Detection and characterization of lung cancer using cell-free DNA fragmentomes.

Non-invasive approaches for cell-free DNA (cfDNA) assessment provide an opportunity for cancer detection and intervention. Here, we use a machine learning model for detecting tumor-derived cfDNA through genome-wide analyses of cfDNA fragmentation in a prospective study of 365 individuals at risk for lung cancer. We validate the cancer detection model using an independent cohort of 385 non-cancer individuals and 46 lung cancer patients. Combining fragmentation features, clinical risk factors, and CEA levels, followed by CT imaging, detected 94% of patients with cancer across stages and subtypes, including 91% of stage I/II and 96% of stage III/IV, at 80% specificity. Genome-wide fragmentation profiles across ~13,000 ASCL1 transcription factor binding sites distinguished individuals with small cell lung cancer from those with non-small cell lung cancer with high accuracy (AUC = 0.98). A higher fragmentation score represented an independent prognostic indicator of survival. This approach provides a facile avenue for non-invasive detection of lung cancer.

Noninvasive stratification of nonalcoholic fatty liver disease by whole transcriptome cell-free mRNA characterization.

Hepatic fibrosis stage is the most important determinant of outcomes in patients with nonalcoholic fatty liver disease (NAFLD). There is an urgent need for noninvasive tests that can accurately stage fibrosis and determine efficacy of interventions. Here, we describe a novel cell-free (cf)-mRNA sequencing approach that can accurately and reproducibly profile low levels of circulating mRNAs and evaluate the feasibility of developing a cf-mRNA-based NAFLD fibrosis classifier. Using separate discovery and validation cohorts with biopsy-confirmed NAFLD (n = 176 and 59, respectively) and healthy subjects (n = 23), we performed serum cf-mRNA RNA-Seq profiling. Differential expression analysis identified 2,498 dysregulated genes between patients with NAFLD and healthy subjects and 134 fibrosis-associated genes in patients with NAFLD. Comparison between cf-mRNA and liver tissue transcripts revealed significant overlap of fibrosis-associated genes and pathways indicating that the circulating cf-mRNA transcriptome reflects molecular changes in the livers of patients with NAFLD. In particular, metabolic and immune pathways reflective of known underlying steatosis and inflammation were highly dysregulated in the cf-mRNA profile of patients with advanced fibrosis. Finally, we used an elastic net ordinal logistic model to develop a classifier that predicts clinically significant fibrosis (F2-F4). In an independent cohort, the cf-mRNA classifier was able to identify 50% of patients with at least 90% probability of clinically significant fibrosis. We demonstrate a novel and robust cf-mRNA-based RNA-Seq platform for noninvasive identification of diverse hepatic molecular disruptions and for fibrosis staging with promising potential for clinical trials and clinical practice.NEW & NOTEWORTHY This work is the first study, to our knowledge, to utilize circulating cell-free mRNA sequencing to develop an NAFLD diagnostic classifier.

High-intensity sequencing reveals the sources of plasma circulating cell-free DNA variants.

Accurate identification of tumor-derived somatic variants in plasma circulating cell-free DNA (cfDNA) requires understanding of the various biological compartments contributing to the cfDNA pool. We sought to define the technical feasibility of a high-intensity sequencing assay of cfDNA and matched white blood cell DNA covering a large genomic region (508 genes; 2 megabases; >60,000× raw depth) in a prospective study of 124 patients with metastatic cancer, with contemporaneous matched tumor tissue biopsies, and 47 controls without cancer. The assay displayed high sensitivity and specificity, allowing for de novo detection of tumor-derived mutations and inference of tumor mutational burden, microsatellite instability, mutational signatures and sources of somatic mutations identified in cfDNA. The vast majority of cfDNA mutations (81.6% in controls and 53.2% in patients with cancer) had features consistent with clonal hematopoiesis. This cfDNA sequencing approach revealed that clonal hematopoiesis constitutes a pervasive biological phenomenon, emphasizing the importance of matched cfDNA-white blood cell sequencing for accurate variant interpretation.

The 17-Gene Genomic Prostate Score Assay Predicts Outcome After Radical Prostatectomy Independent of PTEN Status.

