Have you ever wondered how low a test can go before it stops detecting an analyte? That’s where Detection Limit Experiments - Limit of Blank (LoB), Limit of Detection (LoD), and Limit of Quantitation (LoQ) - come in.
Detection Limit Experiments are fundamental experiments tested in a Method Verification study. These experiments establish how well a method can detect and measure low concentrations of an analyte - a performance characteristic commonly referred to as sensitivity.
While the specific terminology used for detection limit parameters may vary, the goal remains consistent - to establish the method’s ability to reliably detect and quantify analytes at the lowest measurable levels. In this blog, we’ll take a closer look at why Detection Limit is essential in MV and how they help confirm the reliability of a method.
Limit of Blank
Limit of Blank (LoB) is defined as the highest response expected when testing a sample without the analyte - also known as a sample blank.
Purpose
Even samples with no analyte are still capable of producing small, non-zero signals - not due to the presence of the analyte itself, but due to factors such as instrument background noise, reagent impurities and other minor interferences.
LoB determines the signal level obtained from factors aside from the analyte itself. This establishes the signal level of background noise present. Any signal detected above this threshold indicates the potential presence of the analyte.
LoB Experiment Design
Following the recommendations from Clinical and Laboratory Standards Institute (CLSI) EP17-A2 Section 7.2.1, the minimum experiment design includes:
- One reagent lot
- One instrument system
- Two blank samples
- Two replicates per sample per day
- Three days
- Minimum of 20 total blank replicates across all samples and days.
This experiment design does not produce the minimum required 20 total blank replicates. Some design factors need to be increased depending on the Laboratory Director’s discretion.
LoB Procedure and Data Analysis
- Establish a reference value to be used as the maximum LoB threshold. Common practice is to use a manufacturer LoB claim.
- Test the blank samples using the method to be verified ensuring the minimum experiment design is followed.
- Record the obtained test results.
- Compute for the mean and standard deviation (SD) results from the 20 sample blank replicates.
- Compute for the obtained LoB using the formula:
✅ LoB Verification passes if the calculated LoB falls below the manufacturer’s LoB claim.
Value of 1.645:
1.645 used in the calculation of LoB is a statistical constant derived from the normal distribution representing z-score of the 95th percentile in a one-sided distribution.
In simple terms, this means that there is only a 5% probability that a blank sample will produce a signal higher than the calculated LoB.
Limit of Detection
Limit of Detection (LoD) is the lowest concentration of an analyte that a method can reliably detect without quantifying.
Purpose
LoD is tested to ensure that a method can reliably detect the analyte of interest at very low concentrations. In a clinical setting, LoD serves as a decision threshold. Below this level, the presence of the analyte cannot be confidently confirmed. While above this level, values indicate reliable detection although quantification may still remain uncertain.
LoD helps minimize the risk of false-positive results especially in highly sensitive tests where even trace concentrations interpret as significant clinical implications.
LoD Experiment Design
Following Clinical and Laboratory Standards Institute (CLSI) EP17-A2 Section 7.3.1., the minimal experimental design are as follows:
- One reagent lot
- One instrument system
- Two samples whose levels span the LoD claim concentration
- Two replicates per sample per day
- Three days
- 20 total low level replicates across all samples and days.
This experiment design does not produce the minimum required 20 total blank replicates. Some design factors need to be increased depending on the Laboratory Director’s discretion.
LoD Procedure and Data Analysis
- Obtain the minimum detectable concentration of the analyte. Common practice is to use the manufacturer’s LoB claims.
- Conduct the experiment using two level samples and record test results obtained.
- Compare each test result against the minimum detectable concentration. Calculate the percentage of samples whose results fall above this value.
✅ LoD Verification is considered as passed if 95% of the samples fall above the minimum detectable concentration.
Limit of Quantitation
Limit of Quantitation (LoQ) is the lowest concentration of an analyte that can be reliably quantitated with accuracy and precision.
Purpose
By establishing the LoQ, laboratories are able to set the lower limit of calibration curves or reportable ranges. Proper LoQ verification allows labs to define where “reportable quantitation” begins. This is especially crucial for tests whose low concentrations matter in clinical decision making.
LoQ Experiment Design
Following the recommendations of CLSI EP17-A2 Section 7.4.1, the minimal experiment design is:
- One reagent lot
- One instrument system
- Two samples whose levels span the LoQ claim concentration
- Two replicates per sample per day
- Three days
- Total of 20 low level replicates across all samples and days
- This experiment design does not produce the minimum required 20 total blank replicates. Some design factors need to be increased depending the Laboratory Director’s discretion.
LoQ Procedure and Data Analysis
- Establish the experimental design for basis of quality requirement
- Determine the LoQ Goal - usually a manufacturer’s claim
- Test samples in replicates over multiple days until 20 minimum measurements are obtained. Record test results obtained.
- Calculate the total error (TE) between the obtained test results and the established reference value using the formula:
where:
σ = overall experiment standard deviation
- Count the percentage of samples that meet the acceptance criteria.
