Original Article Pan-analysis reveals CACYBP to be a novel prognostic and predictive marker for multiple cancers

Baosen Mo1, Bijun Luo2, Yuesong Wu1

1Department of Cardiothoracic Surgery, The 923 Hospital of Joint Logistics Support Force of Chinese People’s Liberation Army, Nanning, Guangxi Zhuang Autonomous Region, China; 2Department of Anesthesiology, The Region Maternal and Child Health Hospital of Guangxi Zhuang Autonomous, Nanning, Guangxi Zhuang Autonomous Region, China

Received August 3, 2023; Accepted October 26, 2023; Epub January 15, 2024; Published January 30, 2024

Abstract: Objectives: Cancer has emerged as a global issue in terms of public health care and treatment. The signifi- cance of calcyclin binding protein (CACYBP) in various neoplasms suggests that it may serve as a novel biomarker for numerous types of human tumors. Methods: Our research investigated the differences in CACYBP expression between cancer tissues and normal tissues using a total of 18,787 samples from multiple centers. To explore the prognostic factor of CACYBP in cancers, we utilized Cox regression analysis and Kaplan-Meier curves. We also con- ducted Spearman’s rank correlation analyses to determine the associations of CACYBP expression with the immune microenvironment, etc. Additionally, we applied gene set enrichment analysis to explore the underlying mechanisms of CACYBP in cancers. A partial validation of CacyBP expression in cancer tissues was performed through lung adenocarcinoma samples using Western blotting and paired t-test. Results: Compared to normal tissues, CACYBP exhibited high expression levels in 14 cancer types, including breast invasive carcinoma, and low expression levels in six cancers, including glioblastoma multiforme (P < 0.05). CACYBP expression was found to be significantly as- sociated with the prognosis of 13 cancers, including adrenocortical carcinoma (P < 0.05). CACYBP demonstrated a robust ability to distinguish 15 cancers, including cholangiocarcinoma, from their control samples (area under the curve > 0.8). Furthermore, CACYBP expression was correlated with tumor mutational burden, microsatellite instabil- ity, and immune infiltration levels, indicating its potential as an exciting target for cancer treatment. CACYBP may exert its effects on several signaling pathways, including cytokine-cytokine receptor interaction, in various cancers. Compared with paired adjacent specimens, the expression level of CacyBP protein was up-regulated in lung adeno- carcinoma specimens (P < 0.05), partially validating the increased expression of CACYBP in cancers. Conclusions: CACYBP has the potential to serve as a novel prognostic and predictive marker for multiple human cancers.

Keywords: Tumor, prognosis, prediction, immunology, biomarker

Introduction

Cancer has emerged as a global issue in public health care and treatment. According to esti- mated data in 2020, the annual number of newly diagnosed cancer cases exceeded 18.1 million, while the number of cancer-related deaths reached 9.6 million around the world [1]. There exist various conventional approach- es to treating cancer, such as surgical interven- tion, radiation therapy, and chemotherapy. Molecular targeted therapy and immunothera- py have been gradually attracting the attention of clinical practitioners [2, 3]. However, for most tumors, there is still a lack of biomarkers that

can accurately predict patient prognosis and cancer status. Therefore, exploring such novel biomarkers suitable for multiple cancers is like- ly to benefit cancer patients.

The human calcyclin binding protein (CacyBP) was encoded by the gene CACYBP. CacyBP is a highly conserved protein with distinct biological functions, and thus plays a significant role in regulating calcium ion signal transduction, cell proliferation, and apoptosis in cells [4, 5]. Based on this, the relationship between CacyBP and the onset and progression of diverse dis- eases, particularly tumors, has been docu- mented. In rectum adenocarcinoma (READ),

CACYBP in multiple cancers

CacyBP levels are elevated in cancer cells but remain undetected in the normal colonic epi- thelium; the protein also promotes the prolifer- ation of colorectal cancer cells [6]. In the con- text of non-small-cell lung cancer (NSCLC), the expression of CacyBP in cancerous tissue is notably elevated, compared to healthy tissue. Furthermore, this protein has been shown to stimulate the proliferation and invasion of NSCLC cells through the regulation of Akt sig- naling pathway [7]. Therefore, CACYBP has been identified as a crucial factor in the pro- gression and growth of diverse malignancies, underscoring its potential as a promising the- rapeutic target for effective cancer manage- ment. However, a comprehensive evaluation of CACYBP across multiple types of cancer has yet to be conducted, thereby necessitating further investigation.

Using a large sample size, this study compre- hensively investigates CACYBP expression and its clinical value in human cancers. In addition, it explores the correlation between CACYBP and immune filtration levels as well as the underlying mechanisms of CACYBP in tumors, thereby enhancing the understanding of CACYBP as a novel prognostic and predictive marker for human neoplasms.

Materials and methods

Collection of public CACYBP mRNA expression, CACYBP protein level, and clinical characteris- tics data

Transcriptome data to evaluate CACYBP mRNA expression in normal tissues (including 8671 samples) were obtained from the Genotype- Tissue Expression database [8], which contains a multitude of samples from Homo sapiens. The Xena database provided access to CACYBP mRNA expression and clinical information for 33 types of cancer from the Cancer Genome Atlas, and the data from 10,080 samples (n = 9358 cancer samples, n = 722 control sam- ples) were collected for this study. The mRNA expression levels were subjected to log2 (x + 1) transformation using R (v4.2.2). Data on immu- nohistochemical staining were obtained from The Human Protein Atlas [9] to detect CacyBP protein levels in cancer and normal tissues. A total of 30 specimens, consisting of 15 cancer tissue specimens and 15 normal tissue speci- mens, were collected from this database for

further analysis. The Xena database was utilized to retrieve the American Joint Committee on Cancer (AJCC) stage, age, gen- der, overall survival (OS), disease-specific sur- vival (DSS), disease-free interval (DFI), and progression-free interval (PFI) of individuals with cancer.

Collection of tumor mutational burden, micro- satellite instability, neoantigen count, and im- mune microenvironment data

Data on tumor mutational burden (TMB), micro- satellite instability (MSI), and neoantigen count for multiple cancers were applied in this research. The TIMER [10, 11] algorithm can predict the immune abundance of B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages, and dendritic cells for cancer patients. The ESTIMATE [12] algorithm is recognized for its ability to detect immune abundance through the following three score categories: stromal (on stromal cells), immune (on immune cells), and ESTIMATE scores (on tumor purity). The TIMER algorithm data were accessible through the official TIMER website, while the ESTIMATE algorithm data were obtained from Sanger Box (version 3.0) [13].

Signaling pathways that may be affected by CACYBP

The Kyoto Encyclopedia of Genes and Genomes (KEGG) database [14, 15] provides information about multiple signaling pathways. The study utilized the “clusterProfiler” package [16] to investigate the potential KEGG signaling path- ways of CACYBP in 33 cancers through gene set enrichment analysis. Those signaling path- ways with a p-value of < 0.05 were selected in this study.

Collection of internal samples and use of Western blotting

The six samples with pathological confirmation collected in this study (three cases of lung ade- nocarcinoma [LUAD] and their corresponding adjacent non-cancerous tissues) were all from the 923 Hospital of Joint Logistics Support Force of Chinese People’s Liberation Army. The basic clinical information of the LUAD and the adjacent non-cancerous tissue samples can be found in Supplementary Material 1.

Figure 1. A detailed overview of the study. CACYBP, calcyclin binding pro- tein. TMB, tumor mutational burden. MSI, microsatellite instability.

CACYBP mRNA expression between control and cancer tissues

Differential CACYBP expression in various cancers

CacyBP protein levels between control and cancer tissues

Relationship of CACYBP with prognosis

The clinical significance of CACYBP in cancers

Relationship of CACYBP with cancer status

The correlations of CACYBP expression with TMB, MSI, neoantigen count, and immune microenvironment

The signaling pathways of CACYBP in multiple cancers

Partial validation of CacyBP expression in cancer tissues through lung adenocarcinoma

Pan-analysis reveals CACYBP to be a novel prognostic and predictive marker for multiple cancers

ed the relative expression level of CacyBP protein.

