Check for updates
ELSEVIER
ORIGINAL ARTICLE Ferroptosis Biomarkers for Predicting Prognosis and Immunotherapy Efficacy in Adrenocortical Carcinoma
Chengquan Shen,a,b and Yonghua Wanga,b
a Department of Urology, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China
b Key Laboratory of Urology and Andrology, The Affiliated Hospital of Qingdao University, Qingdao, Shandong, China Received for publication April 14, 2022; accepted December 5, 2022 (ARCMED-D-22-00471).
Background. Numerous studies have suggested that ferroptosis plays an important regu- latory role in cancer cell death. Nonetheless, the potential effects of ferroptosis regulators on the prognosis, the expression of immunomodulatory factors in the tumor microen- vironment and on the efficacy of immunotherapy in adrenocortical carcinoma (ACC) remain largely unknown.
Methods. Public ACC datasets were used to investigate the relationship between fer- roptosis regulators and prognosis and clinical features. A ferroptosis scoring system was established for individual cases of ACC using principal component analysis algo- rithms. Hub ferroptosis-related genes involved in immunoregulation and immunotherapy efficacy in ACC were further identified.
Results. Twenty ferroptosis regulators were differentially expressed in ACC and 17 ferroptosis regulators were closely related to prognosis in ACC. A ferroptosis scoring system was developed based on ACSL4, FANCD2, and SLC7A1 expression, and the ferroptosis regulators could serve as an independent prognostic factor for ACC. Further analyses indicated that the ferroptosis score integrated with the tumor mutation burden (TMB), and immune-checkpoint gene expression could predict prognosis in ACC. RNA isolation and reverse transcription-quantitative polymerase chain reaction (RT-qPCR) demonstrated significant differences in the expression levels of ACSL4, FANCD2, and SLC7A1 between ACC and normal tissues. Furthermore, FANCD2 was significantly related to immunotherapy efficacy and prognosis in ACC.
Conclusion. Our study demonstrated that ferroptosis was significantly associated with prognosis, clinical characteristics, immune-checkpoint gene expression, and tumor mi- croenvironment immune cell infiltration in ACC. The current study provides compre- hensive evidence for further research on ferroptosis regulators in ACC and provides new insight into the epigenetic regulation of the antitumor immune response. @ 2022 Instituto Mexicano del Seguro Social (IMSS). Published by Elsevier Inc. All rights reserved.
Key Words: Adrenocortical Carcinoma, Ferroptosis, Prognosis, Immunotherapy, Biomarkers.
Introduction
Adrenocortical carcinoma (ACC) is a rare and highly ag- gressive disease for which diagnostic approaches and thera- peutic strategies have changed only gradually over the past
few decades (1-2). Currently, surgical resection is the main treatment modality to prolong survival in early ACC, but for advanced or metastatic ACC, mitotane, chemotherapy, or other traditional adjuvant treatments often have diffi- culty achieving satisfactory therapeutic effects due to the heterogeneity of the ACC tumor microenvironment.
Emerging discoveries have showcased the potency of ferroptosis on the depression of cancer growth and pro- liferation. Ferroptosis is a new iron-dependent, and reac- tive oxygen species (ROS) form of regulated cell death
Address reprint requests to: Yonghua Wang, Department of Urology, The Affiliated Hospital of Qingdao University, Qingdao 266000, Shan- dong, People’s Republic of China; Phone: (+86) 18661805829; E-mail: wangyonghua@qdu.edu.cn
that is morphologically, biochemically, and genetically dif- ferent from other types of cell death such as apoptosis, necroptosis, autophagy, and pyroptosis (3). Recent studies have suggested that ferroptosis may be an adaptive pro- cess, that is essential in cancer cell eradication (4) and that its activation enhances the therapeutic efficacy of an- ticancer agents (5). Importantly, ferroptosis can influence therapeutic resistance, evasion of immune surveillance, and prognosis by remodeling the immune microenvironment of the tumor (6). Dai E, et al. reported that autophagy- dependent ferroptosis is essential for the polarization of tumor-associated macrophages, which are oriented toward promoting tumor growth, remodeling the tumor microen- vironment, and suppressing tumor immunity (7). Addition- ally, Wang W, et al. found that immunotherapy-activated CD8+ T cells enhance tumor cell ferroptosis, which in turn promotes the antitumor efficacy of immunotherapy (8). Meanwhile, tumor-infiltrating immune cells and im- mune checkpoint inhibitor therapy also play prominent roles in ACC (9-10). Previous studies also indicated that ferroptosis-related genes could predict prognosis in various cancers, such as low-grade gliomas, uveal melanoma, and hepatocellular carcinoma (11-13). Thus, it is particularly important to further study the interaction between ferrop- tosis regulators and immune molecules in the tumor mi- croenvironment, which regulate the function of both tumor cells and immune cells, and then identify the ferroptosis factors closely related to prognosis and immune molecules in ACC.
In our study, we integrated the genomic information of BC samples to comprehensively explore the relation- ship between ferroptosis subgroups and immune-related molecules. The results revealed three different ferroptosis subgroups, and found that the immune cell infiltration, im- mune checkpoint gene expression, and antigen-presenting gene expression were significantly different in three sub- groups. Subsequently, we developed a ferroptosis score to quantify the ferroptosis level in individual patients. Fur- thermore, FANCD2 was a hub biomarker that predicted prognosis and the efficacy of immunotherapy in ACC.
