Using machine learning to improve the accuracy of patient deterioration predictions: Mayo Clinic Early Warning Score (MC-EWS).

OBJECTIVE: We aimed to develop a model for accurate prediction of general care inpatient deterioration. MATERIALS AND METHODS: Training and internal validation datasets were built using 2-year data from a quaternary hospital in the Midwest. Model training used gradient boosting and feature engineering (clinically relevant interactions, time-series information) to predict general care inpatient deterioration (resuscitation call, intensive care unit transfer, or rapid response team call) in 24 hours. Data from a tertiary care hospital in the Southwest were used for external validation. C-statistic, sensitivity, positive predictive value, and alert rate were calculated for different cutoffs and compared with the National Early Warning Score. Sensitivity analysis evaluated prediction of intensive care unit transfer or resuscitation call. RESULTS: Training, internal validation, and external validation datasets included 24 500, 25 784 and 53 956 hospitalizations, respectively. The Mayo Clinic Early Warning Score (MC-EWS) demonstrated excellent discrimination in both the internal and external validation datasets (C-statistic = 0.913, 0.937, respectively), and results were consistent in the sensitivity analysis (C-statistic = 0.932 in external validation). At a sensitivity of 73%, MC-EWS would generate 0.7 alerts per day per 10 patients, 45% less than the National Early Warning Score. DISCUSSION: Low alert rates are important for implementation of an alert system. Other early warning scores developed for the general care ward have achieved lower discrimination overall compared with MC-EWS, likely because MC-EWS includes both nursing assessments and extensive feature engineering. CONCLUSIONS: MC-EWS achieved superior prediction of general care inpatient deterioration using sophisticated feature engineering and a machine learning approach, reducing alert rate.

Dual VEGF inhibition with sorafenib and bevacizumab as salvage therapy in metastatic colorectal cancer: results of the phase II North Central Cancer Treatment Group study N054C (Alliance).

BACKGROUND: Bevacizumab (BEV), a monoclonal antibody against vascular endothelial growth factor-A (VEGF-A), is a standard component of medical therapy of metastatic colorectal cancer (mCRC). Activation of alternative angiogenesis pathways has been implicated in resistance to BEV. This phase II study examines the activity of combined vertical blockade of VEGF signaling with sorafenib and BEV as salvage therapy in patients with progressive disease (PD) on all standard therapy in mCRC. METHODS: mCRC patients with documented PD on standard therapy, received sorafenib (200 mg orally twice daily, days 1-5 and 8-12) and BEV (5 mg/kg intravenously, day 1) every 2 weeks. Primary endpoint was 3-month progression-free survival (PFS) rate and secondary endpoints were overall survival (OS), response rate (RR), safety, and feasibility. RESULTS: Of the 83 patients enrolled, 79 were evaluable. Of these, 42 (53%) were progression-free at 3 months. Median PFS was 3.5 months and median OS was 8.3 months. One patient had a partial response and 50 patients (63.3%) had at least one stable tumor assessment. Of 79 evaluable patients, 54 (68%) experienced grade 3/4 adverse events (AEs) at least possibly related to treatment. Most frequent grade 3/4 AEs were: fatigue (24.1%), hypertension (16.5%), elevated lipase (8.9%), hand-foot skin reaction (8.9%), diarrhea (7.6%), and proteinuria (7.6%). Reasons for treatment discontinuation were PD (72%), AEs (18%), patient refusal (8%), physician decision (1%), and death (1%). CONCLUSIONS: The combination of BEV and sorafenib as salvage therapy in heavily pretreated mCRC patients is tolerable and manageable, with evidence of promising activity. CLINICALTRIALSGOV IDENTIFIER: NCT00826540, URL:http://clinicaltrials.gov/ct2/show/NCT00826540.

Prevalence and Outcome of High-Risk QT Prolongation Recorded in the Emergency Department from an Institution-Wide QT Alert System.

