Construction of a robust prognostic model for adult adrenocortical carcinoma: Results from bioinformatics and real- world data
Xi Tian1,2 iD Wen-Hao Xu1,2 Aihetaimujiang Anwaier1, 1,2 Hong-Kai Wang1,2
Fang-Ning Wan1,2 Da-Long Cao1,2 I Wen-Jie Luo1,2 iD Guo-Hai Shi1,2
Yuan-Yuan Qu1,2 Hai-Liang Zhang1,2 | Ding-Wei Ye1,2
1Department of Urology, Fudan University Shanghai Cancer Center, Shanghai, China
2Department of Oncology, Shanghai Medical College, Fudan University, Shanghai, China
Correspondence
Yuan-Yuan Qu, Hai-Liang Zhang and Ding- Wei Ye, Department of Urology, Fudan University Shanghai Cancer Center, No. 270 Dong’an Road, Shanghai, 200032, China. Emails: quyy1987@163.com (Y .- Y. Q.); zhanghl918@163.com (H .- L. Z.); dwyelie@163.com (D .- W. Y.)
Funding information
This work is supported by Grants from National Key Research and Development Project (No.2019YFC1316000) and the National Natural Science Foundation of China (No.81802525).
Abstract
This study aims to construct a robust prognostic model for adult adrenocortical car- cinoma (ACC) by large-scale multiomics analysis and real-world data. The RPPA data, gene expression profiles and clinical information of adult ACC patients were obtained from The Cancer Proteome Atlas (TCPA), Gene Expression Omnibus (GEO) and The Cancer Genome Atlas (TCGA). Integrated prognosis-related proteins (IPRPs) model was constructed. Immunohistochemistry was used to validate the prognostic value of the IPRPs model in Fudan University Shanghai Cancer Center (FUSCC) cohort. 76 ACC cases from TCGA and 22 ACC cases from GSE10927 in NCBI’s GEO database with full data for clinical information and gene expression were utilized to validate the effectiveness of the IPRPs model. Higher FASN (P = . 039), FIBRONECTIN (P <. 001), TFRC (P <. 001), TSC1 (P <. 001) expression indicated significantly worse overall sur- vival for adult ACC patients. Risk assessment suggested significantly a strong predic- tive capacity of IPRPs model for poor overall survival (P < . 05). IPRPs model showed a little stronger ability for predicting prognosis than Ki-67 protein in FUSCC cohort (P = . 003, HR = 3.947; P = . 005, HR = 3.787). In external validation of IPRPs model using gene expression data, IPRPs model showed strong ability for predicting prog- nosis in TCGA cohort (P = . 005, HR = 3.061) and it exhibited best ability for predict- ing prognosis in GSE10927 cohort (P = . 0898, HR = 2.318). This research constructed IPRPs model for predicting adult ACC patients’ prognosis using proteomic data, gene expression data and real-world data and this prognostic model showed stronger pre- dictive value than other biomarkers (Ki-67, Beta-catenin, etc) in multi-cohorts.
KEYWORDS
adult adrenocortical carcinoma, biomarker, predictive model, proteomics, real-world data
Tian, Xu and Anwaier contribute equally to this work.
@ 2021 The Authors. Journal of Cellular and Molecular Medicine published by Foundation for Cellular and Molecular Medicine and John Wiley & Sons Ltd.
1 INTRODUCTION |
Adrenocortical carcinoma (ACC) is a rare and aggressive endo- crine malignancy with high risk of relapse, poor survival and lim- ited treatment options. The Surveillance, Epidemiology and End Results (SEER) database estimates that the annual incidence rate of ACC is approximately 0.72 per million cancer cases, resulting in 0.2% of all cancer deaths in the United States.1 However, ACC shows highly aggressive biological behaviour with less than 35% of patients surviving 5 years after initial diagnosis.2 Therefore, ap- propriate treatment is extremely important. The current preferred treatment of ACC is based on surgical resection of the primary tumour that is usually the first and most effective therapeutic strategy.3-5 Currently, there are very few drugs to treat this dis- ease and mitotane remains the only medication approved by the US Food and Drug Administration for ACC treatment.6 Thus, new treatment options and drug targets are urgently needed, espe- cially for clinical management of patients with ACC who are resis- tant to mitotane.
Proteomics is a powerful tool for detecting unknown protein species, exploring absolute quantified protein abundance, and iden- tifying biomarkers for pathogenic process.7,8 Proteomics has been used widely to explore biomarkers for various diseases9 and the latest developments in proteomics have made it possible to con- duct more comprehensive examinations of protein biomarkers in various cancers.10 For instance, Bouchal et al11 used transcriptome and proteomic analysis to identify potential biomarkers associated with metastatic breast cancer, and several proteomic studies have focused on identifying new diagnostic biomarkers in patients with prostate cancer.12,13 Thus far, very few studies have used a large- scale sequencing proteomic approach to identify potential protein biomarkers for ACC.14
Bioinformatics studies have generated large amounts of complex biological data through combinations of computer science, informa- tion technology and biology. For example, The Cancer Proteome Atlas (TCPA) database provides researchers with reverse-phase protein array (RPPA) data.15 The RPPA technique is a powerful pro- teomic approach for economical, sensitive and high-throughput evaluation of sizable numbers of selected protein markers, which made it possible to explore protein biomarkers using bioinformat- ics.16,17 Because there is a big difference between adult patients and child patients with ACC, in this study, we focused only on adult pa- tients. This study constitutes the first large-scale proteomic analysis combined with transcriptome data to describe the protein landscape of ACC in adult patients.
To explore novel protein biomarkers of potential prognostic value and develop a protein-derived predictive model in adult patients with ACC, we analysed the survival of proteins and constructed an integrated prognosis-related proteins model on risk assessment. Gene expression profiles also were analysed to reveal the under- lying biological interaction networks. The goal of this study was to provide potential novel therapeutic targets and a high performing prognostic predictive model for clinical management of adult ACC.
2 MATERIALS AND METHODS |
2.1 Data downloading and processing
The RPPA data (level 4) of adult ACC were obtained from The Cancer Proteome Atlas (TCPA). The gene expression profiles and clinical in- formation of patients with ACC were downloaded from The Cancer Genome Atlas (TCGA). Preprocessing and normalization of the raw biological data were performed using R software to remove noise and ensure the integrity of the data. By matching the sample IDs, we obtained 46 ACC cases with full data for clinical information, protein abundance and gene expression. We also obtained 76 ACC cases from TCGA (Table 1) and 22 ACC cases from GSE1092718 (Table 2) in NCBI’s GEO database with full data for clinical information and gene expression. All the cases were patients over 18 years old.
2.2 Survival analysis of candidate proteins
Kaplan-Meier analysis was performed based on the median protein abundance value and univariate Cox regression was used to evalu- ate the prognostic value of candidate proteins. For both statistics, P-values < . 05 were considered significant. The volcano plot was obtained using the ggplot2 package in R software.19 Red indicates negative association between protein abundance and survival, green indicates positive association between protein abundance and sur- vival, and black indicates no statistical significance. Survival curves were drawn using the survival package in R software. Red indicates high-risk group, and blue indicates low-risk group.20
2.3 Screening of candidate proteins and construction of a predictive multivariate Cox model
Lasso Cox regression was used to further narrow the proteins with prognostic significance using the glmnet package in R software.21 Multivariate analysis was performed using the Cox proportional haz- ards regression model to identify candidate proteins and evaluate the risk score based on candidate protein abundance and survival rates. An integrated prognosis-related proteins (IPRPs) model was then constructed (Risk score = 2.743 x fibronectin abundance (ref. Low) + 0.781 x FASN abundance (ref. Low) + 1.091 x TFRC abun- dance (ref. Low) + 3.043 x TSC1 abundance (ref. Low)). Median risk score of the predictive IPRPs model was used as the cut-off value and patients were classified into high-risk or low-risk groups.
