Are We Making PROGRESS in Preventing Graft-versus-Host Disease and Improving Clinical Outcomes? Impact of BMT CTN 1301 Study Results on Clinical Practice

Abstract

The need for prospective randomized clinical trials investigating novel graft-versus-host disease (GVHD) prevention strategies which include other clinical outcomes impacted by GVHD has been highlighted as a priority for the field of hematopoietic cell transplantation. A recently completed study through the Blood and Marrow Transplant Clinical Trials Network (BMT CTN 1301) comparing CD34+ selection and post-transplant cyclophosphamide to tacrolimus/methotrexate (Tac/MTX) for GVHD prevention demonstrated no significant differences in the primary endpoint of Chronic GVHD-Relapse-Free Survival (CRFS) among the three approaches. The trial did not demonstrate a superior approach compared to Tac/MTX; however, it highlights several challenges in determining the best and most relevant approaches to clinical trial design, particularly in the context of current and ongoing changes in real world practices. Here, we review BMT CTN 1301 results and implications for clinical practice and future clinical trial design.

Introduction

Chronic graft-versus-host disease (GVHD) is the leading cause of late morbidity, mortality, and impaired quality of life (QOL) after allogeneic hematopoietic cell transplantation (alloHCT), and is increasing with more frequent use of mobilized peripheral blood stem cells (PBSC) as a graft source.^1,2 The combination of a calcineurin inhibitor (CNI), now commonly tacrolimus (Tac), with methotrexate (MTX) has been standard practice over the past three decades for GVHD prevention; however, this treatment is associated with several toxicities, including mucositis, renal and hepatic toxicities, increased risk of thrombotic microangiopathy, and the potential need for prolonged therapy and monitoring. The latter limits the ability to use post-HCT cellular therapies for treatment or prevention of infection or disease relapse. Furthermore, while CNI with MTX may limit the incidence and severity of acute GVHD, it is not effective in preventing chronic GVHD.³

The National Institutes of Health funded Blood and Marrow Transplant Clinical Trials Network (BMT CTN) is an alliance of transplant centers that conducts multi-institutional studies in the field of HCT and cellular therapy to improve outcomes of HCT.⁴ BMT CTN 1301 (NCT02345850) is a recently completed study comparing two CNI-free interventions to Tac/MTX for GVHD prophylaxis for patients undergoing an alloHCT with myeloablative conditioning: posttransplant cyclophosphamide as a single agent (PTCy) or infusion of a CD34+-selected PBSC graft without additional immune suppressive drugs. Results demonstrated no statistically significant differences in chronic GVHD, relapse-free survival (CRFS) among the experimental arms compared to Tac/MTX; however,there was a significantly inferior overall survival (OS ) with CD34+ selected PBSC grafts.⁵

In 2019, the BMT CTN established an Evidence into Practice Task Force to focus on the implications of BMT CTN trial results for clinical care and to foster translation of those results into practice. The Task Force brought together a group of transplant physicians and providers, researchers, medical directors, administrators, nurses, and patient advocates to review BMT CTN 1301 results and identify opportunities for implementation of the findings into practice. Here we describe the key findings in the context of prior relevant literature in the area, discuss potential implications for clinical practice, and the impact of results on future initiatives to prevent GVHD and optimize patient outcomes.

