FDA SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED) CLINICAL SECTION CHECKLIST OFFICE OF DEVICE EVALUATION Note to FDA PMA Reviewers: The Summary of Safety and Effectiveness (SSED) is a document mandated by the Food, Drug and Cosmetic Act subparagraph 520(h)(1)(A) to be publicly available upon issuance of an approval order of a premarket approval application (PMA). The SSED is applicable for all original PMAs and panel-track supplements. It is an FDA document intended to present a reasoned, objective, and balanced summary of the scientific evidence, both positive and negative, that served as the basis of the decision to approve or deny the PMA. There can be no claims in the SSED that are unsubstantiated by the clinical results of the PMA clinical study(ies). The SSED is not a marketing document, and should not contain marketing/advertising language. NOTE: This checklist contains the essential elements that should, at minimum, be included in the clinical section of the SSED. Additional instructions or examples of what should be included under each topic heading are indicated by italicized text. This document is intended to provide general recommendations for writing the clinical portion of the SSED. However, there will be exceptions, as well as devices that will require additional sections. 1
SUMMARY OF SAFETY AND EFFECTIVENESS DATA (SSED) I. SUMMARY OF PRIMARY CLINICAL STUDY(IES) Provide a definition of terms such as pilot and pivotal. If there are multiple studies, provide a table with all of the studies and their major characteristics EXAMPLE: Clinical Study Study Design Objective Pilot/Feasibility Number of Sites Number of Subjects (Include enrolled, evaluable and lost to follow-up, etc.) Pivotal Continued Access Multi-center, prospective, randomized, controlled clinical trial Single-arm multi-center study Evaluate the safety and effectiveness of the DEVICE Continue to gather additional safety and effectiveness data XX randomized subjects enrolled/treated YY nonrandomized subjects enrolled/treated AA subjects enrolled/treated; data available for BB subjects enrolled/treated as of DATE A. Pivotal Clinical Study Design The clinical study that formed the basis for FDA s finding that [device] is safe and effective for its intended use was a [include all that apply, such as: prospective, multicenter, one/two-arm, cohort study, randomized, controlled, objective performance criteria (OPC), (un)masked, etc.] clinical study. Major design characteristics: level of masking (e.g., double-masked, partially-masked, open-label), type of controls, duration of study, method of allocation to treatment groups (e.g., randomization), and treatment arms. Clinical Endpoints Safety (Include primary safety endpoints and secondary safety endpoints--include only those that are hypothesis driven, and clinically meaningful) Example: The primary safety endpoint was measured at Y months Key secondary endpoints such as Q, R, S were selected to buttress findings of the primary endpoint and lend consistency to trial results. These endpoints were measured at A, B, and C days, weeks, months, years, respectively. 2
Effectiveness (Include primary effectiveness endpoints and secondary effectiveness endpoints--include only those that are hypothesis-driven, and clinically meaningful) Example: The primary effectiveness endpoint was measured at X months. Key secondary endpoints such as Q, R, S were selected to buttress findings of the primary endpoint and lend consistency to trial results. These endpoints were measured at A, B, and C days, respectively. Success/failure criteria This should include a definition of individual patient success, as well as for the study overall. The definition of success should include a targeted value at a specific point in time. It must be specific and quantifiable. Pre-specified statistical analysis plan: study hypothesis(es) comparator (with justification) o OPC? methodology e.g., Frequentist, Bayesian, Adaptive sample size justification statistical test (e.g. superiority, non-inferiority) o delta if non-inferiority method for how missing data will be imputed assumptions External evaluation groups: core laboratory use independent evaluators Data Monitoring Committee (e.g. DSMB) Study design discussion: This design was thought to be appropriate because: Explanation of safety and effectiveness endpoints, control, duration of follow-up, etc. (May use clinical IDE template when available) Explain how thinking on study design evolved and/or changed, if appropriate. Clinical Inclusion and Exclusion Criteria Enrollment in the [name] study was limited to patients who met the following selection criteria Inclusion Exclusion 3
Brief Summary of Treatment and Follow-Up Protocols Follow-up schedule and evaluations Clinical assessments occurred at baseline, operative/discharge and postoperative intervals at X wks, X months and X years. [A detailed chart describing follow-up assessments and time points can be provided here] Prospectively defined subgroup evaluations can be provided here B. Accountability of PMA Cohort At the time of database lock, of N subjects enrolled in PMA study, %(n) subjects are available for analysis at the completion of the study, the xx month/year post-operative visit (final visit evaluated for safety and effectiveness as the basis for the PMA submission). Include an accountability summary table or patient accountability tree. This section should clearly define which set of patients is what analysis cohort (e.g., the intent-totreat cohort, the per protocol cohort, evaluable cohort). C. Study Population Demographics and Baseline Parameters The demographics of the study population were as follows: Include tables showing the proportion of subjects of various races, age distribution, gender, and comorbidities Assessment of appropriateness of study demographics: Identification of demographic groups under or over represented in the study population D. Safety and Effectiveness Results Safety Results Begin with a declarative sentence regarding whether the statistical hypothesis for the primary safety endpoint was met. The analysis of safety was based on the treated (e.g. implanted, ablated) cohort of xxx patients/procedures, etc available for the xxx month evaluation. The key safety outcomes for this study are presented below in tables xx to xx. Adverse effects are reported in tables xx to xx. 4