OBJECTIVE: To compare the ability of loss of phosphatase and tensin homolog (PTEN) and Genomic prostate score assay (GPS) in predicting the biochemical-recurrence (BCR) and clinical-recurrence (CR) after radical prostatectomy (RP) for clinically localized prostate cancer (PCa). METHODS: Three hundred seventy seven patients with and without CR were retrospectively selected by stratified cohort sampling design from RP database. PTEN status (by immunohistochemistry [IHC] and fluorescence in situ hybridization [FISH]) and GPS results were determined for RP specimens. BCR was defined as Prostate Specific Antigen (PSA) ≥ 0.2 ng/mL or initiation of salvage therapy for a rising PSA. CR was defined as local recurrence and/or distant metastases. RESULTS: Baseline mean age, PSA, and GPS score for the cohort were 61.1 years, 8 ng/dL, and 32.8. PTEN loss was noted in 38% patients by FISH and 25% by IHC. The concordance between FISH and IHC for PTEN loss was 66% (Kappa coefficient 0.278; P < .001). On univariable analysis, loss of PTEN by FISH or IHC was associated with BCR and CR (P < .05). However, after adjusting for GPS results, PTEN loss was not a significant predictor for CR or BCR (P > .1). The GPS result remained strongly associated with CR and BCR after adjusting for PTEN status (P < .001). PTEN status and GPS results only weakly correlated. GPS was widely distributed regardless of PTEN status indicating the biological heterogeneity of PCa even in PTEN-deficient cases. CONCLUSION: GPS is a significant predictor of aggressive PCa, independent of PTEN status. After adjustment for GPS results, PTEN was not independently associated with recurrence for PCa.

Clinical outcomes in patients with node-negative breast cancer treated based on the recurrence score results: evidence from a large prospectively designed registry.

The 21-gene Recurrence Score® (RS) assay is a validated prognostic/predictive tool in ER + early-stage breast cancer. However, clinical outcome data from prospective studies in RS ≥ 11 patients are lacking, as are relevant real-life clinical practice data. In this retrospective analysis of a prospectively designed registry, we evaluated treatments/clinical outcomes in patients undergoing RS-testing through Clalit Health Services. The analysis included N0 ER + HER2-negative breast cancer patients who were RS-tested from 1/2006 through 12/2010. Medical records were reviewed to verify treatments/recurrences/survival. The cohort included 1801 patients (median follow-up, 6.2 years). Median age was 60 years, 50.4% were grade 2 and 81.1% had invasive ductal carcinoma; 48.9% had RS < 18, 40.7% RS 18-30, and 10.4% RS ≥ 31, with chemotherapy use of 1.4, 23.7, and 87.2%, respectively. The 5-year Kaplan-Meier estimates for distant recurrence were 0.8, 3.0, and 8.6%, for patients with RS < 18, RS 18-30 and RS ≥ 31, respectively; the corresponding 5-year Kaplan-Meier estimates for breast cancer death were 0.0, 0.9, and 6.2%. Chemotherapy-untreated patients with RS < 11 (n = 304) and 11-25 (n = 1037) (TAILORx categorization) had 5-year Kaplan-Meier estimates for distant recurrence risk/breast cancer death of 1.0%/0.0% and 1.3%/0.4%, respectively. Our results extend those of the prospective TAILORx trial: the 5-year Kaplan-Meier estimates for distant recurrence and breast cancer death rate for the RS < 18 patients were very low supporting the use of endocrine therapy alone. Furthermore, in chemotherapy-untreated patients with RS 11-25 (where TAILORx patients were randomized to chemoendocrine or endocrine therapy alone), 5-year distant recurrence rates were also very low, suggesting that chemotherapy would not have conferred clinically meaningful benefit.

Gene expression in normal-appearing tissue adjacent to prostate cancers are predictive of clinical outcome: evidence for a biologically meaningful field effect.