BACKGROUND: QT prolongation is an independent risk factor for sudden death, stroke, and all-cause mortality. However, additional studies have shown that in certain settings, QT prolongation may be transient and a result of external factors. OBJECTIVE: In this study, we evaluated the clinical characteristics and outcomes of patients seen in the emergency department (ED) with QT prolongation. METHODS: Between November 2010 and June 2011, 7522 patients had an electrocardiogram (ECG) obtained during their evaluation in the ED. Clinical, laboratory, and therapeutic information was collected for all patients with QT prolongation (i.e., ≥ 500 ms and QRS < 120 ms). Potential QT-inciting factors (drugs, electrolyte disturbances, and comorbidities) were synthesized into a pro-QT score. RESULTS: Among the 7522 patients with an ECG obtained in the ED, a QT alert was activated in 93 (1.2%; mean QTc 521 ± 34 ms). The majority of ED patients (64%) had more than one underlying condition associated with QT prolongation, with electrolyte disturbances in 51%, a QT prolonging condition in 56%, and QT-prolonging drugs in 77%. Thirty-day mortality was 13% for patients with QT prolongation noted in the ED. CONCLUSIONS: One percent of patients evaluated with an ECG in the ED activated our prolonged QTc warning system, with most demonstrating > 1 QT-prolonging condition. Thirty-day mortality was significant, but it requires further investigation to determine whether the QTc simply provided a non-invasive indicator of increased risk or heralded the presence of a vulnerable host at risk of a QT-mediated sudden dysrhythmic death.

Providers' Response to Clinical Decision Support for QT Prolonging Drugs.

Commonly used drugs in hospital setting can cause QT prolongation and trigger life-threatening arrhythmias. We evaluate changes in prescribing behavior after the implementation of a clinical decision support system to prevent the use of QT prolonging medications in the hospital setting. We conducted a quasi-experimental study, before and after the implementation of a clinical decision support system integrated in the electronic medical record (QT-alert system). This system detects patients at risk of significant QT prolongation (QTc>500ms) and alerts providers ordering QT prolonging drugs. We reviewed the electronic health record to assess the provider's responses which were classified as "action taken" (QT drug avoided, QT drug changed, other QT drug(s) avoided, ECG monitoring, electrolytes monitoring, QT issue acknowledged, other actions) or "no action taken". Approximately, 15.5% (95/612) of the alerts were followed by a provider's action in the pre-intervention phase compared with 21% (228/1085) in the post-intervention phase (p=0.006). The most common type of actions taken during pre-intervention phase compared to post-intervention phase were ECG monitoring (8% vs. 13%, p=0.002) and QT issue acknowledgment (2.1% vs. 4.1%, p=0.03). Notably, there was no significant difference for other actions including QT drug avoided (p=0.8), QT drug changed (p=0.06) and other QT drug(s) avoided (p=0.3). Our study demonstrated that the QT alert system prompted a higher proportion of providers to take action on patients at risk of complications. However, the overall impact was modest underscoring the need for educating providers and optimizing clinical decision support to further reduce drug-induced QT prolongation.

Evaluating Automated Rules for Rapid Response System Alarm Triggers in Medical and Surgical Patients.

BACKGROUND: The use of rapid response systems (RRS), which were designed to bring clinicians with critical care expertise to the bedside to prevent unnecessary deaths, has increased. RRS rely on accurate detection of acute deterioration events. Early warning scores (EWS) have been used for this purpose but were developed using heterogeneous populations. Predictive performance may differ in medical vs surgical patients. OBJECTIVE: To evaluate the performance of published EWS in medical vs surgical patient populations. DESIGN: Retrospective cohort study. SETTING: Two tertiary care academic medical center hospitals in the Midwest totaling more than 1500 beds. PATIENTS: All patients discharged from January to December 2011. INTERVENTION: None. MEASUREMENTS: Time-stamped longitudinal database of patient variables and outcomes, categorized as surgical or medical. Outcomes included unscheduled transfers to the intensive care unit, activation of the RRS, and calls for cardiorespiratory resuscitation ("resuscitation call"). The EWS were calculated and updated with every new patient variable entry over time. Scores were considered accurate if they predicted an outcome in the following 24 hours. RESULTS: All EWS demonstrated higher performance within the medical population as compared to surgical: higher positive predictive value (P < .0001 for all scores) and sensitivity (P < .0001 for all scores). All EWS had positive predictive values below 25%. CONCLUSIONS: The overall poor performance of the evaluated EWS was marginally better in medical patients when compared to surgical patients. Journal of Hospital Medicine 2017;12:217-223.

