A comparison of GvHD characteristics in recipients of haploidentical vs matched unrelated donor transplantation

The limited availability of human leukocyte antigen (HLA)- matched donors for allogeneic hematopoietic stem cell transplantation (HSCT) can prevent patients from receiving potentially life-saving therapy. T-replete HLA-mismatched haploidentical (haplo) HSCT may offer an alternative donor source where matched donors are unavailable. Previous studies indicate that haplo HSCT combined with posttransplant cyclophosphamide (PTCy) can induce survival rates similar to matched related donor (MRD) and matched unrelated donor (MUD) HSCT, with reduced incidence of acute and chronic graft-versus-host disease (a/cGvHD). However, there is limited data on risk factors, as well as aGvHD and cGvHD characteristics, in patients receiving haplo HSCT.

Sohl et al. evaluated the characteristics of a/cGvHD, including organ distribution, severity, and response to treatment, in patients who received haplo HSCT with PTCy versus those who received MUD HSCT with calcineurin inhibitors (CIs). The results were published in the Biology of Blood and Marrow Transplant,¹ and the GvHD Hub is happy to provide a summary.

Study design

• The analysis included patients aged ≥ 18 years who received initial HSCT between 2005 and 2017 (N = 394) from
  • 10/10 allele MUD (n = 179) + CI GvHD prophylaxis, or
  • haplo (n = 215) + PTCy GvHD prophylaxis.
• Primary endpoints: Cumulative incidence of GvHD, time to GvHD onset, and organ involvement.
• Secondary endpoints: Overall survival, disease free survival, non-relapse mortality, and relapse.

Results

• Median follow-up: 52.2 months.
• Patient characteristics are presented in Table 1.

Table 1. Baseline patient characteristics¹

Characteristic, % unless stated otherwise	Haplo HSCT (n = 215)	MUD HSCT (n = 179)	p
ALL, acute lymphocytic leukemia; AML, acute myeloid leukemia; BM, bone marrow; CLL, chronic lymphocytic leukemia; CML, chronic myelogenous leukemia; DRI, disease risk index; haplo, haploidentical transplantation; HCT-CI, hematopoietic cell transplantation-comorbidity index; HL, Hodgkin lymphoma; HSCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndrome; MPS, myeloproliferative syndrome; MUD, matched unrelated donor; NHL, non-Hodgkin lymphoma; PBSC, peripheral blood stem cell. Bold font signifies statistical significance.
Median age, years (range)	53 (19–75)	56 (20–74)	0.007
Male sex	58	52	0.22
Race			< 0.001
White	58	93
Black	38	7
Other/unknown	4	1
Diagnosis			0.24
AML	40	40
ALL	18	13
MDS/MPS/CML	21	29
NHL/HL/CLL	16	11
Cell source			0.013
BM	30	33
PBSC	70	67
Intensity			0.34
Myeloablative	46	51
RIC/non-myeloablative	54	49
DRI			0.10
Low	13	11
Intermediate	49	57
High	28	18
Very high	5	7
N/A	5	7
HCT-CI			< 0.001
0–2	53	36
≥ 3	47	64

Incidence and onset of GvHD in patients receiving haplo vs MUD HSCT

• No differences in cumulative incidence and time to onset of aGvHD was observed between patients who received haplo vs MUD HSCT (Table 2).
• Higher incidences and earlier onset of cGvHD were observed in patients who received haplo vs MUD HSCT (Table 2).

Table 2. Incidences and time to onset of a/cGvHD in patients who received haplo vs MUD HSCT¹

	Haplo HSCT (n = 215)	MUD HSCT (n = 179)	p
aGvHD, acute graft-versus-host disease; cGvHD, chronic graft-versus-host disease; haplo, T-replete HLA-mismatched haploidentical transplantation; HSCT, hematopoietic stem cell transplantation; MUD, matched unrelated donor; NS, not significant. Bold font signifies statistical significance.
Cumulative incidence of aGvHD, %			NS
Grade 2–4	40	49
Grade 3–4	14	16
Cumulative incidence of cGvHD, %			0.026
All Grade	34	41
Moderate–severe	22	31
Severe	16	29	0.02
Median time to aGvHD onset, days (range)
Grade 2–4	56 (14–607)	49 (9–1133)	0.23
Grade 3–4	84 (14–447)	70 (14–1133)	0.53
Median time to cGvHD onset, days (range)
All grade cGvHD	217.5 (74–618)	274 (78–1240)	0.012
Moderate–severe cGvHD	213.5 (74–618)	280.5 (107–1240)	0.011
Severe cGvHD	212 (74–728)	300 (107–1281)	0.003

aGvHD

• Of the patients who received haplo or MUD HSCT, 97 and 94 developed Grade 2–4 aGVHD, respectively.
• There was no statistical difference in the distribution of most affected organs in patients who developed Grade 2–4 aGVHD. The most affected organs in both groups (haplo vs MUD HSCT) were skin (75% vs 71%), gastrointestinal tract (71% vs 68%), and liver (14% vs 17%).
• There was no difference in all-cause mortality between groups (23% after haplo HSCT vs 19% after MUD HSCT; p = 0.34).
• Among patients who received haplo HSCT, a number of risk factors were associated with increased risk for developing Grade 2–4 aGVHD:
  • Myeloablative conditioning regimen (p = 0.0169)
  • Donor age ≥ 38 years (p = 0.05)
  • Child as donor (vs parent as donor; p = 0.0253).
• Among patients who received MUD HSCT, myeloablative conditioning was associated with increased development of Grade 2–4 aGVHD (HR, 1.48; p = 0.043).

