Search

PURPOSE To assess the association between the Oncotype DX Genomic Prostate Score (GPS) result and long-term oncological outcomes following radical prostatectomy (RP). METHODS We evaluated the association of the GPS result assayed from the index lesion from RP tissue with the risk of distant metastases (DM) and prostate cancer–specific mortality (PCSM) over the 20 years following RP in a stratified cohort sample of 428 patients from 2,641 treated between 1987 and 2004. Cox regression of cause-specific hazards was used to estimate the absolute risk of both end points, with death from other causes treated as a competing risk. A correction for regression to the mean (RM) was applied since the GPS test was developed using this cohort. Exploratory analysis using presurgical parameters and the GPS test as prognostic variables was performed to assess the additional value of the GPS test on 20-year risk of DM and PCSM. Model discrimination was measured using the area under the receiver operating characteristic curve. RESULTS The GPS test appears to be independently associated with both 20-year risk of DM and PCSM with a low false discovery rate. Per 20-unit increase in GPS, multivariable analysis with RM correction estimated hazard ratios of 2.24 (95% CI, 1.49 to 3.53) and 2.30 (95% CI, 1.45 to 4.36) for DM and PCSM, respectively. Accuracy of models including clinical risk factors alone appeared to improve when including the GPS test in assessing risk of both end points. CONCLUSION The results suggest that the GPS test provides information on the risk for the meaningful long-term outcomes of DM and PCSM.

Background: The 21-gene recurrence score (RS) provides prognostic information for distant recurrence (DR) and predictive information for chemotherapy (CT) benefit in early breast cancer, whereas clinicopathologic features provide only prognostic information. We used patient-specific meta-analysis (PSMA) methods (Crager et al, J Appl Stat 2014 [PMID: 26664111]; Tang et al. J Clin Oncol 2011 [PMID: 22010013]) to develop an online tool that assesses individualized estimates of DR risk and absolute CT benefit. We also externally validated this new tool for estimating DR risk in an independent cohort.Methods: The PSMA included 10,004 women with hormone-receptor positive, HER2-negative, node-negative breast cancer from the NSABP B-14 (n=577) trial receiving endocrine therapy (ET) alone and TAILORx trial receiving ET alone (n=4854) or CT plus ET (n=4573). The baseline risk used TAILORx event rates with ET alone. A patient-specific estimator of absolute CT benefit was computed by combining PSMA of the individualized relative CT effect from the TAILORx and NSABP B-20 (n=550 HER2-negative) trials and risk estimates with ET alone. External validation of DR risk estimation was performed in an independent cohort of 1098 women enrolled in the Clalit Health Service registry who had a 21-gene RS performed, of whom 222 (20%) received CT in addition to ET as per the recommendation of their treating physician guided by the RS. Results: The new tool is more informative for DR risk than RS or clinicopathologic factors alone (p Citation Format: Joseph A Sparano, Michael R Crager, Gong Tang, Robert J Gray, Salomon M Stemmer, Steve Shak. Development and validation of a tool integrating the 21-gene recurrence score and clinicopathlogic features to individualize prognosis for distant recurrence and prediction of absolute chemotherapy benefit in early breast cancer [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr GS4-10.

PURPOSE The 17-gene Onco type DX Genomic Prostate Score (GPS) test predicts adverse pathology (AP) in patients with low-risk prostate cancer treated with immediate surgery. We evaluated the GPS test as a predictor of outcomes in a multicenter active surveillance cohort. MATERIALS AND METHODS Diagnostic biopsy tissue was obtained from men enrolled at 8 sites in the Canary Prostate Active Surveillance Study. The primary endpoint was AP (Gleason Grade Group [GG] ≥ 3, ≥ pT3a) in men who underwent radical prostatectomy (RP) after initial surveillance. Multivariable regression models for interval-censored data were used to evaluate the association between AP and GPS. Inverse probability of censoring weighting was applied to adjust for informative censoring. Predictiveness curves were used to evaluate how models stratified risk of AP. Association between GPS and time to upgrade on surveillance biopsy was evaluated using Cox proportional hazards models. RESULTS GPS results were obtained for 432 men (median follow-up, 4.6 years); 101 underwent RP after a median 2.1 years of surveillance, and 52 had AP. A total of 167 men (39%) upgraded at a subsequent biopsy. GPS was significantly associated with AP when adjusted for diagnostic GG (hazards ratio [HR]/5 GPS units, 1.18; 95% CI, 1.04 to 1.44; P = .030), but not when also adjusted for prostate-specific antigen density (PSAD; HR, 1.85; 95% CI, 0.99 to 4.19; P = .066). Models containing PSAD and GG, or PSAD, GG, and GPS may stratify risk better than a model with GPS and GG. No association was observed between GPS and subsequent biopsy upgrade ( P = .48). CONCLUSION In our study, the independent association of GPS with AP after initial active surveillance was not statistically significant, and there was no association with upgrading in surveillance biopsy. Adding GPS to a model containing PSAD and diagnostic GG did not significantly improve stratification of risk for AP over the clinical variables alone.

PURPOSE To assess the association of adverse pathology (AP), defined as high-grade (≥ Gleason Grade Group 3) and/or non-organ confined disease, with long-term oncologic outcomes after radical prostatectomy (RP). MATERIALS AND METHODS Using a stratified cohort sampling design, we evaluated the association of AP with the risk of distant metastasis (DM) and prostate cancer-specific mortality (PCSM) up to 20 years after RP in 428 patients treated between 1987 to 2004. Cox regression of cause-specific hazards was used to estimate the absolute risk of both endpoints, with death from other causes treated as a competing risk. Additionally, subgroup analysis in patients with low and/or intermediate-risk disease, who are potentially eligible for active surveillance (AS), was performed. RESULTS Within the cohort sample, 53% of men exhibited AP at time of RP, with median follow up of 15.5 years (IQR 14.6-16.6 years) thereafter. Adverse pathology was highly associated with DM and PCSM in the overall cohort (HR 12.30, 95% confidence interval [CI] 5.30-28.55, and HR 10.03, 95% CI 3.42-29.47, respectively, both P < 0.001). Adverse pathology was also highly associated with DM and PCSM in the low/intermediate-risk subgroup (HR 10.48, 95% CI 4.18-26.28, and 8.60, 95% CI 2.40-30.48, respectively, both P < 0.001). CONCLUSIONS Adverse pathology at the time of RP is highly associated with future development of DM and PCSM. Accurate prediction of AP may thus be useful for individualizing risk-based surveillance and treatment strategies.