Frequency and cause of transient QT prolongation after surgery.

Patients undergoing surgery are often exposed to QT-inciting factors that may increase the risk for complications. We evaluated the clinical characteristics and outcomes of patients with QTc ≥500 ms within the first 24 hours after surgery as identified by an institution-wide electrocardiogram alert system. From November 2010 to June 2011, 470 patients exhibited an electrocardiographically isolated QTc ≥500 ms. QT prolongation after surgery was the setting for >1 of every 10 QTc alerts (59 patients). We determined the presence of QT prolonging medical conditions, drugs, electrolyte abnormalities, and the surgical patient's clinical outcome. The average preoperative QTc of the 59 patients demonstrating perioperative QT prolongation was 463 ± 56 ms with a postoperative QTc increase of 54 ± 37 ms. Most patients (n = 48, 83%) had ≥1 known QT-inciting factor before surgery. Compared with presurgical findings, there was a significant increase in pro-QTc score after surgery (1.8 ± 1.5 vs 3.5 ± 2.0, p <0.01) indicating a greater burden of perioperative QT-inciting factors. In conclusion, nearly all cases of QT prolongation could be explained by known etiologic or iatrogenic factors suggesting that maladaptive cardiac repolarization is most likely not a transient, postoperative stress response and may be avoided by altering clinical management.

Phenotype of Children with QT Prolongation Identified Using an Institution-Wide QT Alert System.

QT prolongation is an independent risk factor for cardiovascular mortality in adults. However, there is little information available on pediatric patients with QT prolongation and their outcomes. Herein, we evaluated the prevalence of QT prolongation in pediatric patients identified by an institution-wide QT alert system, and the spectrum of their phenotype. Patients with documented QT prolongation on an ECG obtained between November 2010 and June 2011 were included. There were 1303 pediatric ECGs, and 68 children had electrographically isolated QT prolongation. Comprehensive review of medical records was performed with particular attention to QT-prolonging clinical, laboratory, and medication data, which were summarized into a pro-QTc score. Overall, 68 (5 %) pediatric patients had isolated QT prolongation. The mean age of this pediatric cohort was 9 ± 6 years, and the average QTc was 494 ± 42 ms. All children had 1 or more QT-prolonging risk factor(s), most commonly QT-prolonging medications. One patient was identified with congenital long QT syndrome (LQTS), which was not previously diagnosed. In one-year follow-up, only one pediatric death (non-cardiac) occurred (1.5 %). Potentially QT-offending/pro-arrhythmic medications were changed in 80 % of pediatric patients after the physician received the QT alert. Children with QT prolongation had very low mortality and minimal polypharmacy. Still, medications and other modifiable conditions were the most common causes of QT prolongation. Children with a prolonged QTc should be evaluated for modifiable QT-prolonging factors. However, if no risk factors are present or the QTc does not attenuate after risk factor modification/removal, the child should be evaluated for congenital LQTS.

Impact of clinical decision support preventing the use of QT-prolonging medications for patients at risk for torsade de pointes.

