Biomedical subjects

Daniel F Heitjan

Publications and source records attributed to Daniel F Heitjan.

At least 19 recordsLinked to original sources

PubMed2007

Sensitivity of the hazard ratio to nonignorable treatment assignment in an observational study.

In non-randomized studies, estimation of treatment effects generally requires adjustment for imbalances in observed covariates. One such method, based on the propensity score, is useful in many applications but may be biased when the assumption of strongly ignorable treatment assignment is violated. Because it is not possible to evaluate this assumption from the data, it is advisable to assess the sensitivity of conclusions to violations of strong ignorability. Lin et al. (Biomet. 1998; 54:948-963) have implemented this idea by investigating how an unmeasured covariate may affect the conclusions of an observational study. We extend their method to assess sensitivity of the treatment hazard ratio to hidden bias under a range of covariate distributions. We derive simple formulas for approximating the true from the apparent treatment hazard ratio estimated under a specific survival model, and assess the validity of these formulas in simulation studies. We demonstrate the method in an analysis of SEER-Medicare data on the effects of chemotherapy in elderly colon cancer patients.

Aged↗

PubMed2007

Impact of nonignorable coarsening on Bayesian inference.

The coarse data model of Heitjan and Rubin (1991) generalizes the missing data model of Rubin (1976) to cover other forms of incompleteness such as censoring and grouping. The model has 2 components: an ideal data model describing the distribution of the quantity of interest and a coarsening mechanism that describes a distribution over degrees of coarsening given the ideal data. The coarsening mechanism is said to be nonignorable when the degree of coarsening depends on an incompletely observed ideal outcome, in which case failure to properly account for it can spoil inferences. A theme in recent research is to measure sensitivity to nonignorability by evaluating the effect of a small departure from ignorability on the maximum likelihood estimate (MLE) of a parameter of the ideal data model. One such construct is the "index of local sensitivity to nonignorability" (ISNI) (Troxel and others, 2004), which is the derivative of the MLE with respect to a nonignorability parameter evaluated at the ignorable model. In this paper, we adapt ISNI to Bayesian modeling by instead defining it as the derivative of the posterior expectation. We propose the application of ISNI as a first step in judging the robustness of a Bayesian analysis to nonignorable coarsening. We derive formulas for a range of models and apply the method to evaluate sensitivity to nonignorable coarsening in 2 real data examples, one involving missing CD4 counts in an HIV trial and the other involving potentially informatively censored relapse times in a leukemia trial.

Bayes Theorem↗

PubMed2006

A pattern-mixture model for the analysis of censored quality-of-life data.

We propose a pattern-mixture model for describing the joint distribution of incomplete repeated measurements of quality of life (QoL) and right-censored survival times. The model assumes that the survival times follow a multinomial distribution and that the quality of life outcome follows a multivariate normal distribution conditional on the survival time. We estimate the model using a Bayesian approach by importance sampling. We then use simulated parameters to create multiple imputations of the censored QoL outcomes, which can then be used to calculate individual values of quality-adjusted life-years (QALYs). We apply the method to data from the Randomized Evaluation of Mechanical Assistance in the Treatment of Congestive Heart Failure (REMATCH) clinical trial.

Bayes Theorem↗

PubMed2006

Cost-effectiveness of preventive strategies for women with a BRCA1 or a BRCA2 mutation.

