0
We're unable to sign you in at this time. Please try again in a few minutes.
Retry
We were able to sign you in, but your subscription(s) could not be found. Please try again in a few minutes.
Retry
There may be a problem with your account. Please contact the AMA Service Center to resolve this issue.
Contact the AMA Service Center:
Telephone: 1 (800) 262-2350 or 1 (312) 670-7827  *   Email: subscriptions@jamanetwork.com
Error Message ......
Preliminary Communication |

MicroRNA Biomarkers in Whole Blood for Detection of Pancreatic Cancer FREE

Nicolai A. Schultz, MD, PhD1,2,3,4; Christian Dehlendorff, PhD5; Benny V. Jensen, MD1; Jon K. Bjerregaard, MD, PhD6; Kaspar R. Nielsen, MD, PhD7; Stig E. Bojesen, MD, PhD, DMSc8; Dan Calatayud, MD4; Svend E. Nielsen, MD9; Mette Yilmaz, MD10; Niels Henrik Holländer, MD11; Klaus K. Andersen, PhD5; Julia S. Johansen, MD, DMSc1,2
[+] Author Affiliations
1Department of Oncology, Herlev Hospital, Copenhagen University Hospital, Copenhagen, Denmark
2Department of Medicine, Herlev Hospital, Copenhagen University Hospital, Copenhagen, Denmark
3Department of Gastroenterology, Herlev Hospital, Copenhagen University Hospital, Copenhagen, Denmark
4Department of Surgical Gastroenterology and Transplantation, Rigshospitalet, Copenhagen University Hospital, Copenhagen, Denmark
5Danish Cancer Society Research Center, Danish Cancer Society, Copenhagen, Denmark
6Department of Oncology, Odense University Hospital, Odense, Denmark
7Department of Clinical Immunology, Aalborg University Hospital, Aalborg, Denmark
8Department Clinical Biochemistry, Herlev Hospital, Copenhagen University Hospital, Copenhagen, Denmark
9Department of Oncology, Hillerød Hospital, Hillerød, Denmark
10Department of Oncology, Aalborg University Hospital, Aalborg, Denmark
11Department of Oncology, Næstved Hospital, Næstved, Denmark
JAMA. 2014;311(4):392-404. doi:10.1001/jama.2013.284664.
Text Size: A A A
Published online

Importance  Biomarkers for the early diagnosis of patients with pancreatic cancer are needed to improve prognosis.

Objectives  To describe differences in microRNA expression in whole blood between patients with pancreatic cancer, chronic pancreatitis, and healthy participants and to identify panels of microRNAs for use in diagnosis of pancreatic cancer compared with the cancer antigen 19-9 (CA19-9).

Design, Setting, and Participants  A case-control study that included 409 patients with pancreatic cancer and 25 with chronic pancreatitis who had been included prospectively in the Danish BIOPAC (Biomarkers in Patients with Pancreatic Cancer) study (July 2008-October 2012) plus 312 blood donors as healthy participants. The microRNA expressions in pretreatment whole blood RNA samples were collected and analyzed in 3 randomly determined subcohorts: discovery cohort (143 patients with pancreatic cancer, 18 patients with chronic pancreatitis, and 69 healthy participants), training cohort (180 patients with pancreatic cancer, 1 patient with chronic pancreatitis, and 199 healthy participants), and validation cohort (86 patients with pancreatic cancer, 7 patients with chronic pancreatitis, and 44 healthy participants); 754 microRNAs were screened in the discovery cohort and 38 microRNAs in the training cohort and 13 microRNAs in the validation cohort.

Main Outcomes and Measures  Identification of microRNA panels (classifiers) for diagnosing pancreatic cancer.

Results  The discovery cohort demonstrated that 38 microRNAs in whole blood were significantly dysregulated in patients with pancreatic cancer compared with controls. These microRNAs were tested in the training cohort and 2 diagnostic panels were constructed comprising 4 microRNAs in index I (miR-145, miR-150, miR-223, miR-636) and 10 in index II (miR-26b, miR-34a, miR-122, miR-126*, miR-145, miR-150, miR-223, miR-505, miR-636, miR-885.5p). The test characteristics for the training cohort were index I area under the curve (AUC) of 0.86 (95% CI, 0.82-0.90), sensitivity of 0.85 (95% CI, 0.79-0.90), and specificity of 0.64 (95% CI, 0.57-0.71); index II AUC of 0.93 (95% CI, 0.90-0.96), sensitivity of 0.85 (95% CI, 0.79-0.90), and specificity of 0.85 (95% CI, 0.80-0.85); and CA19-9 AUC of 0.90 (95% CI, 0.87-0.94), sensitivity of 0.86 (95% CI, 0.80-0.90), and specificity of 0.99 (95% CI, 0.96-1.00). Performances were strengthened in the validation cohort by combining panels and CA19-9 (index I AUC of 0.94 [95% CI, 0.90-0.98] and index II AUC of 0.93 [95% CI, 0.89-0.97]). Compared with CA19-9 alone, the AUC for the combination of index I and CA19-9 was significantly higher (P = .01). The performance of the panels in patients with stage IA-IIB pancreatic cancer was index I AUC of 0.80 (95% CI, 0.73-0.87); index I and CA19-9 AUC of 0.83 (95% CI, 0.76-0.90); index II AUC of 0.91 (95% CI, 0.87-0.94); and index II and CA19-9 AUC of 0.91 (95% CI, 0.86-0.95).

Conclusions and Relevance  This study identified 2 diagnostic panels based on microRNA expression in whole blood with the potential to distinguish patients with pancreatic cancer from healthy controls. Further research is necessary to understand whether these have clinical implications for early detection of pancreatic cancer and how much this information adds to serum CA19-9.

Figures in this Article

Pancreatic cancer is the fourth most common cause of cancer death in the Western world.1,2 The prognosis is poor, with 1- and 5-year survival rates of only 20% and 6%.1,3 Systemic chemotherapy administered either after tumor resection surgery4 or in patients with metastatic disease5,6 has been shown to prolong survival; however, surgery is the only curative treatment.3 Approximately 20% of patients with pancreatic cancer can be operated on with curative intent because most have locally advanced or metastatic pancreatic cancer at the time of diagnosis. Early diagnosis of pancreatic cancer is difficult, and no biomarkers in blood can identify patients with pancreatic cancer at an early stage.3,7

MicroRNAs are noncoding 17- to 25-nucleotide-long RNAs that regulate gene expression posttranscriptionally. MicroRNAs play important roles in oncogenesis and tumor metastasis.8,9 More than 2500 human microRNAs sequences are known today,10 and several specific microRNA profiles related to pancreatic cancer tissue are described.1114 A sensitive and specific diagnostic noninvasive blood test for pancreatic cancer would be very valuable because it is difficult to get useful biopsies of tissue from patients suspected of having pancreatic cancer. Small retrospective studies have demonstrated that expression of specific microRNAs in plasma or serum can distinguish patients with pancreatic cancer from healthy participants.1517 Most of these microRNAs are not validated in independent case-control studies. Whole blood–derived microRNA profiles have been suggested as a new tool for early detection of pancreatic cancer and other adenocarcinomas.13,18,19 The advantages of whole blood are higher microRNA content, elimination of methodological problems related to the handling of serum and plasma, and the possibility of measuring both tumor-secreted microRNA and changes in microRNA profiling following the host reaction in patients with cancer.19,20

Serum cancer antigen 19-9 (CA19-9) is elevated in approximately 80% of patients with pancreatic cancer and has been proposed as a useful diagnostic tool for the detection of pancreatic cancer.3,7 It is approved for determination of prognosis and as a guide to treatment and follow-up of patients with pancreatic cancer.21 Therefore, we tested the diagnostic accuracy of serum CA19-9 in combination with microRNA profiles for pancreatic cancer compared with either serum CA19-9 alone or microRNA profiles alone.