2.4 | Assessing the prognostic significance of the IPRPs model in TCPA cohort
Besides the risk score of the IPRPs model for the patients with ACC, the covariables for the univariate and multivariate Cox regression models included age, gender, pTstage, pNstage, pMstage and pathologic stage.
| Characteristics | Entire cohort (N = 76) |
|---|---|
| N (%) | |
| Age | |
| < 70 years | 73(96.1) |
| ≥ 70 years | 3 (3.9) |
| Gender | |
| Male | 30 (39.5) |
| Female | 46 (60.5) |
| Laterality | |
| Left | 42 (55.3) |
| Right | 34 (44.7) |
| Stage | |
| I-II | 45 (59.2) |
| III-IV | 29 (38.2) |
| Censored | 2 (2.6) |
| T stageª | |
| T1 - T2 | 48 (63.2) |
| T3 - T4 | 26 (34.2) |
| Censored | 2 (2.6) |
| N stageª | |
| N0 | 66 (86.8) |
| N1 | 8 (10.5) |
| Censored | 2 (2.6) |
| M stageª | |
| M0 | 59 (77.6) |
| M1 | 15 (19.7) |
| Censored | 2 (2.6) |
| Mitotic rate | |
| > 5/50 HPF | 39 (51.3) |
| ≤ 5/50 HPF | 28 (36.8) |
| Censored | 9 (11.8) |
| Weiss score | |
| ≤ 4 | 22 (28.9) |
| > 4 | 36 (47.4) |
| Censored | 18 (23.7) |
| Invasion of tumour capsule | |
| Present | 41 (53.9) |
| Absent | 29 (38.2) |
| Censored | 6 (7.9) |
| Necrosis | |
| Present | 40 (52.6) |
| Absent | 32 (42.1) |
| Censored | 4 (5.3) |
ªTNM scoring system: Tumour size, Lymph Nodes affected, Metastases. AJCC, American Joint Committee on Cancer.
| Characteristics | Entire cohort (N = 22) |
|---|---|
| N (%) | |
| Age | |
| <70 years | 20 (90.9) |
| ≥70 years | 2 (9.1) |
| Gender | |
| Male | 6 (27.3) |
| Female | 16 (72.7) |
| Laterality | |
| Left | 10 (45.5) |
| Right | 9 (40.9) |
| Unknown | 3 (13.6) |
| Stage | |
| I-II | 12 (54.5) |
| III-IV | 10 (45.5) |
| Characteristics | Entire cohort (N = 39) |
|---|---|
| N (%) | |
| Age | |
| <70 years | 34 (87.2) |
| ≥70 years | 5 (12.8) |
| Gender | |
| Male | 19 (48.7) |
| Female | 20 (51.3) |
| AJCC stage | |
| I-II | 14 (35.9) |
| III-IV | 25 (64.1) |
| T Stageª | |
| T1 - T2 | 20 (51.3) |
| T3 - T4 | 19 (48.7) |
| N stageª | |
| N0 | 19 (48.7) |
| N1 | 20 (51.3) |
| M Stageª | |
| M0 | 21 (53.8) |
| M1 | 18 (46.2) |
| Necrosis | |
| Present | 25 (64.1) |
| Absent | 14 (35.9) |
ªTNM scoring system: Tumour size, Lymph Nodes affected, Metastases. AJCC, American Joint Committee on Cancer.
23
22
22
23
23
23
23
19
18
15
11
9
9
8
B
3
A
TFRC
B
4
8
P27_pT157
FIBRONECTIN
Partial Likelihood Deviance
40
ERK2
CYCLINB1
30
CMET
JAB1
PKCALPHA_pS657
3
PRDX1
TSC1
2
ERALPHA_pS118
NRAS
SNAIL
SMAC
PARP1
TAZ
-log10(pvalue)
PKCDELTA_pS664
1
ARAF_pS299
BCL2A1
Sig
P70S6K_pT389
CD31
XBP1
ECADHERIN
RBM15
CASPASE3
High risk
-8
-7
-6
-5
-4
-3
-2
-1
Log(2)
2
CIAP
EPPK1
CABL
Low risk
23
22
23
21
18
10
7
0
RAB25
FASN
DVL3
BID
· Not
C
X1433BETA
CI
INPP4B
YAP_pS127
RAD50
19
5
P53
PARPCLEAVED
PEA15
JAK2
M
ERALPHA
IGF1R_pY1135Y1136
0
1
Coefficients
A
1
3
2
-100
12
-20
-10
0
10
log2(HR)
-8
-7
-6
-5
-4
-3
-2
-1
Log Lambda
A receiver operating characteristic (ROC) curve was constructed to an- alyse the diagnostic accuracy of the logistic model and the area under curve (AUC) was calculated. Co-abundance analysis was performed using Pearson’s test to identify proteins associated with the logistic model with 0.4 set as the correlation coefficient cut-off value. Survival curves and a scatter diagram were used to explore the correlation be- tween risk score and patient’s prognosis, and a heat map of candidate protein abundance in the high-risk and low-risk groups was drawn.
2.5 | Validation of the IPRPs model in a cohort from the Fudan University Shanghai Cancer Center (FUSCC) in China
Real-world data were collected to validate the prognostic value of the IPRPs model. The cohort included 39 adult patients with ACC (Table 3) from the FUSCC between 2013 and 2019, and tumour specimens were obtained with informed consent. Anti-Ki67 (ab16667, Abcam, USA) anti- fatty acid synthase (ab128870, Abcam, USA), anti-fibronectin (ab2413, Abcam, USA), anti-TSC1 (ab217328, Abcam, USA), and anti-transferrin receptor (ab84036, Abcam, USA) antibodies were used to detect the abundance of the corresponding proteins by immunohistochemistry (IHC). Positive or negative staining of a certain protein in one FFPE slide was independently assessed by two experienced pathologists and
determined as follows. The staining intensity level was graded from 0 to 3. Samples with no staining, weak, median and strong staining denote to the level of 0, 1, 2 and 3. Based on the coverage percentage of immuno- reactive tumour cells (0%, 1-25%, 26-50%, 51-75%, 76-100%), the stain- ing extent was ranging from 0 to 4. The overall IHC score grading from 0 to 12 was evaluated according to the multiply of the staining intensity and extent score. Negative staining represented grade 0 to 3 and posi- tive staining from 4 to 12 for each sample. Risk score of each patient was calculated using the formula generated by the IPRPs model. The Kaplan-Meier method was applied to validate the prognostic value of the model, and the median of the risk score was set as the cut-off value.
2.6 Comparing the IPRPs model with other biomarkers using gene expression data
The number of patients with proteomic data was low, and therefore we used the gene expression data for the prognostic validation. The IPRPs model was compared with other biomarkers in the TCGA co- hort (76 cases) and GSE10927 (22 cases). Survival analyses were carried out using the Kaplan-Meier method and median of gene ex- pression was set as the cut-off value. AUC, C-index and net reclassi- fication improvement (NRI) were calculated to compare IPRPs model with other biomarkers.