GVHD Prevention and Novel Endpoints

The need and challenges for prospective, randomized clinical studies comparing new GVHD prevention strategies that include clinical outcomes impacted by GVHD was recognized as a priority at the 2007 BMT CTN State of the Science Symposium.⁶ A benchmark analysis was performed using data from the Center for International Blood and Marrow Transplant Research (CIBMTR) in order to select approaches to be further investigated. ⁷ Six investigational approaches from single-institution clinical trials were compared with contemporary CIBMTR controls of Tac/MTX: 1) etanercept, 2) pentostatin/anti-thymocyte globulin (ATG), 3) PTCy 4) graft modification with CD34+ selection, 5) bortezomib, and 6) maraviroc. A second objective of these analyses was to evaluate novel and clinically meaningful endpoints that not only account for GVHD, but also consider the potential effect of therapies on relapse and mortality. Several composite endpoints were explored, including CRFS, which is defined as survival without development of moderate to severe chronic GVHD, disease relapse or death; and GVHD, relapse-free survival (GRFS) which includes survival without grade III-IV acute GVHD or chronic GVHD requiring systemic therapy, disease relapse or death. In this benchmark analysis, the CRFS rate at 1 year with Tac/MTX was 22%, versus 33% for PTCy and 59% for CD 34 selection.⁷ Improving upon the observed CRFS rate with standard Tac/MTX was recognized as a clear priority for future prospective trials. These results thus led to the development of two multicenter clinical trials for the Prevention and Reduction Of GVHD and Relapse and Enhancing Survival after Stem cell transplantation (PROGRESS) that tested different approaches using composite endpoints: PROGRESS I, BMT CTN 1203 (NCT02208037) and PROGRESS II, BMT CTN 1301. Figure 1 provides an overview and summary of this work, spanning over a decade, from conception to recent completion of these trials. PROGRESS I addressed GVHD prevention after reduced intensity conditioning with a primary endpoint of GRFS and PROGRESS II after myeloablative conditioning with a primary endpoint of CRFS. PROGRESS I was a randomized Phase II study that was followed by a randomized Phase III study comparing PTCy to Tac/MTX, which has completed accrual, with patients still being followed for the primary endpoint of GRFS. This paper focuses on the findings of PROGRESS II.

Figure 1:

Overview of the development of Blood and Marrow Transplant Clinical Trials Network (BMT CTN) clinical trials investigating novel GVHD prevention approaches from 2011-2021.

[CIBMTR- Center for International Blood and Marrow Transplant Research; HLA- human leukocyte antigen; Tac- tacrolimus; MTX- methotrexate; PTCy- post-transplant cyclophosphamide; MMF- mycophenolate mofetil; PBSC- peripheral blood stem cell; BM- bone marrow; GRFS- GVHD-free, relapse-free survival; CRFS- Chronic GVHD, relapse-free survival; HR-hazard ratio; CI- confidence interval]

BMT CTN 1301 was a phase III randomized, open-label, 3-arm trial comparing a CD34+ selected PBSC graft, PTCy (BM graft) and Tac/MTX (BM graft) in matched related and unrelated donor transplants for acute leukemia or myelodysplastic syndrome (MDS). BM was selected as the graft source for the control arm (Tac/MTX) based on results of BMT CTN 0201 that demonstrated similar OS but substantially less chronic GVHD with BM as opposed to PBSC grafts. BM was also the primary graft source used in prior studies of PTCy with myeloablative conditioning in both single and multi-institutional settings.^8,9 CRFS was selected as the primary endpoint as both regimens (CD34+ selection and PTCy) demonstrated their largest effect on the development of chronic GVHD in the benchmark analysis.⁷ With growing emphasis on integrating patient-reported outcomes as endpoints in GVHD trials, QOL was an important secondary endpoint. Assessments were performed immediately prior to HCT and post-HCT at day 100, day 180 and at one and two years.