The NSABP B-20 prospective-retrospective study of the 21-gene Oncotype DX Breast Cancer Recurrence Score® test predicted benefit from addition of chemotherapy to tamoxifen in node-negative, estrogen-receptor positive breast cancer when recurrence score (RS) was ≥31. HER2 is a component of the RS algorithm with a positive coefficient and contributes to higher RS values. Accrual to B-20 occurred prior to routine testing for HER2, so questions have arisen regarding assay performance if HER2-positive patients were identified and excluded. We report an exploratory reanalysis of the B-20, 21-gene study following exclusion of such patients. Patients were considered HER2 positive if quantitative RT-PCR for HER2 was ≥11.5 units, and excluded from re-analyses performed using the original cutoffs: 25. The endpoint remained distant recurrence-free interval (DRFI) as in the original study. Distribution was estimated via the Kaplan–Meier method and compared via log-rank test. Multivariate Cox proportional hazards models estimated chemotherapy benefit in each group. In the RS 25 cohort (HR = 0.28; 95% CI: 0.12–0.64). No benefit from chemotherapy was evident in the other RS groups. Following exclusion of HER2-positive patients based on RT-PCR expression, substantial benefit of chemotherapy remained for RS ≥ 31 as originally employed, and with RS > 25 employed in TAILORx., Genetics: multigene panel predicts chemo benefit regardless of HER2 status A commonly used genetic test helps to predict whether patients with estrogen-receptor positive, node-negative breast cancer stand to gain from chemotherapy. Charles Geyer from NRG Oncology/NSABP and Virginia Commonwealth University and colleagues reanalyzed data from a decades-old study that helped establish a 21-gene panel as a tool for determining whether to add chemotherapy to endocrine therapy for women with estrogen-receptor positive, node-negative breast cancer. The researchers excluded patients with HER2-positive disease because HER2 gene expression is part of the diagnostic test and questions have persisted regarding the performance of the assay, if HER2-positive patients were excluded. This new analysis demonstrated that although chemotherapy did not extend disease-free survival overall among the cohort of patients with HER2-negative disease, it still proved beneficial for those with high recurrence scores on the genetic test. The findings thus validate the clinical utility of the test for patients with estrogen-receptor positive, HER2-negative, node negative breast cancers.

Purpose Better tools are needed to estimate local recurrence (LR) risk after breast-conserving surgery (BCS) for DCIS. The DCIS score (DS) was validated as a predictor of LR in E5194 and Ontario DCIS cohort (ODC) after BCS. We combined data from E5194 and ODC adjusting for clinicopathological factors to provide refined estimates of the 10-year risk of LR after treatment by BCS alone. Methods Data from E5194 and ODC were combined. Patients with positive margins or multifocality were excluded. Identical Cox regression models were fit for each study. Patient-specific meta-analysis was used to calculate precision-weighted estimates of 10-year LR risk by DS, age, tumor size and year of diagnosis. Results The combined cohort includes 773 patients. The DS and age at diagnosis, tumor size and year of diagnosis provided independent prognostic information on the 10-year LR risk (p ≤ 0.009). Hazard ratios from E5194 and ODC cohorts were similar for the DS (2.48, 1.95 per 50 units), tumor size ≤ 1 versus > 1–2.5 cm (1.45, 1.47), age ≥ 50 versus 15%) 10-year LR risk after BCS alone compared to utilization of DS alone or clinicopathological factors alone. Conclusions The combined analysis provides refined estimates of 10-year LR risk after BCS for DCIS. Adding information on tumor size and age at diagnosis to the DS adjusting for year of diagnosis provides improved LR risk estimates to guide treatment decision making. Electronic supplementary material The online version of this article (10.1007/s10549-018-4693-2) contains supplementary material, which is available to authorized users.

The absolute standardized hazard ratio (ASHR) is a scale-invariant scalar measure of the strength of association of a vector of covariates with the risk of an event. It is derived from proportional hazards regression. The ASHR is useful for making comparisons among different sets of covariates. Extensions of the ASHR concept and practical considerations regarding its computation are discussed. These include a new method to conduct preliminary checks for collinearity among covariates, a partial ASHR to evaluate the association with event risk of some of the covariates conditioning on others, and the ASHR for interactions. To put the ASHR in context, its relationship to measures of explained variation and other measures of separation of risk is discussed. A new measure of the contribution of each covariate to the risk score variance is proposed. This measure, which is derived from the ASHR calculations, is interpretable as variable importance within the context of the multivariable model.

3599 Background: The 12-gene Oncotype DX Colon Recurrence Score assay is a clinically validated genomic assay that evaluates recurrence risks in stage II and stage III colon cancer patients independent of clinical-pathologic features. Improved colon cancer care has reduced recurrence rates since the late 1990’s. Methods: Pre-specified patient-specific meta-analysis methods were used to estimate 1-, 3- and 5-year recurrence risk combining the 12-gene colon recurrence score (RS) validation studies CALGB 9581, NSABP C-07 and SUNRISE. Cox models had effects for RS result, number of nodes examined (

219 Background: Adverse pathology (AP) at radical prostatectomy (RP) is often used as a proxy for long-term prostate cancer outcomes. The goal of this study was to assess the association of AP at RP, defined as high-grade (> Grade Group 3) and/or non-organ confined disease (pT3), with distant metastasis and prostate cancer death. Methods: A stratified cohort sample of 428 patients was used to evaluate the association of adverse pathology with the risk of distant metastases and prostate cancer-specific mortality over 20 years after prostatectomy in 2641 patients treated between 1987-2004. Cox regression of cause-specific hazards was used to estimate the absolute risk of both endpoints, with death from other causes treated as a competing risk. Subgroup analysis in patients with low/intermediate risk disease potentially eligible for active surveillance was performed. Results: Among the 428 patients, 343 had AUA Low or Intermediate risk disease and 85 had High risk disease. Median follow-up time was 15.5 years (IQR 14.6–16.6 years). Using the cohort sampling weights for estimation, at RP 29.8% of patients had high-grade disease, 42.3 % had non-organ confined disease, 19.3% had both, and thus 52.8% had AP. Adverse pathology was highly associated with metastasis and prostate cancer mortality in the overall cohort (HR 12.30, 95% CI 5.30-28.55, and 10.03, 95% CI 3.42-29.47, respectively, both p

