The Impact of EBV Status on Characteristics and Outcomes of Posttransplantation Lymphoproliferative Disorder.

We examined the associations of Epstein-Barr virus (EBV) status with characteristics and outcomes of posttransplantation lymphoproliferative disorder (PTLD) by studying 176 adult solid organ transplant recipients diagnosed with PTLD between 1990 and 2013 (58 [33%] EBV-negative; 118 [67%] EBV-positive). The proportion of EBV-negative cases increased over time from 10% (1990-1995) to 48% (2008-2013) (p < 0.001). EBV-negative PTLD had distinct characteristics (monomorphic histology, longer latency) though high-risk features (advanced stage, older age, high lactate dehydrogenase, central nervous system involvement) were not more common compared to EBV-positive PTLD. In multivariable analysis, EBV negativity was not significantly associated with worse response to initial therapy (adjusted odds ratio, 0.84; p = 0.75). The likelihood of achieving a complete remission (CR) was not significantly different for EBV-negative versus EBV-positive PTLD including when therapy was reduction of immunosuppression alone (35% vs. 43%, respectively, p = 0.60) or rituximab (43% vs. 47%, p = 1.0). EBV negativity was also not associated with worse overall survival (adjusted hazard ratio, 0.91; p = 0.71). Our findings indicate that EBV status is not prognostic or predictive of treatment response in adults with PTLD. The high proportion of EBV-negative disease diagnosed in recent years highlights the need for new strategies for prevention and management of EBV-negative PTLD.

High-dose corticosteroids with or without etanercept for the treatment of idiopathic pneumonia syndrome after allo-SCT.

Idiopathic Pneumonia Syndrome (IPS) is a common complication after allo-SCT and results in high mortality rates. Conventional treatment for IPS typically includes supportive care and high-dose corticosteroids (CS). Data suggests that TNF-α is important in the pathogenesis of IPS and that the TNF-α inhibitor etanercept may be useful for IPS treatment. We performed a retrospective comparison of consecutive patients treated at our center for IPS with CS only from 1999 to 2003 (group 1, n=22) or CS plus etanercept from 2004 to 2007 (group 2, n=17). In all, 18% of patients in group 1 vs 53% in group 2 were successfully taken off respiratory support and discharged from the hospital (P=0.039). OS was significantly better for recipients of CS plus etanercept (P=0.003). The estimated survival at 28 days and 2 years after IPS was 36.4% (95% CI 17-56%) and 9.1% (95% CI 2-25%) for group 1 and 88.2% (95% CI 61-97%) and 18% (95% CI 4-38%) for group 2, respectively. Our retrospective comparison suggests that the addition of etanercept to CS for IPS improves response rates and OS. However, outcomes remain limited in both groups, highlighting the need for more effective interventions to treat early and late complications of IPS.

Association of HLA polymorphisms with post-transplant lymphoproliferative disorder in solid-organ transplant recipients.

The association between HLA polymorphisms and PTLD was investigated in a case-control study, comparing 110 predominantly adult solid-organ transplant recipients who developed PTLD to 5601 who did not. Donor and recipient HLA were analyzed. We detected a significant association between recipient HLA-A26 and the development of PTLD (OR 2.74; p = 0.0007). In Caucasian recipients, both recipient and donor HLA-A26 were independently associated with development of PTLD (recipient A26 OR 2.99; p = 0.0004, donor A26 OR 2.81; p = 0.002). Analysis of HLA-A and -B haplotypes revealed that recipient HLA-A26, B38 haplotype was strongly correlated with a higher incidence of EBV-positive PTLD (OR 3.99; p = 0.001). The common ancestral haplotype HLA-A1, B8, DR3, when carried by the donor, was protective against PTLD (OR 0.41; p = 0.05). Several other HLA specificities demonstrated associations with clinical and pathological characteristics as well as survival. These findings demonstrate the importance of HLA polymorphisms in modulating the risk for PTLD, and may be useful in risk stratification and development of monitoring and prophylaxis strategies.

Reduction of immunosuppression as initial therapy for posttransplantation lymphoproliferative disorder(☠).