BACKGROUND: For BRCA1 or BRCA2 mutation carriers, decision analysis indicates that prophylactic surgery or chemoprevention leads to better survival than surveillance alone. OBJECTIVE: To evaluate the cost-effectiveness of the preventive strategies that are available to unaffected women carrying a single BRCA1 or BRCA2 mutation with high cancer penetrance. DESIGN: Markov modeling with Monte Carlo simulations and probabilistic sensitivity analyses. DATA SOURCES: Breast and ovarian cancer incidence and mortality rates, preference ratings, and costs derived from the literature; the Surveillance, Epidemiology, and End Results (SEER) Program; and the Health Care Financing Administration (now the Centers for Medicare & Medicaid Services). TARGET POPULATION: Unaffected carriers of a single BRCA1 or BRCA2 mutation 35 to 50 years of age. TIME HORIZON: Lifetime. PERSPECTIVE: Health policy, societal. INTERVENTIONS: Tamoxifen, oral contraceptives, bilateral salpingo-oophorectomy, mastectomy, both surgeries, or surveillance. OUTCOME MEASURES: Cost-effectiveness. RESULTS OF BASE-CASE ANALYSIS: For mutation carriers 35 years of age, both surgeries (prophylactic bilateral mastectomy and oophorectomy) had an incremental cost-effectiveness ratio over oophorectomy alone of 2352 dollars per life-year for BRCA1 and 100 dollars per life-year for BRCA2. With quality adjustment, oophorectomy dominated all other strategies for BRCA1 and had an incremental cost-effectiveness ratio of 2281 dollars per life-year for BRCA2. RESULTS OF SENSITIVITY ANALYSIS: Older age at intervention increased the cost-effectiveness of prophylactic mastectomy for BRCA1 mutation carriers to 73,755 dollars per life-year. Varying the penetrance, mortality rates, costs, discount rates, and preferences had minimal effects on outcomes. LIMITATIONS: Results are dependent on the accuracy of model assumptions. CONCLUSION: On the basis of this model, the most cost-effective strategies for BRCA mutation carriers, with and without quality adjustment, were oophorectomy alone and oophorectomy and mastectomy, respectively.

Adult↗

PubMed2006

A predictive model for the detection of tumor lysis syndrome during AML induction therapy.

Tumor lysis syndrome (TLS) is defined by metabolic derangements occurring in the setting of rapid tumor destruction. In acute myelogenous leukemia (AML), TLS frequency, risk stratification, monitoring, and management strategies are based largely on case series and data from other malignancies. A single-center, retrospective cohort study was conducted to estimate TLS incidence and identify TLS predictive factors in a patient population undergoing myeloid leukemia induction chemotherapy. This study included 194 patients, aged 18-86 years, with AML or advanced myelodysplastic syndrome undergoing primary myeloid leukemia induction chemotherapy. Nineteen patients (9.8%) developed TLS. In univariate analysis, elevated pre-chemotherapy values for uric acid (P < 0.0001), creatinine (P = 0.0025), lactate dehydrogenase (LDH) (P = 0.0001), white blood cell (P = 0.0058), gender (P = 0.0064) and chronic myelomonocytic leukemia history (P = 0.0292) were significant predictors. In multivariate analysis, LDH (P = 0.0042), uric acid (P < 0.0001) and gender (P = 0.0073) remained significant TLS predictors. A predictive model was then designed using a scoring system based on these factors. This analysis may lay the groundwork for the development of the first evidence-based guidelines for TLS monitoring and management in this patient population.

Adolescent↗

PubMed2006

Predicting event times in clinical trials when treatment arm is masked.

Because power is primarily determined by the number of events in event-based clinical trials, the timing for interim or final analysis of data is often determined based on the accrual of events during the course of the study. Thus, it is of interest to predict early and accurately the time of a landmark interim or terminating event. Existing Bayesian methods may be used to predict the date of the landmark event, based on current enrollment, event, and loss to follow-up, if treatment arms are known. This work extends these methods to the case where the treatment arms are masked by using a parametric mixture model with a known mixture proportion. Posterior simulation using the mixture model is compared with methods assuming a single population. Comparison of the mixture model with the single-population approach shows that with few events, these approaches produce substantially different results and that these results converge as the prediction time is closer to the landmark event. Simulations show that the mixture model with diffuse priors can have better coverage probabilities for the prediction interval than the nonmixture models if a treatment effect is present.

Bayes Theorem↗

PubMed2006

A simple local sensitivity analysis tool for nonignorable coarsening: application to dependent censoring.