We developed and implemented a 'CPOE-QT Alert' system, that is, clinical decision support integrated in the computerized physician order entry system (CPOE), in 2011. The system identifies any attempts to order medications with risk of torsade de pointes (TdP) for patients with a history of significant QT prolongation (QTc ≥500 ms) and alerts the provider entering the order. We assessed its impact by comparing orders and subsequent medication administration before and after activation of the system. We found a significant decrease in the proportion of completed order per ordering attempt after system activation (94% (1293/1379) vs 77% (1888/2453), difference 16.8%; p<0.001). This resulted in a 13.9% reduction in the administration of those medications to patients. A significant decrease was observed across all provider types, educational levels, and specialties. The CPOE-QT Alert system successfully reduced exposure to QT-prolonging medications in high risk patients.

Impact of left ventricular hypertrophy on QT prolongation and associated mortality.

BACKGROUND: QT prolongation on electrocardiogram (ECG) is a risk marker of ventricular arrhythmias and all-cause mortality. Left ventricular hypertrophy (LVH) on ECG is also associated with poor outcome. Patients satisfying ECG voltage criteria for LVH frequently show concomitant QT prolongation. OBJECTIVE: This study aimed to explore the impact of marked QT prolongation on all-cause mortality in patients copresenting with LVH voltage criteria and prolonged QT on ECG. METHODS: We evaluated 3364 ECGs with corrected QT (QTc) interval ≥460 ms detected by Mayo Clinic's QT alert system from November 2010 through June 2011. Every ECG with QTc interval ≥460 ms was evaluated for the presence of LVH voltage criteria by using Sokolow-Lyon voltage, Cornell voltage, and Cornell product. RESULTS: Concomitant LVH voltage criteria were present in 181 of 3364 ECGs (5.3%) with QTc interval ≥460 ms. Mortality during a follow-up period of 217 ± 184 days was 13% (23 of 181). Independent of age and hypertension, the QTc interval predicted mortality in patients with LVH voltage criteria (hazard ratio 1.31 per 10-ms increase; 95% confidence interval 1.09-1.58; P < .01). Patients with LVH voltage criteria and QTc interval ≥500 ms had highest mortality (log rank, P < .001). CONCLUSION: The QTc interval was an independent predictor of mortality in patients with concomitant LVH voltage and prolonged QTc interval on ECG. Mortality was highest in those with QTc interval ≥500 ms. QT prolongation on ECGs with concomitant LVH voltage criteria should not be regarded as a harmless byproduct of LVH, but should be used as a significant marker of increased mortality risk similar to that in patients without LVH voltage criteria.

Widely used track and trigger scores: are they ready for automation in practice?

INTRODUCTION: Early Warning Scores (EWS) are widely used for early recognition of patient deterioration. Automated alarm/alerts have been recommended as a desirable characteristic for detection systems of patient deterioration. We undertook a comparative analysis of performance characteristics of common EWS methods to assess how they would function if automated. METHODS: We evaluated the most widely used EWS systems (MEWS, SEWS, GMEWS, Worthing, ViEWS and NEWS) and the Rapid Response Team (RRT) activation criteria in use in our institution. We compared their ability to predict the composite outcome of Resuscitation call, RRS activation or unplanned transfer to the ICU, in a time-dependent manner (3, 8, 12, 24 and 36 h after the observation) by determining the sensitivity, specificity and positive predictive values (PPV). We used a large vital signs database (6,948,689 unique time points) from 34,898 unique consecutive hospitalized patients. RESULTS: PPVs ranged from less than 0.01 (Worthing, 3 h) to 0.21 (GMEWS, 36 h). Sensitivity ranged from 0.07 (GMEWS, 3 h) to 0.75 (ViEWS, 36 h). Used in an automated fashion, these would correspond to 1040-215,020 false positive alerts per year. CONCLUSIONS: When the evaluation is performed in a time-sensitive manner, the most widely used weighted track-and-trigger scores do not offer good predictive capabilities for use as criteria for an automated alarm system. For the implementation of an automated alarm system, better criteria need to be developed and validated before implementation.

Evaluating implementation of a rapid response team: considering alternative outcome measures.