Reduction of immunosuppression (RI) is commonly used to treat posttransplant lymphoproliferative disorder (PTLD) in solid organ transplant recipients. We investigated the efficacy, safety and predictors of response to RI in adult patients with PTLD. Sixty-seven patients were managed with RI alone and 30 patients were treated with surgical excision followed by adjuvant RI. The response rate to RI alone was 45% (complete response-37%, partial response-8%). The relapse rate in complete responders was 17%. Adjuvant RI resulted in a 27% relapse rate. The acute rejection rate following RI-containing strategies was 32% and a second transplant was feasible without relapse of PTLD. The median survival was 44 months in patients treated with RI alone and 9.5 months in patients who remained on full immunosuppression (p = 0.07). Bulky disease, advanced stage and older age predicted lack of response to RI. Survival analysis demonstrated predictors of poor outcome-age, dyspnea, B symptoms, LDH level, hepatitis C, bone marrow and liver involvement. Patients with none or one of these factors had a 3-year overall survival of 100% and 79%, respectively. These findings support the use of RI alone in low-risk PTLD and suggest factors that predict response and survival.

Cost-effectiveness of pharmacogenetic testing to tailor smoking-cessation treatment.

We evaluated the cost-effectiveness of a range of smoking cessation drug treatments, including varenicline, transdermal nicotine (TN), bupropion and the use of a genetic test to choose between TN and bupropion. We performed Monte Carlo simulation with sensitivity analysis, informing analyses with published estimates of model parameters and current prices for genetic testing and smoking-cessation therapy. The primary outcomes were discounted life-years (LY) and lifetime tobacco-cessation treatment costs. In the base case, varenicline treatment was optimal with an ICER, compared to bupropion, of $2985/LY saved. In sensitivity analyses, varenicline was in all cases (and bupropion in most cases) admissible; only under favorable assumptions was the genetically tailored approach competitive. Our data suggest that an untailored approach of treatment with either bupropion or varenicline is a cost-effective form of tobacco dependence treatment, but a tailored approach for selecting between TN and bupropion can be cost-effective under plausible assumptions.

Predicting analysis times in randomized clinical trials.

Randomized clinical trial designs commonly include one or more planned interim analyses. At these times an external monitoring committee reviews the accumulated data and determines whether it is scientifically and ethically appropriate for the study to continue. With failure-time endpoints, it is common to schedule analyses at the times of occurrence of specified landmark events, such as the 50th event, the 100th event, and so on. Because interim analyses can impose considerable logistical burdens, it is worthwhile predicting their timing as accurately as possible. We describe two model-based methods for making such predictions during the course of a trial. First, we obtain a point prediction by extrapolating the cumulative mortality into the future and selecting the date when the expected number of deaths is equal to the landmark number. Second, we use a Bayesian simulation scheme to generate a predictive distribution of milestone times; prediction intervals are quantiles of this distribution. We illustrate our method with an analysis of data from a trial of immunotherapy in the treatment of chronic granulomatous disease.

Estimating the effect of an intensive surveillance program on stage of breast carcinoma at diagnosis: a propensity score analysis.

BACKGROUND: The authors used propensity score adjustment to investigate the impact of intensive screening on stage of breast carcinoma at diagnosis in women who were at elevated risk for breast carcinoma. METHODS: The authors compared 58 women participating in a surveillance program at the Columbia Presbyterian Medical Center of New York Presbyterian Hospital who developed breast carcinoma with 3022 nonparticipating breast carcinoma patients. A propensity score was constructed for each woman by using important background covariates, and multivariable regression modeling was used to estimate the association of program membership with disease stage after adjusting for the propensity score. RESULTS: Before propensity score adjustment, nine baseline covariates significantly differed between the two groups (number of pregnancies, number of births, age at first delivery, race, how the tumor was discovered, history of prior breast disease, breast carcinoma in mother, breast carcinoma in maternal aunt, and breast carcinoma in sister), and there was a significant difference in stage at diagnosis. After adjustment, no significant differences remained. Program participants were more likely to have lower stage tumors at diagnosis than nonparticipants, but this association did not reach statistical significance (odds ratio, 1.52; 95% confidence interval, 0.94--2.46). CONCLUSIONS: Propensity score methods can remove bias in treatment comparisons in observational studies. An intensive surveillance program at a major cancer center may have had some effect on improving stage at diagnosis, but this effect was not statistically significant.

Ascertainment and anticipation in family studies.

