In vitro diagnostics (IVD) development

In vitro diagnostics occupy a distinct corner of the medical device landscape—one where the engineering challenge is fundamentally about measurement. A diagnostic device does not treat or monitor in real time. It produces a result that a clinician or patient acts on, and the quality of that result is determined by decisions made during development: the choice of assay format, the selection of reference materials, the statistical design of the performance study. Getting the engineering right means getting the measurement right. The regulatory submission is the argument that you did.

MANKAIND is built for IVD engineering teams who understand that analytical validation is not a checklist exercise—it is an evidence-generation program that must be designed with the submission framework in mind from the start. The platform connects the analytical decisions made at the bench to the performance claims made in the submission, maintaining traceability across the full validation program.

Analytical validation is an experimental design problem

The core performance claims for an IVD—sensitivity, specificity, accuracy, precision, linearity, interference susceptibility—are engineering specifications before they are submission data points. Setting the right acceptance criteria requires understanding the clinical context: what does a false negative cost, and what does a false positive cost? For a sepsis marker, the tolerance for false negatives is different than for a low-acuity screening test. The performance specification is a clinical engineering decision, not an arbitrary target.

MANKAIND supports analytical validation planning by connecting intended use and clinical claims to the EP09, EP05, EP06, EP07, and EP15 study designs that generate the necessary evidence. When an engineering team defines the analyte, the specimen type, the measuring range, and the patient population, the platform maps those parameters to the applicable CLSI EP protocols and helps the team design a study plan that will generate data sufficient to support the intended performance claims. This happens before the reagent lots are manufactured—not after the first round of studies reveals a gap in the experimental design.

Precision studies are a common source of delays. A device that meets its within-run precision target but fails its between-laboratory repeatability requirement has a genuine engineering problem—reagent lot-to-lot variability, calibrator stability, or matrix sensitivity that was not characterized early enough. MANKAIND tracks precision data longitudinally through the development program, flagging trends before they become submission failures.

Reference material selection and method comparison

For quantitative IVDs, the accuracy claim depends on the reference method comparison study, and the reference method comparison study depends on having access to appropriate reference materials. This is a practical engineering constraint that affects submission timelines more than most teams anticipate. Reference materials from NIST, the WHO, or commercial sources may have matrix differences, concentration range limitations, or commutability issues that affect the comparison study design.

MANKAIND supports the reference material characterization process by maintaining the provenance chain from reference material to calibrator to measurement result. When the FDA or a notified body asks how the quantitative claims were established—what reference method was used, what the traceability chain looks like, and what the uncertainty budget is—the engineering team can answer with a structured document that reflects the actual development history, not a retrospective reconstruction.

EU IVDR: a technical file engineering problem

EU IVDR 2017/746 substantially increased the evidence burden for in vitro diagnostics in the European market. Class C and Class D devices—which include most clinical chemistry analyzers, most infectious disease assays, and all blood grouping systems—now require notified body involvement at the technical documentation review stage. The technical file requirements under IVDR Annex II and Annex III are extensive, and notified bodies are scrutinizing performance evaluation documentation with a rigor that was not present under the former IVDD framework.

The performance evaluation report (PER) is the centerpiece of IVDR technical documentation. It must integrate the analytical performance data, the clinical performance data, the scientific validity assessment, and the post-market performance follow-up (PMPF) plan into a coherent evidence narrative. Engineering teams that generate each component separately—analytical validation binder, clinical study report, literature review—and then attempt to assemble them into a PER find that the components do not tell a consistent story. MANKAIND generates the PER structure from the underlying engineering data, maintaining consistency between the analytical performance claims, the intended purpose, and the clinical performance evidence from the start of the validation program.

The IVDR common specifications (CS) define minimum performance requirements for specific analyte categories. For devices covered by CS, the engineering team must demonstrate conformance with quantitative performance thresholds—not just demonstrate that performance is adequate for the intended use. MANKAIND maps the device's analyte category to the applicable common specifications early in the development program, so the study design is built to demonstrate CS conformance rather than being retrofitted to meet requirements discovered late.

Companion diagnostics: the intersection of IVD and drug development

Companion diagnostic devices—IVDs developed in parallel with a therapeutic to identify patients likely to respond—face a dual submission pathway: a PMA or 510(k) with FDA's Center for Devices and Radiological Health, and a co-development agreement with the drug sponsor navigating FDA's Center for Drug Evaluation and Research. The engineering challenge is that the diagnostic performance specifications must be established in the context of the drug's clinical trial, using specimens from trial participants, before the clinical evidence for the drug is complete.

MANKAIND supports companion diagnostic programs by maintaining the traceability between the diagnostic device's performance specifications and the clinical trial protocol—the specimen collection procedures, the patient population definition, the clinical endpoint against which diagnostic performance is measured. When the NDA and PMA submissions go in concurrently, the engineering team can demonstrate that the diagnostic claims are anchored to the same clinical data that supports the drug's efficacy claims.

Post-market performance follow-up: a continuous engineering obligation

IVDR's post-market performance follow-up requirements are more demanding than the post-market surveillance requirements most medical device teams are accustomed to. PMPF for Class C and D devices requires a proactive, systematic data collection program that monitors the device's performance in real-world clinical use—not just complaint handling and adverse event reporting.

The PMPF plan must define the performance indicators to be monitored, the data sources to be used, the frequency of evaluation, and the thresholds that would trigger a corrective action. That is an engineering specification. MANKAIND generates the PMPF plan from the analytical performance claims established during development—the precision targets, the accuracy claims, the interference limits—and connects it to the post-market data collection infrastructure that will generate the evidence for periodic performance evaluation reports.

IVD development is a discipline where engineering rigor and measurement science are inseparable from regulatory success. MANKAIND provides the engineering intelligence infrastructure to keep those disciplines aligned—from first-in-lab experiments through post-market surveillance.