Patients with Muscle-Invasive Bladder Cancer with Nonluminal Subtype Derive Greatest Benefit from Platinum Based Neoadjuvant Chemotherapy.

PURPOSE: Neoadjuvant chemotherapy (NAC) prior to radical cystectomy (RC) in patients with nonmetastatic muscle-invasive bladder cancer (MIBC) confers an absolute survival benefit of 5%-10%. There is evidence that molecular differences between tumors may impact response to therapy, highlighting a need for clinically validated biomarkers to predict response to NAC. MATERIALS AND METHODS: Four bladder cancer cohorts were included. Inverse probability weighting was used to make baseline characteristics (age, sex and clinical tumor stage) between NAC-treated and untreated groups more comparable. Molecular subtypes were determined using a commercial genomic subtyping classifier. Survival rates were estimated using weighted Kaplan-Meier curves. Cox proportional hazards models were used to evaluate the primary and secondary study end points of overall survival (OS) and cancer-specific survival, respectively. RESULTS: A total of 601 patients with MIBC were included, of whom 247 had been treated with NAC and RC, and 354 underwent RC without NAC. With NAC, the overall net benefit to OS and cancer-specific survival at 3 years was 7% and 5%, respectively. After controlling for clinicopathological variables, nonluminal tumors had greatest benefit from NAC, with 10% greater OS at 3 years (71% vs 61%), while luminal tumors had minimal benefit (63% vs 65%) for NAC vs non-NAC. CONCLUSIONS: In patients with MIBC, a commercially available molecular subtyping assay revealed nonluminal tumors received the greatest benefit from NAC, while patients with luminal tumors experienced a minimal survival benefit. A genomic classifier may help identify patients with MIBC who would benefit most from NAC.

Transcriptome evaluation of the relation between body mass index and prostate cancer outcomes.

BACKGROUND: Large epidemiological studies indicate that an increased body mass index (BMI) is associated with increased prostate cancer (PCa) mortality. Data indicate that there is no association between elevated metabolic pathway proteins and PCa mortality. There are no published studies evaluating the relation between BMI and metabolic pathways with respect to PCa outcomes with a genomics approach. METHODS: The Decipher Genomic Resource Information Database was queried for patients who had undergone prostatectomy and had BMI information available. These patients came from Thomas Jefferson University (TJU) and Johns Hopkins Medical Institution (JHMI); the latter provided 2 cohorts (I and II). A high-BMI group (≥30 kg/m2 ) and a low-BMI group (<25 kg/m2 ) were identified, and genomic data were interrogated for differentially expressed genes with an interquartile range filter and a Wilcoxon test. P values were adjusted for multiple testing with the Benjamini-Hochberg false-discovery rate method. RESULTS: A total of 477 patients with a median follow-up of 108 months had BMI information available. Two genes were found to interact with BMI in both the JHMI I cohort and the TJU cohort, but there was no statistical significance after adjustments for multiple comparisons. Aberrant metabolic gene expression was significantly correlated with distant metastases (P < .05). No relation was found between BMI and metastases or overall survival (both P values > .05). CONCLUSIONS: In a genomic analysis of prostatectomy specimens, metabolic gene expression, but not BMI, was associated with PCa metastases. Cancer 2017;123:2240-2247. © 2017 American Cancer Society.

External Validation of a Digital Pathology-based Multimodal Artificial Intelligence Architecture in the NRG/RTOG 9902 Phase 3 Trial.

