Search

The purpose of this study was to compare bariatric surgery versus non-surgical treatment on blood pressure for patients with obesity. Nineteen RCTs (1353 total patients) were included. In the pooled analyses, bariatric surgery reduces more systolic blood pressure (WMD: − 3.937 mmHg, CI95%: − 6.000 to − 1.875, p 130/80 mmHg) and diabetes mellitus (HbA1C > 7.0%, FPG > 7.0 mmol/L), elder patients (> 45 years), non-severe obesity (BMI

Intrahepatic cholangiocarcinoma is a disease with grave prognosis due to limited therapeutic regimens. Programmed cell death 1 (PD-1)/programmed death-ligand 1 (PD-L1) inhibitor have shown dramatic clinical effectiveness in multiple solid tumors. Here, we report the case that a patient with metastasis intrahepatic cholangiocarcinoma, being failure of first-line chemotherapy, was enrolled into the Phase I study of a PD-1 inhibitor, sintilimab. The patient achieved complete remission after three cycles of treatment with mild adverse reaction. In addition, the tumor mutational burden and the microsatellite instability status were low in the present case. Hence, PD-1 inhibitor might be a promising therapeutic approach for patients with advanced cholangiocarcinoma.

Ravidasvir (RDV) is a novel oral hepatitis C virus NS5A inhibitor. This study aimed to evaluate the pharmacokinetics and safety of RDV and the drug-drug interactions between RDV and ritonavir-boosted danoprevir (DNVr) in healthy adults. In the 1st study, healthy volunteers were administered single oral doses of 100, 200, and 300 mg of RDV and 200 mg once daily for 7 days. The 2nd study was a randomized, double-blinded, and placebo-controlled sequential design (day 1 for 200 mg of RDV alone, day 7 for 100 mg/100 mg of DNVr, day 13 for 200 mg of RDV plus 100 mg/100 mg DNVr, followed by 200 mg of RDV once daily with 100 mg/100 mg of DNVr twice daily for 10 days). The results showed that RDV exposure increased in a dose-proportional manner following a single dose with no evidence of accumulation with multiple doses. Coadministration with DNVr (100 mg/100 mg, twice daily) resulted in a 2.92-fold and 1.99-fold increase in minimum plasma concentration at steady state (C(min,ss)) and area under the concentration-time curve at steady state (AUC(τ)) of RDV, respectively. With coadministration of RDV, maximum plasma concentration (C(max)) and area under the concentration-time curve from 0 to 12 h (AUC(0–12)) of DNV increased 1.71-fold and 2.33-fold, respectively. We did not observe any significant changes in ritonavir exposure. Both single and multiple doses of RDV with or without DNVr were well tolerated. The favorable pharmacokinetic and safety results support ravidasvir’s continued clinical development and treatment. (The studies described in this paper have been registered at ClinicalTrials.gov under identifiers NCT03430830 and NCT03288636.)

Immune checkpoint inhibitors (ICIs) have shown potential to improve the prognosis of patients with brain metastasis (BM) caused by advanced cancers. However, controversies still exist in regard to its survival benefits. In the present work, a time series-based meta-analysis based on the phase I/II/III trials and observational studies were performed to investigate the differences in mortality of ICI-treated BM patients. A number of public library databases, including MEDLINE, EMBASE, OVID, and COCHRANE, were systemically searched by March 2019. The quality of included studies was evaluated by the Newcastle-Ottawa Scale (NOS) scoring. Outcome measures here established were mortality and progression-free survival (PFS) at different follow-up endpoints. Survival rates and curve data were pooled for further analysis. To detect the data heterogeneity, subgroup analyses were conducted according to tumor and ICI types. Eighteen studies, 6 trials, and 12 controlled cohorts were assessed, involving a total of 1330 ICI-treated BM patients. The 6-month survival rate and PFS were 0.67 (95%CI: 0.59-0.74) and 0.36 (95%CI: 0.24-0.49), respectively. According to the tumor type (melanoma, NSCLC, and RCC), subgroup analyses indicated that melanoma presented the lowest survival rates among the three groups here selected. In regard to the type of ICIs, the anti-CTLA-4 combined with the anti-PD-1/PD-L1 showed the best survival outcome among these groups. The 12-month survival rate and PFS showed a consistent pattern of findings. In the long-term, the 24-month survival rate and PFS were 0.20 (95%CI: 0.12-0.31) and 0.18 (0.05-0.46) in BM patients. Hence, ICI therapy may be associated with an improved prognosis of BM patients. Nevertheless, current research presented a limited study design. Multicenter randomized trials may later assist in validating ICI-based therapies for a better outcome of BM patients.

