NEONATAL TOTAL GALACTOSE KIT; NEONTAL TOTAL GALACTOSE KIT, MODELS 3029-0010 AND 3029-110B

K071649 · Wallac OY · JIA · Aug 7, 2008 · Clinical Chemistry

Device Facts

• Record ID: K071649
• Device Name: NEONATAL TOTAL GALACTOSE KIT; NEONTAL TOTAL GALACTOSE KIT, MODELS 3029-0010 AND 3029-110B
• Applicant: Wallac OY
• Product Code: JIA · Clinical Chemistry
• Decision Date: Aug 7, 2008
• Decision: SESE
• Submission Type: Traditional
• Regulation: 21 CFR 862.1310
• Device Class: Class 1
• Attributes: Pediatric

Intended Use

This kit is intended for the quantitative determination of total galactose (galactose and galactose-1-phosphate) concentrations in blood specimens dried on filter paper as an aid in screening newborns for galactosemia.

Device Story

The Neonatal Total Galactose Kit is an in vitro diagnostic assay for screening newborns for galactosemia. It utilizes dried blood spot specimens collected on filter paper. The device employs a fluorescent enzymatic method: alkaline phosphatase converts galactose-1-phosphate to galactose; galactose oxidase then converts galactose to D-galacto-hexadialdose and hydrogen peroxide; finally, peroxidase reacts hydrogen peroxide with 3-(p-hydroxyphenyl)propionic acid (HPPA) to produce a fluorescent dimer. The fluorescence intensity is proportional to the total galactose concentration. The assay is performed using a Victor2 D Fluorometer or equivalent plate reader. Results are interpreted against a standard curve generated from six calibrator levels. Healthcare providers use these quantitative results to identify newborns requiring further diagnostic follow-up for galactosemia. The device provides a screening tool to facilitate early detection and intervention.

Clinical Evidence

Method comparison study using 2,109 prospective newborn specimens and 7 retrospective high-galactose samples (3 confirmed galactosemia). Compared against a commercially available kit. Overall percent agreement was 95.9% at the 95th percentile and 99.2% at the 99th percentile. Analytical performance included precision (CVs 8.4-12.7%), linearity (1.3-56 mg/dL), and LoD (1.3 mg/dL). Interference noted for lipids (over-recovery) and high glutathione (under-recovery).

Technological Characteristics

Enzymatic fluorescent assay. Components: calibrators/controls (human blood on Whatman 903 paper), Zinc Sulfate, Reconstitution Buffer, Substrate Reagent, Galactose Oxidase, Stop Solution. Requires fluorometer (340/405 nm). Linearity: 1.3-56 mg/dL. Traceable to HPLC-MS/MS. No software algorithm class specified; manual/instrument-based measurement.

Indications for Use

Indicated for the quantitative determination of total galactose (galactose and galactose-1-phosphate) in dried blood spot specimens from newborns as an aid in screening for galactosemia.

Regulatory Classification

Identification

A galactose test system is a device intended to measure galactose in blood and urine. Galactose measurements are used in the diagnosis and treatment of the hereditary disease galactosemia (a disorder of galactose metabolism) in infants.

Predicate Devices

• Quantase Total Galactose Screening Assay (k990654)

Submission Summary (Full Text)

{0}

1

# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY ASSAY ONLY TEMPLATE

A. 510(k) Number:
k071649

B. Purpose for Submission:
New device

C. Measurand:
Galactose and galactose-1-phosphate

D. Type of Test:
Quantitative, fluorescent galactose oxidase method

E. Applicant:
Wallac Oy

F. Proprietary and Established Names:
Neonatal Total Galactose Kit

G. Regulatory Information:

| Product Code | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| JIA | Class I | 21 CFR 862.1310 - Galactose test system | Chemistry |

H. Intended Use:

1. Intended use(s):
This kit is intended for the quantitative determination of total galactose (galactose and galactose-1-phosphate) concentrations in blood specimens dried on filter paper as an aid in screening newborns for galactosemia.

2. Indication(s) for use:
See intended use above.

3. Special conditions for use statement(s):
For prescription use only, presumptive positive results are to be tested using a confirmatory diagnostic method.

{1}

2

4. Special instrument requirements:

The Neonatal Total Galactose kit requires a fluorometer capable of measuring fluorescence using excitation central wavelength of 340 nm (or 320 nm) and emission central wavelength of 405 nm on 96 well microtiter plates. Performance characteristics were determined using the PerkinElmer 1420 Victor D fluorometer.

I. Device Description:

Each Neonatal Total Galactose kit contains reagents for either 960 assays or 4800 assays. The kit contains the following components:

Galactose calibrators and controls- Prepared from washed human blood, with preservative, adjusted to a hematocrit of 55% and spotted onto Whatman 903 specimen collection paper.