BACKGROUND: Accurate risk stratification is critical to guide management decisions in localized prostate cancer (PCa). Previously, we had developed and validated a multimodal artificial intelligence (MMAI) model generated from digital histopathology and clinical features. Here, we externally validate this model on men with high-risk or locally advanced PCa treated and followed as part of a phase 3 randomized control trial. OBJECTIVE: To externally validate the MMAI model on men with high-risk or locally advanced PCa treated and followed as part of a phase 3 randomized control trial. DESIGN, SETTING, AND PARTICIPANTS: Our validation cohort included 318 localized high-risk PCa patients from NRG/RTOG 9902 with available histopathology (337 [85%] of the 397 patients enrolled into the trial had available slides, of which 19 [5.6%] failed due to poor image quality). OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Two previously locked prognostic MMAI models were validated for their intended endpoint: distant metastasis (DM) and PCa-specific mortality (PCSM). Individual clinical factors and the number of National Comprehensive Cancer Network (NCCN) high-risk features served as comparators. Subdistribution hazard ratio (sHR) was reported per standard deviation increase of the score with corresponding 95% confidence interval (CI) using Fine-Gray or Cox proportional hazards models. RESULTS AND LIMITATIONS: The DM and PCSM MMAI algorithms were significantly and independently associated with the risk of DM (sHR [95% CI] = 2.33 [1.60-3.38], p < 0.001) and PCSM, respectively (sHR [95% CI] = 3.54 [2.38-5.28], p < 0.001) when compared against other prognostic clinical factors and NCCN high-risk features. The lower 75% of patients by DM MMAI had estimated 5- and 10-yr DM rates of 4% and 7%, and the highest quartile had average 5- and 10-yr DM rates of 19% and 32%, respectively (p < 0.001). Similar results were observed for the PCSM MMAI algorithm. CONCLUSIONS: We externally validated the prognostic ability of MMAI models previously developed among men with localized high-risk disease. MMAI prognostic models further risk stratify beyond the clinical and pathological variables for DM and PCSM in a population of men already at a high risk for disease progression. This study provides evidence for consistent validation of our deep learning MMAI models to improve prognostication and enable more informed decision-making for patient care. PATIENT SUMMARY: This paper presents a novel approach using images from pathology slides along with clinical variables to validate artificial intelligence (computer-generated) prognostic models. When implemented, clinicians can offer a more personalized and tailored prognostic discussion for men with localized prostate cancer.

Analysis of a Biopsy-Based Genomic Classifier in High-Risk Prostate Cancer: Meta-Analysis of the NRG Oncology/Radiation Therapy Oncology Group 9202, 9413, and 9902 Phase 3 Randomized Trials.

PURPOSE: Decipher is a genomic classifier (GC) prospectively validated postprostatectomy. We validated the performance of the GC in pretreatment biopsy samples within the context of 3 randomized phase 3 high-risk definitive radiation therapy trials. METHODS AND MATERIALS: A prespecified analysis plan (NRG-GU-TS006) was approved to obtain formalin-fixed paraffin-embedded tissue from biopsy specimens from the NRG biobank from patients enrolled in the NRG/Radiation Therapy Oncology Group (RTOG) 9202, 9413, and 9902 phase 3 randomized trials. After central review, the highest-grade tumors were profiled on clinical-grade whole-transcriptome arrays and GC scores were obtained. The primary objective was to validate the independent prognostic ability for the GC for distant metastases (DM), and secondary for prostate cancer-specific mortality (PCSM) and overall survival (OS) with Cox univariable and multivariable analyses. RESULTS: GC scores were obtained on 385 samples, of which 265 passed microarray quality control (69%) and had a median follow-up of 11 years (interquartile range, 9-13). In the pooled cohort, on univariable analysis, the GC was shown to be a prognostic factor for DM (per 0.1 unit; subdistribution hazard ratio [sHR], 1.29; 95% confidence interval [CI], 1.18-1.41; P < .001), PCSM (sHR, 1.28; 95% CI, 1.16-1.41; P < .001), and OS (hazard ratio [HR], 1.16; 95% CI, 1.08-1.22; P < .001). On multivariable analyses, the GC (per 0.1 unit) was independently associated with DM (sHR, 1.22; 95% CI, 1.09-1.36), PCSM (sHR, 1.23; 95% CI, 1.09-1.39), and OS (HR, 1.12; 95% CI, 1.05-1.20) after adjusting for age, Prostate Specific Antigen, Gleason score, cT stage, trial, and randomized treatment arm. GC had similar prognostic ability in patients receiving short-term or long-term androgen-deprivation therapy, but the absolute improvement in outcome varied by GC risk. CONCLUSIONS: This is the first validation of a gene expression biomarker on pretreatment prostate cancer biopsy samples from prospective randomized trials and demonstrates an independent association of GC score with DM, PCSM, and OS. High-risk prostate cancer is a heterogeneous disease state, and GC can improve risk stratification to help personalize shared decision making.

Artificial Intelligence Predictive Model for Hormone Therapy Use in Prostate Cancer.

