Article Text

IGFBP-5 overexpression as a poor prognostic factor in patients with urothelial carcinomas of upper urinary tracts and urinary bladder
  1. Peir-In Liang1,2,3,
  2. Yu-Hui Wang4,
  3. Ting-Feng Wu5,
  4. Wen-Ren Wu6,
  5. Alex C Liao7,
  6. Kun-Hung Shen7,
  7. Chung-Hsi Hsing8,9,
  8. Yow-Ling Shiue6,
  9. Hsuan-Ying Huang10,
  10. Han-Ping Hsu11,
  11. Li-Tzon Chen12,13,14,15,16,
  12. Ching-Yih Lin17,18,
  13. Chein Tai5,
  14. Jui-Yu Wu9,
  15. Chien-Feng Li2,3,5,19
  1. 1Department of Pathology, Chi-Mei Foundation Medical Center, Liouying, Tainan, Taiwan
  2. 2Departments of Pathology, Chi-Mei Foundation Medical Center, Tainan, Taiwan
  3. 3Department of Pathology, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan
  4. 4Institute of Biosignal Transduction, National Cheng Kung University, Tainan, Taiwan
  5. 5Department of Biotechnology, Southern Taiwan University of Science and Technology, Tainan, Taiwan
  6. 6Institute of Biomedical Science, National Sun Yat-Sen University, Kaohsiung, Taiwan
  7. 7Departments of Urology, Chi-Mei Foundation Medical Center, Tainan, Taiwan
  8. 8Departments of Anesthesiology, Chi-Mei Foundation Medical Center, Tainan, Taiwan
  9. 9College of Medicine, Taipei Medical University, Taipei, Taiwan
  10. 10Departments of Pathology, Kaohsiung Chang Gung Memorial Hospital and Chang Gung University College of Medicine, Kaohsiung, Taiwan
  11. 11College of Medicine, China Medical University, Taichung, Taiwan
  12. 12National Institute of Cancer Research, National Health Research Institutes, Tainan, Taiwan
  13. 13Departments of Molecular Medicine, National Cheng Kung University, Tainan, Taiwan
  14. 14Department of Internal Medicine, National Cheng Kung University Hospital, Tainan, Taiwan
  15. 15Department of Internal Medicine, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan
  16. 16Cancer center, Kaohsiung Medical University Hospital, Kaohsiung Medical University, Kaohsiung, Taiwan
  17. 17Division of Gastroenterology and Hepatology, Department of Internal Medicine, Chi-Mei Foundation Medical Center, Tainan, Taiwan
  18. 18Department of Leisure, Recreation, and Tourism Management, Southern Taiwan University of Science and Technology, Tainan, Taiwan
  19. 19Graduate Institute of Medicine, College of Medicine, Kaohsiung Medical University, Kaohsiung, Taiwan
  1. Correspondence to Dr Chien-Feng Li, Department of Pathology, Chi-Mei Medical Center, No. 901, Chunghwa Road, Yongkang Dist, Tainan City 710, Taiwan; angelo.p{at}


Background Urothelial carcinoma (UC) is prevalent worldwide. Dysregulation of cell growth is a critical event of tumorigenesis and has not been assessed systemically in UC. We thus assessed the published transcriptome of urinary bladder urothelial carcinoma (UBUC) and identified insulin-like growth factor-binding protein-5 (IGFBP-5) as the most significantly upregulated gene associated with the regulation of cell growth. Moreover, validated by using public domain data set, IGFBP-5 expression also significantly predicted worse outcome. IGFBP-5 is one of the binding proteins that regulate insulin-like growth factors (IGFs) and its significance has not been comprehensively evaluated in UCs.

Methods Using immunohistochemistry, we evaluated the IGFBP-5 expression status and its associations with clinicopathological features and survival in 340 cases of upper urinary tract urothelial carcinoma (UTUC) and 295 cases of UBUC. Western blot analysis was used to evaluate IGFBP-5 protein expression in human urothelial cell (HUC) lines.

