saliva diagnostics for kidney disease-presentation · creatinine test •typically ordered as part...
TRANSCRIPT
Saliva Diagnostics for Kidney Disease
Linden Heflin & Sarah Walsh
May 3, 2007
5/2/2007 2
Overview
• Advantages of Saliva Diagnostics
• Background
• FDA Approval
• Consumer Satisfaction
• Consumer Preference
• Pricing and Demand
• Net Present Worth
5/2/2007 3
Alternative to Blood Testing
• Saliva contains many of the components of blood– In much lower concentration
– Mostly items that passively diffuse through salivary glands
• Collecting saliva is much less invasive and faster– Reduces risk of exposure for health care professional
– Saliva has fewer components that interfere with assay, reducing steps needed in analysis
• Some people refuse blood testing for cultural or religious reasons
5/2/2007 4
The Beginning of Saliva Diagnostics
• MicrofluidicsMicrofluidicsMicrofluidicsMicrofluidics
– A new field which began in the early 90’s
– Combination of physics, chemistry, biotechnology, and engineering
– Develops a better understanding of how fluids move on a micro and nanoliter scale
• Allows for the design of more sensitive diagnostic devices
5/2/2007 5
Current State of Saliva Diagnostics
• Conditions presently being assessed using saliva– Alcohol consumption
– Drug use
– Hormone levels
– HIV 1 and 2
– Viral hepatitis A, B, and C
• Current research– Cardiovascular disease
– Cancer
– Alzheimer’s
– Osteoporosis
5/2/2007 6
Saliva
• Saliva has many components
– Mostly water with some mucus
– A variety of electrolytes (K+, Na+, Cl-, Ca+)
• Many proteins found in blood also make their way into saliva
5/2/2007 7
Saliva Composition• 98% water• Electrolytes
– Sodium ~32 mmol/L– Potassium ~22 mmol/L– Calcium ~1.7 mmol/L– Magnesium ~0.18 mmol/L– Copper ~0.4 µmol/L– Lead ~0.55 µmol/L– Cobalt ~1.2 µmol/L– Strontium ~1 µmol/L– Hydrogen Carbonate
~20 mmol/L– Iodide ~10 µmol/L– Bromide ~14 mmol/L– Hypothiocyanate ~1.2 µmol/L– Nitrate ~1.1 µmol/L– Nitrite ~178 µmol/L– Fluoride ~68 µmol/L– Sulfate ~5.8 µmol/L
• Mucus– Mucopolysaccharides– Glucose ~175 µmol/L
• Metabolites– Bilirubin ~15 µmol/L– α-ketoglutaric acid
~2.4 µmol/L– Pyruvic acid ~75 µmol/L
• Proteins– α-amylase ~650-800
µg/ml– Peroxidase ~5-6 µg/ml– Secretory IgA ~96-102
µg/ml– Lactoferrin ~1-2 µg/ml– Fibronectin ~0.2-2 µg/ml
• Cells
5/2/2007 8
Our Screening Procedure
• Can biomarker be detected in saliva
• Abnormal levels indicate threat of organ malfunction
• How do you detect abnormal levels
• How accurate are detection methods
• How widely applicable are detection methods
• How helpful is result in medical decision making
• Is test effective in early diagnosis (compared to serum testing)
• Weigh accuracy vs. speed, convenience, portability
• Cost of detection method
• Making product attractive to consumer
5/2/2007 9
Screening Procedure Flow ChartDoes organ malfunction have salivary
biomarkers?No, back to beginning
Yes, investigate possibility of false negatives
No, back to beginning
Is it accurate enough?
Can it compete with serum tests?
Speed
Patient Comfort
AccuracySerum tests more desirable
Saliva is superior
Essentially equal
Is it useful for early diagnosis?
