Pretreatment Dietary Patterns, Serum Carotenoids and Tocopherols Influence Tumor Immune Response in Head and Neck Squamous Cell Carcinoma.

BACKGROUND: Tumor infiltrating lymphocytes (TILs) aid in informing treatment for head and neck squamous cell carcinoma (HNSCC). Nevertheless, little is known about the role of diet on TILs. METHODS: Immunohistologic expression of CD4, CD8, CD68, CD103, CD104 and FOXP3 were assessed in tissue microarrays from 233 previously untreated HNSCC patients. Associations between these markers and pretreatment dietary patterns were evaluated using linear regression. Associations between baseline serum carotenoids, tocopherols and TILs were assessed using logistic regression. Cox models evaluated the association between diet and TILs on overall and recurrence-free survival. RESULTS: Consumption of a Western dietary pattern was associated with lower CD8+ and FOXP3+ infiltrates (p-value:0.03 and 0.02, respectively). Multivariable logistic regression models demonstrated significantly higher CD8+ (OR:2.21;p-value:0.001) and FOXP3+ (OR:4.26;p-value:<0.0001) among patients with high gamma tocopherol. Conversely, high levels of xanthophylls (OR:0.12;p-value:<0.0001), lycopene (OR:0.36;p-value:0.0001) and total carotenoids(OR:0.31;p-value: <0.0001) were associated with significantly lower CD68+. Among those with high CD4+ (HR:1.77;p-value:0.03), CD68+ (HR:2.42;p-value:0.004), CD103+ (HR:3.64;p-value:0.03) and FOXP3+ (HR:3.09;p-value:0.05), having a high Western dietary pattern increased the risk of overall mortality when compared to a low Western dietary pattern. CONCLUSION: Dietary patterns and serum carotenoids may play an important role in modifying TILs, and ultimately, outcome after diagnosis with HNSCC.

A Bayesian dose-finding design for outcomes evaluated with uncertainty.

INTRODUCTION: In some phase I trial settings, there is uncertainty in assessing whether a given patient meets the criteria for dose-limiting toxicity. METHODS: We present a design which accommodates dose-limiting toxicity outcomes that are assessed with uncertainty for some patients. Our approach could be utilized in many available phase I trial designs, but we focus on the continual reassessment method due to its popularity. We assume that for some patients, instead of the usual binary dose-limiting toxicity outcome, we observe a physician-assessed probability of dose-limiting toxicity specific to a given patient. Data augmentation is used to estimate the posterior probabilities of dose-limiting toxicity at each dose level based on both the fully observed and partially observed patient outcomes. A simulation study is used to assess the performance of the design relative to using the continual reassessment method on the true dose-limiting toxicity outcomes (available in simulation setting only) and relative to simple thresholding approaches. RESULTS: Among the designs utilizing the partially observed outcomes, our proposed design has the best overall performance in terms of probability of selecting correct maximum tolerated dose and number of patients treated at the maximum tolerated dose. CONCLUSION: Incorporating uncertainty in dose-limiting toxicity assessment can improve the performance of the continual reassessment method design.

Modeling adverse event counts in phase I clinical trials of a cytotoxic agent.

Background/Aims The goal of phase I clinical trials for cytotoxic agents is to find the maximum dose with an acceptable risk of severe toxicity. The most common designs for these dose-finding trials use a binary outcome indicating whether a patient had a dose-limiting toxicity. However, a patient may experience multiple toxicities, with each toxicity assigned an ordinal severity score. The binary response is then obtained by dichotomizing a patient's richer set of data. We contribute to the growing literature on new models to exploit this richer toxicity data, with the goal of improving the efficiency in estimating the maximum tolerated dose. Methods We develop three new, related models that make use of the total number of dose-limiting and low-level toxicities a patient experiences. We use these models to estimate the probability of having at least one dose-limiting toxicity as a function of dose. In a simulation study, we evaluate how often our models select the true maximum tolerated dose, and we compare our models with the continual reassessment method, which uses binary data. Results Across a variety of simulation settings, we find that our models compare well against the continual reassessment method in terms of selecting the true optimal dose. In particular, one of our models which uses dose-limiting and low-level toxicity counts beats or ties the other models, including the continual reassessment method, in all scenarios except the one in which the true optimal dose is the highest dose available. We also find that our models, when not selecting the true optimal dose, tend to err by picking lower, safer doses, while the continual reassessment method errs more toward toxic doses. Conclusion Using dose-limiting and low-level toxicity counts, which are easily obtained from data already routinely collected, is a promising way to improve the efficiency in finding the true maximum tolerated dose in phase I trials.