Meihua Lin, Duo Lv, Yunliang Zheng, Minglan Wu, Chang Xu, Qiao Zhang, Lihua Wu Research Center of Clinical Pharmacy, State Key Laboratory for Diagnosis and Treatment of Infectious Disease, First Affiliated Hospital, Zhejiang University, Hangzhou, China Background: Cancer cells often have characteristic changes in metabolism. Besides Warburg effect, abnormal lipid metabolism is also considered as one of the most typical metabolic symbols of cancer. Thus, understanding the mechanisms of cell metabolic reprogramming may provide a potential avenue for cancer treatment. Materials and methods: In total, 41 pairs of matched samples of primary hepatocellular carcinoma (HCC) and adjacent non-cancerous liver tissues were collected. Afterward, we performed quantitative reverse transcriptase polymerase chain reaction to investigate carnitine palmitoyltransferase-2 (CPT2) expression and then systematically analyzed its relationship with clinicopathologic features. We further performed proliferation, colony formation, migration and invasion, drug resistance, and lipogenesis assays to determine the function of CPT2 in HCC. Results: In this study, we have identified CPT2 which is the rate-limiting enzyme of fatty acid oxidation, downregulated in HCC and was significantly associated with tumor histological differentiation and venous invasion. In vitro studies demonstrated that knockdown of CPT2 remarkably enhanced the tumorigenic activity and metastatic potential of hepatoma cells. In addition, CPT2 silencing induced chemoresistance to cisplatin. Mechanistically, low expression of CPT2 promoted cancer cell lipogenesis via upregulation of stearoyl-CoA desaturase-1, the key enzyme involved in the synthesis of monounsaturated fatty acids, at both mRNA and protein levels in hepatoma cell line. Conclusion: Altogether, our findings demonstrate that CPT2 has a critical role in HCC progression and chemoresistance and may potentially serve as a novel prognostic marker and therapeutic target for HCC treatment. Keywords: lipid metabolism, CPT2, HCC, SCD1, chemoresistance

Purpose: The aim of this study was to compare the pharmacokinetics and safety between two vinorelbine formulations [a new oil-in-water emulsion formulation (ANX) versus a previously marketed solution formulation (Navelbine)] in Chinese patients with advanced non-small cell lung cancer (NSCLC). Method: This was a single-center, randomized, open-label study. Eligible patients aged 18-70 years who had histologically or cytologically confirmed NSCLC were enrolled. In cycle 1, the patients alternatively received the two formulations (30 mg/m2, given as a 10-min infusion) with a 7-day interval. Samples for pharmacokinetic analysis were taken during cycle 1. For all subsequent 21-day cycles (maximum four cycles), ANX was administered on days 1 and day 8. Bioequivalence analysis was performed on Cmax, AUClast, and AUCinf. The safety profiles and anti-tumor effects were also determined. Results: From March 2013 to January 2015, 24 patients were enrolled and 20 were eligible for pharmacokinetic evaluation. The 20 subjects in the pharmacokinetic analysis set had a median age of 61 years (range, 37-70 years), and 15 patients were male (75%). Mean vinorelbine Cmax values for ANX and Navelbine were 1,317.40 and 1,446.30 ng/mL, respectively. Corresponding AUClast values were 797.08 and 924.26 ng·h/mL, respectively. AUCinf values were 830.14 and 957.16 ng·h/mL, respectively. Treatment ratios of the geometric means were 90.00% (90% CI, 83.22-99.07%) for Cmax, 86.92% (90% CI, 80.91-93.37%) for AUClast, and 87.44% (90% CI, 82.08-93.16%) for AUCinf. These results met the required 80-125% bioequivalence criteria. The most frequently reported adverse events after vinorelbine administration were neutropenia, leucopenia, neutropenic fever, and constipation. Conclusion: At therapeutic dosage levels, pharmacokinetic behavior and safety profiles were similar for both formulations. Chinese National Registry Code: ChiCTR-IPR-15005856.