Useful as precursor to more invasive tests
No, back to beginning
No, back to beginning
Preliminary EconomicAnalysis
Not Profitable, back to beginning
Profitable
FDA Pre-Market Notification
5/2/2007 10
The Kidney• The kidney’s responsibility is to clean the blood
• Most waste in blood passively diffuses in the kidney, just like most of saliva’s components come from passive diffusion
• About 1 in 12 people have some kidney disease– 9th leading cause of death in USA
– 80,000 deaths per year
• About 450,000 people depend on dialysis or kidney transplants to live
5/2/2007 11
Symptoms of Kidney Disease
• High blood pressure
• Fatigue, less energy
• Poor concentration and appetite
• Trouble sleeping and night time muscle cramps
• Swollen feet and ankles
• Puffiness around eyes, particularly in the morning
• Dry, itchy skin
• Frequent urination
5/2/2007 12
Creatinine Test
• Typically ordered as part of a general metabolic panel
• Usually tested for in urine and serum, but correlations now exist between serum creatinine and salivary levels of creatinine
• Has certain ranges the are considered healthy– Correlations exist that relate serum creatinine to saliva creatinine and also to glomerular filtration rate (GFR)
– The GFR is a good indicator of kidney disease progression
• The physician is looking for the creatinine clearance
5/2/2007 13
Creatinine and GFR
• Creatinine is a break down product of creatinine in muscle
• The kidney removes it from the blood
• Presence may indicate kidney failure or dysfunction
• Correlations exist relating it to Glomerular Filtration Rate – GFR mL/min/1.73m2
– Creatinine mg/dL
crSPcr 10=crP
MassAgeGFR
⋅⋅−
=815
)140(
Cockcroft-Gault Equation:
Pcr=Plasma Creatinine Concentration (mmol/L)
Scr=Saliva Creatinine Concentration (mmol/L)
Mass in kilograms
5/2/2007 14
Unhealthy GFR
Dialysis and/or transplantEndstage kidney failure514 down
Plan for endstage renal failure
Severely reduced kidney function415-29
Make a diagnosis with additional testing
Moderately reduced kidney function330-59
Find out why kidney function is reduced
Mildly reduced kidney function, with urine abnormalities, indicates kidney
disease
260-89
Observe, control blood pressure
Normal kidney function190+
TreatmentDescriptionStageGFR
crSP cr 10=crS
MassAgeGFR
⋅⋅−
=8150
)140(
5/2/2007 15
Description of Assay
• The test utilizes the Jafféreaction, which requires certain reagents– NaOH to provide alkalinity
– Picric Acid to react with the creatinine
• The Picric Acid produces a color change upon reaction– The color change can be tracked with spectrophotometry
• Serum tests require more reagents to reduce interference
5/2/2007 16
Saliva vs. Blood Collection
• Blood collection– Requires invasive, expensive needle
– Requires disinfection
• Saliva collection– Only requires spitting into a vial
– Patient needs to rest, not eating, for 5 minutes prior to collection
– Patient must chew inert paraffin gum for 1 minute to stimulate saliva flow
5/2/2007 17
FDA Approval
• FDA approval is an important part of medical device development
• According to FDA regulations, a salivary creatinine test is considered a medical device
• Medical devices are regulated by the FDA’s Center for Devices and Radiological Health
• First step in the approval process is to classify the device
• There are three classifications, requiring different degrees of approval processes
• Creatinine tests fall into Category II– Does not require Pre-Market Approval
– Requires Pre-Market Notification
5/2/2007 18
Pre-Market Notification
• All medical devices being brought to market must submit a Pre-Market Notification
• The Pre-Market Notification must establish Substantial Equivalence:
– has the same intended use as the predicate; andandandand
– has different technological characteristics and the information submitted to FDA;
• does not raise new questions of safety and effectiveness; andandandand
• demonstrates that the device is at least as safe and effective as the legally marketed device.
