Over the past years, probably the entire last decade, there have been several discussions on how to handle multiple subjects’ enrollments in CDISC data packages. Members of the CDISC SDS Multiple Subject Instances (MSI) team also shared some previews of future possible modifications of the SDTM standard to handle multiple subjects’ enrollment,[i] and we might finally have something available with the upcoming future releases of the SDTM standard and Implementation Guidance (IG).
Read more »
The Facts in the Case of Subject X
Preparing Your Integrated Summaries of Safety and Effectiveness: Best Practices
Integrated Summaries of Safety (ISS) and Integrated Summaries of Effectiveness (ISE) bring together in one place data and analyses pertinent to assess the safety and efficacy of a new drug submitted to the regulatory authorities. Let’s take a closer look at these essential documents and important steps for planning the ISS and ISE.
What are the Integrated Summaries of Safety and Effectiveness?
An Integrated Summary of Safety (ISS) and an Integrated Summary of Effectiveness (ISE) are two distinct components of a regulatory submission, often prepared by pharmaceutical companies seeking approval for a new drug or medical product.
The ISS and ISE are used to support the SCS (Module 2.7.4 – Summary of Clinical Safety) and SCE (Module 2.7.3 – Summary of Clinical Efficacy) in the CTD submission. The SCS provides a comprehensive overview of the safety profile of the drug, and is a critical component in any New Drug Application (NDA) or similar market approval request submitted to regulatory authorities such as the U.S. Food and Drug Administration (FDA) or the European Medicines Agency (EMA). It summarizes data relevant to safety in the intended population, integrating the results of individual clinical study reports as well as other reports such as the integrated analyses of safety performed in the ISS.
The SCE provides a comprehensive overview of the efficacy of the drug. It utilizes two kinds of analyses: comparison of results of individual studies, and analysis of data combined from various studies performed within the ISE.
Important considerations when preparing your ISS/ISE
1. Plan early
It is important to plan for an early discussion of your submission strategy with the relevant regulatory agencies.
2. Understand regulatory requirements
Different regulatory authorities may have specific requirements for integrated analyses and efficacy/safety summaries as well as data submissions. Familiarization with the guidelines and expectations of the relevant authorities, such as the FDA or EMA, is critical.
3. Understand when to use data pooling
It is important at the start of a submission project to decide which data, from which studies or pooled analyses, will be used in each section of the Summary of Clinical Safety and Summary of Clinical Efficacy and whether an ISE or ISS will be required. When pooled analyses are needed, a “single database” is formed by pooling the results of all concerned clinical trials. Pooled analyses are not mandatory and should only be performed if they provide additional insights beyond those observed in individual clinical trials. If pooled analyses are done, the objectives/reasons need to be explained and the validity of the pooling has to be justified.
4. Expert advice
Creating an ISS/ISE is a collaborative effort. It’s important to get expert advice from statisticians and programmers who have submission experience to help you understand the regulatory requirements and when to pool data for your specific project/drug.
5. Plan your data integration strategy
The ISS and ISE integrate data from various sources into cohesive documents. Plan for how you will combine and present data. If pooled analyses are required, data integration from relevant studies and relevant details should planned in advance in a detailed Statistical Analysis Plan (SAP), one for ISS and one for ISE. How data will be pooled can also be anticipated in pre-submission meetings, such as Type C or pre-NDA with the FDA. For example, this can be done by sharing your Study Data Standardization Plan (SDSP) where you could explain to the reviewer your planned data integration approach. Details that can be anticipated are but not limited to how you plan to handle subjects participating in more than one trial, medical dictionaries up-versioning, and so on.
6. Prepare your data submission
Health authorities such as the FDA have strict requirements with regards to submitting study data in support of market approval. Like any other “piece” submitted to the HA, submitted data packages should be of good quality. Attention should be paid to completeness of such data packages, traceability, and clarity of accompanying documentation so that the HA reviewer will be able to understand what you have done and eventually reproduce it.
Interested in learning more about data submission?
Download our complimentary new eBook, “The Good Data Submission Doctor on Data Submission and Data Integration to the FDA”:
Presenting Clinical Data for Regulatory Submission: A Stats Perspective
Data submissions are very regulated, but every drug and drug development are different. Therefore, the data presented in the Common Technical Document (CTD) needs to be tailored to the specific submission. It’s important for statisticians involved in submission projects to not only understand the guidelines, but also what makes sense in terms of data integration and pooling. What follows is an introduction to the ISS/ISE from a stats perspective, which I recently shared at the Italian CDISC User Network on May 12, 2023.