NDN and CD1A are novel prognostic methylation markers in patients with head and neck squamous carcinomas.

BACKGROUND: HPV-associated HNSCCs have a distinct etiologic mechanism and better prognosis than those with non-HPV associated HNSCCs. However, even within the each group, there is heterogeneity in survival time. Here, we test the hypothesis that specific candidate gene methylation markers (CCNA1, NDN, CD1A, DCC, p16, GADD45A) are associated with tumor recurrence and survival, in a well-characterized, prospective, cohort of 346 HNSCC patients. METHODS: Kaplan-Meier curves were used to estimate survival time distributions. Multivariable Cox Proportional Hazards models were used to test associations between each methylation marker and OST/RPFT after adjusting for known or identified prognostic factors. Stratified Cox models included an interaction term between HPV and methylation marker to test for differences in the associations of the biomarker with OST or RPFT across HPV status. RESULTS: Methylation markers were differentially associated with patient characteristics. DNA hypermethylation of NDN and CD1A was found to be significantly associated with overall survival time (OST) in all HNSCC patients (NDN hazard ratio (HR): 2.35, 95% CI: 1.40-3.94; CD1A HR: 1.31, 95% CI: 1.01-1.71). Stratification by HPV status revealed hypermethylation of CD1A was associated with better OST and recurrence/persistence-free time (RPFT) (OST HR: 3.34, 95% CI: 1.88-5.93; RPFT HR: 2.06, 95% CI: 1.21-3.49), while hypomethylation of CCNA1 was associated with increased RPFT in HPV (+) patients only (HR: 0.31, 95% CI: 0.13-0.74). CONCLUSIONS: This study is the first to describe novel epigenetic alterations associated with survival in an unselected, prospectively collected, consecutive cohort of patients with HNSCC. DNA hypermethylation of NDN and CD1A was found to be significantly associated with increased overall survival time in all HNSCC patients. However, stratification by the important prognostic factor of HPV status revealed the immune marker, CD1A, and the cell cycle regulator, CCNA1 to be associated with prognosis in HPV (+) patients, specifically. Here, we identified novel methylation markers and specific, epigenetic molecular differences associated with HPV status, which warrant further investigation.

Using joint models for longitudinal and time-to-event data to investigate the causal effect of salvage therapy after prostatectomy.

Prostate cancer patients who undergo prostatectomy are closely monitored for recurrence and metastasis using routine prostate-specific antigen measurements. When prostate-specific antigen levels rise, salvage therapies are recommended in order to decrease the risk of metastasis. However, due to the side effects of these therapies and to avoid over-treatment, it is important to understand which patients and when to initiate these salvage therapies. In this work, we use the University of Michigan Prostatectomy Registry Data to tackle this question. Due to the observational nature of this data, we face the challenge that prostate-specific antigen is simultaneously a time-varying confounder and an intermediate variable for salvage therapy. We define different causal salvage therapy effects defined conditionally on different specifications of the longitudinal prostate-specific antigen history. We then illustrate how these effects can be estimated using the framework of joint models for longitudinal and time-to-event data. All proposed methodology is implemented in the freely-available R package JMbayes2.

