Paper Details

ABSTRACT

Introduction

Post-transplant cyclophosphamide (PTCy) as graft-versus-host disease (GVHD) prophylaxis enables human leukocyte antigen (HLA) mismatched unrelated donor (MMUD) HCT. Yet, intensification of reduced intensity conditioning (RIC) regimens and HLA disparities may affect post-HCT quality of life (QOL) recovery. QOL trajectory comparisons among PTCy MMUD, PTCy matched unrelated donor (MUD), and non-PTCy MUD RIC platforms are lacking.

Methods:

The primary aim is to compare patient-reported physical function and GVHD symptoms in adults (age ≥ 18) undergoing RIC HCT in 3 cohorts: MMUD with PTCy, MUD HCT with PTCy, and MUD HCT with non-PTCy. We secondarily assessed longitudinal QOL trajectories between the MMUD PTCy cohort and a contemporary real-world cohort of patients from the CIBMTR registry. For the primary aim, we assembled 3 cohorts: 1) MMUD (HLA match ≤7/8) PTCy from the RIC arm of ACCESS (NCT04904588),  2) MUD PTCy from BMT CTN 1703 (NCT03959241), and 3) MUD non-PTCy from BMT CTN 1703. The analysis population of the primary aim included patients who completed a baseline (pre-HCT) survey and ≥1 post-HCT survey.  In the secondary aim, we compared the MMUD PTCy cohort (1 above) and a real-world cohort from the CIBMTR Patient-Reported Outcome (PRO) protocol (NCT01166009). Due to low baseline completion in the PRO protocol, the analysis population for the secondary aim were patients who completed ≥1 post-HCT survey. Surveys submitted after a 2^nd transplant or relapse were excluded. PROs collected in ACCESS and CTN1703 included PROMIS physical function and Lee Symptom Scale (LSS) to measure GVHD symptoms. ACCESS and CIBMTR PRO Protocol collected PROMIS physical, social, and cognitive function, anxiety, depression, fatigue, sleep disturbance, pain interference, sexual interest, and sexual satisfaction. For PROMIS scales, Δ5 points is considered clinically meaningful. PROs were collected at baseline, D100, D180, and 1-year post-HCT. To compare longitudinal PRO trajectories, multivariate mixed models with repeated measures adjusting for baseline PRO score (for primary aim only), age, sex, performance status, HCT-CI, race, insurance, and relationship status were performed. For this exploratory analysis, p<0.05 was considered statistically significant.

Results:

Most patients in MMUD PTCy (119/193, 61.7%), MUD PTCy (136/208, 65.4%) and MUD non-PTCy (131/212, 61.8%) completed a baseline survey and at least one post-HCT PRO surveys. Similarly, post-HCT survey completion in MMUD PTCy (136/193, 70.5%) and the real-world cohort (190/256, 74.2%) was high. In the primary analysis cohort (MMUD PTCy vs MUD PTCy | MUD non-PTCy), patients in MMUD PTCy were more likely to be female (51% vs 34%, p=0.005 | 34%, p=0.007), be Hispanic or Black (30% vs 5%, p<0.001 | 7%, p<0.001), live in a metropolitan area (89% vs 76%, p=0.02 | 82%, p=0.25), have Medicaid (17% vs 4%, p=0.004 | 3%, p<0.001), and have lower baseline physical function (mean T-score (SD): 44.5 (9.5) vs 54.8 (7.4), p=0.004  | 48 (7.0), p=0.004). At 1-year post-HCT, the incidence of moderate-severe chronic GVHD in MMUD PTCy (10/105, 9.5%) and MUD PTCy (9/136, 6.6%) were lower than in MUD non-PTCy (27/131, 20.6%). Accounting for baseline physical function, patients in MMUD PTCy reported lower mean physical function at D100 (adjusted mean T-score [95% CI]: 40.1 [38.8, 42.2] vs 43.0 [40.9, 45.1], p=0.02 | 42.8 [40.7, 45.0], p=0.03). Physical function at D180 and 1-year post-HCT were similar across study cohorts. While GVHD symptoms between the MMUD PTCy and MUD PTCy cohorts were similar at all timepoints, those in the MUD non-PTCy group reported higher mean overall LSS scores (Δ3.5 [1.0, 6.0], p =0.007) and higher mean mouth (Δ9.3 [5.1, 13.5), p < 0.001) and nutrition (Δ3.8 [1.2, 6.5], p = 0.005) subscale scores at 1 year post-HCT. In the secondary analysis cohort (MMUD PTCy vs real world cohort), there were no observed differences in post-HCT trajectories for any QOL domains.

Discussion:

Patients undergoing RIC MMUD HCT with PTCy more slowly recover physical function at D100, but otherwise have similar QOL trajectories to those undergoing MUD HCT and experience fewer clinically significant GVHD symptoms compared to the MUD non-PTCy cohort. These data support the routine incorporation of MMUDs in donor search algorithms. Future studies to evaluate QOL in MMUD HCT with reduced dose PTCy GVHD prophylaxis are needed.