Statistical analysis

To assess the disparity in CACYBP expression across distinct nor- mal tissues, the Kruskal-Wallis test was employed. The Wilcoxon rank-sum test was utilized to investigate the relevance of CACYBP expression to the ages, genders, and AJCC stages of indi- viduals with neoplasms. To deter- mine the relevance of CACYBP expression to the prognosis of cancer patients, univariate Cox regression analysis and Kaplan- Meier plots were conducted uti- lizing the “survival” and “forest- plot” packages. The optimal cut- point for high- and low-CACYBP expression levels in each Kaplan- Meier curve was evaluated using the “maxstat” and “survminer” packages.

Western blotting was used to validate the pro- tein expression levels of CacyBP in the LUAD tissues and the adjacent non-cancerous tis- sues. Samples were treated with radio immu- noprecipitation assay lysis buffer. After sodium dodecyl sulfate polyacrylamide gel electropho- resis, proteins were transferred onto a polyvi- nylidene fluoride membrane (Servicebio, Wu- han, China). The membrane was then blocked with 5% milk at room temperature (approxi- mately 20℃) for 30 minutes, followed by over- night incubation at 4℃ with a primary antibody. The primary antibody used was an anti-CacyBP antibody (11745-1-AP, Proteintech, Wuhan, China) diluted at a ratio of 1:3000. Afterward, the membrane was incubated with a secondary antibody (GB23303, Servicebio, Wuhan, China) diluted at a ratio of 1:5000 at room tempera- ture for 30 minutes. Chemiluminescence assay was performed on the washed polyvinylidene fluoride membrane using an enhanced chemi- luminescence kit (Servicebio, Wuhan, China). The exposed original images were analyzed and the grayscale values were outputted using AIWBwell™ software (Servicebio, Wuhan, China). The ratio of the grayscale value of CacyBP to the grayscale value of the internal control (a-tubulin, ab7291, ABCAM) represent-

Using the “pROC” package [17] and Stata (v15.0), the area under the curve (AUC) of the receiver operating characteristics (ROC) curve and a summary ROC curve were calculated to assess the accuracy of CACYBP expression in distinguishing between cancers and controls. To assess the associations between CACYBP expression and TMB, MSI, neoantigen, and the immune environment, Spearman’s rank corre- lation analyses were performed. Paired t-test was used to compare the CacyBP protein levels between the LUAD tissues and the adjacent non-cancerous tissues.

Results

A detailed overview of the study is provided in Figure 1.

Dysregulated expression of CACYBP in human neoplasms

Based on the analysis of Genotype Tissue Expression data, the expression of CACYBP exhibited significant variation across different tissues. Specifically, normal tissues of the brain, nerve, and ovary demonstrated high expression of CACYBP, while expression in some tissue types (e.g., the heart, liver, kidney,

Figure 2. CACYBP expression in normal and pan-cancer tissues at mRNA levels. (A) Distinct CACYBP expression in various normal tissues of humans. (B) Different CACYBP expressions between cancer tissues with their normal tis- sues. For (B), each p-value is based on the Wilcoxon rank-sum test; * P < 0.05, ** P < 0.01, *** P < 0.001.

10.0

Kruskal-Wallis, p < 2.2e-16

CACYBP Expression Log2(TPM+1)

7.5

5.0

2.5

0.0

Adipose Tissue

Adrenal Gland

Bladder

Blood

Blood Vessel

Brain

Breast

Cervix Uteri

Colon

Esophagus

Fallopian Tube

Heart

Kidney

Liver

Lung

Muscle

Nerve

Ovary

Pancreas

Pituitary

Prostate

Salivary Gland

Skin

Small Intestine

Spleen

Stomach

Testis

Thyroid

Uterus

Vagina

Group

$ Normal

Tumor

CACYBP expression

BLCA

BRCA

CESC

CHOL

COAD

ESCA

GBM

HNSCC

KICH

KIRC

KIRP

LIHC

LUAD

LUSC

PAAD

PCPG

PRAD

READ

STAD

THCA

UCEC

muscle, and pancreas) was significantly lower (P < 0.05; Figure 2A). Among 20 of the 21 observed cancer types (except for pheochro- mocytoma and paraganglioma), the distribu- tion of CACYBP in cancer tissues was signifi- cantly different from that in control tissues (P < 0.05; Figure 2B). Upregulation of CACYBP expression was observed in 14 cancers, includ- ing bladder urothelial carcinoma (BLCA), breast invasive carcinoma (BRCA), cervical squamous cell carcinoma and endocervical adenocarci- noma (CESC), cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSCC), liver hepatocellular carcinoma (LIHC), LUAD, lung squamous cell carcinoma (LUSC), pancreatic adenocarcinoma (PAAD), READ, stomach adenocarcinoma (STAD), and uterine corpus endometrioid carcinoma (UCEC), while downregulation was observed in glioblastoma multiforme (GBM), kidney chromophobe (KICH), kidney renal clear cell carcinoma (KIRC), kidney renal papillary

cell carcinoma (KIRP), prostate adenocarcino- ma (PRAD), and thyroid carcinoma (THCA) (P < 0.05; Figure 2B).

For the 20 cancers listed above, the CacyBP protein-level data of 15 were available from the Human Protein Atlas. As shown in Figure 3, there was no difference in CacyBP protein lev- els in the BLCA and READ tissues compared to their control tissues. For other cancers, CacyBP expression at the protein level was consistent with that at the mRNA level; CacyBP protein lev- els were increased in BRCA, CESC, COAD, HNSCC, LIHC, LUAD, LUSC, PAAD, STAD, and UCEC, while they were decreased in kidney can- cer (relevant to KICH, KIRC, and KIRP), PRAD, and THCA (Figure 3).

Correlation between expression of CACYBP and clinical parameters

Variations in clinical parameters can lead to divergent prognoses among cancer patients

CACYBP in multiple cancers

Figure 3. The staining intensity of the anti-CacyBP antibodies in 15 cancers. Under the microscope, the staining intensity of the anti-CacyBP antibodies in 12 cancer tissues (except for the bladder urothelial carcinoma [BLCA] and rectum adenocarcinoma [READ]) is stronger than that in normal tissues, while a weaker staining intensity for anti-CacyBP is observed in PRAD tissues compared to normal prostate tissues. Images are available from v21.0.proteinatlas.org. "N" refers to normal tissues, while "T" indicates tumor tissues.

BLCA

BRCA

CESC

HNSCC

COAD

Kidney cancer

LIHC

LUAD

LUSC

PAAD

PRAD

READ

STAD

THCA

UCEC

[18]. A significant correlation (P < 0.05) between CACYBP expression and AJCC stages was observed in nine cancers, including adrenocor- tical carcinoma (ACC), BLCA, BRCA, HNSCC, KICH, KIRP, LIHC, LUAD, and STAD (Supple- mentary Material 2). In these tumor types, advanced AJCC stages tended to represent higher CACYBP expression (P < 0.05, Supple- mentary Material 2). Elevated CACYBP levels were found in young patients (< 65 years old) in

ESCA, LIHC, LUSC, and PAAD (P < 0.05, Supplementary Material 3). Male patients were observed with higher CACYBP expression in four cancers, namely HNSCC, LUAD, READ, and skin cutaneous melanoma (SKCM) (P < 0.05, Supplementary Material 4). The opposite phe- nomenon was found in four other cancers, namely KIRP, brain lower grade glioma (LGG), sarcoma (SARC), and STAD (P < 0.05, Supplementary Material 4).

