IVD Performance Evaluation in Saudi Arabia: Annex (7) Analytical and Clinical Evidence

1. Key takeaways

Performance evaluation for IVDs in Saudi Arabia is governed by Annex (7) of MDS-REQ 1, titled Performance Evaluation, Performance Studies and Post-Market Performance Follow-Up. The evidence rests on two chains together: analytical performance answers whether the target can be measured at all, and clinical performance answers whether the measurement is accurate and meaningful. Neither chain is dispensable and neither can substitute for the other. Section 4 of MDS-G028 adds an operational test: where the same clinical and performance evaluation data apply to all products cited in the file, the submission is treated as one TF, which directly decides whether multiple products can share a single dossier.

• Annex (7) consolidates performance evaluation, performance studies, and post-market performance follow-up.
• Analytical performance and clinical performance are two independent chains that cannot replace each other.
• The failure modes of the two chains differ, and deficiency requests come from different directions.
• One TF is recognized only when one set of performance data holds for every product cited in the file.
• Bundling caps for IVDs: a maximum of 5 TFs and 50 items, including accessories.
• Studies involving human subjects must be tested against MDS-REQ 2 before any assumption about approval scope.

2. Legal basis

The table below lists the instruments and clauses supporting the performance evaluation obligation. Regulatory information verified as of October 29, 2026.

3. Expert insights

The two chains fail differently, and the deficiency requests come from different directions. Failures of analytical performance typically take the form of precision data covering only one batch, limit of detection verification that does not span all declared sample types, interference studies that miss medications common in Saudi Arabia or the hemolyzed, lipemic, and icteric specimen states, and stability data that do not cover local high-temperature transport conditions.

Failures of clinical performance typically take the form of a comparison study built on an outdated reference method, population samples that do not reflect the epidemiology of the Saudi population, and cut-off values offered without clinical justification.

The two cannot back each other up. An elegant clinical comparison report cannot demonstrate that a reagent is stable across lot numbers, and a complete precision data set cannot demonstrate that the measurement correlates with the clinical conclusion. When GRZAN prepares a performance evaluation report, the two chains each receive their own conclusion page, marked separately with the products and sample types they cover.

The one-TF test shapes the dossier architecture in reverse. The test in MDS-G028 is concrete: where the same clinical and performance evaluation data, covering analytical and clinical performance, apply to all products cited in the file, the submission is treated as one TF. The three categories of official examples, population-specific respiratory virus panels tested on one platform such as Flu A/B together with RSV, panels of drugs of abuse covering cocaine, THC, opiates, methamphetamine, and amphetamine, and tritized quality control materials or calibrators bundled with the kit, share one feature: they draw on the same performance data.

The binding force of this rule is that squeezing three products into one TF does not save two workloads; it requires the performance chapter of that TF to hold for all three products simultaneously. If each of the three products has its own independent performance study reports, the merged file will be split apart again, adding one more deficiency round. The grouping and bundling of IVDs is detailed at https://www.grzan.cn/regulations/saudi/mds-g028-bundling-procedure-pack-and-ivd-29 .

Post-market performance follow-up is the chapter most often left out. The title of Annex (7) itself carries the words Post-Market Performance Follow-Up, which shows that performance evaluation is likewise a chain extending into the post-market phase. Inter-lot variation under real laboratory conditions, stability after transport, and instrument compatibility problems tend to surface only after launch.

For Chinese manufacturers this means the PMPF chapter cannot be a one-line promise. It must state the data sources, complaints, feedback from quality control failures, and external quality assessment results, together with the review frequency and the conditions that trigger re-evaluation, and it must be connected to the post-market surveillance system. PMPF and post-market surveillance are not parallel lines: the outputs of the former, such as inter-lot variation trends and shifts in failure rates, should become inputs for the latter when it decides whether an FSCA or a label change is needed, and the reverse also applies. When GRZAN prepares a PER, PMPF and the post-market surveillance plan are written as two documents that cross-reference each other, preventing two parallel record sets that never communicate after certification. The placement within the technical documentation appears at https://www.grzan.cn/regulations/saudi/ivd-technical-documentation-annex4-22 .

4. Practical recommendations

• Organize the dossier along the twin tracks of analytical and clinical performance, each with its own conclusion page, each stating the covered products and sample types.
• Cover precision studies across multiple lots, multiple instruments, and multiple operators, and never let a single lot stand for the whole.
• Design stability and transport studies for the high-temperature environment of Saudi Arabia and state the storage and transport conditions.
• State the population source and epidemiological background of clinical performance studies and give clinical justification for cut-off values.
• Run the one-TF test before merging dossiers: do the performance data hold for every cited product at the same time?
• For performance studies involving human subjects, first test against MDS-REQ 2 whether prior approval is needed (a practical estimate item to be verified).
• Write the PMPF chapter with data sources, review frequency, and re-evaluation triggers stated explicitly.
• For IVDs on the reliance pathway, prepare the identity statement against the version approved by a recognized regulatory authority in parallel; see https://www.grzan.cn/regulations/saudi/sfda-reliance-pathway-mds-g030-12 .
• Where one TF covers multiple products, place the applicable product matrix together with the bundling classification conclusion at the front of the file so reviewers see the mapping between grouping and evidence at a glance.

5. Verification checklist

• □ Analytical performance and clinical performance stand as independent chapters, each with its own conclusion page
• □ No missing items among precision, limit of detection, linearity, interference, and stability
• □ Stability and transport conditions account for the high-temperature environment of Saudi Arabia
• □ Population source and cut-off justification of the clinical performance studies are stated
• □ The one-TF test has been run and the performance data hold for all cited products
• □ The number of TFs does not exceed 5 and the number of items does not exceed 50, including accessories
• □ The PMPF chapter states data sources, review frequency, and re-evaluation triggers
• □ Studies involving human subjects have been tested for approval against MDS-REQ 2

How GRZAN can help. GRZAN provides medical device and IVD registration, quality system, authorised representative and local testing support for the Saudi Arabian market. Our team prepares SFDA-compliant STAR dossiers, and supports post-market vigilance obligations. Business enquiries: info@grzanglobal.com