Zheng Su
Research Subjects
Zheng Su has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
Yunzhi Lin
Genentech Inc. , South San Francisco , CA , USA
3 shared publicationsBoan Zhang
Genentech Inc. , South San Francisco , CA , USA
1 shared publication- Bryan Irving
Genentech Research and Early Development, 1 DNA Way, South San Francisco, CA 94080, USA
1 shared publication - Clifton O. BinghamIII
Johns Hopkins School of Medicine Baltimore Maryland
1 shared publication - Eric Stefanich
Genentech Inc, 1 DNA Way, South San Francisco, CA 94080 USA
1 shared publication - Heleen Scheerens
Genentech, Inc, South San Francisco, Calif
1 shared publication
Latest Publications
- 17 September 2013
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 43 (2) , 390-399
Dichotomizing a Continuous Baseline Covariate in Randomized Clinical Trials with Binary Outcomes
After a randomized clinical trial is conducted, it is often desired to dichotomize a continuous baseline covariate such that a subgroup of patients who may benefit more from the experimental treatment can be identified. We herein propose an approach to dichotomizing a continuous baseline covariate in randomized clinical trials by generalizing an existing likelihood-based approach for one-sample observational studies. Numerical results show that the proposed approach can accurately detect the cutpoint and is less likely to produce estimates that are markedly different from the true cutpoint value compared with the widely adopted minimum p-value approach.
- 17 September 2013
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 43 (2) , 390-399
Dichotomizing a Continuous Baseline Covariate in Randomized Clinical Trials with Binary Outcomes
After a randomized clinical trial is conducted, it is often desired to dichotomize a continuous baseline covariate such that a subgroup of patients who may benefit more from the experimental treatment can be identified. We herein propose an approach to dichotomizing a continuous baseline covariate in randomized clinical trials by generalizing an existing likelihood-based approach for one-sample observational studies. Numerical results show that the proposed approach can accurately detect the cutpoint and is less likely to produce estimates that are markedly different from the true cutpoint value compared with the widely adopted minimum p-value approach.
- 30 July 2013
- Published by Wiley in Pharmaceutical Statistics
- Vol. 12 (5) , 275-281
A hybrid approach to achieving both marginal and conditional balances for stratification variables in sequential clinical trials
Various methods exist in the literature for achieving marginal balance for baseline stratification variables in sequential clinical trials. One major limitation with balancing on the margins of the stratification variables is that there is an efficiency loss when the primary analysis is stratified. To preserve the efficiency of a stratified analysis one recently proposed approach balances on the crossing of the stratification variables included in the analysis, which achieves conditional balance for the variables. A hybrid approach to achieving both marginal and conditional balances in sequential clinical trials is proposed, which is applicable to both continuous and categorical stratification variables. Numerical results based on extensive simulation studies and a real dataset show that the proposed approach outperforms the existing ones and is particularly useful when both additive and stratified models are planned for a trial. Copyright © 2013 John Wiley & Sons, Ltd.
- 30 July 2013
- Published by Wiley in Pharmaceutical Statistics
- Vol. 12 (5) , 275-281
A hybrid approach to achieving both marginal and conditional balances for stratification variables in sequential clinical trials
Various methods exist in the literature for achieving marginal balance for baseline stratification variables in sequential clinical trials. One major limitation with balancing on the margins of the stratification variables is that there is an efficiency loss when the primary analysis is stratified. To preserve the efficiency of a stratified analysis one recently proposed approach balances on the crossing of the stratification variables included in the analysis, which achieves conditional balance for the variables. A hybrid approach to achieving both marginal and conditional balances in sequential clinical trials is proposed, which is applicable to both continuous and categorical stratification variables. Numerical results based on extensive simulation studies and a real dataset show that the proposed approach outperforms the existing ones and is particularly useful when both additive and stratified models are planned for a trial. Copyright © 2013 John Wiley & Sons, Ltd.