Materials and Method
Data Collection
The RNA sequencing transcriptome data of ACC pa- tients and the corresponding clinical data were extracted from The Cancer Genome Atlas (TCGA) (https://tcga-data. nci.nih.gov/tcga/) and the Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/, GSE19776,) databases. The RNA sequencing transcriptome data of normal samples was downloaded from the Genotype- Tissue Expression (GTEx) database (https://xenabrowser. net/datapages/). One hundred and twenty-seven ACC sam- ples (TCGA: 79; GES19776: 48) and 155 normal sam-
ples were included in our study. Twenty-four ferroptosis regulators (CDKN1A, HSPA5, EMC2, SLC7A1, MT1G, HSPB1, FANCD2, SLC1A5, RPL8, LPCAT3, DPP4, CARS, NFE2L2, GPX4, CISD1, FDFT1, NCOA4, GLS2, CS, ATP5MC3, ACSL4, SAT1, TFRC and ALOX15) were collected from the published literature. Immune cell frac- tion data was obtained from the Tumor Immune Estimation Resource (TIMER) website (http://cistrome.dfci.harvard. edu/TIMER/), which contains the TIMER, CIBERSORT, EPIC, MCPCOUNTER, QUANTISEQ, and XCELL data sets. An immunotherapeutic cohort was included in our study: advanced urothelial cancer with the intervention of atezolizumab, an anti-PD-L1 antibody (IMvigor210 cohort) (14).
Consensus Clustering of Ferroptosis Regulators
To functionally elucidate the biological characteristics of ferroptosis regulators in ACC, consensus clustering analy- sis with “ConsensusClusterPlus” was employed to classify ACC patients into different groups based on the expression levels of twenty-four ferroptosis regulators. Principal com- ponent analysis (PCA) was performed to evaluate gene- expression profiles among distinct ACC subtypes.
Single-sample Gene-set Enrichment Analysis
The single-sample gene-set enrichment analysis (ssGSEA) algorithm has been widely used to quantify the relative abundance of immune cell infiltration in the tumor mi- croenvironment. The gene set for marking each immune cell type was extracted from the study of Charoentong, which contained various immune cell types, such as acti- vated B-cell, activated CD4 T-cell, activated CD8 T-cell, activated dendritic cell, and macrophage (15). ssGSEA cal- culated the enrichment fractions, which were used to rep- resent the relative abundance of each immune cells infil- tration level in the tumor microenvironment for each ACC sample.
Gene Set Variation Analysis
Gene set variation analysis (GSVA) is usually used for the determination of biological process activity divergences in the samples of an expression dataset in a non-parametric and unsupervised manner (16). To identify the difference in biological process between ferroptosis regulators-related subgroups, gene set variation analysis (GSVA) enrichment analysis was performed by using the “GSVA” R package.
Generation of a Ferroptosis Regulator-related Scoring System
To quantify the ferroptosis level of an individual tumor, a ferroptosis regulator-related scoring system was estab- lished to assess the ferroptosis level of individual ACC
patients, and we named it the ferroptosis score. Univari- ate Cox regression analysis was performed on ferroptosis regulators to screen candidate genes that were closely re- lated to the prognosis of patients with ACC. PCA was then conducted to establish a ferroptosis regulator-related scoring system. The advantage of PCA is concentrating scores on the set with the largest block of highly relevant (or inversely relevant) genes in the set while reducing the contribution weight of genes that do not track with other set members (17).
Development of Prognostic Nomograms
Clinicopathological characteristics, TMB (tumor mutation burden), immune-checkpoint gene (PD-L1, programmed death-ligand 1; CTLA4, cytotoxic T-lymphocyte associ- ated protein 4; PD-1, programmed death 1) expression, and ferroptosis score were integrated to create a nomo- gram that was used to assess the probability of 1, 2, and 3 year overall survival (OS) and progression-free sur- vival (PFS) for ACC patients using the R package (https: /cran.r-project.org/web/packages/rms/) (18).
RNA Isolation and Reverse Transcription-Quantitative Polymerase Chain Reaction (RT-qPCR)
To further validate the expression levels of three ferrop- tosis regulators that were utilized to develop the signature in our ACC tissues and normal tissues, RT-qPCR were performed. The RNAprep Pure FFPE Kit (DP439, TIAN- GEN Biotech (Beijing) Co, Ltd, CHN) was used to ex- tract total RNA from tissue specimens based on the man- ufacturer’s instructions. For the detection of mRNA levels, the total RNA (500 ng) was transcribed into cDNA us- ing a PrimeScript™M RT reagent kit (Perfect Real Time) (Takara, code no RR037A, Beijing, China). All primers were synthesized by Huada Gene (Beijing, China) and the sequences are shown in the Supplementary Table 1. The amplification of cDNAs was achieved with a Roche LightCycler 480II real-time PCR detection system (Roche, Basel, Switzerland). Gene expression was nor- malized against B-actin, and the relative expression lev- els of ACSL4, FANCD2, and SLC7A1 were determined by the comparative threshold (Ct) cycle method using the Formula 2-(44Ct).
