www.impactjournals.com/oncotarget/ Oncotarget, 2018, Vol. 9, (No.

10), pp: 8849-8869

Research Paper: Pathology

LRP1 expression in colon cancer predicts clinical outcome
Camille Boulagnon-Rombi1,2, Christophe Schneider2,3, Chloé Leandri4, Albin
Jeanne2,5, Virginie Grybek6, Aude Marchal Bressenot2, Coralie Barbe7, Benjamin
Marquet1, Saviz Nasri8, Christelle Coquelet8, Caroline Fichel1, Nicole Bouland1,
Arnaud Bonnomet9, Reza Kianmanesh10, Anne-Sophie Lebre6, Olivier Bouché4,
Marie-Danièle Diebold1,2, Georges Bellon2,11 and Stéphane Dedieu2,3
1
Laboratoire de Biopathologie, Centre Hospitalier Universitaire et Faculté de Médecine, Reims, France
2
CNRS UMR 7369, Matrice Extracellulaire et Dynamique Cellulaire, MEDyC, Reims, France
3
Université de Reims Champagne-Ardenne, UFR Sciences Exactes et Naturelles, Campus Moulin de la Housse, Reims, France
4
Service de Gastro-entérologie et Cancérologie Digestive, Centre Hospitalier Universitaire, Reims, France
5
SATT Nord, Lille, France
6
Laboratoire de Génétique, Centre Hospitalier Universitaire, Reims, France
7
Unité d’Aide Méthodologique, Centre Hospitalier Universitaire, Reims, France
8
CRB Tumorothèque de Champagne-Ardenne, Reims, France
9
Plateforme d’Imagerie Cellulaire et Tissulaire, Université de Reims Champagne-Ardenne, Reims, France
10
Service de Chirurgie Digestive, Centre Hospitalier Universitaire, Reims, France
11
Laboratoire de Biochimie, Centre Hospitalier Universitaire, Reims, France
Correspondence to: Camille Boulagnon-Rombi, email: camille.boulagnon@gmail.com
Keywords: colorectal cancer; LRP1; miR-205; BRAF; microsatellite instability; Pathology
Received: May 11, 2017 Accepted: January 09, 2018 Published: January 13, 2018
Copyright: Boulagnon-Rombi et al. This is an open-access article distributed under the terms of the Creative Commons Attribution
License 3.0 (CC BY 3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author
and source are credited.

ABSTRACT

LRP1 (low-density lipoprotein receptor-related protein 1), a multifunctional

endocytic receptor, has recently been identified as a hub within a biomarker network
for multi-cancer clinical outcome prediction. As its role in colon cancer has not yet
been characterized, we here investigate the relationship between LRP1 and outcome.
Materials and Methods: LRP1 mRNA expression was determined in colon
adenocarcinoma and paired colon mucosa samples, as well as in stromal and tumor
cells obtained after laser capture microdissection. Clinical potential was further
investigated by immunohistochemistry in a population-based colon cancer series
(n = 307). LRP1 methylation, mutation and miR-205 expression were evaluated and
compared with LRP1 expression levels.
Results: LRP1 mRNA levels were significantly lower in colon adenocarcinoma
cells compared with colon mucosa and stromal cells obtained after laser capture
microdissection. Low LRP1 immunohistochemical expression in adenocarcinomas was
associated with higher age, right-sided tumor, loss of CDX2 expression, Annexin A10
expression, CIMP-H, MSI-H and BRAFV600E mutation. Low LRP1 expression correlated
with poor clinical outcome, especially in stage IV patients. While LRP1 expression was
downregulated by LRP1 mutation, LRP1 promoter was never methylated.
Conclusions: Loss of LRP1 expression is associated with worse colon cancer
outcomes. Mechanistically, LRP1 mutation modulates LRP1 expression.

www.impactjournals.com/oncotarget 8849 Oncotarget

 INTRODUCTION neutralization could abrogate cell motility in both tumor
and non-tumor cells, and this despite an increase in
Colorectal cancer (CRC) is the third most common pericellular proteolytic activities of several extracellular
cancer diagnosed worldwide in men and the second in proteases such as MMP2 (Matrix Metalloproteinase 2),
women. Despite advances in screening, diagnosis and MMP9 and uPA (urokinase Plasminogen Activator)
management of the disease, it remains the fourth cancer in [20, 27]. On the other hand, LRP1 silencing prevents
terms of mortality. Metastatic disease ultimately occurs in spread of glioblastoma cells [28]. Therefore, LRP1
approximately 50–70% of patients presenting colorectal influence on tumor cell migration and invasion
cancer [1–3]. UICC staging is the only prognostic likely depends on the tumor cell type and the specific
classification used in clinical practice to select patients for extracellular proteins involved in these processes [29].
adjuvant chemotherapy [4]. Currently, CRC has relatively In CRC, little is known about LRP1 and its
few established biomarkers to predict patient outcome. putative function. Previous studies on few colon
Molecular markers include microsatellite instability adenocarcinomas samples showed a frequent loss of LRP1
(MSI), RAS and BRAF mutation. RAS and BRAF mutation immunohistochemical expression in adenocarcinomatous
status are used to guide therapeutic decisions in metastatic cells [30, 31]. To further expand our knowledge on the
CRC patients. CRC with RAS or BRAF mutations are relevance of considering LRP1 expression in colon cancer,
unlikely to respond to anti-epidermal growth factor we analyzed LRP1 expression level and distribution in a
receptor (EGFR) antibody therapy [5–7]. Patients with series of 307 colon cancers with follow-up data. We then
nonhereditary MSI tumors have better prognosis than determined whether LRP1 expression is linked to clinical
those with microsatellite stable (MSS) tumors [1, 2, 8–10], characteristics and outcomes while analyzing the role of
and MSI is currently implemented in clinical guidelines as miRNA expression, LRP1 mutation and methylation in
a prognostic biomarker, especially in stage II CRC patients LRP1 expression profile.
[11]. However, these histomolecular parameters hardly
apprehend disease heterogeneity and are insufficient for RESULTS
recurrence and prognostic prediction in an individual
patient. Therefore, robust biomarkers that can stratify Patients and clinicopathological features
patient prognosis groups and improve treatment strategies
are urgently needed. In total, 307 colon cancer patients were included
The low-density lipoprotein receptor (LDLR)- in our study. The population comprised 174 (57%) men
related protein-1 (LRP1), a member of LDLR family, is and 133 (43%) women, whose mean age was 71 years
a large multifunctional endocytic cell surface receptor, (± 11 years). Tumors were right-sided in 136 cases (44%)
which is ubiquitously expressed [12, 13]. This large and left-sided in 171 cases (56%). Follow-up data were
transmembrane receptor recognizes numerous ligands, available for all except 12 patients. The mean follow-up
therefore regulating a wide range of biological functions. time was 43 months (± 32 months). Clinicopathological
It both acts as a signaling and clearance receptor. The features of the cohort are detailed in Table 1.
biological activity of LRP1 was initially characterized
as a clearance receptor for chylomicron remnants LRP1 is lower expressed in adenocarcinoma
and complexes of α2-macroglobulin with proteinases cells compared with normal colon mucosa and
[14]. Subsequent work has revealed that this receptor stromal cells
regulates the cell ability to respond to growth factors, to
interact with extracellular matrix, as well as to respond LRP1 mRNA expression analyses by quantitative
to perturbations that occur within the microenvironment real-time reverse transcriptase polymerase chain reaction
[15–18]. Numerous studies have suggested a role for (qRT-PCR) on 192 colonic adenocarcinoma samples and
LRP1 in regulation of tumor growth and progression. 105 colonic mucosa samples with RQI values ≥5 showed
LRP1 has recently been identified as a hub within a a 4.08-fold decrease in LRP1 expression within tumor
biomarker network for multi-cancer clinical outcome samples when compared with normal colon samples
prediction [19]. However, the role of LRP1 varies from (Figure 1A and 1B). LRP1 was overexpressed when
one tumor type to another. Indeed, several studies have compared with paired normal colon samples in only 9/85
reported that low LRP1 expression was closely related adenocarcinoma cases (10.6%; data not shown).
to advanced tumor stages and poor survival in several To describe LRP1 distribution in colon tissues,
solid tumors, such as hepatocellular carcinoma [20], lung immunohistochemical (IHC) analyses were performed on
adenocarcinoma [21] melanoma [22] and Wilms tumors paired normal colon mucosa samples (n = 117) and colon
[23]. On the contrary, high LRP1 expression was related adenocarcinomas (n = 307). In colon mucosa, epithelial
to advanced tumor stages in endometrial carcinoma [24], cells expressed LRP1 in 86% cases (101/117). In the
breast cancer [25] and prostate carcinomas [26]. Using majority (85/101) of these cases, LRP1 expression was
in vitro models, it has been demonstrated that LRP1 limited to surface epithelium (Figure 1C). Some fibroblasts

www.impactjournals.com/oncotarget 8850 Oncotarget

 Table 1: Clinicopathological features of the cohort
Clinical/pathological features No. available (%)
Gender
Male 174 (57)
Female 133 (43)
Mean age [range] 71 years [41-91 years]
UICC stage
Stage I 35 (11)
Stage II 117 (38)
Stage III 79 (26)
Stage IV 76 (25)
Tumor location
Left colon 171 (56)
Right colon 136 (44)
Occlusion
Yes 35 (11)
No 272 (89)
Tumor perforation
Yes 20 (7)
No 287 (93)
Differentiation grade
Grade 1–2 258 (84)
Grade 3 49 (16)
KRAS status
Wild type 101 (68)
Mutant 48 (32)
BRAF status
Wild type 260 (86)
Mutant 44 (14)
Microsatellite status
MSS 266 (87)
MSI 40 (13)
CIMP status
No CIMP 22 (34)
CIMP-Low 35 (55)
CIMP-High 7 (11)

