Cost Optimization for Korean Medical Device Registration Projects

1. Key takeaways

The optimisation space in a Korean project budget concentrates on 3 levers: the substantial equivalence route (which can waive the technical documents), using MDSAP audit results for KGMP desktop review (which can substitute part of the on-site audit), and item group consolidation (which can reduce the number of applications). The three act on different stages: substantial equivalence acts on dossier preparation, MDSAP acts on the quality system audit, and item group consolidation acts on the number of applications. They do not replace one another and can be stacked.

• By class, the optimisation order is: for Class 1 the leverage comes mainly from item group consolidation; for Class 2 all three levers can apply; for Class 3 and Class 4 the leverage comes mainly from substantial equivalence dossier organisation and MDSAP, with relatively limited room for item group consolidation.
• The precondition for optimisation is not compromising the correctness of the class determination. Suppressing the class to cut fees invites a re-determination request, and the price is redoing the entire dossier.
• MDSAP saves the initial-stage cross-border on-site audit cost but cannot save the periodic audit 3 years later, so benefits must be modelled accordingly.

2. Legal basis

The three levers rest on the following instruments, traceable through the regulation names, document numbers, publication dates and provisions listed below.

3. Expert insights

Insight 1: The greatest value of substantial equivalence is not saving money but saving time. Where the substantial equivalence conditions are met, the technical documents may be waived, directly cutting the dossier preparation and translation workload (translation fees are a practical estimate). But GRZAN regards the more important value as lowering the probability of deficiency letters, because the technical documents are where deficiencies concentrate. The time saved usually affects the launch schedule more than the money saved, which is especially visible when multiple items are filed in parallel.

Insight 2: MDSAP has boundaries; holding an MDSAP certificate is not sufficient on its own. MFDS Public Notice No. 2023-79 allows desktop review based on MDSAP initial, supplemental and change audit results, but the periodic audit must still be conducted on site. The KGMP certificate obtained on this basis is valid for the same period as the MDSAP certificate. This means what MDSAP saves is the cross-border on-site audit cost at the initial stage; it cannot save the periodic audit 3 years later. Companies should model the benefit as savings at the initial stage only, with no savings at the periodic audit.

Insight 3: Item group consolidation gains scale with model count but changes later change costs. Consolidating item groups reduces the number of applications, directly lowering the per-application official fees and the repeated translation workload. The price is that a change to any model within the group may trigger a change assessment for the whole group. GRZAN’s criterion is: products with highly homogeneous models and low change frequency suit consolidation, while products with large model differences or fast iteration are cheaper filed separately.

Insight 4: The optimisation order should be route first, then quality system, then application count. The controllability of the three levers differs: the route choice depends on the product itself and the availability of the predicate device, and is hard to change once fixed; the audit approach depends on the existing MDSAP status; and the application count can be adjusted repeatedly at the dossier stage. GRZAN evaluates item by item in this order, avoiding effort spent on the lever that acts last. The timing windows also differ: the route and dossier organisation are fixed before filing, the audit approach is fixed before the KGMP application, and application count optimisation can extend into dossier preparation, so locking decisions too early forfeits room for adjustment.

4. Practical recommendations

• Lock the class before discussing optimisation. Do not evaluate optimisation options before the class determination is settled, to avoid misjudging the class in pursuit of lower fees.
• Pursue substantial equivalence first. Confirm in advance that the predicate device must be a product already licensed, certified or notified in Korea, and prepare a documentation fallback plan if the comparison fails.
• Model MDSAP benefits as savings at the initial stage only. The initial audit may use desktop review, but the KGMP periodic audit still requires an on-site audit, and the certificate is valid for 3 years.
• Decide on item group consolidation by model homogeneity and change frequency. Consolidate where models are homogeneous and changes infrequent; file separately where differences are large or iteration is fast.
• Write the optimisation plan into the project budget sheet with its basis stated. Enter statutory fees per the Enforcement Rule attached table and mark the service fee portion as a practical estimate.

5. Verification checklist

• □ Correctness of the class determination: check against Attached Table 1 of the Enforcement Rule; the conclusion is traceable and the class was not suppressed to cut fees.
• □ Availability of substantial equivalence: check whether the predicate device is already licensed, certified or notified in Korea; the predicate qualification and the six comparisons all stand.
• □ MDSAP status: check the existing MDSAP certificate and audit types; whether initial, supplemental or change audits can support desktop review is clarified.
• □ Periodic audit arrangement: check the KGMP certificate validity; valid for 3 years, with the periodic audit planned as an on-site audit.
• □ Item group division: check model homogeneity and change frequency; both consolidation and separate filing have written grounds.
• □ Optimisation benefit projection: compare budgets before and after optimisation; official fees and service fees (practical estimate) are listed separately.
• □ Change impact assessment: assess the whole-group change risk after consolidation; the post-change workload is included in the long-term cost.
• □ Regulatory information verified as of 29 October 2026 (pending verification for any later amendments).

How GRZAN can help. GRZAN provides medical device and IVD registration, quality system, Korean licence holder and local testing support for the Republic of Korea. Our team coordinates licence holder arrangements, import business registration, and supports post-market vigilance obligations. Business enquiries: info@grzanglobal.com