Statistical Analysis
Statistical tests were performed using R version 3.6.0 and GraphPad Prism 8.0. The expression levels of ferroptosis regulators in ACC subtypes, and between tumor and nor- mal tissues were compared using one-way analysis of vari- ance (ANOVA) and t-tests. Depending on the relationship between ferroptosis regulators or the ferroptosis score and ACC patient survival, the “survminer” R package was used
to determine the cutoff point for each subgroup of the data set. We used the Kaplan-Meier method to generate survival curves and compared the difference between groups with the log-rank test. The correlation between the ferroptosis score and immune cell infiltration levels was subjected to a Pearson correlation test in GraphPad Prism 8.0. Uni- variate and multivariate Cox regression analyses were per- formed to ascertain the independent prognostic value of the ferroptosis score integrated with other clinicopathological characteristics.
Results
The landscape of Genetic Variation in Ferroptosis Regulators in ACC
To explore the biological function of ferroptosis regulators in the occurrence and development of ACC, we system- atically assessed the expression profiles of 24 ferropto- sis regulators in ACC and normal tissue samples. Com- pared to the expression levels of ferroptosis regulators in normal samples, CDKN1A, HSPA5, SLC7A1, HSPB1, FANCD2, RPL8, DPP4, GPX4, FDFT1, NCOA4, and ACSLA were dramatically upregulated in ACC samples, while EMC2, MT1G, SLC1A5, LPCAT3, CAR3, NFE2L2, GLS2, ATP5MC3, and STA1 were markedly downregu- lated in ACC samples. However, the expression of CISD1, CS, TFRC, and ALOX15 was not significantly different between normal and ACC samples (Figure 1A, Supple- mentary Table 2). Meanwhile, we analyzed the incidence of somatic mutations of 24 ferroptosis regulators in ACC. Among the 92 samples, only 6 (6.52%) had ferroptosis regulator mutations, indicating that somatic mutations in ferroptosis regulators are infrequent in ACC (Figure 1B). SLC1A5 and SAT1 exhibited the highest mutation rate in ACC samples. Moreover, our results indicated that NCOA4 was highly expressed in SLC1A5-mutant types, while MT1G was expressed at low levels in wild-type cells (Supplementary Figure 1). However, no ferroptosis regu- lators were differentially expressed between SAT1-mutant and wild-type cells (Supplementary Figure 2). Copy num- ber variations (CNVs) of RNA regulatory genes have been demonstrated to contribute to the alteration in mRNA ex- pression and were related to the prognosis and progres- sion of several tumors. Our results indicated a prevalent CNV alteration in 24 ferroptosis regulators and 18 reg- ulators were focused on the amplification in copy num- ber, while NCOA4, HSPB1, ALOX15, ACSL4 CISD1, and SLC7A1 had a widespread frequency of CNV dele- tion (Figure 1C). Figure 1D shows the position of CNV- altered ferroptosis regulators on chromosomes. These re- sults suggest that the imbalanced expression of ferroptosis regulators plays critical biological roles in the progression of ACC.
A
B
Altered in 6 (6.52%) of 92 samples.
960
0
2
0
Type
Normal
3
Tumor
SLC1A5
2%
SAT1
FDFT1
2%
CS
1%
CDKN1A
1%
0%
HSPA5
EMC2
0%
SLC7A1
0%
0%
35
MT1G
10
HSPB1
0%
FANCD2
0%
RPL8
0%
Gene expression
LPCAT3
0% 0%
DPP4
CARS
0%
NFE2L2
0%
GPX4
0%
CISD1
0%
0%
0
NCOA4
GLS2
0%
ATP5MC3
0%
0%
ACSL4
TFRC
0%
ALOX15
0%
0%
· C>T = T>A
= C>G = T>C
CDKN1A
HSPA5
EMC2
SLC7A1
MT1G
HSPB1
FANCD2
SLC1A5
RPL8
LPCAT3
DPP4
CARS
NFE2L2
GPX4
CISD1
FDFT1
NCOA4
GLS2
CS
ATP5MC3
ACSL4
SAT1
TFRC
ALOX15
· C>A = T>G
Missense_Mutation = Multi_Hit
In_Frame_Del
C
D
3
×
10
2,
· GAIN . LOSS
20
2
19
18
SAT1
ACSL4
CNV.frequency(%)
8
17
GPX4
TOPP4 NFE2L2 FANCD2
3
6
16
ALOX15
MT1G
TFRC
15
4
4
14
SLC7A1
GLS2
LPCAT3
CDKN1A
5
2
13
CARS
SIEBA4
HSPA5
ERPES
HSPB1
.
FDFT1
12
0
.
GPX4
MT1G
RPL8
CS
GLS2 FDFT1
SLC1A5
TFRC
CDKN1A
FANCD2
LPCAT3
NCOA4
SAT1
CARS
HSPA5
HSPB1
ALOX15
ACSL4
ATP5MC3
CISD1
DPP4
EMC2
NFE2L2
SLC7A1
11
1
10
5
·
Ferroptosis Regulators were Associated with the Prognosis and Clinical Characteristics of ACC
To further investigate the importance of ferroptosis regula- tors in ACC, a survival analysis was performed to assess the prognostic value of each ferroptosis regulator. The re- sults showed that ACSL4, ATP5MC3 CDKN1A, CISD1, CS, EMC2, FANCD2, FDFT1, GLS2, GPX4, HSPA5, HSPB1, NFE2L2, RPL8, SAT1, SLC1A5, SLC7A1, and TFRC were associated with the survival outcomes of ACC patients (all p <0.05) (Supplementary Figure 3, and Supplementary Figure 4). Furthermore, the relation- ship between prognostic ferroptosis regulators and clini- cal characteristics (stage, T stage, N stage, and M stage) was analyzed. We found that patients with high stage
had higher expression of FANCD2, SLC1A5, SLC7A1, and TFRC, but lower expression of ACSL4, ATP5MC3, EMC2, GPX4, and LPCAT3 (Supplementary Figure 5A). FANCD2, SLC7A1, and TFRC had high expression, while ATP5CM3 and EMC2 had low expression in ACC pa- tients with advanced T stage disease (Supplementary Fig- ure 5B). Compared to ACC patients without lymph node metastasis, HSPB1 was downregulated in the patients with lymph node metastasis (Supplementary Figure 5C). More- over, FANCD2 and SLC7A1 may promote the metastasis of ACC, while EMC2 and HSPB1 may inhibit the metas- tasis of ACC (Supplementary Figure 5D). Overall, the ex- pression of ferroptosis regulators was not only closely re- lated to the prognosis of ACC but was also significantly related to the clinical characteristics of ACC.