Background: The Clinical Treatment Score at 5 years (CTS5) is a prognostic tool using clinicopathological parameters (nodal status, tumor size, grade, and age) to estimate distant recurrence (DR) risk after 5 years of endocrine treatment for postmenopausal women with estrogen receptor (ER) positive breast cancer. It was developed and validated in the ATAC and BIG 1-98 trials. Here, we evaluate whether the CTS5 is prognostic in women who were distant recurrence (DR) free 5 years after adjuvant treatment who participated in the TAILORx study. Methods: 10273 women with ER-positive, HER2-negative, and node-negative disease were entered into the TAILORx study and assigned or randomized to four treatment arms according to their Oncotype Recurrence Score (RS) (endocrine therapy alone (ET) or chemo-endocrine therapy (CET)). Primary endpoint of this analysis was DR and the primary objective of this analysis was to evaluate the CTS5 in the TAILORx study, stratified by chemotherapy use. Secondary objectives included validation of the CTS5 in two specific age groups (age≤50 years vs. age>50 years) and in the four separate treatment arms. Cox regression models were used to determine the prognostic value of CTS5. Kaplan-Meier curves were used to estimate conditional 9-year DR risks (%). Hazard ratios are reported for an increase of 1-standard deviation of CTS5 in the overall analysis population. Results: 7353 (71.6%) women were DR free 5 years after diagnosis with follow-up beyond 5 years, had information for all clinical parameters, and were included in this analysis. Median follow-up after 5 years was 2.9 years (1.8-3.9). Overall, mean age was 55.7 years (SD 9.0), mean tumor size 17.1mm (SD 7.8), and the vast majority had low or moderately differentiated disease (84.7%). CTS5 was significantly higher for women in arm D (RS 26-100, CET) compared to the other three arms (mean CTS5 3.09 vs. 2.82, P Conclusions: Low risks of late DR of 3.1% (2.4-4.0) for RS scores 0-25 (ET) and 3.8% (2.9-4.8) for RS scores 11-100 (CET) were observed in this ER-positive, HER2-negative and node negative TAILORx cohort. We confirm the prognostic ability of the CTS5 for late DR in the TAILORx cohort, specifically for patients older than 50 years and/or those who were deemed intermediate or high risk by Oncotype Dx RS (11-100). Our results show that the CTS5 is less prognostic for late DR in women aged 50 years or younger who received 5 years of endocrine therapy only. Further evaluation of the CTS5 in premenopausal cohorts is needed before it can be applied to younger patients. Table: Hazard Ratios (95% CI) for prognostic value of CTS5 overall and according to sub-groups.HR (95% CI )P-valueAll (N=7353)1.57 (1.36-1.82) 50 years (N=5094)1.78 (1.48-2.14) Citation Format: Ivana Sestak, Michael Crager, Jack Cuzick, Mitchell Dowsett, Steven Shak, Gong Tang, Robert Gray, Joseph Sparano. Validation of the clinical treatment score post 5 years (CTS5) in women with hormone receptor positive, HER2-negative, node-negative disease from the TAILORx study [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr GS4-03.

262 Background: The 17-gene Genomic Prostate Score (GPS) test (scale 0-100) predicts adverse surgical pathology (AP) and recurrence in newly diagnosed low- and intermediate-risk PCa. Studies of the predictive value of the GPS test in men initially managed with AS are limited. Methods: Diagnostic biopsy tissue was obtained from 634 men enrolled at 8 sites in PASS. Time to AP (Gleason Grade Group (GG) ≥3, ≥pT3a, or N1) in men who underwent radical prostatectomy (RP) was the primary endpoint. All diagnostic biopsies and RP specimens were centrally reviewed. Multivariate regression models for interval censored data were used to evaluate the association between time to AP and GPS. Inverse probability of censoring weighting was applied to adjust for informative censoring. Association between GPS and time to Gleason score upgrade on surveillance biopsy was also evaluated using a Cox Proportional Hazards model. Results: GPS results were obtained for 432 men (median follow-up 4.6 [IQR: 2.9-6.2] years); 374 and 58 with GG 1 or 2 cancer, respectively; median PSA density (PSAD) was 0.11 [IQR: 0.08-0.15]; 101 men underwent RP with central pathology after a median of 2.1 [IQR: 1.3-4.3] years surveillance, and 52 (52%) men undergoing RP had AP. 167 men upgraded at a subsequent biopsy. No clinico-pathologic covariates were significantly associated with AP other than PSAD. GPS was significantly associated with time to AP (hazards ratio [HR]/20 GPS units: 1.96 [95% CI = 1.17-4.28]; p = 0.030), when adjusted for diagnostic GG, or for dichotomous PSAD ( < vs ≥ 0.15; HR: 1.83, 95% CI = 1.04-3.62; p = 0.046). GPS was not significantly associated with AP (HR: 1.61, 95% CI = 0.87-2.98; p = 0.12) when adjusted for continuous PSAD. No association, either univariable or multivariable, was observed between GPS and subsequent biopsy upgrade. Conclusions: In a cohort of men on AS, GPS was associated with time to AP when adjusted for diagnostic GG or dichotomous PSAD. GPS was not associated with surveillance biopsy GG upgrading or AP at surgery after adjustment for continuous PSAD, although a trend was seen for AP, suggesting an association may be seen in a larger study.

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.

The standardized hazard ratio for univariate proportional hazards regression is generalized as a scalar to multivariate proportional hazards regression. Estimators of the standardized log hazard ratio are developed, with corrections for bias and for regression to the mean in high-dimensional analyses. Tests of point and interval null hypotheses and confidence intervals are constructed. Cohort sampling study designs, commonly used in prospective–retrospective clinical genomic studies, are accommodated.