OBJECTIVE: Determine the prolonged effect of rapid response team (RRT) implementation on failure to rescue (FTR). DESIGN: Longitudinal study of institutional performance with control charts and Bayesian change point (BCP) analysis. SETTING: Two academic hospitals in Midwest, USA. PARTICIPANTS: All inpatients discharged between 1 September 2005 and 31 December 2010. INTERVENTION: Implementation of an RRT serving the Mayo Clinic Rochester system was phased in for all inpatient services beginning in September 2006 and was completed in February 2008. MAIN OUTCOME MEASURE: Modified version of the AHRQ FTR measure, which identifies hospital mortalities among medical and surgical patients with specified in-hospital complications. RESULTS: A decrease in FTR, as well as an increase in the unplanned ICU transfer rate, occurred in the second-year post-RRT implementation coinciding with an increase in RRT calls per month. No significant decreases were observed pre- and post-implementation for cardiopulmonary resuscitation events or overall mortality. A significant decrease in mortality among non-ICU discharges was identified by control charts, although this finding was not detected by BCP or pre- vs. post-analyses. CONCLUSIONS: Reduction in the FTR rate was associated with a substantial increase in the number of RRT calls. Effects of RRT may not be seen until RRT calls reach a sufficient threshold. FTR rate may be better at capturing the effect of RRT implementation than the rate of cardiac arrests. These results support prior reports that short-term studies may underestimate the impact of RRT systems, and support the need for ongoing monitoring and assessment of outcomes to facilitate best resource utilization.

Institution-wide QT alert system identifies patients with a high risk of mortality.

OBJECTIVES: To determine the phenotype and outcome of patients with QTc of at least 500 ms and to create a pro-QTc risk score for mortality. PATIENTS AND METHODS: An institution-wide computer-based QT alert system was developed and implemented at Mayo Clinic in Rochester, Minnesota. This system screens all electrocardiograms (ECGs) performed and alerts the physician if the QTc is 500 ms or greater. Between November 10, 2010, and June 30, 2011, 86,107 ECGs were performed in 52,579 patients. Clinical diagnoses, laboratory abnormalities, and medications known to influence the QT interval were collected from the medical records and summarized in a new pro-QTc score. Survival was compared with that of the 51,434 Mayo Clinic patients with a QTc less than 500 ms during the same period. RESULTS: QT alerts were sent for 1145 patients (2%); of these, 470 (41%) had no other identifiable ECG reason for QT prolongation (eg, pacing). All-cause mortality during a mean ± SD of 224 ± 174 days of follow-up was 19% in those with QTc of 500 ms or greater compared with 5% in patients with QTc less than 500 ms (log-rank P<.001). The pro-QTc score was an age-independent predictor of mortality (pro-QTc score: hazard ratio, 1.18; 95% CI, 1.05-1.32; P=.006; age: hazard ratio, 1.02; 95% CI, 1.01-1.03; P=.004.). QT-prolonging medications accounted for 37% of the pro-QTc score. CONCLUSION: This novel institution-wide QT alert system identified patients with a high risk of mortality. The pro-QTc score, reflecting patients' multimorbidity and multipharmacy, was an independent predictor of mortality. The QT alert system may increase a physician's awareness of a high-risk patient. Potentially lifesaving interventions can be facilitated by reducing the modifiable factors of the pro-QTc score.

Gemcitabine and docetaxel for hepatocellular carcinoma: a phase II North Central Cancer Treatment Group clinical trial.