Tumor Environment Immune Cell Infiltration Characterization in Distinct Ferroptosis Subgroups
To investigate the biology of different functional groups of ferroptosis regulators, we classified ACC patients with qualitatively different ferroptosis subgroups based on the expression levels of ferroptosis regulators, and three dis- tinct subgroups were eventually identified, namely, Cluster A (n = 31), Cluster B (n= 39), and Cluster C (n=57), re- spectively (Supplementary Figure 6A-6D). Box plot and a heatmap showed that most ferroptosis regulators were dif- ferentially expressed in three subgroups (Figure 2A, Sup- plementary Figure 7A). The detailed p value of each fer- roptosis regulator in different clusters is shown in Sup- plementary Table 3. Interestingly, survival analyses for the three subgroups revealed that patients in Cluster C had sig- nificantly longer OS (p <0.001) and PFS (p <0.001) than those in Cluster A or Cluster B (Figure 2B, and 2C).
Furthermore, GSVA was performed to explore the bio- logical behaviors of the different ferroptosis groups. The results showed that Cluster A markedly enriched extracel- lular matrix (ECM) receptor interactions, steroid biosyn- thesis, terpenoid backbone biosynthesis, the cell cycle, and DNA replication (Supplementary Figure 7B). Cluster B presented enrichment pathways associated with steroid biosynthesis, terpenoid backbone biosynthesis, nucleotide excision repair, and the cell cycle (Supplementary Figure 7C). Cluster C was prominently related to full immune activation including the chemokine signaling pathway, nat- ural killer cell-mediated cytotoxicity, T-cell receptor sig- naling pathway, Toll-like receptor signaling pathway, etc (Supplementary Figure 7D). Subsequently, we analyzed the immune cell infiltration and antigen-presenting gene expression among the different ferroptosis groups. Clus- ter C showed higher infiltration levels of immune cells in the ssGSEA data set, such as activated B-cells, activated CD4 T-cells, CD8 T-cells, macrophages, and T-follicular helper cells, and so on (Figure 2D, Supplementary Ta- ble 4). In addition, other data sets (CIBERSORT, MCP- COUNT, QUANTISQ, TIMER, XCELL) were also found which showed that there were significant differences in the compositions of immune cell types, especially CD4+ T-cells, CD8+ T-cells, and macrophages, among the three subgroups (Supplementary Figure 8A-8E, Supplementary Tables 5-9).
Considering the important roles of immune checkpoint genes and antigen-presenting genes in the immune re- sponse to tumors, we assessed the expression of these genes in the three ferroptosis subgroups. The results showed that PD-L1, CTLA4, and most antigen-presenting genes were highly expressed in Cluster C, representing the strong immunogenicity of Cluster C in ACC (Supplemen- tary Figure 8F-8I). Taken together, ferroptosis regulators play critical roles in the immune microenvironment remod- eling and the antitumor immune response in ACC.
Ferroptosis Score Combined with Immune Factors Predicted Prognosis in ACC
To better evaluate the ferroptosis level of each ACC pa- tient, a ferroptosis score based on the expression values of ACSL4, FANCD2, and SLC7A1 was generated. After- ward, patients were divided into high- or low-score groups based on the best cutoff value determined. The distribution of the survival time, survival status, ferroptosis scores, and expression profiles of the three ferroptosis regulators is dis- played in Supplementary Figure 9A, and 9B. The results showed that SLC7A1 and FANCD2 were overexpressed in the high-score group, while ACSL4 was upregulated in the low-score group. Moreover, the ferroptosis score was pos- itively correlated with the OS and PFS of ACC patients (Figure 3A, and 3B). Interestingly, we found that the fer- roptosis score can predict the OS and PFS of ACC patients who received mitotane therapy (Supplementary Figure 9C, and Supplementary Figure 9D). Univariate and multivariate Cox regression analyses indicated that the ferroptosis score could serve as an independent predictor in ACC patients (Figure 3C, and 3D). Moreover, the score was associated with the stage (p = 0.004), T stage (p = 0.017), and metas- tasis status (p = 0.039) of ACC (Supplementary Figure Figure 9E-9G).
