Statistical rules for safety monitoring in clinical trials

BACKGROUND/AIMS

Protecting patient safety is an essential component of the conduct of clinical trials. Rigorous safety monitoring schemes are implemented for these studies to guard against excess toxicity risk from study therapies. They often include protocol-specified stopping rules dictating that an excessive number of safety events will trigger a halt of the study. Statistical methods are useful for constructing rules that protect patients from exposure to excessive toxicity while also maintaining the chance of a false safety signal at a low level. Several statistical techniques have been proposed for this purpose, but the current literature lacks a rigorous comparison to determine which method may be best suitable for a given trial design. The aims of this article are (1) to describe a general framework for repeated monitoring of safety events in clinical trials; (2) to survey common statistical techniques for creating safety stopping criteria; and (3) to provide investigators with a software tool for constructing and assessing these stopping rules.

METHODS

The properties and operating characteristics of stopping rules produced by Pocock and O'Brien-Fleming tests, Bayesian Beta-Binomial models, and sequential probability ratio tests (SPRTs) are studied and compared for common scenarios that may arise in phase II and III trials. We developed the R package "stoppingrule" for constructing and evaluating stopping rules from these methods. Its usage is demonstrated through a redesign of a stopping rule for BMT CTN 0601 (registered at Clinicaltrials.gov as NCT00745420), a phase II, single-arm clinical trial that evaluated outcomes in pediatric sickle cell disease patients treated by bone marrow transplant.

RESULTS

Methods with aggressive stopping criteria early in the trial, such as the Pocock test and Bayesian Beta-Binomial models with weak priors, have permissive stopping criteria at late stages. This results in a trade-off where rules with aggressive early monitoring generally will have a smaller number of expected toxicities but also lower power than rules with more conservative early stopping, such as the O-Brien-Fleming test and Beta-Binomial models with strong priors. The modified SPRT method is sensitive to the choice of alternative toxicity rate. The maximized SPRT generally has a higher number of expected toxicities and/or worse power than other methods.

CONCLUSIONS

Because the goal is to minimize the number of patients exposed to and experiencing toxicities from an unsafe therapy, we recommend using the Pocock or Beta-Binomial, weak prior methods for constructing safety stopping rules. At the design stage, the operating characteristics of candidate rules should be evaluated under various possible toxicity rates in order to guide the choice of rule(s) for a given trial; our R package facilitates this evaluation.

SARS-CoV-2 vaccination in the first year after hematopoietic cell transplant or chimeric antigen receptor T cell therapy: A prospective, multicenter, observational study (BMT CTN 2101)

Background

The optimal timing of vaccination with SARS-CoV-2 vaccines after cellular therapy is incompletely understood.

Objective

To describe humoral and cellular responses after SARS-CoV-2 vaccination initiated <4 months versus 4-12 months after cellular therapy.

Design

Multicenter prospective observational study.

Setting

34 centers in the United States.

Participants

466 allogeneic hematopoietic cell transplant (HCT; n=231), autologous HCT (n=170), or chimeric antigen receptor T cell (CAR-T cell) therapy (n=65) recipients enrolled between April 2021 and June 2022.

Interventions

SARS-CoV-2 vaccination as part of routine care.

Measurements

We obtained blood prior to and after vaccinations at up to five time points and tested for SARS-CoV-2 spike (anti-S) IgG in all participants and neutralizing antibodies for Wuhan D614G, Delta B.1.617.2, and Omicron B.1.1.529 strains, as well as SARS-CoV-2-specific T cell receptors (TCRs), in a subgroup.

Results

Anti-S IgG and neutralizing antibody responses increased with vaccination in HCT recipients irrespective of vaccine initiation timing but were unchanged in CAR-T cell recipients initiating vaccines within 4 months. Anti-S IgG ≥2,500 U/mL was correlated with high neutralizing antibody titers and attained by the last time point in 70%, 69%, and 34% of allogeneic HCT, autologous HCT, and CAR-T cell recipients, respectively. SARS-CoV-2-specific T cell responses were attained in 57%, 83%, and 58%, respectively. Humoral and cellular responses did not significantly differ among participants initiating vaccinations <4 months vs 4-12 months after cellular therapy. Pre-cellular therapy SARS-CoV-2 infection or vaccination were key predictors of post-cellular therapy anti-S IgG levels.