of the lamina propria expressed LRP1 (Figure 1D). In adenoma cells when compared with adenocarcinoma cells
adenocarcinoma, LRP1 was expressed in malignant cells (data not shown), and this whatever the grade.
in 244/307 (79%) of the cases. The mean IHC tumor score
was 6.22 ± 3.62. In these adenocarcinoma samples, stromal Microdissection analyses confirm low LRP1
fibroblasts expressed LRP1 in all cases (Figure 1E–1H). mRNA expression in malignant cells compared
The mean IHC stroma score was near optimal (10.82 ± to stromal cells
2.44). LRP1 was never found to be expressed in stromal
lymphocytes (Figure 1G). Immunohistochemical expression Owing to the difference of LRP1 IHC expression
of LRP1 was inversely correlated in malignant and stromal between stromal and malignant cells, we performed Laser
cells (p = 0.0003; R2 = 0.04). We didn’t find any difference Capture Microdissection (LCM) analyses to distinguish
of IHC scores between the center and the invasive front of between LRP1 mRNA expression arising from malignant
the adenocarcinomas for both tumor and stromal cells. and stromal cells. LCM was performed on available
Furthermore, IHC analyses performed on 14 fresh frozen samples of 32 colon adenocarcinomas. The
conventional adenomas (8 low-grade, 6 high-grade) efficiency of LCM for separating malignant and stromal
revealed that LRP1 IHC score was significantly higher in cell was ensured morphologically (Figure 2A and 2B)

www.impactjournals.com/oncotarget 8851 Oncotarget

 Figure 1: LRP1 expression in colon cancer cells compared to normal colon and stromal cells. (A) qRT-PCR expression
levels of LRP1 mRNA in colon adenocarcinoma fresh frozen samples compared with normal colon mucosa fresh frozen samples. Values
are shown as dCt normalized with RPL32 (B) Comparative quantification analysis of LRP1 mRNA expression levels in tumor samples
compared with paired normal colon mucosa samples. Values are shown as ddCt fold induction. ****p < 0.0001, Mann Whitney test. (C–H)
Representative microphotographs of LRP1 immunohistochemistry on colon mucosa (C–D) and colon adenocarcinoma (E–H). (C) LRP1
expression in surface epithelium (arrows) in normal colon mucosa (×5 magnification). (D) LRP1 expression in fibroblasts of the lamina
propria (arrows) in normal colon mucosa (×10 magnification). Loss of LRP1 expression in malignant cells of a moderately differentiated
adenocarcinoma (E) and a mucinous adenocarcinoma (F) (×20 magnification). (G) Loss of LRP1 expression in malignant cells (asterisks)
and stromal lymphocytes (arrows) of a poorly differentiated adenocarcinoma (×30 magnification). (H) LRP1 expression in malignant and
stromal cells of a moderately differentiated adenocarcinoma (×20 magnification).

www.impactjournals.com/oncotarget 8852 Oncotarget

 and by mRNA quantification of the epithelial marker In summary, low LRP1 expression evaluated by
carcinoembryonic antigen (CEA) (Figure 2C and 2D). immunohistochemistry and by qRT-PCR in two independent
LCM analyses revealed that LRP1 mRNA expression was cohorts is strongly associated with right tumor location,
5.1-fold lower in adenocarcinoma cells than in stromal MSI-H, BRAF mutation and CIMP-H. These characteristics
cells (Figure 2E and 2F). are those of the hypermutated type of the TCGA [36]. Right
Tumor IHC scores were not correlated with LRP1 colonic cancers with this molecular subtype of CRC are
mRNA expression levels of whole adenocarcinoma known to have a poor prognosis [32–35].
samples (p = 0.10; R2 = 0.02) (Figure 2G) but were
correlated with LRP1 mRNA expression in tumor cells Low LRP1 immunohistochemical expression in
obtained after LCM (p = 0.003; R2 = 0.28) (Figure 2H). tumor cells correlates with poor overall survival
Thus, overall LRP1 mRNA expression does not reflect
malignant cells expression but reflects the sum of We subsequently analyzed the relation between
malignant and non-malignant cells expression. LRP1 expression and prognosis. As detailed in Table
4, univariate analysis in our cohort revealed that age,
Adenocarcinomas with low LRP1 metastatic status, histological grade, vascular invasion,
immunohistochemical expression have a distinct perineural invasion and CDX2 expression were predictors
clinicopathological and molecular phenotype of overall survival (OS). Low LRP1 IHC score in tumor
cells (score ≤ 4) was predictor of poor OS (p = 0.003)
Relationship between clinico-pathological and (Figure 4A and Table 4). The value of LRP1 IHC score
molecular parameters with LRP1 immunohistochemical in tumor cells nearly reached statistical significance as
score in malignant and stromal cells were evaluated in a prognosis indicator of OS (p = 0.09) in multivariate
our cohort. As detailed in Table 2, colon adenocarcinomas analyses.
with low tumor IHC score were associated on univariate Metastatic status, vascular invasion and KRAS
analyses with female gender, higher age, right location, mutation were the only independent predictors of event
high differentiation grade, mucinous type, Annexin A10 free survival (EFS) in our cohort (Table 4, Figure 4B).
expression, loss of CDX2 expression, MSI-H status, In our cohort, LRP1 mRNA expression levels were
BRAFV600E mutation, absence of KRAS mutation and not correlated with OS and EFS (Table 4).
CIMP-H. On multivariate analyses, low LRP1 IHC Stage-specific analyses revealed that LRP1
score in tumor cells was associated with right location expression was not a survival predictor of both OS and
(p = 0.0004), MSI-H (p = 0.01) and BRAFV600E mutation EFS in UICC stage II and III patients (data not shown).
(p = 0.009). Moreover, IHC results on tumor cells were Furthermore, as detailed in Table 5, in metastatic patients
confirmed at the mRNA level by qRT-PCR for age (p = (stage IV, n = 76), low LRP1 IHC score in malignant cells
0.01), BRAFV600E mutation (p = 0.05) and CIMP-H was an independent predictor of poor OS on univariate
phenotype (p < 0.001) (Figure 3A–3C). (p = 0.004, Figure 4C) and multivariate (p = 0.03)
Furthermore, low LRP1 stromal IHC score was analyses. Low LRP1 IHC score in malignant cells was
associated on univariate analyses with younger age, predictor of shorter progression-free survival (PFS) on
UICC stage and mucinous type as detailed in Table 3. On univariate analyses only (Table 5, Figure 4D).
multivariate analysis, low LRP1 stromal IHC score was Among stage IV patients from our cohort with
associated with younger age only. available information regarding medical treatment, 48
Thus, LRP1 IHC score on tumor cells was received 5-fluorouracil-based chemotherapies (LV5FU2, 9;
associated with peculiar clinicopathological and molecular FOLFOX, 14; FOLFIRI, 24; FOLFIXIRI, 1). The
characteristics. Despite an inverse correlation between most frequently used targeted therapy was the Vascular
tumor and stromal cells IHC scores, these peculiar Endothelial Growth Factor (VEGF) inhibitor bevacizumab
characteristics were not found in stromal cells. (37/76), followed by the Epidermal Growth Factor Receptor
To further confirm our results in an independent (EGFR) inhibitors cetuximab (10/76) and panitumumab
patients’ cohort, associations between LRP1 mRNA (7/76). In patients treated with bevacizumab (n = 37), low
expression levels and available clinical and molecular LRP1 IHC score in tumor cells was associated with shorter
characteristics were studied in the TCGA cohort (n = 212) OS (Figure 4E). However, low LRP1 IHC tumor score was
[36]. As in our cohort, LRP1 mRNA expression levels not predictor for PFS in these patients (Figure 4F).
among the TCGA cohort were significantly lower in High LRP1 IHC score in stromal cells was predictor
cases with right tumor location (p = 0.0003), MSI-H of shorter PFS only on both univariate and multivariate
(p < 0.0001), BRAF mutation (p = 0.0015) CIMP-H analyses in stage IV patients only (Tables 4 and 5).
(p < 0.0001) and low CDX2 expression (p < 0.0001) To confirm our results, we performed survival
(Figure 3D–3H). In this cohort, LRP1 mRNA expression analyses in the SieberSmith cohort (n = 286) from R2
levels were not associated with patients’ gender, database [37, 38]. In this cohort, high LRP1 mRNA
AnnexinA10 (ANXA10) expression and KRAS mutation. expression was a poor prognostic predicator for EFS

www.impactjournals.com/oncotarget 8853 Oncotarget

 Figure 2: Laser capture microdissection analyses. (A–B) Representative microphotographs of microscopic control of laser capture
microdissection (LCM) (Cresyl violet, ×20 magnification). (A) Microdissection of the malignant cells. (B) Microdissection of the stromal
cells. Residual malignant glands are highlighted with an asterisk. (C) qRT-PCR expression levels of CEA mRNA in adenocarcinoma cells
compared with stromal cells after LCM. Values are shown as dCt normalized with RPL32. (D) Comparative quantification analysis of CEA
mRNA expression levels in tumor cells compared with stromal cells after LCM. Values are shown as ddCt fold induction. (E) qRT-PCR
expression levels of LRP1 mRNA in adenocarcinoma cells compared with stromal cells after LCM. Values are shown as dCt normalized
with RPL32. (F) Comparative quantification analysis of LRP1 mRNA expression levels in tumor cells compared with stromal cells after
LCM. Values are shown as ddCt fold induction. *p < 0.05; **p < 0.01; ****p < 0.0001, Mann Whitney test. (G) Linear regression analysis of
LRP1 mRNA expression levels evaluated by qRT-PCR on complete fresh frozen adenocarcinoma sample against LRP1 IHC score of tumor
cells obtained by multiplying staining intensity (0 to 3) and percentage of positive cells (0 to 4). (H) Linear regression analysis of LRP1
mRNA expression of tumor cells against LRP1 IHC score of tumor cells obtained by multiplying staining intensity (0 to 3) and percentage
of positive cells (0 to 4) after LCM.

