Radiol Oncol 2023; 57(4): 473-486. doi: 10.2478/raon-2023-0061
473
research article
The association of genetic factors with serum 
calretinin levels in asbestos-related diseases
Cita Zupanc
1,2
, Alenka Franko
2,3
, Danijela Strbac
2,4
, Viljem Kovac
2,4
, Vita Dolzan
5
, 
Katja Goricar
5
1 
Military Medical Unit-Slovenian Army, Ljubljana, Slovenia
2
 University of Ljubljana, Faculty of Medicine, Ljubljana, Slovenia
3
 University Medical Centre Ljubljana, Clinical Institute of Occupational Medicine, Ljubljana, Slovenia
4
 Institute of Oncology Ljubljana, Ljubljana, Slovenia
5
 University of Ljubljana, Faculty of Medicine, Institute of Biochemistry and Molecular Genetics, Pharmacogenetics 
Laboratory, Ljubljana, Slovenia
Radiol Oncol 2023; 57(4): 473-486.
Received 19 October 2023 
Accepted 31 October 2023
Correspondence to: Assist. Prof. Katja Goričar, Ph.D., University of Ljubljana, Faculty of Medicine, Institute of Biochemistry and Molecular 
Genetics, Pharmacogenetics Laboratory, Vrazov trg 2, SI-1000 Ljubljana, Slovenia. E-mail: katja.goricar@mf.uni-lj.si
Disclosure: No potential conflicts of interest were disclosed.
This is an open access article distributed under the terms of the CC-BY license (https://creativecommons.org/licenses/by/4.0/).
Background. Asbestos exposure is associated with different asbestos-related diseases, including malignant meso-
thelioma (MM). MM diagnosis is confirmed with immunohistochemical analysis of several markers, including calretinin. 
Increased circulating calretinin was also observed in MM. The aim of the study was to determine if CALB2 polymor-
phisms or polymorphisms in genes that can regulate calretinin expression are associated with serum calretinin levels 
or MM susceptibility.
Subjects and methods. The study included 288 MM patients and 616 occupationally asbestos-exposed subjects 
without MM (153 with asbestosis, 380 with pleural plaques and 83 without asbestos-related disease). Subjects were 
genotyped for seven polymorphisms in CALB2, E2F2, MIR335, NRF1 and SEPTIN7 genes using competitive allele-specific 
polymerase chain reaction (PCR). Serum calretinin was determined with ELISA in 545 subjects. Nonparametric tests, 
logistic regression and receiver operating characteristic (ROC) curve analysis were used for statistical analysis.
Results. Carriers of at least one polymorphic CALB2 rs889704 allele had lower calretinin levels (P = 0.036). Carriers of 
two polymorphic MIR335 rs3807348 alleles had higher calretinin (P = 0.027), while carriers of at least one polymorphic 
NRF1 rs13241028 allele had lower calretinin levels (P = 0.034) in subjects without MM. Carriers of two polymorphic E2F2 
rs2075995 alleles were less likely to develop MM (odds ratio [OR] = 0.64, 95% confidence interval [CI] = 0.43-0.96, P 
= 0.032), but the association was no longer significant after adjustment for age (P = 0.093). Optimal serum calretinin 
cut-off values differentiating MM patients from other subjects differed according to CALB2, NRF1, E2F2, and MIR335 
genotypes.
Conclusions. The results of presented study suggest that genetic variability could influence serum calretinin levels. 
These findings could contribute to a better understanding of calretinin regulation and potentially to earlier MM diag-
nosis.
Key words: malignant mesothelioma; calretinin; CALB2; asbestos-related disease; polymorphism
Introduction
Prolonged asbestos exposure can lead to occur-
rence of different asbestos-related diseases, in-
cluding pleural plaques and asbestosis, as well 
as several cancers. Use and production of asbes-
tos was largely banned after it was classified as 
a carcinogen, but it is still legally used in mostly 
developing countries and it can still be found in 
the environment.
1,2
 Asbestos-related diseases often 

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 474
occur long after initial asbestos exposure and their 
incidence continues to rise.
1
The most problematic asbestos-related disease 
is malignant mesothelioma (MM), a rare but very 
aggressive cancer. However, only a minority of 
asbestos-exposed people develops MM. Other fac-
tors, such as genetic variability may contribute to 
carcinogenesis and development of MM.
3
 Among 
asbestos-exposed workers, several familial cases 
of MM were described, emphasizing that genetic 
factors could contribute to MM development.
4
 In 
recent years, germline BRCA1-associated protein 
1 (BAP1) mutations were shown to predispose 
to the development of MM and other cancers. 
Additionally, studies suggest that numerous chro-
mosomal deletions can accumulate in most MM 
cases, usually associated with the loss or inactiva-
tion of tumor suppressor genes.
5,6
 Despite advanc-
es in treatment, prognosis and survival of MM 
patients remain poor.
7,8
 Therefore, MM diagnosis 
and treatment have become increasingly focused 
on molecular mechanisms.
9
To confirm MM diagnosis, several tumor mark-
ers are routinely analysed using immunohisto-
chemical staining.
10
 One of the established immu-
nohistochemical markers is calretinin
10
, a calcium 
binding protein and calcium sensor crucial for 
neuron function that is also expressed on meso-
thelial cells.
11
 It has been shown to affect mesothe-
lial cell proliferation and migration and epithelial-
to-mesenchymal transition. It was also associated 
with focal adhesion kinase signaling pathway and 
signaling pathways associated with response to 
asbestos.
12
 Calretinin is encoded by the CALB2 
gene.
13
As MM diagnosis is usually made when the dis-
ease is already advanced, blood-based biomarkers 
such as mesothelin and fibulin-3 that would enable 
an earlier diagnosis and better prognosis of MM 
are extensively studied.
14,15
 Recently, calretinin was 
also proposed as a soluble biomarker in MM, as 
increased plasma or serum levels were observed 
in MM patients compared to subjects with other 
asbestos-related diseases or healthy controls.
8,16-18
 
However, interindividual variability limits the 
sensitivity and specificity of calretinin as a diag-
nostic biomarker and several clinical characteris-
tics were previously associated with soluble cal-
retinin levels.
19
 Low tumor calretinin expression 
was associated with lower protein concentration 
in the bloodstream, but there was no clear correla-
tion with tumor size.
20
 Higher calretinin concen-
trations were observed in patients with epithelioid 
or biphasic MM compared to patients with sarco-
matoid MM.
8,20,21
 Calretinin levels were also higher 
in women compared to men and in subjects with 
renal dysfunction.
22
Molecular mechanisms regulating calretinin 
expression in various tissues or in cancer could 
also contribute to interindividual variability of se-
rum calretinin concentration, but the knowledge 
of these processes is limited.
23
 Calretinin expres-
sion may be affected by several factors, including 
transcription factors or miRNAs. Among tran-
scription factors, calretinin expression was found 
to be influenced by septin 7, E2F transcription fac-
tor 2 (E2F2) and nuclear respiratory factor 1 (NRF-
1) in previous studies.
23,24
 Additionally, miR-335-5p 
was proposed as a regulator of CALB2 expression
25
 
and miR-30e-5p was negatively correlated with 
the calretinin expression in pleural MM patient 
samples.
26
 Gene expression can also be modified 
by genetic variability in the promoter 5’ untrans-
lated region (UTR) of the gene affecting binding of 
transcription factors, or genetic variability in the 3’ 
UTR affecting miRNA binding. Polymorphisms in 
genes coding for miRNAs or transcription factors 
involved in calretinin regulation could also influ-
ence calretinin expression. In previous studies, ge-
netic factors affecting expression and circulating 
levels of other important MM biomarkers such as 
mesothelin have already been identified.
27-29
 On 
the other hand, very little is known about the role 
of single nucleotide polymorphisms (SNPs) in the 
CALB2 gene. An intronic polymorphism in CALB2 
gene was previously proposed as a risk factor for 
colon cancer.
30
 To date, no studies have been per-
formed to evaluate if genetic factors influence cal-
retinin expression or if they could modify suscep-
tibility to develop asbestos-related diseases.
Our aim was to determine whether genetic pol-
ymorphisms in the CALB2 gene and in the genes 
coding for miRNA and transcription factors regu-
lating calretinin expression are associated with 
MM susceptibility or serum calretinin levels in 
patients with asbestos-related diseases.
Subjects and methods
Study population
Our retrospective study included patients with 
MM, subjects with asbestosis, subjects with pleu-
ral plaques, and subjects that were occupationally 
exposed to asbestos but, did not develop any as-
bestos-related disease.
Patients with MM were treated at the Institute of 
Oncology Ljubljana between November 2001 and 