CACYBP in multiple cancers

Cancer (sample number)	p value	Hazard ratio (95%CI)
ACC (n = 77)	<0.050	1.677 (1.027-2.737)
BLCA (n = 425)	<0.050	1.233 (1.021-1.489)
BRCA (n = 1203)	0.700	1.044 (0.863-1.263)
CESC (n = 304)	<0.050	1.594 (1.060-2.395)
CHOL (n= 45)	0.800	0.948 (0.634-1.417)
COAD (n = 327)	0.400	0.852 (0.581-1.248)
DLBC (n = 47)	0.700	0.845 (0.364-1.959)
ESCA (n = 194)	0.700	1.068 (0.783-1.457)
GBM (n = 152)	0.900	0.961 (0.634-1.457)
HNSCC (n = 561)	0.700	1.037 (0.867-1.241)
KICH (n = 89)	0.100	2.147 (0.910-5.066)
KIRC (n = 605)	<0.050	0.717 (0.595-0.864)
KIRP (n = 319)	<0.050	2.502 (1.636-3.828)
LAML (n = 161)	0.300	1.204 (0.851-1.703)
LGG (n = 507)	0.600	0.902 (0.601-1.352)
LIHC (n = 418)	0.100	1.165 (0.986-1.375)
LUAD (n = 563)	<0.050	1.457 (1.195-1.776)
LUSC (n = 542)	0.100	0.866 (0.731-1.025)
MESO (n = 86)	<0.050	1.505 (1.044-2.170)
OV (n = 417)	<0.050	0.857 (0.736-0.998)
PAAD (n = 182)	0.100	1.296 (0.949-1.769)
PCPG (n = 180)	0.100	2.888 (0.724-11.525)
PRAD (n = 547)	0.300	1.815 (0.603-5.465)
READ (n = 101)	0.100	0.595 (0.299-1.182)
SARC (n = 260)	0.200	1.181 (0.910-1.532)
SKCM (n = 103)	0.400	1.168 (0.794-1.716)
STAD (n = 443)	0.600	0.954 (0.785-1.161)
TGCT (n = 132)	0.500	0.689 (0.209-2.267)
THCA (n = 563)	0.100	2.190 (0.894-5.364)
THYM (n = 120)	0.600	1.322 (0.445-3.930)
UCEC (n = 192)	0.700	0.934 (0.680-1.283)
UCS (n = 57)	0.900	0.973 (0.611-1.548)
UVM (n = 79)	0.100	1.664 (0.946-2.926)

C
Cancer (sample number)	p value	Hazard ratio (95%CI)
ACC (n = 75)	0.100	1.610 (0.974-2.662)
BLCA (n = 410)	<0.050	1.373 (1.089-1.731)
BRCA (n = 1174)	0.500	1.103 (0.847-1.436)
CESC (n = 303)	0.100	1.442 (0.925-2.246)
CHOL (n= 43)	1.000	0.990 (0.639-1.534)
COAD (n = 312)	0.800	1.058 (0.619-1.807)
DLBC (n = 47)	0.300	0.529 (0.163-1.711)
ESCA (n = 191)	0.600	1.105 (0.737-1.656)
GBM (n = 139)	1.000	0.989 (0.638-1.535)
HNSCC (n = 531)	0.800	1.038 (0.819-1.314)
KICH (n = 89)	<0.050	3.197 (1.145-8.928)
KIRC (n = 588)	<0.050	0.686 (0.549-0.858)
KIRP (n = 315)	<0.050	3.722 (2.258-6.136)
LGG (n = 499)	0.400	0.836 (0.549-1.272)
LIHC (n = 407)	<0.050	1.357 (1.086-1.696)
LUAD (n = 527)	<0.050	1.427 (1.111-1.833)
LUSC (n = 480)	0.800	0.966 (0.735-1.270)
MESO (n = 66)	<0.050	1.852 (1.175-2.918)
OV (n = 387)	<0.050	0.837 (0.713-0.983)
PAAD (n = 176)	<0.050	1.507 (1.051-2.163)
PCPG (n = 180)	0.100	4.624 (0.643-33.273)
PRAD (n = 545)	<0.050	9.059 (1.689-48.582)
READ (n = 95)	0.800	1.195 (0.313-4.560)
SARC (n = 254)	0.200	1.194 (0.898-1.587)
SKCM (n = 103)	0.100	1.551 (0.921-2.613)
STAD (n = 419)	0.100	0.825 (0.638-1.067)
TGCT (n = 132)	0.400	0.624 (0.200-1.943)
THCA (n = 557)	0.200	2.365 (0.628-8.902)
THYM (n = 120)	0.900	0.933 (0.244-3.569)
UCEC (n = 190)	0.700	0.925 (0.628-1.362)
UCS (n = 55)	0.900	1.028 (0.635-1.665)
UVM (n = 79)	0.100	1.833 (0.986-3.405)

Figure 4. Relation of CACYBP expression with overall survival (A and B) and disease-specific survival (C and D) of cancer patients. (A and C) Univariate Cox regression analysis. (B and D) Kaplan-Meier curves with p-values based on log-rank tests.

ACC

BLCA

CESC

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

0.75

0.50

0.25

p = 0.00081

0.25

p = 0.036

0.25

p = 0.012

0.00

0.00-

0.00

2.5

7.5

12.5

Time (Years)

KIRC

KIRP

Survival probability

1.00

Survival probability

1.00

0.75

0.50

0.25

p < 0.0001

0.25

p < 0.0001

0.00

2.5

7.5 10 12.5

Time (Years)

LUAD

MESO

Survival probability

1.00

Survival probability

1.00

0.75

0.50

0.25

p < 0.0001

0.25

0.00028

0.00

0.25

1.0 2.0 4.0 8.0

Time (Years)

Overall survival

BLCA

KICH

KIRC

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

0.75

0.75-

0.50

0.25

p = 0.0073

0.25

p = 0.00057

0.25-

p < 0.0001

0.00

0.00-

2.5

5 7.5 10

2.5

7.5

Time (Years)

12.5

Time (Years)

KIRP

LIHC

LUAD

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

0.75

.0.50

0.50

0.25

p < 0.0001

0.25

p < 0.0001

0.25

p < 0.0001

0.00

0.00-

2.5

7.5

Time (Years)

MESO

PAAD

PRAD

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

0.75

0.50

0.25

0.00027

0.25

= 0.03

037

0.25

p = 0.00019

0.00

0.25 1.0 4.0 16.0

Time (Years)

Disease specific survival

The prognostic significance of CACYBP expres- sion

For OS and/or DSS, high CACYBP expression was significantly correlated with shortened sur- vival time in patients suffering from ACC, BLCA, CESC, KICH, KIRP, LIHC, LUAD, MESO, PAAD,

and PRAD and associated with better progno- sis in patients with KIRC (P < 0.05, Figure 4). With regard to DFI and PFI, increased CACYBP expression was relevant to the poor prognosis in patients with ACC, BLCA, CESC, KIRP, LIHC, LUAD, MESO, SKCM, and uveal melanoma (UVM), while it demonstrated a favorable prog-

CACYBP in multiple cancers

A Cancer (sample number)	p value	Hazard ratio (95%CI)
ACC (n = 44)	0.300	1.458 (0.718-2.960)
BLCA (n = 193)	0.500	1.194 (0.732-1.947)
BRCA (n = 1028)	0.600	1.086 (0.817-1.444)
CESC (n = 176)	0.100	1.894 (0.952-3.770)
CHOL (n= 32)	0.500	1.221 (0.685-2.175)
COAD (n = 122)	0.200	1.647 (0.721-3.762)
DLBC (n = 27)	0.400	0.416 (0.050-3.484)
ESCA (n = 93)	0.700	1.118 (0.642-1.949)
HNSCC (n = 134)	0.200	1.446 (0.781-2.678)
KICH (n = 42)	0.900	1.052 (0.335-3.309)
KIRC (n = 132)	0.300	1.668 (0.692-4.021)
KIRP (n = 186)	<0.050	2.375 (1.386-4.070)
LGG (n = 133)	1.000	0.975 (0.431-2.206)
LIHC (n = 353)	0.100	1.164 (0.988-1.372)
LUAD (n = 336)	<0.050	1.455 (1.083-1.955)
LUSC (n = 323)	0.600	1.090 (0.784-1.516)
MESO (n = 15)	0.300	2.613 (0.483-14.128)
OV (n = 203)	0.100	0.807 (0.636-1.023)
PAAD (n = 72)	0.100	2.502 (0.871-7.192)
PCPG (n = 159)	0.800	1.140 (0.307-4.238)
PRAD (n = 383)	0.600	0.880 (0.507-1.527)
READ (n = 32)	0.900	1.123 (0.224-5.631)
SARC (n = 154)	0.600	1.079 (0.780-1.492)
STAD (n = 261)	0.100	0.757 (0.519-1.105)
TGCT (n = 103)	0.700	1.122 (0.653-1.927)
THCA (n = 395)	0.700	1.116 (0.577-2.158)
UCEC (n = 128)	0.900	0.984 (0.663-1.460)
UCS (n = 27)	0.300	0.669 (0.307-1.459)

