Elecsys Vitamin D total III

{0} FDA U.S. FOOD & DRUG ADMINISTRATION # 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY ASSAY ONLY ## I Background Information: A 510(k) Number K210901 B Applicant Roche Diagnostics C Proprietary and Established Names Elecsys Vitamin D total III D Regulatory Information | Product Code(s) | Classification | Regulation Section | Panel | | --- | --- | --- | --- | | MRG | Class II | 21 CFR 862.1825 - Vitamin D Test System | CH - Clinical Chemistry | ## II Submission/Device Overview: A Purpose for Submission: New device B Measurand: Total 25-hydroxy vitamin D (25-OH vitamin D) C Type of Test: Quantitative immunoassay ## III Intended Use/Indications for Use: A Intended Use(s): See Indications for Use below. Food and Drug Administration 10903 New Hampshire Avenue Silver Spring, MD 20993-0002 www.fda.gov {1} K210901 - Page 2 of 11 ## B Indication(s) for Use: Binding assay for the in vitro quantitative determination of total 25-hydroxyvitamin D in human serum and plasma. This assay is to be used as an aid in the assessment of vitamin D sufficiency in adults. The electrochemiluminescence binding assay is intended for use on cobas e immunoassay analyzers. ## C Special Conditions for Use Statement(s): Rx - For Prescription Use Only ## D Special Instrument Requirements: cobas e 601 analyzer ## IV Device/System Characteristics: ### A Device Description: The Elecsys Vitamin D total III assay is comprised of a kit for multiple use with three ready to use liquid assay reagents and two pretreatment reagents: - Pretreatment reagent 1: Dithiothreitol 1 g/L. - Pretreatment reagent 2: Sodium hydroxide 57.5 g/L. - Ruthenium labeled vitamin D binding protein 150 µg/L; bis-tris propane buffer 200 mmol/L; albumin 25 g/L; preservative. - Biotinylated 25-hydroxyvitamin D 20 µg/L; bis-tris propane buffer 200 mmol/L; preservative. - Streptavidin-coated microparticles: Streptavidin-coated microparticles 0.72 mg/mL and preservative. ### B Principle of Operation: Elecsys Vitamin D total III assay is a back titration immunoassay, comprising several discrete steps: pretreatment step to release 25-OH vitamin D bound to vitamin D binding protein; incubation of pretreated sample with ruthenium tris-bipyridine labeled vitamin D binding protein; and second incubation of sample with streptavidin-coated microparticles and biotinylated 25-OH vitamin D labeled to bind any unbound ruthenium tris-bipyridine labeled vitamin D binding protein. The sample is drawn into the analyzer’s electrochemiluminescence (ECL) detection module, and the ECL signal generated is inversely proportional to amount of 25-OH vitamin D in the sample. {2} V Substantial Equivalence Information: A Predicate Device Name(s): Elecsys Vitamin D total II B Predicate 510(k) Number(s): k162840 C Comparison with Predicate(s): | Device & Predicate Device(s): | k210901 | k162840 | | --- | --- | --- | | Device Trade Name | Elecsys Vitamin D total III | Elecsys Vitamin D total II | | General Device Characteristic Similarities | | | | Intended Use/Indications For Use | In vitro quantitative determination of total 25-hydroxy vitamin D in human serum and plasma. | Same | | General Device Characteristic Differences | | | | Measuring Range | 6.0 - 120 ng/mL | 5.0 - 100 ng/mL | | Biotin interference levels | >600 ng/mL | >30 ng/mL | VI Standards/Guidance Documents Referenced: CLSI EP05-A3 Evaluation of Precision of Quantitative Measurement Procedures; approved guideline- third edition. CLSI EP06-A Evaluation of the Linearity of Quantitative Measurement Procedures. CLSI EP17-A2 Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures. CLSI EP28-A3c Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory; Approved Guideline—Third Edition. VII Performance Characteristics (if/when applicable): A Analytical Performance: 1. Precision/Reproducibility: The precision performance of the Elecsys Vitamin D total III assay on the cobas e 601 analyzer was established by two studies to evaluate within-run (repeatability) precision and within-laboratory precision following the recommendations in the CLSI guideline EP05-A3. K210901 - Page 3 of 11 {3} Study #1 for 21 days In the study, 10 human serum samples and two controls were tested with one lot and one instrument in duplicate per run, with two runs per day for 21 days with a total of 84 measurements per sample. The result