5/2/2007 19
Other Requirements
• Good Manufacturing Practices/Quality System Regulation
• Provides guidance for:– Designing processes and products
– Process control
– Employee training
– Facilities
– Labeling
– Distributing
5/2/2007 20
Consumer Satisfaction
• Relate “consumer” properties to physical properties
• Sensitivity– Ability to detect creatinine
• Likelihood for False Positives– Due to positive interference
• Likelihood for False Negatives– Due to bilirubin interference
• Discomfort– Associated with obtaining sample blood vs. saliva
• Consumers are both patients and medical professionals
5/2/2007 21
Consumer Satisfaction Model
• H is a function of consumer properties related to physical properties
ji
j
jii ywH ,,∑=
H2: consumer satisfaction with existing product
H1: consumer satisfaction with new product
w: weight of property j for product i
y: satisfaction with property j for product i
5/2/2007 22
Weights for Satisfaction Function
• Weights were determined from consumer surveys
• Participants were asked to rate the following factors
– Discomfort
– Sensitivity
– Chance for False Positive Results
– Chance for False Negative Results
0.27False Positive
Rate
0.26False Negative
Rate
0.25Sensitivity
0.22Discomfort
WeightWeightWeightWeightParameterParameterParameterParameter
5/2/2007 23
Discomfort
• Discomfort is a consumer property related to the invasiveness of the test
• Discomfort (D) is a constant dependent on whether or not blood is drawn and is added to the satisfaction function
D = 0.5 if blood is drawn
D = 1 if no blood is drawn
5/2/2007 24
Sensitivity
Consumer satisfaction corresponds to the disease stage that the test can detect
Consumer Satisfaction vs. Disease State at Diagnosis
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Disease State
Satisfaction
Stage 2 Stage 3 Stage 4 Stage 5
5/2/2007 25
Sensitivity
The ability of the test to detect certain disease stages relates to the minimum detectable concentration of
the test
Disease Stage vs. Minimum Detectable Concentration
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
0 10 20 30 40 50 60 70 80
Minimum Detectable Concentration (umol/L)
Disease Stage
5/2/2007 26
Sensitivity
Consumer satisfaction decreases with increasing minimum detectable concentration
Consumer Satisfaction vs. Concentration
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
15 20 25 30 35 40
Detectable Concentration (umol/L)
Satisfaction
5/2/2007 27
Interference
Positive Interference• Cause creatinine test results to be higher than actual
• Interfering compounds
– PyruvicPyruvicPyruvicPyruvic acid, glucose, and acid, glucose, and acid, glucose, and acid, glucose, and alphaalphaalphaalpha----ketoglutaricketoglutaricketoglutaricketoglutaric acidacidacidacid
• Less significant
Negative Interference• Cause creatinine test results to be lower than actual
• Interfering compound
– BilirubinBilirubinBilirubinBilirubin
• Moderately significant
� Certain compounds are known to interfere with the Jaffé
reaction and create misleading results
5/2/2007 28
Distribution of Creatinine in Patients
Percent of Patients with Specific Salivary Creatinine Concentrations
0
0.5
1
1.5
2
2.5
3
12
15
18
21
24
27
30
33
36
39
42
45
48
51
54
57
60
63
66
69
72
75
78
81
84
87
90
93
96
99
102
105
Creatinine Concentration (umol/L)
% Patients
5/2/2007 29
Positive Interference
Satisfaction dramatically drops as the percent of false positives increases
Consumer Satisfaction vs. False Positive Rate
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 10 20 30 40 50 60 70 80 90 100
Fa l se P osi t i v e Ra t e ( % Fa l se P osi t i v e )
5/2/2007 30
Positive Interference
The percentage of tests that give a false positive increases with an increasing concentration of