✅ LoQ Verification is considered as passed if 95% of the samples fall above the minimum detectable concentration.
Detection Limit Verification for Qualitative Methods
Verification of sensitivity performance claims is equally as important for qualitative measurements. Since qualitative results don't provide a continuous signal, this process requires a statistical approach to determine the lowest concentration at which an analyte can be reliably detected.
- Obtain the manufacturer LoD claim to be verified.
- Prepare the following samples to be tested:
Sample | Sample Concentration | Replicates Tested | Expected Result |
C0 | Negative sample/sample blank | 20 replicates | 100% negative |
C95 | LoD Concentration | 20 replicates | 95% positive |
- Record the binary outcome for every replicate at every concentration.
- Calculate the percent of samples that tested positive for each set of replicates.
- Verify the LoD tested during the MV process. This is the lowest concentration tested that achieves at least 95% positive results.
- Compare this to the LoD from manufacturer claims.
✅ Qualitative LoD is verified once C95 is equal to or lower than the manufacturer's claim.
⌛ TL;DR: LoB, LoD and LoQ - Detection Limits
- Limit of Blank (LoB) - defines the highest signal from a blank sample to define values obtained from background noise
- Limit of Detection (LoD) - defines the lowest analyte concentration that can be reliably distinguised from the LoB - meaning at this value, the analyte is reliably detected; but not reliably quantitated.
- Limit of Quantitation (LoQ) - defines the lowest concentration that can be quantified with acceptable accuracy and precision.
LoB, LoD, and LoQ: How Do They Interconnect?
Think of these experiments as an interconnected staircase to define a method’s sensitivity:
- LoB: defined to determine how much noise the blank produces.
- LoD: established above the LoB balancing the acceptable false positive and false negative risks.
- LoQ: LoD is tested to confirm at which point the signal can be quantified accurately and precisely.
LOB vs LOD vs LOQ: Which One To Use?
In the MV process, all three detection limit experiments are important to establish the sensitivity of a method. However, the concept of when each experiment should be verified can differ between qualitative, semi-quantitative, and quantitative methods.
The decision of when to perform the detection limit experiments heavily relies on guidelines provided by accreditation bodies and manufacturers, but depends entirely on the laboratory director’s discretion.
The table below summarizes when LoB, LoD and LoQ are useful in an MV study:
Experiment | LoB | LoD | LoQ |
Qualitative Tests | ⚠️ Not routinely but may be functionally verified to confirm that negative samples truly produce. negative results
If your qualitative method uses a measurable signal to call positive/negative → perform a LoB (blank) verification (EP17 approach) and check low-positive samples. If it’s purely categorical with no numeric signal → verify negative/blank behaviour with confirmed negatives per APHL/CLSI guidance. | ✅ Verify using low-positive samples near the cutoff — LoD verification confirms that positives at or near this concentration are consistently detected. | ❌ LoQ does not apply because the assay does not report quantitative results |
Semi-Quantitative | ⚠️ Not routinely but may be functionally verified to ensure no false positive are detected in the negative or trace category | ✅ Verified to confirm lowest ranges that can be assigned in a semi-quantitative experiment | ✅ Verified to confirm analytical sensitivity by confirming results near the manufacturer’s claimed LoD perform as expected |
Quantitative Test | ✅ Verified to ensure background signal does not exceed the manufacturer’s claimed LoB | ⚠️ LoQ is preferred over LoD for quantitative methods as this provides similar information but with accuracy and precision | ✅ Verified to confirm the lowest concentration that meets precision and bias criteria |
Cut-off limits for most qualitative methods are set equal to or relatively close to the LoD as this establishes the sensitivity and reliability of qualitative methods. This ensures that the target analyte can be detected at specific cut-off limits, making it essential to qualitative methods but not entirely applicable to quantitative methods.
Feeling ready to test your method’s sensitivity? Head on to Cualia to start setting up your own Detection Limit Experiments!
References
Clinical and Laboratory Standards Institute. (2008). CLSI EP12-A2: User Protocol for Evaluation of Qualitative Test Performance; Approved Guideline-Second Edition. Wayne, PA.
Clinical and Laboratory Standards Institute. (2008). CLSI EP17-A2: Evaluation of Detection Capability for Clinical Laboratory Measurement Procedure; Approved Guideline-Second Edition. Wayne, PA.
Westgard, James. (2020). Basic Method Validation, 4th Edition. Wisconsin, Westgard QC, Inc.
Westgard, James. (2008). The Detection Limit Experiment. Basic Method Validation, 1st Edition. Retrieved from: https://www.westgard.com/lesson29.htm
Lister, A. (2005). Limit of Quantitation. Handbook of Pharmaceutical Analysis by HPLC. Retrieved from: https://www.sciencedirect.com/topics/nursing-and-health-professions/limit-of-quantitation
Taleuzzaman, M. T. (2018). Limit of Blank (LOB), Limit of Detection (LOD), and Limit of Quantification (LOQ). Organic and Medicinal Chemistry, Vol. 7 (5), 2-4. DOI: 10.19080/OMCIJ.2018.07.555722