Results IGFBP-5 overexpression was significantly associated with advanced pT stage (p<0.001), high histological grade (UTUC, p<0.001; UBUC, p=0.035), lymph node metastasis (UTUC, p=0.006; UBUC, p=0.004), vascular invasion (UTUC, p<0.001; UBUC, p=0.003), perineural invasion (UTUC, p=0.034; UBUC, p=0.021) and frequent mitosis (UTUC, p<0.001; UBUC, p=0.023). IGFBP-5 overexpression also independently predicted poor disease-specific survival and metastasis-free survival in both groups of patients. Western blot analysis showed IGFBP-5 protein as overexpressed in human urothelial cancer cell lines and not in normal urothelial cancer cells.

Conclusions IGFBP-5 plays an important role in tumour progression in UC. Its overexpression is associated with advanced tumour stage and conferred poorer clinical outcome.

View Full Text

Statistics from


Urothelial carcinoma (UC) is the most common tumour type of urinary tract. It can arise from the upper urinary tract (UT; renal pelvis and ureter) and urinary bladder (UB).1 The aetiology of all UCs regardless of the location share common carcinogens, but certain factors show a stronger impact on the urinary tract urothelial carcinoma (UTUC), such as UC developed in Balkans endemic nephropathy, in Chinese herb nephropathy and in those associated with phenacetin abuse.2 Although previous study shows that promoter hypermethylation is more common in UTUC than urinary bladder urothelial carcinoma (UBUC), the gene expression profile of UCs from both locations are  very similar.3 ,4 Besides, the disease behaviour of UTUC and UBUC is identical when stage and grade are considered.2 This may indicate that tumorigenesis of UC of the entire urinary tract shares a common molecular pathway.

Several evidences have demonstrated that the carcinogenesis of human cancers is a multiple step process, which eventually transforms  a normal cell into a malignant neoplasm.5 Of these, self-sufficiency in growth signals, that is, disruption of cell growth regulation is the first step towards cell transformation. However, the genes associated with dysregulation of cell growth have not been systemically evaluated in UC. By performing data mining on the documented transcript expression profiles in Gene Expression Omnibus (GEO, National Center Biotechnology Information (NCBI), Bethesda, Maryland, USA) with a special focus on regulation of cell growth, we identified that the transcription of the gene, insulin-like growth factor-binding pattern 5 (IGFBP5), was most significantly upregulated in the late-staged and metastatic human UC, suggesting its role in tumorigenesis and progression.

IGFBP-5 is a carrier protein that may prolong half-life and the circulation turnover of insulin-like growth factors (IGFs), which play a role in cellular growth, differentiation and apoptosis.6 IGFBP-5 expression was altered in various cancers, such as breast cancer, neuroblastoma, osteosarcoma, lung cancer, colon cancer, etc, and demonstrated a cell-type-dependent and tissue-type-dependent prognostic impact.7 However few, if any, reports have evaluated IGFBP-5 expression in UC. Therefore, we aimed to comprehensively analyse IGFBP-5 expression and its associations with clinicopathological factors and survivals in our well-characterised cohort of UC.

Materials and methods

Data mining on the GEO to identify overexpressed transcripts in UCs

Data mining on GEO (National Center Biotechnology Information) was performed. One data set, GSE31684 ( analysis of radical cystectomy specimens from 93 UBUC using Affymetrix U133 Plus 2.0 Array, was identified. Statistical software Nexus Expression 3 (BioDiscovery, EI Segundo, California, USA) was used to analyse all probe sets without preselection or filtering. Supervised comparative analysis was performed to examine the statistical significance of differentially expressed transcripts on the basis of primary tumour (pT) status  and the development of metastatic event. Transcriptomes of high-staged (pT2–pT4) UCs with metastases developed and low-staged (pTa–pT1) UCs devoid of metastasis were used to perform functional profiles focusing on those related to regulation of cell growth (GO:0001558). Further survival analysis was performed in all cases by dichotomise cases into high-expression and low-expression clusters to computerise the prognostic impact of selected gene.

Patients and tumour specimens

This study was approved by the institutional review board (IRB971006). For immunohistochemical study and survival analysis, we retrieved 635 cases diagnosed as UC, not otherwise specified (NOS) from the archives of Chi-Mei Medical Center between 1996 and 2004. Of these cases, 340 tumours originated from the UT and 295 arose from the UB. Other histological variants were excluded. The criteria for the clinicopathological evaluation were essentially identical to our previous works.8 Detailed information is provided in the online supplementary data.