Right- and interval-censored data are common special cases of coarsened data (Heitjan and Rubin, 1991, Annals of Statistics19, 2244-2253). As with missing data, standard statistical methods that ignore the random nature of the coarsening mechanism may lead to incorrect inferences. We extend a simple sensitivity analysis tool, the index of local sensitivity to nonignorability (Troxel, Ma, and Heitjan, 2004, Statistica Sinica14, 1221-1237), to the evaluation of nonignorability of the coarsening process in the general coarse-data model. By converting this index into a simple graphical display one can easily assess the sensitivity of key inferences to nonignorable coarsening. We illustrate the validity of the method with a simulated example, and apply it to right-censored data from an observational study of cardiac transplantation and to interval-censored data on time to detectable viral load from a clinical trial in HIV disease.

Biometry↗

PubMed2005

DNA methylation in anal intraepithelial lesions and anal squamous cell carcinoma.

PURPOSE: Anal intraepithelial neoplasia is associated with human papillomavirus infection and may progress to invasive squamous cell carcinoma (SCC), which is increasing in immunocompromised patients. We hypothesize that anal intraepithelial neoplasia is associated with abnormal DNA methylation and that detection of these events may be used to improve screening programs. EXPERIMENTAL DESIGN: Seventy-six patients were identified who underwent anal cytology screening and subsequent biopsy at our institution between 1999 and 2004. The specimens from these patients included 184 anal biopsies [normal, n = 57; low-grade squamous intraepithelial lesion (LSIL), n = 74; high-grade squamous intraepithelial lesion (HSIL), n = 41; and invasive SCC, n = 12] and 37 residual liquid-based anal cytology specimens (normal, n = 11; LSIL, n = 12; HSIL, n = 14). The methylation status of the following genes was determined for each biopsy and cytology sample using real-time methylation-specific PCR: HIC1, RASSF1, RARB, CDKN2A, p14, TP73, APC, MLH1, MGMT, DAPK1, and IGSF4. RESULTS: Methylation-specific PCR analysis of biopsy samples revealed that DNA methylation was more common in SCC and HSIL than LSIL and normal mucosa. Specifically, methylation of IGSF4 and DAPK1 was prevalent in SCC (75% and 75% of cases, respectively) and HSIL (59% and 71%, respectively) but was absent in LSIL and normal biopsy samples. Methylation profiles of cytologic samples were similar to those found in the biopsy samples. CONCLUSIONS: Aberrant DNA methylation is a frequent event in anal HSIL and SCC. Methylation of IGSF4 and DAPK1 is specific for HSIL and SCC, and may serve as a useful molecular biomarker.

Adaptor Proteins, Signal Transducing↗

PubMed2005

Suppression of beta-catenin by antisense oligomers augments tumor response to isolated limb perfusion in a rodent model of adenomatous polyposis coli-mutant colon cancer.

BACKGROUND: Isolated hepatic perfusion has been used in patients with colorectal cancer (CRC) metastatic to the liver. We sought to determine whether perfusion with antisense oligodeoxynucleotides results in the downregulation of beta-catenin and whether this improves tumor response to isolated limb perfusion (ILP) in a heterotopic model of human CRC. METHODS: Adenomatous polyposis coli-mutant human CRC xenografts were implanted into athymic rats. Animals were randomized to the following groups: (1) no treatment, (2) control ILP, (3) melphalan ILP, (4) ILP with antisense specific for beta-catenin, (5) ILP with nonspecific antisense, and (6) melphalan plus beta-catenin-specific antisense ILP. Tumor response and Western blot analysis of protein expression were evaluated. RESULTS: The maximal decrease (mean +/- SE) in tumor volume was 0% +/- 10% for no treatment, 19% +/- 14% for control ILP, 58% +/- 3% for melphalan ILP, 58% +/- 9% for beta-catenin-specific ILP, 13% +/- 19% for nonspecific antisense ILP, and 73% +/- 6% for melphalan plus beta-catenin-specific ILP (P < .05 for melphalan ILP, beta-catenin-specific ILP, and melphalan plus antisense ILP). Tumor regrowth was delayed for 6 days after control ILP, 24 days after melphalan ILP, 20 days after beta-catenin-specific ILP, 10 days after nonspecific antisense ILP, and 60 days after melphalan plus beta-catenin-specific ILP (P < .05 for melphalan plus beta-catenin-specific ILP compared with all others). Western blotting revealed prolonged suppression of beta-catenin expression after beta-catenin-specific ILP. CONCLUSIONS: Short-term beta-catenin antisense treatment improves tumor response rates after ILP in a rodent model of human CRC.