Subsequently, we analyzed the correlation between fer- roptosis regulators and immune factors in ACC. The re- sults showed that the ferroptosis score was positively related to the expression of PD-L1 and CTLA4 and negatively related to the TMB (Supplementary Figure 10A-10F). However, the ferroptosis score was not as- sociated with the expression of PD-1 (Supplementary Figure 10G-10H). Prognostic analyses showed that OS and PFS have significant advantages in patients with low TMB and PD-1, and high expression of PD-L1 and CTLA4 (Supplementary Figure 101-10P). Further analyses suggested that the ferroptosis score integrated with various immune factors including CTLA4, PD-L1, and TMB expression, could predict prognosis in ACC (Figure 4).
To establish a clinically applicable method for mon- itoring the prognosis of ACC patients, we developed a prognostic nomogram by using clinical characteristics (age, gender, stage, T stage, N stage, M stage, invasion of tu- mor capsule, mitotic rate >5/50 HPF, nuclear grade III or IV, Weiss score, radiation therapy, and mitotane ther- apy), TMB, PD-L1 expression, CTLA4 expression, PD- 1 expression, and the ferroptosis score. The results indi- cated that the new prognostic nomogram could better pre- dict the 1, 2, 3, and 5 year OS and PFS of ACC pa- tients (Supplementary Figure 11A, and 10B). The cali- bration plots also validated excellent agreement between prediction and observation for the 3 and 5 year OS and PFS probabilities of ACC patients (Supplementary Figure 11C-11F).
A
Cluster 白AB c
10
Gene expression
O
0
CDKN1A
HSPA5
EMC2
SLC7A1
MT1G
HSPB1
FANCD2
SLC1A5
RPL8
LPCAT3
DPP4
CARS
NFE2L2
GPX4
CISD1
FDFT1
NCOA4
GLS2
CS
ATP5MC3
ACSL4
SAT1
TFRC
ALOX15
B
C
1.00
Cluster
1.00
Cluster
4
A
+
Je
O
Survival probability
.0.75
C
> 0.75
Survival probability
C
0.50
0.50
0.25
p<0.001
0.25
p<0.001
0.00
0.00
0
1
A
CI
8
9 10 11 12
13
14
2
ã
2
00
0
2
0
1
2
O
Time(years)
3
4
5
6
7
8
9
10
11
12
Time(years)
Cluster
Number at risk
Number at risk
NO
ONO
1000
000
Cluster
QUE
4 RAN
0-0
000
0
NOO
NOO
0
1
2
V
4
5
6
1
9
10
1
12
15
1
18
20
Time(years)
0
1
2
3
4
5
6
7
8
9
10
11
12
D
Cluster 3
Time(years)
A
B
3
C
1.00
ns
**
ns
ns
ns
**
**
*
ns
**
:
0.75
Immune infiltration
0.50
0.25
0.00
Activated B cell
Activated CD4 T cell
Activated CD8 T cell
Activated dendritic cell
CD56bright natural killer cell
CD56dim natural killer cell
Eosinophil
Gamma delta T cell
Immature B cell
Immature dendritic cell
MDSC
Macrophage
Mast cell
Monocyte
Natural killer T cell
Natural killer cell
Neutrophil
Plasmacytoid dendritic cell
Regulatory T cell
T follicular helper cell
Type 1 T helper cell
Type 17 T helper cell
Type 2 T helper cell
A
1.00
OS
B
1.00
PFS
Ferroptosis score
Ferroptosis score
Low
Low
Survival probability
0.75
High
Survival probability
0.75
High
0.50
0.50
0.25
p
0.25
p
001
0.00
0.00
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
20
Time(years)
0
1
2
3
4
5
6
7
8
9
10
11
12
Time(years)
C
D
0
2
4
6
8
10
12
14
0
20
40
60
80
Hazard ratio
Hazard ratio
ACSL4, FANCD2, and SLC7A1 in the Role of anti-PD-L1 Immunotherapy
RT-qPCR were performed to validate the expression lev- els of the three selected ferroptosis regulators in human ACC tissues and normal tissues. The results demonstrated significant differences in the expression levels of ACSL4, FANCD2, and SLC7A1 between ACC and normal tissues. Compared to normal tissues, three ferroptosis regulators were upregulated in ACC (Figure 5A).
Immunotherapies represented by PD-L1/PD-1 pathway blockade have emerged as the main treatment in advanced cancer therapy. In this study, we investigated the role of ACSLA, FANCD2, and SLC7A1 in predicting the response of patients to anti-PD-L1 therapy based on the IMvigor210 cohort. Survival analyses showed that patients with low ACSL4, high FANCD2, and high SLC7A1 exhibited sig- nificantly prolonged survival (Figure 5B, Supplementary Figure 12A, and 12B) in the IMvigor210 cohort. In addi- tion, patients with high FANCD2 also showed significant therapeutic advantages and clinical response to anti-PD-L1 immunotherapy, while ACSL4 and SLC7A1 were not as- sociated with clinical response to anti-PD-L1 immunother- apy (Figure 5C; Supplementary Figure 12C-12G). Subse- quently, we evaluated the relationship between FANCD2
and immune cell infiltration, immune checkpoint gene ex- pression (CTLA4, PD-L1, and PD-1), and tumor neoanti- gen burden in ACC. The results indicated that FANCD2 expression was associated with the infiltration of activated CD4 T-cells, neutrophil, plasmacytoid dendritic cell, CD56 bright natural killer cell, type 2 T-helper cell, type 17 helper cell (Supplementary Table 10), while FANCD2 ex- pression was not associated with immune checkpoint gene expression (Supplementary Figure 12H-12M). Moreover, FANCD2 expression was negatively related to neoantigen burden and that patients with a high tumor neoantigen bur- den had a better prognosis than the patients with a low tu- mor neoantigen burden (Supplementary Figure 12N-12P). Moreover, the results also found that patients with the com- bination of high FANCD2 expression and high neoantigen burden exhibited a favorable prognosis (Figure 5D). The above results indicated that FANCD2 was a potential and reliable biomarker for prognosis and clinical response as- sessment of immunotherapy.