| Protein | P value (KM) | P value (unicox) | HR |
|---|---|---|---|
| P27_pT157 | .004 | .000 | 0 (0-0.001) |
| ERALPHA_pS118 | .007 | .001 | 0 (0-0.007) |
| NRAS | .002 | .002 | 0 (0-0.041) |
| X1433BETA | .018 | .022 | 0.001 (0-0.34) |
| CMET | .001 | .001 | 0.003 (0-0.077) |
| SNAIL | .020 | .002 | 0.009 (0-0.183) |
| CD31 | .039 | .004 | 0.01 (0-0.23) |
| ARAF_pS299 | .001 | .003 | 0.02 (0.002-0.254) |
| PRDX1 | .004 | .002 | 0.021 (0.002-0.238) |
| XBP1 | .006 | .003 | 0.026 (0.002-0.291) |
| P70S6K_pT389 | .003 | .004 | 0.027 (0.002-0.311) |
| PKCDELTA pS664 | .023 | .003 | 0.036 (0.004-0.324) |
| PARPCLEAVED | .044 | .034 | 0.039 (0.002-0.781) |
| CIAP | .007 | .010 | 0.055 (0.006-0.498) |
| P53 | .012 | .025 | 0.071 (0.007-0.713) |
| RAB25 | .006 | .010 | 0.075 (0.01-0.542) |
| JAB1 | .007 | .001 | 0.08 (0.018-0.347) |
| ECADHERIN | .006 | .004 | 0.133 (0.033-0.527) |
| INPP4B | .002 | .019 | 0.137 (0.026-0.722) |
| ERALPHA | .004 | .050 | 0.164 (0.027-0.999) |
| SMAC | .003 | .002 | 0.188 (0.065-0.543) |
| PKCALPHA pS657 | .024 | .001 | 0.281 (0.133-0.594) |
| EPPK1 | .027 | .011 | 0.343 (0.151-0.779) |
| FASN | .039 | .015 | 1.943 (1.139-3.316) |
| CYCLINB1 | .000 | .001 | 2.198 (1.407-3.434) |
| PEA15 | .028 | .033 | 2.209 (1.066-4.577) |
| TFRC | .001 | .000 | 2.771 (1.681-4.567) |
| YAP_pS127 | .014 | .030 | 2.984 (1.109-8.031) |
| PARP1 | .002 | .002 | 3.23 (1.553-6.72) |
| ERK2 | .000 | .000 | 5.629 (2.168-14.618) |
| RBM15 | .008 | .008 | 6.531 (1.64-26.011) |
| FIBRONECTIN | .001 | .000 | 7.007 (2.517-19.511) |
| DVL3 | .002 | .010 | 10.978 (1.754-68.72) |
| IGF1R_ pY1135Y1136 | .012 | .047 | 14.431 (1.033-201.583) |
| TSC1 | .001 | .001 | 16.639 (2.965-93.39) |
| JAK2 | .042 | .043 | 18.739 (1.098-319.658) |
| BCL2A1 | .003 | .005 | 20.533 (2.432-173.352) |
| CASPASE3 | .042 | .006 | 51.697 (3.028-882.701) |
| CABL | .007 | .008 | 68.124 (2.962-1566.613) |
| BID | .038 | .010 | 77.659 (2.787-2164.284) |
(Continues)
| Protein | P value (KM) | P value (unicox) | HR |
|---|---|---|---|
| RAD50 | .005 | .019 | 158.062 (2.325-10744.279) |
| TAZ | .013 | .003 | 768.971 (10.143-58295.167) |
2.7 Gene set enrichment analysis (GSEA)
To explore potential associated signal pathways, the TCGA data- sets of the high-risk and low-risk groups (according to risk score of the IPRPs model) were analysed using the GSEA software (version 3.0) with the number of permutations set to 1000. False discovery- adjusted P-values were obtained using the Benjamini and Hochberg method.22 Significant differential expression was defined as an ad- justed P-value of < . 01 and a false discovery rate of < 0.25.
2.8 | Identification of differentially expressed genes (DEGs) related to risk score of the IPRPs model
The DEGs (adjusted P-value < 0.01; fold change at least 2x) be- tween the high-risk and low-risk groups were identified using the Limma package.23 A heat map was drawn according to the expres- sion matrix of the samples to show the differences in gene expres- sion between the two groups. The Search Tool for the Retrieval of Interacting Genes (STRING; http://string-db.org) (version 10.0)24 online database was used to predict protein-protein interaction (PPI) networks of the DEGs. Cytoscape (version 3.5)25 is an open-source bioinformatics software platform for visualizing molecular interac- tion networks. We used MCODE (version 1.4.2),26 a Cytoscape plug- in, to find the most significant hub genes with MCODE Score ≥ 20. Functional enrichment analysis of the hub genes was completed using the ClusterProfiler package.27
|
3 RESULTS
In this work, we aimed to explore new prognostic biomarkers for adult patients with ACC using proteomics and transcriptomics data. A flow chart of the methods used in this study is given in Figure S1.
3.1 | Selection for candidate proteins with significant prognostic value
From the volcano plot (Figure 1A), 42 candidate protein biomarkers with P-values < 0.05 in both the Kaplan-Meier analysis and univari- ate Cox regression analysis were selected and are listed in Table 4. The Lasso Cox regression results for the selected proteins are shown in Figure 1B, C.
| age | 0.236 | 1.025(0.984-1.067) |
| gender | 0.111 | 0.351(0.097-1.271) |
| stage | <0.001 | 3.700(1.833-7.466) |
| T stage | <0.001 | 3.823(1.889-7.736) |
| M stage | 0.001 | 6.856(2.152-21.841) |
| N stage | 0.149 | 2.639(0.706-9.867) |
| riskScore | 0.006 | 1.015(1.004-1.025) |
A
C
pvalue
Hazard ratio
1.0
C-index=0.939
0.8
95%CI:0.916-0.962
True positive rate
NRI =0.235
0.6
95%CI:0-0.597
0.4
0.2
AUC=0.933
P<0.05
0
5
10
15
20
0.0
Hazard ratio
D
0.0
0.2
0.4
0.6
0.8
1.0
B
False positive rate
pvalue
Hazard ratio
FRAN
age
0.295
1.027(0.977-1.080)
0.609(0.135-2.748)
3
gender
0.519
stage
0.216
4.026(0.443-36.579)
PIRRONECTIN
[RALPHA_dit:18
T stage
0.385
1.667(0.526-5.283)
M stage
0.329
0.264(0.018-3.829)
TFRC
MALAS
127_p157
N stage
0.919
1.099(0.177-6.818)
riskScore
0.018
1.009(1.002-1.017)
TSC1
0
5
10
15
20
25
30
35
Hazard ratio
3.2 Construction of the IPRPs model
In the univariate Cox regression analysis (Figure 2A), the pathologi- cal stage (P < . 001), pTstage (P < . 001), pMstage (P = . 001) and risk score of the IPRPs model (P < . 01) were associated with shorter overall survival. However, in the multivariate Cox regression analy- sis, only risk score (P < . 05) was significantly correlated with worse outcome (Figure 2B). C-index (0.939, 95% CI:0.916-0.962) and NRI (0.235, 95% CI:0-0.597) indicated that our model is stable. These results indicate that our IPRPs model has independent prognostic significance. The risk score with AUC of 0.933 indicates the diag- nostic accuracy and consistent predictive ability of our IPRPs model (Figure 2C).
3.3 | Survival analysis of the IPRPs model in the TCPA cohort
Kaplan-Meier survival curves (Figure 3A) revealed that high abun- dances of fatty acid synthase (FASN) (P = . 039), fibronectin (FN) (P < . 001), transferrin receptor (TFRC) (P < . 001) and tuberous scle- rosis 1 (TSC1) (P < . 001) indicated a worse outcome. The formula used to predict overall survival was generated by multivariate Cox
regression models as integrated risk score = 2.743 x FN abundance (ref. Low) + 0.781 x FASN abundance (ref. Low) + 1.091 x TFRC abundance (ref. Low) + 3.043 x TSC1 abundance (ref. Low). The heat map shows that the abundances of FASN, FN, TFRC and TSC1 in the high-risk group were higher than they were in the low-risk group (Figure 3B). The survival time of the high-risk group was significantly shorter than that of the low-risk group (P < . 001), and the increased risk score corresponded to shorter survival (Figure 3C-E).
3.4 | Validation of the prognostic value of the IPRPs model in the FUSCC cohort
Representative IHC plots for the ACC samples are displayed in Figure 4A-E (Abundances of A: Ki-67, B: Fatty acid synthase (FASN), C: Fibronectin (FN), D: Tuberous sclerosis 1 (TSC1) and E: Transferrin receptor (TFRC)). The Ki-67 protein abundance and high-risk (HR) score (Figure 4F, G) were both significantly correlated with worse outcome for patients in the FUSCC cohort (P = . 005, HR = 3.787; P = . 003, HR = 3.947). The IPRPs model predicted the prognosis better than the Ki-67 protein in the FUSCC cohort. A high-risk score was significantly correlated with higher Stage, T stage and N stage (Figure 4H, I).