0.0620.250 1.00 4.00

Disease free interval

B Cancer (sample number)	p value	Hazard ratio (95%CI)
ACC (n = 77)	<0.050	1.617 (1.103-2.370)
BLCA (n = 426)	<0.050	1.376 (1.133-1.672)
BRCA (n = 1203)	0.500	1.082 (0.881-1.329)
CESC (n = 307)	<0.050	1.749 (1.166-2.623)
CHOL (n = 45)	0.700	0.935 (0.625-1.399)
COAD (n = 327)	0.700	1.076 (0.756-1.532)
DLBC (n = 47)	0.500	1.325 (0.637-2.756)
ESCA (n = 194)	0.200	1.230 (0.914-1.656)
GBM (n = 152)	0.100	0.735 (0.511-1.057)
HNSCC (n = 561)	0.100	1.163 (0.953-1.419)
KICH (n = 89)	0.200	1.612 (0.790-3.289)
KIRC (n = 600)	<0.050	0.782 (0.642-0.951)
KIRP (n = 318)	<0.050	2.038 (1.391-2.987)
LGG (n = 507)	0.200	0.811 (0.599-1.097)
LIHC (n = 418)	<0.050	1.218 (1.053-1.409)
LUAD (n = 563)	<0.050	1.341 (1.111-1.618)
LUSC (n = 543)	0.600	0.946 (0.767-1.166)
MESO (n = 84)	<0.050	1.584 (1.080-2.324)
OV (n = 417)	0.100	0.870 (0.750-1.009)
PAAD (n = 182)	0.100	1.285 (0.933-1.770)
PCPG (n = 180)	0.800	1.094 (0.623-1.922)
PRAD (n = 547)	0.700	1.065 (0.762-1.488)
READ (n = 101)	0.900	1.055 (0.534-2.082)
SARC (n = 260)	0.100	1.219 (0.975-1.525)
SKCM (n = 103)	<0.050	1.551 (1.063-2.261)
STAD (n = 445)	0.100	0.837 (0.679-1.033)
TGCT (n = 132)	0.900	1.024 (0.653-1.605)
THCA (n = 563)	0.400	1.207 (0.748-1.947)
THYM (n = 120)	0.500	1.225 (0.668-2.245)
UCEC (n = 192)	0.500	0.906 (0.695-1.180)
UCS (n = 57)	0.900	1.016 (0.652-1.582)
UVM (n = 78)	<0.050	2.492 (1.391-4.463)

0.50

1.0

2.0

4.0

Progression free interval

Figure 5. Relation of CACYBP expression with disease-free interval (A and C) and progression-free interval of cancer patients (B and D). (A and B) Univariate Cox regression analysis. (C and D) Kaplan-Meier curves with p-values based on log-rank tests.

KIRP

ACC

BLCA

CESC

KIRC

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00-

Survival probability

1,00-

0.75

0.75-

0.50

5. 0.50

0.50

0.25

p < 0.0001

0.25

0.002

0.25

p = 0.011

0.25

.00065

0.25

p = 0.00018

0.00

2.5

7.5

2.5

7.5

12.5

Time (Years)

LUAD

KIRP

LIHC

LUAD

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

0.75

0.50

0.25

0015

0.25

p < 0.0001

0.25

0.0002

0.25

p < 0.0001

0.00

2.5

7.5

Time (Years)

MESO

SKCM

UVM

Survival probability

1.00

Survival probability

1.00

Survival probability

1.00

0.75

0.50

0.25

0.00055

0.25

= 0.0029

0.25

p < 0.0001

0.00

Time (Years)

nosis in patients with KIRC (P < 0.05, Figure 5). Thus, CACYBP represented a risk prognosis fac- tor for most cancers.

Predictive significance of CACYBP expression

The prediction of cancer status in patients is of great clinical importance in cancer manage- ment. This study highlights the efficacy of CACYBP in distinguishing between cancer tis-

sues and control tissues in 15 out of 21 can- cers (AUC > 0.80, Figure 6A). A comprehensive analysis of the 21 cancers also reveals that CACYBP expression can distinguish between cancer patients and healthy individuals (AUC = 0.95, 95% CI: 0.92-0.96; Figure 6B). Remarka- bly, CACYBP expression demonstrated excep- tional diagnostic accuracy in eight cancers, namely CESC, CHOL, ESCA, GBM, HNSCC,

CACYBP in multiple cancers

Figure 6. The ability of CACYBP expression to distinguish these cancer tissues from their normal tissues in pan-can- cer. A. In receiver operating characteristic curves, CACYBP demonstrated pronounced effects in identifying cancer tissues from their control counterparts based on the area under the receiver curve (AUC). B. In summary receiver operating characteristic curves, the CACYBP expression can accurately identify 21 types of cancer samples from their controls. C. In summary receiver operating characteristic curves, the CACYBP expression can accurately iden- tify eight types of cancer samples from their controls.

BLCA (n = 19 vs 407)

BRCA (n = 113 vs 1092)

CESC (n = 3 vs 304)

CHOL (n =9 vs 36)

COAD (n = 41 vs 288)

0.8

Sensitivity

0.4

AUC: 0.825

0.4

AUC: 0.871

0.4

AUC: 0.961

0.4

AUC: 0.988

0.4

AUC: 0.894

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

1 - Specificity

ESCA (n = 13 vs 181)

GBM (n = 5 vs 153)

HNSCC (n = 44 vs 518)

KICH (n = 25 vs 66)

KIRC (n = 72 vs 530)

0.8

Sensitivity

0.4

AUC: 0.919

0.4

AUC: 0.936

0.4

AUC: 0.904

0.4

AUC: 0.944

0.4

AUC: 0.630

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

1 - Specificity

KIRP (n = 32 vs 288)

LIHC (n = 50 vs 369)

LUAD (n = 59 vs 513)

LUSC (n = 50 vs 498)

PAAD (n = 4 vs 178)

0.8

Sensitivity

0.4

AUC: 0.651

0.4

AUC: 0.931

0.4

AUC: 0.867

0.4

AUC: 0.950

0.4

AUC: 0.815

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

1 - Specificity

PCPG (n = 3 vs 177)

PRAD (n = 52 vs 495)

READ (n = 10 vs 92)

STAD (n = 36 vs 414)

THCA (n = 59 vs 504)

0.8

Sensitivity

0.4

AUC: 0.740

0.4

AUC: 0.685

0.4

AUC: 0.870

0.4

AUC: 0.846

0.4

AUC: 0.610

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

0.0

0.4

0.8

1 - Specificity

UCEC (n = 23 vs 180)

1.0

0.8

Sensitivity

0.4

AUC: 0.792

Sensitivity

0.0

0.5

0.0

0.4

0.8

1 - Specificity

Observed Data

Summary Operating Point

SENS 0.73 [0.65 0.80]

Summary Operating Point

SPEC = 0.95 [0.92 - 0.97]

SENS 0.86 [0.82 0.90]

SPEC = 0.95 [0.91 - 0.98]

SROC Curve

AUC = 0.95 [0.92 - 0.96]

SROC Curve

AUC = 0.97 [0.95 - 0.98]

- 95% Confidence Contour

95% Prediction Contour

·· 95% Prediction Contour

0.0

1.0

0.5

Specificity

0.0

1.0

0.5

Specificity

0.0

KICH, LIHC, and LUSC (AUC > 0.90, Figure 6A), indicating the significant potential of CACYBP as a predictor for those cancers, with an AUC

value of 0.97 (Figure 6C). Therefore, CACYBP may be a valuable marker for predicting the cancer status of specific neoplasm types.

CACYBP in multiple cancers

Figure 7. Correlations of CACYBP expression with TMB, MSI, and neoantigen count. A. CACYBP expression was positively correlated with TMB. B. CACYBP expression was positively correlated with MSI. C. CACYBP expression is positively associated with neoantigen count in four cancers.

LGG* UVM STAD ***

THCA*

OV ** DLBC ** ÚCEC **

LUAD ***

THCA*

THYM*

DLBC

0.4

PAAD ***

LAML

0.5

SARC **

KIRP

0.2

BLCA ***

LUAD

0.25

STAD ***

ESCA

SARC **

BRCA

UVM

TGCT

HNSCC ***

PRAD

UCS

-0.