interfering compounds
Percent False Positive vs Concentration of Interferents
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 0.5 1 1.5 2
Concentration of Interferents
% False Positive
5/2/2007 31
Positive Interference
Satisfaction vs. Positive Interference
0
0.2
0.4
0.6
0.8
1
0 0.4 0.8 1.2 1.6 2 2.4 2.8 3.2 3.6
Concentration of Interferents
Satisfaction
Satisfaction decreases with the concentration of interfering compounds
5/2/2007 32
Positive Interference
• Challenge– Compounds such as glucose can cause the test to show slightly higher creatinine levels than are actually present
– Occurrence of false positives is ~1%
• Possible Resolution– Include monitors to measure levels of positively interfering compounds• i.e. a glucose meter
5/2/2007 33
Negative Interference
Satisfaction decreases rapidly with the percent false negative
Consumer Satisfaction vs. Percent False Negative
0
20
40
60
80
100
0 0.2 0.4 0.6 0.8 1
Percent False Negative
Consumer Satisfaction
5/2/2007 34
Negative Interference
A higher bilirubin to creatinine concentration ratio indicates higher interference
Bilirubin Interference
0
20
40
60
80
100
0 1 2 3 4 5 6
Bilirubin Conc/Creatinine Conc
% of Actual Creatinine
Conc
5/2/2007 35
Negative Interference
• Challenge
– Excessive amounts of bilirubin cause tests to show lower creatinine levels than are actually present
• Resolution
– Sodium dodecyl sulfate decreases the effects of bilirubin, thereby reducing the likelihood for false negative results
5/2/2007 36
Negative InterferencePercent False Negative vs. [SDS]
0
1
2
3
4
5
6
7
8
9
10
0 40 80 120 160 200 240 280
[SDS] (mmol/L)
Percent False Negative
The percentage of false negative results decreases with increasing SDS concentration
up to 140 mmol/L
5/2/2007 37
Negative Interference
Consumer Satisfaction vs. [SDS]
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
0 20 40 60 80 100 120 140
[SDS]
Satisfaction
Satisfaction increases with increasing SDS concentration
5/2/2007 38
Pricing and Demand Model
0)( 1
1
2
111111 =
−
−=Φ
−
ρ
ρρ
βα
dp
dpYpdpd
• p1: new product price• d1: new product demand• α: describes consumer knowledge of new product• β: describes consumer preference for new product• Y: total consumer budget• p2: existing product price (serum test - $10/test)• ρ: constant of 0.75
5/2/2007 39
Awareness Function
• Awareness (α) is a function of consumer awareness of the product
• Awareness increases with time to a value of 1, indicating total awareness
0)( 1
1
2
111111 =
−
−=Φ
−
ρ
ρρ
βα
dp
dpYpdpd
Awareness Function
0
0.2
0.4
0.6
0.8
1
0 1 2 3 4 5
Year
Alpha
5/2/2007 40
Advertising and Education
Advertising and education can increase awareness
0)( 1
1
2
111111 =
−
−=Φ
−
ρ
ρρ
βα
dp
dpYpdpd
Awareness Function
0
0.2
0.4
0.6
0.8
1
0 1 2 3 4 5
Year
Alpha
Advertising,Education
5/2/2007 41
Influence of Awareness
As the consumer awareness increases for the first three years, the demand for the product increases
Demand vs. Price for the First Three Years of Project
Without SDS or Glucose Monitor
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
2 4 6 8 10
Price ($/test)
Demand,
(new product demand/total
demand)
Year 1
Year 2
Year 3
Product Design
-Low minimum
detectable
concentration
-With SDS to counter
negatively interfering
bilirubin
5/2/2007 42
Beta Function
1
2
H
H=β
H1: Consumer satisfaction for the new product
H2: Consumer satisfaction for the existing product
-Lower beta values indicate a more appealing product
Beta does not factor in price
0)( 1
1
2
111111 =
−
−=Φ
−
ρ
ρρ
βα
dp
dpYpdpd
5/2/2007 43
Product Design
• Minimum Detectable Concentration
– Low • 16.8 umol/L
• Stages 2, 3, 4, and 5
• Spectrophotometer
• Includes Standards
– High• 40 umol/L
• Stages 4 and 5