|
WILEY
A
a
FASIN level
high
low
b
FIBRONECTIN level
high
low
B
type
1.00
1.00
Survival probability
0.75
Survival probability
0.75
FIBRONECTIN
0.50
0.50
type
high
0.25
p=3.8946-02
2.25
P=9.882e-04
low
0.00
0.00
0
1
2
8
9
10
12
FASN
y
+
5
11
A
~
3
m
¥
00
0
1
Time(years)
FASN level
IBRONECTIN level
0
S
S
12
Q
Time(years)
2
53
23
22
15
14
10
23
1
1
1
20
15
1
A
0
0
& 3
23 23
22
16
13
0
4
3
NO
S
1
23
10
5
B
23
2
2
N
0
0
A
3
2
3
4
6
6
7
8
9
10
11
12
0
1
2
3
A
3
6
9
10
11
12
0
Time(years)
Time(years)
C
TFRC level
high
low
d
TSC1 level
hig
low
TFRC
-1
1.00-
1.00-
-2
-3
Survival probability
0.75
Survival probability
0,75
0.50
0.50
TSC1
0.25
p=7.659-04
.25
p=9.532e-04
0.00
0.00
1
TCGA-PA-ASYG
TCGA-OR-A5LH
TCGA-OR-A5JR
TCGA-OR-A5J6
TCGA-OR-A5LK
TCGA-OR-A5JT
TCGA-OR-A5LT
TCGA-PK-A5HA
TCGA-OR-A5JW
TCGA-OR-A5K1
TCGA-OR-A5LP
TCGA-OR-A5LN
TCGA-OR-A5LM
TCGA-OR-A5JZ
TCGA-OR-A5K8
TCGA-OR-A5KU
TCGA-OR-A5J3
TCGA-OR-A5LS
TCGA-OR-A5K3
TCGA-OR-A5JV
TCGA-OR-A5J9
TCGA-PK-A5H8
TCGA-OR-A5K4
TCGA-OR-A5LG
TCGA-OR-A5KO
TCGA-OR-A5KX
TCGA-OR-A5LO
TCGA-OR-A5LL
TCGA-OR-A5KW
0
TCGA-PK-A5H9
TCGA-OR-A5J2
TCGA-OU-A5PI
TCGA-OR-A5JS
TCGA-OR-A5J7
TCGA-OR-A5K6
TCGA-OR-A5JP
TCGA-OR-A5KO
TCGA-OR-A5LD
TCGA-OR-A5JA
TCGA-OR-A5LJ
TCGA-OR-A5KY
TCGA-OR-A5J8
TCGA-OR-A5KZ
TCGA-OR-ASJY
TCGA-OR-A5K5
TCGA-PB-A5OG
4
5
6
1
8
9
10
11
Time(years)
12
0
1
4
S
6
Y
S
11
Time(years)
10
12
TFRC level
TSC1 level
5
23
23
O
6
A
N
e
0
2
15
13
0
0
10
1
9
1
0
1
1
SÌ
0
1
/
2
?
1
0
0
23
20
K
.
N
14
10
1
1
2
1
1
0
f
2
3
4
E
6
1
8
9
10
11
12
0
1
2
3
4
5
S
B
9
10
Time(years)
11
12
Time(years)
C
Risk
High risk
Low risk
D
1.00
10
J
High risk
8
low Risk
Survival probability
Risk score
0.75
4
*
0.50
~
0
·
0.25
0
10
20
30
40
p=3.085e-06
E
Patients (increasing risk socre)
0.00
Survival time (years)
~
Dead
Alive
0
1
2
3
4
5
6
7
8
9
10
11
12
0
Time(years)
8
0
៛
Risk
High risk
23 23
22
13 22
11
6
Low risk
23
2
1
0
05
18
05
03
o-
01
~
14
14
10
7
0
0
1
2
3
4
5
6
7
8
9
10
11
12
0
10
20
30
40
Time(years)
Patients (increasing risk socre)
3.5 External validation of the IPRPs model and comparison with other biomarkers using gene expression data
In the TCGA cohort (Figure 5A-F), the IPRPs model showed stronger ability for predicting prognosis than the expression levels of CTNNB1 (beta-catenin gene), IGF2 and TP53 (P = . 005, HR = 3.061; P = . 012, HR = 2.768; P = . 162, HR = 0.574; P = . 033, HR = 2.336), whereas the MKI67 (Ki-67 protein gene) and NR5A1 (SF-1 protein gene) ex- pression levels had stronger predictive ability than the IPRPs model (P < . 0001, HR = 9.238; P = . 003, HR = 4.084). In the GSE10927 cohort (Figure 5G-K), which lacked IGF2 expression data, the IPRPs model showed better ability for predicting prognosis than the ex- pression levels of TP53, CTNNB1, NR5A1 and MKI67 (P = . 0898, HR = 2.318; P = . 73, HR = 1.187; P = . 16, HR = 1.983; P = . 36, HR = 1.57; P = . 22, HR = 1.824). AUC, C-index and NRI of various biomarkers were listed in Table 5, and it indicated that IPRPs model may act better than other biomarkers in RPPA data and IHC.
3.6 Significantly involved pathways of the IPRPs
The top 100 genes that were most significant positively and neg- atively correlated with the risk score are depicted in a heat map (Figure 6A). Besides an ACC progressive phenotype, the GSEA in- dicated that significant alteration of the IPRPs model involved chro- mosome separation, metaphase-anaphase transition of the cell cycle and protein modification by small protein removal. Hub genes with prognostic implications associated with the IPRPs were involved mainly in regulation of cell-cycle pathways (Figure 6B-D).
3.7 Identification of DEGs associated with the IPRPs
A significant difference was detected between the gene expression in high-risk and low-risk groups as shown in the heat map (Figure 7A). A PPI network of the DEGs was constructed and the identified hub
|
A
Anti-Ki67
B
Anti-FASN
C
Anti-FIBRONECTIN
D
Anti-TSC1
E
Anti-TFRC
F
FUSCC cohort
FUSCC cohort
Low Ki-67 expression
G
Overall survival (%)
Low risk score
100
Overall survival (%)
High Ki-67 expression
100
High risk score
n=19
n=19
50-
50-
n=20
n=20
0
p=0.005, HR(high)=3.7.87.
p=0.003, HR(high)=3.947
Time (years)
0
Time (years)
H
*, ANOVA P=0.02
I
80
80
**
**
ns
Risk score
60.
Risk score
60-
40-
40-
20
20
Stage I
Stage II
Stage III
Stage IV
T1-T2 T3-T4
NO
N1
MO
M1
genes were CENPM, NDC80, DLGAP5, SPC25, CENPF, ZWILCH, AURKB, CENPA, CDC20, CCNA2, KIF4A, BUB1B, CCNB2, UBE2C, AURKA, PLK1, CASC5, RANGAP1, BIRC5, CEP55, NEK2, SGOL2, KIF18A, CCNB1, SKA1, RRM2, ASPM, SGOL1, KIF2C, CDCA8, CENPI, KIF11, BUB1, CDCA5, CDK1, SPC24, SPAG5 and NUF2 (Figure 7B,C). The functional enrich- ment analyses indicated the hub genes were enriched mainly in cell cycle, mitotic nuclear division, chromosome, centromeric region and microtubule binding (Table 6 and Figure 7D, E).
3.8 | Correlation analysis between the IPRPs and other potential signatures
Various types of proteins may be associated with the candidate proteins as shown in Figure 2D. The analysis detected 27 kinds of proteins (correlation coefficients from - 0.64 to 0.66, P < . 001) that were correlated with FN abundance; among them, BID abundance was highly positively correlated (correlation coefficient = 0.66) with FN abundance (Figure S2A). Twenty-one kinds of proteins (correla- tion coefficients from - 0.597 to 0.742) were correlated with TSC1 abundance; among them, PARP1 abundance was highly positively
correlated with TSC1 abundance (correlation coefficient = 0.742) (Figure S2B). Seven kinds of proteins (correlation coefficients from - 0.534 to 0.581) were correlated with FASN abundance; among them, EEF2 abundance was highly positively correlated with FASN abundance (correlation coefficient = 0.0.581) (Figure S2C). Four kinds of proteins (correlation coefficients from - 0.52 to 0.607) were correlated with the TFRC abundance; among them, CYCLINB1 abundance was highly positively correlated with TFRC abundance (correlation coefficient = 0.607) (Figure S2D).
4 DISCUSSION |
The prognosis of ACC is poor because most patients with ACC have locally advanced or metastatic diseases and cannot be treated by surgery. Approximately 66% of patients with localized diseases ex- perience recurrence and usually require systematic treatment.28,29 Although there are diagnostic and prognostic molecular detection methods for ACC, including IGF2, p53, and the Wnt/B-catenin and PI3K signalling pathways, they have not been well applied in mor- phological evaluation, auxiliary diagnosis, or prognostic modelling
|
A
TCGA cohort
B
TCGA cohort
C
TCGA cohort
Overall survival (%)
100-
Low-risk score
Overall survival (%)
- Low IGF2 expression
High-risk score
100
Low CTNNB1 expression
Overall survival (%)
- High CTNNB1 expresion
100
- High IGF2 expresion
n=38
n=38
n=38
50-
50.