PURPOSE: We evaluated gene expression in histologically normal-appearing tissue (NT) adjacent to prostate tumor in radical prostatectomy specimens, assessing for biological significance based on prediction of clinical recurrence (cR - metastatic disease or local recurrence). RESULTS: A total of 410 evaluable patients had paired tumor and NT. Forty-six genes, representing diverse biological pathways (androgen signaling, stromal response, stress response, cellular organization, proliferation, cell adhesion, and chromatin remodeling) were associated with cR in NT (FDR < 20%), of which 39 concordantly predicted cR in tumor (FDR < 20%). Overall GPS and its stromal response and androgen-signaling gene group components also significantly predicted time to cR in NT (RM-corrected HR/20 units = 1.25; 95% CI: 1.01-1.56; P = 0.024). EXPERIMENTAL DESIGN: Expression of 732 genes was measured by quantitative reverse transcriptase polymerase chain reaction (RT-PCR) separately in tumor and adjacent NT specimens from 127 patients with and 374 without cR following radical prostatectomy for T1/T2 prostate cancer. A 17-gene expression signature (Genomic Prostate Score [GPS]), previously validated to predict aggressive prostate cancer when measured in tumor tissue, was also assessed using pre-specified genes and algorithms. Analysis used Cox proportional hazards models, Storey's false discovery rate (FDR) control, and regression to the mean (RM) correction. CONCLUSIONS: Gene expression profiles, including GPS, from NT adjacent to tumor can predict prostate cancer outcome. These findings suggest that there is a biologically significant field effect in primary prostate cancer that is a marker for aggressive disease.

Patient-specific Meta-analysis of 2 Clinical Validation Studies to Predict Pathologic Outcomes in Prostate Cancer Using the 17-Gene Genomic Prostate Score.

OBJECTIVE: To perform patient-specific meta-analysis (MA) of two independent clinical validation studies of a 17-gene biopsy-based genomic assay as a predictor of favorable pathology at radical prostatectomy. MATERIALS AND METHODS: Patient-specific MA was performed on data from 2 studies (732 patients) using the Genomic Prostate Score (GPS; scale 0-100) together with Cancer of the Prostate Risk Assessment (CAPRA) score or National Comprehensive Cancer Network (NCCN) risk group as predictors of the likelihood of favorable pathology (LFP). Risk profile curves associating GPS with LFP by CAPRA score and NCCN risk group were generated. Decision curves and receiver operating characteristic curves were calculated using patient-specific MA risk estimates. RESULTS: Patient-specific MA-generated risk profiles ensure more precise estimates of LFP with narrower confidence intervals than either study alone. GPS added significant predictive value to each clinical classifier. A model utilizing GPS and CAPRA provided the most risk discrimination. In decision-curve analysis, greater net benefit was shown when combining GPS with each clinical classifier compared with the classifier alone. The area under the receiver operating characteristic curve improved from 0.68 to 0.73 by adding GPS to CAPRA, and 0.64 to 0.70 by adding GPS to NCCN risk group. The proportion of patients with LFP >80% increased from 11% using NCCN risk group alone to 23% using GPS with NCCN. Using GPS with CAPRA identified the highest proportion-31%-of patients with LFP >80%. CONCLUSION: Patient-specific MA provides more precise risk estimates that reflect the complete body of evidence. GPS adds predictive value to 3 widely used clinical classifiers, and identifies a larger proportion of low-risk patients than identified by clinical risk group alone.

A 16-gene assay to predict recurrence after surgery in localised renal cell carcinoma: development and validation studies.

BACKGROUND: The likelihood of tumour recurrence after nephrectomy in localised clear cell renal cell carcinoma is well characterised by clinical and pathological parameters. However, these assessments can be improved and personalised by the addition of molecular characteristics of each patient's tumour. We aimed to develop and validate a prognostic multigene signature to improve prediction of recurrence risk in clear cell renal cell carcinoma. METHODS: In the development stage, we investigated the association between expression of 732 genes, measured by reverse-transcription PCR, and clinical outcome in 942 patients with stage I-III clear cell renal cell carcinoma who had undergone a nephrectomy at the Cleveland Clinic (OH, USA). 516 genes were associated with recurrence-free interval. 11 of these genes were selected by further statistical analyses, and were combined with five reference genes (ie, 16 genes in total), from which a recurrence score algorithm was developed. The recurrence score was then validated in an independent cohort of 626 patients from France with stage I-III clear cell renal cell carcinoma who had also undergone nephrectomy. The association between the recurrence score and the risk of recurrence and cancer-specific survival in the first 5 years after surgery was assessed using Cox proportional hazard regression, stratified by tumour stage (stage I vs stage II vs III). FINDINGS: In our primary univariate analysis, the continuous recurrence score (median 37, IQR 31-45) was significantly associated with recurrence-free interval (hazard ratio 3·91 [95% CI 2·63-5·79] for a 25-unit increase in score, p<0·0001). In multivariable analyses, the recurrence score was significantly associated with the risk of tumour recurrence (hazard ratio per 25-unit increase in the score 3·37 [95% CI 2·23-5·08], p<0·0001) after stratification by stage and adjustment for tumour size, grade, or Leibovich score. The recurrence score was able to identify a clinically significant number of both high-risk stage I and low-risk stage II-III patients. A heterogeneity study on separate samples showed little to no intratumoural variability among the 16 genes. INTERPRETATION: Our findings validate the recurrence score as a predictor of clinical outcome in patients with stage I-III clear cell renal cell carcinoma, providing a more accurate and individualised risk assessment beyond existing clinical and pathological parameters. FUNDING: Genomic Health Inc and Pfizer Inc.