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 475
March 2019. The diagnosis of pleural or peritoneal 
MM was performed by thoracoscopy or laparos-
copy, respectively, and confirmed histologically by 
an experienced pathologist, mostly in others ter-
tiary institutions in Slovenia. Stage of MM was de-
termined using the TNM staging system for pleu-
ral MM. Performance status of MM patients was 
determined using Eastern Cooperative Oncology 
Group (ECOG) scores.
Subjects with asbestosis, subjects with pleural 
plaques and asbestos-exposed subjects who did 
not develop any asbestos-related disease were 
selected from a cohort of occupationally exposed 
workers who were evaluated by the State Board 
for the Recognition of Occupational Asbestos 
Diseases at the Clinical Institute of Occupational, 
Traffic and Sports Medicine in Ljubljana between 
September 1998 and April 2007 . The diagnosis of as-
bestos-related diseases was based on the Helsinki 
Criteria for Diagnosis and Attribution of Asbestos 
Diseases
31
 and the American Thoracic Society rec-
ommendations.
32
 Follow-up was performed for all 
subjects in 2018 to confirm they did not develop 
any other asbestos-related disease.
For all subjects, data on demographic (sex, 
age, smoking) and clinical characteristics were 
obtained from the medical records or during an 
interview. All participants provided written in-
formed consent. The study has been approved 
by the National Medical Ethics Committee of 
the Republic of Slovenia (31/07/04, 39/04/06 and 
41/02/09) and was carried out according to the 
Declaration of Helsinki.
Bioinformatic analysis
Using bioinformatic analysis, we identified com-
mon SNPs that could affect calretinin expression: 
SNPs in the 5’ UTR and 3’ UTR of the calretinin 
gene (CALB2) and SNPs in the genes coding for 
miRNAs and transcription factors involved in the 
regulation of calretinin expression. Experimentally 
confirmed miRNAs and transcription factors were 
selected using miRTarBase
33
 and literature screen-
ing.
Using LD Tag SNP Selection tool
34
 and dbSNP 
database
35
, we identified all SNPs in 5’ UTR, 3’ 
UTR and near gene regions (± 1000 base pairs) of 
CALB2 gene and all SNPs in 5’ UTR, 3’ UTR and 
coding regions of transcription factor coding genes 
with minor allele frequency (MAF) in European 
populations above 5%. Additionally, available lit-
erature was screened for SNPs in miRNA coding 
genes.
36
 In silico predicted function of SNPs was 
assessed using SNP Function Prediction tool
34
 as 
well as HaploReg v4.1
37
 and GTEx
38
 for SNPs in 
regulatory regions. Linkage disequilibrium (LD) 
between SNPs in one gene was evaluated using 
LD link tool.
39
 For genotyping, we selected only 
SNPs with in silico predicted functional role (non-
synonymous SNPs, SNPs that influence transcrip-
tion factor or miRNA binding or SNPs that influ-
ence splicing). If more SNPs within one gene were 
in high LD (R
2
 > 0.8), only one SNP was selected for 
genotyping analyses.
DNA extraction and genotyping
Genomic DNA was extracted from peripheral ve-
nous blood samples using Qiagen FlexiGene Kit 
(Qiagen, Hilden, Germany) according to the manu-
facturer’s instructions. For a subset of subjects that 
did not develop any asbestos-related disease, DNA 
was extracted from capillary blood samples collect-
ed on Whatman FTA cards using MagMax
TM
 DNA 
Multi-Sample Kit (Applied Biosystems, Foster City, 
California, USA). The genotyping of all selected 
SNPs was carried out using a fluorescence-based 
competitive allele-specific polymerase chain reac-
tion (KASP) assay, according to the manufacturer’s 
instructions (LGC Genomics, UK). For all SNPs, 
15% of samples were genotyped in duplicates. 
Genotyping quality control criteria were 100% du-
plicate call rate and 95% SNP-wise call rate.
Measurement of serum calretinin
Serum samples were collected at diagnosis for 
MM patients and at inclusion in the study for all 
other subjects. Samples were prepared within 
6 hours of blood collection and stored at -20°C. 
Serum calretinin levels were determined using a 
commercially available enzyme-linked immuno-
sorbent Calretinin ELISA assay (DLD Diagnostika 
GmbH, Germany) according to the manufacturer’s 
instructions as previously described.
8,16,21
Statistical Analysis
Continuous and categorical variables were de-
scribed using median with interquartile range and 
frequencies, respectively. Nonparametric Mann-
Whitney test or Kruskal-Wallis test with post hoc 
Bonferroni corrections for pairwise comparisons 
were used to compare the distribution of continu-
ous variables. Chi square test was used to compare 
the distribution of categorical variables among 
different groups and to evaluate deviation from 

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 476
Hardy-Weinberg equilibrium. For all investigated 
SNPs, both additive and dominant models were 
used in the analysis. Univariate and multivariate 
logistic regression was used to compare genotype 
frequencies between groups and to determine 
odds ratios (ORs) and 95% confidence intervals 
(CIs). Demographic and clinical parameters, sig-
nificantly associated with asbestos-related disease 
susceptibility in univariate analysis, were used for 
adjustment in multivariate models. Receiver oper-
ating characteristic (ROC) curve analysis was used 
to determine area under the curve (AUC), sensitiv-
ity and specificity. Cut-off values were determined 
as the values with the highest sum of sensitivity 
and specificity. All statistical tests were two-sided 
and the level of significance was set at 0.05. The 
statistical analyses were carried out by using IBM 
SPSS Statistics version 27.0 (IBM Corporation, 
Armonk, NY, USA). To assess the combined ef-
fect of all CALB2 SNPs, we reconstructed haplo-
types using Thesias software.
40
 Haplotypes with 
predicted frequency above 0.04 were included in 
the analysis and the most common haplotype was 
used as a reference. 
Results 
Subjects’ characteristics
Among 904 subjects included in our study, 288 
(31.9%) had MM. Among 616 non-MM subjects 
that were occupationally exposed to asbestos, 153 
subjects had asbestosis, 380 subjects had pleural 
plaques and 83 did not develop any asbestos-relat-
ed disease. Characteristics of each subject group 
are presented in Table 1. Patients with MM were 
older than all other groups (P < 0.001), but there 
were no significant differences regarding sex (P = 
0.180) and smoking (P = 0.205).
Among patients with MM, 217 (75.3%) patients 
had epithelioid histological type, 26 (9.0%) patients 
had biphasic type, and 26 (9.0%) patients had sar-
comatoid type, while histological type could not 
be determined in 19 (6.6%) patients. According to 
cancer stage, 19 (6.6%) patients had stage 1 MM, 
63 (22.0%) patients had stage 2 MM, 85 (29.6%) pa-
tients had stage 3 MM, and 87 (30.3%) patients had 
stage 4 MM, while no data were available for one 
patient. Additionally, 33 (11.5%) patients had peri-
toneal MM. Regarding ECOG performance status, 
18 patients (6.3%) had score 0, 142 (49.5%) score 1, 
110 (38.3%) score 2 and 17 (5.9%) score 3, while no 
data was available for one patient.
Bioinformatic analysis
Based on available literature and publicly available 
databases, we identified genes and SNPs that could 
influence calretinin expression and serum levels: 
SNPs in 5’ and 3’ UTR of CALB2 gene and SNPs 
in genes coding for transcription factors and miR-
NAs associated with calretinin expression. Three 
miRNAs were experimentally associated with 
regulation of CALB2 expression: hsa-miR-9, hsa-
miR-30e and hsa-miR-335-5p
26
 but common SNPs 
were only described in MIR335 gene. Additionally, 
three transcription factors were experimentally as-
sociated with regulation of CALB2 expression: E2F 
transcription factor 2 (E2F2), nuclear respiratory 
factor 1 (NRF1), and septin 7 (SEPTIN7).
23,24
In total, seven SNPs fulfilling all inclusion cri-
teria were included in the study: CALB2 rs1862818, 
CALB2 rs889704, CALB2 rs8063760, E2F2 rs2075995, 
MIR335 rs3807348, NRF1 rs13241028, and SEPTIN7 
rs3801339. Their role, predicted function and geno-
type frequencies in the whole study group as well 
as minor allele frequency and agreement with 
Hardy-Weinberg equilibrium (HWE) in controls 
TABLE 1. Clinical characteristics of the subjects included in the study
Characteristic Category/unit
No disease
(N = 83)
Pleural plaques
(N = 380)
Asbestosis
(N = 153)
MM
(N = 288)
P
Sex
Male, N (%) 61 (73.5) 262 (68.9) 119 (77.8) 213 (74.0) 0.180
1
Female, N (%) 22 (26.5) 118 (31.1) 34 (22.2) 75 (26.0)
Age
Median
(25%−75%)
53.4
(48.5−59.2)
54.8
(48.8−62.7)
59.4
(51.3−66.1)
66.0
(59−73)
< 0.001
2
Smoking
No, N (%) 46 (55.4) 187 (49.2) 74 (48.4) 158 (56.4) [8] 0.205
1
Yes, N (%) 37 (44.6) 193 (50.8) 79 (51.6) 122 (43.6)
1 
calculated using chi-square test; 
2
 calculated using Kruskal-Wallis test. Number of missing data is presented in [] brackets.
MM = malignant mesothelioma