-d

LAML

THYM*

ESCA

GBM

-0.4

KIRC

UCEC **

PCPG

CESC*

CESC

BRCA ***

LUSC

COAD*

GBM

ACC

HNSCC

PAAD

LIHC

LUSC ***

LGG

MESO

UCS

COAD*

TGCT

ACC

PCPG

MESO

KIRP

KIRC

READ

SKCM

PRAD*

KICH

KICH CHOL

LIHC

CHOL BLCA

SKCM

READ

C THYM (n = 64)

UCEC (n = 166)

COAD (n = 255)

PAAD (n = 113)

CACYBP Log2(TPM+1)

= 0.25

047

CACYBP Log2(TPM+1)

= 0.23, p = 0.003

CACYBP Log2(TPM+1)

= 0.22, p = 0.00036

CACYBP Log2(TPM+1)

p = 0.2, p = 0.031

Log2(neoantigen count + 1)

Association between CACYBP expression and TMB, MSI, neoantigen, and the immune micro- environment

A significant positive correlation was noted between the levels of CACYBP expression and TMB in BLCA, HNSCC, LUAD, PAAD, SARC, STAD, THYM, and UCEC (p > 0.2, P < 0.05, Figure 7A). The expression levels of CACYBP were positive- ly relevant to MSI in UCEC (p > 0.2, P < 0.05) and negatively associated with MSI in DLBC (p < - 0.2, P < 0.05, Figure 7B). CACYBP expres- sion had a mild positive relationship with neo- antigen number in THYM, UCEC, COAD, and PAAD (p > 0.2, P < 0.05, Figure 7C).

The TIMER and ESTIMATE data were utilized to evaluate the association between CACYBP expression and the tumor immune microenvi- ronment. There was a positive association between the expression levels of CACYBP and almost all of B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages, and dendritic cells

in ACC, KICH, KIRC, PRAD, and THCA (P < 0.05, Figure 8). However, decreased CACYBP ex- pression exhibited a significant negative corre- lation with all immune, stromal, and estimate scores in some cancers-particularly LGG, LUSC, SARC, SKCM, TGCT, and UCEC (P < 0.05, Figure 9). These results demonstrate that CACYBP may affect the immune microenviron- ment through distinct aspects in different cancers.

Underlying signaling pathways of CACYBP

KEGG signaling pathways were utilized to exploit the underlying mechanisms of CACYBP in 32 types of human cancer. As shown in Figure 10A, CACYBP plays a crucial role in the onset and development of ESCA, LGG, and LUSC through complex mechanisms, since it was found to affect multiple signaling path- ways. The analysis identified 11 KEGG signal- ing pathways associated with CACYBP, such as “olfactory transduction”, “neuroactive ligand

CACYBP in multiple cancers

Figure 8. Relevance of CACYBP expression with immune infiltration levels. The Spearman correlation coefficient follows the letter "p".

ACC

CACYBP Log2(TPM+1)

4 -

€

0.44.

7.4e-05

= 0.

.p=

084

.32

= 0.0045

0.013

O .42

0.00018

0.11

0.12

0.13

0.07

0.20

0.25

0.30

0.35

0.12

0.14

0.16

0.18

0.08

0.16

B_cell level

0.09 0.11 0.13 0.15

CD4_Tcell level

CD8_Tcell level

Neutrophil level

0.12

Macrophage level

0.49 0.50 0.51 0.52 0.53

Dendritic level

KICH

CACYBP Log2(TPM+1)

p =0.39, p = 0.0011

011

0.93

p = 0

p = 3.4e-07

p = 0

18.00 = 0.15

= 0.44, p = 0.00021

.P = 0.0034

0.08

0.10

0.12

0.14

0.08

0.12 0.16

0.20

0.24

0.10

0.15

CD4_Tcell level

0.20

0.25

CD8_Tcell level

0.10

0.12

0.14

0.00

0.05

B_cell level

Neutrophil level

5 0.10 0

0.15

0.20

Macrophage level

0.45 0.50 0.55 0.60

Dendritic level

KIRC

CACYBP Log2(TPM+1)

p = 0.33, p = 1.1e-14

p = 0.088, p = 0.043

= 0.26, p = 5.8e-10

p = 0.34, p = 9.9e-16

34,

: 8.2e-16

38, p < 2.2e-16

0.0

0.2

0.4

0.6

0.0

0.2

0.4

0.6

0.8

0.0

0.5

1.0

1.5

0.0

0.1

2 0

0.2

0.3

0.4

0.5

0.0

0.2

0.4

0.6

0.8

0.0

0.5

1.0

1.5

B_cell level

CD4_Tcell level

2.0

CD8_Tcell level

Neutrophil level

Macrophage level

Dendritic level

PRAD

CACYBP Log2(TPM+1)

3 -

= 0

2.2e-

0.045. p = 0.31

O 45

2.2e

D= 0.22, p = 1,3e-06

0.25, p = 2e-08

29,

4.6e

0.00 0.25 0.50 0.75 1.00

0.0

0.3

0.6

0.9

0.0

0.1

.2 0.3 0.4 0

0.5

0.2

0.4

0.6

0.0

0.1

0.2

0.3

0.4

0.0

0.5

1.0

1.5

2.0

B_cell level

CD4_Tcell level

CD8_Tcell level

Neutrophil level

Macrophage level

Dendritic level

THCA

CACYBP Log2(TPM+1)

2.2e-1 2.

0.48

2.2e-16

0.32

DE1

le-13

= 0

0.34. p = 6.8e-15

0.54, p < 2.28-16

p = 1.7e-14

0.0

0.2

0.4

0.6

0.8

0.0 0.1 0.2 0.3 0.4 0.5

CD4_Tcell level

0.00

0.25

0.50

0.75

1.00

0.1

0.2

0.3

0.0

0.1

0.2

0.3

0.4

0.8

CD8_Tcell level

1.2

Neutrophil level

Macrophage level

Dendritic level

1.6

B_cell level

receptor interaction”, “cytokine-cytokine recep- tor interaction”, and “calcium signaling path- way”, in various cancers (Figure 10B).

Partial validation of CacyBP expression in can- cer tissues through LUAD samples

As shown above, CACYBP is highly expressed in various tumors. Considering that lung cancer is the leading cause of cancer-related death worldwide, this study explored the expression of the CacyBP protein in lung adenocarcinoma to verify its expression in cancer tissue using internal samples. Compared with paired adja- cent specimens, the expression level of Cacybp protein was up-regulated in the LUAD speci- mens (Figure 10C). This result was statistically validated by a significance test (P < 0.05, Figure 10D).

Discussion

Cancer remains a pressing issue in the global public health landscape. The absence of reli- able biomarkers presents a considerable hur- dle in the care of cancer patients. While CACYBP has been identified as a biomarker for various cancers, a comprehensive investigation into its potential as a pan-cancer biomarker is lacking and warrants further study.

In the current study, we conducted a pan-can- cer analysis using 10,080 samples to enhance the understanding of the clinical value of CACYBP. Our findings revealed both upregulat- ed expression of CACYBP in 14 cancers (e.g., BLCA) and downregulated expression of CACYBP in six cancers (e.g., GBM); almost all of these expression trends were verified at the

Figure 9. Relevance of CACYBP expression with immune microenvironment scores. The Spearman correlation coef- ficient follows the letter "p".

LGG

LUSC

CACYBP Log2(TPM+1)

₹

₼

-0.36, p < 2.28-16

p =- 0.49, p < 2.2e-16

p = - 0.46, p < 2.2e-16

-0.27.