- 19 March 2013
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 40 (5) , 1132-1139
A note on variance estimation in the Cox proportional hazards model
The Cox proportional hazards model is widely used in clinical trials with time-to-event outcomes to compare an experimental treatment with the standard of care. At the design stage of a trial the number of events required to achieve a desired power needs to be determined, which is frequently based on estimating the variance of the maximum partial likelihood estimate of the regression parameter with a function of the number of events. Underestimating the variance at the design stage will lead to insufficiently powered studies, and overestimating the variance will lead to unnecessarily large trials. A simple approach to estimating the variance is introduced, which is compared with two widely adopted approaches in practice. Simulation results show that the proposed approach outperforms the standard ones and gives nearly unbiased estimates of the variance.
- 19 March 2013
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 40 (5) , 1132-1139
A note on variance estimation in the Cox proportional hazards model
The Cox proportional hazards model is widely used in clinical trials with time-to-event outcomes to compare an experimental treatment with the standard of care. At the design stage of a trial the number of events required to achieve a desired power needs to be determined, which is frequently based on estimating the variance of the maximum partial likelihood estimate of the regression parameter with a function of the number of events. Underestimating the variance at the design stage will lead to insufficiently powered studies, and overestimating the variance will lead to unnecessarily large trials. A simple approach to estimating the variance is introduced, which is compared with two widely adopted approaches in practice. Simulation results show that the proposed approach outperforms the standard ones and gives nearly unbiased estimates of the variance.
- 30 August 2012
- Published by Wiley in Statistics in Medicine
- Vol. 32 (7) , 1239-1245
A sequential testing approach to detecting multiple change points in the proportional hazards model
The semi-parametric proportional hazards model has been widely adopted in clinical trials with time-to-event outcomes. A key assumption in the model is that the hazard ratio function is a constant over time, which is frequently violated as there is often a lag period before an experimental treatment reaches its full effect. One existing approach uses maximal score tests and Monte Carlo sampling to identify multiple change points in the hazard ratio function, which requires the number of change points that exist in the model to be known. We propose a sequential testing approach to detecting multiple change points in the hazard ratio function using likelihood ratio tests, and the distributions of the likelihood ratio statistics under the null hypothesis are evaluated via resampling. An important feature of the proposed approach is that the number of change points in the model is inferred from the data and does not need to be specified. Numerical results based on simulated clinical trials and a real time-to-event study show that the proposed approach can accurately detect the change points in the hazard ratio function. Copyright © 2012 John Wiley & Sons, Ltd.
- 30 August 2012
- Published by Wiley in Statistics in Medicine
- Vol. 32 (7) , 1239-1245
A sequential testing approach to detecting multiple change points in the proportional hazards model
The semi-parametric proportional hazards model has been widely adopted in clinical trials with time-to-event outcomes. A key assumption in the model is that the hazard ratio function is a constant over time, which is frequently violated as there is often a lag period before an experimental treatment reaches its full effect. One existing approach uses maximal score tests and Monte Carlo sampling to identify multiple change points in the hazard ratio function, which requires the number of change points that exist in the model to be known. We propose a sequential testing approach to detecting multiple change points in the hazard ratio function using likelihood ratio tests, and the distributions of the likelihood ratio statistics under the null hypothesis are evaluated via resampling. An important feature of the proposed approach is that the number of change points in the model is inferred from the data and does not need to be specified. Numerical results based on simulated clinical trials and a real time-to-event study show that the proposed approach can accurately detect the change points in the hazard ratio function. Copyright © 2012 John Wiley & Sons, Ltd.
- 25 April 2012
- Published by Wiley in Statistics in Medicine
- Vol. 31 (18) , 1961-1971
Balancing continuous and categorical baseline covariates in sequential clinical trials using the area between empirical cumulative distribution functions
Covariate adaptive allocation is often adopted in sequential clinical trials to maintain the balance of baseline covariates that could potentially confound the outcome of a trial. Several allocation methods exist in the literature that can handle both continuous and categorical covariates. We propose a minimization approach to maintaining the balance of multiple continuous and categorical covariates in sequential clinical trials, which uses the area between the empirical cumulative distribution functions of the observed covariate values as the imbalance metric. Numerical results based on extensive simulation studies and a real dataset show that the proposed approach produces more accurate estimates of the treatment effect and leads to more powerful trials than the existing approaches for trials with binary, continuous, and time‐to‐event outcomes. Copyright © 2012 John Wiley & Sons, Ltd.