50
n=38
n=38
0
p=0.005, HR(high)=3.061
p=0.012, HR(high)=2.768-
p= 0.162. HR(high)=0.574
Time (years)
0
Time (years)
0
Time (years)
D
TCGA cohort
E
TCGA cohort
F
TCGA cohort
Overall survival (%)
100
Low TP53 expression
Overall survival (%)
100
n=38
Low MKI67 expression
n=38
- High TP53 expression
High MKI67 expresion
Overall survival (%)
100
- Low NR5A1(SF-1) expression
n=38
- High NR5A1(SF-1) expresion
50-
n=38
50
50
n=38
n=38.
0
p=0.033, HR(high)=2.336
p< 0.0001, HR(high)=9:238.
p=0.003, HR(high)=4.084
Time (years)
0
Time (years)
0
Time (years)
G
GSE10927
H
GSE10927
I
GSE10927
Overall survival (%)
100-
Low-risk score
Overall survival (%)
100
Low MKI67 expression
Overall survival (%)
- Low NR5A1(SF-1) expression
- High-risk score
High MKI67 expresion
100-
High NR5A1(SF-1) expresion
n=11
n=11
50
n=11
50
50
n=11
n+11
n=11:
0
p=0.0898, HRYhigh
=2:318
Time (years)
0
p=0.22 — HR(high)=1,824
p=0.36; HR(high)=j.57
Time (years)
0
Time (years)
J
GSE10927
K
GSE10927
Overall survival (%)
100
Low TP53 expression
Overall survival (%)
High TP53 expression
100-
Low CTNNB1 expression
High CTNNB1 expresion
n=11:
n=11
50
50
n=11.
n=11
0
p=0.73. HAihigh)=1.187
Time (years)
0
p=0:16 :- HR/high):1,983
Time (years)
| Biomarker evaluation | AUC | C-index | NRI | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Cohorts | TCGA (RPPA) | TCGA (RNAseq) | GSE10927 (RNA seq) | FUSCC IHC) | TCGA (RPPA) | TCGA (RNAseq) | GSE10927 (RNAseq) | FUSCC (IHC) | TCGA (RPPA) | TCGA (RNAseq) | GSE10927 | FUSCC |
| (RNAseq) | (IHC) | |||||||||||
| CTNNB1 | 0.4 | 0.704 | 0.732 | - | 0.564 | 0.659 | 0.657 | - | ns | -0.02 | 0.06 | - |
| IGF2 | - | 0.313 | - | - | - | 0.575 | - | - | - | ns | - | - |
| TP53 | 0.489 | 0.87 | 0.571 | - | 0.676 | 0.659 | 0.551 | - | -0.02 | ns | ns | - |
| MKI67 | - | 0.862 | 0.786 | 0.621 | - | 0.853 | 0.671 | 0.694 | - | 0.37 | ns | ns |
| SF-1 | - | 0.71 | - | - | - | 0.665 | - | - | - | -0.07 | ns | - |
| Risk Score | 0.933 | 0.885 | 0.705 | 0.649 | 0.939 | 0.789 | 0.657 | 0.72 | - | - | - | - |
of ACC.30 Early diagnosis and appropriate treatment play key roles in the management of ACC; thus, effective biomarkers are urgently needed.31 Proteomics has unique advantages and our study is the first large-scale proteomic analysis of ACC with RPPA data. We found that FASN, FN, TFRC and TSC1 abundance levels were of high prognostic value. To explore the underlying biological mecha- nism, we performed a GSEA of high-risk and low-risk groups and the results indicated that the most significant pathways associated with candidate proteins included chromosome separation, metaphase- anaphase transition of cell cycle and protein modification by small protein removal. These pathways are worth further study.
FASN is a key enzyme in mammals that is needed for ab initio palmitic acid synthesis. In most normal non-adipose tissues, the abundance and activity of FASN are largely inhibited by adequate di- etary fat, but in many human cancers, FASN abundance and activity are abnormally increased and are associated with poor prognosis.32 The increased abundance of FASN potentially confers tumour cells an advantage in survival and growth.33 For instance, Ueda et al34 found that FASN expression promoted cell survival and growth of tumour cells in gestational trophoblastic neoplasms, and Nguyen et al35 found that increased intratumoral FASN expression led to more aggressive prostate cancers. In this study, we found that a high
A
B
TCGA-P6-A50G-014-11-A39K-20
TCGA-OR-A5K5-014-21-A39K-20
TCGA-OR-A5JY-01A-21-A39K-20
TCGA-OR-A5K7-01A-21-A39K-20
TCGA-OR-A5J8-014-21-A39K-20
TCGA-OR-A5KY-01A-21-A39K-20
TCGA-OR-A5LJ-01A-21-A39K-20
TCGA-OR-A5JA-01A-21-A39K-20
TCGA-OR-ASLD-01A-21-A39K-20
ICGA-OR-A5KO-01A-21-A39K-20
TCGA-OR-A5JP-01A-21-A39K-20
TCGA-OR-A5K6-01A-21-A39K-20
TCGA-OR-A5J7-01A-21-A39K-20
TCGA-OR-A5JS-01A-21-A39K-20
TCGA-OT-A5PT-01A-21-A39K-20
TCGA-OR-A5J2-01A-21-A39K-20
TCGA-PK-A5H9-01A-21-A39K-20
TCGA-OR-A5KT-01A-21-A39K-20
TCGA-OR-A5LL=01A-21-A39K-20
TCGA-OR-A5L0-01A-21-A39K-20
TCGA-OR-A5KY-01A-21-A39K-20
TCGA-OR-A5K0-014-21-A39K-20
TCGA-OR-ASLG-01A-21-A39K-20
TCGA-OR-A5K4-01A-21-A39K-20
TCGA-PK-A5M8-01A-21-A39K-20
TCGA-OR-A5J9-01A-21-A39K-20
TCGA-OR-A5TH-01A-21-A39K-20
TCGA-OR-A5K3-01A-21-A39K-20
ICGA-OR-ASLS-01A-21-A39K-20
ICGA-OR-A5J3-01A-21-A39K-20
ICGA-OR-A5KUI-01A-21-A39K-20
TCGA-OR-A5K8-01A-21-A39K-20
TCGA-OR-A5J7-01A-21-A39K-20
TCGA-OR-ASIM-01A-21-A39K-20
TCGA-OR-ASLY-01A-21-A39K-20
TCGA-OR-A5LP-01A-21-A39K-20
TCGA-OR-A5K1-01A-21-A39K-20
TCGA-OR-A5JT-01A-21-A39K-20
TCGA-PK-A5HA-01A-21-A39K-20
TCGA-OR-A5LT-01A-21-A39K-20
TCGA-OR-A5JT-014-21-A39K-20
TCGA-OR-A5LK-01A-21-A39K-20
TCGA-OR-A516-01A-41-A39K-20
TCGA-OR-A5 TR-01A-21-A39K-20
TCGA-OR-ASLH-01A-21-A39K-20
TCGA-PA-ASYG-014-21-439K-20
Enrichment plot: GO_CHROMOSOME_SEPARATION
0.6
Enrichment score (ES)
0.5
0.4
0.3
0.2
0.1
SampleName
0.0
HAIISS
KATMBL1
SE3B4
STX5
Ranked list metric (Signal2Noise)
AGAP1
JARID2
1.0
‘h’ (positively correlated)
R3HDM1
TYTS
0.5
TTS1
SAP130
0.0
LIM9
Zero cross at 13137
CHOT9
0.5
ZMF143
ADR74
-1.0
‘T (negatively correlated)
MARK3
RPLA
0
2,500
5,000
7,500
10,000
12,500
15,000
17,500
20,000
7-Mar
Rank in Ordered Dataset
TLKAP
SRPRA
Enrichment profile - Hits
Ranking metric scores
IS¥1
COPS7B
TAPS
C
MAFE
STK25
