microRNA Biomarkers as Potential Diagnostic Markers for Cancer - Genomic Biomarkers for Pharmaceutical Development - page 105

important miRNAs in the serum of ALL patients, along with miR-24 and miR-21 [116,117] .
Tanaka et al. showed that differential expression between miR-92a and miR-638 in plasma
could distinguish ALL patients from healthy control subjects, suggesting that the miRNA
pair could have clinical applications in ALL [131] .
There are a limited number of studies of miRNAs in AML. Costinean et al. showed that
miR-155 was up-regulated in AML [132] . Mi et al. compared miRNA expression profiles in
ALL and AML and found several differentially expressed miRNAs between the two, with a
number of miRNAs found to be expressed at a much higher level in ALL than AML includ-
ing miR-128a, miR-128b, miR-151*, miR-5, miR-130b, and miR-210 [133] . These findings indi-
cate that not only can miRNAs distinguish cancer from normal samples, but that they are
also expressed in a disease-specific manner. Schotte et al. corroborated Mi's finding show-
ing that miRNA expression profiles in leukemia were expressed in a subtype-specific manner
[134] . Because leukemia is a blood disorder, it is reasonable to envision circulating miRNAs
as biomarkers for the disease. In contrast, less is known about circulating miRNAs in patients
with solid tumors and their potential as diagnostic biomarkers. However, miRNA research
in leukemia has paved the way for the study of circulating miRNA in solid tumors, and the
findings of some of these studies are discussed below.
5.3.2.2 Pancreatic Cancer
Blood biomarkers may provide a minimally invasive and cost-effective way of detect-
ing pancreatic cancer; however, as discussed in the previous section, there is currently
no marker with sufficient diagnostic sensitivity and specificity to identify pancreatic can-
cer patients early. Several research groups assessed the feasibility of using plasma for the
detection of miRNAs that were previously reported to be aberrantly expressed in PDAC,
including miR-21, miR-210, miR-155, and miR-196. Wang et al. reported that the combina-
tion of these four miRNAs detected in plasma could discriminate PDAC from healthy indi-
viduals with a sensitivity of 64% and specificity of 89% [135] . Another group evaluated
the supplementary role of miRNAs in plasma to serum measurements of CA 19-9 in early
identification of patients with PDAC [136] . Out of the seven interrogated miRNAs (miR-
16, miR-21, miR-155, miR-181a, miR-181b, miR-196a, and miR-210), miR-16 and miR-196a
independently distinguished between PDAC and chronic pancreatitis, and in combination
with CA 19-9 discriminated patients with PDAC from a combination of normal controls and
chronic pancreatitis with a sensitivity of 92.0% and specificity of 95.6%, and from chronic
pancreatitis only with sensitivity of 88.4% and specificity of 96.3%.
Analysis of miRNA expression levels in serum collected from patients diagnosed with
PDAC led to development of another miRNA signature consisting of miR-20a, miR-21, miR-
24, miR-25, miR-99a, miR-185, and miR-191 [137] . In the study validation phase, the diagnos-
tic sensitivity and specificity of this 7-miRNA classifier was 94% and 93%, respectively. With
the exception of miR-21, there were no other miRNAs that overlapped with those identified
in human plasma, emphasizing the need for molecular diagnostic tools to be developed on a
specimen that is relevant to the clinical question posed.
Finally, a study exploring the potential value of seven fecal miRNAs (miR-16, miR-21,
miR-155, miR-181a, miR-181b, miR-196a, and miR-210) in screening for pancreatic cancer was
reported [138] . miR-181b and miR-210 discriminated PDAC from normal individuals with
receiver operating characteristic (ROC) curves and area under the curve (AUC) of 0.745 and