Brian P. Hobbs

Improved Axillary Evaluation Following Neoadjuvant Therapy for Patients With Node-Positive Breast Cancer Using Selective Evaluation of Clipped Nodes: Implementation of Targeted Axillary Dissection

PURPOSE Placing clips in nodes with biopsy-confirmed metastasis before initiating neoadjuvant therapy allows for evaluation of response in breast cancer. Our goal was to determine if pathologic changes in clipped nodes reflect the status of the nodal basin and if targeted axillary dissection (TAD), which includes sentinel lymph node dissection (SLND) and selective localization and removal of clipped nodes, improves the false-negative rate (FNR) compared with SLND alone. METHODS A prospective study of patients with biopsy-confirmed nodal metastases with a clip placed in the sampled node was performed. After neoadjuvant therapy, patients underwent axillary surgery and the pathology of the clipped node was compared with other nodes. Patients undergoing TAD had SLND and selective removal of the clipped node using iodine-125 seed localization. The FNR was determined in patients undergoing complete axillary lymphadenectomy (ALND). RESULTS Of 208 patients enrolled in this study, 191 underwent ALND, with residual disease identified in 120 (63%). The clipped node revealed metastases in 115 patients, resulting in an FNR of 4.2% (95% CI, 1.4 to 9.5) for the clipped node. In patients undergoing SLND and ALND (n = 118), the FNR was 10.1% (95% CI, 4.2 to 19.8), which included seven false-negative events in 69 patients with residual disease. Adding evaluation of the clipped node reduced the FNR to 1.4% (95% CI, 0.03 to 7.3; P = .03). The clipped node was not retrieved as an SLN in 23% (31 of 134) of patients, including six with negative SLNs but metastasis in the clipped node. TAD followed by ALND was performed in 85 patients, with an FNR of 2.0% (1 of 50; 95% CI, 0.05 to 10.7). CONCLUSION Marking nodes with biopsy-confirmed metastatic disease allows for selective removal and improves pathologic evaluation for residual nodal disease after chemotherapy.

Hierarchical commensurate and power prior models for adaptive incorporation of historical information in clinical trials.

Bayesian clinical trial designs offer the possibility of a substantially reduced sample size, increased statistical power, and reductions in cost and ethical hazard. However when prior and current information conflict, Bayesian methods can lead to higher than expected type I error, as well as the possibility of a costlier and lengthier trial. This motivates an investigation of the feasibility of hierarchical Bayesian methods for incorporating historical data that are adaptively robust to prior information that reveals itself to be inconsistent with the accumulating experimental data. In this article, we present several models that allow for the commensurability of the information in the historical and current data to determine how much historical information is used. A primary tool is elaborating the traditional power prior approach based upon a measure of commensurability for Gaussian data. We compare the frequentist performance of several methods using simulations, and close with an example of a colon cancer trial that illustrates a linear models extension of our adaptive borrowing approach. Our proposed methods produce more precise estimates of the model parameters, in particular, conferring statistical significance to the observed reduction in tumor size for the experimental regimen as compared to the control regimen.

Commensurate Priors for Incorporating Historical Information in Clinical Trials Using General and Generalized Linear Models

Assessing between-study variability in the context of conventional random-effects meta-analysis is notoriously difficult when incorporating data from only a small number of historical studies. In order to borrow strength, historical and current data are often assumed to be fully homogeneous, but this can have drastic consequences for power and Type I error if the historical information is biased. In this paper, we propose empirical and fully Bayesian modifications of the commensurate prior model (Hobbs et al., 2011) extending Pocock (1976), and evaluate their frequentist and Bayesian properties for incorporating patient-level historical data using general and generalized linear mixed regression models. Our proposed commensurate prior models lead to preposterior admissible estimators that facilitate alternative bias-variance trade-offs than those offered by pre-existing methodologies for incorporating historical data from a small number of historical studies. We also provide a sample analysis of a colon cancer trial comparing time-to-disease progression using a Weibull regression model.