• Determined Visually
• Standards Unecessary
• Inclusion of Anti-Interference Components
– Option 1: No Additives
– Option 2: SDS included to counter negative interference
– Option 3: Glucose meter to monitor positive interference
– Option 4: SDS and glucose meter counter interference
5/2/2007 44
Consumer Satisfaction
Satisfaction is higher for options including additives to counteract interference
Consumer Satisfaction vs. Detectable Concentration
for Various Interference Scenarios
0
0.2
0.4
0.6
0.8
1
16.8 26.8 36.8
Detectable Concentration (umol/L)
Consumer
Satisfaction
No Additives
With SDS
With Glucose Meter
With SDS and
Glucose Meter
5/2/2007 45
Consumer Preference
0)( 1
1
2
111111 =
−
−=Φ
−
ρ
ρρ
βα
dp
dpYpdpd
Beta Values Used for Economic Analysis
0.390.46
0.50
0.63
1.08
0
0.2
0.4
0.6
0.8
1
1.2
Beta
With SDS and Glucose
Meter (D.C.=16.8 umol/L)
With SDS (D.C.=16.8
umol/L)
With Glucose Meter
(D.C.=16.8 umol/L)
No Additives (D.C.= 16.8
umol/L)
No Additives (D.C.=40
umol/L)
D.C. = Minimum Detectable Concentration
5/2/2007 46
Price and Demand
The demand is higher for the test with a lower minimum detectable concentration
Demand vs. Price for the First Year
For High and Low Minimum Detectable Concentrations
Product Option 1: Without SDS or Glucose Monitor
0
0.2
0.4
0.6
0.8
1
2 3 4 5 6
Price ($/test)
Demand,
(new product demand/total
demand)
High Min. Detectable Conc.
Beta = 1.08
Low Min. Detectable Conc.
Beta = 0.63
High Min. Detectable Conc.
Spectrophotometer needed,
tests include standards
Low Min. Detectable Conc.
Visual Observations,
standards unnecessary
5/2/2007 47
NPW for Different Sensitivities
The test able to detect low concentrations was more profitable despite added product cost
Net Present Worth Over Three Year Period
For High and Low Minimum Detectable ConcentrationsProduct Option 1: Without SDS or Glucose Monitor
0
2
4
6
8
10
12
2 3 4 5 6
Price ($/test)
Net Present Worth
(millions $)
High Min. Detectable Conc.
Beta = 1.08
Low Min. Detectable Conc.
Beta = 0.63
High Min. Detectable Conc.
Spectrophotometer needed,
tests include standards
Low Min. Detectable Conc.
Visual Observations,
standards unnecessary
5/2/2007 48
Price and Demand
Demand versus price curves were created using the pricing and demand model for the four interference
scenarios
Demand for Various Product Optionsat Low Minimum Detectable Concentration
0
0.2
0.4
0.6
0.8
1
2 3 4 5 6
Price ($/test)
Demand,
(new product demand/total
demand)
No Additives
Beta = 0.63
With SDS
Beta = 0.46
With Glucose
Meter
Beta = 0.50
With SDS and
Glucose Meter
Beta = 0.39
5/2/2007 49
Net Present Worth
The most profitable design is the product with SDS priced at $4/test
Net Present Worth Over Three Year Period
for Various Product Optionsat Low Minimum Detectable Concentration
-15
-10
-5
0
5
10
15
2 4 6 8 10
Price ($/test)
Net Present Worth,
$Millions
Nothing Added
Beta = 0.63
With SDS
Beta = 0.46
With Glucose
Monitor
Beta = 0.50
With SDS and
Glucose Monitor
Beta = 0.39
5/2/2007 50
Conclusions
• The product design including SDS yields the highest NPW
• The product with the lowest beta value was not the most profitable
5/2/2007 51
Acknowledgements
• Dr. Miguel Bagajewicz
– CBME, University of Oklahoma
• Dr. Stephen Kastl
• Noah Abbas
5/2/2007 52
References1. American Association for Clinical Chemistry, “Creatinine”.
http://www.labtestsonline.org/understanding/analytes/creatinine/test.html2. Amarelo, Monica. “Saliva Tests Detect Disease, Cavities, and Drug Use”. American Association for the Advancement of
Science. 30 April 2007. http://www.aaas.org/news/releases/2005/0217saliva.shtml3. Aydin, Suleyman. "A Comparison of Ghrelin, Glucose, Alpha-Amylase and Protein Levels in Saliva From Diabetics."