Many human diseases show anticipation; that is, disease occurs earlier (or with greater severity) in successive generations. In a computer simulation, we assessed the degree of anticipation that one would expect to see in two-generation breast cancer families. Under reasonable assumed distributions for age at cancer onset, number of children, and mortality, we find a consistent earlier mean age at diagnosis in daughters than in mothers, but the same mean age at diagnosis in affected aunts and nieces. We compare these results with published pedigree data for familial breast cancer that show substantial anticipation in affected daughters compared to their mothers. We find that at least some anticipation is expected in human disease families even when the disease is stable and families are ascertained without obvious sampling bias. We further demonstrate that such anticipation is reduced when comparing affected children to the parents' affected siblings.

Long-term use of a left ventricular assist device for end-stage heart failure.

BACKGROUND: Implantable left ventricular assist devices have benefited patients with end-stage heart failure as a bridge to cardiac transplantation, but their long-term use for the purpose of enhancing survival and the quality of life has not been evaluated. METHODS: We randomly assigned 129 patients with end-stage heart failure who were ineligible for cardiac transplantation to receive a left ventricular assist device (68 patients) or optimal medical management (61). All patients had symptoms of New York Heart Association class IV heart failure. RESULTS: Kaplan-Meier survival analysis showed a reduction of 48 percent in the risk of death from any cause in the group that received left ventricular assist devices as compared with the medical-therapy group (relative risk, 0.52; 95 percent confidence interval, 0.34 to 0.78; P=0.001). The rates of survival at one year were 52 percent in the device group and 25 percent in the medical-therapy group (P=0.002), and the rates at two years were 23 percent and 8 percent (P=0.09), respectively. The frequency of serious adverse events in the device group was 2.35 (95 percent confidence interval, 1.86 to 2.95) times that in the medical-therapy group, with a predominance of infection, bleeding, and malfunction of the device. The quality of life was significantly improved at one year in the device group. CONCLUSIONS: The use of a left ventricular assist device in patients with advanced heart failure resulted in a clinically meaningful survival benefit and an improved quality of life. A left ventricular assist device is an acceptable alternative therapy in selected patients who are not candidates for cardiac transplantation.

Decision analysis of tamoxifen for the prevention of invasive breast cancer.

PURPOSE: The recent Breast Cancer Prevention Trial has shown that tamoxifen may prevent invasive breast cancer. We used a Markov model to estimate the long-term effects of chemoprevention with tamoxifen on survival, quality-adjusted survival, and health care costs. METHODS: We used a hypothetical cohort of women with breast-cancer risk similar to that of participants in the Breast Cancer Prevention Trial, and a computer-based decision analysis (Markov model and 500 Monte Carlo simulations) to model the outcomes of interest. Survival calculations were from Surveillance, Epidemiology, and End-Results (SEER) data; preference ratings from a time trade-off questionnaire administered to a group of average-risk women; and cost estimates from the Group Health Cooperative of Puget Sound and the Health Care Financing Administration. We obtained utility measures for quality-adjustment by administering a time trade-off questionnaire to a group of community-based women. RESULTS: Use of tamoxifen prolonged the average survival of cohort members by 69 days (95% probability interval [PI] 27 to 117) for those who started use at age 35 years; 40 days (95% PI 16 to 67) for those who started use at age 50 years; and 27 days (95% PI 14 to 40) for those who started use at age 60 years. Tamoxifen extended quality-adjusted survival by 38 days (95% PI 0.1 to 82) at age 35, 25 days (95% PI 0 to 50) at age 50, and 22 days (95% PI 5 to 39) days at age 60. Chemoprevention with tamoxifen cost $46,619 (95% PI $27,928 to $98,796) per life year life saved for women who started at age 35; for women over age 50, it cost more than $50,000 per life year saved. DISCUSSION: Tamoxifen use may improve long-term survival and quality-adjusted survival among women who are at increased risk of breast cancer, but this benefit diminishes with age. Tamoxifen is cost-effective in comparison with other cancer treatment strategies for younger women only.

Fieller's method and net health benefits.

Statistical and conceptual difficulties complicate the estimation of the incremental cost-effectiveness ratio (ICER). An alternative approach is to measure cost-effectiveness by the incremental net health benefit (INHB), defined as the difference in mean effectiveness of a new treatment compared with a standard, adjusted for cost difference by subtracting the health foregone if purchasing care at the rate of a marginally cost-effective therapy. Because net health benefit (NHB) is dependent on this threshold rate, one can construct confidence intervals for the INHB at various values of the rate. It turns out that the set of rates where new and standard are not significantly different is equal to the Fieller's method confidence set for the ICER. We review the derivation of the Fieller's method confidence set, present numerical examples, and discuss the implications of our result for the calculation and interpretation of NHB analyses.

