Myeloid Leukemia

28 Picard, Silvy, and Gabert
However, like all PCR, inaccurate results may occur because of either falsepositives
due to contamination, or false-negatives as a result of (1) poor RNA
quality, (2) failure of the RT and PCR steps, or (3) inappropriate primers.
In acute myeloid leukemia (AML), molecular diagnosis for the optimal management
of patients and for minimal residual disease (MRD) monitoring is of extreme
importance. In the mid-1990s, competitive RT-PCR was successfully applied
to quantify the level of fusion gene transcripts in CML and AML with either t(8;21)
or inv(16). Based on the results of this method, the risk of relapse among patients
with AML1-ETO and CBFB-MYH11 transcripts detected in clinical remission is
correlated with the relative level of residual disease, and the kinetic of achievement
of molecular remission is an independent prognostic factor (1,2). However, competitive
RT-PCR is end-point PCR, requires many controls, lacks reproducibility,
and necessitates labor-intensive and time-consuming practices, which prohibit both
standardization and large-scale multicenter analysis.
Real-time quantitative PCR (RQ-PCR) is by far the most sensitive assay in
the context of MRD detection. It can detect a single leukemia cell in a background
of 10 5 –10 6 normal cells. Therefore, it is quantitative to seven orders of
magnitude and is up to five orders of magnitude more sensitive than other conventional
methods. The inter-assay and intra-assay sensitivity of RQ-PCR is
controlled by quantification of housekeeping genes. Some laboratories have
demonstrated the reliability of this technology and its potential clinical value
for MRD studies using fusion gene (FG) transcripts.
The detection of fusion transcripts (PML-RARA, AML1–ETO, and CBFB-
MYH11) by RQ-PCR both at diagnosis and for MRD analysis plays an increasing
role in the management of AML patients and is prospectively being
incorporated into many clinical trials. However, the real predictive clinical
value of MRD detection remains to be confirmed.
After an overview of the principles of RQ-PCR encompassing the chemistry,
the instruments, and the primer and probe design, we will focus on more
practical considerations “at the bench” and finish this overview with comments
on RQ-PCR results (their expression and their interpretation).
Fig. 1. Example of amplification on ABI 7700 instrument of a plasmid standard
curve (100,000 to 10 copies) using TaqMan chemistry and following the Europe
Against Cancer protocol. (A) ∆Rn vs cycle in a linear scale—the threshold (black line)
was set in the exponential phase of polymerase chain reaction (PCR); (B) ∆Rn vs
cycle in a logarithmic scale—the threshold (black line) was set in the exponential
phase of the PCR reaction; (C) standard curve obtained by plotting Ct of different
dilutions vs log of initial concentration. The slope is –3.341 with a correlation coefficient
of 0.996, and the y-intercept is 38.833, indicating that the experiment is valid and
allowing further quantification and analysis of this experiment.