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 477
are presented in Table 2. All SNPs were in agree-
ment with HWE in controls without asbestos relat-
ed diseases and variant allele frequencies ranged 
between 14 and 63%.
Association of selected SNPs with MM 
susceptibility
In the whole study group, we evaluated if selected 
polymorphisms were associated with MM suscep-
tibility. Genotype frequencies in MM patients and 
subjects without MM and are presented in Table 3. 
Carriers of two polymorphic E2F2 rs2075995 al-
leles were less likely to develop MM (OR = 0.64, 
95% CI = 0.43−0.96, P = 0.032), but the association 
was no longer significant after adjustment for 
age (OR = 0.68, 95% CI = 0.44−1.07, P = 0.093). No 
other SNP was significantly associated with MM 
susceptibility (Table 3). Additionally, we also com-
pared MM patients to other subject groups sepa-
rately. Genotype frequencies of SNPs among sub-
jects with asbestosis, subjects with pleural plaques 
and subjects without asbestos-related diseases, are 
presented in Supplementary Table 1. When com-
paring MM patients with subjects without any 
asbestos-related disease, carriers of two polymor-
TABLE 2. Genotype frequencies of investigated single nucleotide polymorphisms (SNPs) in the whole study group, their variant allele frequency 
(VAF) and agreement with Hardy-Weinberg equilibrium (HWE) in subjects without any asbestos-related disease (controls)
Gene SNP
Nucleotide or amino 
acid change
Predicted
function
Genotype N (%)
VAF 
(controls)
pHWE 
(controls)
CALB2 rs1862818 c.-828C>T
May influence transcription 
factor binding, may alter 
chromatin states and 
regulatory motifs
CC 479 (53.0) 0.27 0.617
CT 346 (38.3)
TT 79 (8.7)
CALB2 rs889704 c.-634C>A
May influence transcription 
factor binding, may alter 
chromatin states and 
regulatory motifs
CC
708 (78.4) 
[1]
0.14 0.814
CA 182 (20.2)
AA 13 (1.4)
CALB2 rs8063760 c.*138T>C
May influence miRNA 
binding, may alter 
regulatory motifs
CC
527 (58.4) 
[2]
0.23 0.322
CT 319 (35.4)
TT 56 (6.2)
E2F2 rs2075995 c.678C>A, p.Gln226His
Nonsynonymous, may 
influence splicing
CC 187 (20.7) 0.61 0.209
CA 468 (51.8)
AA 249 (27.5)
MIR335 rs3807348 g.130496266G>A
Downstream transcript 
variant, may influence 
transcription factor binding
GG
228 (25.3) 
[3]
0.49 0.376
GA 446 (49.5)
AA 227 (25.2)
NRF1 rs13241028 c.*1321T>C
May influence miRNA 
binding
TT 547 (60.5) 0.22 0.061
TC 313 (34.6)
CC 44 (4.9)
SEPTIN7 rs3801339 c.1168 - 4 4 51T>C
Genic downstream 
transcript variant
1
TT 164 (18.1) 0.63 0.187
TC 401 (44.4)
CC 339 (37.5)
1 
previously classified as a nonsynonymous variant. Number of missing data is presented in [] brackets.
A = adenine; C = cytosine; G = guanine; SNP = single nucleotide polymorphisms; T = thymine

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 478
TABLE 3. Association of investigated single nucleotide polymorphisms (SNPs) with malignant mesothelioma (MM) susceptibility
SNP Genotype
Subjects without 
MM
(N = 616)
N (%)
MM patients
(N = 288)
N (%)
OR (95% CI) P OR (95% CI)
adj
P
adj
CALB2 rs1862818 CC
340
(55.2)
139
(48.3)
Reference Reference
CT
226
(36.7)
120
(41.7)
1.30
(0.97−1.75)
0.084
1.35
(0.97−1.87)
0.073
TT
50
(8.1)
29
(10.1)
1.42 
(0.86−2.34)
0.169
1.34 
(0.77−2.32)
0.299
CT+TT
276
(44.8)
149
(51.7)
1.32 
(1.00−1.75)
0.052
1.35 
(0.99−1.83)
0.059
CALB2 rs889704 CC
485
(78.9) [1]
223
(77.4)
Reference Reference
CA
121
(19.7)
61
(21.2)
1.10 
(0.78−1.55)
0.602
1.03 
(0.70−1.51)
0.899
AA
9
(1.5)
4
(1.4)
0.97 
(0.29−3.17)
0.955
0.55 
(0.15−1.94)
0.349
CA+AA
130
(21.1)
65
(22.6)
1.09 
(0.78−1.52)
0.626
0.98 
(0.67−1.42)
0.912
CALB2 rs8063760 CC
352
(57.3) [2]
175
(60.8)
Reference Reference
CT
222
(36.2)
97
(33.7)
0.88 
(0.65−1.19)
0.398
0.91 
(0.66−1.26)
0.576
TT
40
(6.5)
16
(5.6)
0.80 
(0.44−1.48)
0.483
0.82 
(0.42−1.60)
0.554
CT+TT
262
(42.7)
113
(39.2)
0.87 
(0.65−1.15)
0.329
0.90 
(0.65−1.23)
0.493
E2F2 rs2075995 CC
117
(19.0)
70
(24.3)
Reference Reference
CA
319
(51.8)
149
(51.7)
0.78 
(0.55−1.11)
0.171
0.83 
(0.56−1.23)
0.349
AA
180
(29.2)
69
(24.0)
0.64 
(0.43−0.96)
0.032
0.68 
(0.44−1.07)
0.093
CA+AA
499
(81.0)
218
(75.7)
0.73 
(0.52−1.02)
0.067
0.78 
(0.53−1.13)
0.182
MIR335 
rs3807348
GG
158
(25.8) [3]
70
(24.3)
Reference Reference
GA
307
(50.1)
139
(48.3)
1.02 
(0.72−1.44)
0.902
1.00 
(0.68−1.46)
0.980
AA
148
(24.1)
79
(27.4)
1.20 
(0.81−1.78)
0.352
1.22 
(0.79−1.87)
0.376
GA+AA
455
(74.2)
218
(75.7)
1.08 
(0.78−1.50)
0.636
1.07 
(0.75−1.52)
0.724
NRF1 rs13241028 TT
374
(60.7)
173
(60.1)
Reference Reference
TC
210
(34.1)
103
(35.8)
1.06 
(0.79−1.43)
0.699
1.08 
(0.78−1.50)
0.636
CC
32
(5.2)
12
(4.2)
0.81 
(0.41−1.61)
0.550
0.92 
(0.44−1.93)
0.823
TC+CC
242
(39.3)
115
(39.9)
1.03 
(0.77−1.37)
0.853
1.06 
(0.78−1.45)
0.711
SEPTIN7 
rs3801339
TT
109
(17.7)
55
(19.1)
Reference Reference
TC
266
(43.2)
135
(46.9)
1.01 
(0.68−1.48)
0.976
1.05 
(0.69−1.61)
0.815
CC
241
(39.1)
98
(34.0)
0.81 
(0.54−1.20)
0.291
0.76 
(0.49−1.18)
0.218
TC+CC
507
(82.3)
233
(80.9)
0.91 
(0.64−1.30)
0.610
0.91 
(0.61−1.35)
0.627
Number of missing data is presented in [] brackets.
A = adenine; Adj = adjusted for age; C = cytosine; CI = confidence interval; G = guanine; OR = odds ratio; T= thymine