Immunohistochemical staining and scoring of IGFBP-5

Tissue sections underwent standard procedure for deparaffinisation, rehydration and antigen retrieval. Subsequently, the sections incubated with a primary antibody targeting IGFBP-5 (1 : 100; Abcam, Cambridge, Massachusetts, USA) for an hour. Scoring of IGFBP-5 immunoreactivity was evaluated based on the combination of the percentage and intensity of positively stained tumoural cytoplasm to generate H-score, which was calculated using the following equation: H-score=ΣPi(i+1), where i is the intensity of stained tumour cells (0–4+), and Pi is the percentage of stained tumour cells for each intensity varying from 0% to 100%.9 ,10 Detailed information is provided in the online supplementary data.

Cell lines and culture conditions

Five UBUC cell lines, RT4, J82, TCCSUP, BFTC-905 and TSGH8301, and one UTUC cell line, RT9, were grown based on the suggested media and conditions. A non-tumorigenic urothelial cell, human urothelial cell (HUC), was used as control. Detailed information is provided in the online supplementary data.

Western blot assays

Equal amounts of total cell protein (25 μg) extracted were separated on 4%–12% gradient sodium dodecyl sulphate polyacrylamide gel, transferred to polyvinylidene difluoride membranes and then probed with antibodies against IGFBP-5 (1 : 1000; Abcam). Detailed information is provided in the online supplementary data.

Statistical analysis

Statistics were performed using SPSS V.14.0 software (SPSS Inc., Chicago, Illinois, USA). The median H-score of IGFBP-5 was used as the cut-off to dichotomise the study cohort, separating cases into high expression and low expression. The χ2 test was used to compare the IGFBP-5 expression status and various clinicopathological parameters. The end points analysed were disease-specific survival (DSS) and metastasis-free survival (MeFS), calculated from the starting date of resection of tumor to the date of event developed. Patients lost to follow-up were censored on the latest follow-up date. Univariate survival analyses were performed using Kaplan–Meier plots and compared by the log-rank test. Those parameters with univariate p<0.1 were enrolled into multivariate test that was performed using Cox proportional hazards regression. For all analyses, two-sided tests of significance were used with p<0.05 considered as significant.


IGFBP5 identified as a significant differentially upregulated transcript implicating dysregulation of the cell growth in UBUC

From the transcriptomic profile of 33 high-staged and metastatic cases and 16 low-staged and non-metastatic ones of UBUC, 26 probes covering 21 transcripts associated with regulation of cell growth (GO:0001558) were found (figure 1). The log2 ratios of one transcript met the selection criteria of >1.5 increase (p<0.0001), that is, IGFBP5, with log2 ratio of 1.7363 to 3.3334-fold upregulation (table 1). Of note, the expression level of IGFBP5 transcript significantly predicted DSS (figure 2, p=0.0120). Detailed information is provided in the online supplementary data.

Figure 1

The transcriptome of urinary bladder urothelial carcinomas reconstructed from GSE31684 with focusing on those involving regulation of cell growth (GO: 0001558). The expression levels of 26 probes targeting 21 genes were identified to be significantly different (p<0.01, Log2 ratio >+/ 0.1) by comparing 33 high-staged (pT2-T4) and metastasising tumour and 16 low-staged (pT0-T1) and non-metastasising lesions.

Figure 2

By subdividing 93 cases from GSE31684 into IGFBP5 high-expression (n=53) and low-expression (n=40) clusters, its expression level significantly predicts disease-specific survival (p=0.0120).

Table 1

Summary of differentially expressed genes associated with regulation of cell growth in the transcriptome of urothelial carcinoma of urinary bladder

Clinicopathological findings of the UTUC and UBUC

The clinicopathological findings of the UTUC and UBUC cases are listed in table 2. Detailed information is provided in the online supplementary data.

Table 2

Correlations between IGFBP-5 expression and other important clinicopathological parameters in urothelial carcinomas

Correlations of immunoreactivity of IGFBP-5 with parameters in the UTUC and UBUC

IGFBP-5 shows variable cytoplasmic expression in the UC of both locations with H-scores in the range of 155 to 485 (median 340) for UTUC and 150 to 490 (median 345) for UBUC. After dichotomising the tumour into low and high IGFBP-5 expression (H-score less than median and no less than median, respectively), as demonstrated in table 2, increased IGFBP-5 expression showed significant association with increment of pT status (figure 3), lymph node metastasis, a high histological grade, vascular invasion, perineural invasion and frequent mitosis in UC for either location. Detailed information is provided in the online supplementary data.