Utility of the Oncotype DX® Prostate Cancer Assay in Clinical Practice for Treatment Selection in Men Newly Diagnosed with Prostate Cancer: A Retrospective Chart Review Analysis.

INTRODUCTION: The 17-gene Oncotype DX® prostate cancer assay (Genomic Health Inc., Redwood City, California) is a validated, biopsy based gene expression assay that reports the Genomic Prostate Score. Combined with clinical risk features, Genomic Prostate Score provides an individualized estimation of disease aggressiveness at diagnosis. With this retrospective chart review we assessed the impact of incorporating the Oncotype DX Genomic Prostate Score on treatment recommendations and decisions for men with newly diagnosed low risk prostate cancer in community urology practices. METHODS: A total of 24 urologists who ordered the Oncotype DX prostate cancer assay soon after launch (May 2013) were invited to participate in the study. Clinicopathological data, Genomic Prostate Score results and treatment related information were retrieved from medical records. Data also were collected for a pre-Genomic Prostate Score baseline group diagnosed from May 2012 to April 2013. Descriptive analyses were performed to evaluate the proportion of men for whom active surveillance was recommended and used before and after the availability of Genomic Prostate Score. RESULTS: Overall 15 physicians contributing 211 patients (Genomic Prostate Score group 124, baseline group 87) participated in the chart review. Patients in the Genomic Prostate Score and baseline groups had comparable risk based on traditional clinical pathological features, with 82% with NCCN® very low or low risk disease. With Genomic Prostate Score the relative increase in active surveillance recommended was 22% (baseline 50% and Genomic Prostate Score 61%, absolute increase of 11%) and the relative increase in use of active surveillance was 56% (baseline 43% and Genomic Prostate Score 67%, absolute increase of 24%). Treatment recommendations for active surveillance were directionally consistent with assay reported risk. CONCLUSIONS: Genomic Prostate Score testing was associated with greater physician recommendation of and use of active surveillance in community clinical practices.

A Biopsy-based 17-gene Genomic Prostate Score Predicts Recurrence After Radical Prostatectomy and Adverse Surgical Pathology in a Racially Diverse Population of Men with Clinically Low- and Intermediate-risk Prostate Cancer.

BACKGROUND: Biomarkers that are validated in independent cohorts are needed to improve risk assessment for prostate cancer (PCa). OBJECTIVE: A racially diverse cohort of men (20% African American [AA]) was used to evaluate the association of the clinically validated 17-gene Genomic Prostate Score (GPS) with recurrence after radical prostatectomy and adverse pathology (AP) at surgery. DESIGN, SETTING, AND PARTICIPANTS: Biopsies from 431 men treated for National Comprehensive Cancer Network (NCCN) very low-, low-, or intermediate-risk PCa between 1990 and 2011 at two US military medical centers were tested to validate the association between GPS and biochemical recurrence (BCR) and to confirm the association with AP. Metastatic recurrence (MR) was also evaluated. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Cox proportional hazards models were used for BCR and MR, and logistic regression was used for AP. Central pathology review was performed by one uropathologist. AP was defined as primary Gleason pattern 4 or any pattern 5 and/or pT3 disease. RESULTS AND LIMITATIONS: GPS results (scale: 0-100) were obtained in 402 cases (93%); 62 men (15%) experienced BCR, 5 developed metastases, and 163 had AP. Median follow-up was 5.2 yr. GPS predicted time to BCR in univariable analysis (hazard ratio per 20 GPS units [HR/20 units]: 2.9; p<0.001) and after adjusting for NCCN risk group (HR/20 units: 2.7; p<0.001). GPS also predicted time to metastases (HR/20 units: 3.8; p=0.032), although the event rate was low (n=5). GPS was strongly associated with AP (odds ratio per 20 GPS units: 3.3; p<0.001), adjusted for NCCN risk group. In AA and Caucasian men, the median GPS was 30.3 for both, the distributions of GPS results were similar, and GPS was similarly predictive of outcome. CONCLUSIONS: The association of GPS with near- and long-term clinical end points establishes the assay as a strong independent measure of PCa aggressiveness. Tumor aggressiveness, as measured by GPS, and outcomes were similar in AA and Caucasian men in this equal-access health care system. PATIENT SUMMARY: Predicting outcomes in men with newly diagnosed prostate cancer is challenging. This study demonstrates that a new molecular test, the Genomic Prostate Score, which can be performed on a patient's original prostate needle biopsy, can predict the aggressiveness of the cancer and help men make decisions regarding the need for immediate treatment of their cancer.