The aims of the present study were to describe differences in microRNA expression in whole blood between patients with pancreatic cancer and healthy participants and patients with chronic pancreatitis, and to identify diagnostic panels using a limited number of microRNAs for use in the diagnosis of pancreatic cancer.

The study was conducted according to the Reporting Recommendations for Tumor Marker Prognostic Studies (REMARK) guidelines.22 Patients with pancreatic cancer who were treated in surgical and/or oncological departments at 6 hospitals in Denmark were included in the Danish multicenter BIOPAC (Biomarkers in Patients with Pancreatic Cancer) study. Patients were included from July 1, 2008, to October 18, 2012. Inclusion criteria were age older than 18 years, histologically verified pancreatic cancer (pancreatic ductal adenocarcinoma) in a resection specimen, a computed tomographic (CT) scan showing a solid mass in the pancreas in patients not undergoing surgery, and histology or cytology from this primary tumor or histology from a liver metastasis that confirmed the diagnosis of adenocarcinoma.

The patients included in the BIOPAC study were all consecutive patients who met inclusion criteria and agreed to participate. Blood samples were taken before treatment, during treatment, and at follow-up. All patients provided written informed consent. The study was approved by the regional ethics committee (VEK ref KA-20060113).

Healthy participants were included from the Danish corps of volunteer blood donors, Aalborg University Hospital. Patients with chronic pancreatitis were included in the BIOPAC study because of suspected pancreatic cancer. Diagnoses were made by means of imaging techniques (CT, magnetic resonance, or ultrasound) or on the basis of histological findings (patients with chronic pancreatitis who received surgery).

More details of the design of the study appear in Figure 1. Pretreatment blood samples from patients included in the BIOPAC study until June 30, 2012, were allocated in chronological order to the discovery cohort from centers 1 and 6 and to the training cohort from all 6 participating centers. Patients with pancreatic cancer included after June 30, 2012, and patients with other periampullary cancers were allocated to the validation cohort. We used the discovery cohort for screening (ie, reducing the number of candidate microRNAs for further investigation). Potential microRNAs were measured in the training cohort and used for the derivation of diagnostic indices. In addition, the predictive performance was investigated in the validation cohort.

Place holder to copy figure label and caption
Figure 1.
Study Design

aThere were also 33 other periampullary cancers analyzed as part of this cohort study (15 ampullary, 12 distal common bile duct, and 6 duodenal).bIndicates low RNA yield, low absorbance, and few detectable microRNAs.

Graphic Jump Location

The microRNA was purified from pretreatment whole blood samples collected in PAXgene blood RNA tubes (Qiagen). MicroRNA expression analysis used preconfigured TaqMan Human MicroRNA assay (Applied Biosystem; 754 human microRNAs tested) in the discovery cohort. To increase the number of patients and healthy participants and to increase the number of replicates of each microRNA, the FluidigmBioMark System was used in the training cohort (38 microRNAs tested in duplicates or triplicates) and the validation cohort (13 microRNAs tested in 7 or 8 replicates). This system can perform multiple simultaneous real-time polymerase chain reaction experiments running TaqMan assays. More information appears in the Supplement (eText, see Methods and Statistics and eTable 1).

For the discovery cohort, the raw cycle threshold values (the polymerase chain reaction cycle at which the sample reaches the level of detection or the point at which a reaction reaches a fluorescent intensity above background) for each microRNA were normalized to remove technical bias. To minimize the risk of false-positive results in the evaluation of the 754 microRNAs in the discovery cohort due to multiple testing, we applied an overall test of association by means of the Kolmogorov-Smirnov method followed by a test of individual microRNA at a .001 significance level. The missing values effect was analyzed with imputation (eText, see Statistics and eFigure 1 in the Supplement).

For the training and validation cohorts, the association between microRNA expression and case-control status was estimated by means of bivariable and multivariable logistic regression (including serum CA19-9 cutoff level of >37 KU/L or ≤37 KU/L). In the logistic regression model, we assumed a linear association between the continuous cycle threshold values and disease status on the logit scale.

Based on microRNAs that were significant in the training cohort, we suggested 2 diagnostic indices. Index I consists of 4 selected microRNAs and is designed to be robust to technical variation and contains no model parameters. Index II includes all significant microRNAs from a multivariable model and corresponds to the upper limit in terms of training and is thus potentially overfitted.

Using indices I and II, we evaluated and presented the performance for the discovery, training, and validation cohorts by means of sensitivity, specificity, area under the curve (AUC), and corresponding 95% confidence intervals. For both indices, we considered the performance by defining a cutoff corresponding to fixing the sensitivity to 0.85 to handle the difference in setup between studies (ie, finding the value in which 85% of the cases are correctly classified). We also tested the performance by including serum CA19-9 in the indices.

In all analyses, R version 2.14.0 (R Foundation for Statistical Computing, http://www.R-project.org), 2-sided tests, and a significance level of .05 were used. If not stated otherwise, we considered all patients with pancreatic cancer as a single group regardless of stage. Detailed statistical information appears in the Supplement (eText, see Statistics).

The characteristics of the study participants are presented in Table 1. There were 409 patients with pancreatic cancer included from the Danish BIOPAC study. Blood samples included all pretreatment samples taken before surgery for patients with resectable tumors (n = 44) and before chemotherapy for patients with unresectable tumors (n = 365). The controls included 25 patients with chronic pancreatitis and 312 healthy participants. Due to restrictions in the sampling of healthy participants, there was a significant difference in age (P < .001) between patients with pancreatic cancer and healthy participants and patients with chronic pancreatitis (mean difference, 10.08 years [95% CI, 8.17-11.98 years] in the discovery cohort; 22.22 years [95% CI, 19.82-24.62 years] in the training cohort; and 10.00 years [95% CI, 7.30-12.70 years] in the validation cohort). The purification and microRNA assay quality results appear in the Supplement (eText, see Results).