Adenocarcinoma↗

PubMed2005

An index of local sensitivity to nonignorable drop-out in longitudinal modelling.

In longitudinal studies with potentially nonignorable drop-out, one can assess the likely effect of the nonignorability in a sensitivity analysis. Troxel et al. proposed a general index of sensitivity to nonignorability, or ISNI, to measure sensitivity of key inferences in a neighbourhood of the ignorable, missing at random (MAR) model. They derived detailed formulas for ISNI in the special case of the generalized linear model with a potentially missing univariate outcome. In this paper, we extend the method to longitudinal modelling. We use a multivariate normal model for the outcomes and a regression model for the drop-out process, allowing missingness probabilities to depend on an unobserved response. The computation is straightforward, and merely involves estimating a mixed-effects model and a selection model for the drop-out, together with some simple arithmetic calculations. We illustrate the method with three examples.

Animal Feed↗

PubMed2005

Multiple imputation for model checking: completed-data plots with missing and latent data.

In problems with missing or latent data, a standard approach is to first impute the unobserved data, then perform all statistical analyses on the completed dataset--corresponding to the observed data and imputed unobserved data--using standard procedures for complete-data inference. Here, we extend this approach to model checking by demonstrating the advantages of the use of completed-data model diagnostics on imputed completed datasets. The approach is set in the theoretical framework of Bayesian posterior predictive checks (but, as with missing-data imputation, our methods of missing-data model checking can also be interpreted as "predictive inference" in a non-Bayesian context). We consider the graphical diagnostics within this framework. Advantages of the completed-data approach include: (1) One can often check model fit in terms of quantities that are of key substantive interest in a natural way, which is not always possible using observed data alone. (2) In problems with missing data, checks may be devised that do not require to model the missingness or inclusion mechanism; the latter is useful for the analysis of ignorable but unknown data collection mechanisms, such as are often assumed in the analysis of sample surveys and observational studies. (3) In many problems with latent data, it is possible to check qualitative features of the model (for example, independence of two variables) that can be naturally formalized with the help of the latent data. We illustrate with several applied examples.

Animals↗

PubMed2005

Nonignorable censoring in randomized clinical trials.

BACKGROUND: In a clinical trial, survival may be censored by the end of the study, especially for subjects who enter later in the enrollment period. If there is a trend toward better survival over time then longer survivors experience shorter censoring times (heavier censoring). In such a case, the censoring and survival times are correlated, and thus the censoring is nonignorable in the sense that standard survival models that assume independent censoring could yield incorrect inferences. We will demonstrate a graphical method for analyzing sensitivity of estimates of survival model parameters to small departures from nonignorable censoring. METHODS: We assume a parametric model of survival together with a scaled beta model for the censoring process that incorporates the dependence of censoring time on survival time. We assess sensitivity using an index of local sensitivity to nonignorability (Troxel et al.). High sensitivity indicates a large impact of nonignorable censoring on the parameter of interest and a need for additional modeling. RESULTS: A simulation study shows that the approach is valid for practical use. We apply our method to a clinical trial evaluating the survival benefit of a surgically implanted left ventricular assist device in subjects with end-stage heart failure. Sensitivity is somewhat larger in estimates of the mean survival in the device arm, where survival is better and the fraction censored is therefore larger. The degree of nonignorability required to substantially affect estimates is larger than seems plausible, however. CONCLUSIONS: Our results illustrate how one can apply sensitivity analysis to evaluate the reliability of survival parameter estimates in a clinical trial.

