Results of a multicenter phase I/II trial of TCRαß and CD19-depleted haploidentical hematopoietic stem cell transplantation for adult and pediatric patients.

Hematopoietic stem cell transplantation (HSCT) from haploidentical donors is a viable option for patients lacking HLA-matched donors. Here we report the results of a prospective multicenter phase I/II trial of transplantation of TCRαß and CD19-depleted peripheral blood stem cells from haploidentical family donors after a reduced-intensity conditioning with fludarabine, thiotepa, and melphalan. Thirty pediatric and 30 adult patients with acute leukemia (n = 43), myelodysplastic or myeloproliferative syndrome (n = 6), multiple myeloma (n = 1), solid tumors (n = 6), and non-malignant disorders (n = 4) were enrolled. TCR αß/CD19-depleted grafts prepared decentrally at six manufacturing sites contained a median of 12.1 × 106 CD34+ cells/kg and 14.2 × 103 TCRαß+ T-cells/kg. None of the patients developed grade lll/IV acute graft-versus-host disease (GVHD) and only six patients (10%) had grade II acute GVHD. With a median follow-up of 733 days 36/60 patients are alive. The cumulative incidence of non-relapse mortality at day 100, 1 and 2 years after HSCT was 5%, 15%, and 17% for all patients, respectively. Estimated probabilities of overall and disease-free survival at 2 years were 63% and 50%, respectively. Based on these promising results in a high-risk patient cohort, haploidentical HSCT using TCRαß/CD19-depleted grafts represents a viable treatment option.

D-optimal designs for multiarm trials with dropouts.

Multiarm trials with follow-up on participants are commonly implemented to assess treatment effects on a population over the course of the studies. Dropout is an unavoidable issue especially when the duration of the multiarm study is long. Its impact is often ignored at the design stage, which may lead to less accurate statistical conclusions. We develop an optimal design framework for trials with repeated measurements, which takes potential dropouts into account, and we provide designs for linear mixed models where the presence of dropouts is noninformative and dependent on design variables. Our framework is illustrated through redesigning a clinical trial on Alzheimer's disease, whereby the benefits of our designs compared with standard designs are demonstrated through simulations.

Simultaneous confidence sets for several effective doses.

Construction of simultaneous confidence sets for several effective doses currently relies on inverting the Scheffé type simultaneous confidence band, which is known to be conservative. We develop novel methodology to make the simultaneous coverage closer to its nominal level, for both two-sided and one-sided simultaneous confidence sets. Our approach is shown to be considerably less conservative than the current method, and is illustrated with an example on modeling the effect of smoking status and serum triglyceride level on the probability of the recurrence of a myocardial infarction.

Reproducibility and reliability of the ankle-brachial index as assessed by vascular experts, family physicians and nurses.

The reliability of ankle-brachial index (ABI) measurements performed by different observer groups in primary care has not yet been determined. The aims of the study were to provide precise estimates for all effects influencing the variability of the ABI (patients' individual variability, intra- and inter-observer variability), with particular focus on the performance of different observer groups. Using a partially balanced incomplete block design, 144 unselected individuals aged > or = 65 years underwent double ABI measurements by one vascular surgeon or vascular physician, one family physician and one nurse with training in Doppler sonography. Three groups comprising a total of 108 individuals were analyzed (only two with ABI < 0.90). Errors for two repeated measurements for all three observer groups did not differ (experts 8.5%, family physicians 7.7%, and nurses 7.5%, p = 0.39). There was no relevant bias among observer groups. Intra-observer variability expressed as standard deviation divided by the mean was 8%, and inter-observer variability was 9%. In conclusion, reproducibility of the ABI measurement was good in this cohort of elderly patients who almost all had values in the normal range. The mean error of 8-9% within or between observers is smaller than with established screening measures. Since there were no differences among observers with different training backgrounds, our study confirms the appropriateness of ABI assessment for screening peripheral arterial disease (PAD) and generalized atherosclerosis in the primary case setting. Given the importance of the early detection and management of PAD, this diagnostic tool should be used routinely as a standard for PAD screening. Additional studies will be required to confirm our observations in patients with PAD of various severities.