Blood sampling and detection limits
In one sentence
A blood assay can detect only qualifying material that reaches the sample and survives its workflow; an analytical detection limit does not remove those sampling limits.
The intuition
A very sensitive microphone cannot record a voice that never reaches it. Blood testing has a similar chain: tumor material must enter the circulation, be included in the tube, survive processing and satisfy the detector's rule.
The analogy has limits. DNA fragments are finite molecules, and a call often combines multiple features. Sampling uncertainty and technical error are separate, interacting problems.
How it works
For circulating tumor DNA (ctDNA), at least four gates matter:
Biological availability: tumors differ in how much material reaches peripheral blood. Burden, site, treatment and timing can affect availability. A highly sensitive assay cannot infer a complete body inventory from plasma.
Finite sampling: the analyzed plasma aliquot contains only some circulating molecules. Very rare features can be absent by chance. For a deliberately simplified model, suppose an evaluated aliquot has an average of one target molecule, independently randomly distributed. A Poisson model gives a probability of no target molecules of exp(−1), about 37%. This is a mathematical illustration, not a clinical false-negative rate. Real assays track different features, face losses and use more complex calling rules.
Recovery and representation: extraction, library construction and sequencing retain or observe some starting molecules. Amplifying one molecule into many reads does not create independent original tumor molecules. Multiple tracking targets can add opportunities to detect signal, but their number alone does not establish complete recovery or clinical sensitivity.
Analytical discrimination: errors and background must be separated from true signal. Limit of detection (LoD) describes detection probability under specified conditions. A nominal LoD from high-input validation material may not apply to a low-input clinical sample. A sample that fails quality criteria should not be silently treated as negative.
Why it matters in cancer
Analytical detection capability differs from clinical sensitivity: the proportion of people with a defined clinical condition who test positive under a specified schedule and follow-up. Site mix, missed draws and target availability affect that denominator. Negative results require those limits; a positive result still requires origin and clinical interpretation.
How it is measured
| Assay-card item | Scope to keep attached |
|---|---|
| Measures / how | Qualifying molecular signals in the analyzed aliquot |
| Input and cost | Draw volume, plasma analyzed, recovered DNA and independent molecular coverage; finite material and draw burden |
| Output / units | Detection call and any validated concentration or fraction |
| Thresholds | LoD with detection probability, input and feature/calling conditions |
| Failure modes | Low shedding, chance absence, handling losses, poor recovery, background or target mismatch |
| Cannot tell | Whether every tumor site was sampled or a negative person is cured |
| Validation | Clinical-material confirmation, input-range testing, representative targets and outcome-based evaluation |
Common confusions
- Read depth is not the same as independent molecule count.
- Larger input can help sampling but does not guarantee detectable shedding or remove noise.
- Repeated draws sample different times; their errors are not automatically independent.
Try it
A fictional laboratory has a validated 95%-detection claim at a specified concentration with 30 units of input. A patient sample supplies only 5 units. Can the laboratory attach the same 95% claim without supporting data?
Answer: No. The lower input changes molecular opportunities and may change detection performance. The sample needs an appropriate validated claim, qualification or failure classification.
Explain it back
Why can a detector with excellent laboratory performance still miss a clinically present tumor?
Takeaway
Follow material from the body to the call; every gate has a scope.
Related concepts
Sources and scope
Source check: October 10, 2026. Poisson example and input quantities are fictional; expert and learner review remain pending.
- FDA 2024 circulating tumor DNA (ctDNA) drug-development guidance — shedding, sampling, input and complete-workflow validation.
- Newman et al. 2016 — molecular recovery, input and error-suppression limits in a defined sequencing method.
- CLSI harmonized terminology — specified detection probability and distinction from clinical sensitivity.