Adaptive adjustment of the randomization ratio using historical control data

Background Prospective trial design often occurs in the presence of ‘acceptable’ historical control data. Typically, these data are only utilized for treatment comparison in a posteriori retrospective analysis to estimate population-averaged effects in a random-effects meta-analysis. Purpose We propose and investigate an adaptive trial design in the context of an actual randomized controlled colorectal cancer trial. This trial, originally reported by Goldberg et al., succeeded a similar trial reported by Saltz et al., and used a control therapy identical to that tested (and found beneficial) in the Saltz trial. Methods The proposed trial implements an adaptive randomization procedure for allocating patients aimed at balancing total information (concurrent and historical) among the study arms. This is accomplished by assigning more patients to receive the novel therapy in the absence of strong evidence for heterogeneity among the concurrent and historical controls. Allocation probabilities adapt as a function of the effective historical sample size (EHSS), characterizing relative informativeness defined in the context of a piecewise exponential model for evaluating time to disease progression. Commensurate priors are utilized to assess historical and concurrent heterogeneity at interim analyses and to borrow strength from the historical data in the final analysis. The adaptive trial’s frequentist properties are simulated using the actual patient-level historical control data from the Saltz trial and the actual enrollment dates for patients enrolled into the Goldberg trial. Results Assessing concurrent and historical heterogeneity at interim analyses and balancing total information with the adaptive randomization procedure lead to trials that on average assign more new patients to the novel treatment when the historical controls are unbiased or slightly biased compared to the concurrent controls. Large magnitudes of bias lead to approximately equal allocation of patients among the treatment arms. Using the proposed commensurate prior model to borrow strength from the historical data, after balancing total information with the adaptive randomization procedure, provides admissible estimators of the novel treatment effect with desirable bias-variance trade-offs. Limitations Adaptive randomization methods in general are sensitive to population drift and more suitable for trials that initiate with gradual enrollment. Balancing information among study arms in time-to-event analyses is difficult in the presence of informative right-censoring. Conclusions The proposed design could prove important in trials that follow recent evaluations of a control therapy. Efficient use of the historical controls is especially important in contexts where reliance on preexisting information is unavoidable because the control therapy is exceptionally hazardous, expensive, or the disease is rare.

Determination of Skeletal Tumor Burden on 18F-Fluoride PET/CT

The purpose of this study was to define a method to assess skeletal tumor burden with 18F-labeled sodium fluoride PET/CT (18F-fluoride PET/CT) and evaluate the reproducibility of these measurements. Methods: Ninety-eight consecutive patients (90 men; mean age ± SD, 65.7 ± 14.2 y) underwent 158 18F-fluoride PET/CT scans for evaluation of skeletal metastatic disease. In order to determine the mean normal bone SUV, initially a 1-cm spheric volume of interest (VOI) was placed over 5 bone sites: T12, L5, sacrum, right iliac bone, and right femur. For each patient, the mean SUVmax for all sites was generated. Afterward, a threshold value of normal bone uptake was established. Subsequently, skeletal tumor burden was determined by generating volumetric data using a whole-body segmentation method. Any SUVmax below the normal threshold was excluded from analysis, as were VOIs not related to metastatic disease. Statistics for the remaining VOIs were then generated and defined as the skeletal metastatic tumor burden by 2 parameters: total lesion fluoride uptake above an SUVmax of 10 (TLF10) and fluoride tumor volume above an SUVmax of 10 (FTV10). TLF10 and FTV10 reproducibility was determined using 2 independent and experienced PET/CT interpreters analyzing a subset of 13 18F-fluoride PET/CT scans. Results: Mean (±SD) normal bone SUVmax was 6.62 ± 1.55 for T12, 6.11 ± 1.73 for L5, 4.59 ± 1.74 for sacrum, 5.39 ± 1.72 for right iliac bone, and 3.90 ± 1.57 for right femur. The mean normal SUVmax for all 543 sites was 5.32 ± 0.99. On the basis of these values, an SUVmax threshold of 10 was chosen to exclude normal bone from the volumetric calculations. Semiautomated measurements of TLF10 and FTV10 exhibited high interobserver reproducibility, within ±0.77% and ±3.62% of the interinterpreter average for TLF10 and FTV10, respectively. Conclusion: Determination of skeletal tumor burden with 18F-fluoride PET/CT is feasible and highly reproducible. Using an SUVmax threshold of 10 excludes nearly all normal bone activity from volumetric calculations.