Limitations

The majority of participants were adults and received mRNA vaccines.

Conclusions

These data support starting mRNA SARS-CoV-2 vaccination three to four months after allogeneic HCT, autologous HCT, and CAR-T cell therapy.

Funding

National Marrow Donor Program, Leukemia and Lymphoma Society, Multiple Myeloma Research Foundation, Novartis, LabCorp, American Society for Transplantation and Cellular Therapy, Adaptive Biotechnologies, and the National Institutes of Health.

Incidence and Impact of Fungal Infections in Post-Transplantation Cyclophosphamide-Based Graft-versus-Host Disease Prophylaxis and Haploidentical Hematopoietic Cell Transplantation: A Center for International Blood and Marrow Transplant Research Analysis

Fungal infection (FI) after allogeneic hematopoietic cell transplantation (HCT) is associated with increased morbidity and mortality. Neutropenia, HLA mismatch, graft-versus-host disease (GVHD), and viral infections are risk factors for FI. The objectives of this Center for International Blood and Marrow Transplant Research registry study were to compare the incidence and density of FI occurring within 180 days after HCT in matched sibling (Sib) transplants with either calcineurin inhibitor (CNI)-based or post-transplantation cyclophosphamide (PTCy)-based GVHD prophylaxis and related haploidentical transplants receiving PTCy, and to examine the impact of FI by day 180 on transplantation outcomes.

METHODS

Patients who underwent their first HCT between 2012 and 2017 for acute myeloid leukemia, acute lymphoblastic leukemia, and myelodysplastic syndrome and received a related haploidentical transplant with PTCy (HaploCy; n = 757) or a Sib transplant with PTCy (SibCy; n = 403) or CNI (SibCNI; n = 1605) were analyzed. The incidence of FI by day 180 post-HCT was calculated as cumulative incidence with death as the competing risk. The associations of FI with overall survival, transplant-related mortality, chronic GVHD, and relapse at 2 years post-HCT were examined in Cox proportional hazards regression models. Factors significantly associated with the outcome variable at a 1% level were kept in the final model.

RESULTS

By day 180 post-HCT, 56 (7%) HaploCy, 24 (6%), SibCy, and 59 (4%) SibCNI recipients developed ≥1 FI (P < .001). The cumulative incidence of yeast FI was 5.2% (99% confidence interval [CI], 3.3% to 7.3%) for HaploCy, 2.2% (99% CI, .7% to 4.5%) for SibCy, and 1.9% (99% CI, 1.1% to 2.9%) for SibCNI (P = .001), and that of mold FI was 2.9% (99% CI, 1.5% to 4.7%), 3.7% (99% CI, 91.7% to 6.6%), and 1.7% (99% CI, 1.0% to 2.6%), respectively (P = .040). FI was associated with an increased risk of death, with an adjusted hazard ratio (HR) of 4.06 (99% CI, 2.2 to 7.6) for HaploCy, 4.7 (99% CI, 2.0 to 11.0) for SibCy, and 3.4 (99% CI, 1.8 to 6.4) for SibCNI compared with SibCNI without FI (P < .0001 for all). Similar associations were noted for transplantation-related mortality. FI did not impact rates of relapse or chronic GVHD.

CONCLUSIONS

Rates of FI by day 180 ranged between 1.9% and 5.2% for yeast FI and from 1.7% to 3.7% for mold FI across the 3 cohorts. The use of PTCy was associated with higher rates of yeast FI only in HaploHCT and with mold FI in both HaploHCT and SibHCT. The presence of FI by day 180 was associated with increased risk for overall mortality and transplant-related mortality at 2 years regardless of donor type or PTCy use. Although rates of FI were low with PTCy, FI is associated with an increased risk of death, underscoring the need for improved management strategies.

