We provide end-to-end analytical and clinical testing to support FDA diagnostic submissions, including benchtop testing, analytical performance studies, clinical performance studies, and point-of-care testing, across genomics, cell biology, microbiology, immunoassays, and other advanced technologies. We also support companion diagnostics.

510(k) Premarket Notification (for assays with a predicate device), De Novo classification (for novel assays with no predicate), Premarket Approval (for high-risk diagnostic assays), Emergency Use Authorization (for declared public health emergencies), and CLIA Waiver by Application (for point-of-care and home-use tests).

Limit of detection and limit of quantitation; accuracy and method comparison; precision (repeatability, intermediate precision, and reproducibility); linearity and measuring range; and analytical specificity, including cross-reactivity, interference, and carryover.

Molecular assays (PCR and NGS), immunoassays (ELISA and lateral flow), and clinical chemistry assays (LC-MS/MS). Molecular assays and immunoassays can be supported through benchtop, analytical performance, clinical performance, and point-of-care testing. Clinical chemistry assays are supported through benchtop, analytical performance, and clinical performance testing.

Yes: matrix equivalency (serum, plasma, whole blood), sample type validation, and sample stability under real-time, freeze–thaw, and transport conditions.

Yes: real-time and accelerated stability, open-vial stability, environmental stress, and operator variability studies.

Yes: clinical sensitivity and specificity, comparator method alignment, patient sample testing, multi-site study coordination, and IDE support.

Our operations are 21 CFR-compliant, our facility is FDA-registered and inspected, our testing services are CLIA-registered, and we work to CLSI standards and FDA expectations under GLP/CGMP operations.

FDA reviews the data package, not who ran it, but it expects the study to be scientifically sound and traceable. FDA’s IVD study guidance asks for a protocol with objectives and performance goals tied to your claims, a pre-specified statistical analysis plan, records that preserve raw measurements and specimen traceability, and a primary analysis that includes all observations rather than excluding outliers after the fact.

The pathway depends on your intended use, the analyte, the specimen type, who will run the test, and whether a predicate exists. A 510(k) is built on showing substantial equivalence to a predicate, with analytical performance (bias, imprecision, sensitivity, specificity) as the core of the review. De Novo applies when there is no predicate and general controls are enough to assure safety and effectiveness. PMA applies to the highest-risk tests. Because the wrong pathway assumption can invalidate a study plan, we start from your intended use and design the study package to match the pathway you and FDA settle on.

Analytical studies for IVDs are typically designed around the relevant CLSI evaluation protocols, for example for precision, linearity, detection capability, interference, and method comparison, and FDA recognizes CLSI standards for sensitivity and specificity estimation. We design each study to the CLSI protocol that matches the performance claim, including the number of runs, replicates, days, lots, and operators needed to support it, and we will tell you up front if your proposed claim needs a larger study than you budgeted.

Clinical performance claims need specimens from a representative sample of the intended-use population, though enriched, credentialed, or spiked samples are acceptable for rare conditions or to challenge cut-offs, as long as the performance estimate still reflects real-world use. Because IVD studies using human specimens are human-subject research, informed consent and IRB review (21 CFR Parts 50 and 56) apply, though FDA exercises enforcement discretion for certain studies using non-identifiable leftover specimens. We work with you to define a specimen plan that matches your intended use before testing begins.

We coordinate multi-site studies, align results to the comparator method, and test patient samples under a single protocol so that site-to-site variation is something you can analyze rather than discover afterward. FDA’s guidance is explicit that a sponsor’s own lab counts as a study site and is subject to the same investigational plan and bioresearch monitoring inspection as independent sites, so we plan documentation and traceability on the assumption that an inspector may ask to see it. We also support IDE needs where your study is not exempt, for example when the test result will guide treatment decisions and no medically established diagnostic is available to confirm it.

A companion diagnostic provides information that is essential for the safe and effective use of a corresponding therapeutic product, so its validation is tied to the therapy’s development timeline and clinical trial rather than a standalone product launch. Practically, that means locking the assay’s intended use and performance targets early, planning clinical specimen bridging between trial samples and the final assay, and keeping analytical validation moving in parallel with the clinical program. We support companion diagnostic testing and validation across molecular, immunoassay, and clinical chemistry platforms.

Before the first scoping call, it helps to have your intended-use statement (analyte, specimen type, patient population, user setting), your target claims (for example, the LoD and clinical sensitivity/specificity you want to state), your predicate or comparator if you have one, and whether your sample types and matrices are fixed. We recommend engaging before you finalize your protocol, since sample size, number of lots and operators, and specimen sourcing are the biggest drivers of study cost and timeline, and they are far cheaper to change on paper than after the study has started. FDA’s own process encourages early engagement for novel technology and complex studies for the same reason.