Table Graphic Jump LocationTable 1.  Demographics of Patients and Healthy Participants in the Discovery, Training, and Validation Cohortsa

In the discovery cohort, multivariable analysis demonstrated that 38 microRNAs had the potential to separate patients with pancreatic cancer from healthy participants and patients with chronic pancreatitis by at least 1 of 5 normalization methods (eTable 2 in Supplement). Fourteen of these microRNAs were found by 2, 3, or 4 of the normalization methods (ie, high expression of miR-34a, miR-122, miR-145, miR-199b-5p, miR-582-3p, miR-769-5p, and miR-885-5p; and low expression of miR-31, miR-31*, miR-93, miR-126*, miR-150, miR-636, and miR-935). The relative expressions of the microRNAs that were found to be significantly differently expressed between patients with pancreatic cancer and healthy participants by at least 3 normalization methods are illustrated as box plots in eFigure 2 in the Supplement.

In the training cohort, 19 of the 36 microRNAs (2 of the 38 measured microRNAs were undetectable) selected from the discovery cohort were validated by the Fluidigm method with a P value of less than .05 (Table 2). The box plots of the relative expression of these microRNAs in patients with pancreatic cancer and healthy participants appear in eFigure 3 in the Supplement. The microRNA comparisons between patients with pancreatic cancer and healthy participants in all 3 cohorts appear in eTable 3 in the Supplement.

Table Graphic Jump LocationTable 2.  Prediction by MicroRNA of Patients With Pancreatic Cancer vs Healthy Participants (Controls) in the Training Cohort and vs Those With Chronic Pancreatitis Plus Controls in the Validation Cohort

Based on the results from the training cohort, 2 diagnostic indices were developed. Index I was calculated as miR-150 + miR-636 − miR-145 − miR-223. Index II was calculated as 6.9275 − (0.2134 × miR-122) − (0.3560 × miR-34a) − (0.8577 × miR-145) + (1.0043 × miR-636) − (0.6725 × miR-223) + (0.7018 × miR-26b) − (0.3233 × miR-885.5p) + (1.1304 × miR-150) − (0.2204 × miR-126*) − (0.1730 × miR-505).

In the training cohort, the AUC was 0.86 (95% CI, 0.82-0.90) for index I, 0.93 (95% CI, 0.90-0.96) for index II, and 0.90 (95% CI, 0.87-0.94) for serum CA19-9. In the discovery cohort, the AUC was 0.91 (95% CI, 0.87-0.94) for index I, 0.94 (95% CI, 0.91-0.97) for index II, and 0.88 (95% CI, 0.83-0.93) for CA19-9 when patients with pancreatic cancer were tested against healthy participants. The box plots of indices I and II using the whole blood samples from all 3 cohorts appear in Figure 2.

Place holder to copy figure label and caption
Figure 2.
Box Plots of Indices I and II Using the Whole Blood Samples From the Discovery, Training, and Validation Cohorts

The horizontal line in the middle of each box indicates the median, whereas the top and bottom borders of the box mark the 75th and 25th percentiles, respectively. The whiskers above and below the box extend to the most extreme point no longer than 1.5 times the interquartile range from the box. The points beyond the whiskers are outliers. Index I includes miR-150, miR-636, miR-145, miR-223. Index II includes miR-26b, miR-34a, miR-122, miR-126*, miR-145, miR-150, miR-223, miR-505, miR-636, miR-885.5p. P < .001 for all comparisons (healthy participants vs patients with cancer for indices I and II for each cohort).

Graphic Jump Location

In the discovery cohort, the AUC was 0.88 (95% CI, 0.85-0.92) for index I, 0.93 (95% CI, 0.89-0.96) for index II, and 0.87 (95% CI, 0.82-0.92) for CA19-9. The AUC values, sensitivity, specificity, and accuracy of the indices when patients with pancreatic cancer were tested against both healthy participants and patients with chronic pancreatitis for all 3 cohorts appear in Table 3. The diagnostic accuracies of the indices were significantly improved by combining either of the indices with serum CA19-9 in the discovery cohort (except AUC for index I combined with CA19-9 for patients with pancreatic cancer vs healthy participants), and in the training and the validation cohorts (Table 3). Figure 3 shows receiver operating characteristic curves for indices I and II alone and in combination with serum CA19-9.

Table Graphic Jump LocationTable 3.  Performance of Indices I and II and Serum Cancer Antigen 19-9 in the Differential Diagnosis of Pancreatic Cancer From Healthy Participants (Controls) and Patients With Chronic Pancreatitis (CP)
Place holder to copy figure label and caption
Figure 3.
Reciever Operating Characteristic Curves for Performance of Indices I and II and in Combination With Serum Cancer Antigen 19-9 in the Discovery, Training, and Validation Cohorts

AUC indicates area under curve; CA19-9, cancer antigen 19-9. See legend of Figure 2 for definitions of indicies I and II.

Graphic Jump Location

Box plots of the relative expression in the training cohort of the 10 microRNAs included in the indices for patients with pancreatic cancer and healthy participants appear in eFigure 4 in the Supplement. The indices for different tumor stages appear in eFigure 5 in the Supplement; no association between indices and age was found (eFigure 6 in Supplement). No association between indices and sex was found for patients with pancreatic cancer, but a significant association with sex was found for healthy participants in the training cohort with index I (mean score of 0.31 [95% CI, 0.18-0.43]; P < .001) and with index II (mean score of 0.68 [95% CI, 0.14-1.22], P = .01) (eFigure 6 in Supplement).

Both indices correlated with counts of white blood cells (index I, ρ = 0.63; index II, ρ = 0.48), granulocytes (index I, ρ = 0.66; index II, ρ = 0.53), and platelets (index I ρ = 0.43; index II ρ = 0.28) (all comparisons yielded P < .001).

In the validation cohort, 13 microRNAs (of which 10 met the .05 significance level in both the discovery and training cohorts) were measured and 10 met the significance criteria of P < .05 in this final validation (Table 2). The sensitivity, specificity, accuracy, and AUC of indices I and II tested in the validation cohort appear in Table 3. Using the indices to compare patients with pancreatic cancer with healthy participants, the AUC was 0.83 (95% CI, 0.76-0.90) for index I, 0.82 (95% CI, 0.75-0.89) for index II, and 0.91 (95% CI, 0.86-0.96) for CA19-9. If the indices were combined with serum CA19-9, the AUC increased (index I and CA19-9: 0.94 [95% CI, 0.90-0.98]; index II and CA-19-9: 0.93 [95% CI, 0.89-0.97]; Figure 3 and Table 3).

The sensitivity of indices was fixed at 0.85, and thus 85% (73/86) of patients with pancreatic cancer had a correct diagnosis using index I or II with or without serum CA19-9. In the validation cohort, 29% of patients with chronic pancreatitis had a correct diagnosis using index I or II with or without CA19-9. Of the healthy participants, 48% and 54% had a correct diagnosis using index I or II, increasing to 98% and 89% in combination with CA19-9. Index I in combination with CA19-9 improved AUC significantly compared with CA19-9 alone in the validation cohort of patients with pancreatic cancer vs healthy participants and patients with chronic pancreatitis (P = .01). The combination of index II and CA19-9 also improved AUC compared with CA19-9 alone, although this improvement was not significant (P = .15). Further results appear in eTable 4 in the Supplement.