Sample size and power determination for multiparameter evaluation in nonlinear regression models with potential stratification

Sample size and power determination are crucial design considerations for biomedical studies intending to formally test the effects of key variables on an outcome. Other known prognostic factors may exist, necessitating the use of techniques for covariate adjustment when conducting this evaluation. Moreover, the main interest often includes assessing the impact of more than one variable on an outcome, such as multiple treatments or risk factors. Regression models are frequently employed for these purposes, formalizing this assessment as a test of multiple regression parameters. But, the presence of multiple variables of primary interest and correlation between covariates can complicate sample size/power calculation. Given the paucity of available sample size formulas for this context, these calculations are often performed via simulation, which can be both time-consuming as well as demanding extensive probability modeling. We propose a simpler, general approach to sample size and power determination that may be applied when testing multiple parameters in commonly used regression models, including generalized linear models as well as ordinary and stratified versions of the Cox and Fine-Gray models. Through both rigorous simulations and theoretical derivations, we demonstrate the formulas' accuracy in producing sample sizes that will meet the type I error rate and power specifications of the study design.

Allogeneic Hematopoietic Cell Transplantation Improves Outcome in Myelodysplastic Syndrome Across High-Risk Genetic Subgroups: Genetic Analysis of the Blood and Marrow Transplant Clinical Trials Network 1102 Study

PURPOSE

Allogeneic hematopoietic cell transplantation (HCT) in patients with myelodysplastic syndrome (MDS) improves overall survival (OS). We evaluated the impact of MDS genetics on the benefit of HCT in a biological assignment (donor v no donor) study.

METHODS

We performed targeted sequencing in 309 patients age 50-75 years with International Prognostic Scoring System (IPSS) intermediate-2 or high-risk MDS, enrolled in the Blood and Marrow Transplant Clinical Trials Network 1102 study and assessed the association of gene mutations with OS. Patients with TP53 mutations were classified as TP53multihit if two alleles were altered (via point mutation, deletion, or copy-neutral loss of heterozygosity).

RESULTS

The distribution of gene mutations was similar in the donor and no donor arms, with TP53 (28% v 29%; P = .89), ASXL1 (23% v 29%; P = .37), and SRSF2 (16% v 16%; P = .99) being most common. OS in patients with a TP53 mutation was worse compared with patients without TP53 mutation (21% ± 5% [SE] v 52% ± 4% at 3 years; P < .001). Among those with a TP53 mutation, OS was similar between TP53single versus TP53multihit (22% ± 8% v 20% ± 6% at 3 years; P = .31). Considering HCT as a time-dependent covariate, patients with a TP53 mutation who underwent HCT had improved OS compared with non-HCT treatment (OS at 3 years: 23% ± 7% v 11% ± 7%; P = .04), associated with a hazard ratio of 3.89; 95% CI, 1.87 to 8.12; P < .001 after adjustment for covariates. OS among patients with molecular IPSS (IPSS-M) very high risk without a TP53 mutation was significantly improved if they had a donor (68% ± 10% v 0% ± 12% at 3 years; P = .001).

CONCLUSION

HCT improved OS compared with non-HCT treatment in patients with TP53 mutations irrespective of TP53 allelic status. Patients with IPSS-M very high risk without a TP53 mutation had favorable outcomes when a donor was available.

Post-Transplantation Cyclophosphamide-Based Graft-versus-Host Disease Prophylaxis

BACKGROUND

In patients undergoing allogeneic hematopoietic stem-cell transplantation (HSCT), a calcineurin inhibitor plus methotrexate has been a standard prophylaxis against graft-versus-host disease (GVHD). A phase 2 study indicated the potential superiority of a post-transplantation regimen of cyclophosphamide, tacrolimus, and mycophenolate mofetil.

METHODS

In a phase 3 trial, we randomly assigned adults with hematologic cancers in a 1:1 ratio to receive cyclophosphamide-tacrolimus-mycophenolate mofetil (experimental prophylaxis) or tacrolimus-methotrexate (standard prophylaxis). The patients underwent HSCT from an HLA-matched related donor or a matched or 7/8 mismatched (i.e., mismatched at only one of the HLA-A, HLA-B, HLA-C, and HLA-DRB1 loci) unrelated donor, after reduced-intensity conditioning. The primary end point was GVHD-free, relapse-free survival at 1 year, assessed in a time-to-event analysis, with events defined as grade III or IV acute GVHD, chronic GVHD warranting systemic immunosuppression, disease relapse or progression, and death from any cause.