Background: RNA biomarkers discovered by RT-PCR-based gene expression profiling of archival formalin-fixed paraffin-embedded (FFPE) tissue are the basis for very precise and sensitive clinical diagnostic tests, such as the 21 gene Oncotype DX® breast cancer assay. Both inherent limits of technical scalability and the small amounts of patient FFPE RNA available place practical constraints on the number of transcripts that can be interrogated by RT-PCR. We developed new methods for RNA profiling through massively parallel “next generation” sequencing (RNA-Seq) of archival FFPE specimens. We report here the technical performance of this methodology and compare the results to RT-PCR results obtained in one of the studies that were carried out to develop the 21 gene assay. Methods: RNA was extracted in 2002 from 136 invasive breast tumors that were formalin-fixed and paraffin-embedded between 1990 and 1997. RNA-Seq was carried out using minor modifications to methods we have reported previously (Sinicropi et al., Advances in Genome Biology and Technology Conference, p. 170, 2010 and p. 198, 2011). Briefly, 0.1 mg of total RNA was selectively depleted of ribosomal RNA and sequencing libraries were prepared using a modification of the ScriptSeq™ kit from Epicentre. The libraries were sequenced on an Illumina HiSeq 2000 instrument with multiplexing of two libraries per lane for 50 cycles in one direction. The resulting FASTQ sequences were mapped to version hg19 of the human genome using the Illumina CASAVA pipeline. The total number of sequences (reads) that uniquely mapped to all exons of each RefSeq entry was used for quantification of expression levels. Results: On average, there were 43 million reads per sample (range 31 - 58 million; SD=4.6 million) of which 69% uniquely mapped to the human genome. Ribosomal RNA was effectively removed and accounted for Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr PD03-09.

Background: The 21-gene Oncotype DX® Recurrence Score® (RS) is widely used for assessment of recurrence risk and prediction of chemotherapy benefit in patients with early stage ER-positive breast cancer. Through a meta-analysis on the NSABP B-14 study and the ATAC study, the Recurrence Score-Pathology-Clinical (RSPC) risk assessment has been developed as an integration of RS and clinico-pathologic factors (including tumor size, tumor grade and age) for assessment of recurrence risk in early stage ER-positive, node-negative breast cancer. The value of RSPC in prediction of chemotherapy benefit needs to be evaluated. Methods: In the NSABP trial B-20, patients were node-negative and ERpositive by the ligand binding assay, and were randomized to tamoxifen with or without adjuvant chemotherapy (MF or CMF). B-20 patients who had successful Oncotype DX RS assay and ER score ≥6.5 were included in this study. Cox proportional hazards models with interaction terms were used to evaluate RS and RSPC in prediction of chemotherapy benefit in reducing risk of distant recurrence. Results: The analysis included 625 patients among whom 60 distant recurrence events occurred. RS showed a significant interaction with chemotherapy treatment (p=0.037), as found previously, with a standardized hazard ratio (HR) of 0.836. The interaction of RSPC with treatment was not significant (p=0.10), although the trend was in the same direction as RS (standardized HR 0.833). Conclusions: RS used alone remains the best predictor of chemotherapy benefit in ER-positive, node-negative breast cancer. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr S4-9.

This analysis examined the safety and efficacy of extended-release ranolazine among patients aged 70 years or older (n=363) compared with patients younger than 70 years (n=1024) enrolled in 2 large multinational prospective clinical trials. The primary end points were exercise capacity and number of weekly angina episodes. Beneficial effects of ranolazine, relative to placebo, were generally similar for each of these outcomes among older and younger participants. For example, at a ranolazine dose of 1000 mg bid, mean exercise duration increased by 19.8+/-13.1 seconds (mean +/- SE) relative to placebo in patients younger than 70 years and by 32.4+/-19.7 seconds relative to placebo in patients 70 years or older. Adverse effects were more common in older than in younger patients, but the incidence of serious adverse effects attributable to ranolazine did not differ significantly between age groups. Outcomes were also similar at dosages of either 750 mg or 1000 mg bid. In conclusion, pooled data from 2 large randomized trials indicate that the efficacy of ranolazine is similar in older and younger patients but that adverse effects are more common in the elderly.

5074 Background: Over-treatment of localized prostate cancer (PC) can result from over-estimation of a patient’s risk of CR and PCD. GPS (scale 0-100) has been validated to predict adverse pathology, biochemical recurrence, metastasis, and PCD and provides a more accurate overall assessment of patient risk than clinical risk factors alone. A recent validation study found that no patients with AUA Low- or Intermediate-risk disease and a GPS of 20 units. Methods: Data from Klein et al. European Urology (EU) 2014 and Cullen et al. EU 2014 were analyzed to establish the risk of CR and PCD associated with a pre-established GPS cut-point of 20. Patients were divided based on the value of GPS (≤20, >20). Cox regression analyses accounted for cohort sampling weights. Since GPS was developed using Klein et al, standardized hazard ratios (std HR, HR for 1 SD change in the covariate) for GPS and CR and PCD survival curves for the 2 groups were estimated correcting for regression to the mean (RM). Results: Of the 402 patients in Cullen et al. (median follow up 5.2 years), only 5 patients developed metastases; all 5 had GPS >20. For Klein et al., of 426 patients with a median follow up of 6.6 years, there were 109 CR (metastasis and local recurrence) and 39 PCD; only one patient with events had a GPS

528 Background: Better tools are needed to estimate the risk of local recurrence (LR; DCIS or invasive) after breast-conserving surgery (BCS) for DCIS to inform treatment decisions. The DCIS Score (DS) was validated as a predictor of LR in E5194 and Ontario DCIS Cohort (ODC) after BCS without radiation (Solin,2013; Rakovitch,2015). We performed a meta-analysis (MA) combining data from E5194 and ODC with additional follow-up from E5194 adjusting for pertinent clinico-pathologic factors to provide refined prediction estimates of LR risk after BCS alone. Methods: The MA used data from E5194 and ODC. Patients with positive margins and multifocality were excluded. Identical Cox regression models were fit including age at diagnosis ( < 50, ≥50 yr), tumor size (1cm, > 1cm), DCIS Score and year of surgery (before vs after 2000). Grade was not significant. MA was used to calculate precision-weighted estimates of 10 year LR risk by DS. Results: Combined cohort includes 773 pts (tamoxifen used in 20% E5194, 17% of ODC > 65 yr). The DS and the clinico-pathologic variables age, tumor size and year provided independent prognostic information on 10 yr LR risk (p≤.009). Hazard ratios from E5194 and ODC cohorts were similar for tumor size ≤1 vs. > 1cm (1.45, 1.47), age ≥50 vs. < 50 yr (0.61, 0.84) and surgery year after 2000 (0.67, 0.49). 10 yr LR risks by combinations of age, tumor size, and DS are detailed in Table. For patients ≥50 yr with tumors ≤1cm and low risk DS, the 10 yr LR risks range from 5.3-10.0%. A high risk DS is associated with a higher 10 yr predicted risk of LR in all subsets. 10 yr risk of contralateral BC was 5.4%. Conclusions: This MA provides refined estimates of 10 yr LR risk after BCS alone for DCIS. Adding clinico-pathologic factors to the DCIS Score provides enhanced prognostic LR risk estimates to guide individualized treatment decision-making. [Table: see text]

In 2008, Efron showed that biological features in a high-dimensional study can be divided into classes and a separate false discovery rate (FDR) analysis can be conducted in each class using information from the entire set of features to assess the FDR within each class. We apply this separate class approach to true discovery rate degree of association (TDRDA) set analysis, which is used in clinical-genomic studies to identify sets of biomarkers having strong association with clinical outcome or state while controlling the FDR. Careful choice of classes based on prior information can increase the identification power of the separate class analysis relative to the overall analysis.

