The guidance provides recommendations on the study and evaluation of sex-specific data in medical device clinical studies
-
The purpose of the guidance is to encourage consideration of sex and related covariates (such as body size and plaque morphology) at the study design stage; to recommend study design and conduct practices that promote appropriate enrolment of each sex (e.g. where appropriate, in proportions generally representative of the demographics of the disease distribution); to outline recommended sex-specific statistical analyses of study data and provide a framework for considering sex-specific data when interpreting overall study results; and to clarify the FDA’s expectations for reporting sex-specific information in summaries and labelling of approved or cleared medical devices.
-
The guidance applies to devices that require clinical information to support a marketing application, including premarket notifications (510(k)), premarket approval (PMA) applications, De Novo classification requests, Humanitarian Device Exemption (HDE) applications, and applicable post-approval studies (PAS) and postmarket surveillance (PS) studies under Section 522 of the Federal Food, Drug, and Cosmetic Act.
Exceptions generally include: devices intended for use in only one sex (such as certain obstetrics/gynaecology and urology devices); and some in vitro diagnostic (IVD) device clinical studies using de-identified leftover specimens where demographic information such as sex cannot be obtained.
Chapter 1: Overview of the guidance
Background to the guidance, recommendations for achieving appropriate enrolment, considering sex in study design and data interpretation, and recommendations for reporting sex-specific information in submissions and public documents – in one chart:
Chapter 2: Statistical concepts for assessing heterogeneity between sexes
In the chart, heterogeneity refers to differences in outcomes between sexes, which can be detected by statistical hypothesis tests; treatment-by-sex interactions can be qualitative or quantitative. A device may perform significantly differently in women and men in terms of safety or effectiveness. Heterogeneity between sexes should be thoroughly investigated, especially for the primary safety and effectiveness endpoints. Heterogeneity here means differences in outcomes between the sexes. Statistical hypothesis tests can be performed to detect heterogeneity, and there are also statistical inference methods for estimating its magnitude.
When multiple treatment groups are considered, one form of heterogeneity is the treatment-by-sex interaction, which measures how much the difference in outcomes between treatments in one sex differs from that in the other sex. The concept of treatment-by-sex interaction applies to study endpoints involving a comparison of two treatments (such as survival probability or adverse event rates).
It is important to distinguish between qualitative and quantitative interactions. For a given parameter, a qualitative treatment-by-sex interaction is one where one treatment is better than the other in one sex but not in the other. A quantitative treatment-by-sex interaction is one where one treatment is better than the other in both sexes, but to different degrees (see figure below)
Illustration of quantitative (left) and qualitative (right) interactions
The inherent limitations of statistical tests of the “sex-by-treatment-effect interaction”, and the different handling logic in different study designs, make the relationship between sex and treatment effect complex.
Statistical tests are therefore only tools, and their results cannot directly replace clinical judgement. Study design (e.g. single-arm / controlled), test limitations (power, sample size) and other confounding factors must be taken into account to analyse the effect of sex on treatment more rigorously. This reflects the FDA’s emphasis on “patient-centred” differentiated analysis, requiring conclusions to be grounded in clinical reality rather than relying solely on statistical results.
Chapter 3: Three decision frameworks provided in the guidance
Decision framework 1: Recommendations for sex-specific statistical design
Use existing scientific data (e.g. previous studies, disease science) to determine whether there is a hypothesis of a clinically meaningful sex difference for your device. Where such a hypothesis exists, the decision tree below provides a framework for determining when various sex-specific statistical recommendations apply to different clinical study designs.
“ONE-ARM” in the figure above refers to a “single-arm study” – a study design without a control group (such as a placebo, active comparator or no-treatment control group).
Decision framework 2: Recommendations for sex-specific statistical analysis of completed studies – single-arm studies (objective performance criteria, performance goals, observational studies)
Decision framework 3: Recommendations for sex-specific statistical analysis of completed studies (comparative studies)
Chapter 4: International standards, regulatory trends and compliance recommendations
International standards and regulatory trends on sex-specific data in medical device clinical studies
Compliance recommendations on sex-specific data in medical device clinical studies
This is a series of educational articles introducing FDA, NMPA, GMP, CE MDR/IVDR and MDSAP regulations and standards. Everbiz is happy to share medical device expertise with industry clients and hopes our extensive hands-on experience in quality systems and product registration will be of help to you. Let’s contribute together to the development of the medical device industry.
More articles:
Clinical Evaluation Routes and Methods in MDR Certification
Guide to Writing MDR Technical Documentation
EU Safety Rules for Medical Device Apps on Online Platforms
Introduction to EU MDR Class IIa Products
Introduction to EU MDR Class IIb Products
How to Look Up US 21 CFR Regulations
FDA Medical Device Product Codes
Finding Core Information on Predicate Devices on the FDA Website