Results for discrimination between patients with other periampullary cancers and healthy participants appear in the Supplement (eText, see Results and eFigure 7).

To assess the performance of the indices in a sufficient number of cases of pancreatic cancer vs healthy participants, we combined patients with low-stage pancreatic cancer (stages I and II) from the 3 study cohorts. The performances of indices I and II were compared with serum CA19-9 alone or combined with CA19-9 using the cutoff values defined in Table 3.

For CA19-9, the AUC was 0.81 (95% CI, 0.73-0.88), the sensitivity was 0.74 (95% CI, 0.62-0.84), the specificity was 0.99 (95% CI, 0.97-1.00), and the accuracy was 0.94 (95% CI, 0.92-0.96).

For index I, the AUC was 0.80 (95% CI, 0.73-0.87), the sensitivity was 0.77 (95% CI, 0.66-0.86), the specificity was 0.66 (95% CI, 0.61-0.71), and the accuracy was 0.68 (95% CI, 0.63-0.73). For CA19-9 and index I, the AUC was 0.83 (95% CI, 0.76-0.90), the sensitivity was 0.74 (95% CI, 0.62-0.84), the specificity was 0.96 (95% CI, 0.93-0.98), and the accuracy was 0.92 (95% CI, 0.89-0.95).

For index II, the AUC was 0.91 (95% CI, 0.87-0.94), the sensitivity was 0.80 (95% CI, 0.69-0.89), the specificity was 0.82 (95% CI, 0.77-0.86), and the accuracy was 0.82 (95% CI, 0.77-0.85). For CA19-9 and index II, the AUC was 0.91 (95% CI, 0.86-0.95), the sensitivity was 0.73 (95% CI, 0.61-0.83), the specificity was 0.97 (95% CI, 0.94-0.99), and the accuracy was 0.93 (95% CI, 0.90-0.95). eTable 5 in the Supplement provides results for patients with stage I and stage II pancreatic cancer and also as a combined group.

Sensitive and specific biomarkers to identify patients with pancreatic cancer at an early stage are needed. This study describes 2 novel panels of microRNA for diagnosing pancreatic cancer using the combination of 4 or 10 microRNAs in whole blood.

When this microRNA biomarker study was designed in 2008, the aim was to identify microRNAs in whole blood that could identify pancreatic cancer (local, locally advanced, and metastatic disease) in individuals thought to be healthy. We first wanted to test the extremes (ie, patients with known pancreatic cancer vs healthy blood donors) because if panels of microRNA could not differentiate patients with pancreatic cancer and high tumor burden from these healthy individuals, it would be difficult to develop a diagnostic microRNA test for patients with small tumors and low-stage cancer. In the future, we will assess the accuracy of indices I and II in combination with serum CA19-9 in large cohorts of patients either seen by a family physician or referred to hospitals on suspicion of cancer or pancreatic cancer and assess whether it is possible to identify patients with pancreatic cancer at an early stage.

The microRNA candidates were selected in the discovery cohort in which the expressions of 754 microRNAs were tested. The indices were then developed using results from a training cohort in which the diagnostic ability was increased by the data-driven index II based on 10 microRNAs. In addition, both indices were validated in the discovery and validation cohorts.

Index II performed better than index I in the training cohort, but the indices performed almost identically in the validation cohort. The data-driven index II may have a tendency to overfit the data and lose power when tested in other populations (eg, as seen in the validation cohort), whereas index I was designed to reduce the influence from technical variation and is thus more robust.

Both indices performed better (according to AUC results) than serum CA19-9 in the discovery cohort; index II was also the best index in the training cohort. However, CA19-9 had the highest diagnostic accuracy compared with the indices alone in the validation cohort. When patients with pancreatic cancer were compared with healthy participants in the training cohort, the combination of index I and CA19-9 increased AUC (0.93) significantly compared with CA19-9 alone (0.90). The combination of index II and CA19-9 did not significantly increase AUC (0.97).

Although the increase in AUC was modest, combining CA19-9 with each of the indices increased the AUC in all 3 cohorts compared with CA19-9 alone. However, the increase in AUC was only significant for index I (pancreatic cancer vs healthy participants and patients with pancreatic cancer vs healthy participants plus patients with chronic pancreatitis) (Figure 3 and Table 3; eTable 4 in Supplement).23Figure 2 shows that the indices do achieve some separation of pancreatic cancer from healthy participants but not a complete separation, as also indicated by the performance of the indices.

Although the analyses do not show that these microRNA panels provide significant information over serum CA19-9, they raise the possibility that the indices could be useful in combination with CA19-9 or in situations in which CA19-9 is normal. All microRNAs included in index I were present in index II but with different weights. This suggests that index I is the best panel to be tested in combination with serum CA19-9 in future large case-control studies.

Our findings of significantly dysregulated microRNAs in the discovery cohort were consistent through the training and validation cohorts. The results obtained with our indices are similar, particularly when combined with CA19-9. There is also agreement with the only other study of whole blood microRNA expression profiles in patients with pancreatic cancer.13 Others have reported a microRNA classifier in serum for pancreatic cancer with an accuracy of 83.6%, which was higher than CA19-9 (56.4%),17 and a combination of 2 microRNAs in plasma with serum CA19-9 (AUC of 0.98), but this panel has not been validated.16

Several of the microRNAs found to be dysregulated in whole blood from patients with pancreatic cancer are associated with tumor or stem cell biology (let7g, miR-34a, miR-122, miR-145, miR-150, miR-223, miR-636).2433 The let-7 family is dysregulated in many types of cancer and is involved in Ras and Myc oncogene signaling and in JAK pathways.24 The miR-34a has been found to be related to cell cycle, differentiation, and apoptosis, regulation of p53 tumor suppressor function, and also dysregulation in plasma from patients with colorectal and breast cancer.2535 The miR-636 has been found to be involved in Ras signaling.33

Circulating microRNAs in whole blood can originate from distant sites of tissue damage, such as solid cancers and inflammatory foci, and from the neutrophils, monocytes, platelets, and mature red blood cells.3638 Whether microRNAs in blood from patients with pancreatic cancer also originate from circulating pancreatic adenocarcinoma cells is not known. The microRNA miR-223 in plasma is related to neutrophil and platelet counts and miR-150 is related to lymphocyte count.38 We found that both indices correlated with leukocytes and platelet counts, whereas microRNAs from mature red blood cells (miR-16, miR-92a, miR-451, miR-486)38,39 were not included in our indices.

The strengths of our study are the relatively large number of patients and controls included in the discovery and training cohorts, the large number of microRNAs analyzed in the discovery cohort for selection of microRNA candidates, and the validation of the indices in 2 other populations using another assay platform.

In the discovery cohort, we studied the sensitivity of results by applying different methods for normalization and imputation of missing values. In all experiments, we distributed samples such that factors like age, sex, and diagnosis were balanced with respect to day of purification and day of analysis or plate number and randomized within each day and plate. This aspect is important to reduce confounding from technical variation such as plate-to-plate variation and variation due to purification.

