|
|
|
|
|
|
|
· Dynamic
treatment sequencing |
·
Methods |
· Invited · Others |
|
Treatment of complex diseases such as cancer, leukemia, AIDS and depression usually follows complex treatment regimes consisting of a sequence of pre- and post-remission therapies. An adaptive treatment strategy (ATS), sometimes referred to as a dynamic treatment regime is a sequence of individually tailored treatments. Under an ATS, during the treatment period individuals receive time-varying treatment based on their health status and other eligibility criteria specified prior to the start of the treatment. An example might be the cases where patients are treated with one of several available treatments (or different doses of same drug) for a fixed period of time and then based on the intermediate response are switched to a different treatment. Another example of the application of such dynamic treatment regime is where patients receive an induction therapy and among those who achieve remission receive some form of maintenance therapy.
While making treatment decisions at each stage, a physician looks at multiple factors including (i) treatments assigned in prior stages (ii) response to the treatments in prior stage (iii) health status (quality of life) of the patient and (iv) possible treatments the patient is eligible for at that particular stage. The goal at each stage is to decide on the treatment which will result in largest short/long-term benefit. If the number of stages and the number of treatment options at each stage are small, sequential multiple assignment randomization trials could be used to compare different treatment strategies. My main research interest is efficient estimation of survival distributions under treatment strategies and comparison of multiple treatment strategies based on observational or randomized studies.
In the year 2007, we have formed a reading group to accelerate our research in this area. Please visit the reading group website Adaptive Treatment Strategy Reading Group.
In recent years, there has been considerable interest among
investigators in the construction of general class of skewed distributions
which includes the standard symmetric distributions such as the normal, t,
logistic and Cauchy distributions. The key is to introduce additional parameters
or parametric functions in the distributional form that accounts for the skewness of the distribution. The idea became
institutionalized when Azzalini (1985) defined a
class of distribution (which he referred to as skew-normal) by introducing an
additional skewness parameter that included the
normal distribution as a special case. The name suggests that this
distribution, unlike the normal distribution, is asymmetric in general and
allows both positively and negatively skewed distributions. Subsequently, Azzalini and Dalla Valle (1996)
came up with the multivariate version of the skew-normal distribution. A
statistical application of the multivariate skew-normal distribution was
considered by Azzalini and Capitanio
(1999). This paper popularized the application of such distributions and led
the way for others to define similar families of distributions based on other
symmetric distributions such as skew-Cauchy (
My research with HCV is a result of my collaboration in the
clinical trials/studies coordinated by the
In addition, currently I am a co-investigator in a Phase I/II clinical trial (SYNCH) sponsored by NIDDK and NCAM to investigate the safety and efficacy of silymarin as a treatment for hepatitis C.
The Longitudinal Assessment of Bariatric Surgery (LABS) is a National Institutes of Health (NIH)-funded consortium of six clinical centers and a data coordinating center working in cooperation with NIH scientific staff to plan, develop, and conduct coordinated clinical, epidemiological, and behavioral research in the field of bariatric surgery. Recently I joined this group as a statistician.
Recently I started collaborating with Clinical Research, Investigation, and Systems Modeling of Acute Illness Laboratory (CRISMA) as a statistician. CRISMA is directed by Derek C. Angus MD, MPH, and co-directed by Gilles Clermont MD, MSc. This large team of clinicians, mathematicians, and epidemiologists enjoys superb funding from the NIH and multiple industrial sponsors. Regarded by many around the world as the leading investigative team carrying out studies of the clinical epidemiology of critical illness, Dr. Angus and his colleagues are actively studying the genetics of human sepsis, a syndrome that affects about 750,000 Americans every year and carries a mortality rate of almost 30 percent. Dr. Angus and his team of scientists have published papers in leading journals such as JAMA, Lancet, Critical Care Medicine, and the American Journal of Respiratory and Critical Care Medicine. I mainly collaborate with John Kellum in HIDONOR and MONiTOR study.
1. Kidwell, KM and Wahed, AS (2011). Weighted Log-rank Statistic to Compare
Shared-Path Adaptive Treatment Strategies. In revision, Technical Draft
2. Wahed,
AS and Thall, PF (2012). Evaluating
Joint E_ects of Induction-Salvage Treatment Regimes
on Overall Survival in Acute Leukemia. In Press, Journal of Royal Statistical
Society, Series C.
3. Ko, JH, and Wahed, AS (2012). Up-front vs. Sequential Randomizations for
Inference on Adaptive Treatment Strategies, Statistics in Medicine, Volume 31,
Issue 9, pages 812–830, 30 April 2012.
4. Tang, X,
and Wahed, AS (2011).
Comparison of Treatment Regimes with Adjustment for Auxiliary Variables.
Journal of Applied Statistics, 2011; 38: 2925-2938.
