Assessment of the British Society of Blood and Marrow Transplantation and Cellular Therapy (BSBMTCT) registry using the Registry Evaluation and Quality Standards Tool (REQueST)
The National Institute for Health and Care Excellence (NICE) commissioned the Newcastle External Assessment Centre (EAC) to undertake an evaluation of registries to support the Cancer Drugs Fund (CDF) and Highly Specialised Technologies (HST) real world evidence development using EUnetHTA’s Registry Evaluation and Quality Standards Tool (REQueST).
The REQueST tool is a standardised set of questions contained within an Excel file which enables:
The tool comprises 3 sections:
The BSBMTCT registry is to be used to provide information to support the Managed Access Agreement (MAA) for Tisagenlecleucel (Kymriah, Novartis) for treating relapsed or refractory B-cell acute lymphoblastic leukaemia in people aged up to 25 years (see NICE Technology Assessment TA554 [published Dec 2018 – to be reviewed June 2023]). Tisagenlecleucel is a chimeric antigen receptor (CAR) T‑cell therapy. It contains the patient’s own T cells that have been modified to attach to and kill cancer cells. It is recommended for use under the CDF only if the conditions in the MAA are followed: https://www.nice.org.uk/guidance/ta554/resources/managed-access-agreement-december-2018-pdf-6651288397.
The primary sources of data collection are ongoing clinical trials: ELIANA, ENSIGN and B2101J, however a secondary source of data collection was identified as the BSBMTCT data registry which is held at Guy’s and St Thomas’ NHS Trust and collects data on all bone marrow transplants in the UK, and more recently cell therapies. This could be used to determine the rate and time to subsequent stem cell transplants.
The REQueST tool was used to assess the suitability of the registry for this purpose.
The tool was completed and returned to the EAC by Julia Lee – Head of the BSBMTCT Data Registry. The information provided was independently assessed by the EAC and NICE. A meeting was held between EAC/NICE to agree feedback and further specific questions to be asked. After the first submission, the majority of items in the tool were assessed as‘Needs development / Clarification’ (see output below).
After the second submission, the majority of items were assessed as ‘Satisfactory’ however with the information available, 5 items were assessed as ‘Not suitable’ (see output below).
The items assessed as ‘Not suitable’ are summarised below with an explanation and recommendation:
|Item 10 (Governance) Within this item, there was expected to be a statement regarding declarations of interest and how this is checked.||BSBMTCT should advise whether there are any declarations of interest. This should include as a minimum, all those on the Board or Steering Group of the registry and those involved in the data analysis. They should also describe the process in place (if one exists) for declarations of interest to be checked. If no process exists, then it is recommended that one is set up and incorporated into the registry governance. We would also expect a funding statement describing financial support for the previous and next 3 years.|
|Item 11 (Informed Consent) Q3 “Will the consent forms in use assure consent for NHSE MAA purposes?” was answered as “Yes, as is for research purposes”. However, we are unsure whether only summary data can be shared within the current consent model and are aware that linking with other datasets is not permitted as yet. There is also a query as to whether sharing with pharmaceutical companies is acceptable – this is mentioned in the consent form but is not completely explicit. In addition, we are still unsure of the process for the management of data in the absence of informed consent. The registry has not been through ethics approval but there should still be evidence of a review process having taken place for the consent forms for example by patient representatives or governance bodies of the contributing data centres.||All of these points relating to consent should be discussed further with NICE. Evidence that the consent forms have been through a review process should be provided. Assurance is needed that the consent forms will allow the data use required for the MAA. For example, it should be confirmed that the consent includes the collection of data concerning “illness and ongoing treatment” as this would enable NICE to assess whether patients in the database require subsequent stem-cell therapy. As the BSBMTCT is a research registry (without ethics approval) we would expect that each centre contributing data has performed a local data management review prior to submitting data, including approval by the Senior Information Risk Officer (SIRO), and has sent confirmation of their agreement/R&D approval to contribute data to the BSBMTCT. We would recommend that the BSBMTCT shares the R&D approval from each contributing centre with NICE.|
|Item 15 (Data Collection) It is stated that one centre (not a BSBMTCT member) does not allow registry staff to view data. Information regarding which centre this is and how many people included in the registry have been recruited via this centre is required in order to assess whether it could affect data available for the MAA. How are registry staff performing quality assurance work and analytics required by NICE/NHSE if they are unable to see all of the data?||The centre in question and the number of people recruited via this centre should be confirmed and the impact of this considered further. A discussion with NICE is needed to establish the extent to which this will affect the analyses for the MAA and whether an agreement can be obtained with the centre for the data to be made available for viewing by the registry staff.|
|Item 18 (Missing Data) More information is required in order to confirm the suitability of the registry data for the MAA: specifically relating to the statistical analysis plan (SAP) and the percentage completeness for the key fields for the MAA.||NICE should be provided with the SAP for any analyses that it requires for evidence development. The percentage completeness for the key fields for the MAA should also be provided.|
|Item 21 (Interoperability and readiness for data linkage) The data sharing agreement (DSA) between BSBMTCT and other organisations wasn’t provided.||NICE should be shown any DSA that relates to the analyses that NICE requires in order to give assurance of the contractual terms and conditions that the registry has agreed to with relevant third parties. In particular a DSA is required to link with NHSE/I data and it would also be expected that the BSBMTCT would be linking to ONS mortality data – confirmation is needed that governance is in place for this to happen. As for item 11 above, any Data Sharing Agreement should be approved by the SIRO in each centre.|
The information provided in the REQueST tool enabled 18 out of the 23 items to be assessed as ‘satisfactory’ for the purposes of the MAA. For the 5 remaining items (described above), an ‘unsuitable’ assessment was made as there was no further opportunity within this study to request further information and/or clarification. Recommendations have been made for each of the 5 items and these should be followed up between NICE and the BSBMTCT at the earliest opportunity.