www.impactjournals.com/oncotarget 8854 Oncotarget

 Table 2: Clinicopathological characteristics associated with low LRP1 immunohistochemical expression in
adenocarcinomatous cells
n Mean LRP1 tumor score p univariate p multivariate
Sex 307 0.0006 ‡ n.s
Male 174 5.64 ± 4.01
Female 133 4.03 ± 4.03
Age 307 0.0009 ‡ 0.09
≤ 71 years 140 5.78 ± 4.2
> 71 years 167 4.24 ± 3.89
Tumor location 307 <0.0001 ‡ 0.0004
Right 136 3.34 ± 3.60
Left 171 6.22 ± 4.02
UICC stage 304 0.20 † NA
Stage I 35 6.2 ± 3.53
Stage II 115 5.07 ± 4.37
Stage III 79 4.76 ± 4.12
Stage IV 75 4.45 ± 3.79
Vascular invasion 300 0.30 ‡ NA
Yes 129 4.71 ± 4.1
No 171 5.2 ± 4.07
Perineural invasion 300 0.51 ‡ NA
Yes 87 4.75 ± 3.97
No 213 5.09 ± 4.14
Budding score 286 0.49 ‡ NA
High 14 4.21 ± 4.28
Low 272 4.99 ± 4.08
Differentiation grade 307 <0.0001 ‡ n.s
Grade 1-2 258 5.44 ± 4.03
Grade 3 49 2.35 ± 3.39
CDX2 303 0.0003 ‡ n.s
Positive 278 5.19 ± 3.98
Negative 25 2.16 ± 4.19
Mucinous type 287 0.004 ‡ n.s
Yes 19 2.37 ± 3.58
No 268 5.14 ± 4.06
Annexin A10 305 <0.0001 ‡ n.s
Positive 39 1.90 ± 3.31
Negative 266 5.41 ± 4.01
KRAS status 149 0.003 ‡ n.s
Wild type 101 3.35 ± 3.72
Mutant 48 5.31 ± 3.82
BRAF status 303 <0.0001 ‡ 0.009
Wild type 259 5.5 ± 3.93
Mutant 44 1.29 ± 2.69
Microsatellite status 305 <0.0001 ‡ 0.01
MSS 265 5.52 ± 3.96
MSI 40 1.1 ± 2.47
CIMP status 62 0.02 † NA
No CIMP 23 5.09 ± 3.94
CIMP-Low 32 3.75 ± 4.54
CIMP-High 7 0±0
NA : Not adopted ; n.s : not significant; ‡ T test † Linear regression.

www.impactjournals.com/oncotarget 8855 Oncotarget

 (p = 0.0006) in the entire cohort (Figure 5A). Stage- In the SieberSmith cohort, overall LRP1 mRNA
specific analyses in this cohort revealed that LRP1 was an expression may have little significance because it reflects
indicator of EFS in stage III patients only (Figure 5B–5D). the ratio of epithelial cells (normal and malignant) versus
In this patient’s group (n = 75), high LRP1 expression was non-epithelial cells.
associated with shorter EFS (p = 0.006). OS data were not
available for this cohort. Analyses of LRP1 expression regulation by
Despite apparent conflicting results of LRP1 IHC mutation, methylation and microRNA
and mRNA survival analyses, LRP1 expression was found
to be a strong prognosis indicator. IHC analyses allowed In order to explain the decrease of LRP1 expression
to distinguish LRP1 expression between malignant and in malignant cells, we analyzed genetic and epigenetic
stromal cells. In our cohort, LRP1 IHC score in malignant modifications that could be involved in LRP1 expression
cells was a strong prognosis indicator for OS especially regulation. First, mutation analysis among the TCGA cohort
in stage IV patients, whereas LRP1 IHC score in stromal dataset [36] revealed that LRP1 gene mutation was rare (6%;
cells was an indicator of PFS in stage IV patients only. 12/212), without particular hotspot mutation site (Figure 6A).

Figure 3: Correlation of LRP1 mRNA levels with clinical and molecular findings. Left panel: LRP1 mRNA levels analyses
by qRT-PCR (dCt normalized with RPL32) on fresh frozen colon adenocarcinoma samples from our cohort compared with age (A),
BRAFV600E mutation (B) and CpG island methylator phenotype (CIMP-H) (C). Right panel: Correlation analysis of LRP1 mRNA
expression levels extracted from the colorectal cancer cohort of the TCGA, as retrieved using cBioportal for Cancer Genomics (http://
cbioportal.org) web resources with sided adenocarcinomas (D), BRAF mutation (E), CIMP status (F), MSI status (G). and CDX2 mRNA
expression (H). *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001, Mann Whitney test. Abbreviations: H, high; L, Low; MSI, microsatellite
instability; MSS, microsatellite stability; CIMP, CpG island methylator phenotype.

www.impactjournals.com/oncotarget 8856 Oncotarget

 Table 3: Clinicopathological characteristics associated with LRP1 immunohistochemical expression in stromal cells
n Mean LRP1 stromal score p univariate p multivariate
Sex 307 0.20‡ NA
Male 174 10.80 ± 2.37
Female 133 11.14 ± 2.22
Age 307 0.03‡ 0.004
≤71 years 140 10.63 ± 2.63
>71 years 167 11.22 ± 1.96
Tumor location 307 0.53 ‡ NA
Right 136 11.04 ± 2.20
Left 171 10.88 ± 2.39
UICC stage 304 0.02 † n.s
Stage I 35 9.94 ± 3.07
Stage II 115 11.16 ± 2.11
Stage III 79 11.30 ± 1.75
Stage IV 75 10.85 ± 2.44
Vascular invasion 300 0.24 ‡ NA
Yes 129 11.12 ± 2.08
No 171 10.81 ± 2.47
Perineural invasion 300 0.61‡ NA
Yes 87 10.84 ± 2.31
No 213 10.99 ± 2.32
Budding score 286 0.41 ‡ NA
High 14 10.21 ± 3.24
Low 272 10.96 ± 2.27
Differentiation grade 307 0.17 ‡ NA
Grade 1–2 258 10.83 ± 2.47
Grade 3 49 11.51 ± 1.41
CDX2 303 0.32 ‡ NA
Positive 278 10.36 ± 3.07
Negative 25 10.99 ± 2.24
Mucinous type 287 0.003 ‡ n.s
Yes 19 11.68 ± 0.94
No 268 10.87 ± 2.38
Annexin A10 305 0.81 ‡ NA
Positive 39 11.02 ± 2.03
Negative 266 10.93 ± 2.36
KRAS status 149 0.11 ‡ NA
Wild type 101 11.22 ± 2.06
Mutant 48 10.60 ± 2.39
BRAF status 303 0.89 ‡ NA
Wild type 259 10.97 ± 2.32
Mutant 44 11.02 ± 2.10
Microsatellite status 305 0.81 ‡ NA
MSS 265 10.93 ± 2.34

www.impactjournals.com/oncotarget 8857 Oncotarget

 MSI 40 11.02 ± 2.18
CIMP status 62 0.36 † NA
No CIMP 23 10.61 ± 3.00
CIMP-Low 32 11.28 ± 1.59
CIMP-High 7 10.00 ± 3.46
NA : not adopted ; n.s : not significant; ‡ T test † Linear regression.

Table 4: Univariate and multivariate analyses of factors associated with overall and event-free survival in our entire
cohort of 307 patients
Overall Survival Event Free Survival
Variables Univariate Multivariate Univariate Multivariate
p value HR 95%CI p value p value HR 95%CI p value
Age 0.005 1.03 1.01–1.05 0.0004 0.37 N.A
Metastasis (M0 vs. M+) <0.0001 2.10 1.40–3.13 <0.0001 <0.0001 1.57 1.01–2.45 0.04
Vascular invasion (yes vs. no) <0.0001 1.56 1.10–2.23 0.01 <0.0001 1.89 1.17–3.02 0.008
Perineural invasion (yes vs. no) 0.002 n.s <0.0001 n.s
Differentiation grade (3 vs. 1-2) 0.003 n.s 0.003 n.s
CDX2 IHC expression (yes vs. no) 0.0005 1.59 0.93–2.72 0.09 0.11 N.A
KRAS mutation (yes vs. no) 0.22 N.A 0.003 1.62 1.06–2.49 0.03
LRP1 IHC tumor score (low vs. high) 0.003 1.35 0.95–1.93 0.09 0.46 N.A
LRP1 IHC stroma score (low vs. high) 0.42 N.A 0.92 N.A
LRP1 mRNA 0.12 N.A 0.59 N.A
n.s: not significant; N.A: not adopted; HR: hazard ratio. Results were adjusted on T and N.

