The AquaLite® Prolactin Bioluminescent Immunoassay (BIA) Kit (or the AquaLite® Prolactin assay) is intended to be used in clinical laboratories for the quantitative determination of human prolactin levels in sera and plasma. The AquaLite® Prolactin assay is for in vitro diagnostic use.

Device Description

The AquaLite® Prolactin Bioluminescent Immunoassay Kit uses a mixture of mouse monoclonal and rabbit polyclonal with anti-prolactin activity that is pre-coated onto polystyrene tubes (solid phase). Samples (serum or plasma), or appropriate calibrators or controls, are pipetted (25 uL) into the pre-coated tubes. Anti-prolactin conjugate consisting of mouse monoclonal antibody covalently linked to AquaLite® (150µL) is then added to the tubes. The conjugate uses the photoprotein, AquaLite® (recombinant aequorin: U.S. Patent Nos. 5,422,266 and 5,486,455) which is covalently linked to an anti-prolactin polyclonal antibody. Prolactin in the sample simultaneously combines with polyclonal antibody on the solid phase and conjugate antibody to form an immune complex or "sandwich" bound to the solid phase. Complex formation is complete after a 60-minute incubation period at room temperature (18° to 25°C) on a standard orbital shaker. The tubes are then washed to remove unbound conjugate.

The washed tubes are placed in a luminometer that is capable of reading a triggered. flash-type reaction in 12 x 75 mm tubes. An injected calcium trigger solution causes AquaLite® to oxide its self-contained luciferin molecule. This reaction produces a twosecond flash of light at 469 nm, which is measured by the luminometer. The intensity of the light is directly proportional to the concentration of the prolactin in the sample. To calculate results, the luminometer uses a cubic spline curve fit applied to a logit-log transformation of the light intensity (in relative light units, RLU) of the prolactin calibrators versus prolactin concentration (in ng/mL).

Note: Samples that generate signals greater than the signal from the highest calibrator are off-scale. These samples must be diluted and re-assayed. Remember to multiply the results from diluted samples by the dilution factor used.

AI/ML Overview

Here's a breakdown of the acceptance criteria and study details for the SeaLite Sciences, Inc. AquaLite® Prolactin Bioluminescent Immunoassay (BIA) Kit:

1. Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria Category	Specific Metric (Unit)	Acceptance Criterion	Reported Device Performance
Sensitivity	Detection Limit (ng/mL)	Not explicitly stated (implied to be low)	0.01 ng/mL
Specificity	Cross-reactivity (%)	Not explicitly stated (implied to be low for non-prolactin hormones)	rh: <0.007 FSH: Not detectable TSH: <0.001 hCG: Not detectable hPL: 0.38% hGH: 0.52%
High Dose Hook Effect	Prolactin Level (ng/mL)	No hook effect below the highest calibrator level (implied)	No high dose hook effect occurs prior to 1,000 ng/mL prolactin.
Precision (Intra-assay)	% CV	Not explicitly stated (implied to be acceptably low)	13.6 ng/mL: 7.3% 29.1 ng/mL: 7.2% 68.7 ng/mL: 8.9%
Precision (Inter-assay)	% CV	Not explicitly stated (implied to be acceptably low)	12.8 ng/mL: 8.17% 28.1 ng/mL: 8.96% 67.8 ng/mL: 9.76%
Method Comparison	Correlation CoefficientSlopeY-intercept	Not explicitly stated (implied to be strong correlation with existing assay)	0.9669 0.8665 -1.9869
Linearity & Nonparallelism	Recovery (%)	Not explicitly stated (implied to be within an acceptable range for diluted samples)	Sample #22: 100%-124% Sample #51: 99%-111% Sample #96: 84%-109%
Spike and Recovery	Percent Recovered (%)	Not explicitly stated (implied to be within an acceptable range)	Sample 1: 94% Sample 2: 104% Sample 3: 91.5% Sample 4: 94.5% Sample 5: 92.5%
Recovery in Serum and Plasma	Relative Recovery (%)	No significant differences among serum and SST serum nor serum and heparin, EDTA, and citrate plasmas. Oxalate plasma is not recommended.	Ranges from 82% to 116% observed across different matrix types, with specific recommendation against oxalate plasma.

2. Sample Size Used for the Test Set and Data Provenance

Sensitivity: 20 replicates of the zero-level calibrator. Data provenance is internal to SeaLite Sciences, Inc.
Specificity: Not explicitly stated how many aliquots or how many times each substance was tested, but they were supplied by the World Health Organization's National Institute for Biological Standards and Control (London, England).
Precision (Intra-assay): N = 20 per concentration level for tri-level commercial controls. Data provenance is internal to SeaLite Sciences, Inc.
Precision (Inter-assay): N = 2 x 10 = 20 (duplicate assays across 10 sets of calibration values) for tri-level commercial controls. Data provenance is internal to SeaLite Sciences, Inc.
Method Comparison: N = 100 patient samples. Data provenance is internal to SeaLite Sciences, Inc., as these were patient samples previously assayed by another method. This would be considered retrospective in the context of the AquaLite Prolactin study.
Linearity and Nonparallelism: Three human serum samples, each diluted at 4 levels, assayed in duplicate. Data provenance is internal to SeaLite Sciences, Inc.
Spike and Recovery: Five normal male human serum samples. Data provenance is internal to SeaLite Sciences, Inc.
Recovery in Serum and Plasma: Blood samples from 4 normal subjects, prepared into various matrices. Data provenance is internal to SeaLite Sciences, Inc.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and the Qualifications of Those Experts