Metastases to the Liver from Neuroendocrine Tumors: Effect of Duration of Scan Acquisition on CT Perfusion Values

PURPOSE To assess the effects of acquisition duration on computed tomographic (CT) perfusion parameter values in neuroendocrine liver metastases and normal liver tissue. MATERIALS AND METHODS This retrospective study was institutional review board approved, with waiver of informed consent. CT perfusion studies in 16 patients (median age, 57.5 years; range, 42.0-69.7 years), including six men (median, 54.1 years; range, 42.0-69.7), and 10 women (median, 59.3 years; range 43.6-66.3), with neuroendocrine liver metastases were analyzed by means of distributed parametric modeling to determine tissue blood flow, blood volume, mean transit time, permeability, and hepatic arterial fraction for tumors and normal liver tissue. Analyses were undertaken with acquisition time of 12-590 seconds. Nonparameteric regression analyses were used to evaluate the functional relationships between CT perfusion parameters and acquisition duration. Evidence for time invariance was evaluated for each parameter at multiple time points by inferring the fitted derivative to assess its proximity to zero as a function of acquisition time by using equivalence tests with three levels of confidence (20%, 70%, and 90%). RESULTS CT perfusion parameter values varied, approaching stable values with increasing acquisition duration. Acquisition duration greater than 160 seconds was required to obtain at least low confidence stability in any of the CT perfusion parameters. At 160 seconds of acquisition, all five CT perfusion parameters stabilized with low confidence in tumor and normal tissues, with the exception of hepatic arterial fraction in tumors. After 220 seconds of acquisition, there was stabilization with moderate confidence for blood flow, blood volume, and hepatic arterial fraction in tumors and normal tissue, and for mean transit time in tumors; however, permeability values did not satisfy the moderate stabilization criteria in both tumors and normal tissue until 360 seconds of acquisition. Blood flow, mean transit time, permeability, and hepatic arterial fraction were significantly different between tumor and normal tissue at 360 seconds (P < .001). CONCLUSION CT perfusion parameter values are affected by acquisition duration and approach progressively stable values with increasing acquisition times. Online supplemental material is available for this article.

Pre-radiotherapy FDG PET predicts radiation pneumonitis in lung cancer

BackgroundA retrospective analysis is performed to determine if pre-treatment [18 F]-2-fluoro-2-deoxyglucose positron emission tomography/computed tomography (FDG PET/CT) image derived parameters can predict radiation pneumonitis (RP) clinical symptoms in lung cancer patients.Methods and MaterialsWe retrospectively studied 100 non-small cell lung cancer (NSCLC) patients who underwent FDG PET/CT imaging before initiation of radiotherapy (RT). Pneumonitis symptoms were evaluated using the Common Terminology Criteria for Adverse Events version 4.0 (CTCAEv4) from the consensus of 5 clinicians. Using the cumulative distribution of pre-treatment standard uptake values (SUV) within the lungs, the 80th to 95th percentile SUV values (SUV80 to SUV95) were determined. The effect of pre-RT FDG uptake, dose, patient and treatment characteristics on pulmonary toxicity was studied using multiple logistic regression.ResultsThe study subjects were treated with 3D conformal RT (n = 23), intensity modulated RT (n = 64), and proton therapy (n = 13). Multiple logistic regression analysis demonstrated that elevated pre-RT lung FDG uptake on staging FDG PET was related to development of RP symptoms after RT. A patient of average age and V30 with SUV95 = 1.5 was an estimated 6.9 times more likely to develop grade ≥ 2 radiation pneumonitis when compared to a patient with SUV95 = 0.5 of the same age and identical V30. Receiver operating characteristic curve analysis showed the area under the curve was 0.78 (95% CI = 0.69 – 0.87). The CT imaging and dosimetry parameters were found to be poor predictors of RP symptoms.ConclusionsThe pretreatment pulmonary FDG uptake, as quantified by the SUV95, predicted symptoms of RP in this study. Elevation in this pre-treatment biomarker identifies a patient group at high risk for post-treatment symptomatic RP.

