Skip Ribbon Commands
Skip to main content


In urothelial cancer, we study both upper and lower urinary tract disease. Upper tract urothelial cancer (UTUC) is known to have the highest incidence in Taiwan. By collaborating with the Chang Gung Memorial Hospital, we have identified the aristolochic acid (AA) carcinogen signature in a subset of UTUC, suggesting that consumption of this herbal carcinogen may contribute to tumorigenesis. These AA-related tumours harbour an incredible amount of mutations, specifically in the form of A-T or T-A transversion. Among the most frequently mutated genes is lysine (K)-specific demethylase 6A (KDM6A, also known as UTX), which is also the second-most frequently mutated gene (second to p53) in lower tract urothelial cancer (also known as bladder cancer). We therefore undertook studies on KDM6A mutations with the goal of targeting them therapeutically. We demonstrated that these mutations are indeed driver mutations but difficult to target directly because they are inactivating tumour suppressor mutations. However, by targeting the oncogenic counterpart of KDM6A, enhancer of zeste homolog 2 (EZH2) using its inhibitor, we can inhibit KDM6A-null urothelial cancer cell lines effectively (Figure 1, Ler et al., Sci Transl Med, 2017). Our goal is to explore the mechanistic roles of chromatin enzyme mutations in both urothelial and renal cell carcinoma (see below) and identify ways to therapeutically target them.

Figure 1: Pathogenic KDM6A-null urothelial bladder carcinoma can be targeted using EZH2 inhibitors.

Figure 2: At NCCS, our prostate cancer research is conducted along the three themes.

For renal cell carcinoma, we have explored the epigenomic landscape (e.g., histone modifications, promoters, and enhancers) associated with distinct mutations found in different histological subtypes. For example, Von Hippel-Lindau (VHL) mutations are almost exclusively found in clear cell renal cell carcinoma (ccRCC), whereas cMET activating mutations appear to be specifically involved in type 1 papillary RCC and mutations in fumarate hydratase and downstream genes related to the Keap1-NRF2 pathway in type 2 papillary RCC. Recently, we reported the first study of super-enhancers related to VHL mutations in ccRCC, and, from this data, we further identified a novel ccRCC-specific master regulator, zinc-finger gene 395 (ZNF395). Subsequent studies confirmed that inhibition of ZNF395 can effectively suppress ccRCC growth in both in vitro and in vivo models. Besides histological subtype-specific genes, members of chromatin modifiers or remodellers are also frequently mutated in these tumours, making epigenomic studies of these tumours even more timely and appropriate. For example, polybromo 1 (PBRM1), an accessory subunit of the switch/sucrose nonfermenting (SWI/ SNF) chromatin remodelling complexes, is the second-most commonly mutated gene (after VHL) in ccRCC followed by SETD2 (a histone H3 lysine 36 methyltransferase) and BAP1 (which encodes the BRCA1 associated protein-1). It is likely that these mutations may impact gene regulation, and consequently contribute to cancer phenotypes. The technologies employed for these studies include CHIP sequencing, RNA sequencing, methylation profiling, as well as a series of cellular and molecular assays.

Prostate cancer has emerged as the third most common cancer among men in Singapore. At the NCCS, our research is focused along three themes: 1) To characterise the germline and somatic genomic landscape of Singaporean men with prostate cancer; 2) To investigate the molecular drivers of therapeutic resistance in high-risk and advanced prostate cancers; and 3) To conduct novel treatment-intensification trials with a new generation of anti-androgens. For theme (1), we are interested to ascertain whether germline genetics for prostate cancer susceptibility differ among the different racial demographics in Singapore (Chinese vs Malays vs Indians), and also between Asians and Caucasians. To this end, we have a prospective cohort of >1,000 men on follow-up, for which we are running whole-exome sequencing and single nucleotide polymorphism genotyping; paired clinical data is available for all patients. This research is performed in collaboration with the Ontario Institute of Cancer Research (OICR), Toronto. We are also a member of the PRACTICAL ( consortium, which seeks to investigate the risk genes for prostate cancer in 200,000 men globally. Additionally, we are investigating the prevalence of specific germline mutations in DNA repair genes and HSD3B1, which may have a clinical relevance
in predicting therapeutic sensitivity to chemotherapeutic agents and hormonal therapy (HT), respectively. Separately, on the somatic genomic landscape, we have collaborated with OICR to publish one of the largest comprehensive molecular datasets on 500 localised prostate tumours.

Similar to the findings by The Cancer Genome Atlas (TCGA), we, again, observed that prostate cancer is a C-class tumour but, more importantly, found that epigenomic dysregulation of the transcription elongation regulator 1-like gene (TCERG1L) carries a greater impact on prognosis than recurrent copy number aberrations, like c-MYC and TP53 (Fraser et al., Nature, 2017). We further performed subclonal reconstruction in these tumours, and found that polyclonal tumours have a higher risk of relapse post-treatment than monoclonal tumours (Espiritu et al., Cell, 2018). Finally, we have made an interesting discovery of recurrent mitochondrial mutations that appear to be linked to recurrent nuclear mutations in prostate cancer, and these compound mutational events may affect prognosis in patients (Hopkins et al., Nat Comm, 2017).

For theme (2), we adopted the same next-generation sequencing approaches to determine which mutational events are enriched in recurrent and high-risk prostate cancers that are resistant to radiotherapy and surgery, using a matched case control design. We observed that clonal selection of resistant tumour clones predominates in prostate cancer at the time of recurrence (Chua et al., Int J Radiat Oncol Biol Phys 2017). To perform functional characterisation, we have established radioresistant prostate cancer cell lines (RR-22RV1 and RRDU145). Next, we are investigating the differential molecular genetics in high Gleason’s grade (GS9-10) prostate cancers between tumours that recur within 18 months post-therapy and those that are controlled with single modality surgery or a combination of hormonal treatment and radiotherapy. This will allow us to hopefully identify the molecular drivers associated with highly aggressive prostate cancers. Lastly, we are conducting a prospective study to validate the role of circulating androgen-receptor splice variant 7 (ARV7) in predicting resistance to second-generation anti-androgen therapy of enzalutamide and abiraterone in Asian men with metastatic castrate-resistant prostate cancer; the NCCS is the lead institution in this international, multicentre study, which involves five other institutions from Asia.

To complement these molecular genetics studies, we are currently conducting a novel Phase II clinical trial to investigate the role of neoadjuvant apalutamide administered prior to radical prostatectomy in unfavourable intermediate and highrisk prostate cancers. Biopsies were obtained pre- and postapalutamide treatment, which will allow for the study of the tumour response to profound castration and provide insight into possible mechanisms driving early adaptation to a castrate environment in resistant clones. These discoveries could lead to the potential identification of novel therapeutic pathways. Separately, we have commenced an oligometastatic programme, whereby men with hormone-sensitive, treatment-naïve or recurrent oligometastatic prostate cancer are managed by local ablative therapies, and prospectively surveyed for response to treatment and long-term biochemical control. 68Ga-PSMA-PET imaging is incorporated into the clinical workflow to detect occult metastases, and to assist with radiotherapy planning for more precise tumour targeting. We hope that incorporating high-quality imaging with aggressive local therapies could potentially help us define a subgroup of men with metastatic disease, who may be “cured” with such intensive therapies.


​Dr. Melvin CHUA​Dr. Ravindran KANESVARAN ​Dr. Winnie LAM​Dr. Lui Shiong LEE
​Prof Puay Hoon TAN​Prof Bin Tean TEH