Enrichment plot: GO_METAPHASE_ANAPHASE_TRANSITION_OF_CELL_C YCLE
SPTAM1
CKAP2L
RIT1
ZIF 496
AMMECR1L
SIIV39H2
0.7
CCHIB2
SPC25
Enrichment score (ES)
0.6
AKTRTU2
0.5
PHESA
DDx50
0.4
HIST1H4T
0.3
HTRA2
KCTD20
0.2
MITFK
0.1
REM1 7
SIIRPD3
0.0
TET20
SART1
PRPF40A
Ranked list metric (Signal2Noise)
KUSTRIL
CCDC9OR
MICAPH
1.0
‘h’ (positively correlated)
CASC10
CIDSPL2
0.5
PPP2BSC
0.0
TIMES
Zero cross at 13137
GRTA3
0.5
TOUR
-1.0
TMEM232
‘I’ (negatively correlated)
CISA
0
2,500
5,000
7,500
10,000
12,500
15,000
17,500
20,000
PALM
SSBP2
Rank in Ordered Dataset
CYSTM1
EMPP4
Enrichment profile - Hits
Ranking metric scores
ARAT
AMIKEF1
SPATA4
D
CASC1
GLIM2
Enrichment plot: GO_PROTEIN_MODIFICATION_BY_SMALL_PROTEIN_RE MOVAL
PAKS
AMIE3858
CYP27A1
COXZR
TTC25
PRKG1
0.5
STAC
SERPTIT1
Enrichment score (ES)
0.4
SMTM1012B
PIFO
0.3
GIGZ
PRLR
EMPPS
0.2
ISCIL
TIGD4
0.1
CPME4
HS6ST3
0.0
PTPRA
DMAT1
TSC2204
Ranked list metric (Signal2Noise)
ATFM1
TAK3
PI3K
1.0
‘h’ (positively correlated)
AMY2R
MSRA
0.5
SYCP3
0.0
SPEF1
Zero cross at 13137
MAGED2
0.5
LACTE
-1.0
ECHTH2
‘T (negatively correlated)
ICEP2L
0
2,500
5,000
7,500
10,000
12,500
15,000
17,500
20,000
APSS1
IMT12
Rank in Ordered Dataset
SPAG17
ZDHHC17
Enrichment profile - Hits
Ranking metric scores
PYROXD1
A
Type
15
Type
high
B
C
10
low
5
0
CDCA5
-5
CENPI
SPC25
SGOL2
RANGAP1
CENPM
CENPA
SKA1
ZWILCH
BUB1
NUF2
BUB1B
CASC5
CDCA8
NDC80
SPC24
BIRC5
PLK1
SGOL1
CCNA2
CCNB1
CENPF
CDC20
CDK1
KIF18A
CEP55
KIF2C
KIF4A
CCNB2
AURKB
RRM2
AURKA
DLGAP5
KIF11
ASPM
SPAG5
UBE2C
NEK2
mitotic nuclear division
TCGA-OR-ASK4-01A-21-A39K-20
TCGA-PK-ASH8-01A-21-A39K-20
TCGA-OR-A519-01A-21-A39K-20
TCGA-OR-A5JV-01A-21-A39K-20
BAJOR-A5K3-01A-21-A39K-20
TCGA-OR-ASLS-01A-21-A39K-20
TCGA-OR-A5J3-01A-21-A39K-20
TCGA-OR-A5KB-01A-21-A39K-20
TOGA-OH NABOK-20
TCGA-OR-A5IZ-01A-21-A39K-20
PRIM-018-21-A19K-20 TCGA-OR-ASLN-01A-21-A39K-20
TCGA-OR-ASLP-01A-21-A39K-20
TCGA-OR-A5JW-01A-21-A39K-20
MORAM04-014-21-A39K-20
TCGA-PK-A5HA-01A-21-A39K-20
ePAOP-ALT-01A-21-A39K-20
TOGA-OR-ASJT-01A-21-A38K-20
TCGA-OR-A5LK-01A-21-A39K-20
TCGA-OR-A5J5-01A-41-A39K-20
TCGA-OR-A5JR-01A-21-A39K-20
TCGA-OR-A5LH-01A-21-A39K-20
TCGA-PA-ASYG-01A-21-A39K-20
TCGA-P6-ASOG-01A-11-A39K-20
NON 005-01A-21-A39K-20
TCGA-OR-A5JY-01A-21-A39K-20
TCGA-OR-A5KZ-01A-21-A39K-20
TOGA-OR-ASKY-01A-21-A39K-20
TCGA-OR-ASLJ-01A-21-A39K-20
TOGA-OR-A5LD-01A-21-A39K-20
TCGA-OR-45JA-014-21-A39K-20
TCGA-OR-A5KO-01A-21-A39K-20
TCGA-OR-ASJP-01A-21-A39K-20
TCGA-OR-ASK6-01A-21-A39K-20
TCGA-OR-A5J7-01A-21-A39K-20
TCGA-OR-A5J8-01A-21-A39K-20
TCGA-OU-A5PI-01A-21-A39K-20
TCGA-OR-A5.12-01A-21-A39K-20
TCGA-PK-ASH9-01A-21-A39K-20
TOGA-OR-A5KW-01A-21-A39K-20
TOUR OR POLY-04-21-A39K-20 TCGA-OR-A5LO-01A-21-A39K-20
TCGA-OR-A5KX-01A-21-A39K-20
TCGA-OR-ASLG-01A-21-A39K-20
TOGA-OR-A5K0-01A-21-A3BK-20
E
nuclear division
organelle fission-
mitotic sister chromatid segregation
chromosome segregation-
sister chromatid segregation
4
nuclear chromosome segregation-
regulation of chromosome segregation-
regulation of mitotic nuclear division-
D
spindle organization
Cell cycle
chromosome, centromeric region
Progesterone-mediated oocyte maturation
kinetochore
Oocyte meiosis
chromosomal region
p.adjust
p53 signaling pathway
Cellular senescence
condensed chromosome, centromeric region-
Human T-cell leukemia virus 1 infection
condensed chromosome
0.0025
FoxO signaling pathway
condensed chromosome kinetochore
8
0.0050
Viral carcinogenesis
Human immunodeficiency virus 1 infection-
spindle
p.adjust
condensed nuclear chromosome, centromeric region
0.0075
Ubiquitin mediated proteolysis-
Hepatitis B
Apoptosis - multiple species
midbody
Glutathione metabolism-
0.1
condensed chromosome outer kinetochore
Pyrimidine metabolism-
Acute myeloid leukemia
Platinum drug resistance
0.2
microtubule binding
histone kinase activity
Drug metabolism - other enzymes
Colorectal cancer
0.3
tubulin binding-
Gap junction
protein serine/threonine kinase activity
AMPK signaling pathway
microtubule motor activity
Purine metabolism
Apoptosis
cyclin-dependent protein serine/threonine kinase regulator activity-
휴
Hippo signaling pathway
motor activity
RNA transport
Transcriptional misregulation in cancer-
microtubule plus-end binding
Epstein-Barr virus infection
ATP-dependent microtubule motor activity, plus-end-directed-
0
2
4
6
8
cyclin-dependent protein kinase activity
0
5
10
15
20
25
abundance of FASN also was significantly correlated with worse prognosis of ACC. Previous studies have established the anti-tumour effects of the first-generation FASN inhibitors.36,37 Thus, FASN may be a potential therapeutic target in ACC.
FN is a large extracellular matrix protein in bones, which can combine with itself and collagen to form a network.38 Studies have shown that the abundance of FN in breast cancer is higher than in normal tissues and FN abundance is significantly related to the in- vasiveness of the disease.39 Knowles et al40 found that FN matrix formation was associated with kidney tumour cell spreading. Besides the prognostic value of FN in ACC, we also found that the abun- dances of JAB1, SCD1 and PRDX1 were negatively correlated with FN abundance, whereas the abundances of HEREGULIN, TIGAR and BID were positively correlated with FN abundance. Thus, FN is also a candidate target for new therapeutic drugs.
