Ct value in real-time PCR: what it says and what it does not say about a sample

The number that appears next to each curve

Anyone looking for the first time at the report of a real-time PCR finds, next to each sample, a number with decimals: 24.3; 31.8; 37.5. It is the Ct value, and it is probably the most cited and most misunderstood piece of data in the technique. It is often read as if it were a measure of "how much pathogen there is". In part it is, but with conditions worth knowing before drawing conclusions about a batch or a surface.

This article explains how that value is obtained, how it relates to the amount of starting genetic material, why late values call for caution and why, in food microbiology, a Ct does not translate directly into CFU.

Ct, Cp, Cq: three names for the same thing

Ct comes from threshold cycle. Other instrument manufacturers used names such as crossing point (Cp) or take-off point. The MIQE guidelines, published in 2009 by Bustin and colleagues in Clinical Chemistry, pointed out that all those terms refer to the same instrument value and that they arose from commercial differentiation, not scientific clarity. That is why they proposed a single name: Cq, quantification cycle. In industry people still say "Ct" and in this article we use both as synonyms.

How the value is obtained

In each PCR cycle the instrument measures the fluorescence of the tube. At the beginning the signal from the amplified product is too low to be distinguished from background noise; that zone is called the baseline. Then comes a phase in which the amount of product grows exponentially and the fluorescence rises clearly, and finally a plateau in which the reagents are exhausted.

The software sets a fluorescence threshold above the baseline. The Ct is the cycle, with decimals because it is interpolated, at which a sample's curve crosses that threshold. Two technical decisions influence that number:

• The baseline estimate. Ruijter and colleagues (2009) showed that errors in estimating the baseline are reflected directly in the observed efficiency of the reaction and propagate to subsequent calculations.
• The position of the threshold. The same work found that the best reproducibility of the Ct is obtained when the threshold is placed in the upper part of the log-linear phase, where noise weighs less.

The practical consequence is that a Ct is not a property of the sample in isolation: it also depends on the instrument, the chemistry and how the analysis was configured. That is why commercial kits fix those parameters, or an interpretation software applies them, and the laboratory should not modify them on its own.

An inverse, logarithmic relationship

The more target genetic material there is at the start, the sooner the threshold is reached and the lower the Ct. The relationship is not linear but logarithmic. If the reaction doubles the product in each cycle (100% efficiency), a sample with ten times more target crosses the threshold about 3.3 cycles earlier. The guide by Kralik and Ricchi (2017) on real-time PCR in microbial diagnostics puts it this way: between ten-fold serial dilutions, the theoretical Cq difference is always 3.322 cycles.

That has a misleading effect. The same authors warn that, although the numerical difference between a Cq of 20 and one of 35 seems small, in real quantity it represents almost five orders of magnitude. A Ct of 35 is not "a little less" than one of 20: it is about a hundred thousand times less starting material.

Reaction efficiency

The 3.3-cycle rule only holds if efficiency is close to 100%. MIQE establishes that efficiency must be calculated from a calibration curve: serial dilutions of a known amount of target are analyzed, the Cq is plotted against the logarithm of the concentration and the slope of the line is used. The formula is efficiency = 10^(-1/slope) - 1; a slope of -3.32 corresponds to 100%. Kralik and Ricchi indicate that, in practice, this parameter usually falls between 90% and 105%.

The calibration curve also defines the linear dynamic range: the interval of concentrations in which the relationship between Cq and quantity holds. MIQE asks that it cover at least three orders of magnitude and ideally five or six. Outside that interval, the Cq still appears on screen, but it no longer allows quantities to be estimated with confidence.

The MIQE 2.0 revision, published in 2025, reinforced this idea: Cq values must be converted into efficiency-corrected target quantities and reported with prediction intervals, together with the limits of detection and the dynamic range of each target. The Cq is an intermediate instrument datum, not the final result.

Late Ct values: the zone of uncertainty

The difficult decisions appear at the end of the run, with curves that cross the threshold at cycles 36, 38 or 40. Three ideas help to read them:

• There is a physical floor. MIQE reminds us that, because of the Poisson distribution, the most sensitive theoretically possible limit of detection is about 3 copies per reaction. With so few molecules, one replicate amplifying and another not is an effect of pipetting chance, not necessarily an error.
• The limit of detection is defined with replicates. MIQE describes it as the lowest concentration at which 95% of positive samples are detected. An isolated Ct near that limit is, by definition, a result that would not always be repeated. We go deeper into this concept in limit of detection, LOD50 and uncertainty.
• Fixed cut-offs are conventions. MIQE considers Cq values above 40 suspect, because of the low efficiency they imply, but it also warns that arbitrary cut-offs can be too low (discarding valid results) or too high (adding false positives).

That is why each validated kit comes with its own interpretation criterion: up to which cycle a signal is considered positive, what shape the curve must have and what to do with a doubtful result. That criterion is part of the method validation and should not be replaced by a cut-off chosen in the laboratory. A late curve can also come from contamination, which the no-template control makes it possible to monitor, as we explain in PCR controls.

Why a Ct does not equal CFU

In pathogen detection it is tempting to use the Ct as an indicator of the level of contamination in the food. There are at least three reasons not to do so.

Enrichment changes the starting quantity

In most methods for Salmonella or Listeria monocytogenes, PCR is performed on a broth that has already gone through a many-hour enrichment. The Ct reflects how much target there was in that broth at the end of incubation, which depends on how many cells there were at the start, their physiological state and competition with the accompanying microbiota. Two samples with the same initial contamination can give very different Ct values after enrichment. That is why these methods are qualitative: they answer "detected" or "not detected" in the portion analyzed.

DNA does not distinguish live cells from dead ones

Kralik and Ricchi stress that PCR on DNA cannot differentiate live from dead cells, and that a processed food can retain amplifiable DNA even though all the bacteria have been inactivated. A low Ct in a heat-treated product does not prove that viable pathogens are present. It is the problem we address in RNA vs. DNA.

The Ct is an instrument reading

The same authors remind us that Cq values are instrumental readings that must be recalculated into concrete units. A Ct of 30 on one instrument, with one chemistry and one threshold, is not comparable with a Ct of 30 obtained in another assay. Comparing values between different kits, or between laboratories, without a common calibration curve has no quantitative meaning.

When the Ct is useful for quantification

Real-time PCR can be quantitative when the assay is designed for that: the sample is analyzed without enrichment, a calibration curve or a verified calibrator is included and work is done within the linear dynamic range. The standard ISO 22174, in its 2024 edition, establishes the general requirements for nucleic acid amplification in the detection and quantification of microorganisms in the food chain. Even so, what is quantified are copies of the target or genome equivalents, not colony-forming units: one CFU may correspond to one or several cells, and the number of target copies per cell varies depending on the gene chosen.

Good practices for reading a Ct

• Look at the curve, not just the number. A curve without a sigmoid shape that crosses the threshold through baseline drift can generate a Ct that does not correspond to a real amplification. In intercalating dye chemistry, MIQE warns that primer dimers can produce false positives, and the melting curve helps rule them out.
• Read the Ct together with the controls. A sample Ct is only interpretable if the positive control, the no-template control and the internal amplification control behaved as expected.
• Do not average or compare Ct values as if they were concentrations. MIQE considers it inappropriate to report coefficients of variation on Cq values, because they are always smaller (and therefore misleading) than those calculated on copy number.
• Respect the kit's criterion. The thresholds, baseline parameters and cut-off cycle are part of the validated method.
• Automate the application of rules. In laboratories with many runs, an interpretation software such as TxA always applies the same validity and cut-off criteria, and reduces variation between analysts.