Rationale for Studying CD34+ Selection and Post-transplant Cyclophosphamide

CD34+ Selection

Several different strategies have been used over the years to reduce the risk of GVHD by removing T-cells from the hematopoietic cell graft. The predominant recent approach is removal through positive selection of CD34+ cells using immunomagnetic beads (CliniMACS® Cell Separation System [Miltenyi Biotec, Gladbach, Germany]).^10-12 CD34+ selected PBSC grafts for HCT are consistently associated with decreased rates of acute and/or chronic GVHD; the GVHD that occurs is most commonly low grade or mild.¹³ Although early studies of CD34+ selected allografts reported high rates of graft failure,^14,15 use of ablative conditioning regimens and addition of ATG now produce consistent engraftment.¹⁶ Early concerns for an increased risk of relapse with T-cell depletion (most prominent in chronic myeloid leukemia) were not confirmed in studies of acute myelogenous leukemia (AML)^17-19, acute lymphoblastic leukemia (ALL)²⁰, or myelodysplastic syndromes (MDS)²¹. BMT CTN 0303, a phase 2 single-arm mutli-center study investigating CD34 selection in AML resulted in a disease-free survival for patients in first complete remission of 72.8% at 1 year.¹⁰ Use of T cell depleted grafts, however, does result in delayed immune reconstitution,^22,23 thereby increasing rates of opportunistic infections (such as cytomegalovirus, Epstein Barr Virus, adenovirus and human herpes virus-6) and infection-related mortality.^18,21 Finally, the use of CD34+ selection offers a promising platform for post-transplant immunotherapy with adoptive cell therapies targeting infection, minimal residual disease and disease relapse, since use of immunosuppressive agents that might affect the efficacy of these therapies is avoided.

Post-transplant Cyclophosphamide

High dose cyclophosphamide in combination with CNI and mycophenolate mofetil was initially introduced as an effective strategy to prevent GVHD in the human leukocyte antigen (HLA) mismatched/haploidentical setting.²⁴ Preclinical and clinical studies demonstrated that Cy administered early in the post-HCT period (days +3 and +4) preferentially kills activated allo-reactive T-cells while sparing non-allo-reactive T-cells, such as regulatory T-cells, leading to the attenuation of GVHD without compromising a graft-versus-tumor effect.²⁵ PTCy was subsequently investigated as the sole prophylaxis (without CNI or other drugs) after myeloablative fully HLA-matched related or unrelated donor transplants and demonstrated low rates of severe acute and chronic GVHD, with more than half of patients never requiring additional immunosuppressive medications even with long term follow up.⁸ This approach also resulted in rapid immune reconstitution and a low incidence of opportunistic infections.^8,9,26 Additional studies with longer follow up are still needed regarding the impact of additional high dose cyclophosphamide, with or without other drugs, on late effects such as secondary malignancies.

BMT CTN 1301 Key Results

The primary aim of BMT CTN 1301 was to determine whether CNI-free approaches offer advantages over Tac/MTX, as reflected in the CRFS primary endpoint. The patient population included patients up to 65 years of age, with acute leukemia in morphologic remission or MDS with <5% blasts in bone marrow with plans to undergo a myeloablative conditioning regimen with an 8/8 HLA matched related or unrelated donor graft. The median age of study participants was 51 years (range 21-66) in the CD34+ selection arm, 50 years (19-65) in the PTCy arm and 51 years (13-64) in the Tac/MTX arm. There were no statistically significant differences in CRFS among the three GVHD prevention approaches. Importantly, although CD34+ selection was associated with significantly lower rates of acute and chronic GVHD than the other two arms, it was also associated with worse OS, primarily driven by higher transplant-related mortality (TRM) due to increased infection- and organ failure-related deaths. PTCy was associated with comparable chronic GVHD and survival outcomes relative to Tac/MTX. Despite reported differences in GVHD and other outcomes, QOL measured were similar with all approaches. Although BMT CTN 1301 did not demonstrate one clear optimal approach to GVHD prophylaxis, there are several important findings that may inform change in current clinical practices and clinical trial design.