RESULTS

In a multivariate Cox regression analysis, GVHD-free, relapse-free survival was significantly more common among the 214 patients in the experimental-prophylaxis group than among the 217 patients in the standard-prophylaxis group (hazard ratio for grade III or IV acute GVHD, chronic GVHD, disease relapse or progression, or death, 0.64; 95% confidence interval [CI], 0.49 to 0.83; P = 0.001). At 1 year, the adjusted GVHD-free, relapse-free survival was 52.7% (95% CI, 45.8 to 59.2) with experimental prophylaxis and 34.9% (95% CI, 28.6 to 41.3) with standard prophylaxis. Patients in the experimental-prophylaxis group appeared to have less severe acute or chronic GVHD and a higher incidence of immunosuppression-free survival at 1 year. Overall and disease-free survival, relapse, transplantation-related death, and engraftment did not differ substantially between the groups.

CONCLUSIONS

Among patients undergoing allogeneic HLA-matched HSCT with reduced-intensity conditioning, GVHD-free, relapse-free survival at 1 year was significantly more common among those who received cyclophosphamide-tacrolimus-mycophenolate mofetil than among those who received tacrolimus-methotrexate. (Funded by the National Heart, Lung, and Blood Institute and others; BMT CTN 1703 ClinicalTrials.gov number, NCT03959241.).

A Multicenter Phase II, Double-Blind, Placebo-Controlled Trial of Maintenance Ixazomib After Allogeneic Transplantation for High-Risk Multiple Myeloma: Results of the Blood and Marrow Transplant Clinical Trials Network 1302 Trial

The role of allogeneic hematopoietic cell transplantation (allo-HCT) followed by maintenance therapy in high-risk multiple myeloma (MM) remains controversial. We evaluated the efficacy of ixazomib maintenance therapy after reduced-intensity conditioning allo-HCT from HLA-matched donors in patients with high-risk MM. The primary study endpoint was progression-free survival (PFS) postrandomization, treated as a time to event. Secondary endpoints were grade II-IV and grade II-IV acute graft-versus-host-disease (GVHD), chronic GVHD, best response, disease progression, nonrelapse mortality (NRM), overall survival (OS), toxicity, infection, and health-related quality of life. In this phase 2, double-blinded, prospective multicenter trial, we randomized patients with high-risk MM (ie, those with poor-risk cytogenetics, plasma cell leukemia, or relapsing within 24 months after autologous HCT) to ixazomib (3 mg on days 1, 8, and 15) or placebo after allo-HCT. The conditioning regimen included fludarabine/melphalan/bortezomib with tacrolimus plus methotrexate for GVHD. Fifty-seven patients were enrolled, of whom 52 (91.2%) underwent allo-HCT and 43 (82.7%) were randomized to ixazomib versus placebo. At 21 months postrandomization, the ixazomib and placebo groups had similar PFS (55.3% versus 59.1%; P = 1.00) and OS (94.7% versus 86.4%; P = .17). The cumulative incidences of grade III-IV acute GVHD at 100 days (9.5% versus 0%) and chronic GVHD at 12 months (68.6% versus 63.6%) also were similar in the 2 groups. The secondary analysis showed that at 24 months post-allo-HCT, PFS and OS were 52% and 82%, respectively, with a corresponding NRM of 11.7%. These results demonstrate the safety and durable disease control with allo-HCT in high-risk MM patients. We could not adequately assess the efficacy of ixazomib maintenance because the trial terminated early owing to enrollment delays, but there was no indication of any impact on outcomes.