were analyzed for within-run and within-laboratory precision: | Sample | Mean (ng/mL) | Within-run | | Within-laboratory | | | --- | --- | --- | --- | --- | --- | | | | SD (ng/mL) | CV% | SD (ng/mL) | CV% | | HS 1 | 12.3 | 0.905 | 7.4 | 1.20 | 9.8 | | HS 2 | 15.3 | 0.809 | 5.3 | 1.10 | 7.2 | | HS 3 | 28.7 | 1.28 | 4.4 | 1.74 | 6.0 | | HS 4 | 33.0 | 1.39 | 4.2 | 1.85 | 5.6 | | HS 5 | 52.3 | 1.68 | 3.2 | 2.09 | 4.0 | | HS 6 | 61.0 | 1.39 | 2.3 | 2.23 | 3.7 | | HS 7 | 96.5 | 2.6 | 2.8 | 2.92 | 3.0 | | HS 8 | 99.6 | 2.45 | 2.5 | 2.76 | 2.8 | | HS 9 | 111 | 3.00 | 2.7 | 3.63 | 3.3 | | HS 10 | 112 | 3.37 | 3.0 | 3.65 | 3.3 | | Control 1 | 23.0 | 1.27 | 5.5 | 1.49 | 6.5 | | Control 2 | 41.9 | 1.52 | 3.6 | 1.87 | 4.5 | Study #2 for five days In the study, human serum samples and controls were tested with each of three lots and one instrument in replicates of five per run for five days with a total of 25 measurements per sample. Ten serum samples were tested with two of the lots and 9 serum samples with a third lot. For each lot, the result were analyzed for within-run and within-laboratory precision. The results from one representative lot are shown in the following table: | Sample | Mean (ng/mL) | Within-run | | Within-lab | | | --- | --- | --- | --- | --- | --- | | | | SD (ng/mL) | CV (%) | SD (ng/mL) | CV (%) | | HS 1 | 11.6 | 1.11 | 9.6 | 1.16 | 10.0 | | HS 2 | 15.0 | 1.36 | 9.1 | 1.36 | 9.1 | | HS 3 | 24.4 | 1.85 | 7.6 | 1.85 | 7.6 | | HS 4 | 32.8 | 1.92 | 5.9 | 1.92 | 5.9 | | HS 5 | 56.8 | 3.13 | 5.5 | 3.13 | 5.5 | | HS 6 | 55.0 | 2.60 | 4.7 | 2.93 | 5.3 | | HS 7 | 69.0 | 2.55 | 3.7 | 2.97 | 4.3 | | HS 8 | 102 | 3.37 | 3.3 | 3.37 | 3.3 | | HS 9 | 107 | 3.45 | 3.2 | 3.59 | 3.4 | | HS 10 | 119 | 3.41 | 2.9 | 3.78 | 3.2 | | Control 1 | 24.0 | 1.48 | 6.2 | 1.48 | 6.2 | | Control 2 | 45.1 | 2.47 | 5.5 | 2.47 | 5.5 | 2. Linearity: The linearity performance of the Elecsys Vitamin D total III assay on the cobas e 601 analyzer was established in a study conducted following the recommendations of CLSI guideline EP06-A. In the study, 11 samples were prepared by intermixing known volumes of K210901 - Page 4 of 11 {4} a high analyte serum sample of known concentration with a depleted low analyte serum sample, and covering a range of 1.67 to 133 ng/mL. Each sample was assayed in triplicate using one lot and one instrument. The mean of the observed results were compared by regression analysis to the expected results based on the dilution volumes. Regression analysis identified that the coefficients for polynomial fit were statistically significant, and therefore the polynomial better than the linear model. Further analysis found that the deviation from linearity at each concentration (i.e. difference between polynomial and linear fit) to be insignificant as defined by the sponsor with absolute bias less than ±2 ng/mL and percent bias less than ±10%. Based on these results, the sponsor concluded that the test system demonstrated a proportional response over the claimed measuring range of 6 – 120 ng/mL. ## Sample Dilution study A sample dilution study was conducted to support the claim in the product labeling that 25-OH vitamin D samples with results above the measuring range may be manually diluted with the sponsor’s Diluent Universal or serum with a low analyte concentration. In the study, four samples were spiked with 25-OH vitamin D to prepare concentrations with expected values ranging from 125 - 183 ng/mL. Each of the four samples were diluted 1:2 with Universal Diluent, and diluted 1:2 with analyte depleted serum. Each diluted sample was assayed in triplicate, and analyzed for recovery after correcting for dilution. The recoveries ranged from 95% to 109%. The dilution study results support the sponsor’s labeling claims that samples with 25-OH vitamin D concentrations above 120 ng/mL may be manually diluted 1:2 to obtain results up to 240 ng/mL. ## 3. Analytical Specificity/Interference: The analytical specificity performance of the Elecsys Vitamin D total III assay on the cobas e 601 analyzer was established by conducting a cross-reactivity study and interference testing for endogenous and exogenous substances. ## Endogenous Interference from certain endogenous substances was assessed using three serum samples with 25-OH Vitamin D concentrations of low, mid, and high. Each of the three samples was further divided into