5. Wahed, AS (2011). On the Equivalence of
Inverse-Probability-of-Censoring-Weighted and Kaplan-Meier Estimators. Invited
Article, Journal of the Applied Statistical Science, Volume 18, Issue 4.
6. Miyahara, S, and Wahed, AS (2010). Weighted Kaplan-Meier Estimators for Two-Stage
Treatment Regimes. Statistics
in Medicine, 2010; 29: 2581-2591.
7. Wahed
, AS (2010). Inference for Two-Stage Adaptive Treatment
Strategies Using Mixture Distributions. Journal of Royal Statistical
Society Series C (Applied Statistics) Appl. Statist. (2010) 59.
8. Feng, W and Wahed, AS (2009).
Sample Size for Two-Stage Studies with Maintenance Therapy. Statistics
in Medicine, 2009; 28: 2028-2041
9. Wahed,
AS ,Luong, T, and Jeong, J-H (2009). A new
generalization of Weibull distribution with
application to a breast cancer data set. Accepted, Statistics
in Medicine, 2009; 28: 2077-2094
12. Ali, MM, Woo, J, Pal, M and Wahed,
AS. (2008) Some Skew-Symmetric Double Inverted Distributions.
International Journal of Statistical Sciences. Vol. 7, pp.
1-12.
13. Wahed,
AS. (2007) The family of curvi-triangular
distributions. International Journal of Statistical Sciences, Vol 6, pp 7-18.
14. Wahed,
AS and Tsiatis, A.
A. (2006) Semi-parametric Efficient Estimation of The Survival Distribution for
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials with
Censored Data. Biometrika Vol
93: pp. 147-161.
15. Wahed,
AS. (2006)
Bayesian Inference Using Burr Model Under Asymmetric
Loss Function: An Application to Carcinoma Survival Data. Journal of
Statistical Research, 2006, Vol. 40, No. 1, pp. 45-57.
16. Wahed,
AS. (2006). A
General Method of Constructing Extended Families of Distributions from an
Existing Continuous Class. Journal of Probability and Statistical Science 4(2),
165-177.
17. Feng, W and Wahed, AS
(2006). A Review of Inferential Procedures for Survival Analysis in Two-Stage
Randomization Designs. Far East Journal of Theoretical Statistics Vol. 19 (1), pp 117 – 139.
18. Wahed,
AS and Tsiatis, A.
A.(2004). Optimal estimator for the survival
distribution and related quantities for treatment policies in two-stage
randomization designs in clinical trials, Biometrics, Vol. 60, No. 1. pp 124-133.
19. Wahed,
AS and Ali, MM
(2001). The Skew-Logistic Distribution. Journal of Statistical Research, Vol.
35, No,2, pp. 71-80.
20. Wahed,
AS and Uddin, B (1998). Bayes Estimation Under
Asymmetric Loss. Dhaka University Journal of Science, Vol. 46.
13. December,
2009. Assessing the effect of treatment regimes on longitudinal outcome data. Department
of Applied Statistics,
14. December,
2009. A new generalization to the Weibull
distribution with an application to a breast cancer dataset. 7th triennial statistics and probability
conference,
15. November,
2009. Statistical methods for comparing dynamic treatment regimes with
time-to-event endpoints. Department of Statistics,
16. November,
2009. Statistical methods for comparing dynamic treatment regimes with
time-to-event endpoints. Center for Statistical Sciences,
17. August,
2009. Dynamic Treatment regimes in
Leukemia Treatment. Joint Statistical Meetings 2009, American Statistical
Association.
18. February,
2009. Adaptive Designs in Clinical Trials.
19. September, 2008. Comparing adaptive treatment
strategies following sequential multiple assignment randomization trials,
Clinical and Translational Sciences Research Institute, CHRC, RAND-Pittsburgh
Institute, VA-CHERP.
20. July, 2008. Inference on dynamic treatment
regimes following sequential multiple assignment randomization trials,
Center for Statistics at Queen Mary, University of London.
21. February,
2008. Adaptive treatment strategies – one step forward towards individualized
treatment rules. Dean’s Junior Faculty Seminar Series, Graduate School of
Public Health, Pittsburgh.