TRAF-binding domain (Trabid), one of deubiquitination enzymes, was recently reported to activate Wnt/ β-catenin signaling pathway. However, the role of Trabid in tumors including hepatocellular carcinoma (HCC) and the underlying mechanisms controlling its activity remain poorly understood. Here, we report that Trabid is significantly downregulated in HCC tumor samples and cell lines compared with normal controls and that its expression level is negatively correlated with HCC pathological grading, recurrence, and metastasis. The reintroduction of Trabid expression in tumor cells significantly decreases HCC progression as well as pulmonary metastasis. The effect of Trabid on HCC development occurs at least partially through regulation of Twist1 activity. Mechanistically, Trabid forms a complex with Twist1 and specifically cleaves RNF8-induced K63-linked poly-ubiquitin chains from Twist1, which enhances the association of Twist1 with β-TrCP1 and allows for subsequent K48-linked ubiquitination of Twist1. Knockdown of Trabid increases K63-linked ubiquitination, but abrogates K48-linked ubiquitination and degradation of Twist1, thus enhancing HCC growth and metastasis. Interestingly, Twist1 negatively regulates the promoter activity of Trabid, indicating that a double-negative feedback loop exists. Our findings also identify an essential role for activation of Trabid by AKT-mediated phosphorylation at Ser78/Thr117 in negatively regulating Twist1 signaling, which further provides insights into the mechanisms by which Trabid regulates Twist1 ubiquitination. Our results reveal that Trabid is a previously unrecognized inhibitor of HCC progression and metastasis, which sheds light on new strategies for HCC treatment.

// Juan Liu 1,4 , Cen Zhang 1,4 , Meihua Lin 1 , Wei Zhu 1,2 , Yingjian Liang 1 , Xuehui Hong 1,2 ,Yuhan Zhao 1,2 , Ken H. Young 3 , Wenwei Hu 1,2 , Zhaohui Feng 1 1 Department of Radiation Oncology, 2 Department of Pediatrics, Rutgers Cancer Institute of New Jersey, Rutgers University, New Brunswick, NJ, USA 3 Department of Hematopathology, University of Texas MD Anderson Cancer Center, Houston , TX , USA 4 These two authors contributed equally to this work Correspondence: Wenwei Hu, email: // Zhaohui Feng, email: // Keywords :p53; GLS2; tumor suppression; PI3K/AKT; Hepatocellular carcinoma Received : February 15, 2014 Accepted : March 24, 2014 Published : March 26, 2014 Abstract The tumor suppressor p53 and its signaling pathway play a critical role in tumor prevention. As a direct p53 target gene, the role of glutaminase 2 (GLS2) in tumorigenesis is unclear. In this study, we found that GLS2 expression is significantly decreased in majority of human hepatocellular carcinoma (HCC). Restoration of GLS2 expression in HCC cells inhibits the anchorage-independent growth of cells and reduces the growth of HCC xenograft tumors. Interestingly, we found that GLS2 negatively regulates the PI3K/AKT signaling, which is frequently activated in HCC. Blocking the PI3K/AKT signaling in HCC cells largely abolishes the inhibitory effect of GLS2 on the anchorage-independent cell growth and xenograft tumor growth. The GLS2 promoter is hypermethylated in majority of HCC samples. CpG methylation of GLS2 promoter inhibits GLS2 transcription, whereas reducing the methylation of GLS2 promoter induces GLS2 expression. Taken together, our results demonstrate that GLS2 plays an important role in tumor suppression in HCC, and the negative regulation of PI3K/AKT signaling contributes greatly to this function of GLS2. Furthermore, hypermethylation of GLS2 promoter is an important mechanism contributing to the decreased GLS2 expression in HCC.