The six calibrators contain galactose at the following concentrations:

A 0 mg/dL
D 9.0 mg/dL
B 1.5 mg/dL
E 18 mg/dL
C 4.0 mg/dL
F 40 mg/dL

The two Normal and Abnormal Controls contain galactose and galactose-1-phosphate at approximately 190 µmol/L blood (3.5 mg/dL) and 530 µmol/L blood (9.5 mg/dL) respectively.

All human source materials used in the preparation of kit components was tested and found to be non-reactive for the presence of HBsAg, anti-HIV 1 and 2, and HCV by FDA approved methods.

Zinc Sulfate Reagent -1 bottle, 30 mL
Galactose Reconstitution Buffer- 1 bottle, 240 mL
Galactose Substrate Reagent -10 bottles, dried
Galactose Oxidase Reagent- 10 bottles, lyophilized
Stop Solution- 1 bottle, 240 mL

J. Substantial Equivalence Information:

1. Predicate device name(s):
Quantase Total Galactose Screening Assay

2. Predicate K number(s):
k990654

3. Comparison with predicate:

{2}

| Similarities | | |
| --- | --- | --- |
| Item | Device | Predicate |
| Intended use | For the quantitative determination of total galactose (galactose and galactose-1-phosphate) concentrations in blood specimens dried on filter paper as an aid in screening newborns for galactosemia | Same |
| Test methodology | Enzymatic assay | Same |
| Differences | | |
| --- | --- | --- |
| Item | Device | Predicate |
| Reportable range | 1.3-56 mg/dL (approximate) | 0.6-55 mg/dL |
| Detection method | Fluorescence- measured at 340 nm and 405 nm wavelengths | Colorimetric- absorbance read at 570 nm and 690 nm wavelengths |

# K. Standard/Guidance Document Referenced (if applicable):

-Evaluation of the Linearity of Quantitative Measurement Procedures: A Statistical Approach; Approved Guideline (CLSI EP6-A)
- Protocols for Determination of Limits of Detection and Limits of Quantitation; Approved Guideline (CLSI EP17-A).

# L. Test Principle:

The Neonatal Total Galactose assay measures total galactose, i.e. both galactose and galactose-1-phosphate, using a fluorescent galactose oxidase method. The fluorescence is measured using excitation central wavelength of $340\mathrm{nm}$ (or $320\mathrm{nm}$ ) and emission central wavelength of $405\mathrm{nm}$ .

# M. Performance Characteristics (if/when applicable):

# 1. Analytical performance:

# a. Precision/Reproducibility:

Two studies were performed to evaluate precision of the Neonatal Total Galactose test. Samples were prepared from adult human whole blood spiked with different concentrations of galactose. The samples were then dispensed onto filter paper and dried at room temperature overnight. Since the testing was performed using dried blood spots, the extraction procedure was captured

{3}

in the evaluation process.

In study 1, three levels of spiked dried blood samples were assayed in replicates of four in 18 runs over 6 days. The testing was conducted using three kit lots, two fluorometers, and three operators. The total number of replicates was $n = 144$ . The results are summarized in the table below.

Study 1

| Sample | Galactose mg/dL | Within plate variation (CV%) | Total within lot variation (CV%) | Between lot variation (CV%) | Total variation (CV%) |
| --- | --- | --- | --- | --- | --- |
| 1 | 7.1 | 6.6 | 12.1 | 2.1 | 12.3 |
| 2 | 13.8 | 6.5 | 10.6 | 2.4 | 10.8 |
| 3 | 20.2 | 5.3 | 8.6 | 0.4 | 8.6 |

In study 2, four levels of spiked dried blood samples were assayed in replicates of four in 20 runs over 20 days. The testing was conducted using one kit lot, one fluorometer, and one operator. The results are summarized in the table below.

Study 2

| Sample | Galactose mg/dL | Within plate variation (CV%) | Total variation (CV%) |
| --- | --- | --- | --- |
| A | 3.5 | 7.8 | 12.7 |
| B | 11 | 7.2 | 11.8 |
| C | 23 | 6.0 | 8.4 |
| D | 36 | 5.2 | 8.6 |

# b. Linearity/assay reportable range:

The study was performed following the CLSI protocol EP6-A. Whole blood heparinized sample pools were spiked with analyte to create samples with high concentrations of galactose and galactose-1-phosphate. The high samples were mixed in varying proportions with the sample pools containing very low galactose to create a dilution series. Each test sample was spotted on filter paper in quadruplicate $(n = 4)$ , dried overnight, and tested.