Then, LRP1 mutation was strongly associated with female To evaluate the putative contribution of miRNA,
gender (p < 0.0001), right tumor location (p = 0.04), MSI-H two of the most important miRNA implicated in LRP1
(p < 0.0001) and CIMP-H status (p = 0.0006) (Figure 6B). expression regulation i.e. miR-205 and miR-338-5p
Besides, LRP1 mRNA expression was lower expressed in the were assessed on available fresh frozen samples of 49
LRP1-mutated group when compared with LRP1 wild type adenocarcinomas and 29 paired normal colon mucosa.
group (p = 0.003) (Figure 6C). Hence, although infrequent, In these samples, both miR-205 and miR-338-5p were
LRP1 mutations may partly explain the decrease in LRP1 significantly higher expressed in adenocarcinomas than
mRNA expression in some CRC. in normal colon (Figure 7A and 7B). Moreover, linear
Due to the low rate of LRP1 mutation, it is likely regression analyses revealed that miR-205 tended to
that other phenomenon, such as epigenetic modifications, stimulate LRP1 mRNA expression (p = 0.06; R2 = 0.10)
may be involved in LRP1 gene expression regulation. (Figure 7C) despite the absence of correlation with LRP1
To explore LRP1 epigenetic modifications, we analyzed IHC score in tumor cells (Figure 7D). Additionally, no
both intronic and promoter methylation on available fresh
correlation was found between LRP1 mRNA level or
frozen samples of 64 adenocarcinomas and 39 normal
IHC score in tumor cells and miR-338-5p expression
colon mucosa. Surprisingly, LRP1 promoter or intronic
(Figure 7E and 7F). Thus, miR-205 expression does not
methylation levels were very low in all these samples.
appear to be implied in the low expression of LRP1 in
Moreover, LRP1 mRNA expression levels and LRP1 IHC
adenocarcinomatous cells.
score in tumor cells were neither correlated with LRP1
intronic or promoter levels nor with global methylation as
evaluated by LINE1 methylation levels (Figure 6D–6F). DISCUSSION
Available LRP1 methylation analyses from the
TCGA cohort (n = 212) [36] confirm the low level of LRP1 has been attributed a role in cancer. Such
LRP1 gene methylation (Figure 6G). In this cohort, no multifunctional endocytic receptor has both endocytic
correlation was found between LRP1 mRNA expression and signaling activities. LRP1 expression levels are often
and LRP1 methylation levels (p = 0.08). dysregulated in cancer, while LRP1 role varies from one
Thus, LRP1 methylation does not seem to be tumor type to another. In CRC, the role and impact of
involved in the regulation of LRP1 gene expression. LRP1 expression remained however unknown so far.

www.impactjournals.com/oncotarget 8858 Oncotarget

 In this report, we made several important and and stromal cells. Second, analysis of two independent
previously unrecognized findings regarding the role of patient cohorts revealed that low LRP1 expression
LRP1 in colon cancer. First, qRT-PCR, LCM and IHC correlated with right tumor location and specific molecular
analyses revealed that LRP1 was significantly lower profile. Third, low LRP1 IHC score in tumor cells was
expressed in adenocarcinoma cells than in normal mucosa associated with poor OS in non-metastatic colon cancer

Figure 4: Survival analysis in colon cancer patients from our cohort compared with LRP1 immunohistochemical
expression in tumor cells. Kaplan-Meier curves of overall survival and event or progression free-survival probability for low (red
line) and high (blue line) LRP1 immunohistochemical (IHC) score in adenocarcinoma cells whatever the tumor stage (A, B), in stage IV
(metastatic) patients (C, D) and in stage IV patients treated with bevacizumab (E, F). IHC score were evaluated by multiplying staining
intensity (0 to 3) and percentage of positive malignant cells (0 to 4) obtained with anti-LRP1 clone 8G1 immunolabelling. Median IHC
score was used to separate low (score 0 to 4) and high (score 6 to 12) LRP1 IHC score. All p values were calculated using the log rank test.

www.impactjournals.com/oncotarget 8859 Oncotarget

 Table 5: Univariate and multivariate analyses of factors associated with overall and progression survival in 76 UICC
stage IV patients
Overall survival Progression free survival
Variables Univariate Multivariate Univariate Multivariate
p value HR 95%CI p value p value HR 95%CI p value
Age 0.05 1.03 1.00–1.05 0.04 0.52 N.A
Tumor location (right vs. left) 0.01 n.s 0.01 n.s
Occlusion (yes vs. no) 0.003 1.96 1.40–3.69 0.04 0.006 n.s
Tumor perforation (yes vs. no) 0.10 2.23 1.02–4.86 0.04 0.07 n.s
Differentiation grade (3 vs. 1-2) 0.0002 2.18 1.07–4.17 0.03 <0.0001 3.45 1.73–6.88 0.0004
LRP1 IHC tumor score (high vs. low) 0.004 0.55 0.32–0.96 0.03 0.05 n.s
LRP1 IHC stroma score (high vs. low) 0.11 N.A 0.03 2.58 1.35–4.94 0.004
LRP1 mRNA 0.12 N.A 0.22 N.A
n.s: not significant; N.A: not adopted; HR: hazard ratio. Results were adjusted on T and N.

patients, and was an important prognostic and predictive occurred in 6% of the TCGA cohort cases. These LRP1-
factor in metastatic patients. Finally, we found that LRP1 mutated cases shared the same clinical and molecular
expression could be partly regulated by LRP1 mutation. profile as those with low LRP1 IHC score in tumor cells
In our cohort, LRP1 was lower expressed, both at and low LRP1 mRNA expression: right location, MSI-H
mRNA and protein levels, in malignant cells compared and CIMP-H. Right colonic cancers with this molecular
with colonic mucosa and stromal cells. In colon mucosa, pattern correspond to the hypermutated type of the
we observed that LRP1 expression seems to be restricted TCGA molecular type of CRC [36]. Hypermutated CRC
to surface epithelium, which is the most specialised part of had a higher mutation rate than non-hypermutated CRC
the epithelium. However, the surface epithelium is about [36], this being mainly due to mismatch repair system
to undergo apoptosis. Thus, staining of these cells should deficiency related to MLH1 methylation. Thus, in this
be interpreted with caution. Some cells of the lamina molecular subgroup of CRC, loss of LRP1 expression
propria, especially myofibroblasts, expressed LRP1 as can partly be explained by LRP1 gene mutation. BRAF
previously described [30, 31]. In adenocarcinomas, IHC mutation is found in around 80–90 % of sporadic MSI-H
and LCM analyses highlighted the differential expression colorectal cancers [36]. Thus, low LRP1 mRNA might be
pattern of LRP1 between tumor and stromal cells. Such a correlated to BRAF mutation through the hypermutator
loss of LRP1 expression in tumor cells as well as its strong type of CRC. Furthermore, hypermutated CRC are also
expression in stromal fibroblast were previously described known as displaying frequent gene hypermethylation. Due
in small cohorts of CRC [30, 31] and in other types of to the abundance of CpG islands in LRP1 gene promoter
cancer such as pancreatic ductal adenocarcinoma [39] and and frequent hypermethylation of LRP1B, another member
lung adenocarcinoma [21]. These previous studies showed of the LRP family, in various cancer types [40, 41], it
that this differential expression between tumor and stromal could be possible that LRP1 gene methylation might
cells seems to play a role in tumor aggressiveness. In regulate its expression. However, in our cohort as well as
pancreatic carcinoma, high stromal expression of LRP1 in the TCGA cohort, the methylation level of both intronic
was correlated with a decreased activation of caspase 3 in and promoter region of LRP1 was very low, suggesting
tumor cells and increased level of SNAIL, a transcription that epigenetic regulation by methylation was not involved
factor promoting epithelial-mesenchymal transition and in the regulation of LRP1 expression.
cell migration [39]. In CRC, high stromal expression of We therefore investigated the role of microRNAs
LRP1 was correlated with high u-PA expression in stromal (miRNAs) regulation on LRP1 expression. Previous
cells [30]. In hepatocellular carcinoma and Wills tumor studies on vascular smooth muscle cells, glioma and lung
cells, the diminished expression of LRP1 in tumor cells carcinoma cells showed that expression of LRP1 was
correlated with increased levels of MMP9, probably due negatively regulated by miR-205 [42, 43]. This reduced
to loss of LRP1-mediated endocytosis [20, 23]. Thus, expression of LRP1 by miR-205 led to decreased tumor
the differential expression of LRP1 between tumor and cell migration [42]. In colon cancer, miR-205 expression
stromal cells might confer survival and spreading benefits findings are conflicting. In one study [44], miR-205 was
for tumor cell in some tumor types including CRC. higher expressed in colon cancer than in paired normal
The loss of LRP1 expression in tumor cells is colon. Another study found inverse results [45]. These
partly explained by mutations in LRP1 gene. Indeed, studies also found conflicting results regarding the role
we observed a loss of LRP1 IHC expression in 21% of of miR-205 in regulation of cell proliferation [44, 45].
the cases in our cohort, while LRP1 gene mutation only Moreover, contrary to previous studies, we found that

www.impactjournals.com/oncotarget 8860 Oncotarget

 miR-205 tended to regulate LRP1 expression positively. regulation by other miRNA, by transcriptional factors or
However, this correlation is weak. Thus, miR-205 could by histone methylation or acethylation. Some microRNA
regulate LRP1 expression in colon cancer but its precise might directly target effectors of the epigenetic machinery
role needs to be furher clarified. (such as DNA methyltransferases, histone deacetylases,
LRP1 IHC score and LRP1 mRNA expression and polycomb repressive complex genes) and indirectly
after LCM in malignant cells were correlated. Thus, pre- affect the expression of tumor suppressor genes [46].
transcriptional processes seem to be involved in LRP1 In colon cancer, the low expression of LRP1 in
down-regulation. Indeed, in our study, LRP1 mutation that tumor cells was strongly associated with right tumor
is found in 6% of the cases might partly explain the loss location, poor differentiation, BRAF mutation, MSI-H and
of LRP1 expression observed in 21% of the cases. LRP1 CIMP-H status in our cohort as well as in an independent
methylation and miR-205 do not seem to be involved in CRC cohort. These molecular findings correspond to the
LRP1 expression down-regulation. Thus, other epigentic hypermutated subtype of the TCGA classification [36]
pre-transcriptional processes might be involved such as and the MSI-immune subtype according to the Consensus