A 17-gene assay to predict prostate cancer aggressiveness in the context of Gleason grade heterogeneity, tumor multifocality, and biopsy undersampling.

BACKGROUND: Prostate tumor heterogeneity and biopsy undersampling pose challenges to accurate, individualized risk assessment for men with localized disease. OBJECTIVE: To identify and validate a biopsy-based gene expression signature that predicts clinical recurrence, prostate cancer (PCa) death, and adverse pathology. DESIGN, SETTING, AND PARTICIPANTS: Gene expression was quantified by reverse transcription-polymerase chain reaction for three studies-a discovery prostatectomy study (n=441), a biopsy study (n=167), and a prospectively designed, independent clinical validation study (n=395)-testing retrospectively collected needle biopsies from contemporary (1997-2011) patients with low to intermediate clinical risk who were candidates for active surveillance (AS). OUTCOME MEASURES AND STATISTICAL ANALYSIS: The main outcome measures defining aggressive PCa were clinical recurrence, PCa death, and adverse pathology at prostatectomy. Cox proportional hazards regression models were used to evaluate the association between gene expression and time to event end points. Results from the prostatectomy and biopsy studies were used to develop and lock a multigene-expression-based signature, called the Genomic Prostate Score (GPS); in the validation study, logistic regression was used to test the association between the GPS and pathologic stage and grade at prostatectomy. Decision-curve analysis and risk profiles were used together with clinical and pathologic characteristics to evaluate clinical utility. RESULTS AND LIMITATIONS: Of the 732 candidate genes analyzed, 288 (39%) were found to predict clinical recurrence despite heterogeneity and multifocality, and 198 (27%) were predictive of aggressive disease after adjustment for prostate-specific antigen, Gleason score, and clinical stage. Further analysis identified 17 genes representing multiple biological pathways that were combined into the GPS algorithm. In the validation study, GPS predicted high-grade (odds ratio [OR] per 20 GPS units: 2.3; 95% confidence interval [CI], 1.5-3.7; p<0.001) and high-stage (OR per 20 GPS units: 1.9; 95% CI, 1.3-3.0; p=0.003) at surgical pathology. GPS predicted high-grade and/or high-stage disease after controlling for established clinical factors (p<0.005) such as an OR of 2.1 (95% CI, 1.4-3.2) when adjusting for Cancer of the Prostate Risk Assessment score. A limitation of the validation study was the inclusion of men with low-volume intermediate-risk PCa (Gleason score 3+4), for whom some providers would not consider AS. CONCLUSIONS: Genes representing multiple biological pathways discriminate PCa aggressiveness in biopsy tissue despite tumor heterogeneity, multifocality, and limited sampling at time of biopsy. The biopsy-based 17-gene GPS improves prediction of the presence or absence of adverse pathology and may help men with PCa make more informed decisions between AS and immediate treatment. PATIENT SUMMARY: Prostate cancer (PCa) is often present in multiple locations within the prostate and has variable characteristics. We identified genes with expression associated with aggressive PCa to develop a biopsy-based, multigene signature, the Genomic Prostate Score (GPS). GPS was validated for its ability to predict men who have high-grade or high-stage PCa at diagnosis and may help men diagnosed with PCa decide between active surveillance and immediate definitive treatment.