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 479
TABLE 4. Association of selected SNPs with serum calretinin concentration
SNP Genotype All subjects Subjects without MM MM patients
Calretinin 
(ng/ml)
Median 
(25 −75%)
P
add
P
dom
Calretinin 
(ng/ml)
Median 
(25 −75%)
P
add
P
dom
Calretinin 
(ng/ml)
Median 
(25 −75%)
P
add
P
dom
CALB2 
rs1862818
CC
0.18 
(0.11−0.34)
0.622 0.422
0.15 
(0.09−0.22)
0.751 0.865
0.64 
(0.22−1.45)
0.952 0.802
CT
0.19 
(0.11−0.41)
0.16 
(0.09−0.24)
0.51 
(0.23−1.41)
TT
0.18 
(0.10−0.37)
0.13 
(0.08−0.20)
0.38 
(0.21−3.57)
CT+TT
0.19 
(0.11−0.40)
0.15 
(0.09−0.24)
0.48 
(0.23−1.43)
CALB2 
rs889704
CC
0.19 
(0.11−0.37)
0.099 0.036
0.15 
(0.10−0.23)
0.130 0.069
0.52 
(0.25−1.43)
0.508 0.441
CA
0.17 
(0.08−0.27)
0.16 
(0.08−0.21)
0.44 
(0.14−1.35)
AA
0.21 
(0.05−0.77)
0.10 
(0.02−0.21)
1.07 
(0.28−1.84)
CA+AA
0.17 
(0.08−0.28)
0.14 
(0.07−0.21)
0.50 
(0.15−1.51)
CALB2 
rs8063760
CC
0.18 
(0.11−0.38)
0.955 0.770
0.14 
(0.09−0.22)
0.382 0.647
0.53 
(0.24−1.44)
0.326 0.768
CT
0.18 
(0.12−0.32)
0.16 
(0.1−0.24)
0.44 
(0.19−1.30)
TT
0.21 
(0.06−0.51)
0.12 
(0.05−0.22)
0.86 
(0.50−2.30)
CT+TT
0.19 
(0.11−0.34)
0.16 
(0.09−0.24)
0.51 
(0.21−1.43)
E2F2 
rs2075995
CC
0.19 
(0.10−0.46)
0.512 0.481
0.14 
(0.08−0.2)
0.161 0.059
0.72 
(0.33−1.45)
0.189 0.117
CA
0.18 
(0.12−0.34)
0.16 
(0.1−0.23)
0.53 
(0.20−1.48)
AA
0.18 
(0.10−0.33)
0.14 
(0.09−0.24)
0.40 
(0.18−0.90)
CA+AA
0.18 
(0.11−0.34)
0.15 
(0.1−0.23)
0.48 
(0.20−1.44)
MIR335 
rs3807348
GG
0.18 
(0.09−0.34)
0.057 0.151
0.14 
(0.08−0.2)
0.027 0.081
0.44 
(0.26−1.43)
0.400 0.978
GA
0.18 
(0.11−0.34)
0.14 
(0.09−0.22)
AA vs. 
GG P = 
0.029
0.50 
(0.18−1.16)
AA
0.21 
(0.13−0.39)
0.18 
(0.11−0.26)
0.65 
(0.27−1.80)
GA+AA
0.19 
(0.11−0.37)
0.15 
(0.1−0.23)
0.52 
(0.22−1.44)
NRF1 
rs13241028
TT
0.19 
(0.12−0.36)
0.272 0.144
0.16 
(0.1−0.23)
0.096 0.034
0.52 
(0.21−1.15)
0.381 0.672
TC
0.18 
(0.10−0.33)
0.14 
(0.08−0.21)
0.64 
(0.25−1.67)
CC
0.17 
(0.07−0.36)
0.15 
(0.07−0.3)
0.24 
(0.07−1.18)
TC+CC
0.18 
(0.09−0.34)
0.14 
(0.08−0.21)
0.46 
(0.24−1.53)
SEPTIN7 
rs3801339
TT
0.18 
(0.11−0.34)
0.403 0.419
0.14 
(0.09−0.2)
0.424 0.288
0.35 
(0.17−1.05)
0.079 0.080
TC
0.18 
(0.11−0.33)
0.15 
(0.09−0.22)
0.51 
(0.21−1.23)
CC
0.20 
(0.11−0.45)
0.16 
(0.09−0.25)
0.72 
(0.38−1.48)
TC+CC
0.19 
(0.11−0.37)
0.15 
(0.09−0.23)
0.64 
(0.26−1.45)
A = adenine; Add = additive model, calculated using Kruskal-Wallis test; C = cytosine; Dom = dominant model, calculated using Mann-Whitney test; G = guanine;  
MM = malignant mesothelioma, SNP = single nucleotide polymorphism, T = thymine

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 480
phic E2F2 rs2075995 alleles were less likely to de-
velop MM (OR = 0.35, 95% CI = 0.16−0.78, P = 0.010), 
even after adjustment for age (OR = 0.35, 95% CI 
= 0.14−0.84, P = 0.019). The association with MM 
susceptibility was significant also in the domi-
nant model, both in univariate (OR = 0.43, 95% CI 
= 0.21−0.87, P = 0.019) and multivariate (OR = 0.43, 
95% CI = 0.19−0.94, P = 0.033) analysis. Compared to 
subjects with asbestosis, carriers of two polymor-
phic MIR335 rs3807348 alleles were more likely 
to develop MM (OR = 1.82, 95% CI = 1.05−3.16, P 
= 0.033), even after adjustment for age (OR = 0.35, 
95% CI = 1.10−3.50, P = 0.022). After adjustment for 
age, the association with MM susceptibility was 
significant also in the dominant model (OR = 1.62, 
95% CI = 1.03−2.55, P = 0.037). None of the other 
SNPs was significantly associated with MM sus-
ceptibility (Supplementary Table 2).
Association of selected SNPs with serum 
calretinin levels
Serum calretinin concentration was determined in 
545 subjects. Calretinin concentration significantly 
differed among subject groups (P < 0.001): MM pa-
tients (N = 163) had median calretinin concentra-
tion 0.52 (0.23−1.43) ng/ml, subjects with asbestosis 
(N = 117) 0.13 (0.08−0.20) ng/ml, subjects with pleu-
ral plaques (N = 195) 0.18 (0.12−0.25) ng/ml and sub -
jects without disease (N = 70) 0.12 (0.07−0.19) ng/ml.
TABLE 5. Receiver operating characteristic (ROC) curve analysis according to individual genotypes for selected single nucleotide polymorphisms: 
comparison of malignant mesothelioma (MM) patients with all other subjects
SNP Genotype AUC (95% CI) P
Calretinin
cut-off (ng/ml)
1
Sensitivity Specificity
Overall 
analysis in 
the whole 
group
/
0.825 
(0.781−0.868)
< 0.001 0.32 0.681 0.887
CALB2 
rs889704
CC
0.831 
(0.782−0.880)
< 0.001 0.32 0.695 0.876
CA
0.779 
(0.667−0.891)
< 0.001 0.31 0.607 0.935
AA
2
0.958 
(0.837−1.000)
0.019 0.21 1.000 0.833
CA+AA
0.801 
(0.702−0.901)
< 0.001 0.31 0.625 0.940
E2F2 
rs2075995
CC
0.906 
(0.845−0.968)
< 0.001 0.26 0.810 0.903
CA
0.803 
(0.736−0.869)
< 0.001 0.32 0.671 0.888
AA
0.781 
(0.686−0.876)
< 0.001 0.33 0.615 0.877
CA+AA
0.797 
(0.742−0.851)
< 0.001 0.32 0.653 0.881
MIR335 
rs3807348
GG
0.853 
(0.766−0.940)
< 0.001 0.29 0.757 0.872
GA
0.803 
(0.739−0.867)
< 0.001 0.32 0.643 0.892
AA
0.845 
(0.765−0.925)
< 0.001 0.35 0.738 0.881
GA+AA
0.815 
(0.764−0.866)
< 0.001 0.32 0.675 0.881
NRF1 
rs13241028
TT
0.812 
(0.754−0.871)
< 0.001 0.32 0.693 0.884
TC
0.868 
(0.804−0.931)
< 0.001 0.23 0.818 0.798
CC
3
0.664 
(0.406−0.922)
0.203 0.18 0.714 0.700
TC+CC
0.842 
(0.777−0.907)
< 0.001 0.23 0.790 0.785
1
 Cut-off with the highest sum of sensitivity and specificity; 
2
 based on 10 subjects, 3 based on 27 subjects.
A = adenine; AUC = area under the curve; C = cytosine; G = guanine; SNP = single nucleotide polymorphism; T = thymine