Background: Androgen deprivation therapy (ADT) with radiotherapy can benefit patients with localized prostate cancer. However, ADT can negatively impact quality of life and there remain no validated predictive models to guide its use. Methods: Digital pathology image and clinical data from pre-treatment prostate tissue from 5,727 patients enrolled on five phase III randomized trials treated with radiotherapy +/- ADT were used to develop and validate an artificial intelligence (AI)-derived predictive model to assess ADT benefit with the primary endpoint of distant metastasis. After the model was locked, validation was performed on NRG/RTOG 9408 (n = 1,594) that randomized men to radiotherapy +/- 4 months of ADT. Fine-Gray regression and restricted mean survival times were used to assess the interaction between treatment and predictive model and within predictive model positive and negative subgroup treatment effects. Results: In the NRG/RTOG 9408 validation cohort (14.9 years of median follow-up), ADT significantly improved time to distant metastasis (subdistribution hazard ratio [sHR] = 0.64, 95%CI [0.45-0.90], p = 0.01). The predictive model-treatment interaction was significant (p-interaction = 0.01). In predictive model positive patients (n = 543, 34%), ADT significantly reduced the risk of distant metastasis compared to radiotherapy alone (sHR = 0.34, 95%CI [0.19-0.63], p < 0.001). There were no significant differences between treatment arms in the predictive model negative subgroup (n = 1,051, 66%; sHR = 0.92, 95%CI [0.59-1.43], p = 0.71). Conclusions: Our data, derived and validated from completed randomized phase III trials, show that an AI-based predictive model was able to identify prostate cancer patients, with predominately intermediate-risk disease, who are likely to benefit from short-term ADT.

Artificial Intelligence Predictive Model for Hormone Therapy Use in Prostate Cancer.

BACKGROUND: Androgen deprivation therapy (ADT) with radiotherapy can benefit patients with localized prostate cancer. However, ADT can negatively impact quality of life, and there remain no validated predictive models to guide its use. METHODS: We used digital pathology images from pretreatment prostate tissue and clinical data from 5727 patients enrolled in five phase 3 randomized trials, in which treatment was radiotherapy with or without ADT, as our data source to develop and validate an artificial intelligence (AI)­derived predictive patient-specific model that would determine which patients would develop the primary end point of distant metastasis. The model used baseline data to provide a binary output that a given patient will likely benefit from ADT or not. After the model was locked, validation was performed using data from NRG Oncology/Radiation Therapy Oncology Group (RTOG) 9408 (n=1594), a trial that randomly assigned men to radiotherapy plus or minus 4 months of ADT. Fine­Gray regression and restricted mean survival times were used to assess the interaction between treatment and the predictive model and within predictive model­positive, i.e., benefited from ADT, and ­negative subgroup treatment effects. RESULTS: Overall, in the NRG/RTOG 9408 validation cohort (14.9 years of median follow-up), ADT significantly improved time to distant metastasis. Of these enrolled patients, 543 (34%) were model positive, and ADT significantly reduced the risk of distant metastasis compared with radiotherapy alone. Of 1051 patients who were model negative, ADT did not provide benefit. CONCLUSIONS: Our AI-based predictive model was able to identify patients with a predominantly intermediate risk for prostate cancer likely to benefit from short-term ADT. (Supported by a grant [U10CA180822] from NRG Oncology Statistical and Data Management Center, a grant [UG1CA189867] from NCI Community Oncology Research Program, a grant [U10CA180868] from NRG Oncology Operations, and a grant [U24CA196067] from NRG Specimen Bank from the National Cancer Institute and by Artera, Inc. ClinicalTrials.gov numbers NCT00767286, NCT00002597, NCT00769548, NCT00005044, and NCT00033631.)

PRECISION.array: An R Package for Benchmarking microRNA Array Data Normalization in the Context of Sample Classification.

We present a new R package PRECISION.array for assessing the performance of data normalization methods in connection with methods for sample classification. It includes two microRNA microarray datasets for the same set of tumor samples: a re-sampling-based algorithm for simulating additional paired datasets under various designs of sample-to-array assignment and levels of signal-to-noise ratios and a collection of numerical and graphical tools for method performance assessment. The package allows users to specify their own methods for normalization and classification, in addition to implementing three methods for training data normalization, seven methods for test data normalization, seven methods for classifier training, and two methods for classifier validation. It enables an objective and systemic evaluation of the operating characteristics of normalization and classification methods in microRNA microarrays. To our knowledge, this is the first such tool available. The R package can be downloaded freely at https://github.com/LXQin/PRECISION.array.

Making External Validation Valid for Molecular Classifier Development.