This device is an in vitro diagnostic (IVD) assay for measuring prolactin levels. The "ground truth" for such assays is typically established by:

Reference materials: For specificity, WHO/NIBSC standards were used.
Known concentrations: For sensitivity, linearity, spike and recovery, and precision, predefined calibrators, controls, or spiked samples with known prolactin concentrations were used.
Comparison to a legally marketed predicate device: For method comparison, the results were compared to a "commercially available automated chemiluminescent immunoassay."

Therefore, the concept of "experts establishing ground truth for a test set" as one might find in image analysis (e.g., radiologists reviewing images) does not directly apply here. Instead, it relies on the chemical and biological properties of the assay and comparison to established and validated reference methods/materials.

4. Adjudication Method for the Test Set

Not applicable. As this is an IVD assay measuring an analyte, there isn't a subjective "adjudication" process in the way there would be for image interpretation by multiple human readers. The results are quantitative measurements.

5. If a Multi Reader Multi Case (MRMC) Comparative Effectiveness Study Was Done, What was the Effect Size of How Much Human Readers Improve with AI vs Without AI Assistance

Not applicable. This is not an AI-assisted diagnostic device or an imaging study that involves human readers interpreting cases. It is a standalone in vitro diagnostic assay.

6. If a Standalone (i.e. Algorithm Only Without Human-in-the-Loop Performance) Was Done

Yes, the studies presented are all standalone performance evaluations of the AquaLite® Prolactin assay itself. The device is designed to quantitatively determine prolactin levels without direct human interpretation of the results beyond reading the numerical output.

7. The Type of Ground Truth Used

Known concentrations: For sensitivity, linearity, spike and recovery, and precision, the ground truth was based on samples with experimentally determined or certified prolactin concentrations (e.g., calibrators, commercial controls, spiked samples).
Reference materials: For specificity, internationally recognized reference preparations from the World Health Organization's National Institute for Biological Standards and Control (WHO/NIBSC) were used.
Predicate device comparison: For method comparison, a "commercially available automated chemiluminescent immunoassay" was used as a reference for patient sample measurements.

8. The Sample Size for the Training Set

Not applicable. This is a traditional immunoassay, not a machine learning or AI-based device that requires a "training set" in the computational sense. The "training" of the assay refers to its chemical and biological formulation and optimization during development, before the performance studies presented here.

9. How the Ground Truth for the Training Set Was Established

Not applicable. See point 8.

Summary

{0}------------------------------------------------

K970188

510(k) SUMMARY

I. GENERAL INFORMATION

Trade or proprietary name - SeaLite Sciences, Inc. AquaLite® Prolactin

Common or usual name - Bioluminescent immunoassay (BIA)

Classification name - FDA has classified prolactin test systems intended for the measurement of prolactin in the diagnosis of hypothalamic and pituitary function as Class I devices. 21 C.F.R. §862.1625.

Submitter's Name and Address:	Cathryn N. CambriaDirector, Regulatory affairsand Quality AssuranceSeaLite Sciences, Inc.3000 Northwoods ParkwaySuite 200Norcross, GA 30071(800) 874-4471, x227
Submission Date:	January 15, 1997
Legally Market Device to WhichClaim Substantial Equivalence:	Chiron Diagnostics ACS 180Prolactin assay

II. INDICATIONS FOR USE

{1}------------------------------------------------

ਹ ਅੰ

SeaLite Sciences, Inc.

DEVICE DESCRIPTION III.

IV. SUMMARY OF STUDIES AND TECHNOLOGICAL CHARACTERISTICS

Studies on SeaLite Sciences, Inc. AquaLite® Prolactin were conducted at SeaLite Sciences. The results are summarized below:

Performance Characteristics

1. Sensitivity

The sensitivity or detection limit of the AquaLite® Prolactin is 0.01 ng/mL. Sensitivity is determined by adding the mean signal of twenty (20) replicates of

{2}------------------------------------------------

the zero level calibrator plus two (2) standard deviations above this mean. The prolactin concentration (ng/mL) associated with this calculated signal is defined as the sensitivity of the assay.

2. Specificity

The AquaLite® Prolactin measures intact prolactin. The following human sialoglycoprotein hormones were supplied by the World Health Organization's National Institute for Biological Standards and Control (London, England). Aliquots of these preparations were diluted to the following levels in Calibrator A and assayed. Percent cross-reactivity (%) is reported below:

		Tested
Substance	Lot Number		Value	% Cross-reactivity
rh		2nd I.S. 80/552	1000 ng/mL	<0.007
FSH	15t I.S.	83/575	1000 ng/mL	Not detectable
TSH	2nd IRP	80/228	1000 ng/mL	<0.001
hCG		3rd I.S. 75/537	1000 ng/mL	Not detectable
hPL		1st I.R.P. 73/545	12 ug/mL	0.38
hGH	156 I.S. I	80/505	500 ng/mL	0.52

3. High Dose Hook Effect

No high dose hook effect occurs prior to 1,000 ng/mL prolactin.