All 3 study populations came from the same Danish cohort, which is a limitation in relation to genetics, ethnicity, and geography. A better control group would have been age-matched patients included in the BIOPAC study with symptoms of pancreatic cancer in whom the pancreatic cancer diagnosis was ruled out. This group is small, but the performance of both index I and index II in patients with chronic pancreatitis included in the BIOPAC study was reasonable.

A limitation of our study is that the outcome was confounded by age due to an upper limit of 67 years for Danish volunteer healthy blood donors. Thus, cases were significantly older than controls in all 3 cohorts. However, separate analysis for cases and controls revealed no correlation between age and the 2 indices (eFigure 6 in Supplement). This finding suggests that the indices might be useful diagnostic biomarkers independent of age even under more realistic clinical conditions (eg, in a population of elderly people with chronic and inflammatory diseases and in patients either seen by a primary care physician or referred to the hospital for a pancreatic cancer diagnostic workup). However, the specificity reported in this study may be an overestimate of the specificity in an older control population (ie, may give more false-positives).

To our knowledge, no effect estimates of microRNAs in whole blood from patients with pancreatic cancer have previously been reported in the literature. A limitation of our study is that it could not be designed with a statistical power that was certain to detect differences in microRNA expression between patients with pancreatic cancer and healthy participants. Our study should therefore be seen as an exploratory study.

We have included more than 400 patients with pancreatic cancer and 300 healthy participants, which to our knowledge, is the largest study of microRNA in whole blood from patients with pancreatic cancer. The statistical precision is thus evaluated in terms of magnitude of confidence intervals of effect estimates and in terms of predictive ability of the proposed indices.

The use of a screening test with moderate diagnostic accuracy in a low-prevalence population, such as patients with pancreatic cancer, will result in a low positive predictive value, meaning that many persons with a positive test will not have cancer. It is likely that in a different control population (eg, in older persons with chronic and inflammatory diseases) that the specificity would be lower than in our study. This limitation raises the question of whether the test will be useful among typical patients. The harms of a high number of false-positives in screening for pancreatic cancer using an inexpensive, noninvasive blood sample from individuals with or without symptoms should be quantified in the future.

However, pancreatic cancer is a very lethal disease, and today most patients are diagnosed too late for surgery to be performed. Late diagnosis of pancreatic cancer results from a lack of early symptoms of the disease and the fact that even late symptoms are often not characteristic for pancreatic cancer.3,40 Although there is a risk of generating false-positive test results using our panels of microRNAs in combination with serum CA19-9, the test could refer more individuals with characteristic or uncharacteristic symptoms to CT, magnetic resonance, or ultrasound imaging. The test could thereby diagnose more patients with pancreatic cancer, some of them at an early stage, and thus have a potential to increase the number of patients that can be operated on and possibly cured of pancreatic cancer.

This study identified 2 diagnostic microRNA panels in whole blood that had the ability to distinguish, to a certain degree, patients with pancreatic cancer from healthy younger controls. Although we validated the panels, our findings are preliminary. Further research is necessary to understand whether these microRNAs have clinical implications as a screening test for early detection of pancreatic cancer and how much this information adds to serum CA19-9.

Corresponding Author: Julia S. Johansen, MD, DMSc, Departments of Oncology and Medicine, Herlev Hospital, Herlev Ringvej 75, DK-2730 Herlev, Denmark (julia.johansen@post3.tele.dk).

Author Contributions: Drs Schultz and Johansen had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Schultz, Dehlendorff, Jensen, Andersen, Johansen.

Acquisition of data: Schultz, Jensen, Bjerregaard, K. Nielsen, Bojesen, Calatayud, S. Nielsen, Yilmaz, Holländer, Johansen.

Analysis and interpretation of data: Schultz, Dehlendorff, Bjerregaard, Andersen, Johansen.

Drafting of the manuscript: Schultz, Dehlendorff, Andersen, Johansen.

Critical revision of the manuscript for important intellectual content: Schultz, Dehlendorff, Jensen, Bjerregaard, K. Nielsen, Bojesen, Calatayud, S. Nielsen, Yilmaz, Holländer, Andersen, Johansen.

Statistical analysis: Schultz, Dehlendorff, Andersen, Johansen.

Obtained funding: Jensen, Johansen.

Administrative, technical, and material support: Schultz, Jensen, K. Nielsen, Bojesen, Calatayud, S. Nielsen, Yilmaz, Holländer, Johansen.

Study supervision: Jensen, Bjerregaard, Holländer, Johansen.

Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Drs Schultz, Jensen, Andersen, and Johansen reported that they are named in a European patent application, which has been submitted by the Copenhagen Region, Copenhagen University Hospital.

Funding/Support: The study was supported by grants from the Research Council at Herlev Hospital, the Joint Proof-of-Concept Fund (Ministry of Science, Technology and Innovation, Denmark), Region Hovedstadens forsknings fond til Sundhedsforskning, the Danish Cancer Society, Aase og Ejnar Danielsens Fond, and Prosektor Axel Søeborg Ohlsen og Else Søeborg Ohlsen Mindelegat.

Role of the Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Additional Contribution: We thank the biomedical laboratory scientists at the participating departments for excellent technical assistance with handling of blood samples. We also thank Astrid Z. Johansen and Lærke M. Nelson, MD, for registration of clinical data in the BIOPAC database. In addition, we thank the nurses and physicians at the participating departments for inclusion of patients in the BIOPAC study. We also thank the patients for their participation in this study.