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 481
The association of selected SNPs with serum 
calretinin concentration is presented in Table 4 
and Figure 1. In all subjects, carriers of at least one 
polymorphic CALB2 rs889704 A allele had lower 
calretinin than carriers of two wild-type alleles 
in the dominant model (P = 0.036), but no signifi-
cant differences were observed if subjects without 
MM and MM patients were evaluated separately 
(P = 0.069 and 0.441, respectively). In the group of 
subjects without MM, carriers of two polymorphic 
MIR335 rs3807348 alleles had higher calretinin 
than carriers of two wild-type alleles (P = 0.027). In 
this group also carriers of at least one polymorphic 
NRF1 rs13241028 C allele had lower calretinin than 
TABLE 6. Association of CALB2 haplotypes with malignant mesothelioma (MM) susceptibility and serum calretinin concentration
Haplotype
Subjects  
without MM
Predicted 
frequency
MM patients
Predicted 
frequency
OR
(95% CI)
P
OR
(95% CI)
adj
P
adj
Serum calretinin 
concentration
P
CCC 0.457 0.431 Reference Reference
TCC 0.245 0.294
1.26
(0.0−991.60)
0.061
1.26 
(0.97−1.64)
0.084 0.272
CCT 0.176 0.147
0.88 
(0.65−1.20)
0.415
0.94 
(0.66−1.34)
0.731 0.125
CAT 0.058 0.066
1.21 
(0.77−1.89)
0.408
1.08 
(0.64−1.81)
0.782 0.731
CAC 0.045 0.047
1.11 
(0.64−1.91)
0.713
0.99 
(0.55−1.79)
0.974 0.852
The SNPs are ordered from the 5′- to 3′-end as follows: rs1862818, rs889704, rs8063760. 
A = adenine; Adj = adjusted for age, C = cytosine; CI = confidence interval; MM = malignant mesothelioma; OR = odds ratio; SNP = single nucleotide polymorphism;  
T = thymine
FIGURE 1. Association of selected single nucleotide polymorphisms (SNPs) with serum calretinin concentration: CALB2 rs889704 (A), E2F2 rs2075995 
(B), MIR335 rs3807348 (C), NRF1 rs13241028 (D).
A B
C D

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 482
carriers of two wild-type alleles in the dominant 
model (P = 0.034), but no significant differences 
were observed in group of MM patients. 
Association of selected SNPs with serum cal-
retinin concentration in subjects with asbestosis, 
subjects with pleural plaques and subjects with-
out disease is shown in Supplementary Table 3. 
In subjects without asbestos-related disease, car-
riers of at least one polymorphic CALB2 rs889704 
A allele had lower calretinin than carriers of two 
wild-type alleles in the additive model (P = 0.014) 
and dominant model (P = 0.004), but no significant 
differences were observed in subjects with pleu-
ral plaques (P
add
 = 0.060, P
dom
 = 0.300) and subjects 
with asbestosis (P
add
 = 0.290, P
dom
 = 0.279). In sub-
jects with pleural plaques, carriers of at least one 
polymorphic NRF1 rs13241028 C allele had lower 
calretinin than carriers of two wild-type alleles 
in the dominant model (P = 0.025). In subjects 
with asbestosis, carriers of at least one polymor-
phic E2F2 rs2075995 A allele had higher calretinin 
than carriers of two wild-type alleles in the ad-
ditive model (P = 0.049) and dominant model (P 
= 0.017). With ROC curve analysis, we compared 
serum calretinin levels in MM patients with all 
other subjects according to individual genotypes 
for SNPs, which affected calretinin levels in at 
least one group. In almost all groups, calretinin 
concentration could significantly discriminate be-
tween MM patients and other subjects with good 
sensitivity and specificity (Table 5). Optimal cal-
retinin cut off values differed according to geno-
type, even though the differences were small. For 
CALB2 rs889704, lower cut off was observed in 
carriers of two polymorphic alleles (0.21 vs. 0.32 
ng/ml). For E2F2 rs2075995, higher cut off was ob-
served in carriers of two polymorphic alleles (0.33 
vs. 0.26 ng/ml). For MIR335 rs3807348, higher cut 
off was observed in carriers of two polymorphic 
alleles (0.35 vs. 0.29 ng/ml). For NRF1 rs13241028, 
lower cut off was observed in carriers of at least 
one polymorphic alleles (0.23 vs. 0.32 ng/ml) 
(Table 5).
Haplotype analysis
Analysis of CALB2 haplotypes identified eight 
SNP combinations. The most common haplotype 
was CCC with predicted frequency 0.449, fol-
lowed by TCC (0.261), CCT (0.167), CAT (0.060), 
CAC (0.045), TCT (0.009), TAC (0.007) and TAT 
(0.003). Haplotype TCC was more common in MM 
patients, but the association was not statistically 
significant (P = 0.061, Table 6). CALB2 haplotypes 
were not associated with serum calretinin concen-
trations (Table 6).
Discussion
In the present study, we evaluated the role of ge-
netic variability in CALB2 and its regulatory miR-
NA and transcription factors genes with serum 
calretinin levels and MM susceptibility. Genetic 
variability of CALB2 was associated with cal-
retinin concentration, but not with MM suscep-
tibility. For SNPs in genes regulating calretinin 
expression, differences in genotype frequencies 
among MM and other subjects were also observed. 
Additionally, genetic factors influenced optimal 
serum calretinin cut off values differentiating MM 
patients from other asbestos-exposed subjects.
Using bioinformatic analysis, we identified sev-
en common putatively functional SNPs that could 
affect calretinin expression: three SNPs in CALB2 
gene, one SNP in transcription factor E2F2, one 
SNP in transcription factor NRF1, one SNP in tran-
scription factor SEPTIN7 and one SNP in miRNA 
MIR335. In previous studies, demographic and 
clinical factors such as sex and renal function af-
fecting plasma or serum calretinin concentration 
in asbestos-related diseases were already identi-
fied
21,22,41
, but the role of genetic variability is large-
ly unexplored.
Among CALB2 SNPs investigated in our study, 
CALB2 rs1862818 and CALB2 rs889704 may influ-
ence transcription factor binding, while CALB2 
rs8063760 may influence miRNA binding. In our 
study, CALB2 rs889704 was associated with lower 
serum calretinin levels in all subjects and subjects 
without asbestos-related diseases, while there 
was no association in patients with MM. None of 
the selected CALB2 SNPs or haplotypes were sig-
nificantly associated with MM susceptibility. To 
the best of our knowledge, the functional role of 
CALB2 SNPs and their association with asbestos-
related diseases was not investigated yet. However, 
one intronic SNP in CALB2 was previously associ-
ated with calretinin expression in tumor cell lines 
and the development of colon cancer, but no asso-
ciation with lung cancer was observed.
30
 Data on 
CALB2 genetic variability are therefore lacking 
and further studies are needed to evaluate its role 
in MM and serum calretinin levels.
Three important transcription factors were pre-
viously associated with regulation of calretinin.
23,24
 
E2F2 is a transcription factor that binds to CALB2 
promoter and was associated with calretinin ex-