4. Precision

(a) Intra-assay precision. Tri-level commercial controls containing prolactin at the following concentrations were assayed to determine intra-assay precision. (Total N = 20 per concentration level.)

Prolactin Level (ng/ml)	% CV
13.6	7.3%
29.1	7.2%
68.7	8.9%

{3}------------------------------------------------

SeaLite Sciences, Inc.

Inter-assay precision. Tri-level commercial controls containing prolactin (b) at the following concentrations were assayed in duplicate repetitively. Ten assays were performed using ten sets of calibration values. The interassay precision observed for the solutions (Total n = 2 x 10 = 20) are shown below.

Prolactin Level(ng/ml)	% CV
12.8	8.17%
28.1	8.96%
67.8	9.76%

5. Method Comparison

The AquaLite® Prolactin was used to assay patient samples (N=100) that were previously assayed by a commercially available automated chemiluminescent immunoassay. A slope of 0.8665 with a y-intercept of -1.9869 was obtained. The correlation coefficient was 0.9669.

6. Linearity and Nonparallelism

Three human serum samples containing the levels of endogenous prolactin shown below were diluted in parallel as indicated using Calibrator A (0 ng/mL) and assayed in duplicate using AquaLite® Prolactin. All concentrations are in ng/mL.

{4}------------------------------------------------

SampleID	DilutionFactor	Found(ng/mL)	Expected(ng/mL)	Recovery(%)
#22	Undiluted1:21:41:8	93.847.0929.0513.44	--46.9023.4511.73	--100%124%115%
#51	Undiluted1:21:41:8	109.3854.3830.4514.41	--54.6927.3513.67	--99%111%105%
#96	Undiluted1:21:41:8	97.5343.7826.6610.29	--48.7724.3812.19	--90%109%84%

7. Spike and Recovery

Five normal male human serum samples were spiked to 25.7ng/mL prolactin using WHO prolactin (3rd I.S. 84/500). The spiked samples were assayed using the AquaLite® Prolactin. All values are in ng/mL.

SampleID	Prolactinng/mL	ProlactinObserved	ProlactinExpected	PercentRecovered
1	13.4	36.8	39.1	94%
2	7.63	34.5	33.3	104%
3	17.7	39.7	43.4	91.5%
4	21.9	45.0	47.6	94.5%
5	12.5	35.9	38.8	92.5%

{5}------------------------------------------------

8. Recovery in Serum and Plasma

Blood samples from 4 normal subjects were prepared as sera (standard technique and SST tubes) as well as heparin, EDTA, citrate and oxalate plasmas. Prolactin was quantified using the AquaLite® Prolactin. Recovered prolactin was compared with the prolactin recovered in serum (standard technique). The data demonstrate that there are no significant differences among serum and SST serum nor serum and heparin, EDTA, and citrate plasmas when using the AquaLite® Prolactin. Oxalate plasma is not recommended.

Sample	Serum	SST	%	Hep	%	EDTA	%	Citrate	%	Oxalate	%
A	19.4	21.4	110	18.6	96	19.9	103	18.3	94	16.0	82
B	11.7	12.1	103	11.3	97	10.7	91	10.7	91	9.93	85
C	15.0	15.5	103	17.4	116	13.2	88	13.4	89	14.5	97
D	12.1	11.8	98	12.1	100	11.1	92	10.3	85	10.3	85

V. ALTERNATIVE PRACTICES AND PROCEDURES

There are several assay technologies commonly employed to measure the presence of human prolactin in serum. They include: radioimmunoassay (RIA) and enzyme-linked immunosorbent immunoassay (ELISA).

VI. POTENTIAL ADVERSE EFFECTS OF THE DEVICE ON HEALTH

Use Universal Precautions. No known test method can offer complete assurance that products derived from human serum are pathogen-free; therefore, handle all materials of human origin as though they were potentially infectious.

Sodium azide is used as a preservative. This preservative may react with metallic plumbing to from explosive metal azides. Flush with large volumes of water when disposing of materials containing sodium azide.

As an in vitro diagnostic test, there are not direct adverse effects on the health of a patient from the use of this product. However, failure of the device to perform as indicated, the contamination of reagents, the use of reagents past the labeled expiration dates, the use of improper specimens, or human error during the performance of the test may lead to erroneous results and possible improper patient management.

Regulation Number and Section

§ 862.1625 Prolactin (lactogen) test system.

(a)
Identification. A prolactin (lactogen) test system is a device intended to measure the anterior pituitary polypeptide hormone prolactin in serum and plasma. Measurements obtained by this device are used in the diagnosis and treatment of disorders of the anterior pituitary gland or of the hypothalamus portion of the brain.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 862.9.