Using information criteria to select smoothing parameters when analyzing survival data with time-varying coefficient hazard models.

Analyzing the large-scale survival data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program may help guide the management of cancer. Detecting and characterizing the time-varying effects of factors collected at the time of diagnosis could reveal important and useful patterns. However, fitting a time-varying effect model by maximizing the partial likelihood with such large-scale survival data is not feasible with most existing software. Moreover, estimating time-varying coefficients using spline based approaches requires a moderate number of knots, which may lead to unstable estimation and over-fitting issues. To resolve these issues, adding a penalty term greatly aids estimation. The selection of penalty smoothing parameters is difficult in this time-varying setting, as traditional ways like using Akaike information criterion do not work, while cross-validation methods have a heavy computational burden, leading to unstable selections. We propose modified information criteria to determine the smoothing parameter and a parallelized Newton-based algorithm for estimation. We conduct simulations to evaluate the performance of the proposed method. We find that penalization with the smoothing parameter chosen by a modified information criteria is effective at reducing the mean squared error of the estimated time-varying coefficients. Compared to a number of alternatives, we find that the estimates of the variance derived from Bayesian considerations have the best coverage rates of confidence intervals. We apply the method to SEER head-and-neck, colon, prostate, and pancreatic cancer data and detect the time-varying nature of various risk factors.

Impact of Nodal Metastases in HPV-Negative Oropharyngeal Cancer.

BACKGROUND: The updated American Joint Committee on Cancer (AJCC) 8th Edition staging manual restructured nodal classification and staging by placing less prognostic emphasis on nodal metastases for human papillomavirus (HPV)-positive oropharyngeal squamous cell carcinoma (OPSCC). However, there was no change for HPV-negative OPSCC. The purpose of our study is to examine the impact of nodal metastases on survival in HPV-negative OPSCC. METHODS: HPV-negative OPSCC was queried from the National Cancer Database (NCDB) and Surveillance, Epidemiology and End Results program (SEER) databases. Univariable and multivariable models were utilized to determine the impact of nodal status on overall survival. These patients were reclassified according to AJCC 8 HPV-positive criteria (TNM8+) and risk stratification was quantified with C-statistic. RESULTS: There were 11,147 cases of HPV-negative OPSCC in the NCDB and 3,613 cases in SEER that were included in the nodal classification analysis. Unlike nonoropharyngeal malignancies, increased nodal stage is not clearly associated with survival for patients with OPSCC independent of HPV status. When the TNM8+ was applied to HPV-negative patients, there was improved concordance in the NCDB cohort, 0.561 (plus minus) 0.004 to 0.624 (plus minus) 0.004 (difference +0.063) and the SEER cohort, 0.561 (plus minus) 0.008 to 0.625 (plus minus) 0.008 (difference +0.065). CONCLUSIONS: We demonstrated a reduced impact of nodal metastasis on OPSCC survival, independent of HPV status and specific to OPSCC. IMPACT: We demonstrate, for the first time that when nodal staging is deemphasized as a part of overall staging, we see improved concordance and risk stratification for HPV-negative OPSCC. The exact mechanism of this differential impact remains unknown but offers a novel area of study.

Using multiple imputation to classify potential outcomes subgroups.

With medical tests becoming increasingly available, concerns about over-testing, over-treatment and health care cost dramatically increase. Hence, it is important to understand the influence of testing on treatment selection in general practice. Most statistical methods focus on average effects of testing on treatment decisions. However, this may be ill-advised, particularly for patient subgroups that tend not to benefit from such tests. Furthermore, missing data are common, representing large and often unaddressed threats to the validity of most statistical methods. Finally, it is often desirable to conduct analyses that can be interpreted causally. Using the Rubin Causal Model framework, we propose to classify patients into four potential outcomes subgroups, defined by whether or not a patient's treatment selection is changed by the test result and by the direction of how the test result changes treatment selection. This subgroup classification naturally captures the differential influence of medical testing on treatment selections for different patients, which can suggest targets to improve the utilization of medical tests. We can then examine patient characteristics associated with patient potential outcomes subgroup memberships. We used multiple imputation methods to simultaneously impute the missing potential outcomes as well as regular missing values. This approach can also provide estimates of many traditional causal quantities of interest. We find that explicitly incorporating causal inference assumptions into the multiple imputation process can improve the precision for some causal estimates of interest. We also find that bias can occur when the potential outcomes conditional independence assumption is violated; sensitivity analyses are proposed to assess the impact of this violation. We applied the proposed methods to examine the influence of 21-gene assay, the most commonly used genomic test in the United States, on chemotherapy selection among breast cancer patients.