Implications of Results for Clinical Practice and Future Directions

Current Practices for GVHD Prophylaxis: CIBMTR Data Analysis

As challenges exist in determining and implementing appropriate practice changes in response to results of randomized controlled trials, ^27,28 we sought to better understand the current practices for GVHD prophylaxis. We queried CIBMTR data as to the approaches to GVHD prevention in myeloablative alloHCT being used by U.S. centers before (2013-2015) and after (2018-2020) the BMT CTN 1301 study period for both BM and PBSC grafts (Table 1). Tac/MTX was the most common approach to GVHD prophylaxis in both matched related donor transplants (61% in 2013-2015 and 63% in 2018-2020) and matched unrelated donor transplants (72% in 2013-2015 and 64% in 2018-2020).^27,28 CD34+ selection accounted for only 1-3% of HLA matched related and HLA-matched unrelated donor transplants, and only 2% of all PBSC grafts; its use was limited to just a few centers with expertise. PTCy without CNI accounted for only 2% of HLA matched related donor (both time periods) and <1% (2013-2015) and 1% (2018-2020) of matched unrelated donor transplants, and only 4% of all BM grafts within the time periods queried, likely due to studies demonstrating higher rates of GVHD with PTCy as a single agent.^27,28 Thus while the study design of BMT CTN 1301 may have reflected what was expected to be best practice with the use of BM, it did not, ultimately, reflect transplant center real-world current practice.

Table 1.

Characteristics of patients who underwent first myeloablative matched related or unrelated donor HCT for AML, ALL, or MDS/MF in the United States and reported to the CIBMTR, 2013-2015 and 2018-2020

Characteristic	Matched related 2013- 2015	Matched unrelated 2013- 2015	Total 2013- 2015	Matched related 2018- 2020	Matched unrelated 2018-20	Total 2018- 2020
No. of patients	2441	3085	5526	1697	2434	4131
Age at HCT, years - no. (%)
Median (min-max)	47 (1-66)	45 (1-66)	46 (1-66)	43 (1-66)	46 (1-66)	45 (1-66)
<10	140 (6)	188 (6)	328 (6)	98 (6)	116 (5)	214 (5)
10-17	146 (6)	220 (7)	366 (7)	148 (9)	139 (6)	287 (7)
18-29	330 (14)	466 (15)	796 (14)	257 (15)	361 (15)	618 (15)
30-39	292 (12)	439 (14)	731 (13)	257 (15)	354 (15)	611 (15)
40-49	506 (21)	580 (19)	1086 (20)	287 (17)	468 (19)	755 (18)
50-59	768 (31)	771 (25)	1539 (28)	461 (27)	639 (26)	1100 (27)
60-69	259 (11)	421 (14)	680 (12)	189 (11)	357 (15)	546 (13)
Recipient sex - no. (%)
Male	1354 (55)	1693 (55)	3047 (55)	937 (55)	1360 (56)	2297 (56)
Female	1087 (45)	1392 (45)	2479 (45)	760 (45)	1074 (44)	1834 (44)
KPS - no. (%)
90-100	1586 (65)	2040 (66)	3626 (66)	1103 (65)	1586 (65)	2689 (65)
≤ 80	820 (34)	1016 (33)	1836 (33)	569 (34)	820 (34)	1389 (34)
Not reported	35 (1)	29 (1)	64 (1)	25 (1)	28 (1)	53 (1)
Primary disease for HCT - no. (%)
AML	1332 (55)	1729 (56)	3061 (55)	891 (53)	1382 (57)	2273 (55)
ALL	771 (32)	864 (28)	1635 (30)	572 (34)	682 (28)	1254 (30)
MDS/MF	338 (14)	492 (16)	830 (15)	234 (14)	370 (15)	604 (15)
Graft type - no. (%)
Bone marrow	448 (18)	806 (26)	1254 (23)	391 (23)	633 (26)	1024 (25)
Peripheral blood	1993 (82)	2279 (74)	4272 (77)	1306 (77)	1801 (74)	3107 (75)
Planned GVHD prophylaxis - no. (%)
CD34 selection ± other	62 (3)	77 (2)	139 (3)	17 (1)	37 (2)	54 (1)
PTCY alone	43 (2)	9 (<1)	52 (1)	27 (2)	29 (1)	56 (1)
PTCY ± others (including TAC/MMF)	30 (1)	42 (1)	72 (1)	183 (11)	339 (14)	522 (13)
TAC + MTX ± others	1494 (61)	2231 (72)	3725 (67)	1066 (63)	1566 (64)	2632 (64)
Other GVHD prophylaxis	812 (33)	726 (24)	1538 (28)	404 (24)	463 (19)	967 (23)