OBJECTIVES: Few effective options are available for the treatment of unresectable hepatocellular carcinoma (HCC). Several phase I trials suggest promising activity of a combination of gemcitabine and docetaxel. METHODS: Patients with unresectable or metastatic HCC were treated with docetaxel 40 mg/m (later reduced to 30 mg/m) and gemcitabine 800 mg/m on days 1 and 8 every 3 weeks. Twenty-five patients were enrolled in 26 months. Median age was 64 (range, 27-078), 17 were male, 14 had liver-only disease, and 11 had extrahepatic disease. RESULTS: Of 25 patients evaluable for the primary end point (response), 2 (8%) have a confirmed partial response. The median time to progression is 2.76 months (95% confidence interval, 1.84-6.64 mo). Median survival was 12.8 months (95% confidence interval, 5.26-28.00). Two patients died on study owing to adverse events (1 hepatic and 1 renal failure), neither of which were attributed to the study medications. Twenty patients (81%) have experienced grade 3+ adverse events, including 11 with grade 4+ adverse events, primarily neutropenia, thrombocytopenia, diarrhea, and fatigue. CONCLUSIONS: Although this combination seems to have potential benefit, as measured by overall survival, its toxicity and the recent introduction of sorafenib has further limited the use of chemotherapy. Approaches other than chemotherapy are likely to be of the greatest potential benefit.

Cediranib (AZD2171) in patients with advanced hepatocellular carcinoma: a phase II North Central Cancer Treatment Group Clinical Trial.

OBJECTIVES: Vascular endothelial growth factor has been shown to be overexpressed in several studies of hepatocellular carcinoma (HCC). Cediranib is a potent inhibitor of vascular endothelial growth factor signaling. We assessed the efficacy and toxicity of cediranib in patients with HCC. METHODS: Twenty-eight patients with unresectable or metastatic HCC were enrolled in this study. Patients received 45 mg of cediranib orally, once daily, for 28-day cycles. The primary objective of this phase II study was to assess 6-month survival. Secondary objectives were to assess tumor response, time to progression, and toxicity. RESULTS: All 28 patients were evaluable for efficacy outcomes. Twelve patients (42.9%) survived 6 months, 15 (53.6%) died within 6 months, and 1 (3.6%) was lost to follow-up before 6 months. The median overall survival was 5.8 months (95% confidence interval, 3.4-7.3 mo). No patients experienced confirmed response. The median time to progression was 2.8 months (95% confidence interval, 2.3-4.4 mo). Twenty-six patients (93%) experienced a grade 3+ adverse event with the most common adverse event s being fatigue (46%), anorexia (25%), hypertension (21%), and elevated alanine aminotransferase (18%). CONCLUSIONS: Owing to the toxicity, cediranib at this dose and schedule is not an effective treatment in patients with unresectable or metastatic HCC.

miRNA expression in colon polyps provides evidence for a multihit model of colon cancer.

Changes in miRNA expression are a common feature in colon cancer. Those changes occurring in the transition from normal to adenoma and from adenoma to carcinoma, however, have not been well defined. Additionally, miRNA changes among tumor subgroups of colon cancer have also not been adequately evaluated. In this study, we examined the global miRNA expression in 315 samples that included 52 normal colonic mucosa, 41 tubulovillous adenomas, 158 adenocarcinomas with proficient DNA mismatch repair (pMMR) selected for stage and age of onset, and 64 adenocarcinomas with defective DNA mismatch repair (dMMR) selected for sporadic (n = 53) and inherited colon cancer (n = 11). Sporadic dMMR tumors all had MLH1 inactivation due to promoter hypermethylation. Unsupervised PCA and cluster analysis demonstrated that normal colon tissue, adenomas, pMMR carcinomas and dMMR carcinomas were all clearly discernable. The majority of miRNAs that were differentially expressed between normal and polyp were also differentially expressed with a similar magnitude in the comparison of normal to both the pMMR and dMMR tumor groups, suggesting a stepwise progression for transformation from normal colon to carcinoma. Among the miRNAs demonstrating the largest fold up- or down-regulated changes (≥4), four novel (miR-31, miR-1, miR-9 and miR-99a) and two previously reported (miR-137 and miR-135b) miRNAs were identified in the normal/adenoma comparison. All but one of these (miR-99a) demonstrated similar expression differences in the two normal/carcinoma comparisons, suggesting that these early tumor changes are important in both the pMMR- and dMMR-derived cancers. The comparison between pMMR and dMMR tumors identified four miRNAs (miR-31, miR-552, miR-592 and miR-224) with statistically significant expression differences (≥2-fold change).