=. 7

9e-10

0.29

1.1e-10

p =- 0.3, p = 2.5e-11

-2000 -1000

1000

-1000

1000 2000

-2000

2000

1000

ESTIMATE_score

-2000 -1000

-1000

Stromal_score

1000 2000 3000

-2000 0 2000 4000

Stromal_score

Immune_score

ESTIMATE_score

SARC

SKCM

CACYBP Log2(TPM+1)

- 0.34, p = 1.6e-08 0

0.25, 0=

1e-05

p =

1.3, P=

.58-07

.34,

0.00059

p = 0.32, p= 0:0012

0.33

00068

-1000

1000 2000

-2000

Stromal_score

-20001000 0 100020003000

Immune_score

2000 4000

ESTIMATE_score

-1500 -1000 -500

-1000

1000

2000

-2000-1000

Stromal_score

1000 2000

Immune_score

ESTIMATE_score

TGCT

UCEC

CACYBP Log2(TPM+1)

0.54

2.2e-

0.19

.034

0.37

8e-05

3e-06

-2000

-1000

1000

-1000

1000 2000 3000

-2000

2000

4000

-1000

Stromal_score

Immune_score

ESTIMATE_score

-200015001000-500 0 500

1000 2000 3000

-2000

Stromal_score

2000

Immune_score

ESTIMATE_score

protein level. CACYBP was identified as a signifi- cant prognostic indicator for patients with 11 cancer types (ACC, and so on) in terms of OS and DSS. For DFI and PFI, increased CACYBP expression was associated with the prognosis of patients in 10 cancers (e.g., BLCA). CACYBP was also demonstrated as a predictive marker of cancer status for 15 cancers (e.g., CESC). Furthermore, the associations observed be- tween CACYBP expression and TMB, MSI, neo- antigen, and the immune microenvironment indicate that it may be an attractive target for multiple neoplasms.

For various types of cancers, the expression of CACYBP varies between cancer and control groups. Previous studies identified differential- ly expressed CACYBP/CacyBP in certain can- cers, with increased expression in BLCA, BRCA, COAD, glioblastoma, LUAD, NSCLC, osteosar- coma, PAAD, prostate cancer, and STAD [6, 7, 19-25] and decreased expression in KIRC and chronic lymphocytic leukemia cells [26, 27]. Our study identified the results for BLCA, COAD, LUAD, PAAD, STAD, and KIRC. Additionally, it revealed overexpression of CACYBP in CESC, CHOL, ESCA, HNSCC, LIHC, LUSC, READ, and UCEC and low expression of CACYBP in GBM, KICH, KIRP, PRAD, and THCA. CacyBP protein levels were consistent with most of these can- cers, based on our study. Among the listed can-

cers, the expression of CACYBP/CacyBP in BRCA remains disputed; one report identified low levels of CACYBP/CacyBP in BRCA [28], while another study [29] and our research dis- played upregulated CACYBP mRNA and CacyBP protein expression levels in this disease. Unfortunately, the mechanisms of CACYBP expression in BRCA have not been explored, warranting further study. In summary, upregu- lation of CACYBP/CacyBP expression was observed in most cancers, while the opposite was found in several cancers.

CACYBP has been found to have varying prog- nostic implications and can serve as an excel- lent predictive marker for cancer status in cer- tain types of cancers. Previous studies demon- strated a correlation between elevated CACYBP expression and unfavorable prognosis in indi- viduals with BLCA, BRCA, glioblastoma, LUAD, or osteosarcoma [20, 22-24, 30]. In our study, we investigated the potential prognostic roles of CACYBP in multiple human cancers by ana- lyzing survival data, including OS, DSS, DFI, and PFI. Our results showed overexpression of CACYBP to be associated with unfavorable OS and/or DSS in patients with ACC, BLCA, CESC, KICH, KIRP, LIHC, LUAD, MESO, PAAD, or PRAD, while it was linked to a favorable OS and/or DSS in patients with KIRC. Increased CACYBP expression was associated with a good progno- sis in patients with KIRC, while it was also relat-

CACYBP in multiple cancers

Figure 10. The potential mechanisms of CACYBP in pan-cancer and the protein expression of CacyBP in lung adeno- carcinoma (LUAD). A. Gene set enrichment analysis suggests that CACYBP is identified to involve no fewer than five signaling pathways in some cancers. B. All signaling pathways that CACYBP may affect in cancers. C. The protein expression levels of CacyBP in LUAD detected by Western blot. D. Comparison of protein expression levels of CacyBP in LUAD and adjacent non-cancerous tissues. Paired t-test is used in this plot.

ESCA

KEGG_ALDOSTERONE_REGULATED_SODIUM_REABSORPTION

0.00

Running Enrichment Score

KEGG_CALCIUM_SIGNALING_PATHWAY

Olfactory transduction

KEGG_NATURAL_KILLER_CELL_MEDIATED_CYTOTOXICITY

Neuroactive ligand receptor interaction

-0.25

KEGG_NEUROACTIVE_LIGAND_RECEPTOR_INTERACTION

-0.50

KEGG_OLFACTORY_TRANSDUCTION

Cytokine-cytokine receptor interaction

-0.75

Signaling pathway

Ribosome

Maturity onset diabetes of the young

-1.00

Hematopoietic cell lineage

Taste transduction

Ranked List Metric

Retinol metabolism

Pentose and glucuronate interconversions

-10

Calcium signaling pathway

10000

20000

30000

40000

Rank in Ordered Dataset

50000

Asthma

LGG

KEGG_ASTHMA

Cancer type count

1.0

Running Enrichment Score

KEGG_CALCIUM_SIGNALING_PATHWAY

0.5

KEGG_CYTOKINE_CYTOKINE_RECEPTOR_INTERACTION

KEGG_JAK_STAT_SIGNALING_PATHWAY

0.0

KEGG_NEUROACTIVE_LIGAND_RECEPTOR_INTERACTION

-0.5

-1.0

CacyBP

Ranked List Metric

a-tubulin

-10

10000

20000

30000

Rank in Ordered Dataset

40000

50000

LUSC

KEGG_ALPHA_LINOLENIC_ACID_METABOLISM

0.00

Running Enrichment Score

KEGG_ETHER_LIPID_METABOLISM

Internal cohrt

KEGG_HEMATOPOIETIC_CELL_LINEAGE

-0.25

KEGG_LINOLEIC_ACID_METABOLISM

CacyBP protein expression

-0.50

KEGG_OLFACTORY_TRANSDUCTION

1.1

p = 0.007

1.0

-0.75

-1.00

0.9

0.8

Ranked List Metric

0.7

-10

Control

LUAD

10000

20000

30000

40000

50000

Rank in Ordered Dataset

ed to a poor prognosis in patients in nine can- cers, namely ACC, BLCA, CESC, KIRP, LIHC, LUAD, MESO, SKCM, and UVM. Additionally, our study revealed that CACYBP could distinguish between 15 types of cancer tissues and their

normal tissues; this novel finding highlights the significant predictive value of CACYBP in can- cers. Therefore, CACYBP may be an excellent prognostic and predictive biomarker for multi- ple cancers.

CACYBP in multiple cancers

The mechanisms of CACYBP in cancers remain complex and unclear. TMB and MSI are consid- ered to be effective biomarkers for various tumors because they aid in the diagnosis and treatment of cancers [31, 32]. CACYBP expres- sion levels have been associated with TMB and/or MSI in BLCA, DLBC, HNSCC, LUAD, PAAD, SARC, STAD, THYM, and UCEC, suggest- ing that the gene may play a role in influencing TMB and MSI levels in these tumors. Additionally, CACYBP expression levels have been found to be correlated with neoantigen count in patients with COAD, PAAD, THYM, and UCEC, indicating that the gene may also impact the immune microenvironment of certain tumors. Notably, our study also revealed a sig- nificant positive association between CACYBP expression and the immune microenvironment. On the one hand, the expression of CACYBP showed a correlation with the infiltration levels of almost all six types of immune cells, namely B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages, and dendritic cells, in ACC, KICH, KIRC, PRAD, and THCA. This suggests that the gene may act as an oncogene contributing to the stimulation of immune response [20]. On the other hand, in LGG, LUSC, SARC, SKCM, TGCT, and UCEC, there was a negative correla- tion between CACYBP expression and immune, stromal, and estimate scores, which represents a prognosis risk factor for cancer patients and affects the immune response. These findings highlight that CACYBP may affect the immune microenvironment through distinct aspects in different cancers. Moreover, CACYBP may play a role in affecting up to 11 signaling pathways, such as “olfactory transduction”, “neuroactive ligand receptor interaction”, “cytokine-cytokine receptor interaction”, and “calcium signaling pathway”. Even in single cancer, CACYBP may play a crucial role in the occurrence and devel- opment of ESCA, LGG, and LUSC through sev- eral signaling pathways. These findings provide a clue for further experimental validation of the potential mechanisms of CACYBP in cancers.