Designing and conducting a clinical trial in blood and marrow transplantation

Clinical trials form the cornerstone of the science-based approach to improving patient outcomes. A trial needs to be designed and performed carefully to provide valid evidence to inform medical science and to protect the safety and well-being of its participants. The development of a clinical trial involving blood and marrow transplant (BMT) requires special considerations, including the rare disease populations involved and transplant-specific outcomes of interest that necessitate appropriate analysis techniques to evaluate. This article reviews key considerations and best practices for the design and conduct of a clinical trial in BMT, including the selection of patient population, treatment groups, objectives and endpoints, targeted sample size, statistical analysis strategy, provisions for monitoring patient safety and trial progress, and dissemination of trial results. The practical application of these principles is demonstrated using BMT CTN 1301, a recently completed clinical trial evaluating regimens for chronic graft-versus-host disease prevention in transplant patients.

SARS-CoV-2 vaccination in the first year after allogeneic hematopoietic cell transplant: a prospective, multicentre, observational study

Background

The optimal timing for SARS-CoV-2 vaccines within the first year after allogeneic hematopoietic cell transplant (HCT) is poorly understood.

Methods

We conducted a prospective, multicentre, observational study of allogeneic HCT recipients who initiated SARS-CoV-2 vaccinations within 12 months of HCT. Participants were enrolled at 22 academic cancer centers across the United States. Participants of any age who were planning to receive a first post-HCT SARS-CoV-2 vaccine within 12 months of HCT were eligible. We obtained blood prior to and after each vaccine dose for up to four vaccine doses, with an end-of-study sample seven to nine months after enrollment. We tested for SARS-CoV-2 spike protein (anti-S) IgG; nucleocapsid protein (anti-N) IgG; neutralizing antibodies for Wuhan D614G, Delta B.1.617.2, and Omicron B.1.1.529 strains; and SARS-CoV-2-specific T-cell receptors (TCRs). The primary outcome was a comparison of anti-S IgG titers at the post-V2 time point in participants initiating vaccinations <4 months versus 4-12 months after HCT using a propensity-adjusted analysis. We also evaluated factors associated with high-level anti-S IgG titers (≥2403 U/mL) in logistic regression models.

Findings

Between April 22, 2021 and November 17, 2021, 175 allogeneic HCT recipients were enrolled in the study, of whom all but one received mRNA SARS-CoV-2 vaccines. SARS-CoV-2 anti-S IgG titers, neutralizing antibody titers, and TCR breadth and depth did not significantly differ at all tested time points following the second vaccination among those initiating vaccinations <4 months versus 4-12 months after HCT. Anti-S IgG ≥2403 U/mL correlated with neutralizing antibody levels similar to those observed in a prior study of non-immunocompromised individuals, and 57% of participants achieved anti-S IgG ≥2403 U/mL at the end-of-study time point. In models adjusted for SARS-CoV-2 infection pre-enrollment, SARS-CoV-2 vaccination pre-HCT, CD19+ B-cell count, CD4+ T-cell count, and age (as applicable to the model), vaccine initiation timing was not associated with high-level anti-S IgG titers at the post-V2, post-V3, or end-of-study time points. Notably, prior graft-versus-host-disease (GVHD) or use of immunosuppressive medications were not associated with high-level anti-S IgG titers. Grade ≥3 vaccine-associated adverse events were infrequent.

Interpretation

These data support starting mRNA SARS-CoV-2 vaccination three months after HCT, irrespective of concurrent GVHD or use of immunosuppressive medications. This is one of the largest prospective analyses of vaccination for any pathogen within the first year after allogeneic HCT and supports current guidelines for SARS-CoV-2 vaccination starting three months post-HCT. Additionally, there are few studies of mRNA vaccine formulations for other pathogens in HCT recipients, and these data provide encouraging proof-of-concept for the utility of early vaccination targeting additional pathogens with mRNA vaccine platforms.

Funding

National Marrow Donor Program, Leukemia and Lymphoma Society, Multiple Myeloma Research Foundation, Novartis, LabCorp, American Society for Transplantation and Cellular Therapy, Adaptive Biotechnologies, and the National Institutes of Health.