Prevention with tamoxifen or other hormones versus prophylactic surgery in BRCA1/2-positive women: a decision analysis.

PURPOSE: Recent randomized controlled trials have shown that tamoxifen and raloxifene may prevent invasive breast cancer. This decision analysis study compares the outcomes of chemoprevention with tamoxifen, raloxifene, or oral contraceptives with the outcomes of prophylactic surgery among women with high-risk BRCA1/2 mutations. PATIENTS AND METHODS: We used a simulated cohort of 30-year-old women who tested positive for BRCA1/2 mutations and constructed a Markov model with Monte Carlo simulations, incorporating cumulative breast and ovarian cancer incidence rates from the literature and survival figures from SEER data. We assumed that prophylactic surgery reduces ovarian cancer risk by 45% and breast cancer risk by 90%, that tamoxifen reduces invasive breast cancer risk by 49%, and that raloxifene has similar efficacy and safety in premenopausal and postmenopausal women. We used data obtained from high-risk women by a time trade-off questionnaire to calculate quality-adjusted life-years. We based our cost estimates for hospital and ambulatory care on Health Care Financing Administration payments, the SEER-HCFA database, and the Pharmacy Fundamental Reference. RESULTS: In our model, a 30-year-old BRCA1/2+ woman could prolong survival by 0.9 years (95% probability interval, 0.4-1.2 years) by having bilateral oophorectomy, 3.4 years (2.7-3.7 years) by having bilateral mastectomy, and 4.3 years (3.6-4.6 years) by having both procedures instead of surveillance alone. Chemoprevention with tamoxifen and raloxifene increased survival by 1.6 years (1.0-2.1 years) and 2.2 years (1.3-2.8 years), respectively. Chemoprevention yielded more quality-adjusted life-years than did prophylactic surgery, even when treatment was delayed to age 40 or 50 years. All these procedures were cost-effective or cost-saving compared with surveillance alone. DISCUSSION: Our model suggests that although surgery may yield more substantial survival and cost benefits, quality of life issues may make chemoprevention a more attractive option for young women at high genetic risk.

Mixed-effects models in psychophysiology.

The current methodological policy in Psychophysiology stipulates that repeated-measures designs be analyzed using either multivariate analysis of variance (ANOVA) or repeated-measures ANOVA with the Greenhouse-Geisser or Huynh-Feldt correction. Both techniques lead to appropriate type I error probabilities under general assumptions about the variance-covariance matrix of the data. This report introduces mixed-effects models as an alternative procedure for the analysis of repeated-measures data in Psychophysiology. Mixed-effects models have many advantages over the traditional methods: They handle missing data more effectively and are more efficient, parsimonious, and flexible. We described mixed-effects modeling and illustrated its applicability with a simple example.

Ignorability and bias in clinical trials.

Patient non-compliance and drop-out can bias analyses of clinical trial data. I describe a parametric model for treatment cross-over and drop-out and demonstrate how the concept of ignorability, originally defined for incomplete-data problems, can elucidate sources of bias in clinical trials. I discuss some implications of the theory and present simulation examples that illustrate the potential effects of non-ignorable cross-over and drop-out on bias and power.

Causal inference in a clinical trial: a comparative example.

Recently there has been much interest in methods for analyzing clinical trials of treatments that are subject to noncompliance. In this paper I study a small, simple dataset from a clinical trial of immunosuppressive therapy in the treatment of multiple sclerosis. I apply and compare a range of methods: the as-randomized (intention-to-treat) analysis, the as-treated analysis, estimates based on a nonignorable selection model, and Rubin's causal model. The results differ substantially even in this small dataset that exhibits modest noncompliance. For this reason, data analysts should be clear about which parameters are of greatest importance in the analysis of a clinical trial.

Geographic variations in the rates of operative procedures involving the shoulder, including total shoulder replacement, humeral head replacement, and rotator cuff repair.