In this study, some limitations should be noted. Initially, we failed to collect body fluid-related samples (e.g., blood specimens) to detect the ability of CACYBP expression in directly screen- ing cancer patients from individuals without cancers. The sample size for investigating CacyBP protein levels was relatively small. It

would be worthwhile to conduct experimental validation of the underlying mechanisms of CACYBP.

In summary, the expression of CACYBP varies across different types of human can- cers, and this gene may be utilized as a prog- nostic and predictive marker for multiple can- cer types.

Acknowledgements

The results shown in the study are based upon data generated by the TCGA, GTEx, The Human Protein Atlas, and Sanger Box (version 3.0). Data on the internal cohort can be obtained from the corresponding author.

Disclosure of conflict of interest

The authors declare that the research was con- ducted in the absence of any commercial or financial relationships that could be considered as a potential conflict of interest.

Address correspondence to: Yuesong Wu, Depart- ment of Cardiothoracic Surgery, The 923 Hospital of Joint Logistics Support Force of Chinese People’s Liberation Army, No. 52, Zhiwu Road, Qingxiu Dis- trict, Nanning 530021, Guangxi Zhuang Autonomous Region, China. E-mail: wuyuesong303@163.com

References

[1] Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A and Bray F. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 2021; 71: 209-249.

[2] Miller KD, Fidler-Benaoudia M, Keegan TH, Hipp HS, Jemal A and Siegel RL. Cancer statis- tics for adolescents and young adults, 2020. CA Cancer J Clin 2020; 70: 443-459.

[3] Tariq S, Kim SY, Monteiro de Oliveira Novaes J and Cheng H. Update 2021: management of small cell lung cancer. Lung 2021; 199: 579- 587.

[4] Lian YF, Huang YL, Zhang YJ, Chen DM, Wang JL, Wei H, Bi YH, Jiang ZW, Li P, Chen MS and Huang YH. CACYBP enhances cytoplasmic re- tention of P27(Kip1) to promote hepatocellular carcinoma progression in the absence of RNF41 mediated degradation. Theranostics 2019; 9: 8392-8408.

[5] Topolska-Wos AM, Chazin WJ and Filipek A. CacyBP/SIP — structure and variety of functions. Biochim Biophys Acta 2016; 1860: 79-85.

CACYBP in multiple cancers

[6] Zhai H, Shi Y, Chen X, Wang J, Lu Y, Zhang F, Liu Z, Lei T and Fan D. CacyBP/SIP promotes the proliferation of colon cancer cells. PLoS One 2017; 12: e0169959.

[7] Xu YJ, Hu YM, Qin C, Wang F, Cao W, Yu YW, Zhao L, Li J, Chen WQ, Li N and He J. CacyBP promotes the proliferation and invasion of non- small cell lung cancer. Zhonghua Zhong Liu Za Zhi 2021; 43: 924-931.

[8] GTEx Consortium. The Genotype-Tissue Ex- pression (GTEx) project. Nat Genet 2013; 45: 580-585.

[9] Ghandi M, Huang FW, Jané-Valbuena J, Kryukov GV, Lo CC, McDonald ER 3rd, Barretina J, Gelfand ET, Bielski CM, Li H, Hu K, Andreev- Drakhlin AY, Kim J, Hess JM, Haas BJ, Aguet F, Weir BA, Rothberg MV, Paolella BR, Lawrence MS, Akbani R, Lu Y, Tiv HL, Gokhale PC, de Weck A, Mansour AA, Oh C, Shih J, Hadi K, Rosen Y, Bistline J, Venkatesan K, Reddy A, Sonkin D, Liu M, Lehar J, Korn JM, Porter DA, Jones MD, Golji J, Caponigro G, Taylor JE, Dunning CM, Creech AL, Warren AC, McFarland JM, Zamanighomi M, Kauffmann A, Stransky N, Imielinski M, Maruvka YE, Cherniack AD, Tsherniak A, Vazquez F, Jaffe JD, Lane AA, Weinstock DM, Johannessen CM, Morrissey MP, Stegmeier F, Schlegel R, Hahn WC, Getz G, Mills GB, Boehm JS, Golub TR, Garraway LA and Sellers WR. Next-generation characteriza- tion of the cancer cell line encyclopedia. Nature 2019; 569: 503-508.

[10] Li T, Fan J, Wang B, Traugh N, Chen Q, Liu JS, Li B and Liu XS. TIMER: a web server for compre- hensive analysis of tumor-infiltrating immune cells. Cancer Res 2017; 77: e108-e110.

[11] Li T, Fu J, Zeng Z, Cohen D, Li J, Chen Q, Li B and Liu XS. TIMER2.0 for analysis of tumor-in- filtrating immune cells. Nucleic Acids Res 2020; 48: W509-W514.

[12] Yoshihara K, Shahmoradgoli M, Martinez E, Vegesna R, Kim H, Torres-Garcia W, Trevino V, Shen H, Laird PW, Levine DA, Carter SL, Getz G, Stemke-Hale K, Mills GB and Verhaak RG. Inferring tumour purity and stromal and im- mune cell admixture from expression data. Nat Commun 2013; 4: 2612.

[13] Shen W, Song Z, Zhong X, Huang M, Shen D, Gao P, Qian X, Wang M, He X, Wang T, Li S and Song X. Sangerbox: a comprehensive, interac- tion-friendly clinical bioinformatics analysis platform. iMeta 2022; 1: e36.

[14] Kanehisa M, Furumichi M, Sato Y, Ishiguro- Watanabe M and Tanabe M. KEGG: integrating viruses and cellular organisms. Nucleic Acids Res 2021; 49: D545-D551.

[15] Kanehisa M. Toward understanding the origin and evolution of cellular organisms. Protein Sci 2019; 28: 1947-1951.

[16] Yu G, Wang LG, Han Y and He QY. clusterProfil- er: an R package for comparing biological themes among gene clusters. OMICS 2012; 16: 284-287.

[17] Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez JC and Muller M. pROC: an open- source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics 2011; 12: 77.

[18] Olawaiye AB, Baker TP, Washington MK and Mutch DG. The new (Version 9) American Joint Committee on Cancer tumor, node, metastasis staging for cervical cancer. CA Cancer J Clin 2021; 71: 287-298.

[19] Li Q, Liu Z, Ma L, Yin W and Zhang K. CACYBP knockdown inhibits progression of prostate cancer via p53. J Cancer Res Clin Oncol 2023; 149: 5761-5772.

[20] Zheng H and Chen C. Downregulation of CacyBP by CRISPR/dCas9-KRAB prevents bladder cancer progression. Front Mol Biosci 2021; 8: 692941.

[21] Chen Y, Zhang K, Wang X, Li Q, Wu Q and Ning X. Cell cycle-dependent translocation and reg- ulatory mechanism of CacyBP/SIP in gastric cancer cells. Anticancer Drugs 2018; 29: 19- 28.

[22] Zhang P, Pei S, Gong Z, Feng Y, Zhang X, Yang F and Wang W. By integrating single-cell RNA- seq and bulk RNA-seq in sphingolipid metabo- lism, CACYBP was identified as a potential therapeutic target in lung adenocarcinoma. Front Immunol 2023; 14: 1115272.

[23] Zhao M, Zhang RZ, Qi DW, Chen HY and Zhang GC. CacyBP/SIP promotes tumor progression by regulating apoptosis and arresting the cell cycle in osteosarcoma. Exp Ther Med 2020; 20: 1397-1404.

[24] Yan S, Li A and Liu Y. CacyBP/SIP inhibits the migration and invasion behaviors of glioblas- toma cells through activating Siah1 mediated ubiquitination and degradation of cytoplasmic p27. Cell Biol Int 2018; 42: 216-226.

[25] Chen X, Mo P, Li X, Zheng P, Zhao L, Xue Z, Ren G, Han G, Wang X and Fan D. CacyBP/SIP pro- tein promotes proliferation and G1/S transi- tion of human pancreatic cancer cells. Mol Carcinog 2011; 50: 804-810.

[26] Fu C, Wan Y, Shi H, Gong Y, Wu Q, Yao Y, Niu M, Li Z and Xu K. Expression and regulation of CacyBP/SIP in chronic lymphocytic leukemia cell balances of cell proliferation with apopto- sis. J Cancer Res Clin Oncol 2016; 142: 741- 748.