We propose a method for assessing an individual patient's risk of a future clinical event using clinical trial or cohort data and Cox proportional hazards regression, combining the information from several studies using meta-analysis techniques. The method combines patient-specific estimates of the log cumulative hazard across studies, weighting by the relative precision of the estimates, using either fixed- or random-effects meta-analysis calculations. Risk assessment can be done for any future patient using a few key summary statistics determined once and for all from each study. Generalizations of the method to logistic regression and linear models are immediate. We evaluate the methods using simulation studies and illustrate their application using real data.

The impact of cytomegalovirus (CMV) on human immunodeficiency virus type 1 (HIV-1) disease progression has been controversial. In this study, we sought to determine if CMV viral load is independent of HIV-1 viral load in predicting CMV disease and survival. Our findings indicate that in patients with advanced AIDS, CMV DNA load is an independent marker of CMV disease and survival and is more predictive than HIV-1 RNA load. Moreover, patients who respond to preemptive therapy with oral ganciclovir, with resulting undetectable levels of CMV DNA, in their plasma, have a significantly lower risk of developing CMV disease and higher rates of survival, despite stable or increasing HIV-1 RNA loads. These data provide support for CMV as an independent risk factor for mortality in persons with advanced AIDS and further suggest that effective preemptive therapy for CMV can improve patient survival rates.

100 Background: We validated a biopsy-based assay as a predictor of PCa aggressiveness in two independent patient cohorts. Using GPS with NCCN risk group, the test provides an estimate of likelihood of favorable pathology (LFP) in low-intermediate risk PCa based on data from the first validation study. The second study re-confirmed the association between GPS and LFP. We combined information from both studies to provide more precise estimates of LFP, and examined predictive models using GPS with other clinical risk tools. Methods: Patient-specific MA provides precision-weighted predictions for individual patients using data from multiple studies (Crager and Tang, J. Appl. Stat. 2014). MA was performed on the two validation studies (732 patients total) using GPS (scale 0-100) with CAPRA score, NCCN risk group, or AUA/EAU risk group as predictors of LFP (Gleason score 3+3 or 3+4, organ-confined disease). Decision curves were calculated using the MA risk estimates. Results: MA provided more precise estimates of LFP with narrower confidence intervals (CIs) than either study alone (median width 24% narrower than the smaller of the two individual study CIs). GPS added significant predictive value for LFP to each of the 3 clinical classifiers. 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 clinical classifier alone. Over a wide range of threshold probabilities, incorporation of GPS would be expected to lead to fewer treatments of patients with favorable pathology without increasing the number of patients with adverse pathology left untreated. The proportion of men with MA-estimated LFP > 80% was 31% with GPS and CAPRA, and 23% with GPS and NCCN risk group; 11% were identified with LFP > 80% using NCCN risk group alone. Conclusions: Patient-specific MA of two independent studies provided more precise risk estimates reflecting the complete body of evidence. GPS adds predictive value to three clinical risk tools. The MA-estimated LFP identified more patients with an LFP > 80% than identified by clinical classifiers alone.

104 Background: The biopsy-based GPS (scale 0-100) is validated as an independent predictor of adverse pathology in men with low- and intermediate-risk PCa. We wished to examine the association of GPS with BCR and CR in higher-risk disease. Methods: We performed exploratory analyses of data from a prior development study of radical prostatectomies from 441 men with AUA low-, intermediate- and high-risk disease, using a cohort sampling design (Klein et al. Eur Urol 2015). Multivariable Cox proportional hazards models were employed with the cohort sampling weights. Since these data were used in the selection of genes and coefficients for GPS, hazard ratios (HR) and other estimates based on GPS were corrected for regression to the mean (RM), and Storey’s method was applied to hypothesis tests for GPS to control the false discovery rate (FDR). Results: All estimates are RM-corrected. Broad, overlapping ranges of GPS values were observed across all AUA risk groups. GPS was strongly associated with BCR (HR 1.64 for 20 GPS units, p 40, who represented 41% of all intermediate-risk patients, had estimated 3-year BCR risk and 10-year CR risk similar to high-risk patients (Table). Conversely, high-risk patients with GPS≤ 40, who represented 63% of all high-risk patients, had a 3-year BCR risk and a 10-year CR risk similar to men with intermediate-risk disease. High-risk patients with GPS > 40 had 3-year BCR risk of almost 50% and 10-year CR risk of 35%. Conclusions: GPS appears to provide improved risk stratification for BCR and CR in AUA intermediate- and high-risk PCa. These findings require confirmation in an independent cohort of patients. [Table: see text]