KPS- Karnofsky Performance Status; AML- acute myelogenous leukemia; ALL- acute lymphoblastic leukemia; MDS- myelodysplastic syndrome; MF- myelofibrosis; GVHD- graft-versus-host disease; PTCY- Post-transplant cyclophosphamide; TAC- tacrolimus; CSA- cyclosporine; MTX- methotrexate; CNI- calcineurin inhibitor; MMF- mycophenolate mofetil

CD34+ Selection and Graft Manipulation

One of the goals of BMT CTN 1301 was to identify a prevention strategy that would minimize the need for longer term immunosuppression. Although CD34+ selection achieved this goal using a PBSC graft, it was also associated with inferior OS primarily from excess TRM due to infection and organ failure. While CNI free approaches using graft manipulation still represents a potentially viable platform for post-transplant maintenance as well as other cell therapies, current strategies require further improvement to bring into standard practice. Graft manipulation technologies require relatively new expertise and resource utilization and there may be a learning curve to successfully apply these strategies, although results of BMT CTN 0303 did indicate that this was feasible.¹² Also, the high TRM associated with CD34+selection was postulated to be due, at least in part, to the very high intensity of the conditioning regimens used with this approach, which could be improved upon with less intensive but still myeloablative approaches. Future studies using novel graft manipulation methods, less toxic preparative regimens, and novel approaches to infection prevention are warranted. This is especially important since PBSC continues to be the preferential graft source used in clinical practice. While we do not anticipate use of CD34+ selection specifically as a GVHD prophylaxis modality based on results of this trial, we expect graft manipulation to continue to emerge as an important strategy for GVHD prevention in the future once limitations are addressed. Other types of novel selective T cell depletion strategies (e.g. naïve T cell depletion, alpha-beta T cell depletion, T cell depletion with regulatory T cells) are being explored for GVHD prophylaxis.^29,30

Post-transplant Cyclophosphamide

PTCy also represents a viable platform for post-transplant maintenance and other cell therapy approaches. PTCy as a single agent with a BM graft resulted in similar GVHD and survival outcomes to Tac/MTX without the additional toxicities of MTX or long-term CNI. It is important to note, however, that the BMT CTN 1301 analysis was not powered to determine the equivalency of PTCy and Tac/MTX. While this approach to GVHD prophylaxis may still represent an attractive alternative for patients with pre-existing co-morbidities that place them at higher risk for post-HCT complications, caution must be made in over-interpreting these results. Whether the addition of CNI or other agents to PTCy improves upon GVHD and survival outcomes in this setting remains unknown and requires further prospective study^31,32

Peripheral Blood Stem Cells versus Bone Marrow as Graft Source

BMT CTN 0201 was a phase III, multicenter, randomized trial that compared PBSC and BM grafts for unrelated donor HCT in patients with hematologic malignancies, demonstrating similar survival but significantly less chronic GVHD with BM²⁸. Results of this trial were published in 2012; however, despite the findings, there was no significant change in use of BM versus PBSC after this time point. Fewer than one-fifth of HCT physicians reported a practice change in response to BMT CTN 0201 results on a survey done in 2016, which was confirmed by CIBMTR data.²⁷ Evaluation of practices before and after the BMT CTN 1301 study period (2013-2015 and 2018-2020) confirms the predominant use of PBSC for both matched related (82% and 77%, respectively) and matched unrelated donor HCT (74% in both time periods) (Table 1).