Siegel  R, Naishadham  D, Jemal  A.  Cancer statistics, 2013. CA Cancer J Clin. 2013;63(1):11-30.
PubMed   |  Link to Article
Ferlay  J, Steliarova-Foucher  E, Lortet-Tieulent  J,  et al.  Cancer incidence and mortality patterns in Europe. Eur J Cancer. 2013;49(6):1374-1403.
PubMed   |  Link to Article
Hidalgo  M.  Pancreatic cancer. N Engl J Med. 2010;362(17):1605-1617.
PubMed   |  Link to Article
Oettle  H, Neuhaus  P, Hochhaus  A,  et al.  Adjuvant chemotherapy with gemcitabine and long-term outcomes among patients with resected pancreatic cancer. JAMA. 2013;310(14):1473-1481.
PubMed   |  Link to Article
Conroy  T, Desseigne  F, Ychou  M,  et al.  FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817-1825.
PubMed   |  Link to Article
Von Hoff  DD, Ervin  T, Arena  FP,  et al.  Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med. 2013;369(18):1691-1703.
PubMed   |  Link to Article
Locker  GY, Hamilton  S, Harris  J,  et al; ASCO.  ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol. 2006;24(33):5313-5327.
PubMed   |  Link to Article
Iorio  MV, Croce  CM.  MicroRNAs in cancer. J Clin Oncol. 2009;27(34):5848-5856.
PubMed   |  Link to Article
Farazi  TA, Spitzer  JI, Morozov  P, Tuschl  T.  miRNAs in human cancer. J Pathol. 2011;223(2):102-115.
PubMed   |  Link to Article
miRBase. miRBase: the microRNA database. http://www.mirbase.org. Accessibility verified December 20, 2013.
Bloomston  M, Frankel  WL, Petrocca  F,  et al.  MicroRNA expression patterns to differentiate pancreatic adenocarcinoma from normal pancreas and chronic pancreatitis. JAMA. 2007;297(17):1901-1908.
PubMed   |  Link to Article
Szafranska  AE, Doleshal  M, Edmunds  HS,  et al.  Analysis of microRNAs in pancreatic fine-needle aspirates can classify benign and malignant tissues. Clin Chem. 2008;54(10):1716-1724.
PubMed   |  Link to Article
Bauer  AS, Keller  A, Costello  E,  et al.  Diagnosis of pancreatic ductal adenocarcinoma and chronic pancreatitis by measurement of microRNA abundance in blood and tissue. PLoS One. 2012;7(4):e34151.
PubMed   |  Link to Article
Schultz  NA, Werner  J, Willenbrock  H,  et al.  MicroRNA expression profiles associated with pancreatic adenocarcinoma and ampullary adenocarcinoma. Mod Pathol. 2012;25(12):1609-1622.
PubMed   |  Link to Article
Wang  J, Chen  J, Chang  P,  et al.  MicroRNAs in plasma of pancreatic ductal adenocarcinoma patients as novel blood-based biomarkers of disease. Cancer Prev Res (Phila). 2009;2(9):807-813.
PubMed   |  Link to Article
Liu  J, Gao  J, Du  Y,  et al.  Combination of plasma microRNAs with serum CA19-9 for early detection of pancreatic cancer. Int J Cancer. 2012;131(3):683-691.
PubMed   |  Link to Article
Liu  R, Chen  X, Du  Y,  et al.  Serum microRNA expression profile as a biomarker in the diagnosis and prognosis of pancreatic cancer. Clin Chem. 2012;58(3):610-618.
PubMed   |  Link to Article
Häusler  SFM, Keller  A, Chandran  PA,  et al.  Whole blood-derived miRNA profiles as potential new tools for ovarian cancer screening. Br J Cancer. 2010;103(5):693-700.
PubMed   |  Link to Article
Patnaik  SK, Yendamuri  S, Kannisto  E,  et al.  MicroRNA expression profiles of whole blood in lung adenocarcinoma. PLoS One. 2012;7(9):e46045.
PubMed   |  Link to Article
Chen  X, Liang  H, Zhang  J,  et al.  Secreted microRNAs. Trends Cell Biol. 2012;22(3):125-132.
PubMed   |  Link to Article
Ballehaninna  UK, Chamberlain  RS.  The clinical utility of serum CA 19-9 in the diagnosis, prognosis and management of pancreatic adenocarcinoma. J Gastrointest Oncol. 2012;3(2):105-119.
PubMed
McShane  LM, Hayes  DF.  Publication of tumor marker research results. J Clin Oncol. 2012;30(34):4223-4232.
PubMed   |  Link to Article
Attia  J, Ioannidis  JP, Thakkinstian  A,  et al.  How to use an article about genetic association. JAMA. 2009;301(3):304-308.
PubMed   |  Link to Article
Wang  X, Cao  L, Wang  Y,  et al.  Regulation of let-7 and its target oncogenes (review). Oncol Lett. 2012;3(5):955-960.
PubMed
Raver-Shapira  N, Marciano  E, Meiri  E,  et al.  Transcriptional activation of miR-34a contributes to p53-mediated apoptosis. Mol Cell. 2007;26(5):731-743.
PubMed   |  Link to Article
Chen  F, Hu  SJ.  Effect of microRNA-34a in cell cycle, differentiation, and apoptosis. J Biochem Mol Toxicol. 2012;26(2):79-86.
PubMed   |  Link to Article
Xu  J, Zhu  X, Wu  L,  et al.  MicroRNA-122 suppresses cell proliferation and induces cell apoptosis in hepatocellular carcinoma by directly targeting Wnt/β-catenin pathway. Liver Int. 2012;32(5):752-760.
PubMed   |  Link to Article
Jung  CJ, Iyengar  S, Blahnik  KR,  et al.  Epigenetic modulation of miR-122 facilitates human embryonic stem cell self-renewal and hepatocellular carcinoma proliferation. PLoS One. 2011;6(11):e27740.
PubMed   |  Link to Article
Zhang  J, Sun  Q, Zhang  Z, Ge  S, Han  ZG, Chen  WT.  Loss of microRNA-143/145 disturbs cellular growth and apoptosis of human epithelial cancers by impairing the MDM2-p53 feedback loop. Oncogene. 2013;32(1):61-69.
PubMed   |  Link to Article
Garzon  R, Croce  CM.  MicroRNAs in normal and malignant hematopoiesis. Curr Opin Hematol. 2008;15(4):352-358.
PubMed   |  Link to Article
Vasilatou  D, Papageorgiou  S, Pappa  V,  et al.  The role of microRNAs in normal and malignant hematopoiesis. Eur J Haematol. 2010;84(1):1-16.
PubMed   |  Link to Article
Jia  CY, Li  HH, Zhu  XC,  et al.  MiR-223 suppresses cell proliferation by targeting IGF-1R. PLoS One. 2011;6(11):e27008.
PubMed   |  Link to Article
Jang  JY, Lee  YS, Jeon  YK,  et al.  ANT2 suppression by shRNA restores miR-636 expression, thereby downregulating Ras and inhibiting tumorigenesis of hepatocellular carcinoma. Exp Mol Med.2013;45(e3).
PubMed
Nugent  M, Miller  N, Kerin  MJ.  Circulating miR-34a levels are reduced in colorectal cancer. J Surg Oncol. 2012;106(8):947-952.
PubMed   |  Link to Article
Eichelser  C, Flesch-Janys  D, Chang-Claude  J,  et al.  Deregulated serum concentrations of circulating cell-free microRNAs miR-17, miR-34a, miR-155, and miR-373 in human breast cancer development and progression. Clin Chem. 2013;59(10):1489-1496.
PubMed   |  Link to Article
Taylor  DD, Gercel-Taylor  C.  MicroRNA signatures of tumor-derived exosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol. 2008;110(1):13-21.
PubMed   |  Link to Article
Kong  XY, Du  YQ, Li  L,  et al.  Plasma miR-216a as a potential marker of pancreatic injury in a rat model of acute pancreatitis. World J Gastroenterol. 2010;16(36):4599-4604.
PubMed   |  Link to Article
Pritchard  CC, Kroh  E, Wood  B,  et al.  Blood cell origin of circulating microRNAs: a cautionary note for cancer biomarker studies. Cancer Prev Res (Phila). 2012;5(3):492-497.
PubMed   |  Link to Article
Kirschner  MB, Kao  SC, Edelman  JJ,  et al.  Haemolysis during sample preparation alters microRNA content of plasma. PLoS One. 2011;6(9):e24145.
PubMed   |  Link to Article
Seufferlein  T, Bachet  JB, Van Cutsem  E, Rougier  P; ESMO Guidelines Working Group.  Pancreatic adenocarcinoma. Ann Oncol. 2012;23(suppl 7):vii33-vii40.
PubMed   |  Link to Article

Figures

Place holder to copy figure label and caption
Figure 1.
Study Design

aThere were also 33 other periampullary cancers analyzed as part of this cohort study (15 ampullary, 12 distal common bile duct, and 6 duodenal).bIndicates low RNA yield, low absorbance, and few detectable microRNAs.