Noninvasive detection of urine tumor DNA (utDNA) has the potential to augment the diagnosis and management of bladder cancer. We sought to identify whether patient specific molecular events identified initially in the primary tumor could be detected in their utDNA. We interrogated both genetic and epigenetic events by a number of different technologies. Single nucleotide variants (SNVs) were analyzed by Next Generation Sequencing (NGS) on both the Ion Torrent platform (using amplicon enrichment) and the Illumina platform (via hybridization capture). Thirteen differentially methylated regions (DMRs) in the genome were analyzed by methylation-specific quantitative polymerase chain reaction (MS-qPCR). The cohort included patients with newly diagnosed NMIBC and those undergoing surveillance cystoscopy following a prior diagnosis of NMIBC. DNA was isolated from fixed paraffin embedded (FPE) primary tumors, matched blood buffy coat (BC), urine sediment (US) and, in some cases, clarified urine (CU-a.k.a. urine supernatant). Tumor specific molecular events were identified by comparing primary tumor tissue to BC (for SNVs with NGS) or to pooled data from urine DNA from healthy controls (for DMRs with MS-qPCR). Initially, 8 patients were evaluated with all three technologies. A larger cohort of 66 patients undergoing surveillance for NMIBC was then analyzed using MS-qPCR and a subset was also analyzed using NGS on the Ion Torrent platform. Tumor specific DMRs and/or SNVs were detected in tumor tissue from 73 of 74 patients. In the 8 patients assessed using both NGS platforms, SNVs were found in the primary tumor for 8/8 (Ion Torrent) and 6/8 (Illumina) patients, and both NGS platforms identified utDNA in 4 of 5 biopsy confirmed NMIBC patients. In patients with sufficient DNA to be assessed with all three technologies, in both US and CU, there was 100% agreement for all positive/negative calls for utDNA. Targeted genomic regions methylated in NMIBC tissue were seen to be methylated in the corresponding utDNA at the time of diagnosis or recurrence. In the cohort of 66 patients undergoing surveillance for NMIBC using MS-qPCR, a negative predictive value for high risk recurrences of 95% (with a prevalence of 10%) was achievable. These results provide evidence that both genetic and epigenetic events are evident in utDNA and reflect the presence and genetics of the NMIBC. This proof of concept study demonstrates the potential for a personalized non-invasive means to detect and monitor bladder cancer, thus making a valuable contribution to the field of precision medicine. Citation Format: Matthew J. Resnick, Margarita Lopatin, Neal D. Shore, Po N. Lam, Brian Helfand, Richard D. Abramson, Michael Crager, Michael Bonham, Haluk Tezcan, Kim M. Clark-Langone, Chris Silk, Seth Michelson, Gregory Alexander, Phillip G. Febbo. Analysis of tumor DNA in urine as a highly sensitive liquid biopsy for patients with non-muscle invasive bladder cancer (NMIBC). [abstract]. In: Proceedings of the AACR Precision Medicine Series: Integrating Clinical Genomics and Cancer Therapy; Jun 13-16, 2015; Salt Lake City, UT. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(1_Suppl):Abstract nr 14.

Recent work on prospective power and sample size calculations for analyses of high-dimension gene expression data that control the false discovery rate (FDR) focuses on the average power over all the truly nonnull hypotheses, or equivalently, the expected proportion of nonnull hypotheses rejected. Using another characterization of power, we adapt Efron's ([2007] Ann Stat 35:1351-1377) empirical Bayes approach to post hoc power calculation to develop a method for prospective calculation of the "identification power" for individual genes. This is the probability that a gene with a given true degree of association with clinical outcome or state will be included in a set within which the FDR is controlled at a specified level. An example calculation using proportional hazards regression highlights the effects of large numbers of genes with little or no association on the identification power for individual genes with substantial association.

Purpose The 21-gene breast cancer assay recurrence score (RS) is widely used for assessing recurrence risk and predicting chemotherapy benefit in patients with estrogen receptor (ER) –positive breast cancer. Pathologic and clinical factors such as tumor size, grade, and patient age also provide independent prognostic utility. We developed a formal integration of these measures and evaluated its prognostic and predictive value. Patients and Methods From the National Surgical Adjuvant Breast and Bowel (NSABP) B-14 and translational research cohort of the Arimidex, Tamoxifen Alone or in Combination (TransATAC) studies, we included patients who received hormonal monotherapy, had ER-positive tumors, and RS and traditional clinicopathologic factors assessed (647 and 1,088, respectively). Individual patient risk assessments from separate Cox models were combined using meta-analysis to form an RS-pathology-clinical (RSPC) assessment of distant recurrence risk. Risk assessments by RS and RSPC were compared in node-negative (N0) patients. RSPC was compared with RS for predicting chemotherapy benefit in NSABP B-20. Results RSPC had significantly more prognostic value for distant recurrence than did RS (P < .001) and showed better separation of risk in the study population. RSPC classified fewer patients as intermediate risk (17.8% v 26.7%, P < .001) and more patients as lower risk (63.8% v 54.2%, P < .001) than did RS among 1,444 N0 ER-positive patients. In B-20, the interaction of RSPC with chemotherapy was not statistically significant (P = .10), in contrast to the previously reported significant interaction of RS with chemotherapy (P = .037). Conclusion RSPC refines the assessment of distant recurrence risk and reduces the number of patients classified as intermediate risk. Adding clinicopathologic measures did not seem to enhance the value of RS alone nor the individual biology RS identifies in predicting chemotherapy benefit.

This prospective, randomized, multicenter, controlled trial was designed to evaluate the efficacy and safety of intravenous ganciclovir for the treatment of peripheral cytomegalovirus (CMV) retinitis in patients with AIDS. Patients were randomly assigned to receive either immediate treatment, intravenous ganciclovir, 5 mg/kg twice daily for 14 days followed by 5 mg/kg once daily for 14 weeks, or deferred treatment. Patients randomized to deferred treatment whose retinitis progressed were offered ganciclovir. Of the 22 patients randomized to deferred treatment who were included in the final analysis, 20 were found to have progressive CMV retinitis compared with 10 of the 13 randomized to immediate treatment. The median time to progression in the deferred treatment group, as determined by a masked fundus photography reading center, was 13.5 days compared with 49.5 days in the immediate treatment group (mean +/- SD, 19.3 +/- 4.1 vs. 66.4 +/- 14.0; P = .001, log rank test). These data indicate that ganciclovir delays the progression of CMV peripheral retinitis in persons with AIDS.

Analyses intended to identify genes with expression that is associated with some clinical outcome or state are often based on ranked p-values from tests of point null hypotheses of no association. Van de Wiel and Kim take the innovative approach of testing the interval null hypotheses that the degree of association for a gene is less than some value of interest against the alternative that it is greater. Combining this idea with the false discovery rate controlling methods of Storey, Taylor and Siegmund gives a computationally simple way to identify true discovery rate degree of association (TDRDA) sets of genes among which a specified proportion are expected to have an absolute association of a specified degree or more. This leads to a gene ranking method that uses the maximum lower bound degree of association for which each gene belongs to a TDRDA set. Estimates of each gene's actual degree of association with approximate correction for 'selection bias' due to regression to the mean (RM) can be derived using simple bivariate normal theory and Efron and Tibshirani's empirical Bayes approach. For a given data set, all possible TDRDA sets can be displayed along with the gene ranking and the RM-corrected estimates of degree of association in a concise graphical summary.