Human epidermal growth factor receptor 2 assessment in a case-control study: comparison of fluorescence in situ hybridization and quantitative reverse transcription polymerase chain reaction performed by central laboratories.

PURPOSE: The optimal method to assess human epidermal growth factor receptor 2 (HER2) status remains highly controversial. Before reporting patient HER2 results, American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guidelines mandate that laboratories demonstrate ≥ 95% concordance to another approved laboratory or methodology. Here, we compare central laboratory HER2 assessed by fluorescence in situ hybridization (FISH) and quantitative reverse transcriptase polymerase chain reaction (RT-PCR) using Oncotype DX in lymph node-negative, chemotherapy-untreated patients from a large Kaiser Permanente case-control study. PATIENTS AND METHODS: Breast cancer specimens from the Kaiser-Genomic Health study were examined. Central FISH assessment of HER2 amplification and polysomy 17 was conducted by PhenoPath Laboratories (ratios > 2.2, 1.8 to 2.2, and < 1.8 define HER2 positive, HER2 equivocal, and HER2 negative, respectively). HER2 expression by RT-PCR was conducted using Oncotype DX by Genomic Health (normalized expression units ≥ 11.5, 10.7 to < 11.5, and < 10.7 define HER2 positive, HER2 equivocal, and HER2 negative, respectively). Concordance analyses followed ASCO/CAP guidelines. RESULTS: HER2 concordance by central FISH and central RT-PCR was 97% (95% CI, 96% to 99%). Twelve percent (67 of 568 patients) and 11% (60 of 568 patients) of patients were HER2 positive by RT-PCR and FISH, respectively. HER2-positive patients had increased odds of dying from breast cancer compared with HER2-negative patients. Polysomy 17 was demonstrated in 12.5% of all patients and 33% of FISH-positive patients. Nineteen of 20 FISH-positive patients with polysomy 17 were also RT-PCR HER2 positive. Although not statistically significantly different, HER2-positive/polysomy 17 patients tended to have the worst prognosis, followed by HER2-positive/eusomic, HER2-negative/polysomy 17, and HER2-negative/eusomic patients. CONCLUSION: There is a high degree of concordance between central FISH and quantitative RT-PCR using Oncotype DX for HER2 status, and the assay warrants additional study in a trastuzumab-treated population.

Estrogen- and progesterone-receptor status in ECOG 2197: comparison of immunohistochemistry by local and central laboratories and quantitative reverse transcription polymerase chain reaction by central laboratory.

PURPOSE: Central and local laboratory concordance for hormone receptor measurement is therapeutically important. This study compares estrogen receptor (ER) and progesterone receptor (PR) measured by local laboratory immunohistochemistry (IHC), central IHC, and central reverse-transcriptase polymerase chain reaction (RT-PCR) using a proprietary 21-gene assay. PATIENTS AND METHODS: A case-control sample of 776 breast cancer patients from Eastern Cooperative Oncology Group (ECOG) study E2197 was evaluated. Central IHC Allred score for ER and PR was obtained using tissue microarrays and 1D5 ER antibody and 636 PR antibody. Quantitative RT-PCR for ER and PR in whole sections was performed using the 21-gene assay. RESULTS: For ER, the concordance between local and central IHC was 90% (95% CI, 88% to 92%), between local IHC and central RT-PCR was 91% (95% CI, 89% to 93%), and between central IHC and central RT-PCR was 93% (95% CI, 91% to 95%). For PR, the concordance between local IHC and central IHC was 84% (95% CI, 82% to 87%), between local IHC and central RT-PCR was 88% (95% CI, 85% to 90%), and between central IHC and central RT-PCR was 90% (95% CI, 88% to 92%). Although concordance was high, IHC ER-negative cases that were RT-PCR positive were more common than IHC ER-positive cases that were RT-PCR negative. In ER-positive patients, ER expression by central IHC Allred score was marginally associated with recurrence (P = .091), and ER expression by central RT-PCR was significantly associated with recurrence (P = .014). However, recurrence score, which incorporates additional genes/pathways, was a highly significant predictor of recurrence (P < .0001). CONCLUSION: There is a high degree of concordance among local IHC, central IHC, and central RT-PCR by the proprietary gene assay for ER and PR status. Although ER expression is marginally associated with relapse in ER-positive patients treated with chemohormonal therapy, recurrence score is a highly significant predictor of recurrence.