Many cancers apparently showing similar phenotypes are actually distinct at the molecular level, leading to very different responses to the same treatment. It has been recently demonstrated that pathway-based approaches are robust and reliable for genetic analysis of cancers. Nevertheless, it remains unclear whether such function-based approaches are useful in deciphering molecular heterogeneities in cancers. Therefore, we aimed to test this possibility in the present study. First, we used a NCI60 dataset to validate the ability of pathways to correctly partition samples. Next, we applied the proposed method to identify the hidden subtypes in diffuse large B-cell lymphoma (DLBCL). Finally, the clinical significance of the identified subtypes was verified using survival analysis. For the NCI60 dataset, we achieved highly accurate partitions that best fit the clinical cancer phenotypes. Subsequently, for a DLBCL dataset, we identified three hidden subtypes that showed very different 10-year overall survival rates (90%, 46% and 20%) and were highly significantly (P=0.008) correlated with the clinical survival rate. This study demonstrated that the pathway-based approach is promising for unveiling genetic heterogeneities in complex human diseases.

The present study aimed to investigate the therapeutic efficacy and clinical value of recombinant human adenovirus-p53 (rAd-p53) perfusion via the pancreatic artery for the treatment of mid-late stage pancreatic cancer. rAd-p53 (2×1012 virus particles) in 6 ml normal saline was pushed (intravenous bolus) into the gastroduodenal and superior pancreaticoduodenal arteries via interventional superselection, with the catheter retained for subsequent drug administration at a 3-day interval for 4 cycles. Tumor changes in all patients were observed to evaluate tumor response by computed tomography (CT) at 2, 8 and 16 weeks post-treatment. The following improvements were noted in the 23-patient cohort: A total of 73.9% (17/23) of patients demonstrated significant tumor shrinkage (>20%); the symptoms of abdominal and back pain were relieved in 15 patients; the survival time was >12 months in 1 patient and >6 months in 14 patients; the patient's general condition, including appetite, was improved in 13 patients; body weight was increased in 9 patients; jaundice was attenuated in 12 patients; and ascites subsided in 10 patients. However, the therapeutic outcome was poor in 2 patients whose tumors size did not show significant change after treatment as detected by CT. These 2 patients succumbed within 6 months. In conclusion, rAd-p53 perfusion via the pancreatic artery is a safe and minimally invasive option for the treatment of mid-late stage pancreatic cancer.

All-trans retinoic acid (ATRA) has been successfully used in differentiation therapy for acute promyelocytic leukemia (APL) in the clinic. ATRA-induced differentiation of leukemia cells is accompanied by a G0/G1 arrest, yet how ATRA couples cell cycle arrest to differentiation remains largely unknown. Here we observed that the ubiquitin-proteasome pathway (UPP) was activated upon ATRA treatment in the human acute myeloid leukemia cell lines, NB4 and HL-60, as represented by the accumulation of ubiquitinated proteins, the up-regulation of ubiquitin mRNA and increased 20S proteasome activity. Interestingly, we found that complete inhibition of proteasome activity suppressed ATRA-induced proliferation/differentiation (P/D) transition in both cell lines. Furthermore, we demonstrate that the exact protein contributing to this phenomenon is different in these two cell lines. Cyclin-dependent kinase 2 (CDK2) and Cyclin E were degraded by the UPP; they accumulated significantly after complete inhibition of the proteasome in ATRA-treated NB4 and HL-60 cells, respectively. These findings suggested that the UPP might be indispensable in the ATRA-induced G0/G1 arrest and differentiation of leukemia cells. The exact protein degraded by the UPP to promote the myeloid maturation program set in motion by the retinoid may be cell type dependent.