Expression of CD74 in high grade gliomas: a potential role in temozolomide resistance.

Temozolomide (TMZ) is the most effective chemotherapeutic agent for glioblastoma (GBM). Resistance to this methylating agent is linked to DNA repair enzyme O6-methylguanine-DNA methyltransferase (MGMT). However, in recent studies MGMT status was not completely accurate as a predictor of TMZ response in GBM, suggesting other mechanisms of resistance. As part of an effort aimed at discovery of genes involved in TMZ resistance in GBM, the expression of CD74 was evaluated in GBM patient samples and the influence of CD74 on TMZ response was evaluated in GBM tumor models. Reverse transcription-polymerase-chain reaction (RT-PCR) demonstrated differential expression of CD74 mRNA among the GBM xenografts; 8 of 20 (40%) expressed CD74 mRNA. In a preliminary evaluation of whether CD74 expression might influence TMZ response, CD74 mRNA expression levels were inversely associated with in vivo TMZ resistance in 20 GBM xenograft lines (median survival 122 vs. 62.5 days; r = -0.48, P = 0.032). In follow up to this observation, CD74 shRNA knock down in U87 cells significantly suppressed in vitro proliferation and increased TMZ sensitivity as compared to a non-specific control shRNA. Consistent with an effect on proliferation and survival, silencing of CD74 by shRNA was associated with reduced Akt and Erk1/2 activation in response to stimulation by CD74 ligand macrophage-migration inhibition factor (MIF). Lastly, expression of CD74 protein was assessed in patient samples [nine anaplastic astrocytoma (AA), and 62 GBM] by immunohistochemistry, and appreciable expression was observed in 28% of samples. Collectively, these findings suggest that CD74 is expressed in a subset of high grade gliomas and may contribute to TMZ resistance.

Human colon cancer profiles show differential microRNA expression depending on mismatch repair status and are characteristic of undifferentiated proliferative states.

BACKGROUND: Colon cancer arises from the accumulation of multiple genetic and epigenetic alterations to normal colonic tissue. microRNAs (miRNAs) are small, non-coding regulatory RNAs that post-transcriptionally regulate gene expression. Differential miRNA expression in cancer versus normal tissue is a common event and may be pivotal for tumor onset and progression. METHODS: To identify miRNAs that are differentially expressed in tumors and tumor subtypes, we carried out highly sensitive expression profiling of 735 miRNAs on samples obtained from a statistically powerful set of tumors (n = 80) and normal colon tissue (n = 28) and validated a subset of this data by qRT-PCR. RESULTS: Tumor specimens showed highly significant and large fold change differential expression of the levels of 39 miRNAs including miR-135b, miR-96, miR-182, miR-183, miR-1, and miR-133a, relative to normal colon tissue. Significant differences were also seen in 6 miRNAs including miR-31 and miR-592, in the direct comparison of tumors that were deficient or proficient for mismatch repair. Examination of the genomic regions containing differentially expressed miRNAs revealed that they were also differentially methylated in colon cancer at a far greater rate than would be expected by chance. A network of interactions between these miRNAs and genes associated with colon cancer provided evidence for the role of these miRNAs as oncogenes by attenuation of tumor suppressor genes. CONCLUSION: Colon tumors show differential expression of miRNAs depending on mismatch repair status. miRNA expression in colon tumors has an epigenetic component and altered expression that may reflect a reversion to regulatory programs characteristic of undifferentiated proliferative developmental states.

Evaluation of a new high-dimensional miRNA profiling platform.