- 25 April 2012
- Published by Wiley in Statistics in Medicine
- Vol. 31 (18) , 1961-1971
Balancing continuous and categorical baseline covariates in sequential clinical trials using the area between empirical cumulative distribution functions
Covariate adaptive allocation is often adopted in sequential clinical trials to maintain the balance of baseline covariates that could potentially confound the outcome of a trial. Several allocation methods exist in the literature that can handle both continuous and categorical covariates. We propose a minimization approach to maintaining the balance of multiple continuous and categorical covariates in sequential clinical trials, which uses the area between the empirical cumulative distribution functions of the observed covariate values as the imbalance metric. Numerical results based on extensive simulation studies and a real dataset show that the proposed approach produces more accurate estimates of the treatment effect and leads to more powerful trials than the existing approaches for trials with binary, continuous, and time‐to‐event outcomes. Copyright © 2012 John Wiley & Sons, Ltd.
- 1 November 2011
- Published by Springer Nature in Drug Information Journal
- Vol. 45 (6) , 725-729
Optimal Allocation of Prognostic Factors in Randomized Preclinical Animal Studies
- 1 November 2011
- Published by Springer Nature in Drug Information Journal
- Vol. 45 (6) , 725-729
Optimal Allocation of Prognostic Factors in Randomized Preclinical Animal Studies
- Open Access
- 26 October 2011
- Published by Springer Nature in Arthritis Research & Therapy
- Vol. 13 (5) , R177
MTRX1011A, a humanized anti-CD4 monoclonal antibody, in the treatment of patients with rheumatoid arthritis: a Phase I randomized, double-blind, placebo-controlled study incorporating pharmacodynamic biomarker assessments
Introduction: The purpose of this study was to evaluate the safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD) of the humanized anti-CD4 monoclonal antibody MTRX1011A in a randomized, double-blind placebo-controlled Phase 1 study in patients with rheumatoid arthritis (RA). Methods: In the single ascending dose (SAD) portion of the study, patients received single doses of a placebo or MTRX1011A at 0.3, 1.0, 3.5 and 7.0 mg/kg intravenously (IV) or 1.0 and 3.5 mg/kg subcutaneously (SC), followed by five weeks of evaluation. In the multi-dose (MD) portion of the study, placebo or MTRX1011A was administered weekly for eight doses at 1.5 or 3.5 mg/kg SC, or 5 mg/kg IV, followed by eight weeks of evaluation. Results: MTRX1011A was well tolerated in the SAD phase up to 7 mg/kg IV and in the MD phase up to 1.5 mg/kg SC. At weekly doses of 3.5 mg/kg SC and 5 mg/kg IV, a moderate pruritic papular rash was observed in some MTRX1011A-treated patients, which was considered a dose-limiting toxicity for this clinical indication. No serious adverse events occurred in any cohort. Reduction in disease activity was modest. PD assessments demonstrated that MTRX1011A induced a dose-dependent down-modulation of CD4 expression on peripheral blood CD4 T cells, CD4 receptor occupancy, increases in serum sCD4-MTRX1011A complexes and up-regulation of CD69 on T cells, but was non-depleting. Conclusions: The maximum tolerated dose of MTRX1011A was 1.5 mg/kg SC administered weekly. At this dose MTRX1011A did not achieve maximum PD activity expected to be required for reduction in disease activity.