Multiple imputation with missing data indicators.

Multiple imputation is a well-established general technique for analyzing data with missing values. A convenient way to implement multiple imputation is sequential regression multiple imputation, also called chained equations multiple imputation. In this approach, we impute missing values using regression models for each variable, conditional on the other variables in the data. This approach, however, assumes that the missingness mechanism is missing at random, and it is not well-justified under not-at-random missingness without additional modification. In this paper, we describe how we can generalize the sequential regression multiple imputation imputation procedure to handle missingness not at random in the setting where missingness may depend on other variables that are also missing but not on the missing variable itself, conditioning on fully observed variables. We provide algebraic justification for several generalizations of standard sequential regression multiple imputation using Taylor series and other approximations of the target imputation distribution under missingness not at random. Resulting regression model approximations include indicators for missingness, interactions, or other functions of the missingness not at random missingness model and observed data. In a simulation study, we demonstrate that the proposed sequential regression multiple imputation modifications result in reduced bias in the final analysis compared to standard sequential regression multiple imputation, with an approximation strategy involving inclusion of an offset in the imputation model performing the best overall. The method is illustrated in a breast cancer study, where the goal is to estimate the prevalence of a specific genetic pathogenic variant.

Increasing efficiency for estimating treatment-biomarker interactions with historical data.

Detecting a treatment-biomarker interaction, which is a task better suited for large sample sizes, in a phase II trial, which has a small sample size, is challenging. In this paper, we investigate how two plausibly available sources of historical data may contain partial information to help estimate the treatment-biomarker interaction parameter in a randomized phase II study. The parameter is not identified in either historical dataset alone; nonetheless, both can provide some information about the parameter and, consequently, increase the precision of its estimate. To illustrate the potential for gains in efficiency and implications for the design of the study, we consider Gaussian outcomes and biomarker data and calculate the asymptotic variance using the expected Fisher information matrix. We quantify the gain in efficiency both through a numerical study and, in a simplified setting, insights derived from an algebraic development of the problem. We find that a non-negligible gain in precision is possible, even if the historical and prospective data do not arise from identical underlying models.

A Phase I/II trial design when response is unobserved in subjects with dose-limiting toxicity.

We propose a Phase I/II trial design in which subjects with dose-limiting toxicity are not followed for response, leading to three possible outcomes for each subject: dose-limiting toxicity, absence of therapeutic response without dose-limiting toxicity, and presence of therapeutic response without dose-limiting toxicity. We define the latter outcome as a 'success,' and the goal of the trial is to identify the dose with the largest probability of success. This dose is commonly referred to as the most successful dose. We propose a design that accumulates information on subjects with regard to both dose-limiting toxicity and response conditional on no dose-limiting toxicity. Bayesian methods are used to update the estimates of dose-limiting toxicity and response probabilities when each subject is enrolled, and we use these methods to determine the dose level assigned to each subject. Due to the need to explore doses more fully, each subject is not necessarily assigned the current estimate of the most successful dose; our algorithm may instead assign a dose that is in a neighborhood of the current most successful dose. We examine the ability of our design to correctly identify the most successful dose in a variety of settings via simulation and compare the performance of our design to that of competing approaches.