Clinical Trials as Topic↗

PubMed2004

DNA methylation profiling of cervical squamous intraepithelial lesions using liquid-based cytology specimens: an approach that utilizes receiver-operating characteristic analysis.

BACKGROUND: Cervical carcinoma is a common malignancy among women worldwide, and its pathogenesis is related causally to human papillomavirus infection. The progression from precursor squamous intraepithelial lesions to cervical carcinoma requires additional genetic and epigenetic alterations that have not been characterized fully. The authors examined aberrant promoter methylation of multiple tumor suppressor genes in precursor squamous intraepithelial lesions. METHODS: A multiplex, nested, methylation-specific polymerase chain reaction approach was used to examine promoter methylation of 15 tumor suppressor genes in high-grade squamous intraepithelial lesions (HSIL, n = 11), low-grade squamous intraepithelial lesions (LSIL, n = 17), and negative tissues (n = 11) from liquid-based cytology samples. The area under the receiver-operating characteristic (ROC) curve was determined for individual methylated tumor suppressor genes and for gene combinations to evaluate test performance for the ability of methylation profiles to distinguish HSIL cytology samples from combined LSIL/negative cytology samples. RESULTS: Aberrant promoter methylation of DAPK1 and IGSF4 occurred at a high frequency in HSIL samples and was absent in LSIL and negative samples. There was a significant trend toward increased methylation with the increased severity of lesions, and the mean number of methylated genes was significantly higher in HSIL samples compared with LSIL and negative samples. Using the area under the ROC curve as a measure of test performance, the methylation of IGSF4 and DAPK1 had areas that were significantly greater than 0.5; thus, each had the ability to distinguish HSIL samples from combined LSIL/negative samples. The areas under the curve for the best two-gene combination (IGSF4/DAPK1) and the best three-gene combination (IGSF4/DAPK1/HIC1) were not statistically different from the best individual tumor suppressor gene (IGSF4) in distinguishing HSIL samples from combined LSIL/negative samples. CONCLUSIONS: Aberrant promoter methylation of tumor suppressor genes is an epigenetic alteration that occurs during neoplastic progression to cervical carcinoma. The methylation status of multiple tumor suppressor genes can be evaluated using ROC analysis to determine methylation profiles that can distinguish HSIL samples from combined LSIL/negative samples.

Adolescent↗

PubMed2004

Bayesian estimation of cost-effectiveness from censored data.

We describe a Bayesian methodology for estimating the cost-effectiveness of a new treatment compared to a standard in a clinical trial, when censoring of survival, the effectiveness variable, induces censoring of total cost. The statistical model assumes that survival follows a Weibull distribution and that total health care cost follows a gamma distribution whose mean has a linear regression on survival time. We summarize the posterior distributions of key parameters by importance sampling. We illustrate the method with an analysis of data from a randomized clinical trial of a treatment for cardiovascular disease.

Bayes Theorem↗

PubMed2004

Bayesian estimation of cost-effectiveness: an importance-sampling approach.

We describe a method for estimating the cost-effectiveness of a new treatment compared to a standard, using data from a comparative clinical trial. We quantify the clinical effectiveness as a binary variable indicating success or failure. The underlying statistical model assumes that costs are uncensored and follow separate gamma distributions in each of the groups defined by the four possible combinations of treatment arm and effectiveness outcome. The method is subjectivist, in that it represents prior uncertainty about model parameters with a probability distribution, which we update via Bayes's theorem to produce a posterior distribution. We approximate the posterior by importance sampling, a straightforward simulation method. We illustrate the method with an analysis of cost (derived from resource usage data) and effectiveness (measured by one-year survival) in a clinical trial in heart disease. The example demonstrates that the method is practical and provides for a flexible data analysis.