Raising the Awareness for Additional FDA Data Submission Recommendations (Part I)
For years CDISC data standards implementers have struggled to find good implementation examples and use cases beside those provided in the CDISC Implementation Guidance (IG). However, in the recent years, thanks to the efforts of several different experts, such as clinicians in the different Therapeutic Areas or data standards experts, several CDISC Therapeutic Area User Guidance (TAUG) have been made available [1, 2, 3]. As I write this blog, 46 TAUGs are available from various therapeutic areas, such as Oncology, Neurology, Endocrinology, Cardiovascular, Infective Diseases, Autoimmune Diseases, Mental Health, Gastrointestinal, and others. The use of and adherence to many of these TAUGs is recommended by the FDA in its Study Data Technical Conformance Guidance (SDTCG) [4], and as per the March-2022 version, 24 TAUGs are “supported”.
Moreover, regulatory agencies such as the FDA have also released some additional guidances either specific to an FDA division or covering any specific medical aspects, with the aim of reducing the variability in interpretation of CDISC IG. In a previous blog , I discussed the detailed ADaM specifications provided by the FDA Office of Oncologic Diseases in support of Real Time Oncology Review [5].
Metadata Repositories: Overcoming Challenges with Automation
Written by Angelo Tinazzi, Nicolas Rouillé, and Sebastià Barceló
In the realm of standards management, companies of all sizes are increasingly exploring the potential of metadata repositories (MDR). From protocol development to eCRF, SDTM, and ADaM to Analysis Results, these repositories are being used to speed up study set up and delivery. Sponsors leverage metadata using a framework that involves putting together a data governance team that establishes and pilots company standards, defines roles, establishes workflows, develops standard operating procedures, and provides necessary training. This structured framework is supported by selecting or building a metadata repository that aligns with the established infrastructure.
However, unlike sponsors, CROs encounter specific challenges when implementing standards management or metadata use, given that each sponsor has unique processes and not all aspects of clinical trial are always managed by a single CRO.
Here, we share a new approach to automation to address the challenges inherent in metadata repositories.
Mind the Health Gap: Is It Time to Reliably Measure the Impact of Health Inequity in Product Development and Assessment? Yes, It Is.
The spotlight for this year’s World Evidence-Based Healthcare Day (EBHC) (October 20, 2023), is on the generation and application of evidence to improve global health equity.
According to the World Health Organization (WHO), “health equity is the absence of unfair, avoidable or remediable differences in health among population groups, defined by social, economic, demographic or geographic characteristics.”
Health inequity (or disparity) is not a new health topic; it has been widely documented across the world for the last four decades. What is new are the recent efforts to make health equity a strategic priority in healthcare decision-making. A couple of things are clearer now more than ever — 1) there is a need to
Negative Binomial Distribution in Group Sequential Designs
In clinical trials based on count data, the aim is to compare independent treatment groups in terms of the rate of occurrence of a particular outcome, such as number of times a subject responds to a therapy, develops a certain adverse experience, requires specialized care, or takes medication to achieve a particular response — for example, the number of migraines, seizures, recurrent infections, hospitalizations, episodes of diarrhea, and so on.
Negative binomial probability distribution can be used to model the number of times a particular outcome occurs during a clinical trial. Here, I explain this statistical methodology and its application in adaptive group sequential clinical trial designs.
New FDA Data Submission Requirements and Substantial Changes
Ten years ago this month, in January 2014, the FDA issued the first version of its Technical Conformance Guide (by chance I also found comments provided by my friend Jozef Aerts already complaining about the use of SAS XPORT). The following December, a second version was released after public comments and, on December 16, 2014, the FDA stopped the clock, providing sponsors with a pivotal two-year window to adapt their methods of creating clinical dataset packages to comply with the FDA’s new required data standards for any study commencing after December 16, 2016.
Fast forward through approximately 30 subsequent versions, with the latest version, 5.6, being released in December 2023. The guidance has undergone significant changes, not only in terms of length (from 38 to 88 pages), but also in its content, consequently impacting the requirements for sponsors set forth by the FDA.
More specifically, between 2021 and 2023, the FDA released 13 versions. If you find yourself fatigued from comparing the differences, fear not! In this post, I aim to simplify things for you by highlighting the most substantial changes or new requirements.
Latest Updates: December 2023 Version
Version 5.6 was released while I was completing this article. The major update is in section 5.3, where the FDA lists additional FDA Technical Specification Documents. This includes two new guidances with SDTM and ADaM recommendations: “Submitting Clinical Trial Datasets and Documentation for Clinical Outcome Assessments (COA) Using Item Response Theory” and “Submitting Patient-Reported Outcome (PRO) Data in Cancer Clinical Trials,” which contains analysis display examples.