Adverse effects that occurred in the PMA clinical study: List all adverse effects observed during the PMA clinical study, with incidence rate and number, in descending order of clinical importance, as determined by their severity and/or incidence. It is helpful to present these data in tabular form with a comparative column listing the adverse events that occurred in the control treatment group. Specifics of device or procedure related events should be further discussed in the clinical studies section. You may choose a cut-off such as events occurring at a rate > 1% or >5%. However, for novel devices with small studies, including adaptive and bayesian designs, consider the clinical importance and device relatedness of low rates of occurrence in a larger proportion of the general population with the disease being treated. Consider whether the lack of effectiveness as manifested by worsening of the original condition should or should not be counted as an adverse event. This should be evaluated on a case by case basis and may be condition specific. Include all tables deemed important in the determination of device safety. A short narrative that describes significant findings should be associated with each table. A time course of the occurrence of adverse events as compared to the control treatment is recommended in relation to the initial treatment [Option for PMA Supplements where updating SSED is warranted: Since PMA cohort tends to be smaller: It should be noted that the safety of the device for [indication] was not based on this sample alone, but rather on all the available for the device to date. The safety data from this study were for confirmatory purposes.] Include any additional observations by the clinical reviewer. If adverse events led to any device design modifications during the PMA clinical study, they should be described briefly. Identify the cause of any device failures and the number of occurrences. Effectiveness Results Begin with a declarative sentence regarding whether the statistical hypothesis for the primary effectiveness endpoint was met. The analysis of effectiveness was based on the xxx evaluable at the xxx-month timepoint. Key effectiveness outcomes are presented in tables xx to xx. Continue with FDA s clinical assessment as to the acceptability of these rates. Include all tables on all cohorts deemed important in the determination of device effectiveness. 5
Subgroup Analyses The following preoperative characteristics were evaluated for potential association with outcomes: (e.g., gender, site, age). (Describe associations found with outcomes.) II. SUMMARY OF SUPPLEMENTAL CLINICAL INFORMATION Note: Only include discussion when information contributed to FDA s decision in a major way. This section should describe results for other information considered such as Continued Access, European, or other Outside-the-US (OUS) clinical data, as well as potentially relevant subsets of the study cohort of potential interest to physicians and/or patients (e.g., pediatric, diabetic or other disease/condition of interest, ethnicity, and sex). Include relevant covariate analyses, if performed, and why they were done. III. OVERALL CONCLUSIONS FROM CLINICAL DATA Should include a discussion of the overall clinical risk and benefit analysis decision. If applicable, your conclusions should consider the panel deliberations and conclusions and FDA s post-panel actions Implied Claims If applicable, should consider implied claims and advertising and promotional claims. SSED vs. Device Labeling Although the SSED is a different document and has a different purpose compared to labeling, the clinical trials data will have an important impact on both the SSED and labeling. In writing the SSED, bear in mind specific details of labeling that were influenced by the clinical outcomes data and provide a discussion of these outcomes and how they affected device labeling. The following examples illustrate situations where the SSED should explain important details of labeling: Off-Label Use During its review, FDA may seek a statement in the labeling that there is a lack of evidence that a device is effective for an off-label use or indication. The SSED should contain an explanation of the basis for such limitations. 6
Contraindications Based on Clinical Trials Data If any of the clinical trials data led to a requirement for a Contraindication in the device labeling, provide a thorough explanation in the SSED of what led to the Contraindication. Special Patient Populations Limitations on the usage of a device, based on the clinical data, may be necessary for various reasons including lack of long-term safety and effectiveness data, lack of safety and effectiveness data for specific patient populations (e.g., pregnant women), growth processes still occurring in the body, and anatomical or physiological limitations on the effectiveness of the device. If the clinical data indicate that safety and effectiveness of the device for use in specific patient populations have not been established, provide a discussion of these patient populations in the SSED. Such information is available in the "Indications for Use" section of the labeling. Similarly, the "Precautions" section of the labeling will generally include information about populations for whose use the device s safety and effectiveness have not been established. If use of the device in a certain patient population is associated with a specific hazard, the hazard shall be described in the SSED. This information should be available in the applicable section - Precautions, Warnings, or Contraindications sections of the labeling. 7
Additional Review Tips Helpful Hints General Avoid unsubstantiated claims often included in the draft SSED submitted by the applicant in the PMA. Words or phrases that lack a commonly understood meaning (e.g., imprecise quantitative terms), are not easily defined, are vague, misleading, or promotional in tone should be avoided. Examples include large or small (instead, use actual size or amount), welldesigned (instead, provide specifics about the study design), extensively studied (instead, provide specifics about the database), rapid (instead, specify change/unit time), trend (instead, provide specifics about the outcome), very effective (instead, give the size of the effect), and highly significant (instead, provide the confidence interval). Use only the necessary detail in tables and graphs to summarize large amounts of data that support FDA s decision about the safety and effectiveness of the device. The SSED should not be a rehash of the review memos, but should tell the story of the preclinical and clinical testing in a logical order that led to FDA s decision. Present the data as it is without judgments as to the impact of that data. That information should go into the reviews not the SSED. Number the pages. Tables and Graphs Make sure each is adequately titled and numbered to identify the data summarized. Include the numerator and denominator for ratios and percentages. Include the sample size for each treatment group and the analysis performed to obtain the summarized result. Include baseline data where applicable for comparison. Include footnotes or legends to simplify graphics. Label graph axes with appropriate labels and units of scale. Include p-values and indicate levels of significance reached only if applicable (i.e. prespecified safety and effectiveness analyses with alpha allocation). Clinical and biostatistical reviewers should discuss relevance of p-values in the context of informing for review and use of the product in question. 8