PURPOSE: Accurate assessment of a molecular classifier that guides patient care is of paramount importance in precision oncology. Recent years have seen an increasing use of external validation for such assessment. However, little is known about how it is affected by ubiquitous unwanted variations in test data because of disparate experimental handling and by the use of data normalization for alleviating such variations. METHODS: In this paper, we studied these issues using two microarray data sets for the same set of tumor samples and additional data simulated by resampling under various levels of signal-to-noise ratio and different designs for array-to-sample allocation. RESULTS: We showed that (1) unwanted variations can lead to biased classifier assessment and (2) data normalization mitigates the bias to varying extents depending on the specific method used. In particular, frozen normalization methods for test data outperform their conventional forms in terms of both reducing the bias in accuracy estimation and increasing robustness to handling effects. We make available our benchmarking tool as an R package on GitHub for performing such evaluation on additional methods for normalization and classification. CONCLUSION: Our findings thus highlight the importance of proper test-data normalization for valid assessment by external validation and call for caution on the choice of normalization method for molecular classifier development.

Impact of Decipher on use of post-operative radiotherapy: Individual patient analysis of two prospective registries.

Objective: To assess the association between Genomic Classifier (GC)-risk group and post-radical prostatectomy treatment in clinical practice. Methods: Two prospective observational cohorts of men with prostate cancer (PCa) who underwent RP in two referral centers and had GC testing post-prostatectomy between 2013 and 2018 were included. The primary endpoint of the study was to assess the association between GC-risk group and time to secondary therapy. Univariable (UVA) and multivariable (MVA) Cox proportional hazards models were constructed to assess the association between GC-risk group and time to receipt of secondary therapy after RP, where secondary therapy is defined as receiving either RT or ADT after RP. Results: A total of 398 patients are included in the analysis. Patients with high-GC risk were more likely to receive any secondary therapy (OR: 6.84) compared to patients with low/intermediate-GC risk. The proportion of high-GC risk patients receiving RT at 2 years post-RP was 31.5%, compared to only 6.3% among the low/intermediate-GC risk patients. Conclusion: This study demonstrates that physicians in routine practice used GC to identify high risk patients who might benefit the most from secondary treatment. As such, GC score was independent predictor of receipt of secondary treatment.

Validation of a 22-Gene Genomic Classifier in Patients With Recurrent Prostate Cancer: An Ancillary Study of the NRG/RTOG 9601 Randomized Clinical Trial.

Importance: Decipher (Decipher Biosciences Inc) is a genomic classifier (GC) developed to estimate the risk of distant metastasis (DM) after radical prostatectomy (RP) in patients with prostate cancer. Objective: To validate the GC in the context of a randomized phase 3 trial. Design, Setting, and Participants: This ancillary study used RP specimens from the phase 3 placebo-controlled NRG/RTOG 9601 randomized clinical trial conducted from March 1998 to March 2003. The specimens were centrally reviewed, and RNA was extracted from the highest-grade tumor available in 2019 with a median follow-up of 13 years. Clinical-grade whole transcriptomes from samples passing quality control were assigned GC scores (scale, 0-1). A National Clinical Trials Network-approved prespecified statistical plan included the primary objective of validating the independent prognostic ability of GC for DM, with secondary end points of prostate cancer-specific mortality (PCSM) and overall survival (OS). Data were analyzed from September 2019 to December 2019. Intervention: Salvage radiotherapy (sRT) with or without 2 years of bicalutamide. Main Outcomes and Measures: The preplanned primary end point of this study was the independent association of the GC with the development of DM. Results: In this ancillary study of specimens from a phase 3 randomized clinical trial, GC scores were generated from 486 of 760 randomized patients with a median follow-up of 13 years; samples from a total of 352 men (median [interquartile range] age, 64.5 (60-70) years; 314 White [89.2%] participants) passed microarray quality control and comprised the final cohort for analysis. On multivariable analysis, the GC (continuous variable, per 0.1 unit) was independently associated with DM (hazard ratio [HR], 1.17; 95% CI, 1.05-1.32; P = .006), PCSM (HR, 1.39; 95% CI, 1.20-1.63; P < .001), and OS (HR, 1.17; 95% CI, 1.06-1.29; P = .002) after adjusting for age, race/ethnicity, Gleason score, T stage, margin status, entry prostate-specific antigen, and treatment arm. Although the original planned analysis was not powered to detect a treatment effect interaction by GC score, the estimated absolute effect of bicalutamide on 12-year OS was less when comparing patients with lower vs higher GC scores (2.4% vs 8.9%), which was further demonstrated in men receiving early sRT at a prostate-specific antigen level lower than 0.7 ng/mL (-7.8% vs 4.6%). Conclusions and Relevance: This ancillary validation study of the Decipher GC in a randomized trial cohort demonstrated association of the GC with DM, PCSM, and OS independent of standard clinicopathologic variables. These results suggest that not all men with biochemically recurrent prostate cancer after surgery benefit equally from the addition of hormone therapy to sRT. Trial Registration: ClinicalTrials.gov identifier: NCT00002874.