Iron is a basic trace element involved in cell metabolism, division and proliferation, and iron also has been considered as an important factor in the development of cancer.41 TFRC is a cell surface receptor that is responsible for transferrin-mediated iron uptake; thus, TFRC may play a key role in the energy supply of cancer cells.42 Shpyleva
et al43 found a high abundance of TFRC in breast cancer, and TFRC antibodies have been used to inhibit tumour growth. 44 We found mu- tual inhibition between TFRC and SMAC, and that the abundances of X1433ZETA, ERK2 and CYCLINB1 were positively correlated with TFRC abundance. Modulation of PPIs is a promising new idea in drug development45,46; thus, the design of TFRC inhibitors based on the interaction modes may create new therapeutic drugs.
TSC1, in a complex with tuberous sclerosis 2, inhibits the nutrient- mediated or growth factor-stimulated phosphorylation of S6K1 and EIF4EBP1 by negative regulation of mTORC1 signal transduction.47,48 We also found interactions between various types of proteins and TSC1. Among them, PARP1 abundance showed the highest correla- tion with TSC1 (correlation coefficient = 0.742) and it attracted our attention because of its key role in DNA repair.49 Maintaining the integrity of the genome is the basis of cell survival, and PARP inhibi- tors kill tumours mainly by inhibiting DNA repair and destroying the genomes of tumour cells.50 Inhibiting PARP1 also may inhibit TSC1, suggesting a potential strategy for the treatment of ACC.
Moon et al also established a model for predicting prognosis using RPPA data. They focused on the patients with distant metastasis. But
| Term | Description | Count in gene set | P value |
|---|---|---|---|
| GO:0 140 014 | Mitotic nuclear division | 21 | 1.26E-30 |
| GO:0 000 280 | Nuclear division | 22 | 1.81E-28 |
| GO:0 048 285 | Organelle fission | 22 | 1.61E-27 |
| GO:0 000 775 | Chromosome, centromeric region | 24 | 5.20E-41 |
| GO:0 000 776 | Kinetochore | 21 | 1.48E-37 |
| GO:0 098 687 | Chromosomal region | 25 | 1.04E-36 |
| GO:0 008 017 | Microtubule binding | 8 | 1.65E-08 |
| GO:0 035 173 | Histone kinase activity | 4 | 2.65E-08 |
| GO:0 015 631 | Tubulin binding | 8 | 1.83E-07 |
| hsa04110 | Cell cycle | 8 | 7.26E-12 |
| hsa04914 | Progesterone- mediated oocyte maturation | 7 | 1.13E-10 |
| hsa04114 | Oocyte meiosis | 7 | 7.02E-10 |
Note: Abbreviations: DEGs, differentially expressed genes; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes.
the C-index of their model (maximum:0.86) is much lower than ours (0.939). Guo J et al identified 9 hub genes (CCNB1, CDK1, TOP2A, CCNA2, CDKN3, MAD2L1, RACGAP1, BUB1 and CCNB2) with prog- nostic value. But the data they used were different from our research. They only focused on gene expression data, which is usually consid- ered unstable than protein data. And they just identified 9 hub genes with prognostic value without any further validation. In our study, we used protein data to establish a model for predicting prognosis and validate its value successfully in multiple cohorts. The main strength of this study lies in the first attempt to explore the prognostic role of protein biomarkers based on quantitative proteomic analysis of ACC in adult patients. An IPRPs model was constructed with AUC values equal to 0.933 and our results show that it distinguished itself from previous prognostic predictive models of ACC.
This study had several limitations. The validation of IRPPs model in the transcriptome dataset may lead some bias as the model derived from the proteomic data. The nature of retrospective research limits the clinical value of this work. Further validation cohorts in multicentre or prospective studies are needed to verify the findings. And the more advanced ACC patients in FUSCC cohort may lead to unbalanced baseline. However, it is difficult to conduct randomized controlled tri- als for ACC because of the rarity of these tumours. There is also an urgent need for in vitro and in vivo experiments to explore potential effective functions of IPRPs and reveal the underlying mechanisms.
5 CONCLUSION
We constructed an IPRPs model for predicting the prognosis of adult patients with ACC using proteomic data, gene expression data and
real-world data. The prognostic model showed a stronger predic- tive value for prognosis than other biomarkers (eg Ki-67 and beta- catenin) in multi-cohorts. Our results distinguished FASN, FN, TFRC and TSC1 from previously identified tumour promoters and revealed novel prediction model IPRPs that outperformed the currently es- tablished prognostic parameters for anticipating disease course and better clinical management of adult ACC.
ACKNOWLEDGEMENTS
We thank the TCPA, TCGA and GEO database for providing RPPA data and gene expression profiles of ACC.
CONFLICT OF INTERESTS
The authors declare no competing interests.
AUTHORS’ CONTRIBUTIONS
All authors: Work present carry out in collaboration. YDW, ZHL and QYY: Define the theme of the study and discussed analysis, inter- pretation and presentation. TX and XWH: Manuscript draft; data analysis; development of the algorithm; and explanation of the results. Aihetaimujiang, WHK and WFN: Participation in the collection of rele- vant data and manuscript draft. CDL and LWJ: Help to perform the sta- tistical analysis. SGH: Help to revise the manuscript and provide guiding suggestions. All the authors: Read and approval of the final manuscript.
ETHICAL APPROVAL
The Ethics approval and consent to participate of the current study was approved and consented by the ethics committee of Fudan University Shanghai Cancer center.
DATA AVAILABILITY STATEMENT
The datasets analysed during the current study available from the corresponding author on reasonable request.
ORCID Xi Tian İD https://orcid.org/0000-0003-1965-0647
Wen-Jie Luo İD
https://orcid.org/0000-0003-1133-9870
REFERENCES
1. Kebebew E, Reiff E, Duh Q-Y, et al. Extent of disease at presen- tation and outcome for adrenocortical carcinoma: have we made progress? World J Surg. 2006;30(5):872-878.
2. Else T, Williams AR, Sabolch A, et al. Adjuvant therapies and pa- tient and tumor characteristics associated with survival of adult patients with adrenocortical carcinoma. J Clin Endocrinol Metab. 2014;99(2):455-461.
3. Icard P, Goudet P, Charpenay C, et al. Adrenocortical carcinomas: surgical trends and results of a 253-patient series from the French Association of Endocrine Surgeons study group. World J Surg. 2001;25(7):891-897.
4. Bilimoria KY, Shen WT, Elaraj D, et al. Adrenocortical carcinoma in the United States: treatment utilization and prognostic factors. Cancer. 2008;113(11):3130-3136.
5. Wängberg B, Khorram-Manesh A, Jansson S, et al. The long-term sur- vival in adrenocortical carcinoma with active surgical management and use of monitored mitotane. Endocr Relat Cancer. 2010;17(1):265-272.
|
6. Schteingart DE, Doherty GM, Gauger PG, et al. Management of patients with adrenal cancer: recommendations of an international consensus conference. Endocr Relat Cancer. 2005;12(3):667-680.
7. Rahim MAA, Rahim ZHA, Ahmad WAW, et al. Can saliva proteins be used to predict the onset of acute myocardial infarction among high-risk patients? Int J Med Sci. 2015;12(4):329-335.
8. Li X, Li S, Lu M, et al. Proteomic profiling of iron overload-induced human hepatic cells reveals activation of TLR2-mediated inflamma- tory response. Molecules (Basel, Switzerland). 2016;21(3):322.
9. WuL, Sun Y, Wan J, et al. A proteomic analysis identifies candidate early biomarkers to predict ovarian hyperstimulation syndrome in polycys- tic ovarian syndrome patients. Mol Med Rep. 2017;16(1):272-280.
10. Borrebaeck CAK. Precision diagnostics: moving towards protein biomarker signatures of clinical utility in cancer. Nat Rev Cancer. 2017;17(3):199-204.
11. Bouchal P, Dvořáková M, Roumeliotis T, et al. Combined pro- teomics and transcriptomics identifies carboxypeptidase B1 and nuclear factor KB (NF-KB) associated proteins as putative biomark- ers of metastasis in low grade breast cancer. Mol Cell Proteomics. 2015;14(7):1814-1830.
12. Geisler C, Gaisa NT, Pfister D, et al. Identification and validation of potential new biomarkers for prostate cancer diagnosis and prog- nosis using 2D-DIGE and MS. Biomed Res Int. 2015;2015:454256.