Bayes Theorem↗

PubMed2004

Sensitivity analysis of causal inference in a clinical trial subject to crossover.

In many clinical trials it is possible for some subjects to cross over between treatment arms. One can evaluate the effect of crossover by modeling it as a missing-data problem, where for subjects who cross over, one treats the unobserved value of the outcome in the original randomization arm as the missing data. The as-treated analysis is invalid if the crossover is nonignorable, in the sense that the crossovers represent a nonrandom sample of the randomized subjects. A recent area of general interest is the development of methods for measuring the sensitivity of inferences to nonignorability in the missing-data mechanism; one such approach is that of Troxel et al. In this paper we apply their method to the problem of measuring sensitivity to nonignorable crossover in randomized trials, extending it to the case where the crossover mechanism may differ between arms. Our method allows us to identify circumstances under which the as-treated analysis may be more or less sensitive to nonignorable crossover. We illustrate it with the example of a randomized clinical trial (RCT) in multiple sclerosis and a study of the effect of military service on income.

Causality↗

PubMed2004

Nonparametric prediction of event times in randomized clinical trials.

In clinical trials with planned interim analysis, it can be valuable for logistical reasons to predict the times of landmark events such as the 50th and 100th event. Bagiella and Heitjan (Stat Med 2001; 20: 2055-63) proposed a parametric prediction model for failure-time outcomes assuming exponential survival and Poisson enrollment. When little is known about the distributions of interest, there is concern that parametric prediction methods may be biased and inefficient if their underlying distributional assumptions are invalid. We propose nonparametric approaches to make point and interval predictions for landmark dates during the course of the trial. We obtain point predictions using the Kaplan-Meier estimator to extrapolate the survival probability into the future, selecting the time when the expected number of events is equal to the landmark number. To construct prediction intervals, we use a simulation strategy based on the Bayesian bootstrap. Monte Carlo results demonstrate the superiority of the nonparametric method when the assumptions underlying the parametric model are incorrect. We demonstrate the methods using data from a trial of immunotherapy of chronic granulomatous disease.

Granulomatous Disease, Chronic↗

PubMed2004

Metaiodobenzylguanidine and hyperglycemia augment tumor response to isolated limb perfusion in a rodent model of human melanoma.

BACKGROUND: Perfusate acidification with dilute hydrochloric acid augments tumor response rates in a rodent model of isolated limb perfusion (ILP). This study investigates the combination of metaiodobenzylguanidine (MIBG), a mitochondrial inhibitor, and systemic hyperglycemia as a strategy to selectively acidify tumors and thereby sensitize them to ILP. METHODS: Human melanoma xenografts were implanted into the hind limbs of athymic rats. When tumors reached 12 to 15 mm in diameter, animals were randomized to ILP with or without melphalan, with or without systemic MIBG, and hyperglycemia of 485 +/- 35 mg/dL. Intratumoral pH was measured during MIBG and glucose treatment by using magnetic resonance spectroscopy. RESULTS: MIBG at 30 mg/kg plus hyperglycemia decreased intracellular pH by.6 units and extracellular pH by.8 units. MIBG at 22.5 mg/kg plus hyperglycemia decreased intracellular and extracellular pH by.4 and.5 units, respectively. Tumor growth was unaffected by systemic MIBG and hyperglycemia alone. When MIBG at 30 mg/kg and hyperglycemia were combined with ILP, tumor growth was delayed for 33 days after control ILP and for 44 days after melphalan ILP. However, this dose of MIBG was complicated by a 40% mortality rate after ILP. MIBG at 22.5 mg/kg, in combination with MIBG in the perfusate, did not cause mortality and delayed tumor growth by 51 days after melphalan ILP. CONCLUSIONS: MIBG and hyperglycemia improve tumor response rates after ILP in a rodent model of human melanoma. Selective tumor acidification with MIBG and hyperglycemia may offer added benefit to current regional perfusion strategies.

3-Iodobenzylguanidine↗