Background: Analytically and clinically validated non-invasive blood tests that quantify breast cancer burden and clinical drug response/resistance are greatly needed. Many groups have successfully detected tumor markers in blood using a variety of technologies, including next generation sequencing (NGS). We performed a comprehensive NGS study on a small number of patients to evaluate the value of global versus individual markers for the quantitation of tumor-derived cell free DNA (cfDNA) in plasma. Methods: DNA isolated from formalin-fixed primary tumor, buffy coat cells, and plasma from 2 patients with metastatic breast cancer were characterized simultaneously for copy number aberrations (CNAs) and differentially methylated regions (DMRs) using whole genome bisulfite sequencing (WBGS), and targeted sequencing-based genotyping of 346 cancer-associated single nucleotide variations (SNVs). CNA and DMR regions were identified from log normalized, GC content corrected counts and DMR data using Poisson and binomial distribution theory and false discovery rate controlling methods. Percent tumor in cfDNA was estimated from the normalized ratio (plasma: primary tumor) of CNA or DMR compared to buffy coat, aggregating over genomic regions. Sample sets from 8 non-metastatic patients were also profiled using the targeted SNV panel in order to compare SNVs between samples and estimate percent tumor cfDNA. Results: WGBS detected tumor specific alterations in each primary tumor compared to buffy coat. By analyzing the genome using 100 Kb bins, we observed over 1000 bins with detectable CNA signal and, among 56 million CpG sites, over 30,000 DMRs. As expected, 5 or fewer informative somatic SNVs were detected in each patient. Analysis of these somatic changes in plasma revealed that the tumor fraction estimated from SNV detected in cfDNA varied widely between sites originally discovered in the patient’s primary tumor. In contrast, similar estimates of tumor fraction in cfDNA were obtained using CNA and DMR profiles within each patient; both methods yielded similar estimates of over 50% in one patient and less than 10% in the other. For the patient with high tumor fraction, both CNA and DMR profiles contained examples of individual large genomic regions that displayed additional clear aberrations in the plasma compared to the original tumor, such as a striking loss of a >25 Mb region of chromosome 4. Conclusions: Although individual somatic SNV in cfDNA can be detected in metastatic disease, calculated allelic fraction based on individual SNVs varies greatly within the same patient. Measuring and integrating CNA or DMR across the genome provided more consistent and reliable estimates of tumor DNA fraction in plasma, and also revealed alterations in plasma from patients with metastatic disease that were not prominent in the primary tumor. Citation Format: Ellen M Beasley, Richard D Abramson, Gregory E Alexander, David Chan, Kristen Bradley, Francois Collin, Michael Crager, Andrew Dei Rossi, Joseph Dorado, Adam Friedman, William J Gibb, Jennie Jeong, Col Jones, C J Ku, Yan Ma, John Morlan, Kunbin Qu, Aibing Rao, Aaron Scott, Haluk Tezcan. Global quantitative measures using next-generation sequencing for breast cancer presence outperform individual tumor markers in plasma [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P4-02-08.

Aims The anti-anginal efficacy and safety of ranolazine in diabetic and non-diabetic patients included in the Combination Assessment of Ranolazine In Stable Angina (CARISA) trial ( JAMA 2004; 291 :309) were studied. Glycaemic control was also assessed in CARISA and its long-term open-label extension study. Methods and results Patients with chronic angina enrolled in CARISA (189 with diabetes, 634 without diabetes) on background atenolol, diltiazem, or amlodipine therapy were randomized to placebo, ranolazine 750 or 1000 mg twice daily for 12 weeks, during which exercise tolerance, angina frequency, nitroglycerin usage, glucose, HbA1c, and lipids were measured. Patients completing the randomized study could enroll in an ongoing open-label extension study and were evaluated every 3 months. Ranolazine produced similar improvements in exercise parameters, nitroglycerin use, and angina frequency in diabetic and non-diabetic patients. Adverse events were similar between groups. Fasting glucose and lipids remained unaltered in diabetic patients after 12 weeks of therapy. In a post hoc analysis, ranolazine 750 and 1000 mg reduced HbA1c vs. placebo by 0.48±0.18% ( P =0.008) and 0.70±0.18% ( P =0.0002), respectively; the HbA1c levels appeared to remain unchanged over time during long-term therapy. Conclusion Anti-anginal efficacy and safety of ranolazine for angina were similar between diabetic and non-diabetic patients. Ranolazine significantly improved glycaemic control in diabetic patients.

e14569 Background: Appropriate use of prognostic and predictive markers depends on quality of evidence for clinical utility in addition to clinical validity. In stage II CC, RR assessment is based traditionally on clinicopathologic factors (CP) with limited clinical validation and varies substantially across providers. We hypothesized that the validated Oncotype DX Colon Cancer RS and MMR tests may impact clinical decision-making by decreasing variation in RR assessment across MDs. We conducted a survey to compare the level of MD agreement on stage II CC RR assessment using CP alone vs CP+RS+MMR (CP+). Methods: A clinical trial database was randomly sampled for 100 cases of stage II CC stratified on T-stage, MMR and RS groups. Anonymous internet surveys asked each MD to assess 3-year RR for 10 cases with CP and 10 different cases with CP+. MDs were divided into panels of 5; all members of each panel reviewed the same cases. Agreement in RR among MDs was assessed using within-panel mean squared difference in RR assessments and analyzed using generalized linear models. Results: 30 community (C) and 20 university-based (U) MDs (Assoc of Northern California Oncologists or NCCN, respectively) completed evaluable surveys April-June 2012. For C vs U, median years in practice were 5 and 2; prior use of RS 13% vs 25%; and routine MMR use 21% vs 55%. The standard deviation of the differences (SDD) in RR assessments for CP alone was high overall (7%), and greater for U vs C MDs (8.2% vs 6.4%). CP+ produced higher agreement in RR assessments vs CP for both C (p