13. Khan AP, Poisson LM, Bhat VB, et al. Quantitative proteomic profil- ing of prostate cancer reveals a role for miR-128 in prostate cancer. Mol Cell Proteomics. 2010;9(2):298-312.
14. Kim HM, Lee YK, Koo JS. Proteome analysis of adrenal cortical tu- mors. Expert Review of Proteomics. 2016;13(8):747-755.
15. Li J, Akbani R, Zhao W, et al. Explore, visualize, and analyze func- tional cancer proteomic data using the cancer proteome atlas. Can Res. 2017;77(21):e51-e54.
16. Nishizuka S, Charboneau L, Young L, et al. Proteomic profiling of the NCI-60 cancer cell lines using new high-density reverse-phase lysate microarrays. Proc Natl Acad Sci USA. 2003;100(24):14229-14234.
17. Tibes R, Qiu Y, Lu Y, et al. Reverse phase protein array: validation of a novel proteomic technology and utility for analysis of primary leukemia specimens and hematopoietic stem cells. Mol Cancer Ther. 2006;5(10):2512-2521.
18. Giordano TJ, Kuick R, Else T, et al. Molecular classification and prog- nostication of adrenocortical tumors by transcriptome profiling. Clin Cancer Res. 2009;15(2):668-676.
19. Wickham H. Ggplot2: Elegant Graphics for Data Analysis. Incorporated: Springer Publishing Company; 2009.
20. Lin H, Zelterman DJT. Modeling survival data: extending the cox model. Technometrics. 2000;44(1):85-86.
21. Friedman J, Hastie T, Tibshirani R. Regularization paths for general- ized linear models via coordinate descent. J Stat Softw. 2010;33(1).
22. Subramanian A, Tamayo P, Mootha VK, et al. Gene set enrich- ment analysis: a knowledge-based approach for interpret- ing genome-wide expression profiles. Proc Natl Acad Sci USA. 2005;102(43):15545-15550.
23. Ritchie ME, Phipson B, Wu D, et al. limma powers differential ex- pression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47.
24. Franceschini A, Szklarczyk D, Frankild S, et al. STRING v9.1: protein- protein interaction networks, with increased coverage and integra- tion. Nucleic Acids Res. 2013;41(Database issue):D808-D815.
25. Smoot ME, Ono K, Ruscheinski J, et al. Cytoscape 2.8: new fea- tures for data integration and network visualization. Bioinformatics (Oxford, England). 2011;27(3):431-432.
26. Bandettini WP, Kellman P, Mancini C, et al. MultiContrast Delayed Enhancement (MCODE) improves detection of subendocardial myocardial infarction by late gadolinium enhancement cardiovas- cular magnetic resonance: a clinical validation study. J Cardiovasc Magn Reson. 2012;14:83.
27. Yu G, Wang L-G, Han Y, et al. clusterProfiler: an R package for compar- ing biological themes among gene clusters. Omics. 2012;16(5):284-287.
28. Miller BS, Ammori JB, Gauger PG, et al. Laparoscopic resection is inappropriate in patients with known or suspected adrenocortical carcinoma. World J Surg. 2010;34(6):1380-1385.
29. Leboulleux S, Deandreis D, Al Ghuzlan A, et al. Adrenocortical car- cinoma: is the surgical approach a risk factor of peritoneal carcino- matosis? Eur J Endocrinol. 2010;162(6):1147-1153.
30. Mete O, Gucer H, Kefeli M, et al. Diagnostic and prognos- tic biomarkers of adrenal cortical carcinoma. Am J Surg Pathol. 2018;42(2):201-213.
31. Lacombe AMF, Soares IC, Mariani BMP, et al. Sterol O-Acyl transfer- ase 1 as a prognostic marker of adrenocortical carcinoma. Cancers. 2020;12(1):247.
32. Liu H, Liu J-Y, Wu X, et al. Biochemistry, molecular biology, and phar- macology of fatty acid synthase, an emerging therapeutic target and diagnosis/prognosis marker. Int J Biochem Mol Biol. 2010;1(1):69-89.
33. Albiges L, Hakimi AA, Xie W, et al. Body mass index and metastatic renal cell carcinoma: clinical and biological correlations. J Clin Oncol. 2016;34(30):3655-3663.
34. Ueda SM, Mao T-L, Kuhajda FP, et al. Trophoblastic neoplasms ex- press fatty acid synthase, which may be a therapeutic target via its inhibitor C93. Am J Pathol. 2009;175(6):2618-2624.
35. Nguyen PL, Ma J, Chavarro JE, et al. Fatty acid synthase polymor- phisms, tumor expression, body mass index, prostate cancer risk, and survival. J Clin Oncol. 2010;28(25):3958-3964.
36. Kridel SJ, Axelrod F, Rozenkrantz N, et al. Orlistat is a novel in- hibitor of fatty acid synthase with antitumor activity. Can Res. 2004;64(6):2070-2075.
37. Knowles LM, Axelrod F, Browne CD, et al. A fatty acid synthase blockade induces tumor cell-cycle arrest by down-regulating Skp2. J Biol Chem. 2004;279(29):30540-30545.
38. Sottile J, Hocking DC. Fibronectin polymerization regulates the composition and stability of extracellular matrix fibrils and cell- matrix adhesions. Mol Biol Cell. 2002;13(10):3546-3559.
39. Park J, Schwarzbauer JE. Mammary epithelial cell interactions with fibronectin stimulate epithelial-mesenchymal transition. Oncogene. 2014;33(13):1649-1657.
40. Knowles LM, Gurski LA, Maranchie JK, et al. Fibronectin matrix formation is a prerequisite for colonization of kidney tumor cells in fibrin. J Cancer. 2015;6(2):98-104.
41. Jones NR, Ashmore JH, Lee SY, et al. Association studies of HFE C282Y and H63D variants with oral cancer risk and iron homeosta- sis among whites and blacks. Cancers. 2015;7(4):2386-2396.
42. Richardson DR, Baker E. The effect of desferrioxamine and ferric ammonium citrate on the uptake of iron by the membrane iron- binding component of human melanoma cells. Biochem Biophys Acta. 1992;1103(2):275-280.
43. Shpyleva SI, Tryndyak VP, Kovalchuk O, et al. Role of ferritin al- terations in human breast cancer cells. Breast Cancer Res Treat. 2011;126(1):63-71.
44. Brooks D, Taylor C, Dos Santos B, et al. Phase Ia trial of murine im- munoglobulin A antitransferrin receptor antibody 42/6. Clin Cancer Res. 1995;1(11):1259-1265.
45. Lu S, Jang H, Zhang J, et al. Inhibitors of Ras-SOS interactions. ChemMedChem. 2016;11(8):814-821.
46. Jiang H, Deng R, Yang X, et al. Peptidomimetic inhibitors of APC- Asef interaction block colorectal cancer migration. Nat Chem Biol. 2017;13(9):994-1001.
47. Tee AR, Fingar DC, Manning BD, et al. Tuberous sclerosis complex-1 and -2 gene products function together to inhibit mammalian tar- get of rapamycin (mTOR)-mediated downstream signaling. Proc Natl Acad Sci USA. 2002;99(21):13571-13576.
48. LimJS, Gopalappa R, Kim SH, et al. Somatic mutations in TSC1 and TSC2 cause focal cortical dysplasia. Am J Hum Genet. 2017;100(3):454-472.
49. Ahel I, Ahel D, Matsusaka T, et al. Poly(ADP-ribose)-binding zinc finger motifs in DNA repair/checkpoint proteins. Nature. 2008;451(7174):81-85.
50. Fong PC, Yap TA, Boss DS, et al. Poly(ADP)-ribose polymerase inhibition: frequent durable responses in BRCA carrier ovar- ian cancer correlating with platinum-free interval. J Clin Oncol. 2010;28(15):2512-2519.
How to cite this article: Tian X, Xu W-H, Anwaier A, et al. Construction of a robust prognostic model for adult adrenocortical carcinoma: Results from bioinformatics and real-world data. J Cell Mol Med. 2021;25:3898-3911. https:// doi.org/10.1111/jcmm.16323
SUPPORTING INFORMATION
Additional supporting information may be found online in the Supporting Information section.