All-trans retinoic acid (ATRA) is nowadays considered to be the sole efficient agent for differentiation-based therapy in leukemia; however, the mechanisms of ATRA's biological effects remain largely unknown. Here we first reported that ATRA-induced myeloid leukemia differentiation was accompanied with the increased level of ubiquitin-protein conjugates and the upregulation of proteasome activity. To explore the functional role of the activated proteasome in retinoic acid (RA) signaling, the effects of proteasome inhibitors on RA-induced cell differentiation were determined. Our results demonstrated that inhibition of ATRA-elevated proteasome activity obviously promoted the myeloid maturation program triggered by ATRA, suggesting that the overactivated proteasome is not beneficial for ATRA's effects. Further studies demonstrated that the synergistic differentiating effects of ATRA and proteasome inhibitors might be associated with the protection of retinoic acid receptor alpha (RARα) from degradation by the ubiquitin-proteasome pathway (UPP). Moreover, the accumulated RARα was able to enhance the transcription of its target gene, which might also contribute to the enhanced differentiation of leukemia cells. Together, by linking the UPP to ATRA-dependent signaling, our data provide a novel insight into studying the mechanisms of ATRA-elicited cellular effects and imply the possibility of combination of ATRA and proteasome inhibitors in leukemia therapy.

All-trans-retinoic acid (ATRA) is a morphogenetic signalling molecule derived from vitamin A and is used clinically to target acute promyelocytic leukemia by inducing differentiation of immature blood cells. Retinoid signals are mediated by retinoic acid (RA) receptors (RARs) and retinoid X receptors (RXRs). Retinoic acid receptors consist of RARalpha, RARbeta and RARgamma isotypes. Among these components, RARalpha is preferentially bound to ATRA, which is used to treat acute T-lymphoblastic leukemia, yet the conditions and mechanisms remain unknown. In this study, we have demonstrated that, in human acute T-lymphoblastic leukemia Molt3 cells, inhibition of RA-induced proliferation results from massive cell death characterised by apoptosis. The effect of ATRA:RARalpha binding on apoptosis in Molt3 cells has been investigated. Consequently, it has been shown that, in RA-treated Molt3 cells, upregulation of p21 due to RA accompanies caspase 3/PARP activation which precedes the occurrence of apoptosis.

// Cen Zhang 1,* , Juan Liu 1,* , Rui Wu 1 , Yingjian Liang 1 , Meihua Lin 1 , Jia Liu 1 , Chang S. Chan 2 , Wenwei Hu 1 and Zhaohui Feng 1 1 Department of Radiation Oncology, Rutgers Cancer Institute of New Jersey, Rutgers, State University of New Jersey, New Brunswick, USA 2 Department of Medicine, Rutgers Cancer Institute of New Jersey, Rutgers, State University of New Jersey, New Brunswick, USA * These two authors contributed equally to this work Correspondence: Wenwei Hu, email: // Zhaohui Feng, email: // Keywords : p53, glycolysis, RRAD, hypoxia, lung cancer Received : May 17, 2014 Accepted : June 24, 2014 Published : June 26, 2014 Abstract Cancer cells display enhanced glycolysis to meet their energetic and biosynthetic demands even under normal oxygen concentrations. Recent studies have revealed that tumor suppressor p53 represses glycolysis under normoxia as a novel mechanism for tumor suppression. As the common microenvironmental stress for tumors, hypoxia drives the metabolic switch from the oxidative phosphorylation to glycolysis, which is crucial for survival and proliferation of cancer cells under hypoxia. The p53’s role and mechanism in regulating glycolysis under hypoxia is poorly understood. Here, we found that p53 represses hypoxia-stimulated glycolysis in cancer cells through RRAD, a newly-identified p53 target. RRAD expression is frequently decreased in lung cancer. Ectopic expression of RRAD greatly reduces glycolysis whereas knockdown of RRAD promotes glycolysis in lung cancer cells. Furthermore, RRAD represses glycolysis mainly through inhibition of GLUT1 translocation to the plasma membrane. Under hypoxic conditions, p53 induces RRAD, which in turn inhibits the translocation of GLUT1 and represses glycolysis in lung cancer cells. Blocking RRAD by siRNA greatly abolishes p53’s function in repressing glycolysis under hypoxia. Taken together, our results revealed an important role and mechanism of p53 in antagonizing the stimulating effect of hypoxia on glycolysis, which contributes to p53’s function in tumor suppression.