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 483
pression in mesothelioma cell lines.
23
 In our study, 
E2F2 rs2075995 was associated with decreased 
MM risk. When comparing MM patients to only 
subjects without disease, the association remained 
significant even after taking into account the age 
of the subjects. E2F2 rs2075995 was also associated 
with higher serum calretinin level among subjects 
with asbestosis. E2F2 has an important role in the 
regulation of cell cycle, but also affects other impor-
tant processes such as cell proliferation, apoptosis 
and inflammation.
42
 In cancer, it was mostly asso-
ciated with promoting tumor progression in vari-
ous malignancies, including lung cancer.
42
 E2F2 
could also contribute to the cell cycle-dependent 
differences observed for calretinin expression.
23
 
E2F2 rs2075995 is a nonsynonymous SNP and may 
influence splicing. So far, E2F2 rs2075995 was only 
evaluated in patients with colorectal cancer, where 
no association with cancer risk was observed.
43,44
 
However, no studies evaluated the association of 
E2F2 rs2075995 with MM. Still, the E2F gene fami-
ly was often associated with different types of can-
cer. Several other E2F2 polymorphisms were asso-
ciated with oral and oropharyngeal squamous cell 
carcinoma risk and might also affect the course 
of the disease.
45
 Combinations of different E2F2 
gene SNPs were proposed as a risk factor for squa-
mous cell carcinoma of the head and neck.
46
 The 
E2F2 gene was also associated with ovarian cancer 
risk.
47
 Additionally, E2F2 genetic variability was 
proposed as recurrence biomarker in squamous 
cell carcinoma of the oropharynx.
48
 Among other 
E2F2 SNPs, rs3218211 was in very high LD with 
rs2075995 investigated in our study. E2F2 rs3218211 
was associated with T stage in oral and oropharyn-
geal squamous cell carcinoma and decreased head 
and neck squamous cell carcinoma risk, consistent 
with our results.
45,46
 Taken together, this suggests 
further studies regarding the role of E2F2 genetic 
variability in asbestos-related diseases and its as-
sociation with calretinin are needed.
The second important calretinin-related tran-
scription factor is NRF-1. It binds to CALB2 pro-
moter and might be important for the transcrip-
tional control of calretinin expression in MM.
23
 In 
our study, NRF1 rs13241028 was associated with 
lower serum calretinin level in subjects without 
MM, but it was not associated with MM suscepti-
bility. NRF-1 regulates expression of various genes 
involved in oxidative phosphorylation, mitochon-
drial biogenesis and other mitochondrial pro-
cesses, including transcription of mitochondrial 
DNA.
49
 Additionally, NRF-1 can modify different 
aspects of carcinogenesis, including proliferation, 
invasion, and apoptosis.
50
 NRF1 rs13241028 may 
influence miRNA binding.
51
 So far, NRF1 genetic 
variability has been associated primarily with 
increased susceptibility to diabetes.
52,53
 NRF1 has 
also been associated with epithelial ovarian cancer 
risk.
54
 Further studies are needed to better evalu-
ate the role of NRF-1 and its genetic variability in 
asbestos-related diseases.
Septin 7 has also been identified as a factor that 
binds to the CALB2 promoter region, resulting in 
decreased calretinin expression in mesothelioma 
cell lines.
24
 Septin 7 is a GTP-binding protein that is 
involved in cytokinesis, cytoskeleton organization 
and other cellular processes.
24,55
 It was also impli-
cated in calcium homeostasis.
56
 Several studies al-
so reported that septin 7 plays an important role in 
cancer development, especially glioma.
55,56
 In our 
study, SEPTIN7 rs3801339 was not significantly 
associated with MM susceptibility or with serum 
calretinin levels. The functional role of SEPTIN7 
rs3801339 is not yet understood: it was previously 
classified as a non-synonymous variant, while it is 
now described as a genic downstream transcript 
variant. Interestingly, SEPTIN7 rs1143149 in mod-
erate LD with rs3801339 was proposed as a risk 
factor for the development of non-small cell lung 
cancer and was associated with shorter survival in 
long-term smokers.
55
 SEPTIN7 was often mutated 
in breast ductal carcinoma in situ cell lines and 
these mutations might participate in the progres-
sion of breast ductal carcinoma.
57
 Recent studies 
therefore suggest that SEPTIN7 variability may 
play a role in some cancers, but it was not an im-
portant risk factor in asbestos-related diseases in 
our study.
MiRNAs affect gene expression on the post-
transcriptional level and are often deregulated in 
cancer.
58
 Among miRNAs predicted to modify cal-
retinin expression, common polymorphisms were 
only described for miR-335. In our study, carriers 
of two polymorphic MIR335 rs3807348 alleles were 
more likely to develop MM compared to subjects 
with asbestosis, even after adjustment for age. 
MIR335 rs3807348 was also associated with serum 
calretinin level in subjects without MM. MiR-335 
can modulate cell proliferation, apoptosis, migra-
tion and invasion through various signaling path-
ways. It mostly acts as a tumor suppressor and is 
downregulated in different cancer types.
58
 MIR335 
rs3807348 may influence transcription factor bind-
ing, but its role has not been experimentally con-
firmed. To date, no research has been done on 
the association of rs3807348 with MM. MIR335 
rs3807348 was not associated with breast cancer 