two aliquots for a control sample (with no added interferent) and test sample (with added interferent). For each of the three original serum samples, the control and test samples were further prepared into a series of 11 samples of varying substance concentration by inter-mixing the control and test samples with known volumes. Each sample of the dilution series was assayed in five replicates using one lot and one instrument. From measurement of the dilution series for the three original serum samples of low, mid, and high 25-OH Vitamin D concentration, the highest substances concentration within the allowable error was found, and identified as the highest concentration at which no significant interference is observed. The sponsor defined the allowable error as: LoQ to 20.0 ng/mL: ≤ 2.5 ng/mL >20.0 ng/mL to 50 ng/mL: ≤ 10% >50.0 ng/mL: ≤ 15% K210901 - Page 5 of 11 {5} The results of the study are summarized as follows: | Substance | Highest concentration tested at which no significant interference is observed | | --- | --- | | Albumin | 7 g/dL | | Bilirubin | 66 mg/dL | | Hemoglobin | 600 mg/dL | | Intralipid | 300 mg/dL | | IgG | 7 g/dL | | IgA | 1.3 g/dL | | IgM | 1 g/dL | | Rheumatoid Factor | 1200 IU/mL | | Triglyceride | 300 mg/dL | ## Biotin Interference Biotin interference was tested with three serum samples containing 25-OH Vitamin D concentrations of 10, 28, and 77 ng/mL. Each of the three samples were spiked at two levels of biotin at 1550 and 3600 ng/mL, along with controls samples with no added biotin. For each of the serum samples, the control and test samples were further prepared into a series of 11 samples of varying substance concentration by inter-mixing the control and test samples with known volumes. Each sample of the dilution series was assayed in five replicates using one lot and one instrument. From measurement of the dilution series for the serum samples of low, mid, and high 25-OH Vitamin D concentration, the highest biotin concentration within the allowable error of ±10% was found. The results of the study support the labeling claim that specimens containing biotin up to a concentration of 600 ng/mL demonstrated ≤ 10% change in 25-hydroxyvitamin D assay results. | 25-OH Vitamin D conc. | | Biotin concentration (ng/mL) | | | | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | 155 | 310 | 620 | 930 | 1240 | 1395 | 1550 | | 9.8 ng/mL | Absolute difference | 0.45 | 0.48 | 2.05 | 2.63 | 3.63 | 4.18 | 4.67 | | 28.3 ng/mL | % difference | 3% | 4% | 8% | 11% | 12% | 15% | 16% | | 77.0 ng/mL | | -1% | -2% | -1% | -2% | -2% | -1% | -2% | | 25-OH Vitamin D conc. | | Biotin concentration (ng/mL) | | | | | | | | --- | --- | --- | --- | --- | --- | --- | --- | --- | | | | 360 | 620 | 1240 | 1860 | 2480 | 2790 | 3600 | | 10.5 ng/mL | Absolute difference | 0.16 | 0.58 | 3.68 | 4.41 | 8.55 | 11.4 | 13.2 | | 36.3 ng/mL | % difference | -1% | 1% | 15% | 18% | 36% | 47% | 55% | | 76.7 ng/mL | | 0% | 5% | 7% | 14% | 27% | 31% | 36% | ## Exogenous Interference from 20 drugs was assessed using two serum samples with 25-OH Vitamin D concentrations of low and high. Each of the samples was further divided into two aliquots for a control sample with no added interferent and test sample with added interferent. For each of the substances, the test sample and control sample were assayed in 10 replicates using one lot and one instrument. The results were analyzed as percent recovery between the test sample and control sample. The substance under test was considered to have no significant K210901 - Page 6 of 11 {6} interference if the recovery was $100 \pm 10\%$. The substances and concentrations tested which did not caused significant interference are listed below. | Substance | Concentration tested at which no significant interference was observed | | --- | --- | | Acetaminophen | 15.6 mg/dL | | Acetylcysteine | 15 mg/dL | | Acetylsalicylic acid | 3.0 mg/dL | | Alphacalcidol | 0.00018 mg/dL | | Ampicillin-Na | 7.5 mg/dL | | Ascorbic acid | 5.25 mg/dL | | Cefoxitin | 75 mg/dL | | Cyclosporine | 0.18 mg/dL | | Doxycycline | 1.8 mg/dL | | Heparin | 3300 IU/L | | Ibuprofen | 21.9 mg/dL | | Itraconazole | 3.0 mg/dL | | Levodopa | 0.75 mg/dL | | Methyldopa | 2.25 mg/dL | | Metronidazole | 12.3 mg/dL | | Phenylbutazone | 32.1 mg/dL | | Rifampicin | 4.8 mg/dL | | Rocaltrol (Calcitriol) | 0.000102 mg/dL | | Theophylline | 6.0 mg/dL | | Zemplar (Paricalcitol) | 0.00012 mg/dL | ## Cross-reactivity A cross-reactivity study was conducted to quantify the level of cross-reactivity from certain vitamin D metabolites with the assay. In the study, three test samples were prepared by spiking individually various vitamin D metabolites into native human serum samples with approximate 25-OH vitamin D concentrations of 20, 40, and $60\mathrm{ng/mL}$. The concentrations of the cross-reactants were chosen to ensure results within the measuring range depending on their expected cross-reactivity and to reflect at least $>10$ times higher concentration as typically found in native serum. These samples were paired with three reference samples of the same serum that were not spiked and with the same 25-OH vitamin D concentrations. Each sample was assayed in replicates of five using one lot and one instrument. The results from each of the three test samples and reference samples were averaged. The difference in results between test and reference samples was calculated for each of the three 25-OH vitamin D concentrations, and each analyzed for $\%$ cross-reactivity using the below equation. The table summarizes the cross-reactivities from the study as the mean of the three results. The product labeling lists $\%$ cross-reactivity and $\%$ cross-reactivity normalized to the 25-hydroxyvitamin D3 $\%$ cross-reactivity. $\%$ Cross-Reactivity = 25-OH vitamin D result from test sample – 25-OH vitamin D result from reference sample $\times 100\%$ / Concentration of cross-reactive substance K210901 - Page 7 of 11 {7} | Cross-Reactant | Concentration of Cross-Reactant, ng/mL | %Cross Reactivity | %Cross Reactivity normalized to 25-hydroxyvitamin D3 | | --- | --- | --- | --- | | 25-hydroxyvitamin D3 | 50 | 72.2% | 100.0% | | 25-hydroxyvitamin D2 | 50 | 75.7% | 103.3% | | 24,25-dihydroxyvitamin D3 | 100 | 5.8% | 8.1% | | 3-epi-25-hydroxyvitamin D3 | 50 | 88.2% | 121.6% | | 3-epi-25-hydroxyvitamin D2 | 50 | 74.7% | 102.7% | | 1,25-dihydroxyvitamin D3 | 100 | None detected | - | | 1,25-dihydroxyvitamin D2 | 100 | 0.3% | 0.9% | | Vitamin D3 (Cholecalciferol) | 1000 | 0.6% | 0.9% | | Vitamin D2 (Ergocalciferol) | 1000 | 0.5% | 0.7% | Human Anti-Mouse Antibodies (HAMA) The effect of human anti-mouse antibodies with the Elecsys Vitamin D total III assay on the cobas e 601 analyzer was assessed. In the study, serum containing 805 ng/mL HAMA and control serum without HAMA were each measured in duplicate. The results demonstrated that there was no significant interference from HAMA with the assay when using the sponsor's criteria of 100 ± 10% recovery. 4. Assay Reportable Range: See Linearity A.2 5. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods): Metrological traceability: The device has been standardized using internal standards which are traceable to the ID-LC-MS/MS 25-hydroxyvitamin D Reference Measurement Procedure. The ID-LC-MS/MS is traceable to the National Institute of Standards and Technology Standard Reference Material 2972. 6. Detection Limit: Detection capability studies of the Elecsys Vitamin D total III assay on the cobas e 601 analyzer were conducted for limit of blank (LoB), limit of detection (LoD), and limit of quantitation (LoQ) following the recommendations in CLSI EP17-A2. LoB The LoB of the assay was evaluated using Vitamin D depleted human serum sample pool, and three reagent kit lots and one instrument. The sample was measured in replicates of 10 per run, across six runs over four days for a total of 60 replicates per lot. The LoB was analyzed as the 95th percentile of measurements by the non-parametric method of the guidance by ranking ordering of the results. The highest LoB of the three lots was found to be 1.6 ng/mL; supporting the claimed LoB in the product labeling of 2.0 ng/mL. K210901 - Page 8 of 11 {8} LoD The LoD was evaluated using five low-level samples prepared from diluted serum pools using three reagent lots and one instrument. The samples were tested in replicates of two per run, with six runs over four days for a total of 60 measurements per lot. For each lot, the LoD was analyzed using a parametric data analysis which was the lowest concentration at which the analyte can be detected with 95% probability. The LoD of the assay was selected from the highest value obtained across the three lots, which was 2.8 