- 26 October 2011
- Published by Springer Nature in Arthritis Research & Therapy
- Vol. 13 (5) , R177
Open AccessMTRX1011A, a humanized anti-CD4 monoclonal antibody, in the treatment of patients with rheumatoid arthritis: a Phase I randomized, double-blind, placebo-controlled study incorporating pharmacodynamic biomarker assessments
Introduction: The purpose of this study was to evaluate the safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD) of the humanized anti-CD4 monoclonal antibody MTRX1011A in a randomized, double-blind placebo-controlled Phase 1 study in patients with rheumatoid arthritis (RA). Methods: In the single ascending dose (SAD) portion of the study, patients received single doses of a placebo or MTRX1011A at 0.3, 1.0, 3.5 and 7.0 mg/kg intravenously (IV) or 1.0 and 3.5 mg/kg subcutaneously (SC), followed by five weeks of evaluation. In the multi-dose (MD) portion of the study, placebo or MTRX1011A was administered weekly for eight doses at 1.5 or 3.5 mg/kg SC, or 5 mg/kg IV, followed by eight weeks of evaluation. Results: MTRX1011A was well tolerated in the SAD phase up to 7 mg/kg IV and in the MD phase up to 1.5 mg/kg SC. At weekly doses of 3.5 mg/kg SC and 5 mg/kg IV, a moderate pruritic papular rash was observed in some MTRX1011A-treated patients, which was considered a dose-limiting toxicity for this clinical indication. No serious adverse events occurred in any cohort. Reduction in disease activity was modest. PD assessments demonstrated that MTRX1011A induced a dose-dependent down-modulation of CD4 expression on peripheral blood CD4 T cells, CD4 receptor occupancy, increases in serum sCD4-MTRX1011A complexes and up-regulation of CD69 on T cells, but was non-depleting. Conclusions: The maximum tolerated dose of MTRX1011A was 1.5 mg/kg SC administered weekly. At this dose MTRX1011A did not achieve maximum PD activity expected to be required for reduction in disease activity.
- 1 July 2011
- Published by Elsevier in Contemporary Clinical Trials
- Vol. 32 (4) , 547-550
Balancing multiple baseline characteristics in randomized clinical trials
- 1 July 2011
- Published by Elsevier in Contemporary Clinical Trials
- Vol. 32 (4) , 547-550
Balancing multiple baseline characteristics in randomized clinical trials
- 3 June 2011
- Published by Taylor & Francis in International Journal of Computer Mathematics
- Vol. 88 (13) , 2743-2751
Bounds for an estimate of the optimal backward error for linear least squares problems
Algorithms for calculating sequences of upper and lower bounds for an estimate of the optimal backward error for linear least squares problems are developed, based on the Gauss quadrature theory. Numerical results show that the bounds converge quickly and are therefore useful in practice.
- 3 June 2011
- Published by Taylor & Francis in International Journal of Computer Mathematics
- Vol. 88 (13) , 2743-2751
Bounds for an estimate of the optimal backward error for linear least squares problems
Algorithms for calculating sequences of upper and lower bounds for an estimate of the optimal backward error for linear least squares problems are developed, based on the Gauss quadrature theory. Numerical results show that the bounds converge quickly and are therefore useful in practice.