Figure 3

Representative H&E sections of low-stage urothelial carcinoma (UC) (A) demonstrated low cytoplasmic IGFBP-5 expression (C), whereas the high-stage, infiltrating UC (B) shows bright IGFBP-5 expression (D).

Survival analysis of UTUC

The univariate and multivariate analyses of the associations of the clinical outcomes with various clinicopathological in UTUC and UBUC cases are demonstrated in tables 3 and 4, respectively. Detailed information is provided in the online supplementary data.

Table 3

Univariate log-rank and multivariate analyses for disease-specific and metastasis-free survivals in upper urinary tract urothelial carcinoma

Table 4

Univariate log-rank and multivariate analyses for disease-specific and metastasis-free survivals in urinary bladder urothelial carcinoma

Prognostic impact of IGFBP-5 expression in UCs

Univariate and multivariate analyses (tables 3 and 4 and figure 4) showed that high IGFBP-5 cytoplasmic expression was significantly associated with worse DSS and MeFS in patients with UTUC or UBUC. Detailed information is provided in the online supplementary data.

Figure 4

Kaplan-Meier plots show that IGFBP-5 overexpression conferred significant prognostic impacts in disease-specific survival and metastasis-free survival of patients with upper urinary tract urothelial carcinoma (A and B, respectively) and urinary bladder urothelial carcinoma (C and D, respectively).

IGFBP-5 is more abundant in urothelial tumour cells than primary urothelial cells

According to the western blot analysis, IGFBP-5 protein is detected in urothelial cancer cells including RT4, TSGH8301, J82, RT9, BFTC905 and TCCSUP. However, its expression level in HUC is negligible. This result suggested that IGFBP-5 might participate in urothelial oncogenesis (figure 5).

Figure 5

Western blotting was used to assess IGFBP-5 protein expressing in human urothelial cancer cell lines. IGFBP-5 expression was abundant in cancer cells (RT4, TSGH8301, J82, RT9, BFTC905 and TCCSUP), but was barely detected in normal human urothelial cells.


Carcinogenesis is a complex process. It takes multiple steps to transform normal cells, via various precancerous states, into invasive tumours. Disruption of the regulation of normal cell growth is the fundamental concept for malignant tumour growth.5 Data mining of the GEO data set GSE31684 identified that the alteration of IGFBP5 transcript has the strongest positive correlation with advance disease condition. Using immunohistochemistry study, we have demonstrated that the overexpression of IGFBP-5 strongly correlates with advance disease in UTUC and UBUC, including high tumour stage, high tumour grade, frequent mitosis, presence of vascular and perineural invasion and nodal metastasis. Besides, IGFBP-5 was found to independently correlate with poor outcome, DSS and MeFS, in UTUC and UBUC. These findings are comparable with those observed in malignant tumour of other organs, such as glioblastoma,11 ovarian serous carcinoma12 and breast cancer.13

The IGF axis is an important pathway for regulating cell proliferation, differentiation and apoptosis.14 ,15 Located in chromosome 2, IGFBP5 has a length of 33 kb and encoded for a 29-kDa soluble protein. It binds with IGFs with high affinity and modulates their biological functions. IGFBP-5 has both IGF-dependence and IGF-independence biological functions. The former is started as IGF­I and IGF­II binds to IGF I receptor (IGFIR), and initiates multiple cellular response to promote tissue growth, including induced proliferation, suppress apoptosis and promote differentiation.6 By controlling the binding affinity to IGFs, IGFBP-5 prolongs IGFs half-life and restricts their function, affecting the IGF signalling pathway.16 ,17 On the other hand, IGFBP-5 can enter the targeted cells via Impβ-mediated pathway or through diffusion, interact with intracellular proteins and induce acquired novel function such as anti-apoptosis and metastasis.7 However, the IGF-independent functions are more complex and less well characterised that are supposed to be associated with the activation of Akt and mitogen-activated protein kinase (MAPK) and their downstream signalling.18 Although the details of IGFBP-5-related signalling remains uninvestigated in UC, the importance of Akt and MAPK activation in urothelial carcinogenesis has been comprehensively analysed.19 ,20 Moreover, a recent study also suggests a link between IGF-related pathway to Akt and MAPK to promote cell motility, providing a potential mechanism of IGFBP-5 in the development of urothelial cancer.21

