Disparities in Allograft Access in the Era of Post-Transplant Cyclophosphamide (PTCy)-Based Graft-Versus-Host Disease Prophylaxis

Abstract

Introduction

Although PTCy based GVHD prophylaxis has extended allograft access to patients (pts) from underserved populations, the extent to which disparities in allograft access persist is unknown.

Methods

We examined racial & socioeconomic (SES) disparities in graft source/characteristics & time to BMT in all adult 1^st allograft recipients 1/2018-6/2024 transplanted for AML, a diagnosis for which BMT is urgent (since 1/2018, our center routinely used PTCy GVHD ppx for adult donor grafts). We defined: 1) low SES as the composite of area deprivation index (ADI) ≥60%, Medicaid/Indigent insurance, &/or pt self report of cost of living/medical expense insecurity; 2) BMT indication date as date of diagnosis if ELN 2017 int/high risk or high risk molecular mutations or 2^o AML (n=672), else date of relapse (n=45); & 3) delayed BMT indication to transplant time as ≥180 days. Associations between delayed times to BMT & pt characteristics were evaluated using multivariable logistic regression models.

Results

Of 717 AML pts (median age 57 yrs, range 18-78; 87% received PTCy GVHD ppx), 485 (68%) were White/non-Hispanic & 232 (32%) non-White [49 Black, 97 Hispanic, 81 Asian, 5 Other]. Compared with White pts, fewer older non-White pts were transplanted (median age 60 vs 52 yrs, p<.001).

Examining SES, 225 (31%) pts had low SES by ADI, 33 (5%) by insurance & 29 (4%) by self report. However, each measure classified different pts as low (Fig 1), with 256 (36%) pts having low SES by ≥1 measure (including 53% of Black pts).

By donor type, 199 (28%) pts received HLA-identical related, 345 (48%) 8/8 unrelated donor (URD), 12 (2%) cord blood (CB), 129 (18%) haplo & 32 (4%) 6-7/8 URD grafts. Compared with White pts, non-White pts received >2x as many mismatched grafts (ie haplo/ CB/ 6-7/8 URDs), 40% vs 17%, p<.001. Over half (54%) of Black low SES pts received mismatched grafts. Considering graft characteristics, non-White pts received older adult donors including >2x as many haplos >40 yrs, 39% vs 18%, p=.010.

Overall, 40% of pts faced delays to BMT ≥180 days. Analysis of factors associated with time from BMT indication to transplant revealed marked disparities, with non-White pts more likely to face delays to BMT ≥180 days [OR (95% CI) 1.65 (1.17, 2.33), p=.004], due to the combinations of delays from indication to consult ≥90 days (1.62 (1.11, 2.36), p=.013), consult to financial approval ≥ 45 days (1.59 (1.13, 2.25), p=.008) & financial approval to BMT ≥90 days (1.65 (1.17, 2.32), p=.004). 8/8 URD recipients also faced delays from BMT indication to transplant [1.61 (1.10, 2.35), p=.014] due to delays from financial approval to BMT [1.53 (1.05, 2.21), p=.025]. In subgroup analysis, non-White 6-8/8 URDs faced greatest delays to BMT (Fig 2).

Conclusions

We demonstrate multiple levels of disparities in allograft access impacting underserved pts in the modern era, which will require significant, multifactorial interventions to address.