[27] Smereczanska M, Domian N, Mlynarczyk G and Kasacka I. The effect of CacyBP/SIP on the phosphorylation of ERK1/2 and p38 ki- nases in clear cell renal cell carcinoma. Int J Mol Sci 2023; 24: 10362.

CACYBP in multiple cancers

[28] Nie F, Yu XL, Wang XG, Tang YF, Wang LL and Ma L. Down-regulation of CacyBP is associated with poor prognosis and the effects on COX-2 expression in breast cancer. Int J Oncol 2010; 37: 1261-1269.

[29] Wang N, Wang Y, Wang H, Luo N, Yang W and Zhao Z. Knockout of calcyclin binding protein impedes the growth of breast cancer cells by regulating cell apoptosis and beta-Catenin sig- naling. DNA Cell Biol 2021; 40: 1317-1324.

[30] Wang N, Ma Q, Wang Y, Ma G and Zhai H. CacyBP/SIP expression is involved in the clini- cal progression of breast cancer. World J Surg 2010; 34: 2545-2552.

[31] Jardim DL, Goodman A, de Melo Gagliato D and Kurzrock R. The challenges of tumor muta- tional burden as an immunotherapy biomark- er. Cancer Cell 2021; 39: 154-173.

[32] Schumacher TN, Scheper W and Kvistborg P. Cancer neoantigens. Annu Rev Immunol 2019; 37: 173-200.

CACYBP in multiple cancers

Supplementary Material 1. Basic clinical information for internal samples

Patient ID	Gender	Age (year)	Sample type
1	Male	59	LUAD & adjacent non-cancerous tissues
2	Male	67	LUAD & adjacent non-cancerous tissues
3	Female	59	LUAD & adjacent non-cancerous tissues

CACYBP in multiple cancers

ACC

BLCA

BRCA

12.5-

CACYBP expression

10.0

7.5

5.0

Stage | Stage II Stage III Stage IV AJCC_stage (n = 75 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 405 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 1067 )

HNSCC

KICH

KIRC

10.0

CACYBP expression

ns **

CACYBP expression

7.5

5.0

2.5

Stage | Stage II Stage III Stage IV AJCC_stage (n = 443 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 66 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 527 )

LUSC

MESO

PAAD

12.5-

CACYBP expression

10.0

7.5

5.0

Stage | Stage II Stage III Stage IV AJCC_stage (n = 494 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 87)

2.5

Stage | Stage II Stage III Stage IV AJCC_stage (n = 175)

TGCT

THCA

UVM

10-

CACYBP expression

.. .

Stage I

Stage II

AJCC_stage (n = 79 )

Stage III

Stage | Stage II Stage III Stage IV AJCC_stage (n = 502 )

Stage II

Stage III AJCC_stage (n = 78 )

Stage IV

CHOL

COAD

ESCA

Stage | Stage II Stage III Stage IV AJCC_stage (n = 36 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 276 )

LIHC

LUAD

12-

Stage | Stage II Stage III Stage IV AJCC_stage (n = 258 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 345 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 505 )

READ

SKCM

Stage | Stage II Stage III Stage IV AJCC_stage (n = 82 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 97)

Supplementary Material 2. The correlations between CACYBP expression and AJCC stages found in the cancers. All p-values were based on the Wilcoxon rank-sum test.

STAD

11-

CACYBP expression

Stage | Stage II Stage III Stage IV AJCC_stage (n = 389 )

Stage | Stage II Stage III Stage IV AJCC_stage (n = 158 )

KIRP

CACYBP expression

CACYBP in multiple cancers

ACC

BLCA

BRCA

CESC

CHOL

COAD

CACYBP expression

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

Age_in_year (n = 77 )

Age_in_year (n = 407 )

Age_in_year (n = 1090 )

Age_in_year (n = 304 )

Age_in_year (n = 36 )

Age_in_year (n = 286 )

DLBC

ESCA

GBM

HNSCC

KICH

KIRC

CACYBP expression

··

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

Age_in_year (n = 47 )

Age_in_year (n = 181 )

Age_in_year (n = 152 )

Age_in_year (n = 517 )

Age_in_year (n = 66 )

Age_in_year (n = 530 )

KIRP

LAML

LGG

LIHC

LUAD

LUSC

CACYBP expression

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

Age_in_year (n = 285 )

Age_in_year (n = 173 )

Age_in_year (n = 508 )

Age_in_year (n = 368 )

Age_in_year (n = 494 )

Age_in_year (n = 489 )

MESO

PAAD

PCPG

PRAD

READ

CACYBP expression

7.5

CACYBP expression

5.0

2.5

0.0

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

Age_in_year (n = 87 )

Age_in_year (n = 419 )

Age_in_year (n = 178 )

Age_in_year (n = 177 )

Age_in_year (n = 495 )

Age_in_year (n = 91 )

SARC

SKCM

STAD

TGCT

THCA

THYM

CACYBP expression

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

<65

>=65

Age_in_year (n = 258 )

Age_in_year (n = 102 )

Age_in_year (n = 409 )

Age_in_year (n = 132 )

Age_in_year (n = 504 )

Age_in_year (n = 118 )

UCEC

UCS

UVM

CACYBP expression

Supplementary Material 3. The correlations between CA- CYBP expression and ages found in the cancers. All p-val- ues were based on the Wilcoxon rank-sum test.

<65

>=65

<65

>=65

<65

>=65

Age_in_year (n = 177 )

Age_in_year (n = 57 )

Age_in_year (n = 79 )

CACYBP in multiple cancers

ACC

BLCA

BRCA

CHOL

COAD

DLBC

CACYBP expression

Female

Male

Gender (n = 77 )

Female

Gender (n = 407 )

Male

Female

Gender (n = 1091 )

Male

Female

Gender (n = 36 )

Male

Female

Gender (n = 286 )

Male

Female

Gender (n = 47 )

Male

ESCA

GBM

HNSCC

KICH

KIRC

KIRP

CACYBP expression

Female

Male

Female

Gender (n = 152 )

Male

Female

Gender (n = 518 )

Male

Female

Male

Female

Male

Female

Gender (n = 288 )

Male

Gender (n = 181 )

Gender (n = 66 )

Gender (n = 530 )

LAML

LGG

LIHC

LUAD

LUSC

MESO

CACYBP expression

Female

Male

Female

Male

Female

Gender (n = 369 )

Male

Female

Gender (n = 513 )

Male

Gender (n = 173 )

Gender (n = 508 )

Female Gender (n = 498 )

Male

Female

Gender (n = 87 )

Male

PAAD

PCPG

READ

SARC

SKCM

STAD

CACYBP expression

Female

Male

Female

Male

Gender (n = 177 )

Female

Male

Female

Male

Female

Male

Gender (n = 178 )

Female

Male

Gender (n = 91 )

Gender (n = 258 )

Gender (n = 102 )

Gender (n = 414 )

THCA

THYM

UVM

CACYBP expression

Supplementary Material 4. The correlations between CACYBP expres- sion and genders found in the cancers. All p-values were based on the Wilcoxon rank-sum test.

Female

Male

Female

Male

Female

Male

Gender (n = 504 )

Gender (n = 119 )

Gender (n = 79 )

Quartz 4

Explorer

38322570

Original Article Pan-analysis reveals CACYBP to be a novel prognostic and predictive marker for multiple cancers

Introduction

CACYBP in multiple cancers

Materials and methods

Signaling pathways that may be affected by CACYBP

Collection of internal samples and use of Western blotting

Statistical analysis

Results

CACYBP in multiple cancers

CACYBP in multiple cancers

CACYBP in multiple cancers

CACYBP in multiple cancers

CACYBP in multiple cancers

CACYBP in multiple cancers

Discussion

CACYBP in multiple cancers

CACYBP in multiple cancers

Acknowledgements

Disclosure of conflict of interest

References

CACYBP in multiple cancers

CACYBP in multiple cancers

CACYBP in multiple cancers

Supplementary Material 1. Basic clinical information for internal samples

CACYBP in multiple cancers

CACYBP in multiple cancers

Graph View

Table of Contents

Backlinks