Heath-related Quality of Life in Double Umbilical Cord Blood vs. Haploidentical Marrow Transplantation: a QOL Analysis Report of BMT CTN 1101

INTRODUCTION

The Blood and Marrow Transplant Clinical Trials Network study 1101 (BMT CTN 1101, NCT01597778) was a multicenter phase III randomized trial comparing the clinical outcomes and quality of life (QOL) of patients with hematological malignancies receiving double umbilical cord blood transplants (dUCBT) or HLA-haploidentical bone marrow transplants (haplo-BM) after reduced intensity conditioning. Five-year follow up found no significant differences in progression free survival (PFS) or overall survival (OS) between the two cohorts. The impact of alternative donor source on QOL, however, is unknown.

METHODS

English and Spanish speaking patients completed the Functional Assessment of Cancer Therapy-General (FACT-G), SF-36, EQ-5D, and Global QOL patient-reported outcome (PRO) assessments pre-transplant and at 12 and 24 months post-transplant. We compared longitudinal QOL measures between dUCBT and haplo-BMT and investigated the association of QOL and clinical outcomes using an inverse probability weighted-independent estimating equations (IPW-IEE) method accounting for missingness and baseline variables.

RESULTS

We found no significant differences in any of the QOL scores between the treatment arms at pre-transplant, 12, and 24 months. Pre-transplant scores were the only significant predictors of post-transplant QOL scores. Relapse and Grade III-IV acute graft-versus-host disease (GVHD) were associated with significant declines in mean FACT-BMT and SF-36 physical component scores. Chronic GVHD was associated with a decline in mean EQ-5D utility scores. We found no significant association between pre-transplant QOL scores and OS or PFS.

CONCLUSIONS

Donor type did not impact post-transplant QOL. Pre-transplant QOL scores and clinical events of GVHD and relapse were the only predictors of post-transplant QOL. QOL was not associated with survival in either treatment arm. PROs may be a valuable tool in pre-transplant risk assessment strategies to improve QOL outcomes.

Health-related quality of life in reduced-intensity hematopoietic cell transplantation based on donor availability in patients aged 50-75 with advanced myelodysplastic syndrome: BMT CTN 1102

For myelodysplastic syndrome (MDS), allogeneic hematopoietic cell transplantation (alloHCT) is the only available curative therapy. The Blood and Marrow Transplant Clinical Trials Network study 1102 (BMT CTN 1102, NCT02016781) was a multicenter, biologic assignment trial based on matched donor availability in adults aged 50-75 with higher risk de novo MDS who were candidates for reduced-intensity conditioning (RIC) alloHCT. The primary analysis showed that those who received alloHCT had a survival benefit, but whether this is at the cost of worse quality of life (QOL) has not been described in detail. English- and Spanish-speaking trial participants completed the Functional Assessment of Cancer Therapy-General (FACT-G), the SF-36, and the EQ-5D, at enrollment, every 6 months until 24 months, and 36 months. We compared patient-reported outcome (PRO) scores between study arms using an inverse probability weighted-independent estimating equation (IPW-IEE) model. Between January 2014 and November 2018, 384 subjects (median age 66.7 years, range: 50.1-75.3) enrolled at 34 centers. PRO completion rates were generally high at 65%-78%. The PRO trajectories for both arms were similar, with most decreasing or stable from baseline to 6 months and improving thereafter. Baseline PRO scores were the most consistent independent predictors of subsequent QOL outcomes and survival, even after controlling for clinical and patient-level factors. For older adults with MDS, the survival advantage associated with donor availability and alloHCT did not come at the cost of worse QOL. These results should reassure older patients and clinicians who prefer a curative approach to treating MDS.