Discussion
Ferroptosis, a programmed necrosis factor, plays a vital role in killing cancer cells and inhibiting cancer growth
A
1.00
+ H-CTLA4+H-Ferroptosis score
+ H-CTLA4+L-Ferroptosis score
0.75
+ L-CTLA4+H-Ferroptosis score
Survival probability
+ L-CTLA4+L-Ferroptosis score
0.50
0.25
p <0.001
0.00
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18 1
20
Time(years)
B
1.00
+ H-PD-L1+H-Ferroptosis score
+ H-PD-L1+L-Ferroptosis score
0.75
+ L-PD-L1+H-Ferroptosis score
Survival probability
+ L-PD-L1+L-Ferroptosis score
0.50
0.25
p <0.001
0.00
0
1
3
4
5
6
7
8
9
10
11 1
12
13
14
15
16
17
18
19
20
Time(years)
C
1.00
H-TMB+H-Ferroptosis score
+ H-TMB+L-Ferroptosis score
0.75
+ L-TMB+H-Ferroptosis score
Survival probability
+ L-TMB+L-Ferroptosis score
0.50
0.25
p<0.001
0.00
0
1
2
3
4
5
6
7
8
9
10
11
12
Time(years)
by mediating multiple biological pathways, including the P53 and Ras-MEK pathways (19). In addition, ferroptosis- related genes can serve as prognostic biomarkers in mul- tiple cancers, such as head and neck squamous cell carci- noma, gastric cancer, and colorectal cancer (20-22). Pre- vious studies have also demonstrated that ferroptosis was significantly related to immune microenvironment remod- eling and immune cell infiltrating in hepatocellular can- cer, breast cancer, glioma, and so on (23-25). However, the role of ferroptosis regulators in the prognosis, immune checkpoint gene expression, and immune cell infiltration of ACC remains largely unknown. The identification of ferroptosis-related biomarkers is important for the preven- tion, diagnosis, and treatment of ACC.
The present study showed that most ferroptosis regula- tors were differentially expressed in ACC and were signif- icantly related to prognosis and clinical characteristics in ACC. The results indicated that the imbalanced expression of ferroptosis regulators is significant in the tumorigene- sis and progression of ACC. Further analyses found that ferroptosis subgroups were correlated with prognosis, im- mune cell infiltration, and many immune pathways, such as the chemokine signaling pathway, natural killer cell- mediated cytotoxicity, T-cell receptor signaling pathway, and Toll-like receptor signaling pathway. These results in- dicated that the imbalanced expression of ferroptosis reg- ulators may play critical biological roles in the tumorigen- esis, progression, and immune microenvironment of ACC.
Considering the individual heterogeneity of ferroptosis regulators, we constructed a scoring system to assess the ferroptosis modification pattern of individual ACC patients based on the expression of SC7A1, FANCD2, and ACSL4. The results demonstrated that the ferroptosis score was sig- nificantly related to prognosis, PD-L1 expression, TMB, and immune cell infiltration, especially CD4+ T-cell in- filtration. Previous studies have validated that PD-L1 and TMB are the main markers for predicting the efficacy of immunotherapy in tumors, including ACC (26). The above results imply that ferroptosis regulators could influence the efficacy of immune checkpoint blockade and shape the im- mune microenvironment in ACC.
Furthermore, our study showed that overexpressed SC7A1, FANCD2, and ACSL4 were associated with a poor prognosis in ACC. Recent studies showed that SLC7A1, FANCD2, and ACSL4 were highly expressed and pre- dicted a worse prognosis for tumors (27-29). To further evaluate the correlation between the three genes and im- munotherapy, the IMvigor210 cohort was used. However, we found that only FANCD2 was associated with the ef- ficacy of immunotherapy. Further analyses indicated that FANCD2 was associated with the tumor neoantigen burden and immune cell infiltration in the tumor microenvironment of ACC. FANCD2, a regulator involved in the negative regulation of ferroptosis, can regulate gene expression in- volved in iron metabolism (FTH1, TFRC, and STEAP3)
A
6-p = 0.0105
B
Relative expression of mRNA Ref to GAPDH
O
p = 0.0191
O
1.00
FANCD2
Low
4
p = 0.0199
Survival probability
0.75
High
0.50
2
0.25
p=0.038
0
0.00
0
1
2
FANCD2
SLC7A1
ACSL4
Time(years)
Normal
ACC
C
D
Best Confirmed Overall Response
CR
PD
PR
1.00
10-
0.1
0.2
- H-Neo+H-FANCD2
H-Neo+L-FANCD2
0.27
+ L-Neo+H-FANCD2
0.003
0.75
8
+ L-Neo+L-FANCD2
0.15
0.0036
FANCD2
Survival probability
0.50
6
0.25
4
p<0.001
0.00
CR
PD
PR
SD
0
1
2
Best Confirmed Overall Response
Time(years)
and lipid peroxidation (GPX4) (30). Lei LC, et al. showed that FANCD2 promoted cancer cell proliferation and tumor colony formation and metastasis potential, and cell cycle progression by modulating cyclin-CDK and ATR/ATM sig- naling (31). Sun S, et al. also found that FANCD2 may be a novel ferroptosis-related biomarker to predict prognosis and immunotherapeutic effects in lung adenocarcinomas (32). Moreover, a study showed that FANCD2 can limit the formation of replication stress-induced P-bodies that are capable of regulating activation of the innate immune response after prolonged replication stress (33). The above results showed that the ferroptosis-related gene FANCD2 may serve as a reliable biomarker for prognosis and im- munotherapy efficacy in ACC by regulating the ferroptosis level of the tumor microenvironment and immune-related signaling pathway.