Graphic Jump Location
Place holder to copy figure label and caption
Figure 2.
Box Plots of Indices I and II Using the Whole Blood Samples From the Discovery, Training, and Validation Cohorts

The horizontal line in the middle of each box indicates the median, whereas the top and bottom borders of the box mark the 75th and 25th percentiles, respectively. The whiskers above and below the box extend to the most extreme point no longer than 1.5 times the interquartile range from the box. The points beyond the whiskers are outliers. Index I includes miR-150, miR-636, miR-145, miR-223. Index II includes miR-26b, miR-34a, miR-122, miR-126*, miR-145, miR-150, miR-223, miR-505, miR-636, miR-885.5p. P < .001 for all comparisons (healthy participants vs patients with cancer for indices I and II for each cohort).

Graphic Jump Location
Place holder to copy figure label and caption
Figure 3.
Reciever Operating Characteristic Curves for Performance of Indices I and II and in Combination With Serum Cancer Antigen 19-9 in the Discovery, Training, and Validation Cohorts

AUC indicates area under curve; CA19-9, cancer antigen 19-9. See legend of Figure 2 for definitions of indicies I and II.

Graphic Jump Location

Tables

Table Graphic Jump LocationTable 1.  Demographics of Patients and Healthy Participants in the Discovery, Training, and Validation Cohortsa
Table Graphic Jump LocationTable 2.  Prediction by MicroRNA of Patients With Pancreatic Cancer vs Healthy Participants (Controls) in the Training Cohort and vs Those With Chronic Pancreatitis Plus Controls in the Validation Cohort
Table Graphic Jump LocationTable 3.  Performance of Indices I and II and Serum Cancer Antigen 19-9 in the Differential Diagnosis of Pancreatic Cancer From Healthy Participants (Controls) and Patients With Chronic Pancreatitis (CP)

References

Siegel  R, Naishadham  D, Jemal  A.  Cancer statistics, 2013. CA Cancer J Clin. 2013;63(1):11-30.
PubMed   |  Link to Article
Ferlay  J, Steliarova-Foucher  E, Lortet-Tieulent  J,  et al.  Cancer incidence and mortality patterns in Europe. Eur J Cancer. 2013;49(6):1374-1403.
PubMed   |  Link to Article
Hidalgo  M.  Pancreatic cancer. N Engl J Med. 2010;362(17):1605-1617.
PubMed   |  Link to Article
Oettle  H, Neuhaus  P, Hochhaus  A,  et al.  Adjuvant chemotherapy with gemcitabine and long-term outcomes among patients with resected pancreatic cancer. JAMA. 2013;310(14):1473-1481.
PubMed   |  Link to Article
Conroy  T, Desseigne  F, Ychou  M,  et al.  FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817-1825.
PubMed   |  Link to Article
Von Hoff  DD, Ervin  T, Arena  FP,  et al.  Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med. 2013;369(18):1691-1703.
PubMed   |  Link to Article
Locker  GY, Hamilton  S, Harris  J,  et al; ASCO.  ASCO 2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol. 2006;24(33):5313-5327.
PubMed   |  Link to Article
Iorio  MV, Croce  CM.  MicroRNAs in cancer. J Clin Oncol. 2009;27(34):5848-5856.
PubMed   |  Link to Article
Farazi  TA, Spitzer  JI, Morozov  P, Tuschl  T.  miRNAs in human cancer. J Pathol. 2011;223(2):102-115.
PubMed   |  Link to Article
miRBase. miRBase: the microRNA database. http://www.mirbase.org. Accessibility verified December 20, 2013.
Bloomston  M, Frankel  WL, Petrocca  F,  et al.  MicroRNA expression patterns to differentiate pancreatic adenocarcinoma from normal pancreas and chronic pancreatitis. JAMA. 2007;297(17):1901-1908.
PubMed   |  Link to Article
Szafranska  AE, Doleshal  M, Edmunds  HS,  et al.  Analysis of microRNAs in pancreatic fine-needle aspirates can classify benign and malignant tissues. Clin Chem. 2008;54(10):1716-1724.
PubMed   |  Link to Article
Bauer  AS, Keller  A, Costello  E,  et al.  Diagnosis of pancreatic ductal adenocarcinoma and chronic pancreatitis by measurement of microRNA abundance in blood and tissue. PLoS One. 2012;7(4):e34151.
PubMed   |  Link to Article
Schultz  NA, Werner  J, Willenbrock  H,  et al.  MicroRNA expression profiles associated with pancreatic adenocarcinoma and ampullary adenocarcinoma. Mod Pathol. 2012;25(12):1609-1622.
PubMed   |  Link to Article
Wang  J, Chen  J, Chang  P,  et al.  MicroRNAs in plasma of pancreatic ductal adenocarcinoma patients as novel blood-based biomarkers of disease. Cancer Prev Res (Phila). 2009;2(9):807-813.
PubMed   |  Link to Article
Liu  J, Gao  J, Du  Y,  et al.  Combination of plasma microRNAs with serum CA19-9 for early detection of pancreatic cancer. Int J Cancer. 2012;131(3):683-691.
PubMed   |  Link to Article
Liu  R, Chen  X, Du  Y,  et al.  Serum microRNA expression profile as a biomarker in the diagnosis and prognosis of pancreatic cancer. Clin Chem. 2012;58(3):610-618.
PubMed   |  Link to Article
Häusler  SFM, Keller  A, Chandran  PA,  et al.  Whole blood-derived miRNA profiles as potential new tools for ovarian cancer screening. Br J Cancer. 2010;103(5):693-700.
PubMed   |  Link to Article
Patnaik  SK, Yendamuri  S, Kannisto  E,  et al.  MicroRNA expression profiles of whole blood in lung adenocarcinoma. PLoS One. 2012;7(9):e46045.
PubMed   |  Link to Article
Chen  X, Liang  H, Zhang  J,  et al.  Secreted microRNAs. Trends Cell Biol. 2012;22(3):125-132.
PubMed   |  Link to Article
Ballehaninna  UK, Chamberlain  RS.  The clinical utility of serum CA 19-9 in the diagnosis, prognosis and management of pancreatic adenocarcinoma. J Gastrointest Oncol. 2012;3(2):105-119.
PubMed
McShane  LM, Hayes  DF.  Publication of tumor marker research results. J Clin Oncol. 2012;30(34):4223-4232.
PubMed   |  Link to Article
Attia  J, Ioannidis  JP, Thakkinstian  A,  et al.  How to use an article about genetic association. JAMA. 2009;301(3):304-308.
PubMed   |  Link to Article
Wang  X, Cao  L, Wang  Y,  et al.  Regulation of let-7 and its target oncogenes (review). Oncol Lett. 2012;3(5):955-960.
PubMed
Raver-Shapira  N, Marciano  E, Meiri  E,  et al.  Transcriptional activation of miR-34a contributes to p53-mediated apoptosis. Mol Cell. 2007;26(5):731-743.
PubMed   |  Link to Article
Chen  F, Hu  SJ.  Effect of microRNA-34a in cell cycle, differentiation, and apoptosis. J Biochem Mol Toxicol. 2012;26(2):79-86.
PubMed   |  Link to Article
Xu  J, Zhu  X, Wu  L,  et al.  MicroRNA-122 suppresses cell proliferation and induces cell apoptosis in hepatocellular carcinoma by directly targeting Wnt/β-catenin pathway. Liver Int. 2012;32(5):752-760.
PubMed   |  Link to Article
Jung  CJ, Iyengar  S, Blahnik  KR,  et al.  Epigenetic modulation of miR-122 facilitates human embryonic stem cell self-renewal and hepatocellular carcinoma proliferation. PLoS One. 2011;6(11):e27740.
PubMed   |  Link to Article
Zhang  J, Sun  Q, Zhang  Z, Ge  S, Han  ZG, Chen  WT.  Loss of microRNA-143/145 disturbs cellular growth and apoptosis of human epithelial cancers by impairing the MDM2-p53 feedback loop. Oncogene. 2013;32(1):61-69.
PubMed   |  Link to Article
Garzon  R, Croce  CM.  MicroRNAs in normal and malignant hematopoiesis. Curr Opin Hematol. 2008;15(4):352-358.
PubMed   |  Link to Article
Vasilatou  D, Papageorgiou  S, Pappa  V,  et al.  The role of microRNAs in normal and malignant hematopoiesis. Eur J Haematol. 2010;84(1):1-16.
PubMed   |  Link to Article
Jia  CY, Li  HH, Zhu  XC,  et al.  MiR-223 suppresses cell proliferation by targeting IGF-1R. PLoS One. 2011;6(11):e27008.
PubMed   |  Link to Article
Jang  JY, Lee  YS, Jeon  YK,  et al.  ANT2 suppression by shRNA restores miR-636 expression, thereby downregulating Ras and inhibiting tumorigenesis of hepatocellular carcinoma. Exp Mol Med.2013;45(e3).
PubMed
Nugent  M, Miller  N, Kerin  MJ.  Circulating miR-34a levels are reduced in colorectal cancer. J Surg Oncol. 2012;106(8):947-952.
PubMed   |  Link to Article
Eichelser  C, Flesch-Janys  D, Chang-Claude  J,  et al.  Deregulated serum concentrations of circulating cell-free microRNAs miR-17, miR-34a, miR-155, and miR-373 in human breast cancer development and progression. Clin Chem. 2013;59(10):1489-1496.
PubMed   |  Link to Article
Taylor  DD, Gercel-Taylor  C.  MicroRNA signatures of tumor-derived exosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol. 2008;110(1):13-21.
PubMed   |  Link to Article
Kong  XY, Du  YQ, Li  L,  et al.  Plasma miR-216a as a potential marker of pancreatic injury in a rat model of acute pancreatitis. World J Gastroenterol. 2010;16(36):4599-4604.
PubMed   |  Link to Article
Pritchard  CC, Kroh  E, Wood  B,  et al.  Blood cell origin of circulating microRNAs: a cautionary note for cancer biomarker studies. Cancer Prev Res (Phila). 2012;5(3):492-497.
PubMed   |  Link to Article
Kirschner  MB, Kao  SC, Edelman  JJ,  et al.  Haemolysis during sample preparation alters microRNA content of plasma. PLoS One. 2011;6(9):e24145.
PubMed   |  Link to Article
Seufferlein  T, Bachet  JB, Van Cutsem  E, Rougier  P; ESMO Guidelines Working Group.  Pancreatic adenocarcinoma. Ann Oncol. 2012;23(suppl 7):vii33-vii40.
PubMed   |  Link to Article