Sensitive detection of cancer cells plays a critically important role in the early detection of cancer and cancer metastasis. However, because circulating tumor cells are extremely rare in peripheral blood, the detection of cancer cells with high analytical sensitivity and specificity remains challenging. Here, we have demonstrated a simple, sensitive and specific detection of cancer cells with the detection sensitivity of four cancer cells, which is lower than the cutoff value with respect to correlation with survival outcomes as well as predictive of metastatic disease in clinical diagnostics. We re-engineered the hybridization chain reaction (HCR) to multibranched HCR (mHCR) that can produce long products with multiple biotins for signal amplification and multiple branched arms for multivalent binding. The capturing gold surface is modified with DNA tetrahedral probes, which provide superior hybridization conditions for the multivalent binding. The synergetic effect of mHCR amplification and multivalent binding lead to the high sensitivity of our detection platform.

Dasatinib (BMS-354825) is a FDA-approved multitargeted kinase inhibitor of BCR/ABL and Src kinases. It is now used in the treatment of chronic myelogenous leukemia (CML) with resistance or intolerance to prior therapies, including imatinib. Here we report a novel effect of dasatinib on inducing the differentiation of acute myeloid leukemia (AML) cells through MEK/ERK-dependent activation of signal transducer and activator of transcription 1 (STAT1). We found that dasatinib could induce the differentiation of AML cells as demonstrated by the expression of differentiation marker CD11b, G0/G1 phase arrest and decreased ratio of nucleus to cytoplasm. Of note, dasatinib induced robust phosphorylation of STAT1 both at Tyr701 and Ser727 as well as the redistribution of STAT1 from the cytoplasm to the nucleus, thus leading to the transcription of STAT1-targeted genes. Knocking down STAT1 expression by shRNA significantly attenuated dasatinib-induced differentiation, indicating an important role of STAT1 in myeloid maturation. We further found that dasatinib-induced activation of STAT1 was regulated by the MEK/ERK kinases. The phosporylation of MEK and ERK occurred rapidly upon dasatinib treatment and increased progressively as differentiation was induced. MEK inhibitors PD98059 and U0216 not only inhibited the phosphorylation of STAT1, but also abrogated dasatinib-induced myeloid differentiation, suggesting that MEK/ERK dependent phosphorylation of STAT1 might be indispensable for the differentiating effect of dasatinib in AML cells. Taken together, our study suggests that STAT1 is an important mediator in dasatinib-induced differentiation of AML cells, whose activation requires the activation of MEK/ERK cascades.