Figure 5: Event-free survival analyses in an independant cohort. Publicly available SieberSmith gene expression dataset was
obtained from R2 microarray analysis and visualization platform (http://r2.amc.nl), and used for survival analyses. Event-free survival
Kaplan-Meier curves for LRP1 mRNA expression in all stages (A), in stage II (B) and in stage III patients (C). (D) Progression-free
survival Kaplan-Meier curve for LRP1 mRNA expression in stage IV patients. All p values were calculated using the log rank test and
computed using R2 online tools.

www.impactjournals.com/oncotarget 8861 Oncotarget

 Figure 6: Analysis of LRP1 expression regulation by LRP1 gene mutation or methylation. (A) Somatic mutation data from
the complete length of LRP1 gene obtained from colorectal cancer of the TCGA cohort using cBioportal for Cancer Genomics (http://
cbioportal.org) web resources. Colored boxes present on the LRP1 gene representation correspond to exons encoding functional domains
of LRP1. Green domain, low-density lipoprotein receptor domains; blue, low-density lipoprotein receptor repeats; yellow, coagulation
factor Xa inhibitory site; orange, domain of unknown function; red, calcium-binding EGF domain; violet, complement Clr-like EGF-like.
(B) Graphical representation of association of LRP1 mutational status with clinical and molecular tracks and LRP1 mRNA expression.

www.impactjournals.com/oncotarget 8862 Oncotarget

 (C) LRP1 mRNA expression levels comparison between LRP1 mutated and LRP1 wild type colorectal cancer. ***p = 0.003, Mann Whitney
test. Linear regression analyses between LRP1 mRNA expression levels and promoter methylation (D), intronic methylation (E), global
DNA methylation levels approximated by LINE1 (F) in our cohort. (G) Correlation of LRP1 mRNA expression levels and LRP1 promoter
methylation in data extracted from the TCGA.

Molecular Subtype consortium [32]. These subtypes were in tumor cells had shorter OS, even when treated with
found in several studies to be associated with serrated bevacizumab. However, our results in metastatic patients
pathway and to have a poor prognosis, particularly after are limited by the small number of patients treated with
relapse [32–35]. In our cohort, low LRP1 IHC score in bevacizumab in our cohort (n = 37) and the lack of NRAS
tumor cells was associated with poor OS particularly in status data. Nevertheless, in view of our promising results,
metastatic (stage IV) patients. Inverse results were found we believe the potential role of LRP1 IHC for predicting
in the SieberSmith cohort, in which LRP1 expression bevacizumab benefit in metastatic CRC patient needs to be
was assessed by qRT-PCR. In this cohort, low LRP1 studied in larger and prospective cohorts.
mRNA expression was related to better EFS. However, The prognosis impact of low LRP1 IHC expression in
mRNA expression reflects combined stromal and tumor malignant cells from stage IV patients may only be partly
cells expression. Conversely, our LCM analyses showed explained by its association with microsatellite instability.
first that LRP1 was overexpressed in stromal cells Indeed, stage II-III MSI CRC had a better prognosis than
when compared with tumor cells and second that LRP1 stage II-III MSS CRC. However, stage IV MSI CRC are
mRNA expression in tumor cells obtained by LCM were associated with poor prognosis and chemoresistance,
correlated to LRP1 IHC score on tumor cells. Thus, LRP1 especially to 5FU-based chemotherapy [51]. Thus, the
mRNA expression levels on whole tumor samples is more association of low LRP1 expression with MSI might explain
likely to reflect stromal cell expression rather than being the pejorative prognosis impact of low LRP1 expression on
representative of tumor cell expression. Moreover, in metastatic patients only. Moreover, a recent study had shown
our cohort, high stromal LRP1 IHC expression in stage that stage IV CRC that were non-responders to bevacizumab
IV patients was associated with poor PFS. Thus, the therapy had a higher level of MMP12 expression than
results of the SieberSmith cohort might more reflect the responders [52]. This increase in MMP12 expression may be
prognosis impact of LRP1 expression in stromal cells. So, favored by the decrease of LRP1 expression. However, this
we think that LRP1 mRNA expression results obtained hypothesis remains to be demonstrated.
from the SieberSmith cohort should not be completely In summary, our study show that low LRP1 IHC
superimposed with our IHC findings. In addition, the expression in malignant colon adenocarcinoma cells is
IHC score on tumor cells can be easily and routinely a strong prognosis predicator, especially in metastatic
performed on formalin-fixed and paraffin-embedded CRC patients, in which it predicts a shorter OS in patients
tissue, while mRNA analyses requires high quality fresh treated by anti-VEGF therapies. The lower expression of
frozen tissue. Thus, from a practical point of view, LRP1 LRP1 in malignant cells is partly explained by LRP1 gene
IHC score assessed in malignant cells seems to be more mutation through the hypermutator type of CRC.
informative for clinical outcome rather than global mRNA
expression. Other studies are needed to clarify our results MATERIALS AND METHODS
regarding LRP1 IHC expression in malignant and stromal
cells. Patients
To date, the biologic agents that have been proven
as having clinical benefits in metastatic CRC mainly The study was conducted on adult patients who
target VEGF and EGFR. In particular, bevacizumab underwent surgery for sporadic colon cancer in the Digestive
targeting VEGF and cetuximab or panitumumab targeting Surgery Department of the Academic Hospital of Reims
EGFR have demonstrated significant survival benefits between September 2006 and December 2012. Patients
in combination with cytotoxic chemotherapy in first- with rectal cancer were excluded. All patients had given
line, second-line, or salvage setting. However, recent their consent for biospecimen use. The study was performed
retrospective analyses have shown that KRAS or NRAS in accordance with the ethical standards laid down in
mutations were negative predictive markers for anti-EGFR the Declaration of Helsinki. Written patients’ consent for
therapy [47]. The mechanisms of action of anti-VEGF are biospecimen use was obtained in all cases. Approval for the
not completely understood, and apart from right tumor study was previously obtained from the local Institutional
location, no predictive factor has yet been validated [48, Review Board and the Tissue Bank Management Board.
49]. The role of KRAS or NRAS mutation for bevacizumab Study design was published on clinicaltrials.gov web site in
therapy efficiency prediction has not been defined yet May 2016 (#NCT02788669).
[50]. In our study, low LRP1 IHC score in tumor cells was Clinical data including age at the time of surgery,
an indicator of poor OS and PFS in metastatic patients. sex, performance status, surgical circumstances (tumor
Indeed, stage IV patients with low LRP1 IHC score perforation, occlusion), tumor location, synchronous or

www.impactjournals.com/oncotarget 8863 Oncotarget

 metachronous metastases, tumor recurrence, treatment, flexure, descending colon, sigmoid colon or rectosigmoid
death and pathological and molecular data including junction. Mismatched repair (MMR) status of tumors was
adenocarcinoma type, grade and pTNM stage were performed by immunohistochemistry with anti-MLH1,
collected. Patients were classified as having a right colonic PMS2, MSH2 and MSH6 proteins on tissue microarrays,
cancer if the primary tumor was located in the caecum, completed when necessary by microsatellite instability
ascending colon, hepatic flexure or transverse colon, and analysis, as already reported [53]. Mutations within exon
left colonic cancer if the tumor site was within the splenic 2 of KRAS and of the codon 600 of BRAF were detected as

Figure 7: Comparison of miR-205 and miR-338-5p expression with LRP1 expression. Analyses of miR-205 (A) and miR-
338-5p (B) expression by qRT-PCR in fresh frozen colon cancer adenocarcinoma compared with normal colon mucosa from our cohort (*p
< 0.05, ***p < 0.001, Mann Whitney test). Linear regression analysis of LRP1 mRNA expression levels evaluated by qRT-PCR on complete
fresh frozen adenocarcinoma sample against miR-205 (C) and miR-338-5p (E) expression. Linear regression analysis of miR-205 (D)
and miR-338-5p (F) expression against LRP1 immunohistochemical (IHC) score of tumor cells. IHC score was assessed by multiplying
staining intensity (0 to 3) and percentage of positive tumor cells (0 to 4) with anti-LRP1 clone 8G1 immunolabelling.