ng/mL, supporting the claimed LoD in the product labeling of 3.0 ng/mL. LoQ The LoQ was evaluated using 8 low-level diluted serum samples using three reagent lots and one instrument. Each of the samples was tested in replicates of five across each lot on five different days for a total of 25 measurements per sample per lot. The LoQ was defined as the lowest concentration of analyte which has imprecision was less than or equal to 20% CV. The LoQ of the assay was selected from the highest value obtained across the three lots, which was 5.1 ng/mL, supporting the claimed LoQ in the product labeling of 6.0 ng/mL. 7. Assay Cut-Off: Not applicable. B Comparison Studies: 1. Method Comparison: The accuracy performance of the Elecsys Vitamin D total III assay on the cobas e 601 analyzer was established by a method comparison against a reference method for vitamin D. In the study, 157 native, single donor patient serum samples were tested in singlicate on the candidate device and the reference method. The samples were CDC Verification Samples provided by the Vitamin D Standardization and Certification Program with assigned values by the candidate Reference Method Procedure: ID-LCMS/MS at the CDC Vitamin D Reference Laboratory. The results of the study were analyzed graphically for slope and intercept by Deming regression analysis, and summarized as follows: | Concentration range, comparator device | Regression equation | r | | --- | --- | --- | | 5.6 to 118.4 ng/mL | y = 0.981x + 0.795 | 0.982 | The predicted bias at medical decision point (30 ng/mL) was calculated as the percent difference between the Deming regression line and the unity line at the medical decision point and was found to be 0.8%. 2. Matrix Comparison: A matrix equivalency study to serum was conducted to support use of the Elecsys Vitamin D total III assay on the cobas e 601 analyzer with additional specimen matrix types claimed in the product labeling: Serum Separation Tubes, and Li-Heparin plasma, K2-EDTA plasma, and K3-EDTA plasma. In the study, 46 donor matched venous specimens of the aforementioned were collected. Each specimen was tested in singlicate using one lot of the reagent kit and one instrument, and the results compared to the mean of singlicate serum measurements. The results were analyzed by Passing-Bablok linear regression with the K210901 - Page 9 of 11 {9} concentration from the first replicate of each donor's evaluation tube (y-axis) versus the mean concentration of the serum results (x-axis). The slope and intercept of the regression line were calculated, and summarized as follows: | Specimen | Range | r | Intercept | Slope | | --- | --- | --- | --- | --- | | Li-Heparin plasma | 8.28 to 103 ng/mL | 0.985 | -0.62 | 1.03 | | K2-EDTA plasma | 9.27 to 99.8 ng/mL | 0.957 | 0.072 | 0.973 | | K3-EDTA plasma | 9.32 to 100 ng/mL | 0.990 | 0.14 | 0.92 | | Serum Separation Tube | 8.88 to 107 ng/mL | 0.987 | -0.26 | 1.02 | The study results support the sponsor's claim that human serum and plasma (lithium heparin, K2-EDTA and K3 EDTA) are acceptable sample types to be used with this assay. # C Clinical Studies: 1. Clinical Sensitivity: Not applicable. 2. Clinical Specificity: Not applicable. 3. Other Clinical Supportive Data (When 1. and 2. Are Not Applicable): Not applicable. # D Clinical Cut-Off: Not applicable. # E Expected Values/Reference Range: A reference range study with Elecsys Vitamin D total III assay on the cobas e 601 analyzer was conducted testing serum samples collected from apparently healthy donors from southern, middle, and northern sites of the United States during both summer and winter. There were approximately equal numbers of males and females, and approximately $30\%$ of the donors had dark skin complexion. The age range was 22 to 79 years. The results of the study are summarized as follows: | | | Seasonal | | | --- | --- | --- | --- | | | All (n = 463) | Summer (n = 245) | Winter (n = 218) | | Mean, ng/mL | 26.6 | 29.2 | 23.6 | | Median | 25.7 | 27.7 | 22.8 | | 2.5th percentile | 10.2 | 12.5 | 9.4 | | 97.5th percentile | 49.4 | 52.4 | 44.1 | K210901 - Page 10 of 11 {10} VIII Proposed Labeling: The labeling supports the finding of substantial equivalence for this device. IX Conclusion: The submitted information in this premarket notification is complete and supports a substantial equivalence decision. K210901 - Page 11 of 11