Individualized dynamic prediction of prostate cancer recurrence with and without the initiation of a second treatment: Development and validation.

With the emergence of rich information on biomarkers after treatments, new types of prognostic tools are being developed: dynamic prognostic tools that can be updated at each new biomarker measurement. Such predictions are of interest in oncology where after an initial treatment, patients are monitored with repeated biomarker data. However, in such setting, patients may receive second treatments to slow down the progression of the disease. This paper aims to develop and validate dynamic individual predictions that allow the possibility of a new treatment in order to help understand the benefit of initiating new treatments during the monitoring period. The prediction of the event in the next x years is done under two scenarios: (1) the patient initiates immediately a second treatment, (2) the patient does not initiate any treatment in the next x years. Predictions are derived from shared random-effect models. Applied to prostate cancer data, different specifications for the dependence between the prostate-specific antigen repeated measures, the initiation of a second treatment (hormonal therapy), and the risk of clinical recurrence are investigated and compared. The predictive accuracy of the dynamic predictions is evaluated with two measures (Brier score and prognostic cross-entropy) for which approximated cross-validated estimators are proposed.

Retrospective Study of Sirolimus and Cyclophosphamide in Patients with Advanced Differentiated Thyroid Cancers.

BACKGROUND: We hypothesize that the combination of an mTOR inhibitor, sirolimus, with a well-known cytotoxic agent, cyclophosphamide, provides a well-tolerated and promising alternative treatment for advanced, differentiated thyroid cancers (DTC). METHODS: This retrospective review extracted data from patients treated for advanced DTC at the University of Michigan Comprehensive Cancer Center from 1995 through 2013. Fifteen patients treated with combination sirolimus and cyclophosphamide were identified as the sirolimus+cyp group. Seventeen patients treated with standard of care and enrolled in clinical trials were identified as the comparison group. RESULTS: The one-year progression free survival rate (PFS) was 0.45, 95% CI [0.26, 0.80] in the sirolimus+cyp population and 0.30, 95% CI [0.13, 0.67] in the comparison population. The hazard ratio for PFS from initiation of treatment was 1.47, 95% CI [0.57, and 3.78]. In patients treated as first line, one-year PFS rate was 0.57, 95% CI [0.30, 1.00] in the sirolimus+cyp group and relatively unchanged at 0.29, 95% CI [0.11, 0.74] in the comparison group. The hazard ratio for PFS for first line patients was 1.10, 95% CI[ 0.4, and 3.5]. In patients with 3 or fewer sites of metastases, the one year PFS was 0.58, 95% CI [0.33, 1.00] in the sirolimus+cyp group, and 0.37, 95% CI [0.17, 0.80] in the comparison group. The average number of toxicities was 0.87 in the sirolimus+cyp patients and 1.71 in the comparison group. CONCLUSIONS: The combination of sirolimus and cyclophosphamide was generally well tolerated with similar progression free survival, highlighting its applicability in patients with limited options.

BAYESIAN SHRINKAGE METHODS FOR PARTIALLY OBSERVED DATA WITH MANY PREDICTORS.

Motivated by the increasing use of and rapid changes in array technologies, we consider the prediction problem of fitting a linear regression relating a continuous outcome Y to a large number of covariates X , eg measurements from current, state-of-the-art technology. For most of the samples, only the outcome Y and surrogate covariates, W , are available. These surrogates may be data from prior studies using older technologies. Owing to the dimension of the problem and the large fraction of missing information, a critical issue is appropriate shrinkage of model parameters for an optimal bias-variance tradeoff. We discuss a variety of fully Bayesian and Empirical Bayes algorithms which account for uncertainty in the missing data and adaptively shrink parameter estimates for superior prediction. These methods are evaluated via a comprehensive simulation study. In addition, we apply our methods to a lung cancer dataset, predicting survival time (Y) using qRT-PCR ( X ) and microarray ( W ) measurements.