- 19 April 2011
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 40 (8) , 1163-1170
A Note on Importance Resampling for Multi-Dimensional Statistics
Johns ( 1988 Johns , M. V. (1988). Importance sampling for bootstrap confidence intervals. Journal of the American Statistical Association 83:709–714. [Taylor & Francis Online], [Web of Science ®] [Google Scholar] ), Davison ( 1988 Davison , A. C. ( 1988 ). Discussion of paper by D. V. Hinkley . Journal of the Royal Statistical Society Series B 50 : 356 – 357 . [Google Scholar] ), and Do and Hall ( 1991 Do , K. A. , Hall , P. ( 1991 ). On importance sampling for the bootstrap . Biometrika 78 : 161 – 167 . [Crossref], [Web of Science ®] [Google Scholar] ) used importance sampling for calculating bootstrap distributions of one-dimensional statistics. Realizing that their methods can not be extended easily to multi-dimensional statistics, Fuh and Hu ( 2004 Fuh , C. D. , Hu , I. ( 2004 ). Efficient importance sampling for events of moderate deviations with applications . Biometrika 91 : 471 – 490 . [Crossref], [Web of Science ®] [Google Scholar] ) proposed an exponential tilting formula for statistics of multi-dimension, which is optimal in the sense that the asymptotic variance is minimized for estimating tail probabilities of asymptotically normal statistics. For one-dimensional statistics, Hu and Su ( 2008 Hu , J. , Su , Z. ( 2008 ). Adaptive resampling algorithms for estimating bootstrap distributions . Journal of Statistical Planning and Inference 138 ( 6 ): 1763 – 1777 . [Crossref], [Web of Science ®] [Google Scholar] ) proposed a multi-step variance minimization approach that can be viewed as a generalization of the two-step variance minimization approach proposed by Do and Hall ( 1991 Do , K. A. , Hall , P. ( 1991 ). On importance sampling for the bootstrap . Biometrika 78 : 161 – 167 . [Crossref], [Web of Science ®] [Google Scholar] ). In this article, we generalize the approach of Hu and Su ( 2008 Hu , J. , Su , Z. ( 2008 ). Adaptive resampling algorithms for estimating bootstrap distributions . Journal of Statistical Planning and Inference 138 ( 6 ): 1763 – 1777 . [Crossref], [Web of Science ®] [Google Scholar] ) to multi-dimensional statistics, which applies to general statistics and does not resort to asymptotics. Empirical results on a real survival data set show that the proposed algorithm provides significant computational efficiency gains.
- 19 April 2011
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 40 (8) , 1163-1170
A Note on Importance Resampling for Multi-Dimensional Statistics
Johns ( 1988 Johns , M. V. (1988). Importance sampling for bootstrap confidence intervals. Journal of the American Statistical Association 83:709–714. [Taylor & Francis Online], [Web of Science ®] [Google Scholar] ), Davison ( 1988 Davison , A. C. ( 1988 ). Discussion of paper by D. V. Hinkley . Journal of the Royal Statistical Society Series B 50 : 356 – 357 . [Google Scholar] ), and Do and Hall ( 1991 Do , K. A. , Hall , P. ( 1991 ). On importance sampling for the bootstrap . Biometrika 78 : 161 – 167 . [Crossref], [Web of Science ®] [Google Scholar] ) used importance sampling for calculating bootstrap distributions of one-dimensional statistics. Realizing that their methods can not be extended easily to multi-dimensional statistics, Fuh and Hu ( 2004 Fuh , C. D. , Hu , I. ( 2004 ). Efficient importance sampling for events of moderate deviations with applications . Biometrika 91 : 471 – 490 . [Crossref], [Web of Science ®] [Google Scholar] ) proposed an exponential tilting formula for statistics of multi-dimension, which is optimal in the sense that the asymptotic variance is minimized for estimating tail probabilities of asymptotically normal statistics. For one-dimensional statistics, Hu and Su ( 2008 Hu , J. , Su , Z. ( 2008 ). Adaptive resampling algorithms for estimating bootstrap distributions . Journal of Statistical Planning and Inference 138 ( 6 ): 1763 – 1777 . [Crossref], [Web of Science ®] [Google Scholar] ) proposed a multi-step variance minimization approach that can be viewed as a generalization of the two-step variance minimization approach proposed by Do and Hall ( 1991 Do , K. A. , Hall , P. ( 1991 ). On importance sampling for the bootstrap . Biometrika 78 : 161 – 167 . [Crossref], [Web of Science ®] [Google Scholar] ). In this article, we generalize the approach of Hu and Su ( 2008 Hu , J. , Su , Z. ( 2008 ). Adaptive resampling algorithms for estimating bootstrap distributions . Journal of Statistical Planning and Inference 138 ( 6 ): 1763 – 1777 . [Crossref], [Web of Science ®] [Google Scholar] ) to multi-dimensional statistics, which applies to general statistics and does not resort to asymptotics. Empirical results on a real survival data set show that the proposed algorithm provides significant computational efficiency gains.