I have the principle investigator has disseminated information on planned research and also on methodological developments that will inform subsequent analyses.

Unchanged: Expected measurable benefits.