October 2023 Version Updates
Aside from the text additions to enhance clarity in certain sections and various alterations related to SEND (which won’t be discussed here), here are the most significant changes I have identified:
- In section 4.1.1.3, there are now specific requirements for the PC and PP domains. The agency emphasizes that in both datasets, such as when referencing timepoints like visits, consistency with other domains is crucial. Additionally, there should be coherence between the two datasets when referencing analyte names. PCLLOQ should now also include a lower limit of quantitation.
- Moreover, in section 4.1.1.3, the FDA establishes new requirements for the SV domain, necessitating the inclusion of all scheduled visits, regardless of their occurrence. Additionally, the FDA advocates for the incorporation of new variables that are now standard in SDTM version 2.0 / IG 3.4, namely SVREASOC (Reason for Occur Value), SVEPCHGI (Epi/Pandemic Related Change Indicator), and SVCNTMOD (Contact Mode). The agency strongly prefers these variables to be included in SV and not in SUPPSV, irrespective of the SDTM version in use. Consequently, the VE domain, initially proposed in the CDISC “Guidance for Ongoing Studies Disrupted by COVID-19 Pandemic,” is no longer recommended.
- Appendix C lists expected FDA parameters to be included in TS. I strongly recommend regularly assessing any differences and updating your data standards accordingly.
- In Appendix D, the agency provides a list of additional documents not found in the data standards catalogue but encouraged by the FDA. These include the ADaM OCCDS, ADaM Examples of Traceability, ADaM Metadata Submission Guideline v1.0 (note that version 2 has been available since 20214), and “CDISC Documents: Interim User Guide for COVID-19” (of note, last September, CDISC released a consolidated version5) and the “CDISC Guidance for Ongoing Studies Disrupted by COVID-19.”
June 2023 Version Updates
In the June 2023 version, a significant change was introduced. Notably, the FDA mandated the submission of an additional custom laboratory dataset. This dataset should mirror the structure of LB, with standard results variables (LBSTRESU, LBSTRESC, and LBSTRESN) along with corresponding normal range variables, containing results with US Conventional Unit instead of Standard International Unit (SI).
March 2023 Version Updates
In the March 2023 version, the FDA introduced in section “4.1.1.3 – SDTM Domain Specifications” specific requirements for submitting Immunogenicity data through the IS domain.
March 2022 Version Updates
In the March 2022 version, the FDA formally references the requirement to adhere to additional FDA Technical Specification Documents.
This version also introduces a new section, “7.2 Electronic File Directory,” providing instructions for sponsors on organizing study datasets and their supporting files within a designated eCTD directory. Within this section, the FDA explicitly clarifies that when datasets are placed in the /split folder, there is no requirement for a second define.xml.
Moreover, this version set the official go-live of the “FDA Technical Rejection Criteria.”
Other Significant Changes Worth Remembering
It’s been quite some time since October 2018 when the FDA, in section 4.1.1.3, introduced additional requirements regarding the handling of subjects with multiple enrollments in SDTM. Specifically, there was a directive to submit the primary enrollment in the DM domain and any additional enrollments in a demographic custom domain structured similarly to DM. Despite the passage of time, no definitive guidelines have been issued, even with the latest SDTM version (anticipated in SDTM Ig 4.0). Nevertheless, it has become increasingly common practice to submit a custom domain named DC, denoting Demographics as Collected. Moreover, for studies with multiple screenings and/or multiple enrollments per subject, the guidance recommends the inclusion of SUBJID in other related domains besides DM even though it may cause validation errors.
Interested in learning more?
Download Angelo Tinazzi’s new ebook, “The Good Data Submission Doctor on Data Submission and Data Integration to the FDA”:
Looking to the Future — Improving Diagnosis and Prognosis of Eye Conditions with Artificial Intelligence
Written by Alind Gupta, Cytel; Haridarshan Patel, Horizon Therapeutics; and Jason Simeone, Cytel
Ophthalmology is well-suited to using artificial intelligence (AI) methods because clinical decisions often rely on complex data-rich information from medical images of the eye and patient health status. AI has revolutionized image data analysis over the last decade and is promising to improve healthcare delivery and clinical decision-making while reducing healthcare costs. In fact, some AI-based diagnostic platforms for early detection of diabetic retinopathy have received FDA clearance and have been introduced in resource-constrained healthcare settings worldwide to screen for early signs of disease and to accelerate patient access to correct therapies. Read more »
Insights on the New ADaM guidelines and Europe Interchange 2022
I am excited to see you all at the CDISC Europe Interchange, April 27 – 28 but unfortunately, it will be a virtual event (hopefully, for the last time). The program designed by the CDISC Europe Committee and the CDISC team looks promising as always! Silvia Faini and I (members of the committee) will lead the “Tech-Enabled Standards” and “ADaM” streams, respectively.