BACKGROUND: Although geographic variations in the rates of orthopaedic procedures have been well documented, considerable controversy remains regarding the factors that drive these variations, particularly the role of the availability of orthopaedic surgeons. Moreover, little attention has been specifically focused on variations in the rates of commonly performed shoulder procedures. METHODS: The current study documents state-to-state variations in the rates of total shoulder replacement, humeral head replacement, and rotator cuff repair and examines factors that might account for these variations. The regional incidences of these three procedures were analyzed with use of the Health Care Financing Administration Medicare database (MEDPAR, 1992). The rates were age-adjusted, and variations were measured with use of high:low ratios, variation coefficients, and systematic components of variation. Potential causes of variation were analyzed with use of Spearman and partial correlations as well as with Poisson regression. RESULTS: Rates for the three procedures that were studied varied from one state to another by as much as tenfold. Humeral head replacement had the lowest rate of variation according to all three measures. All three procedures were performed less often in states that were more densely populated. With the numbers available for study, no consistent, significant relationship was found between the density of orthopaedists and shoulder surgeons and the rates of any procedure. CONCLUSIONS: The striking variations that were noted for these commonly performed procedures showed that there is a clear need for well designed clinical research to further define the factors that account for the variations and to examine the effectiveness and appropriate indications for the procedures.

Bayesian estimation of cost-effectiveness ratios from clinical trials.

Estimation of the incremental cost-effectiveness ratio (ICER) is difficult for several reasons: treatments that decrease both cost and effectiveness and treatments that increase both cost and effectiveness can yield identical values of the ICER; the ICER is a discontinuous function of the mean difference in effectiveness; and the standard estimate of the ICER is a ratio. To address these difficulties, we have developed a Bayesian methodology that involves computing posterior probabilities for the four quadrants and separate interval estimates of ICER for the quadrants of interest. We compute these quantities by simulating draws from the posterior distribution of the cost and effectiveness parameters and tabulating the appropriate posterior probabilities and quantiles. We demonstrate the method by re-analysing three previously published clinical trials.

The REMATCH trial: rationale, design, and end points. Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure.

BACKGROUND: Because left ventricular assist devices have recently been approved by the Food and Drug Administration to support the circulation of patients with end-stage heart failure awaiting cardiac transplantation, these devices are increasingly being considered as a potential alternative to biologic cardiac replacement. The Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure (REMATCH) trial is a multicenter study supported by the National Heart, Lung, and Blood Institute to compare long-term implantation of left ventricular assist devices with optimal medical management for patients with end-stage heart failure who require, but do not qualify to receive cardiac transplantation. METHODS: We discuss the rationale for conducting REMATCH, the obstacles to designing this and other randomized surgical trials, the lessons learned in conducting the multicenter pilot study, and the features of the REMATCH study design (objectives, target population, treatments, end points, analysis, and trial organization). CONCLUSIONS: We consider what will be learned from REMATCH, expectations for expanding the use of left ventricular assist devices, and future directions for assessing clinical procedures.

Benefits and costs of screening Ashkenazi Jewish women for BRCA1 and BRCA2.

PURPOSE: To determine the survival benefit and cost-effectiveness of screening Ashkenazi Jewish women for three specific BRCA1/2 gene mutations. METHODS: We used a Markov model and Monte Carlo analysis to estimate the survival benefit and cost-effectiveness of screening for three specific mutations in a population in which their prevalence is 2.5% and the associated cancer risks are 56% for breast cancer and 16% for ovarian cancer. We assumed that the sensitivity and specificity of the test were 98% and 99%, respectively, that bilateral prophylactic oophorectomy would reduce ovarian cancer risk by 45%, and that bilateral prophylactic mastectomy would reduce breast cancer risk by 90%. We used Medicare payment data for treatment costs and Surveillance, Epidemiology, and End Results data for cancer survival. RESULTS: Our model suggests that genetic screening of this population could prolong average nondiscounted survival by 38 days (95% probability interval, 22 to 57 days) for combined surgery, 33 days (95% probability interval, 18 to 43 days) for mastectomy, 11 days (95% probability interval, 4 to 25 days) for oophorectomy, and 6 days (95% probability interval, 3 to 8 days) for surveillance. The respective cost-effectiveness ratios per life-year saved, with a discount rate of 3%, are $20,717, $29,970, $72,780, and $134,273. CONCLUSION: In this Ashkenazi Jewish population, with a high prevalence of BRCA1/2 mutations, genetic screening may significantly increase average survival and, depending on costs and screening/treatment strategies, may be cost-effective by the standards of accepted cancer screening tests. According to our model, screening is cost-effective only if all women who test positive undergo prophylactic surgery. These estimates require confirmation through prospective observational studies and clinical trials.