Validation of a neuroendocrine-like classifier confirms poor outcomes in patients with bladder cancer treated with cisplatin-based neoadjuvant chemotherapy.

PURPOSE: Neuroendocrine (NE)-like carcinoma is a newly recognized molecular subtype of conventional urothelial carcinoma of the bladder with transcriptomic profiles and clinical outcomes highly similar to histological NE carcinoma. The identification of NE-like tumors is challenging, as these tumors often appear histologically like urothelial carcinoma and can be missed by routine morphological criteria. We previously developed a single-sample classifier to identify NE-like tumors, which we aimed to validate in an independent cohort. MATERIALS AND METHODS: A single-sample genomic classifier was performed on transurethral specimens from a retrospective multicenter cohort of 234 patients who underwent cisplatin-based neoadjuvant chemotherapy and subsequent radical cystectomy. Outcomes were compared for NE-like vs. non-NE-like. RESULTS: We identified 10 patients with urothelial tumors of the NE-like subtype, all of which had robust gene expression of neuronal markers, but did not express markers associated with basal or luminal tumors. The cancer-specific mortality rates were significantly higher compared to non-NE-like tumors (P < 0.001), with 5 of the 10 patients dying within 12 months from surgery. CONCLUSIONS: The single-sample classifier was able to identify urothelial carcinomas with NE-like subtype. These NE-like tumors have demonstrated transcriptomic profiles and clinical behavior similar to histological NE tumors across multiple patient cohorts. We propose that NE-like tumors should be managed similarly to histological NE tumors, and that standard treatments for small cell lung cancer as well as novel strategies may be evaluated in these patients.

Decipher identifies men with otherwise clinically favorable-intermediate risk disease who may not be good candidates for active surveillance.

BACKGROUND: We aimed to validate Decipher to predict adverse pathology (AP) at radical prostatectomy (RP) in men with National Comprehensive Cancer Network (NCCN) favorable-intermediate risk (F-IR) prostate cancer (PCa), and to better select F-IR candidates for active surveillance (AS). METHODS: In all, 647 patients diagnosed with NCCN very low/low risk (VL/LR) or F-IR prostate cancer were identified from a multi-institutional PCa biopsy database; all underwent RP with complete postoperative clinicopathological information and Decipher genomic risk scores. The performance of all risk assessment tools was evaluated using logistic regression model for the endpoint of AP, defined as grade group 3-5, pT3b or higher, or lymph node invasion. RESULTS: The median age was 61 years (interquartile range 56-66) for 220 patients with NCCN F-IR disease, 53% classified as low-risk by Cancer of the Prostate Risk Assessment (CAPRA 0-2) and 47% as intermediate-risk (CAPRA 3-5). Decipher classified 79%, 13% and 8% of men as low-, intermediate- and high-risk with 13%, 10%, and 41% rate of AP, respectively. Decipher was an independent predictor of AP with an odds ratio of 1.34 per 0.1 unit increased (p value = 0.002) and remained significant when adjusting by CAPRA. Notably, F-IR with Decipher low or intermediate score did not associate with significantly higher odds of AP compared to VL/LR. CONCLUSIONS: NCCN risk groups, including F-IR, are highly heterogeneous and should be replaced with multivariable risk-stratification. In particular, incorporating Decipher may be useful for safely expanding the use of AS in this patient population.

Validation of the Decipher Test for predicting adverse pathology in candidates for prostate cancer active surveillance.