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 484
risk in a previous study
59
, but more studies would 
be needed in this field.
As several genetic factors were associated with 
calretinin, we also evaluated how these factors in-
fluence serum calretinin cut off values. We found 
that four SNPs, CALB2 rs889704, E2F2 rs2075995, 
MIR335 rs3807348, and NRF1 rs13241028 could be 
used to fine tune serum calretinin cut off values 
predicting MM. Calretinin as a biomarker could 
thus have higher sensitivity and specificity in indi-
viduals with known genetic variability. Similar re-
sults were observed for mesothelin, where predic-
tive value was improved when taking into account 
polymorphisms located in 5’ UTR and 3’ UTR of 
the MSLN gene.
27-29
 In the future, combination of 
clinical and genetic factors could thus help guide 
calretinin cut-off values and decrease false nega-
tive or positive results.
This is the first study to show that genetic fac-
tors can affect serum calretinin levels and that 
accounting for these genetic factors may improve 
the predictive value of serum calretinin. We have 
also shown that genetic factors associated with 
calretinin may play a role in the development of 
mesothelioma. A limitation of our study is that we 
only had serum calretinin concentrations avail-
able for a subgroup of participants included in the 
study. On the other hand, we performed a com-
prehensive analysis of the factors that could affect 
calretinin expression using literature review and 
detailed bioinformatics analysis. Genetic variabil-
ity was evaluated in a large cohort, which gives 
additional power to the study. However, other pol-
ymorphisms in the investigated genes could also 
affect calretinin concentration and other factors 
could affect calretinin regulation. In the future, 
further studies in this field and validation of these 
results in an independent population are needed.
Conclusions
The present study showed that genetic variability 
in CALB2 gene and genes coding for transcrip-
tion factors and miRNAs that regulate calretinin 
expression could contribute to interindividual dif-
ferences in serum calretinin levels in MM patients 
or asbestos-exposed subjects. These results could 
contribute to a better understanding of calretinin 
regulation and could potentially contribute to an 
earlier diagnosis of MM.
Acknowledgments 
This study was supported by the Slovenian 
Research Agency (ARRS), research grants P1-0170, 
L3-8203 and L3-2622.
References
1. Chapman SJ, Cookson WO, Musk AW, Lee YC. Benign asbestos pleural 
diseases. Curr Opin Pulm Med 2003; 9: 266-71. doi: 10.1097/00063198-
200307000-00004
2. IARC monographs on the evaluation of the carcinogenic risk of chemicals to 
man: asbestos. IARC Monogr Eval Carcinog Risk Chem Man 1977; 14: 1-106. 
PMID: 863456
3. Weiner SJ, Neragi-Miandoab S. Pathogenesis of malignant pleural mesothe-
lioma and the role of environmental and genetic factors. J Cancer Res Clin 
Oncol 2009; 135: 15-27. doi: 10.1007/s00432-008-0444-9
4. Melaiu O, Gemignani F, Landi S. The genetic susceptibility in the develop-
ment of malignant pleural mesothelioma. J Thorac Dis 2018; 10: S246-52. 
doi: 10.21037/jtd.2017.10.41
5. Pylkkänen L, Sainio M, Ollikainen T, Mattson K, Nordling S, Carpén O, et al. 
Concurrent LOH at multiple loci in human malignant mesothelioma with 
preferential loss of NF2 gene region. Oncol Rep 2002; 9: 955-9. doi: 10.3892/
or.9.5.955
6. Murthy SS, Testa JR. Asbestos, chromosomal deletions, and tumor sup-
pressor gene alterations in human malignant mesothelioma. J Cell Physiol 
1999; 180: 150-7. doi: 10.1002/(sici)1097-4652(199908)180:2<150::Aid-
jcp2>3.0.Co;2-h
7. Kovac V, Zwitter M, Zagar T. Improved survival after introduction of chemo-
therapy for malignant pleural mesothelioma in Slovenia: population-based 
survey of 444 patients. Radiol Oncol 2012; 46: 136-44. doi: 10.2478/v10019-
012-0032-0
8. Johnen G, Gawrych K, Raiko I, Casjens S, Pesch B, Weber DG, et al. Calretinin 
as a blood-based biomarker for mesothelioma. BMC Cancer 2017; 17: 386. 
doi: 10.1186/s12885-017-3375-5
9. Carbone M, Adusumilli PS, Alexander HR, Jr., Baas P , Bardelli F, Bononi A, et 
al. Mesothelioma: scientific clues for prevention, diagnosis, and therapy. CA 
Cancer J Clin 2019; 69: 402-29. doi: 10.3322/caac.21572
10. Husain AN, Colby TV, Ordóñez NG, Allen TC, Attanoos RL, Beasley MB, et 
al. Guidelines for pathologic diagnosis of malignant mesothelioma 2017 
update of the consensus statement from the International Mesothelioma 
Interest Group. Arch Pathol Lab Med 2018; 142: 89-108. doi: 10.5858/
arpa.2017-0124-RA
11. Rogers JH. Calretinin: a gene for a novel calcium-binding protein expressed 
principally in neurons. J Cell Biol 1987; 105: 1343-53. doi: 10.1083/
jcb.105.3.1343.
12. Worthmuller J, Blum W, Pecze L, Salicio V, Schwaller B. Calretinin promotes 
invasiveness and EMT in malignant mesothelioma cells involving the ac-
tivation of the FAK signaling pathway. Oncotarget 2018; 9: 36256-72. doi: 
10.18632/oncotarget.26332
13. Parmentier M, Passage E, Vassart G, Mattei MG. The human calbindin 
D28k (CALB1) and calretinin (CALB2) genes are located at 8q21.3----q22.1 
and 16q22----q23, respectively, suggesting a common duplication with the 
carbonic anhydrase isozyme loci. Cytogenet Cell Genet 1991; 57: 41-3. doi: 
10.1159/000133111
14. Cristaudo A, Bonotti A, Guglielmi G, Fallahi P, Foddis R. Serum mesothelin 
and other biomarkers: what have we learned in the last decade? J Thorac 
Dis 2018; 10: S353-9. doi: 10.21037/jtd.2017.10.132
15. Hollevoet K, Reitsma JB, Creaney J, Grigoriu BD, Robinson BW, Scherpereel 
A, et al. Serum mesothelin for diagnosing malignant pleural mesothelioma: 
an individual patient data meta-analysis. J Clin Oncol 2012; 30: 1541-9. doi: 
10.1200/JCO.2011.39.6671