Ease of collection, faster recovery, and improvements in engraftment and disease-free and OS in high risk hematologic malignancies are cited as reasons for the increasing use of PBSC since the 1990s, though data supporting improved survival in most patient groups are lacking.^33,34 Despite increased moderate-severe chronic GVHD and subsequent poor QOL compared with BM grafts, there remains an overall perceived benefit for PBSC grafts, which includes ease of acquisition, lower rates of graft failure, and lower donor burden.²⁷ Continued preference for PBSC has resulted in fewer transplant practitioners with BM collection experience, and logistical challenges such as surgical room scheduling. Several studies report a progressive downward drift in total nucleated cell (TNC) cell/ul counts in BM harvests. ^35-37 A CIBMTR study of more than 15,000 collections indicates a decline in marrow quality, as measured by total TNC concentration, over the past several decades, with counts decreasing from 21.8 x 10⁶/ml in 1994-1996 to 18.7 x 10⁶/ml in 2012-2016, (p<0.001). The only factor associated with TNC counts in multivariate regression analysis was the number of harvests performed, with centers performing more than 30 harvests per era having better quality harvests. Such factors have led to a lower use of BM grafts despite the strong evidence that they may be better than PBSC grafts for long term outcomes.

Factors beyond our control may also impact the implementation of evidence into clinical practice. The COVID-19 pandemic has further increased the use of PBSC grafts. Given the need to ensure timely arrival of stem cell products in a situation where disruptions to collection and travel were frequent, the National Marrow Donor Program (NMDP) strongly recommended cryopreservation.³⁸ Because of concern about the effect of freezing on cell dose (always lower in BM grafts), there was a significant decrease in requests for marrow grafts. Relative to December 2019 to February 2020, domestic transplant center requests were 37% less likely to be for marrow during March to May 2020.³⁰

Nevertheless, an important result of BMT CTN 1301 is the reaffirmation of BM as a suitable graft for HLA matched HCT with myeloablative conditioning for patients using Tac/MTX or PTCy for GVHD prophylaxis, with two-year survival rates of >75%. Of note, BMT CTN 1301 included both related and unrelated donors in contrast to BMT CTN 0201²⁸. Although 1301 did not formally compare BM with PBSC for PTCy and Tac/MTX, the observed outcomes with BM were excellent and consistent with studies that generally demonstrate that BM is associated with a lower risk of chronic GVHD even with a single agent PTCy platform.³⁹ Whether the results of the current study will lead to higher utilization of BM grafts especially once the pandemic is over, remains to be seen.

Benchmark CRFS and Endpoints

BMT CTN 1301 demonstrated significantly superior CRFS rates of 60.2% (95% CI 50.3-68.7%) for CD34+ selection, 60.3% (95% CI 50.5-68.7%) for PTCy, and 52.6% (95% CI 43.1-61.3%) for Tac/MTX compared to the CIBMTR estimated 1-year CRFS of 22%, which was used at the baseline in designing the study. Notably, the benchmark cohort included transplants from 2000 to 2011, both myeloablative (72%) and reduced intensity conditioning, all disease statuses, and only unrelated donors using both BM and PBSC (66%) grafts.⁷ In contrast, BMT CTN 1301 was a more contemporary cohort that had strictly defined eligibility criteria (e.g. only patients with <5% blasts), allowed a restricted number of conditioning regimens specified the graft type (BM only for PTCy and Tac/MTX and PBSC for CD34+ selection) and included both matched related (38.1%) and unrelated donors. It is also possible that clinical trial participants simply have better outcomes compared to those not enrolled in trials, though a previous study evaluating BMT CTN study participants and non-participants in BMT CTN 0201 demonstrated similar survival.³⁶ Improvements in patient selection, deeper remissions with measurable residual disease assessments and novel therapies, and continued progress in supportive care may also have contributed to the improved survival in BMT CTN 1301. While these factors likely contributed to the significantly better outcome than predicted by the registry estimated CRFS benchmark, the differences noted between the BMT CTN 1301 Tac/MTX cohort and the CIBMTR cohort likely had a significant impact on this CRFS benchmark. It is thus critically important for future BMT CTN trials using historical benchmarks to be able to re-analyze an updated cohort similar to patients eligible for the trial to confirm best benchmarks for clinical trial endpoints. In considering the critical question of the best novel endpoints for future studies, we also must bear in mind the challenges in determining the priorities for success, including whether these priorities are the same for both patients and providers. As we continue to move toward more “personalized medicine”, the need to better understand the impact of new therapies and outcomes on QOL and its potential use as a novel endpoint remains of critical importance. While there were no QOL differences detected between the three approaches in BMT CTN 1301, it remains critical to continue to investigate how we can optimize the use of patient-reported outcomes and endpoints to study for clinical trials. Ongoing efforts within the BMT CTN are seeking to improve patient-reported outcome reporting and compliance, track and report missing data, as well as incorporate QOL expertise early within protocol and study design to be able to optimally use data for analyses.