Journal of Biochemistry and Molecular Biology 40 (2007): 29-35. 4. Bagajewicz, M: “The ‘Best’ Product is Not the Best Product. Integration of Product Design with Multiscale Planning,
Finances and Microeconomics”. 20075. Butler, Anthony R. "Jaffe Reaction Interference." Editorial. Clinical Chemistry 1988: 642-643. 6. Caldwell, M T., P J. Byrne, N Brazil, V Crowley, S E. Attwood, T N. Walsh, and T P. Hennessy. "An Ambulatory Bile Reflux
Monitoring System: an in Vitro Appraisal." Physiological Measurements 15 (1994): 57-65. 7. "Device Advice." U.S. Food and Drug Administration. FDA. 20 Feb. 2007 <http://www.fda.gov/cdrh/devadvice/>. 8. “Facts about kidney disease.” American Kidney Fund. 25 Feb. 2007 www.kidneyfund.org.9. Kaufman, Eliaz and Ira B. Lamster, “The Diagnostic Applications of Saliva-A Review”. Oral Biol Med 2002; 13(2): 197-
21210. Levey, A; Bosch, J; Lewis, J; Greene, T; Rogers, N; Roth, D: “A More Accurate Method to Estimate Glomerular Filtration
Rate from Serum Creatinine: A New Prediction Equation.” Annals of Internal Medicine 1999; 130(6): 461-47011. Li et al. “The Oral Fluid MEMS/NEMS Chip (OFMNC): Diagnostic & Transitional Applications.” Advances in Dental Research.18:3-5,
June 2005.12. http://adr.iadrjournals.org/cgi/content/full/18/1/3
13. Lloyd, J; Broughton, A; Selby, C: “Salivary Creatinine Assays as a Potential Screen for Renal Disease.” Annals in Clinical Biochemistry 1996; 33: 428-431
14. Lolekha, P H., and N Sritong. "Comparison of Techniques for Minimizing Interference of Bilirubin on Serum Creatinine Determined by the Kinetic Jaffe Reaction." Journal of Clinical Laboratory Analysis 8 (1994): 391-399.
15. National Kidney Foundation, “Glomerular Filtration Rate (GFR)” http://www.kidney.org/kidneydisease/ckd/knowGFR.cfm16. Peters, M. S., Klaus D. Timmerhaus, and Ronald E. West. (2003). Plant Design and Economics for Chemical Engineers.
New York, NY: McGraw-Hill. 17. Sanders, B, R Slotcavage, D Scheerbaum, C Kochansky, and T Strein. "Increasing the Efficiency of in-Capillary
Electrophoretically Mediated Microanalysis Reactions Via Rapid Polarity Switching." Analytical Chemistry 77 (2005): 2332-2337.
18. Tenovuo, Jorma O., ed. Human Saliva: Clinical Chemistry and Microbiology. Vol. 1. Boca Raton: CRC P Inc., 1989. 19. United States of America. Centers for Disease Control and Prevention. U.S. Department of Health and Human Services.
Summary Health Statistics for US Adults: National Health Interview Survey, 2005. 2005. 20. Weber, J A., and A P. Van Zanten. "Interference in Current Methods for Measurements of Creatinine." Clinical Chemistry
37 (1991): 695-700. 21. Young, Stacy. ScienCentral, Inc., “Saliva Diagnostics”. 30 April 2007.
http://www.sciencentral.com/articles/view.php3?type=article&article_id=218392664