However, there are several limitations to our study. First, the expression levels of ferroptosis regulators need to be further validated in ACC samples or ACC cells by im- munohistochemical analysis and Western blotting. Second, the ferroptosis score and the interaction between immune factors and ferroptosis regulators should be subjected to ex- ternal verification in multicenter cohorts. In addition, the regulatory mechanisms of ferroptosis regulators in ACC need to be further explored to remodel the immune mi- croenvironment and improve the precision of immunother- apy for ACC.
Conclusions
In conclusion, our study demonstrated that ferroptosis reg- ulators were significantly associated with the prognosis,
clinical characteristics, immune-checkpoint gene expres- sion, and tumor microenvironment immune cell infiltra- tion in ACC. This study provides comprehensive evidence for further research on ferroptosis regulators in ACC and provides new insight into the epigenetic regulation of the antitumor immune response.
Ethics Approval and Consent to Participate
Under the ethical guidelines as required by the Declaration of Helsinki, informed consent was obtained from each pa- tient, and the research protocol was approved by the Ethics Committee of the Affiliated Hospital of Qingdao Univer- sity.
Conflicts of Interest
The authors report no conflicts of interest.
Acknowledgments
This work was supported by the National Natural Science Foundation of China (Nos. 81972378, 81101932).
Supplementary Materials
Supplementary material associated with this article can be found, in the online version, at doi: 10.1016/j.arcmed.2022. 12.003.
References
1 Crona J, Baudin E, Terzolo M, et al. ENSAT registry-based random- ized clinical trials for adrenocortical carcinoma. Eur J Endocrinol 2021;184:R51-R59. doi:10.1530/EJE-20-0800.
2 Crona J, Beuschlein F. Adrenocortical carcinoma - towards genomics guided clinical care. Nat Rev Endocrinol 2019;15:548-560. doi:10. 1038/s41574-019-0221-7.
3 Stockwell BR, Friedmann Angeli JP, Bayir H, et al. Ferroptosis: A Regulated Cell Death Nexus Linking Metabolism, Redox Biology, and Disease. Cell 2017;171:273-285. doi:10.1016/j.cell.2017.09.021.
4 Dixon SJ. Ferroptosis: bug or feature? Immunol Rev 2017;277:150- 157. doi:10.1111/imr.12533.
5 Eling N, Reuter L, Hazin J, et al. Identification of artesunate as a specific activator of ferroptosis in pancreatic cancer cells. Oncoscience 2015;2:517-532. doi:10.18632/oncoscience.160.
6 Friedmann Angeli JP, Krysko DV, Conrad M. Ferroptosis at the cross- roads of cancer-acquired drug resistance and immune evasion. Nat Rev Cancer 2019;19:405-414. doi:10.1038/s41568-019-0149-1.
7 Dai E, Han L, Liu J, et al. Autophagy-dependent ferroptosis drives tumor-associated macrophage polarization via release and uptake of oncogenic KRAS protein. Autophagy 2020;16:2069-2083. doi:10. 1080/15548627.2020.171 4209.
8 Wang W, Green M, Choi JE, et al. CD8+ T cells regulate tumour ferroptosis during cancer immunotherapy. Nature 2019;569:270-274. doi:10.1038/s41586-019-1170-y.
9 Tian X, Xu W, Wang Y, et al. Identification of tumor-infiltrating immune cells and prognostic validation of tumor-infiltrating mast cells in adrenocortical carcinoma: results from bioinformatics and real-world data. Oncoimmunology 2020;9:1784529. doi:10.1080/ 2162402X.2020.1784529.
10 Brabo EP, Moraes AB, Neto LV. The role of immune checkpoint inhibitor therapy in advanced adrenocortical carcinoma revisited: re- view of literature. J Endocrinol Invest 2020;43:1531-1542. doi:10. 1007/s40618-020-01306-5.
11 Luo H, Ma C. A Novel Ferroptosis-Associated Gene Signature to Pre- dict Prognosis in Patients with Uveal Melanoma. Diagnostics (Basel) 2021;11:219. doi:10.3390/diagnostics 11020219.
12 Zheng Y, Ji Q, Xie L, et al. Ferroptosis-related gene signature as a prognostic marker for lower-grade gliomas. J Cell Mol Med 2021;25:3080-3090. doi:10.1111/jcmm.16368.
13 Liang JY, Wang DS, Lin HC, et al. A Novel Ferroptosis-related Gene Signature for Overall Survival Prediction in Patients with Hepatocel- lular Carcinoma. Int J Biol Sci 2020;16:2430-2441. doi:10.7150/ijbs. 45050.