From Protocols to Publications: A Study in Selective Reporting of Outcomes in Randomized Trials in Oncology

PURPOSE The decision by journals to append protocols to published reports of randomized trials was a landmark event in clinical trial reporting. However, limited information is available on how this initiative effected transparency and selective reporting of clinical trial data. METHODS We analyzed 74 oncology-based randomized trials published in Journal of Clinical Oncology, the New England Journal of Medicine, and The Lancet in 2012. To ascertain integrity of reporting, we compared published reports with their respective appended protocols with regard to primary end points, nonprimary end points, unplanned end points, and unplanned analyses. RESULTS A total of 86 primary end points were reported in 74 randomized trials; nine trials had greater than one primary end point. Nine trials (12.2%) had some discrepancy between their planned and published primary end points. A total of 579 nonprimary end points (median, seven per trial) were planned, of which 373 (64.4%; median, five per trial) were reported. A significant positive correlation was found between the number of planned and nonreported nonprimary end points (Spearman r = 0.66; P < .001). Twenty-eight studies (37.8%) reported a total of 65 unplanned end points; 52 (80.0%) of which were not identified as unplanned. Thirty-one (41.9%) and 19 (25.7%) of 74 trials reported a total of 52 unplanned analyses involving primary end points and 33 unplanned analyses involving nonprimary end points, respectively. Studies reported positive unplanned end points and unplanned analyses more frequently than negative outcomes in abstracts (unplanned end points odds ratio, 6.8; P = .002; unplanned analyses odd ratio, 8.4; P = .007). CONCLUSION Despite public and reviewer access to protocols, selective outcome reporting persists and is a major concern in the reporting of randomized clinical trials. To foster credible evidence-based medicine, additional initiatives are needed to minimize selective reporting.

Evaluation of 4D CT acquisition methods designed to reduce artifacts

Four‐dimensional computed tomography (4D CT) is used to account for respiratory motion in radiation treatment planning, but artifacts resulting from the acquisition and postprocessing limit its accuracy. We investigated the efficacy of three experimental 4D CT acquisition methods to reduce artifacts in a prospective institutional review board approved study. Eighteen thoracic patients scheduled to undergo radiation therapy received standard clinical 4D CT scans followed by each of the alternative 4D CT acquisitions: 1) data oversampling, 2) beam gating with breathing irregularities, and 3) rescanning the clinical acquisition acquired during irregular breathing. Relative values of a validated correlation‐based artifact metric (CM) determined the best acquisition method per patient. Each 4D CT was processed by an extended phase sorting approach that optimizes the quantitative artifact metric (CM sorting). The clinical acquisitions were also postprocessed by phase sorting for artifact comparison of our current clinical implementation with the experimental methods. The oversampling acquisition achieved the lowest artifact presence among all acquisitions, achieving a 27% reduction from the current clinical 4D CT implementation (95% confidence interval=34−20). The rescan method presented a significantly higher artifact presence from the clinical acquisition (37%; p<0.002), the gating acquisition (26%; p<0.005), and the oversampling acquisition (31%; p<0.001), while the data lacked evidence of a significant difference between the clinical, gating, and oversampling methods. The oversampling acquisition reduced artifact presence from the current clinical 4D CT implementation to the largest degree and provided the simplest and most reproducible implementation. The rescan acquisition increased artifact presence significantly, compared to all acquisitions, and suffered from combination of data from independent scans over which large internal anatomic shifts occurred. PACS numbers: 87.57.C‐, 87.57.cp, 87.57.Q‐, 87.55.Gh

Semiparametric Bayesian commensurate survival model for post-market medical device surveillance with non-exchangeable historical data

Trial investigators often have a primary interest in the estimation of the survival curve in a population for which there exists acceptable historical information from which to borrow strength. However, borrowing strength from a historical trial that is non-exchangeable with the current trial can result in biased conclusions. In this article we propose a fully Bayesian semiparametric method for the purpose of attenuating bias and increasing efficiency when jointly modeling time-to-event data from two possibly non-exchangeable sources of information. We illustrate the mechanics of our methods by applying them to a pair of post-market surveillance datasets regarding adverse events in persons on dialysis that had either a bare metal or drug-eluting stent implanted during a cardiac revascularization surgery. We finish with a discussion of the advantages and limitations of this approach to evidence synthesis, as well as directions for future work in this area. The articles Supplementary Materials offer simulations to show our procedures bias, mean squared error, and coverage probability properties in a variety of settings.