The data was analyzed using linear regression as well as second and third order non-linear fitted polynomial regression. The second order model fit the data better than the linear model. However, for all dilution points the relative differences between the two models were within the sponsor's defined acceptance criterion of $+/- 25\%$ . The results are summarized below.

{4}

Fitted regression models are:

Linear: $y = 1.6 + 52x$

Second order: $y = 1.3 + 58x - 7.4x^{2}$

Third order: $y = 1.2 + 60x - 16x^{2} + 7.5x^{3}$

Where $y =$ Total galactose concentration (mg/dL) and $x =$ Dilution point.

| Dilution point | Predicted 1storder mg/dL | Predicted 2ndorder mg/dL | Absolute difference mg/dL | Relative difference (%) |
| --- | --- | --- | --- | --- |
| 1 | 53.5 | 51.4 | -2.1 | -4 |
| 0.9 | 48.3 | 47.0 | -1.3 | -3 |
| 0.85 | 45.7 | 44.8 | -0.9 | -2 |
| 0.8 | 43.1 | 42.6 | -0.6 | -1 |
| 0.6 | 32.7 | 33.1 | 0.4 | 1 |
| 0.35 | 19.7 | 20.5 | 0.8 | 4 |
| 0.2 | 12.0 | 12.5 | 0.5 | 5 |
| 0.12 | 7.79 | 8.08 | 0.29 | 4 |
| 0.08 | 5.72 | 5.84 | 0.12 | 2 |
| 0.05 | 4.16 | 4.14 | 0.12 | 0 |
| 0.03 | 3.12 | 3 | -0.11 | -4 |
| 0 | 1.56 | 1.29 | -0.27 | -17 |

The maximum observed percent difference $(-17\%)$ between the models was seen with the undiluted low level sample which is near the limit of detection of the test. The relative differences $(\%)$ of all other dilutions points were within $\pm 5\%$ . Therefore, the Neonatal Total Galactose kit is linear in the range $1.3\mathrm{mg / dL}$ to approximately $56\mathrm{mg / dL}$ .

The values of the highest kit calibrator may vary slightly but does not exceed $56\mathrm{mg / dL}$ . The package insert states that samples with values above the highest calibrator are not accurate and should be reported as presumptive positives.

c. Traceability, Stability, Expected values (controls, calibrators, or methods): The kit calibrators are traceable to a validated HPLC-MS/MS method. Six dried blood spot calibrators and two dried blood spot controls, normal and abnormal, are provided with the kit. Real-time stability including shelf-life, transport, and in-use stability studies for the entire kit are performed, including the controls and calibrators. Additionally, accelerated stability studies are also performed separately for the kit calibrators and controls. The study protocols and statistically calculated acceptance limits were reviewed and found to be acceptable.

{5}

d. Detection limit:

The sponsor states that the limit of blank (LoB) study was performed in accordance with the recommendations of CLSI EP-17. The LoB was calculated from data using two kit lots over five testing days and a total of 100 measurements of the zero calibrator. Calibrators and controls were tested in duplicate for each assay. Since the assay does not report values less than zero, the LoB was estimated non-parametrically, as CLSI EP-17 advises, as the 95th percentile of the measurements. The limit of blank was observed to be 0.7 mg/mL.

To estimate the limit of detection (LoD), five samples with low levels of galactose were also assayed in the same study for a total of 100 measurements at each level. These samples were prepared from heparinized whole blood with added galactose and galactose-1-phosphate, spotted onto filter paper, and dried. Statistical analysis yielded a limit of detection of 1.3 mg/dL.

The package insert states that values below 1.3 mg/dL should be reported as “<1.3 mg/dL.”

e. Analytical specificity:

Interference from elevated lipid, bilirubin, and hemoglobin was evaluated at a galactose concentration of approximately 10 mg/dL. The sponsor used a two sample t-test at the 95% confidence interval to evaluate the significance of any difference seen between test and control samples.

Bilirubin at concentrations ranging from 10-40 mg/dL, hemoglobin at 20 and 25 g/dL, and Intralipid at concentrations ranging from 0.25-1.0 g/dL were separately was added to whole blood specimens containing galactose, spotted onto filter paper, dried, and tested. Specimens without added interferant were also tested. Specimens containing the highest levels of bilirubin and hemoglobin did not interfere with the galactose test using the sponsor's acceptance criteria. Conversely, specimens containing lipid interfered with the accurate measurement of galactose across all levels tested and resulted in an over-recovery of galactose up to 30%. This observation is likely due to the physical nature of lipid in that it interferes with the transmission of light. The results are summarized in the package insert.