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 485
16. Raiko I, Sander I, Weber DG, Raulf-Heimsoth M, Gillissen A, Kollmeier J, et al. 
Development of an enzyme-linked immunosorbent assay for the detection 
of human calretinin in plasma and serum of mesothelioma patients. BMC 
Cancer 2010; 10: 242. doi: 10.1186/1471-2407-10-242
17. Aguilar-Madrid G, Pesch B, Calderón-Aranda ES, Burek K, Jiménez-Ramírez 
C, Juárez-Pérez CA, et al. Biomarkers for Predicting Malignant Pleural 
Mesothelioma in a Mexican Population. Int J Med Sci 2018; 15: 883-91. doi: 
10.7150/ijms.23939
18. Jiménez-Ramírez C, Casjens S, Juárez-Pérez CA, Raiko I, Del Razo LM, Taeger 
D, et al. Mesothelin, calretinin, and megakaryocyte potentiating factor as 
biomarkers of malignant pleural mesothelioma. Lung 2019; 197: 641-9. doi: 
10.1007/s00408-019-00244-1
19. Li D, Wang B, Long H, Wen F. Diagnostic accuracy of calretinin for malignant 
mesothelioma in serous effusions: a meta-analysis. Sci Rep 2015; 5: 9507. 
doi: 10.1038/srep09507
20. Lehnert M, Weber DG, Taeger D, Raiko I, Kollmeier J, Stephan-Falkenau S, 
et al. Determinants of plasma calretinin in patients with malignant pleural 
mesothelioma. BMC Res Notes 2020; 13: 359. doi: 10.1186/s13104-020-
05187-y
21.	 Zupanc	 C,	 Fr ank o	 A,	 Štrbac	 D ,	 Dodič	 Fikf ak	 M,	 K ov ač	 V ,	 Dolž an	 V ,	 et	 al.	 Serum 	
calretinin as a biomarker in malignant mesothelioma. J Clin Med 2021; 10: 
4875. doi: 10.3390/jcm10214875
22. Casjens S, Weber DG, Johnen G, Raiko I, Taeger D, Meinig C, et al. 
Assessment of potential predictors of calretinin and mesothelin to improve 
the diagnostic performance to detect malignant mesothelioma: results 
from a population-based cohort study. BMJ Open 2017; 7: e017104. doi: 
10.1136/bmjopen-2017-017104
23. Kresoja-Rakic J, Kapaklikaya E, Ziltener G, Dalcher D, Santoro R, Christensen 
BC, et al. Identification of cis- and trans-acting elements regulating calretinin 
expression in mesothelioma cells. Oncotarget 2016; 7: 21272-86. doi: 
10.18632/oncotarget.7114
24. Blum W, Pecze L, Rodriguez JW, Steinauer M, Schwaller B. Regulation of 
calretinin in malignant mesothelioma is mediated by septin 7 binding to 
the CALB2 promoter. BMC Cancer 2018; 18: 475. doi: 10.1186/s12885-018-
4385-7
25. Tavazoie SF, Alarcon C, Oskarsson T, Padua D, Wang Q, Bos PD, et al. 
Endogenous human microRNAs that suppress breast cancer metastasis. 
Nature 2008; 451: 147-52. doi: 10.1038/nature06487
26. Kresoja-Rakic J, Sulemani M, Kirschner MB, Ronner M, Reid G, Kao S, et 
al. Posttranscriptional regulation controls calretinin expression in ma-
lignant pleural mesothelioma. Front Genet 2017; 8: 70. doi: 10.3389/
fgene.2017.00070
27. Goricar K, Kovac V, Dodic-Fikfak M, Dolzan V, Franko A. Evaluation of soluble 
mesothelin-related peptides and MSLN genetic variability in asbestos-relat-
ed diseases. Radiol Oncol 2020; 54: 86-95. doi: 10.2478/raon-2020-0011
28. Garritano S, De Santi C, Silvestri R, Melaiu O, Cipollini M, Barone E, et al. 
A common polymorphism within MSLN affects miR-611 binding site and 
soluble mesothelin levels in healthy people. J Thorac Oncol 2014; 9: 1662-8. 
doi: 10.1097/jto.0000000000000322
29. De Santi C, Pucci P, Bonotti A, Melaiu O, Cipollini M, Silvestri R, et al. 
Mesothelin promoter variants are associated with increased soluble meso-
thelin-related peptide levels in asbestos-exposed individuals. Occup Environ 
Med 2017; 74: 456-63. doi: 10.1136/oemed-2016-104024
30. Vonlanthen S, Kawecki TJ, Betticher DC, Pfefferli M, Schwaller B. 
Heterozygosity of SNP513 in intron 9 of the human calretinin gene (CALB2) 
is a risk factor for colon cancer. Anticancer Res 2007; 27: 4279-88. PMID: 
18214032
31. Tossavainen A. Asbestos, asbestosis, and cancer: the Helsinki criteria for 
diagnosis and attribution. Scand J Work Environ Health 1997; 23: 311-6. doi: 
10.5271/sjweh.226
32. American Thoracic S. Diagnosis and initial management of nonmalignant 
diseases related to asbestos. Am J Respir Crit Care Med 2004; 170: 691-715. 
doi:10.1164/rccm.200310-1436ST
33. Chou CH, Shrestha S, Yang CD, Chang NW , Lin YL, Liao KW , et al. miRTarBase 
update 2018: a resource for experimentally validated microRNA-target in-
teractions. Nucleic Acids Res 2018; 46: D296-302. doi: 10.1093/nar/gkx1067
34. Xu Z, Taylor JA. SNPinfo: integrating GWAS and candidate gene information 
into functional SNP selection for genetic association studies. Nucleic Acids 
Res 2009; 37: W600-5. doi: 10.1093/nar/gkp290
35. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, et al. 
dbSNP: the NCBI database of genetic variation. Nucleic Acids Res 2001; 29: 
308-11. doi: 10.1093/nar/29.1.308
36. Srivastava K, Srivastava A. Comprehensive review of genetic association 
studies and meta-analyses on miRNA polymorphisms and cancer risk. PloS 
One 2012; 7: e50966. doi: 10.1371/journal.pone.0050966
37. Ward LD, Kellis M. HaploReg: a resource for exploring chromatin states, con-
servation, and regulatory motif alterations within sets of genetically linked 
variants. Nucleic Acids Res 2012; 40: D930-4. doi: 10.1093/nar/gkr917
38. The Genotype-Tissue Expression (GTEx) project. Nat Genet 2013; 45: 580-5. 
doi: 10.1038/ng.2653
39. Machiela MJ, Chanock SJ. LDlink: a web-based application for exploring 
population-specific haplotype structure and linking correlated alleles of 
possible functional variants. Bioinformatics 2015; 31: 3555-7. doi: 10.1093/
bioinformatics/btv402
40. Tregouet DA, Garelle V. A new JAVA interface implementation of THESIAS: 
testing haplotype effects in association studies. Bioinformatics 2007; 23: 
1038-9. doi: 10.1093/bioinformatics/btm058
41. Lehnert M, Weber DG, Taeger D, Raiko I, Kollmeier J, Stephan-Falkenau S, 
et al. Determinants of plasma calretinin in patients with malignant pleural 
mesothelioma. BMC Res Notes 2020; 13: 359. doi: 10.1186/s13104-020-
05187-y
42. Li L, Wang S, Zhang Y, Pan J. The E2F transcription factor 2: what do we 
know? Biosci Trends 2021; 15: 83-92. doi: 10.5582/bst.2021.01072
43. Guo AY, Zhai K, Xu JL, Hu JL, Gao L. Identification of a low-frequency mis-
sense variant in E2F transcription factor 7 associated with colorectal cancer 
risk in a chinese population. Asian Pac J Cancer Prev 2017; 18: 271-5. doi: 
10.22034/apjcp.2017.18.1.271
44. Chen J, Etzel CJ, Amos CI, Zhang Q, Viscofsky N, Lindor NM, et al. Genetic 
variants in the cell cycle control pathways contribute to early onset colorec-
tal cancer in Lynch syndrome. Cancer Causes Control 2009; 20: 1769-77. doi: 
10.1007/s10552-009-9416-x
45.	 Gołąbek	 K,	 Biernacki	 K,	 Gaź dzick a	 J ,	 Strz elczyk	 JK,	 Miśkiewicz-Or czyk	 K, 	
K r ak owczyk	 Ł,	 et	 al.	 Selected	E2F2 polymorphisms in oral and oropharyn-
geal squamous cell carcinoma. BioMed Res Int 2021; 2021: 8098130. doi: 
10.1155/2021/8098130
46. Lu M, Liu Z, Yu H, Wang LE, Li G, Sturgis EM, et al. Combined effects of E2F1 
and E2F2 polymorphisms on risk and early onset of squamous cell carci-
noma of the head and neck. Mol Carcinog 2012; 51(Suppl 1): E132-41. doi: 
10.1002/mc.21882
47. Cunningham JM, Vierkant RA, Sellers TA, Phelan C, Rider DN, Liebow M, et 
al. Cell cycle genes and ovarian cancer susceptibility: a tagSNP analysis. Br J 
Cancer 2009; 101: 1461-8. doi: 10.1038/sj.bjc.6605284
48. Li Y , Sturgis EM, Zhu L, Cao X, Wei Q, Zhang H, et al. E2F transcription factor 
2 variants as predictive biomarkers for recurrence risk in patients with squa-
mous cell carcinoma of the oropharynx. Mol Carcinog 2017; 56: 1335-43. 
doi: 10.1002/mc.22595
49. Scarpulla RC. Nuclear control of respiratory chain expression by nuclear 
respiratory factors and PGC-1-related coactivator. Ann N Y Acad Sci 2008; 
1147: 321-34. doi: 10.1196/annals.1427.006
50. Bhawe K, Roy D. Interplay between NRF1, E2F4 and MYC transcription fac-
tors regulating common target genes contributes to cancer development 
and progression. Cell Oncol (Dordr) 2018; 41: 465-84. doi: 10.1007/s13402-
018-0395-3
51. Crocco P , Montesanto A, Passarino G, Rose G. Polymorphisms falling within 
putative miRNA target sites in the 3’UTR region of SIRT2 and DRD2 genes 
are correlated with human longevity. J Gerontol A Biol Sci Med Sci 2016; 71: 
586-92. doi: 10.1093/gerona/glv058
52. Qu L, He B, Pan Y, Xu Y, Zhu C, Tang Z, et al. Association between poly-
morphisms in RAPGEF1, TP53, NRF1 and type 2 diabetes in Chinese Han 
population. Diabetes Res Clin Pract 2011; 91: 171-6. doi: 10.1016/j.dia-
bres.2010.11.019
53. Liu Y, Niu N, Zhu X, Du T, Wang X, Chen D, et al. Genetic variation and as-
sociation analyses of the nuclear respiratory factor 1 (nRF1) gene in Chinese 
patients with type 2 diabetes. Diabetes 2008; 57: 777. doi: 10.2337/db07-
0008

Radiol Oncol 2023; 57(4): 473-486.
Zupanc C et al. / Genetic factors and serum calretinin in asbestos diseases 486
54. Permuth-Wey J, Chen YA, Tsai YY , Chen Z, Qu X, Lancaster JM, et al. Inherited 
variants in mitochondrial biogenesis genes may influence epithelial ovar-
ian cancer risk. Cancer Epidemiol Biomarkers Prev 2011; 20: 1131-45. doi: 
10.1158/1055-9965.Epi-10-1224
55. Shen S, Wei Y, Li Y, Duan W, Dong X, Lin L, et al. A multi-omics study links 
TNS3 and SEPT7 to long-term former smoking NSCLC survival. NPJ Precis 
Oncol 2021; 5: 39. doi: 10.1038/s41698-021-00182-3
56. Wang X, Fei F, Qu J, Li C, Li Y , Zhang S. The role of septin 7 in physiology and 
pathological disease: a systematic review of current status. J Cell Mol Med 
2018; 22: 3298-307. doi: 10.1111/jcmm.13623
57. Zhu C, Hu H, Li J, Wang J, Wang K, Sun J. Identification of key differentially 
expressed genes and gene mutations in breast ductal carcinoma in situ us-
ing RNA-seq analysis. World J Surg Oncol 2020; 18: 52. doi: 10.1186/s12957-
020-01820-z
58. Ye L, Wang F, Wu H, Yang H, Yang Y, Ma Y, et al. Functions and targets of 
miR-335 in cancer. Onco Targets Ther 2021; 14: 3335-49. doi: 10.2147/ott.
S305098
59. Yang R, Dick M, Marme F, Schneeweiss A, Langheinz A, Hemminki K, et 
al. Genetic variants within miR-126 and miR-335 are not associated with 
breast cancer risk. Breast Cancer Res Treat 2011; 127: 549-54. doi: 10.1007/
s10549-010-1244-x