22. February,
2008. Adaptive Designs in Clinical Trials.
23. January
2008. Supremum weighted log-rank test and sample size for comparing two-stage
adaptive treatment strategies. Department of Epidemiology, Biostatistics and
Occupational Health,
24. June
2007. Semi-parametric methods for estimating causal effect of treatment
strategies in two-stage randomization clinical trials. ICSA 2007 Applied statistics Symposium, Raleigh,
25. May
2007. Weibull-based approaches to survival analysis
with applications to breast cancer data. Department of Mathematical
Sciences,
26. February
2007. Comparing Adaptive Treatment Strategies: Challenges and Solutions. Center
for Health Equity, Research and Promotions (CHERP), VA Health Care System,
27. December
2006. Survival Analysis for Comparing Adaptive Treatment Strategies. Department
of Applied Statistics,
28. April
2006. Survival Analysis in Two-stage Randomization Designs in Leukemia Trials. Department
of Mathematics and Statistics,
29. February
2006. Survival Analysis in Two-stage Randomization Designs in Oncolgy Trials. Invited presentation, Department of
Biostatistics,
30. March
2005. Survival Analysis in Two-stage Randomization Designs. Invited lecture in
the session CENSORED DATA IN THE ENVIRONMENTAL, AGRICULTURAL AND MEDICAL
SCIENCES. International Biometric Society (ENAR) spring meetings, 2006,
Tampa,
31. March
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials. University
of
32. March
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials. Department
of Biostatistics, and Department of Statistics and Actuarial Science,
33. March
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials. University
of
34. February
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials.
35. February
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials. Department
of Biostatistics,
36. February
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials. Department
of Health Studies,
37. February
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials.
38. January
2003. Survival Analysis in Two-Stage Randomization Designs in Clinical Trials. Department
of Mathematics and Statistics,
39. January
2003. Optimal Estimator of the Survival Distribution and Related Quantities of
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials. Department
of Biostatistics and Epidemiology,
1.
(with Kelley M. Kidwell) Weighted Log-rank Statistic to
Compare Shared-path Adaptive Treatment Strategies. Joint Statistical Meetings,
2012, San Diego, CA.
2.
(with Xinxin
Dong and Lan Kong) Accelerated Failure Time Model for Case-Cohort Design with
Longitudinal Covariates Measured with Error. Joint Statistical Meetings, 2012, San
Diego, CA.
3.
(with Kelley M. Kidwell) Weighted Log-rank Statistic to
Compare Shared-path Adaptive Treatment Strategies. Society for Clinical Trials
Annual Meeting, 2012, Orlando, FL.
4.
(with Abidemi K. Adeniji) Discrete Survival Analysis
with Misclassified Events. International Biometric Society (ENAR) spring
meetings, 2012, Washington, DC.
5.
(with Xinxin Dong and Lan Kong) Accelerated Failure
Time Model for Case-Cohort Design with Longitudinal Covariates Measured with ErrorInternational Biometric Society (ENAR) spring
meetings, 2012, Washington, DC.
6.
(with Kelley M.
Kidwell) Weighted Log-rank Statistic to Compare Shared-path Adaptive Treatment
Strategies. International Biometric Society (ENAR) spring meetings, 2012,
Washington, DC.
7.
(poster with Chetachi A. Emeremni) Analysis of Variance
for Right Censored Survival Data. International Biometric Society (ENAR) spring
meetings, 2012, Washington, DC.
8.
(poster with
Jesse Hsu) Causal Inference for Treatment Strategies from Two-Stage
Randomization Designs. 34th Midwest Biopharmaceutical Statistical
Workshop, May 2010,
9. (With Jinhui Ko) Nonparametric Estimation of Median Residual Life Function for Two-Stage randomization Designs. ENAR Spring meeting 2010.
10. (With Xinyu Tang) Cox Proportional Hazard Model for Dynamic Treatment Regimes. ENAR Spring meeting 2010.
11.
(poster with Jinhui Ko) Nonparametric Estimation of
Median Residual Life Function for Two-Stage randomization Designs. 33rd Midwest
Biopharmaceutical Statistical Workshop, May 2009,
12.
(poster with Xinyu Tang) Statistical Methods for
Sequentially Randomized Trials: An Application to High-Risk Neuroblastoma
Study. 33rd
Midwest Biopharmaceutical Statistical Workshop, May 2009,
13.
Supremum Weighted
Log-rank Test and Sample Size for Comparing Two-stage Adaptive Treatment
Strategies, July 2008, International Biometric Conference,
14.
Inverse-probability-weighting-based sample size formula
for comparing two-stage adaptive treatment strategies. ENAR Spring Meetings
2008,
15. Discussion on Two-Stage Treatment Strategies Based on Sequential Failure Times by Peter Thall (with Patricia Houck, and Jinhui Ko), ATSRG reading group meeting, November 2007, Department of Biostatistics, University of Pittsburgh Graduate School of Public Health.
16.
(poster) Weighted Kaplan-Meier Estimator for Adaptive
Treatment Strategies. SAMSI workshop on dynamic treatment regimes, June 2007,
Statistical and Applied Mathematical Sciences Institute, RTP,
17.
(poster) Weibull-based approaches to survival analysis: an
application to a breast cancer data set. 30th
18.
A supremum log-rank test for two-stage adaptive
treatment strategies and corresponding sample size formula, ENAR Spring
Meetings 2007,
19. Introduction to Adaptive Treatment Strategies with Examples (with Sachiko Miyahara). ATSRG reading group meeting, March 2007, Department of Biostatistics, University of Pittsburgh Graduate School of Public Health.