BACKGROUND: MicroRNAs (miRNAs) are a class of approximately 22 nucleotide long, widely expressed RNA molecules that play important regulatory roles in eukaryotes. To investigate miRNA function, it is essential that methods to quantify their expression levels be available. METHODS: We evaluated a new miRNA profiling platform that utilizes Illumina's existing robust DASL chemistry as the basis for the assay. Using total RNA from five colon cancer patients and four cell lines, we evaluated the reproducibility of miRNA expression levels across replicates and with varying amounts of input RNA. The beta test version was comprised of 735 miRNA targets of Illumina's miRNA profiling application. RESULTS: Reproducibility between sample replicates within a plate was good (Spearman's correlation 0.91 to 0.98) as was the plate-to-plate reproducibility replicates run on different days (Spearman's correlation 0.84 to 0.98). To determine whether quality data could be obtained from a broad range of input RNA, data obtained from amounts ranging from 25 ng to 800 ng were compared to those obtained at 200 ng. No effect across the range of RNA input was observed. CONCLUSION: These results indicate that very small amounts of starting material are sufficient to allow sensitive miRNA profiling using the Illumina miRNA high-dimensional platform. Nonlinear biases were observed between replicates, indicating the need for abundance-dependent normalization. Overall, the performance characteristics of the Illumina miRNA profiling system were excellent.

Characterization and gene expression profiling in glioma cell lines with deletion of chromosome 19 before and after microcell-mediated restoration of normal human chromosome 19.

Nearly 10% of human gliomas are oligodendrogliomas. Deletion of chromosome arm 19q, often in conjunction with deletion of 1p, has been observed in 65-80% of these tumors. This has suggested the presence of a tumor suppressor gene located on the 19q arm. Chromosome 19 deletion is also of interest due to the better prognosis of patients with deletion, including longer survival and better response to chemotherapy, compared with patients without deletion. Two glioma cell lines with deletion of 19q were used for chromosome 19 microcell-mediated transfer, to assess the effect of replacing the deleted segment. Complementation with chromosome 19 significantly reduced the growth rate of the hybrid cells compared with the parental cell lines. Affymetrix U133 Plus 2.0 Gene Chip analysis was performed to measure and compare the expression of the chromosome 19 genes in the chromosome 19 hybrid cell lines to the parental cell line. Probes were considered significantly different when a P value <0.01 was seen in all of the cell line comparisons. Of 345 probes within the commonly deleted 19q region, seven genes (APOE, RCN3, FLJ10781, SAE1, STRN4, CCDC8, and BCL2L12) were identified as potential candidate genes. RT-PCR analysis of primary tumor specimens showed that several genes had significant differences when stratified by tumor morphology or deletion status. This suggests that one or more of these candidates may play a role in glioma formation or progression.

A tissue biomarker panel predicting systemic progression after PSA recurrence post-definitive prostate cancer therapy.

BACKGROUND: Many men develop a rising PSA after initial therapy for prostate cancer. While some of these men will develop a local or metastatic recurrence that warrants further therapy, others will have no evidence of disease progression. We hypothesized that an expression biomarker panel can predict which men with a rising PSA would benefit from further therapy. METHODOLOGY/PRINCIPAL FINDINGS: A case-control design was used to test the association of gene expression with outcome. Systemic (SYS) progression cases were men post-prostatectomy who developed systemic progression within 5 years after PSA recurrence. PSA progression controls were matched men post-prostatectomy with PSA recurrence but no evidence of clinical progression within 5 years. Using expression arrays optimized for paraffin-embedded tissue RNA, 1021 cancer-related genes were evaluated-including 570 genes implicated in prostate cancer progression. Genes from 8 previously reported marker panels were included. A systemic progression model containing 17 genes was developed. This model generated an AUC of 0.88 (95% CI: 0.84-0.92). Similar AUCs were generated using 3 previously reported panels. In secondary analyses, the model predicted the endpoints of prostate cancer death (in SYS cases) and systemic progression beyond 5 years (in PSA controls) with hazard ratios 2.5 and 4.7, respectively (log-rank p-values of 0.0007 and 0.0005). Genes mapped to 8q24 were significantly enriched in the model. CONCLUSIONS/SIGNIFICANCE: Specific gene expression patterns are significantly associated with systemic progression after PSA recurrence. The measurement of gene expression pattern may be useful for determining which men may benefit from additional therapy after PSA recurrence.