ABSTACT: BACKGROUND: Many men diagnosed with prostate cancer are active surveillance (AS) candidates. However, AS may be associated with increased risk of disease progression and metastasis due to delayed therapy. Genomic classifiers, e.g., Decipher, may allow better risk-stratify newly diagnosed prostate cancers for AS. METHODS: Decipher was initially assessed in a prospective cohort of prostatectomies to explore the correlation with clinically meaningful biologic characteristics and then assessed in diagnostic biopsies from a retrospective multicenter cohort of 266 men with National Comprehensive Cancer Network (NCCN) very low/low and favorable-intermediate risk prostate cancer. Decipher and Cancer of the Prostate Risk Assessment (CAPRA) were compared as predictors of adverse pathology (AP) for which there is universal agreement that patients with long life-expectancy are not suitable candidates for AS (primary pattern 4 or 5, advanced local stage [pT3b or greater] or lymph node involvement). RESULTS: Decipher from prostatectomies was significantly associated with adverse pathologic features (p-values < 0.001). Decipher from the 266 diagnostic biopsies (64.7% NCCN-very-low/low and 35.3% favorable-intermediate) was an independent predictor of AP (odds ratio 1.29 per 10% increase, 95% confidence interval [CI] 1.03-1.61, p-value 0.025) when adjusting for CAPRA. CAPRA area under curve (AUC) was 0.57, (95% CI 0.47-0.68). Adding Decipher to CAPRA increased the AUC to 0.65 (95% CI 0.58-0.70). NPV, which determines the degree of confidence in the absence of AP for patients, was 91% (95% CI 87-94%) and 96% (95% CI 90-99%) for Decipher thresholds of 0.45 and 0.2, respectively. Using a threshold of 0.2, Decipher was a significant predictor of AP when adjusting for CAPRA (p-value 0.016). CONCLUSION: Decipher can be applied to prostate biopsies from NCCN-very-low/low and favorable-intermediate risk patients to predict absence of adverse pathologic features. These patients are predicted to be good candidates for active surveillance.

Empirical evaluation of data normalization methods for molecular classification.

BACKGROUND: Data artifacts due to variations in experimental handling are ubiquitous in microarray studies, and they can lead to biased and irreproducible findings. A popular approach to correct for such artifacts is through post hoc data adjustment such as data normalization. Statistical methods for data normalization have been developed and evaluated primarily for the discovery of individual molecular biomarkers. Their performance has rarely been studied for the development of multi-marker molecular classifiers-an increasingly important application of microarrays in the era of personalized medicine. METHODS: In this study, we set out to evaluate the performance of three commonly used methods for data normalization in the context of molecular classification, using extensive simulations based on re-sampling from a unique pair of microRNA microarray datasets for the same set of samples. The data and code for our simulations are freely available as R packages at GitHub. RESULTS: In the presence of confounding handling effects, all three normalization methods tended to improve the accuracy of the classifier when evaluated in an independent test data. The level of improvement and the relative performance among the normalization methods depended on the relative level of molecular signal, the distributional pattern of handling effects (e.g., location shift vs scale change), and the statistical method used for building the classifier. In addition, cross-validation was associated with biased estimation of classification accuracy in the over-optimistic direction for all three normalization methods. CONCLUSION: Normalization may improve the accuracy of molecular classification for data with confounding handling effects; however, it cannot circumvent the over-optimistic findings associated with cross-validation for assessing classification accuracy.

A pair of datasets for microRNA expression profiling to examine the use of careful study design for assigning arrays to samples.

We set out to demonstrate the logistic feasibility of careful experimental design for microarray studies and its level of scientific benefits for improving the accuracy and reproducibility of data inference. Towards this end, we conducted a study of microRNA expression using endometrioid endometrial tumours (n=96) and serous ovarian tumours (n=96) that were primary, untreated, and collected from 2000 to 2012 at Memorial Sloan Kettering Cancer Center. The same set of tumour tissue samples were profiled twice using the Agilent microRNA microarrays: once under an ideal experimental condition with balanced array-to-sample allocation and uniform handling; a second time by mimicking typical practice, with arrays assigned in the order of sample collection and processed by two technicians in multiple batches. This paper provides a detailed description of the generation and validation of this unique dataset pair so that the research community can re-use it to investigate other statistical questions regarding microarray study design and data analysis, and to address biological questions on the relevance of microRNA expression in gynaecologic cancer.

Cautionary Note on Using Cross-Validation for Molecular Classification.