The expression status and prognostic significance of IGFBP-5 are controversial and are in a cell-type and tissue-type-specific manner. In breast cancer, experiments conducted on surgical specimen show that overexpression of IGFBP-5 is positively correlated with lymph node metastasis and poor prognosis.13 ,22–24 On the other hand, studies performed on in-vitro cell culture assay show that IGFBP-5 may induce apoptosis and inhibit cell migration.25 ,26 These contradicted observations can be partly explained by subcellular localisation of IGFBP-5, which will change the phenotypes of the cancer cells. Akkiprik et al27 found that wild-type IGFBP-5 would be localised in the nucleolus of MDA-MB-435 and inhibited cancer cell growth and motility. In contrast, mutant IGFBP-5 with deletion of 5 amino acids (214–218) in the C-domain is found in the cytoplasm and induces cell proliferation and cell motility in the cancer cells. Similar findings were also observed by Sureshbabu et al.28 Besides, poor survival was also associated with downregulation of IGFBP-5 in some tumours, including pleomorphic malignant fibrous histiocytoma sarcoma NOS and squamous cell carcinoma of head and neck origin.29 ,30 Similarly, its expression in various tumours may be upregulated (such as lung, thyroid and pancreas) or downregulated (such as cervix, colon and kidney).7

Recently, Holyoake et al31 evaluated IGFBP-5 mRNA expression in urine samples coupled with CDC2, MDK and HOXA13 to detect UC. Using quantitative real-time PCR, the author shows that the mRNA level of these makers detected in urine has higher sensitivity to detect transitional cell carcinoma, in contrast with urine cytology. OGX-225 (OncoGenex Technologies), a second-generation antisense oligonucleotide that targets IGFBP2 and IGFBP-5 mRNA, has been tested on breast cancer cell lines. The result shows that cancer cells transfected with OGX-225 will decrease cell viability and increase cell susceptibility to chemotherapy.32 Whether such promising results apply on IGFBP-5 and UC warrants further studies.  In conclusion, we have identified IGFBP5 transcript is significantly upregulated in adverse UCs. Moreover, its overexpression is an independent prognosticatory factor for UTUC and UBUC. Our findings suggest that IGFBP-5 is not only a diagnostic biomarker but also a prognostic biomarker in UC. Further evaluation to elucidate the biological significance of IGFBP-5 protein expression in UC is mandatory to explore the potential of IGFBP-5-targeted therapy in UC.

Take-home messages

  • The expression of IGFBP-5, a binding protein that regulates insulin-like growth factors, correlates with the clinicopathological features of upper urinary tract urothelial carcinoma and urinary bladder urothelial carcinoma. Its overexpression strongly correlates with several adverse prognostic factors.

  • IGFBP-5 cytoplasmic overexpression is predictive of poor disease-specific survival and metastatic-free survival in patients with urothelial carcinoma.

  • The overexpression of IGFBP-5 protein in upper urinary tract urothelial cancer cells and urinary bladder urothelial cancer cells suggest its role in urothelial carcinoma tumorigenesis.


The authors are grateful to Translational Research Laboratory of Human Cancers of Chi-Mei Medical Center for providing critical technical assistance.


View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:


  • J-YW and C-FL contributed equally as senior authors.

  • Contributors P-IL, Y-HW, T-FW, W-RW, ACL, K-HS, C-HH, Y-LS, H-YH, H-PH, L-TC, C-YL, CT, J-YW and C-FL participated in the conception and design, acquisition, analysis and interpretation of data. P-IL, J-YW and C-FL drafted the article and all authors revised it critically for important intellectual content. J-YW and C-FL contributed equally to this work.

  • Funding This work was supported by grants (NSC101-2632-B-218-001-MY3) from the National Science Council, Taiwan, (DOH101-TD-C-111-004) Department of Health, Taiwan and (101CM-TMU-01) Chi Mei Medical Center, Tainan, Taiwan.

  • Competing interests None.

  • Ethics approval Institutional Review Board of Chi-Mei Foundation Medical Center.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement The original immunostaining and statistical data are available from the corresponding author.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.