Letters

CME
Meets CME requirements for:
Browse CME for all U.S. States
Accreditation Information
The American Medical Association is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The AMA designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 CreditTM per course. Physicians should claim only the credit commensurate with the extent of their participation in the activity. Physicians who complete the CME course and score at least 80% correct on the quiz are eligible for AMA PRA Category 1 CreditTM.
Note: You must get at least of the answers correct to pass this quiz.
You have not filled in all the answers to complete this quiz
The following questions were not answered:
Sorry, you have unsuccessfully completed this CME quiz with a score of
The following questions were not answered correctly:
Commitment to Change (optional):
Indicate what change(s) you will implement in your practice, if any, based on this CME course.
Your quiz results:
The filled radio buttons indicate your responses. The preferred responses are highlighted
For CME Course: A Proposed Model for Initial Assessment and Management of Acute Heart Failure Syndromes
Indicate what changes(s) you will implement in your practice, if any, based on this CME course.

Multimedia

Supplement.

eText. Patients, Methods, MiRNA Expression Analysis, Statistics, Results

eReferences

eTable 1. Quality of Polymerase Chain Reaction Assays

eTable 2. Odds Ratios and 95% Confidence Intervals for MicroRNAs From the Discovery Study

eTable 3. Differences in MicroRNAs in Patients With Pancreatic Cancer vs Healthy Participants in the 3 Cohorts and in Other Studies

eTable 4. Comparison of the Area Under the Curve for Serum Cancer Antigen (CA) 19-9 vs Indices I and II and Serum CA19-9 vs Index I Plus CA19-9 and Index II Plus CA19-9

eTable 5. Performance of Indices I and II Alone, Serum Cancer Antigen (CA) 19-9 Alone, and Both Indices I and II in Combination With Serum CA19-9

eFigure 1. Cycle Threshold Values of Specific MiRNA Imputed by Random Sampling

eFigure 2. Box Plots of Relative Expression of Several MiRNAs

eFigure 3. Box Plots of 19 MiRNAs Significantly Differently Expressed in Whole Blood in Patients With Pancreatic Cancer vs Healthy Participants

eFigure 4. Box Plots of 10 MiRNAs Included in Indices I and II

eFigure 5. Indices I and II by Tumor Stage in the 3 Cohorts

eFigure 6. Indices I and II by Sex and Age in the 3 Cohorts

eFigure 7. Box Plots of Indices I and II Using the Other Periampullary Cancer Sample

Supplemental Content

Some tools below are only available to our subscribers or users with an online account.

Web of Science® Times Cited: 4

Related Content

Customize your page view by dragging & repositioning the boxes below.

See Also...
Articles Related By Topic
Related Collections
PubMed Articles