New Medical Queries: FMQ and AFMQ Implemented in JMP® Clinical

[#1]

The FDA announced the release of the FDA Medical Query (FMQ) and Algorithmic FDA Medical Queries (AFMQ) at the Duke Margolis-FDA Workshop: Advancing Premarket Safety Analytics¹ on Sept 14, 2022. The description of FMQ and AFMQ can be found in this FDA presentation: Advancing Premarket Safety Analytics².

Traditional medical queries (Standard Medical Queries (SMQ))³ focused primarily on MedDRA preferred terms, which clinical trial sponsors used to classify adverse events that occur during the clinical trial. The FMQ consists of groupings of these terms that correlate to specific broader medical conditions. For example, the FMQ for “Dizziness” consists of 31 different MedDRA preferred terms in below Dizziness sheet in the FMQ  2.1 criteria can be downloaded here⁴. Any trial subject that experiences one or more of these adverse events would be captured as experiencing “Dizziness”. The FMQ version 2.1 has over 100 of these groupings. 

Algorithm FMQs include using additional data from patient demographics, concomitant medications, laboratory findings, and correspondent timing of events with lab results and medications. For example, the AFMQ for Hyperglycemia has multiple criteria as showed below:

Hyperglycemia AFMQ Criteria

Any Preferred Term from the Hyperglycemia FMQ (Narrow Terms only)

A Fasting Plasma Glucose ≥126 mg/dL

Two or more Plasma Glucoses >180 mg/dL

Any New Diabetes Concomitant Medication

A Post Baseline HbA1c ≥6.5%

A HbA1c Increase ≥0.3% with Post Baseline HbA1c ≥5.7%

A Change from Baseline Fasting Plasma Glucose ≥20 mg/dL with Post Baseline FPG >100 mg/dL

Prior to the release of the FMQ and AFMQ criteria, an FDA reviewer used similar criteria to create a  Zegalogue’s Clinical Review (CR) in 2020⁵ to create queries for a FMQ was used by an FDA reviewer to evaluate the common Treatment Emergent Adverse Events (>2%) (shown below). This FMQ was created by JMP, as indicated in the table legend. We mentioned this table but did not show in the Significant AE section in our recent paper, CDISC Enables Efficient Streamlining of Clinical Trial Safety Evaluation⁶.

Dr. Veronica Pei, FDA-CDER, presented “Advancing Pre-Market Safety Analytics” at the CDISC Japan Interchange on July 10, 2023, in conjunction with my talk. Her excellent presentation gave a complete overview about the newly released FMQ and AFMQ criteria that made my presentation much easier to understand. In my keynote, “Efficient Evaluation of Clinical Trial Data Following FDA Guides Using FDA NDAs and CRs as Examples,” I included a few slides to talk about how JMP Clinical takes advantage of FMQ, AFMQ, and SMQ to understand drug safety. (In six months, all the CDISC Japan Interchange presentations⁷ will be available to CDISC members on the CDISC website.)
I am sharing a few of my FMQ and AFMQ slides here to show how JMP Clinical delivers the risk report with FMQ and AFMQ. 
The FMQ and SMQ (standard medical query) from MedDRA have different formats, terminology, and grouping. FMQ is in Excel with multiple sheets, while SMQ has multiple files in .asc and .seq formats. Currently, the FMQ only supports English, but SMQ supports multiple languages.In the FDA presentation² shown below, FMQ picked up a signal for common AE (>2%) for anxiety.I compared FMQ and SMQ in the Medical Query Risk Report in JMP Clinical using our sample data, Nicardipine. Nicardipine is a calcium channel blocker that can treat high blood pressure and control chest pain. Both FMQ and SMQ can adjust either Risk Difference or Absolute Risk Difference to identify any significant difference in adverse events or the risk between the treatment and placebo groups. FMQ recognized hypotension, systemic hypertension, and heart failure, while SMQ named hypertension, cardiac failure, tubulointerstitial diseases, and disorders of sinus node function as having significant differences between the two groups. All risks by FMQ are related to blood pressure and heart problems, but two of the four risks by SMQ were not related to blood pressure or heart problems. We can also see a difference in the terminology: “heart failure” in FMQ and “cardiac failure” in SMQ. The slide below shows risks with significant differences by narrow scope of FMQ and SMQ in the risk difference plots.The algorithmic FMQ is included in FMQ data set for four medical terms for both genders and seven medical terms for each gender. The algorithmic FMQ for each medical term combined information from multiple CDISC domains such as AE, CM, DM, LB, and MH, as well from as various priorities or conditions.JMP Clinical displays AFMQ with a risk plot and table, just as FMQ and SMQ. The tables for the medical queries (AFMQ, FMQ, and SMQ), as are many other tables from JMP Clinical reports, are consistent with FDA Standard Safety Tables and Figures: Integrated Guide⁸ that were published in September 2022.FMQs and AFMQs are already available in JMP Clinical 17. Our amazing development team was able to quickly implement these new queries during the version 17 development cycle. Thanks to Geoffrey Mann for his help in putting these slides together. 

Blog posts in this series:

• New Medical Queries: FMQ and AFMQ Implemented in JMP® Clinical
• Summary of clinical trials using JMP® Clinical
• Safety Results I: Using the Adverse Events Narrative in JMP® Clinical to describe deaths and SAEs

References

1. Duke Margolis-FDA Workshop: Advancing Premarket Safety Analytics 2022: https://healthpolicy.duke.edu/sites/default/files/2022-09/Advancing%20Premarket%20Safety%20Analytics...
2. FDA Advancing Premarket Safety Analytics 2022: https://healthpolicy.duke.edu/sites/default/files/2022-09/Advancing%20Premarket%20Safety%20Analytics...
3. MedDRA https://www.meddra.org/
4. FMQ 2.1 download 2022: https://downloads.regulations.gov/FDA-2022-N-1961-0001/attachment_1.xlsm.
5. FDA Zegalogue’s Clinical Review 2020 https://www.fda.gov/media/147791/download
6. Mann, G. & Pedersen, T. J. & Lyzinski, R. & Scott, A. & Foglia, A. J. & Cromer, J. & Dong, M. & Varga, N. & Gardner, S. & Kirchberg, C. J. & Wingerd, B. A. & Wolfinger, R. D. & Bao, W., (2023) “CDISC Enables Efficient Streamlining of Clinical Trial Safety Evaluation”, Journal of the Society for Clinical Data Management 3(1). https://doi.org/10.47912/jscdm.169  
7. CDISC Japan Interchange 2023 https://www.cdisc.org/members-only/interchange-presentations
8. FDA Standard Safety Tables and Figures: Integrated Guide 2022 https://www.regulations.gov/document/FDA-2022-N-1961-0046