21 Background: We previously identified messenger RNAs (mRNAs) whose expression can distinguish aggressive from indolent prostate cancer. Representing multiple key genomic pathways, these mRNAs are significantly associated with cR and PCSS after RP, providing prognostic information beyond PSA, cT stage and Gleason Score. We evaluated microRNA expression in the same specimens for association with cR and PCSS. Methods: All cT1/cT2 prostate cancer pts treated with RP at Cleveland Clinic from 1987-2004 were identified (n~2,600), of which 127 pts w/ cR and 374 pts w/o cR after RP were randomly selected using cohort sampling. RNA was purified from 2 macrodissected FPE tumor specimens per pt. Expression of 76 test and 5 reference microRNAs was quantified by RT-PCR. Cox regression and control of the false discovery rate (FDR) was used to assess reference-normalized microRNA and mRNA expression for association with cR and PCSS. Results: 106 pts with cR and 310 without cR had sufficient RNA and successful microRNA assays. Analysis of primary Gleason pattern tumor tissue for each pt identified 21 microRNAs associated with cR and 13 microRNAs associated with PCSS, with FDR at 10%; 8 microRNAs were associated with both endpoints. Similar analysis of highest Gleason pattern tumor tissue for each pt identified 22 microRNAs associated with cR (17 overlapping with those for the primary Gleason pattern) and 7 microRNAs associated with PCSS, with FDR at 10%; 4 were associated with both endpoints. miR-1, miR-21, miR-93, and miR-106b were associated with both cR and PCSS in primary and highest Gleason pattern specimens. The 76 microRNAs in this study tended to have weaker association with cR and PCSS than the 732 mRNAs. In multivariate analyses, mRNAs and microRNAs provided prognostic information beyond PSA, cT stage, and biopsy Gleason score. MicroRNAs co-express more frequently with each other than with mRNAs, which may indicate distinct biological regulation. Conclusions: Expression of some microRNAs assayed in FPE prostate tumor tissue was associated with cR and PCSS after RP in this study, and may retain prognostic value in the face of tumor heterogeneity.

509 Background: The 21-gene Oncotype DX Recurrence Score (RS) is widely used for assessment of recurrence risk and prediction of chemotherapy benefit in patients with early stage ER-positive breast cancer. Pathologic and clinical factors such as tumor size and grade, and patient age also provide independent prognostic utility and might be combined with the RS to achieve more prognostic power. Methods: All patients in the NSABP trial B-14 and the ATAC study who had ER-positive tumor specimens with successful Oncotype DX RS assay were included. B-14 patients were node-negative and were treated with tamoxifen. ATAC patients were node-positive or node-negative, and were treated with tamoxifen or anastrozole. Meta-analysis methods were used to assess the risk of distant recurrence by combining the individual study multivariate risk assessments using RS-pathologic-clinical (RSPC) information. A RSPC risk index was defined a priori as the predicted risk of distant recurrence at 10 years with predetermined cut-of...

SUMMARY A method is presented for analyzing ambulatory blood pressure monitoring (ABPM) time series data obtained from well-controlled clinical trials. The method uses running averages based on fixed time-of-day intervals (rather than a fixed number of neighboring measurements). These "interval running averages" effectively estimate average blood pressure during the specified time intervals, adjusting for unequal spacing between measurements, embedded missing data, varying measurement times-of-day, and doses of study medication taken during ABP monitoring. Blood pressure changes from baseline may be computed using the interval running averages in order to separate treatment effects from patients' normal daily blood pressure cycles. To ensure valid estimation of treatment effects over time, study medication dosing times should be rigorously controlled in the trial design and conduct. Interval running average curves may be presented graphically, and from them summary statistics may be computed for purposes of statistical analysis. By allowing for the inherent complications of ABP data collection, the effect of antihypertensive treatment in well-controlled clinical trials can be discerned. Use of ambulatory blood pressure monitoring (ABPM) is becoming increasingly common in the evaluation of the efficacy of antihypertensive agents. ABPM provides a method for measuring blood pressure over the course of an entire day, in settings other than a physician's office. A portable ABPM device is worn by patients in clinical trials over an extended period of time (typically 24 hours). At preset intervals during this time the device automatically inflates a blood pressure measuring cuff. Systolic and diastolic blood pressure are recorded, along with heart rate. Lightweight ABPM devices and ease of data transmission have now made it possible to include ambulatory monitoring in routine clinical studies. In-office hypertension has long been considered diagnostic of disease needing treatment. Control of blood pressure as measured in the physician's office has been considered evidence of effective antihypertensive treatment, as patients with high in-office blood pressure have been found more likely to go on to have adverse cardiovascular events such as strokes and heart attacks (Kannel, 1974). Several studies have suggested, however, that whole-day blood pressure monitoring may provide greater insight into hypertension and the effect of antihypertensive therapy (Pickering and Devereux, 1987; Kain, Hinman, and Sokolow, 1964; Porchet et al., 1986, Sokolow et al., 1966). It may become evident that the morbidity associated with hypertension is better reduced by ensuring 24-hour control of blood pressure. In spite of the many medical publications on the clinical applications of ABP monitoring, there has been little statistical work published to date on the analysis of ABP data. Data from studies

The response to acetylcholine (ACh) administered microiontophoretically was determined for single units in CA1 and CA3–4 areas of the hippocampi of both young (3–5 months) and old (24–26 months) Fisher 344 rats. Single units in CA1 showed a greater response to ACh (20 nA) than single units in CA3–4 in young and in old rats. However, the response to ACh in single units of young rats was significantly greater than in single units of old rats. Baseline firing rates in old rats were lower than those in young rats. However, there was no difference in baseline firing rates between the two areas of the hippocampus indicating that the differential response to ACh as a function of area cannot be attributed to a difference in baseline firing rate.

Analysis of covariance (ANCOVA) techniques are often employed in the analysis of clinical trials to try to account for the effects of varying pretreatment baseline values of an outcome variable on posttreatment measurements of the same variable. Baseline measurements of outcome variables are typically random variables, which violates the usual ANCOVA assumption that covariate values are fixed. Therefore, the usual ANCOVA hypothesis tests of treatment effects may be invalid, and the ANCOVA slope parameter estimator biased, for this application. We show, however, that if the pretreatment - posttreatment measurements have a bivariate normal distribution, then (i) the ANCOVA model with residual error independent of the covariate is a valid expression of the relationship between pretreatment and posttreatment measurements; (ii) the usual (fixed-covariate analysis) ANCOVA estimates of the slope parameter and treatment effect contrasts are unbiased; and (iii) the usual ANCOVA treatment effect contrast t-tests are valid significance tests for treatment effects. Moreover, as long as the magnitudes of the treatment effects do not depend on the "true" pretreatment value of the outcome variable, the true slope parameter must lie in the interval (0, 1) and the ANCOVA model has a clear interpretation as an adjustment (based on between- and within-subject variability) to an analysis of variance model applied to the posttreatment-pretreatment differences.