Clinical Characteristics and Outcomes of COVID-19 in Pediatric and Early Adolescent and Young Adult Hematopoietic Stem Cell Transplant Recipients: A Cohort Study

Adult hematopoietic stem cell transplantation (HSCT) recipients are at a high risk of adverse outcomes after COVID-19. Although children have had better outcomes after COVID-19 compared to adults, data on risk factors and outcomes of COVID-19 among pediatric HSCT recipients are lacking. We describe outcomes of HSCT recipients who were ≤21 years of age at COVID-19 diagnosis and were reported to the Center for International Blood and Marrow Transplant Research between March 27, 2020, and May 7, 2021. The primary outcome was overall survival after COVID-19 diagnosis. We determined risk factors of COVID-19 as a secondary outcome in a subset of allogeneic HSCT recipients. A total of 167 pediatric HSCT recipients (135 allogeneic; 32 autologous HSCT recipients) were included. Median time from HSCT to COVID-19 was 15 months (interquartile range [IQR] 7-45) for allogeneic HSCT recipients and 16 months (IQR 6-59) for autologous HSCT recipients. Median follow-up from COVID-19 diagnosis was 53 days (range 1-270) and 37 days (1-179) for allogeneic and autologous HSCT recipients, respectively. Although COVID-19 was mild in 87% (n = 146/167), 10% (n = 16/167) of patients required supplemental oxygen or mechanical ventilation. The 45-day overall survival was 95% (95% confidence interval [CI], 90-99) and 90% (74-99) for allogeneic and autologous HSCT recipients, respectively. Cox regression analysis showed that patients with a hematopoietic cell transplant comorbidity index (HCT-CI) score of 1-2 were more likely to be diagnosed with COVID-19 (hazard ratio 1.95; 95% CI, 1.03-3.69, P = .042) compared to those with an HCT-CI of 0. Pediatric and early adolescent and young adult HSCT recipients with pre-HSCT comorbidities were more likely to be diagnosed with COVID-19. Overall mortality, albeit higher than the reported general population estimates, was lower when compared with previously published data focusing on adult HSCT recipients.

Biologic Assignment Trial of Reduced-Intensity Hematopoietic Cell Transplantation Based on Donor Availability in Patients 50-75 Years of Age With Advanced Myelodysplastic Syndrome

PURPOSE

Allogeneic hematopoietic cell transplantation (HCT) is the only potentially curative therapy for myelodysplastic syndromes (MDS), although it is infrequently offered to older patients. The relative benefits of HCT over non-HCT therapy in older patients with higher-risk MDS have not been defined.

METHODS

We conducted a multicenter biologic assignment trial comparing reduced-intensity HCT to hypomethylating therapy or best supportive care in subjects 50-75 years of age with intermediate-2 or high-risk de novo MDS. The primary outcome was overall survival probability at 3 years. Between January 2014 and November 2018, we enrolled 384 subjects at 34 centers. Subjects were assigned to the Donor or No-Donor arms according to the availability of a matched donor within 90 days of study registration.

RESULTS

The median follow-up time for surviving subjects was 34.2 months (range: 2.3-38 months) in the Donor arm and 26.9 months (range: 2.4-37.2 months) in the No-Donor arm. In an intention-to-treat analysis, the adjusted overall survival rate at 3 years in the Donor arm was 47.9% (95% CI, 41.3 to 54.1) compared with 26.6% (95% CI, 18.4 to 35.6) in the No-Donor arm (P = .0001) with an absolute difference of 21.3% (95% CI, 10.2 to 31.8). Leukemia-free survival at 3 years was greater in the Donor arm (35.8%; 95% CI, 29.8 to 41.8) compared with the No-Donor arm (20.6%; 95% CI, 13.3 to 29.1; P = .003). The survival benefit was seen across all subgroups examined.

CONCLUSION

We observed a significant survival advantage in older subjects with higher-risk MDS who have a matched donor identified and underwent reduced-intensity HCT, when compared with those without a donor. HCT should be included as an integral part of MDS management plans in fit older adults with higher-risk MDS.

Acute GVHD Diagnosis and Adjudication in a Multicenter Trial: A Report From the BMT CTN 1202 Biorepository Study

PURPOSE

Accurate and reproducible methods to diagnose, grade, and report acute graft-versus-host disease (GVHD) are critical for the evaluation of therapies and biomarkers.

PATIENTS AND METHODS

The Blood and Marrow Transplant Clinical Trials Network 1202 study is an observational study of 1,709 allogeneic hematopoietic cell transplantation recipients that implemented weekly data reporting and near real-time data adjudication by an end point review committee (ERC), assigning a confidence level (confirmed, probable, possible, or negative) to the diagnosis of acute GVHD at onset.