The following substances were tested and found not to interfere at the levels stated: glucose (1200 mg/dL), mannose (1-10 mg/dL), fructose (25 mg/dL), ascorbate (0.1-3 mg/dL), protein BSA (12 g/dL), and acetaminophen (0.1-1 mg/dL). Glutathione at concentrations up to 6 mg/dL did not interfere, however a concentration of 60 mg/dL, which far exceeds the expected concentration in newborns, caused under-recovery of approximately 40%. This is likely due to glutathione's chemical properties and interference in the assay's enzymatic reaction. The results are summarized in the package insert.

{6}

f. Assay cut-off: Not applicable.

## 2. Comparison studies:

a. Method comparison with predicate device:

A total of 2109 specimens from newborns were obtained prospectively from a state public health laboratory's routine newborn screening program. Additionally, seven (7) specimens were also tested that were retrospective high galactose samples of which three (3) were from newborns with diagnosed galactosemia. Dried blood spots were assayed using the Perkin Elmer total galactose kit and a commercially available comparator method. The presumptive positive cut-off was determined from the results using each kit's 95th and 99th percentiles. For the Perkin Elmer kit these values were 5.1 mg/dL and 8.3 mg/dL respectively, while the values for the comparator method were 4.2 mg/dL and 9.4 mg/dL, respectively. The results are summarized in the tables below.

## 95th Percentile Results

| Screening summary using the 95th percentile | Commercially available kit (4.2 mg/dL) | | | |
| --- | --- | --- | --- | --- |
| | | + | - | Total |
| PerkinElmer kit (5.1 mg/dL) | + | 73 | 46 | 119 |
| | - | 41 | 1956 | 1997 |
| | Total | 114 | 2002 | 2116 |
| Screening summary using the 95th percentile | | | | |
| --- | --- | --- | --- | --- |
| Commercially available kit | PerkinElmer kit | Total | Diagnosed galactosemia | No diagnosed galactosemia |
| + | + | 73 | 3 | 70 |
| + | - | 41 | 0 | 41 |
| - | + | 46 | 0 | 46 |
| - | - | 1956 | 0 | 1956 |
| Total | | 2116 | 3 | 2113 |

The overall percent agreement using the 95th percentile is 2029/2116 or 95.9%.

{7}

8

# 99th Percentile Results

| Screening summary using the 99th percentile | Commercially available kit (9.4 mg/dL) | | | |
| --- | --- | --- | --- | --- |
| | | + | - | Total |
| PerkinElmer kit (8.3 mg/dL) | + | 21 | 9 | 30 |
| | - | 8 | 2078 | 2086 |
| | Total | 29 | 2087 | 2116 |
| Screening summary using the 99th percentile | | | | |
| --- | --- | --- | --- | --- |
| Commercially available kit | PerkinElmer kit | Total subjects | Diagnosed galactosemia | No diagnosed galactosemia |
| + | + | 21 | 3 | 18 |
| + | - | 8 | 0 | 8 |
| - | + | 9 | 0 | 9 |
| - | - | 2078 | 0 | 2078 |
| Total | | 2116 | 3 | 2113 |

The overall percent agreement using the 99th percentile is 2099/2116 or 99.2%.

The results of the three confirmed true galactosemia samples are shown below.

| Total Galactose concentrations of the true positive specimens | | |
| --- | --- | --- |
| Specimen no. | PerkinElmer kit (mg/dL) | Commercially available kit (mg/dL) |
| 1 | 16.6 | 23.9 |
| 2 | 12.1 | 14.6 |
| 3 | 22.2 | 39.1 |

b. Matrix comparison:

Not applicable. The device is to be used only with neonatal whole blood.

3. Clinical studies:

a. Clinical Sensitivity:

Not applicable.

{8}

b. Clinical specificity:

Not applicable.

c. Other clinical supportive data (when a. and b. are not applicable):

4. Clinical cut-off:

Not applicable.

5. Expected values/Reference range:

The summary of the results from the 2109 newborn specimens tested with this kit is shown in the table below:

| | μmol/L | mg/dL |
| --- | --- | --- |
| Median | 94 | 1.7 |
| 95th percentile | 283 | 5.1 |
| 97.5th percentile | 355 | 6.4 |
| 99th percentile | 461 | 8.3 |
| 99.5th percentile | 549 | 9.9 |

The package insert includes precautionary language that each laboratory should establish their own reference range and cut-off values and that cut-offs from another galactose screening test should not be used. Also included in the labeling is the recommendation that presumptive positive samples be tested with a confirmatory diagnostic method and to follow local requirements for follow up testing.

N. Proposed Labeling:

The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.

O. Conclusion:

The submitted information in this premarket notification is complete and supports a substantial equivalence decision.