Purpose Reproducibility of scientific experimentation has become a major concern because of the perception that many published biomedical studies cannot be replicated. In this article, we draw attention to the connection between inflated overoptimistic findings and the use of cross-validation for error estimation in molecular classification studies. We show that, in the absence of careful design to prevent artifacts caused by systematic differences in the processing of specimens, established tools such as cross-validation can lead to a spurious estimate of the error rate in the overoptimistic direction, regardless of the use of data normalization as an effort to remove these artifacts. Methods We demonstrated this important yet overlooked complication of cross-validation using a unique pair of data sets on the same set of tumor samples. One data set was collected with uniform handling to prevent handling effects; the other was collected without uniform handling and exhibited handling effects. The paired data sets were used to estimate the biologic effects of the samples and the handling effects of the arrays in the latter data set, which were then used to simulate data using virtual rehybridization following various array-to-sample assignment schemes. Results Our study showed that (1) cross-validation tended to underestimate the error rate when the data possessed confounding handling effects; (2) depending on the relative amount of handling effects, normalization may further worsen the underestimation of the error rate; and (3) balanced assignment of arrays to comparison groups allowed cross-validation to provide an unbiased error estimate. Conclusion Our study demonstrates the benefits of balanced array assignment for reproducible molecular classification and calls for caution on the routine use of data normalization and cross-validation in such analysis.

Differential Expression Analysis for RNA-Seq: An Overview of Statistical Methods and Computational Software.

Deep sequencing has recently emerged as a powerful alternative to microarrays for the high-throughput profiling of gene expression. In order to account for the discrete nature of RNA sequencing data, new statistical methods and computational tools have been developed for the analysis of differential expression to identify genes that are relevant to a disease such as cancer. In this paper, it is thus timely to provide an overview of these analysis methods and tools. For readers with statistical background, we also review the parameter estimation algorithms and hypothesis testing strategies used in these methods.

24/7 Registered Nurse Staffing Coverage in Saskatchewan Nursing Homes and Acute Hospital Use.

RÉSUMÉ La législation, dans de nombreuses juridictions, nécessite les établissements des soins de longue durée (SLD) d'avoir une infirmière en service 24 heures par jour, 7 jours par semaine. Bien que la recherche considérable existe sur l'intensité SLD de la dotation en personnel infirmier, il n'existe pas de la recherche empirique relative à cette exigence. Notre étude rétrospectif d'observation a comparé des installations en Saskatchewan avec 24/7 RN couverture aux établissements offrant moins de couverture, complétées par divers modèles de dotation des postes de nuit. Les ratios de risque associés à moins de 24/7 couverture RN complété de la dotation infirmière autorisé de nuit, ajusté pour l'intensité de dotation en personnel infirmier et d'autres facteurs de confusion potentiels, étaient de 1,17, IC 95% [0,91, 1,50] et 1.00, IC à 95% [0,72, 1,39], et avec moins de couverture 24/7 RN complété avec soin par aides personnels de nuit, les ratios de risque étaient de 1,46, IC 95% [1,11, 1,91] et 1,11, IC 95% [0,78, 1,58], pour les patients hospitalisés et de visites aux services d'urgence, respectivement. Ces résultats suggèrent que l'utilisation des soins de courte durée peut être influencée négativement par l'absence de la couverture 24/7 RN.

Preprocessing Steps for Agilent MicroRNA Arrays: Does the Order Matter?

MOTIVATION/BACKGROUND: Previous publications on microarray preprocessing mostly focused on method development or comparison for an individual preprocessing step. Very few, if any, focused on recommending an effective ordering of the preprocessing steps, in particular, normalization in relationship to log transformation and probe set summarization. In this study, we aim to study how the relative ordering of the preprocessing steps influences differential expression analysis for Agilent microRNA array data. METHODS: A set of 192 untreated primary gynecologic tumor samples (96 endometrial tumors and 96 ovarian tumors) were collected at Memorial Sloan Kettering Cancer Center during the period of 2000-2012. From this same sample set, two datasets were generated: one dataset had no confounding array effects by experimental design and served as the benchmark, and another dataset exhibited array effects and served as the test data. We preprocessed our test dataset using different orderings between the following three steps: quantile normalization, log transformation, and median summarization. Differential expression analysis was performed on each preprocessed test dataset, and the results were compared against the results from the benchmark dataset. True positive rate, false positive rate, and false discovery rate were used to assess the effectiveness of the orderings. RESULTS: The ordering of log transformation, quantile normalization (on probe-level data), and median summarization slightly outperforms the other orderings. CONCLUSION: Our results ease the anxiety over the uncertain effect that the orderings could have on the analysis of Agilent microRNA array data.