Tumor suppressor p53 plays a crucial role in preventing tumor formation. p53 responds to stress signals and transcriptionally regulates its target genes to start various cellular responses, thus preventing tumor initiation and/or progression. p53 is the most frequently-mutated gene in tumors; approximately 50% of human tumors harbor inactivating mutations in p53. The p53 protein is under tight regulation in cells. Loss or the attenuation of p53 function affects cancer risk and tumor progression. Epidemiological studies strongly suggest that chronic psychological stress has negative influences on the onset, progression and mortality of cancers. Studies employing chronic restraint or social isolation in mice, two well-established chronic stress models, have shown that chronic stress enhances tumor growth and metastasis in xenograft tumor assays. However, the direct evidence that chronic stress promotes tumorigenesis in vivo is still lacking, and furthermore, the molecular mechanisms by which chronic stress promotes tumorigenesis remain unclear. To investigate the impact of chronic stress upon tumorigenesis in vivo, we established a mouse model that combines chronic restraint and ionizing radiation (IR)-induced tumorigenesis. IR induces tumorigenesis in both mice and humans. p53+/- mice are susceptible to IR-induced tumorigenesis; a single dose of 4 Gy IR results in the development of tumors, mainly lymphomas, in p53+/- mice. By employing this mouse model system, we found that in p53+/- C57BL6/J mice, chronic restraint stress greatly promoted IR-induced tumorigenesis; chronic restraint reduced IR-induced tumor latency from ∼49 weeks to ∼38 weeks of median survival age (p=0.004). This result provides direct evidence that chronic stress enhances tumorigenesis in vivo. Furthermore, chronic restraint decreased p53 protein levels and function in mice, and promoted the growth of human xenograft tumors in a largely p53-dependent manner. Chronic restraint greatly promotes the growth rate of HCT116 p53+/+ tumors (increased by 3.77-fold) in mice; and has significantly less pronounced promoting effect on the growth of HCT116 p53-/- tumors (increased by 1.52-fold) (p Citation Format: Cen Zhang, Meihua Lin, Yuhan Zhao, Tongsen Zheng, Jiabei Wang, Arnold Levine, Zhaohui Feng, Wenwei Hu. Chronic stress promotes tumorigenesis through attenuation of p53 function . [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 784. doi:10.1158/1538-7445.AM2013-784

BackgroundIt becomes increasingly clear that our current taxonomy of clinical phenotypes is mixed with molecular heterogeneity. Of vital importance for refined clinical practice and improved intervention strategies is to define the hidden molecular distinct diseases using modern large-scale genomic approaches. Microarray omics technology has provided a powerful way to dissect hidden genetic heterogeneity of complex diseases. The aim of this study was thus to develop a bioinformatics approach to seek the transcriptional features leading to the hidden subtyping of a complex clinical phenotype. The basic strategy of the proposed method was to iteratively partition in two ways sample and feature space with super-paramagnetic clustering technique and to seek for hard and robust gene clusters that lead to a natural partition of disease samples and that have the highest functionally conceptual consensus evaluated with Gene Ontology.ResultsWe applied the proposed method to two publicly available microarray datasets of diffuse large B-cell lymphoma (DLBCL), a notoriously heterogeneous phenotype. A feature subset of 30 genes (38 probes) derived from analysis of the first dataset consisting of 4026 genes and 42 DLBCL samples identified three categories of patients with very different five-year overall survival rates (70.59%, 44.44% and 14.29% respectively;p= 0.0017). Analysis of the second dataset consisting of 7129 genes and 58 DLBCL samples revealed a feature subset of 13 genes (16 probes) that not only replicated the findings of the important DLBCL genes (e.g.JAW1andBCL7A), but also identified three clinically similar subtypes (with 5-year overall survival rates of 63.13%, 34.92% and 15.38% respectively;p= 0.0009) to those identified in the first dataset. Finally, we built a multivariate Cox proportional-hazards prediction model for each feature subset and definedJAW1as one of the most significant predictor (p= 0.005 and 0.014; hazard ratios = 0.02 and 0.03, respectively for two datasets) for both DLBCL cohorts under study.ConclusionOur results showed that the proposed algorithm is a promising computational strategy for peeling off the hidden genetic heterogeneity based on transcriptionally profiling disease samples, which may lead to an improved diagnosis and treatment of cancers.