14 Rosenberg JE, Hoffman-Censits J, Powles T, et al. Atezolizumab in patients with locally advanced and metastatic urothelial car- cinoma who have progressed following treatment with platinum- based chemotherapy: a single-arm, multicentre, phase 2 trial. Lancet 2016;387:1909-1920. doi:10.1016/S014 0-6736(16)00561-4.
15 Charoentong P, Finotello F, Angelova M, et al. Pan-cancer Im- munogenomic Analyses Reveal Genotype-Immunophenotype Rela- tionships and Predictors of Response to Checkpoint Blockade. Cell Rep 2017;18:248-262. doi:10.1016/j.celrep.2016.12.019.
16 Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analy- sis for microarray and RNA-seq data. BMC Bioinformatics 2013;14:7. doi:10.1186/1471-2105-14-7.
17 Zhang B, Wu Q, Li B, et al. m6A regulator-mediated methylation modification patterns and tumor microenvironment infiltration char- acterization in gastric cancer. Mol Cancer 2020;19:53. doi:10.1186/ s12943-020-01170-0.
18 Shen C, Liu J, Yang X, et al. Development and Validation of an m6A RNA Methylation Regulators-Based Signature for Predicting the Prognosis of Adrenocortical Carcinoma. Front Endocrinol (Lausanne) 2021;12:568397. doi:10.3389/fendo.2021.568397.
19 Mou Y, Wang J, Wu J, et al. Ferroptosis, a new form of cell death: opportunities and challenges in cancer. J Hematol Oncol 2019;12:34. doi:10.1186/s13045-019-0720-y.
20 He F, Chen Z, Deng W, et al. Development and validation of a novel ferroptosis-related gene signature for predicting prognosis and immune microenvironment in head and neck squamous cell carci- noma. Int Immunopharmacol 2021;98:107789. doi:10.1016/j.intimp. 2021.107789.
21 Wang C, Shi M, Ji J, et al. Stearoyl-CoA desaturase 1 (SCD1) facilitates the growth and anti-ferroptosis of gastric cancer cells and predicts poor prognosis of gastric cancer. Aging (Albany NY) 2020;12:15374-15391. doi:10.18632/a ging.103598.
22 Shao Y, Jia H, Huang L, et al. An Original Ferroptosis-Related Gene Signature Effectively Predicts the Prognosis and Clinical Status for Colorectal Cancer Patients. Front Oncol 2021;11:711776. doi: 10.3389/ fonc.2021.711776.
23 Tang B, Zhu J, Li J, et al. The ferroptosis and iron-metabolism sig- nature robustly predicts clinical diagnosis, prognosis and immune mi- croenvironment for hepatocellular carcinoma. Cell Commun Signal 2020;18:174. doi:10.1186/s12964-020-00663-1.
24 Zhang K, Ping L, Du T, et al. A Ferroptosis-Related lncRNAs Sig- nature Predicts Prognosis and Immune Microenvironment for Breast Cancer. Front Mol Biosci 2021;8:678877. doi:10.3389/fmolb.2021. 678877.
25 Zheng J, Zhou Z, Qiu Y, et al. A Prognostic Ferroptosis-Related lncRNAs Signature Associated With Immune Landscape and Radio- therapy Response in Glioma. Front Cell Dev Biol 2021;9:675555. doi:10.3389/fcell.2021.675555.
26 Araujo-Castro M, Pascual-Corrales E, Molina-Cerrillo J, et al. Immunotherapy in Adrenocortical Carcinoma: Predictors of Re- sponse, Efficacy, Safety, and Mechanisms of Resistance. Biomedicines 2021;9:304. doi:10.3390/biomed icines9030304.
27 Gong W, Chen Y, Zhang Y. Prognostic and clinical significance of Solute Carrier Family 7 Member 1 in ovarian cancer. Transl Cancer Res 2021;10:602-612. doi:10.21037/tcr-20-2744.
28 Wu G, Wang Q, Xu Y, et al. A new survival model based on ferroptosis-related genes for prognostic prediction in clear cell renal cell carcinoma. Aging (Albany NY) 2020;12:14933-14948. doi:10. 18632/aging. 103553.
29 Feng J, Lu PZ, Zhu GZ, et al. ACSL4 is a predictive biomarker of sorafenib sensitivity in hepatocellular carcinoma. Acta Pharmacol Sin 2021;42:160-170. doi:10.1038/s41401-020-0439-x.
30 Song X, Xie Y, Kang R, et al. FANCD2 protects against bone marrow injury from ferroptosis. Biochem Biophys Res Commun 2016;480:443-449. doi:10.1016/j.bbrc.2016.10.068.
31 Lei LC, Yu VZ, Ko JMY, et al. FANCD2 Confers a Malignant Phenotype in Esophageal Squamous Cell Carcinoma by Regulating Cell Cycle Progression. Cancers (Basel) 2020;12:2545. doi: 10.3390/ cancers 12092545.
32 Sun S, Guo W, Lv F, et al. Comprehensive Analysis of Fer- roptosis Regulators in Lung Adenocarcinomas Identifies Prog- nostic and Immunotherapy-Related Biomarkers. Front Mol Biosci 2021;8:587436. doi: 10.3389/fmol b.2021.587436.
33 Emam A, Wu X, Xu S, et al. Stalled replication fork protection limits cGAS-STING and P-body-dependent innate immune signalling. Nat Cell Biol 2022;24:1154-1164. doi:10.1038/s41556-022-00950-8.