Costs and Resources

Novel GVHD prophylaxis platforms and approaches such as CD34+ selection may require higher upfront costs and resources; however, this needs to be weighed against the resource implications of the downstream effects of GVHD and related complications for both the patient and the healthcare system. A limited number of studies have evaluated long term costs of chronic GVHD. A study from Europe demonstrated that patients with moderate-severe chronic GVHD had significantly higher healthcare resource costs, as well as work-related productivity loss compared to patients with no or mild chronic GVHD. Cumulative costs (a sum of direct medical and indirect productivity) during the first 3 years for moderate-severe chronic GVHD was up to EUR 47,811,835 compared to EUR 14,887,599 for those without chronic GVHD.⁴⁰ A recent analysis of United States data using claims from the Optum Research Database also demonstrated that patients with steroid-refractory chronic GVHD had significantly higher health care utilization and costs with 2-year costs of $532,673 compared to $252,909 in patients without GVHD.⁴¹ Although it is clear that there are significant long-term costs incurred by chronic GVHD, the impact of novel GVHD prevention approaches on long-term costs remains unknown. While cost analyses can be challenging to accomplish, this is another critical aspect of clinical practice that requires further study and should be an important consideration to incorporate into future BMT CTN prospective clinical trials⁴².

Conclusions

Multi-center prospective randomized clinical trials such as BMT CTN 1301 offer innovation and progress to our field and highlight the challenges and tension of determining the effectiveness of a transplant approach in an academic clinical research trial within the context of real-world influences and clinical practices. The CIBMTR remains a critical resource to reflect on outcomes data from real-world practices, but prospective randomized controlled trials remain the gold standard for practice change. This trial did not show a superior approach compared to standard Tac/MTX but did show excellent outcomes with two approaches (including the standard) when combined with a BM graft. The choice of approach may be influenced by the risk profile of individual patients – or center preference and experience. A key question that continues to be asked is whether BM should be re-adopted as the “standard”. Or, given the fact that the community has adopted PBSC as the preferred graft source, should future trials instead focus on improving outcomes for PBSC transplants? Ultimately,while evidence has consistently supported the use of BM as the graft source associated with a lower risk of chronic GVHD, given the many factors that influence center choice, as well as patient outcomes, it is likely that a “standard” will never be fully accepted. However, as we have learned, in order to successfully adopt evidence into practice, it is imperative to consider real world influences and current practices into clinical trial design, especially when selecting the control arm. While it remains crucial to continue to investigate novel approaches to build upon and advance our current transplant strategies, multi-center randomized clinical trials are complex and require extensive time and resources underscoring the practical importance of such considerations. Despite the fact that BMT CTN 1301 failed to demonstrate an improvement in the primary endpoint of CRFS with CNI -free approaches, and thus ultimately may not directly translate into clinical practice; it still provides valuable lessons for future clinical trial design. As we face the challenges of translating evidence from clinical research trials into standard clinical practice, it is upon us as transplant investigators to critically examine our current real-world practices and influences to develop and study the most valuable and applicable interventions including their impact on patient reported outcomes and costs.

Highlights.

• BMT CTN 1301 did not demonstrate a superior calcineurin inhibitor-free approach to graft-versus-host disease prevention compared to tacrolimus and methotrexate.

• BMT CTN 1301 reaffirmed use of bone marrow as the graft associated with a lower risk of chronic GVHD, but the use of bone marrow does not reflect current practice.

• Current practice and historical benchmarks must align with clinical trial patient eligibility to confirm best benchmarks for clinical trial endpoints in graft-versus-host disease prevention.