www.impactjournals.com/oncotarget 8864 Oncotarget

 previously described [53]. Follow-up data were obtained all immunohistochemistry, the counterstain used was
from oncologist or attending physicians. hematoxylin. Staining was rated binarly as either positive
or negative for these 2 markers by the same pathologists.
Pathology All tumors in which the tumor cells completely lacked
immunostaining were scored as negative. Cases were rated
All colon adenocarcinomas were classified and as positive when the tumor cells were unequivocally stained
subtyped according to The World Health Organization in the nucleus.
criteria [1] and staged according to the International
Union Against Cancer 2009 guidelines [4]. All slides LRP1 mRNA analyses
were retrieved from the archives of the Department of
Pathology of the Academic Hospital of Reims and were mRNA analyses were performed on fresh frozen
reviewed and classified by two pathologists (CBR and colon adenocarcinoma and normal colon tissues sampled on
MDD). Tumor budding was assessed on Hematoxylin- colectomies received at the Pathology Department of Reims
Eosin-Saffron slides as previously described [54]. University Hospital (France) and stored in the Champagne-
Ardenne Biobank as previously described [57]. Total RNAs
Immunohistochemistry were isolated and purified with Maxwell® 16 LEV simply
RNA tissue kit (Promega, Madison, USA) according to
All tissue samples were analyzed via tissue the manufacturer’s instructions on the Promega’s robotics
microarrays. For each tumor, 3 cores were punched in the platform Maxwell® 16 Research Instrument (Promega,
central part and 3 cores at the invasive front of the tumor Madison, USA). The concentration of total RNA (ng/μL)
from the same original formalin-fixed paraffin-embedded was determined by a Picodrop uL spectrophotometer
tumor block. The cores were precisely arrayed into a (Picodrop, Hinxton, United Kingdom).
recipient paraffin block using the MiniCore Tissue Arrayer RNA quality index (RQI) was determined using
(Excilone, Elancourt, France). Sections of 4-μm thickness the Experion™ automated electrophoresis system (Bio-
were cut and mounted on SuperFrost Plus Gold adhesive Rad, Marnes-la-Coquette, France) according to the
slides (Thermofisher Scientific, Waltham, MA, USA). manufacturer protocol. Only RNA with RQI values ≥5
Immunohistochemistry using anti-LRP-1 α-chain (1/1000, were used for futher analyses.
mouse, clone 8G1, Merck, Darmstadt, Germany) and RNA were reverse-transcribed using VERSO
control isotype mouse IgGs (Agilent Technologies, Santa cDNA kit (Thermo Fisher Scientific, Waltham, MA,
Clara, CA, USA) was performed using Novolink Polymer USA) according to the manufacturer’s instructions using
Detection System (Leica Biosystems, Wetzlar, Germany) random hexamer primers. Real-time PCR was performed
after heat-induced epitope retrieval in citrate pH 6 buffer using an Absolute SYBR Green Rox mix (Thermo Fisher
(95° C, 40 min) and overnight antibody incubation at 4° C. Scientific), on a CFX 96 real time PCR detection system
Staining intensity (SI) was graded by two (Bio-Rad). RS18 and RPL32 were used for LRP1 expression
pathologists (CBR, AMB) as 0 (negative), 1 (weak), 2 normalization. The sequences of the pairs of primers used
(moderate) and 3 (strong). The percentage of positive were: LRP1 (5′–AGA AGT AGC AGG ACC AGA GGG –
cells (PPC), was graded as follows: 0 (<5%), 1 (5–25%), 3′ and 3′–TCA GTA CCC AGG CAG TTA TGC - 5′),
2 (26–50%), 3 (51–75%) and 4 (76–100%). In case of CEA (5′–TTT CTC CCT ATG TGG TCG CTC CAG - 3′
discrepancies a consensus diagnosis was reached. Then, and 3′–AGC AGA TTT TTA TTG AAC TTG TGC _- 5′),
an immunostaining score was generated independently for RS18 (5′- GCA GAA TCC ACG CCA GTA CAA -3′ and
malignant and stromal cells of each case by multiplying 3′–GCC AGT GGT CTT GGT GTG CT– 5′) and RPL32
SI and PPC. The median score was used to distinguish (5′–CAT TGG TTA TGG AAG CAA CAA A- 3′ and 3′–
low (0–4) and high (6–12) LRP1 expression levels for TTC TTG GAG GAA ACA TTG TGA G-5′). All primers
adenocarcinomatous cells. were synthesized by Eurogentec (Eurogentec, Liège,
Additionnaly, imunohistochemistry for the intestinal Belgium). PCR conditions were set as 15 min at 95° C,
differentiation marker CDX2 [55] (RTU, rabbit monoclonal, followed by 40 cycles each consisting of 15 s at 95° C
clone EPR2764Y, Zytomed System, Berlin, Germany) and (denaturation) and 1 min at 60° C (annealing/extension).
the marker of serrated subtype of adenocarcinoma Annexin The specificity of PCR amplification was checked using a
A10 [56] (1/400, rabbit polyclonal, Novus Biologicals, heat dissociation curve from 65° C to 95° C following the
Littleton, CO, USA) were performed with the BenchMark final cycle. The cycle threshold (Ct) values were recorded
XT automated slide stainer (Ventana Medical Systems, with Bio-Rad CFX Manager™ 3.0 software (Bio-Rad).
Tucson, AZ, USA). Antibody retrieval was performed
with Cell Conditioner 1 (EDTA, pH 8.4) incubation for Laser capture microdissection
64 minutes, followed by preprimary peroxidase inhibition,
and incubation with the corresponding antibody at 37° C Fresh frozen colon adenocarcinoma specimens
for 32 minutes. UltraView Universal DAB v3 Kit (Ventana were cut into 12 μm serial sections and mounted on
Medical Systems) was used for staining reaction. For PALM membrane slides (Zeiss, Oberkochen, Germany).

www.impactjournals.com/oncotarget 8865 Oncotarget

 The slides were immediately stained with cresyl violet on the LightCycler 480 II High Resolution Melting
from the LCM staining kit (Thermo Fischer Scientific) instrument (Roche, Pleasanton, CA, USA). All primers
and laser capture microdissection (LCM) was performed were synthesized by Eurogentec (Eurogentec). No-
immediately thereafter. Adenocarcinomatous and stromal CIMP status was defined as no methylated locus, CIMP-
areas were selected during the LCM procedure by a Low status as one to three methylated loci, and CIMP-
pathologist (CBR). Laser capture microdissection was High status as four or five methylated loci as previously
performed with the PALM MicroBeam instrument (Zeiss). described [58].
At least 5 mm2 of tumor tissue or stromal tissue were LINE-1 methylation analyses were performed by
collected from each sample. This required from nine to pyrosequencing analysis using the Pyromark Q96MA
twelve 12 μm sections. instrument (Qiagen) as previously described [59]. The
RNA from tumor and stromal microdissected average LINE-1 methylation level was calculated as the mean
tissues were isolated and purified with the RNeasy of the proportions of C (%) at the 3 CpG sites analyzed and
micro kit (Qiagen GmbH, Hilden, Germany) according this indicated the level of methylation of LINE-1 elements.
to the manufacturer’s instructions. RNA concentrations LRP1 methylation analyses were performed by
were measured using NanoDrop system (Thermo Fisher pyrosequencing using the Pyromark Q96MA instrument
Scientific). RT-PCR analyses were performed as detailed (Qiagen). Promoter and intronic region of LRP1 were
above. amplified from bisulfited DNA using commercialy
available primers (Hs_LRP1_01_PM for intronic region,
miRNA analyses Hs_LRP1_02_PM for promoter region, Qiagen) for PCR
amplification and pyrosequencing. PCR amplifications
miRNA were extracted from fresh frozen colon were performed using PyroMark PCR kit (Qiagen)
adenocarcinoma and normal colon tissues using according to manufacturer’s instructions. Methylated and
miRNeasy mini kit (Qiagen) according to manufacturer’s unmethylated converted and unmethylated unconverted
instructions. RNA concentrations were measured using a controls from the EpiTect PCR Control DNA Set (Qiagen)
NanoDrop spectrophotometer (Thermo Fisher Scientific). were used for each experiment. Each experiment was
cDNA was synthesized using miScript II RT Kit (Qiagen) performed in duplicate.
in accordance with the manufacturer’s instructions.
Expression of miR-205 and miR-338-5p was Data mining and bioinformatic analyses
determined by real time PCR using an Absolute SYBR
Green Rox mix (Thermo Fisher Scientific), on a CFX 96 Mutation and expression data from the colorectal
real time PCR detection system (Bio-Rad) and normalized carcinoma dataset of the The Cancer Genome Atlas
using U6 small nuclear RNA. miR-205 is known to down- (TCGA; https://tcga-data.nci.nih.gov) [36] were analyzed
regulate LRP1 expression [42, 43]. miR-338-5p is not using cBioportal for Cancer Genomics (http://cbioportal.
implicated in LRP1 expression regulation and was used org) web resources [37, 38].
as control as previously described [43]. All primers were Publicly available SieberSmith gene expression
purchased as 10x miScript Primer Assay (Qiagen). PCR dataset was obtained from R2 microarray analysis and
conditions were 15 min at 95° C, followed by 40 cycles visualization platform (http://r2.amc.nl), and used for
each consisting of 15 s at 95° C (denaturation), 30 s at survival analyses. Cut-off value for separating high and
55° C (annealing) and 30 s at 70° C (extension). The low LRP1 expression groups was determined by the online
specificity of PCR amplification was checked using a heat algorithm.
dissociation curve from 65° C to 95° C following the final
cycle. The cycle threshold (Ct) values were recorded with Statistical and survival analyses
Bio-Rad CFX Manager™ 3.0 software (Bio-Rad).
Data are here described using mean and standard
Methylation analyses deviation for quantitative variables and number and
percentage for qualitative variables. Factors associated with
All methylation analyses were performed on DNA mRNA and immunohistochemical expression of LRP1 were
extracted from fresh frozen tissues with the QIAamp studied using univariate analysis (Chi2 test, Fisher’s exact
DNA microkit (Qiagen) according to manufacturer’s test, Student’s t test, linear regression or Wilcoxon test, as
instructions. Bisulfite conversion was performed with the appropriate) and multivariate analysis (linear regression
EZ DNA Methylation gold kit (Zymo Research, Irvine, with stepwise selection, with an exit threshold of 0.10
CA, USA) following manufacturer’s instructions. and factors significant at p = 0.10 included). Overall and
The CpG island methylator phenotype (No event-free survivals were studied. The survival curves were
CIMP, CIMP-Low and CIMP-High) was determined by established by the Kaplan-Meier method. For each analysis,
Methylation Sensitive High Resolution Melting for 5 prognostic factors were identified by univariate analysis
markers (MLH1, CDKN2A, MINT1, MINT2, and MINT31) using log rank tests and by multivariate analysis using a