An experiment on simplifying conjoint analysis designs for measuring preferences.

In conjoint analysis (CA) studies, choosing between scenarios with multiple health attributes may be demanding for respondents. This study examined whether simplifying the choice task in CA designs, by using a design with more overlap of attribute levels, provides advantages over standard minimal-overlap methods. Two experimental conditions, minimal and increased-overlap discrete choice CA designs, were administered to 353 respondents as part of a larger HIV testing preference survey. In the minimal-overlap survey, all six attribute levels were allowed to vary. In the increased-overlap survey, an average of two attribute levels were the same between each set of scenarios. We hypothesized that the increased-overlap design would reduce cognitive burden, while minimally impacting statistical efficiency. We did not find any significant improvement in consistency, willingness to trade, perceived difficulty, fatigue, or efficiency, although several results were in the expected direction. However, evidence suggested that there were differences in stated preferences. The results increase our understanding of how respondents answer CA questions and how to improve future surveys.

Measuring what people value: a comparison of "attitude" and "preference" surveys.

OBJECTIVE: To compare and contrast methods and findings from two approaches to valuation used in the same survey: measurement of "attitudes" using simple rankings and ratings versus measurement of "preferences" using conjoint analysis. Conjoint analysis, a stated preference method, involves comparing scenarios composed of attribute descriptions by ranking, rating, or choosing scenarios. We explore possible explanations for our findings using focus groups conducted after the quantitative survey. METHODS: A self-administered survey, measuring attitudes and preferences for HIV tests, was conducted at HIV testing sites in San Francisco in 1999-2000 (n = 365, response rate = 96 percent). Attitudes were measured and analyzed using standard approaches. Conjoint analysis scenarios were developed using a fractional factorial design and results analyzed using random effects probit models. We examined how the results using the two approaches were both similar and different. RESULTS: We found that "attitudes" and "preferences" were generally consistent, but there were some important differences. Although rankings based on the attitude and conjoint analysis surveys were similar, closer examination revealed important differences in how respondents valued price and attributes with "halo" effects, variation in how attribute levels were valued, and apparent differences in decision-making processes. CONCLUSIONS: To our knowledge, this is the first study to compare attitude surveys and conjoint analysis surveys and to explore the meaning of the results using post-hoc focus groups. Although the overall findings for attitudes and preferences were similar, the two approaches resulted in some different conclusions. Health researchers should consider the advantages and limitations of both methods when determining how to measure what people value.

Measuring preferences for health care interventions using conjoint analysis: an application to HIV testing.

OBJECTIVE: To examine preferences for HIV test methods using conjoint analysis, a method used to measure economic preferences (utilities). DATA SOURCES: Self-administered surveys at four publicly funded HIV testing locations in San Francisco, California, between November 1999 and February 2000 (n = 365, 96 percent response rate). STUDY DESIGN: We defined six important attributes of HIV tests and their levels (location, price, ease of collection, timeliness/accuracy, privacy/anonymity, and counseling). A fractional factorial design was used to develop scenarios that consisted of combinations of attribute levels. Respondents were asked 11 questions about whether they would choose "Test A or B" based on these scenarios. DATA ANALYSIS: We used random effects probit models to estimate utilities for testing attributes. Since price was included as an attribute, we were able to estimate willingness to pay, which provides a standardized measure for use in economic evaluations. We used extensive analyses to examine the reliability and validity of the results, including analyses of: (1) preference consistency, (2) willingness to trade among attributes, and (3) consistency with theoretical predictions. PRINCIPAL FINDINGS: Respondents most preferred tests that were accurate/timely and private/anonymous, whereas they had relatively lower preferences for in-person counseling. Respondents were willing to pay an additional $35 for immediate, highly accurate results; however, they had a strong disutility for receiving immediate but less accurate results. By using conjoint analysis to analyze new combinations of attributes, we found that respondents would most prefer instant, highly accurate home tests, even though they are not currently available in the U.S. Respondents were willing to pay $39 for a highly accurate, instant home test. CONCLUSIONS: The method of conjoint analysis enabled us to estimate utilities for specific attributes of HIV tests as well as the overall utility obtained from various HIV tests, including tests that are under consideration but not yet available. Conjoint analysis offers an approach that can be useful for measuring and understanding the value of other health care goods, services, and interventions.