RESULTS

During the first 100 days, symptoms consistent with GVHD developed in 90% of cases but were often determined by centers to be due to causes other than GVHD. Indeed, GVHD was under consideration in only 23% of cases at symptom onset. Diagnostic biopsies were obtained in 40% of cases, but treatment often was incongruous with biopsy findings and 10.5% of biopsies were equivocal. Importantly, more than 40% of steroid courses were started for reasons other than GVHD. The ERC modified the determination of GVHD diagnosis and/or grade in 12.3% of onset cases. The cumulative incidence of acute GVHD as reported by the centers was 62%. When the ERC adjudicated GVHD onset to be present only if the confidence level was probable or confirmed, the incidence of GVHD declined to 49%.

CONCLUSION

This study demonstrates that the incidence of GVHD may be overestimated at symptom onset, establishes a contemporary benchmark for acute GVHD, and suggests a structured framework for reporting and adjudication of GVHD that could be used in prospective trials.

A unified approach to sample size and power determination for testing parameters in generalized linear and time-to-event regression models

To ensure that a study can properly address its research aims, the sample size and power must be determined appropriately. Covariate adjustment via regression modeling permits more precise estimation of the effect of a primary variable of interest at the expense of increased complexity in sample size/power calculation. The presence of correlation between the main variable and other covariates, commonly seen in observational studies and non-randomized clinical trials, further complicates this process. Though sample size and power specification methods have been obtained to accommodate specific covariate distributions and models, most existing approaches rely on either simple approximations lacking theoretical support or complex procedures that are difficult to apply at the design stage. The current literature lacks a general, coherent theory applicable to a broader class of regression models and covariate distributions. We introduce succinct formulas for sample size and power determination with the generalized linear, Cox, and Fine-Gray models that account for correlation between a main effect and other covariates. Extensive simulations demonstrate that this method produces studies that are appropriately sized to meet their type I error rate and power specifications, particularly offering accurate sample size/power estimation in the presence of correlated covariates.

Group sequential tests for treatment effect on survival and cumulative incidence at a fixed time point

Medical research frequently involves comparing an event time of interest between treatment groups. Rather than comparing the entire survival or cumulative incidence curves, it is sometimes preferable to evaluate these probabilities at a fixed point in time. Performing a covariate adjusted analysis can improve efficiency, even in randomized clinical trials, but no currently available group sequential test for fixed point analysis provides this adjustment. This paper introduces covariate adjusted group sequential pointwise comparisons of survival and cumulative incidence probabilities. Their test statistics have an asymptotic distribution with independent increments, permitting use of common stopping boundary specification methods. These tests are demonstrated through a redesign of BMT CTN 0402, a clinical trial that evaluated a prophylactic treatment for adverse outcomes following blood and marrow transplantation. A simulation study demonstrates that these tests maintain the type I error rate and power at nominal levels under a variety of settings involving influential covariates.

A group sequential test for treatment effect based on the Fine-Gray model

Competing risks endpoints arise when patients can fail therapy from several causes. Analyzing these outcomes allows one to assess directly the benefit of treatment on a primary cause of failure in a clinical trial setting. Regression models can be used in clinical trials to adjust for residual imbalances in patient characteristics, improving the power to detect treatment differences. But, none of the competing risks methods currently available for use in group sequential trials adjust for covariates. We propose a group sequential test for treatment effect that, because it is based on the Fine-Gray model, permits adjustment for covariates. Our derivations show that its sequence of test statistics has an asymptotic distribution with an independent increments structure, which allows standard techniques such as O'Brien-Fleming designs and error spending functions to be employed to meet type I error rate and power specifications. We demonstrate the test in a reanalysis of BMT CTN 0402, a phase III clinical trial that evaluated an experimental treatment for the prevention of adverse outcomes following blood and marrow transplant. Moreover, using a simulation study of randomized group sequential trials, we demonstrate that the proposed method preserves the type I error rate and power at their nominal levels in the presence of influential covariates.