www.impactjournals.com/oncotarget 8866 Oncotarget

 Cox proportional hazard model. Factors significant at the 3. Centre National de Recherche Scientifique. All these
0.10 level in univariate analysis were included in a stepwise study sponsors have no roles in the study design, in the
regression multivariate analysis with entry and removal collection, analysis, and interpretation of data.
limits set at 0.10. Statistical analyses were performed with
SAS version 9.4 (SAS institute Inc, Cary, North California). REFERENCES
For all tests, p < 0.05 were considered to be statistically
significant. 1. Hamilton SR, Bosman FT, Boffetta P, Ilyas M, Morreau H,
Nakamura SI, Quirke P, Riboli E, Sobin LH. Carcinoma of
Abbreviations the colon and rectum. WHO classification of tumours of the
digestive system. Lyon, France: IARC Press. 2010: 132–46.
CIMP-H: High CpG Island Methylator phenotype; 2. Stewart BW, Wild CP. World cancer report 2014. Lyon,
CRC: Colorectal Carcinoma; EGFR: Epidermal Growth France: International Agency for Research on Cancer,
Factor Receptor; IHC: Immunohistochemistry; LCM: World Health Organization; 2014.
Laser Capture Microdissection; MSI-H: High Microsatellite 3. Ronnekleiv-Kelly SM, Burkhart RA, Pawlik TM. Molecular
instability; MSS: Microsatellite stable; TCGA: The Cancer markers of prognosis and therapeutic targets in metastatic
Genome Atlas; UICC: Union Internationale Contre le colorectal cancer. Surg Oncol. 2016; 25:190–9.
Cancer; VEGF: Vascular Endothelial Growth Factor. 4. Sobin L, Gospodarowicz M, Wittekind C. Colon and
Rectum. TNM Classification of Malignant Tumours. 7th ed.
Author contributions Chicheter (West Sussex): Blackwell Publishing Ltd. 2010:
100–105.
Conception or design of the work: CBR, CS, OB, 5. Lech G, Słotwiński R, Słodkowski M, Krasnodębski IW.
MDD, GB, SD; Data collection: CBR, CL, SN, CC, CF, Colorectal cancer tumour markers and biomarkers: Recent
NB; Data analysis and interpretation: CBR, CS, AJ, VG, therapeutic advances. World J Gastroenterol. 2016; 22:
AMB, CB, BM, AB, ASL; Drafting the article: CBR, CS, 1745–55.
AJ, VG, CB; Critical revision of the article: CS, AJ, AMB,
6. Gonzalez-Pons M, Cruz-Correa M. Colorectal Cancer
MDD, SD; Final approval of the version to be published: Biomarkers: Where Are We Now? Biomed Res Int. 2015;
AMB, MDD, RK, ASL, OB, GB, SD. 2015: 149014.
7. Hutchins G, Southward K, Handley K, Magill L,
ACKNOWLEDGMENTS Beaumont C, Stahlschmidt J, Richman S, Chambers P,
Seymour M, Kerr D, Gray R, Quirke P. Value of mismatch
We thank Mrs Arlette Thomachot for her valuable repair, KRAS, and BRAF mutations in predicting recurrence
editing assistance. and benefits from chemotherapy in colorectal cancer. J Clin
We thank Doctors Benjamin Tournier and Caroline Oncol. 2011; 29:1261–70.
Chapusot from the Pathology Department of the 8. Popat S, Hubner R, Houlston RS. Systematic review of
University Hospital of Dijon, France for their help in microsatellite instability and colorectal cancer prognosis. J
methylation analyses. We thank the Image Core Facility Clin Oncol. 2005;23:609–18.
(PICT) of the University of Reims Champagne-Ardenne 9. Bertagnolli MM, Redston M, Compton CC, Niedzwiecki D,
and the Regional Platform of Innovative Biology (PRBI) Mayer RJ, Goldberg RM, Colacchio TA, Saltz LB,
for their excellent technical assistance. Warren RS. Microsatellite instability and loss of
heterozygosity at chromosomal location 18q: prospective
CONFLICTS OF INTEREST evaluation of biomarkers for stages II and III colon
cancer—a study of CALGB 9581 and 89803. J Clin Oncol.
The authors declare that they have no further 2011; 29:3153–62.
financial or other conflicts of interest in relation to this 10. Gray RG, Quirke P, Handley K, Lopatin M, Magill L,
research and its publication. Baehner FL, Beaumont C, Clark-Langone KM,
Yoshizawa CN, Lee M, Watson D, Shak S, Kerr DJ.
FUNDING Validation study of a quantitative multigene reverse
transcriptase-polymerase chain reaction assay for
This study was funded by grants from: 1. Ligue assessment of recurrence risk in patients with stage II colon
Nationale Contre le Cancer Conférence de Coordination cancer. J Clin Oncol. 2011; 29:4611–9.
Interrégionale du Grand Est, Appel d’Offre 2014 et 11. Duffy MJ, Lamerz R, Haglund C, Nicolini A, Kalousova M,
2015. Grant Numbers 61/2014/KM/ML/Recherche and Holubec L, Sturgeon C. Tumor markers in colorectal
2015/KM/NM/PFRS/232 (SD). 2. Centre Hospitalier cancer, gastric cancer and gastrointestinal stromal cancers:
Universitaire de Reims, Appel d’Offre Local 2014 et European group on tumor markers (EGTM) 2014 guidelines
2015. Grant Numbers AU14-03 and AOL 2015-11 (CBR). update. Int J Cancer. 2014;134:2513–22.

www.impactjournals.com/oncotarget 8867 Oncotarget

 12. Lillis AP, Van Duyn LB, Murphy-Ullrich JE, Strickland DK. 25. Catasús L, Gallardo A, Llorente-Cortés V, Escuin D,
LDL receptor-related protein 1: unique tissue-specific Muñoz J, Tibau A, Peiro G, Barnadas A, Lerma E. Low-
functions revealed by selective gene knockout studies. density lipoprotein receptor-related protein 1 is associated
Physiol Rev. 2008; 88:887–918. with proliferation and invasiveness in Her-2/neu and triple-
13. Herz J, Strickland DK. LRP: a multifunctional scavenger negative breast carcinomas. Hum Pathol. 2011; 42:1581–88.
and signaling receptor. J Clin Invest. 2001;108:779–84. 26. McGarvey T, Hussain MM, Stearns ME. In situ
14. Strickland DK, Ashcom JD, Williams S, Burgess WH, hybridization studies of alpha 2-macroglobulin receptor and
Migliorini M, Argraves WS. Sequence identity between receptor-associated protein in human prostate carcinoma.
the alpha 2-macroglobulin receptor and low density Prostate. 1996; 28:311–17.
lipoprotein receptor-related protein suggests that this 27. Dedieu S, Langlois B, Devy J, Sid B, Henriet P, Sartelet H,
molecule is a multifunc- tional receptor. J Biol Chem. 1990; Bellon G, Emonard H, Martiny L. LRP-1 silencing prevents
265:17401–4. malignant cell invasion despite increased pericellular
15. Van Gool B, Dedieu S, Emonard H, Roebroek AJ. The proteolytic activities. Mol Cell Biol. 2008; 28:2980–95.
Matricellular Receptor LRP1 Forms an Interface for 28. Song H, Li Y, Lee J, Schwartz AL, Bu G. Low-density
Signaling and Endocytosis in Modulation of the Extracellular lipoprotein receptor-related protein 1 promotes cancer cell
Tumor Environment. Front Pharmacol. 2015; 6:271. migration and invasion by inducing the expression of matrix
16. Gonias SL, Campana WM. LDL receptor-related protein-1: metalloproteinases 2 and 9. Cancer Res. 2009; 69:879–86.
a regulator of inflammation in atherosclerosis, cancer, 29. Li Y, Reynolds RC. LRP1: A Tumor and Metastasis Promoter
and injury to the nervous system. Am J Pathol. 2014; or Suppressor? Biochem Pharmacol. 2012; 1:e121.
184:18–27.
30. Obermeyer K, Krueger S, Peters B, Falkenberg B,
17. Xing P, Liao Z, Ren Z, Zhao J, Song F, Wang G, Chen K,
Roessner A, Röcken C. The expression of low density
Yang J. Roles of low-density lipoprotein receptor-related
lipoprotein receptor-related protein in colorectal carcinoma.
protein 1 in tumors. Chin J Cancer. 2016; 35:6.
Oncol Rep. 2007; 17:361–7.
18. Langlois B, Emonard H, Martiny L, Dedieu S. [Multiple
31. Toquet C, Jarry A, Bou-Hanna C, Bach K, Denis MG,
involvements of LRP-1 receptor in tumor progression].
Mosnier JF, Laboisse CL. Altered Calreticulin expression in
[Article in French]. Pathol Biol (Paris). 2009; 57:548–54.
human colon cancer: maintenance of Calreticulin expression
19. Martinez-Ledesma E, Verhaak RG, Treviño V. Identification is associated with mucinous differentiation. Oncol Rep. 2007;
of a multi-cancer gene expression biomarker for cancer
17:1101–7.
clinical outcomes using a network-based algorithm. Sci
32. Guinney J, Dienstmann R, Wang X, de Reynies A,
Rep. 2015; 5:11966.
Schlicker A, Soneson C, Marisa L, Roepman P,
20. Huang XY, Shi GM, Devbhandari RP, Ke AW, Wang Y,
Nyamundanda G, Angelino P, Bot BM, Morris JS, Simon
Wang XY, Wang Z, Shi YH, Xiao YS, Ding ZB, Dai Z,
IM, et al. The consensus molecular subtypes of colorectal
Xu Y, Jia WP, et al. Low level of low-density lipoprotein
cancer. Nat Med. 2015;21:1350–6.
receptor-related protein 1 predicts an unfavorable prognosis
33. Bae JM, Kim JH, Kang GH. Molecular subtypes of
of hepatocellular carcinoma after curative resection. PLoS
colorectal cancer and their clinicopathologic features, with
One. 2012;7:e32775.
an emphasis on the serrated neoplasia pathway. Arch Pathol
21. Meng H, Chen G, Zhang X, Wang Z, Thomas DG,
Lab Med. 2016; 140:406–12.
Giordano TJ, Beer DG, Wang MM. Stromal LRP1 in lung
adenocarcinoma predicts clinical outcome. Clin Cancer Res. 34. Phipps AI, Limburg PJ, Baron JA, Burnett-Hartman AN,
2011; 17:2426–33. Weisenberger DJ, Laird PW, Sinicrope FA, Rosty C,
22. de Vries TJ, Verheijen JH, de Bart AC, Weidle UH, Buchanan DD, Potter JD, Newcomb PA. Association
Ruiter DJ, van Muijen GN. Decreased expression of between molecular subtypes of colorectal cancer and patient
both the low-density lipoprotein receptor-related protein/ survival. Gastroenterology. 2015; 148:77–87.
alpha(2)-macroglobulin receptor and its receptor-associated 35. Müller MF, Ibrahim AE, Arends MJ. Molecular pathological
protein in late stages of cutaneous melanocytic tumor classification of colorectal cancer. Virchows Arch. 2016;
progression. Cancer Res. 1996; 56:1432–9. 469:125–34.
23. Desrosiers RR, Rivard ME, Grundy PE, Annabi B. Decrease 36. Muzny DM, Bainbridge MN, Chang K, Dinh HH,
in LDL receptor-related protein expression and function Drummond JA, Fowler G, Kovar CL, Lewis LR,
correlates with advanced stages of Wilms tumors. Pediatr Morgan MB, Newsham IF, Reid JG, Santibanez J,
Blood Cancer. 2006; 46:40–49. Shinbrot E, et al, and Cancer Genome Atlas Network.
24. Catasús L, Llorente-Cortés V, Cuatrecasas M, Pons C, Comprehensive molecular characterization of human colon
Espinosa I, Prat J. Low-density lipoprotein receptor- and rectal cancer. Nature. 2012; 487:330–37.
related protein 1 (LRP-1) is associated with highgrade, 37. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B,
advanced stage and p53 and p16 alterations in endometrial Sumer SO, Sun Y, Jacobsen A, Sinha R, Larsson E,
carcinomas. Histopathology. 2011; 59:567–71. Cerami E, Sander C, Schultz N. Integrative analysis of