20.
Likelihood Inference for Survival Analysis in Two-stage
Randomization Designs, Joint Statistical Meetings, August 2006,
21. (poster) Inferences for Treatment Regimes in Two Stage Clinical Trials, Midwest Biopharmaceutical statistics workshop, May 2006, Muncie, Indiana.
22. (Poster) Insulin Resistance Is Independent Of Hepatic Steatosis and Is Augmented By Environmental Factors Such As Obesity in Patients With HCV Genotype 1 Infection, DDW, Chicago, Illinois, May 2005.
23.
Genetic variation in an interferon-stimulated gene, mxyovirus-1 (MxA), has a significant
protective effect from fibrosis in genotype-1 chronic hepatitis C virus
infection, DDW,
24.
A non-linear mixed effect model for hepatitis C viral
dynamics, International Biometric Society (ENAR) spring meetings, 2005,
25.
A non-linear mixed effect model for hepatitis C viral
dynamics, Joint Statistical Meetings 2004,
26.
Presented in the Faculty Seminar Series, Department of
Biostatistics,
27.
Race, Insulin Resistance, Visceral Adiposity and
Hepatic Steatosis in Genotype 1 Patients with Chronic
Hepatitis C, AASLD, November 2004
(poster).
28.
Efficient Estimation of the Survival Distribution for
Treatment Policies in Two-Stage Randomization Designs in Clinical Trials with
Censored Data. International Biometric
Society (ENAR) meeting 2004,
29.
Optimal Estimator of the Survival Distribution and
Related Quantities of Treatment Policies in Two-Stage Randomization Designs in
Clinical Trials. Joint Statistical
Meetings 2003, American Statistical Association,
30.
Optimal Estimator of the Survival Distribution and
Related Quantities of Treatment Policies in Two-Stage Randomization Designs in
Clinical Trials. International Biometric
Society (ENAR) meeting 2003,
· Shekhar Mehta, MS (2006)
Thesis
title: Longitudinal
Analysis of Renal Function using ZIP GEE on OLT Transplant Patients Undergoing
NAC Prophylaxis
Current
position: Entry position: Pharm.D program, University
of Maryland School of Pharmacy
· Wentao Feng, PhD (2008)
Thesis
title: Inference, Power and
Sample Size Adaptive Treatment Strategies
Current
position: Sr. Biostatistician, Novartis
Pharmaceuticals Corporation
· Sachiko Miyahara, PhD (2009)
Thesis
title: Statistical Inferences for
Two-Stage Treatment Regimes for Time-To-Event and Longitudinal Data
Current
position: Research Associate, Center for Biostatistics in AIDS Research,
Department of Biostatistics,
· Jinhui Ko, PhD (2010)
Thesis
title: Statistical Issues in the
Design and Analysis of Sequentially Randomized Trials
Entry position: Statistician, Biostatistics Solutions;
Current position: Statistician, GlaxoSmithKline
·Xinyu Tang, PhD (2010)
Thesis title: Analyzing Survival Data For Sequentially Randomized Designs
Current position: Assistant Professor, Department of
Pediatrics,
·Jesse Hsu, PhD (2011)
Thesis title: Longitudinal Data Analysis in
Depression Studies: Assessment of Intermediate-Outcome-Dependent Dynamic
Interventions
Current position: Post-doctoral Associate, University
of Pennsylvania, Wharton School
·Zhen Jiang, PhD (2011)
Thesis title: Joint Modeling of Multivariate
Ordinal Longitudinal Outcome
Current position: Division of Biostatistics, Office of
Biostatistics and Epidemiology, CBER, U.S. Food and Drug Administration
·Kelley Kidwell, PhD (2012)
Thesis title: Survival Analysis of Shared-Path Adaptive
Treatment Strategies
Current position: Research Assistant Professor, University
of Michigan, Department of Biostatistics.
·Abidemi Adeniji, PhD (2012).
Thesis title: Incorporating Diagnostic Accuracy into the
Estimation of Discrete Survival Function
Current position: Statistician, Boehringer
Ingelheim
·Chetachi Ememerni, PhD (2012).
Thesis title: Inference for Right Censored, and Right
Censored Length Biased Data through Inverse Weighting
Current position: Assistant Professor of Pediatrics,
and Preventive Medicine, The University of Tennessee Health Science Center
·Xinxin Dong (co-advisor with Dr. Lan Kong)
Expected Ph.D. graduation date: December 2012.
Thesis title: TBD
Current position: TBD
·Semhar Ogbagaber
Expected
Ph.D. graduation date: 2013
Thesis title: TBD
Current position: TBD
·Emmanuel Sampene
Expected
Ph.D. graduation date: 2013
Thesis title: TBD
Current position: TBD