www.impactjournals.com/oncotarget 8868 Oncotarget

 complex cancer genomics and clinical profiles using the Predictive Factor for First-line Bevacizumab Effectiveness
cBioPortal. Sci Signal. 2013;6:pl1. in Metastatic Colorectal Cancer Patients. J Cancer. 2017;
38. Cerami E, Gao J, Dogrusoz U, Gross BE, Sumer SO, Aksoy BA, 8:388–394.
Jacobsen A, Byrne CJ, Heuer ML, Larsson E, Antipin Y, Reva B, 50. Zhou M, Yu P, Qu J, Chen Y, Zhou Y, Fu L, Zhang J.
Goldberg AP, et al. The cBio cancer genomics portal: an open Efficacy of Bevacizumab in the First-Line Treatment of
platform for exploring multidimensional cancer genomics data. Patients with RAS Mutations Metastatic Colorectal Cancer:
Cancer Discov. 2012; 2:401–04. a Systematic Review and Network Meta-Analysis. Cell
39. Leca J, Martinez S, Lac S, Nigri J, Secq V, Rubis M, Bressy C, Physiol Biochem. 2016; 40:361–9.
Sergé A, Lavaut MN, Dusetti N, Loncle C, Roques J, 51. Tougeron D, Sueur B, Sefrioui D, Gentilhomme L,
Pietrasz D, et al. Cancer-associated fibroblast-derived Lecomte T, Aparicio T, Guetz G, Artru P, De La
annexin A6+ extracellular vesicles support pancreatic cancer Fouchardiere C, Moulin V, Hautefeuille V, Coriat R,
aggressiveness. J Clin Invest. 2016; 126:4140–56. Touchefeu Y, et al. A large multicenter study evaluating
40. Lu YJ, Wu CS, Li HP, Liu HP, Lu CY, Leu YW, Wang CS, prognosis and chemosensitivity of metastatic colorectal
Chen LC, Lin KH, Chang YS. Aberrant methylation impairs cancers with microsatellite instability. J Clin Oncol.
low density lipoprotein receptor-related protein 1B tumor 2017; 35.
suppressor function in gastric cancer. Genes Chromosomes 52. Zuurbier L, Rahman A, Cordes M, Scheick J, Wong TJ,
Cancer. 2010; 49:412–24. Rustenburg F, Joseph JC, Dynoodt P, Casey R, Drillenburg P,
41. Sonoda I, Imoto I, Inoue J, Shibata T, Shimada Y, Chin K, Gerhards M, Barat A, Klinger R, et al. Apelin: A putative
Imamura M, Amagasa T, Gray JW, Hirohashi S, Inazawa J. novel predictive biomarker for bevacizumab response in
Frequent silencing of low density lipoprotein receptor- colorectal cancer. Oncotarget. 2017; 8:42949–61. https://doi.
related protein 1B (LRP1B) expression by genetic and org/10.18632/oncotarget.17306.
epigenetic mechanisms in esophageal squamous cell 53. Boulagnon C, Dudez O, Beaudoux O, Dalstein V, Kianmanesh
carcinoma. Cancer Res. 2004; 64:3741–7. R, Bouché O, Diebold MD. BRAFV600E Gene Mutation
42. Song H, Bu G. MicroRNA-205 inhibits tumor cell in Colonic Adenocarcinomas. Immunohistochemical
migration through down-regulating the expression of the Detection Using Tissue Microarray and Clinicopathologic
LDL receptor-related protein 1. Biochem Biophys Res Characteristics: An 86 Case Series. Appl Immunohistochem
Commun. 2009; 388:400–5. Mol Morphol. 2016;24:88–96.
43. Chan CY, Chan YC, Cheuk BL, Cheng SW. Clearance 54. Wang LM, Kevans D, Mulcahy H, O’Sullivan J, Fennelly D,
of matrix metalloproteinase-9 is dependent on low- Hyland J, O’Donoghue D, Sheahan K. Tumor budding
density lipoprotein receptor-related protein-1 expression is a strong and reproducible prognostic marker in T3N0
downregulated by microRNA-205 in human abdominal colorectal cancer. Am J Surg Pathol. 2009; 33:134–41.
aortic aneurysm. J Vasc Surg. 2017; 65:509–20.
55. Saad RS, Ghorab Z, Khalifa MA, Xu M. CDX2 as a marker
44. Eyking A, Reis H, Frank M, Gerken G, Schmid KW, for intestinal differentiation: Its utility and limitations.
Cario E. MiR-205 and MiR-373 Are Associated with World J Gastrointest Surg. 2011; 3: 159–66.
Aggressive Human Mucinous Colorectal Cancer. PLoS
56. Sajanti SA, Väyrynen JP, Sirniö P, Klintrup K, Mäkelä J,
One. 2016;11:e0156871.
Tuomisto A, Mäkinen MJ. Annexin A10 is a marker for the
45. Nguyen-Vu T, Wang J, Mesmar F, Mukhopadhyay S, serrated pathway of colorectal carcinoma. Virchows Arch.
Saxena A, McCollum CW, Gustafsson JÅ, Bondesson M, 2015; 466:5–12.
Williams C. Estrogen receptor beta reduces colon cancer
57. Galissier T, Schneider C, Nasri S, Kanagaratnam L, Fichel C,
metastasis through a novel miR-205-PROX1 mechanism.
Coquelet C, Diebold MD, Kianmanesh R, Bellon G,
Oncotarget. 2016; 7:42159–71. https://doi.org/10.18632/
Dedieu S, Marchal Bressenot A, Boulagnon-Rombi C.
oncotarget.9895.
Biobanking of Fresh-Frozen Human Adenocarcinomatous
46. Fabbri M, Calin GA. Epigenetics and miRNAs in human and Normal Colon Tissues: Which Parameters Influence
cancer. Adv Genet. 2010; 70:87–99. RNA Quality? PLoS One. 2016;11:e0154326.
47. Ohhara Y, Fukuda N, Takeuchi S, Honma R, Shimizu Y, 58. Barault L, Charon-Barra C, Jooste V, de la Vega MF,
Kinoshita I, Dosaka-Akita H. Role of targeted therapy in Martin L, Roignot P, Rat P, Bouvier AM, Laurent-Puig P,
metastatic colorectal cancer. World J Gastrointest Oncol. Faivre J, Chapusot C, Piard F. Hypermethylator phenotype
2016; 8:642–55. in sporadic colon cancer: study on a population-based series
48. Stintzing S, Stremitzer S, Sebio A, Lenz HJ. Predictive and of 582 cases. Cancer Res. 2008; 68:8541–6.
prognostic markers in the treatment of metastatic colorectal 59. Tournier B, Chapusot C, Courcet E, Martin L, Lepage C, Faivre
cancer (mCRC): personalized medicine at work. Hematol J, Piard F. Why do results conflict regarding the prognostic
Oncol Clin North Am. 2015; 29:43–60. value of the methylation status in colon cancers? The role of
49. He WZ, Liao FX, Jiang C, Kong PF, Yin CX, Yang Q, the preservation method. BMC Cancer. 2012;12:12.
Qiu HJ, Zhang B, Xia LP. Primary Tumor Location as a

www.impactjournals.com/oncotarget 8869 Oncotarget