cGvHD

• In patients who developed cGVHD:
  • Moderate–severe cGvHD was seen after haplo and MUD HSCT in 65% and 77% of patients, respectively
  • Two of the most affected organs, eyes and fasciae, differed significantly with eyes being more affected after MUD HSCT, while involvement of fasciae was more frequently seen after haplo HSCT (Table 3)
  • A greater proportion of haplo vs MUD recipients were off immunosuppressants at 2 years (63% vs 43%, p = 0.03)
  • Onset of cGvHD was later in MUD vs haplo recipients (274 days vs 217 days, p = 0.012)
  • No difference in 2-year all-cause mortality was observed between donor types.
• Risk factors associated with increased development of cGvHD after haplo HSCT:
  • White race (HR, 1.43; p = 0.029)
  • History of grade 2–4 aGvHD (HR, 2.34; p = 0.002)
  • Year of transplant (2012–2017 vs 2005–2011; HR, 1.68; p = 0.002)
  • Graft source (PBSC vs BM; HR, 1.41; p = 0.03).
• Risk factors associated with increased development of cGvHD after MUD HSCT:
  • Recipient age (≥ 55 vs < 55; HR, 1.55; p = 0.03)
  • Graft source PBSC (HR, 1.69; p = 0.002)
  • Year of transplant 2012–2017 (HR, 1.48; p = 0.003).

Table 3. Organs with moderate–severe (stage II–III) cGvHD in ≥ 10% of patients¹

Organ involvement, %	Haplo HSCT (n = 74)	MUD HSCT (n = 75)	p
cGvHD, chronic graft-versus-host disease; GI, gastrointestinal; haplo, T-replete HLA-mismatched haploidentical transplantation; HSCT, hematopoietic stem cell transplantation; MUD, matched unrelated donor. Bold font signifies statistical significance.
Skin	46	46	0.65
GI Upper	20	24	0.11
GI Lower	16	14	0.51
Liver	23	18	0.89
Eyes	12	28	< 0.001
Mouth	26	33	0.14
Joints/Fascia	48	14	0.001

Patient survival outcomes

• Multivariable analysis identified factors associated with significantly poorer survival outcomes (Table 4).

Table 4. Risk factors for poorer OS, DFS, NRM and relapse rates¹

Variable	HR	95% CI	p
aGvHD, acute graft-versus-host disease; cGvHD, chronic graft-versus-host disease; CI, confidence interval; DFS, disease-free survival; DRI, disease risk index; HR, hazard ratio; NRM, non-relapse mortality; OS, overall survival. Bold font signifies statistical significance.
OS
Grade 3–4 aGVHD	1.66	1.11–2.47	0.013
Severe cGVHD	1.79	1.20–2.69	0.005
Age (≥ 55 vs < 55)	1.72	1.23–2.41	0.001
Race (White vs Black)	1.66	1.08–2.57	0.022
DRI (high/very high vs low/intermediate)	2.08	1.51–2.86	< 0.001
DFS
Grade 3–4 aGVHD	1.53	1.03–2.28	0.038
Severe cGVHD	1.89	1.25–2.87	0.003
Age (≥ 55 vs < 55)	1.65	1.21–2.25	0.002
Race (White vs Black)	1.70	1.13–2.55	0.01
DRI (high/very high vs low/intermediate)	1.95	1.44–2.64	< 0.001
NRM
Grade 3–4 aGVHD	2.62	1.42–4.81	0.002
Severe cGVHD	4.72	2.49–8.94	<0.001
Age (≥ 55 vs < 55)	2.28	1.34–3.88	0.002
Relapse
Race (White vs Black)	2.22	1.31–3.76	0.003
DRI (high/very high vs low/intermediate)	2.59	1.78–3.77	< 0.001

Conclusion

Overall, the data suggest that haplo HSCT with PTCy leads to superior outcomes compared to MUD HSCT in patients undergoing their first transplantation. GvHD prophylaxis with PTCy resulted in lower incidences of cGvHD, faster cGvHD onset, and improved responses to immunosuppressive therapy in patients who received haplo HSCT. cGvHD organ distribution differed significantly between the two treatment groups, and the study highlighted a number of risk factors for patient survival outcomes following HSCT, such as development of severe aGvHD or cGvHD, higher age, White race, and high DRI, which were irrespective of donor type.

Limitations:

• The transplanting physician had the final decision on certain factors, including donor type, regimen intensity, and graft source.
• MUD HSCT was accompanied by CI-based GvHD prophylaxis as the most commonly used regimen (rather than PTCy, which accompanied haplo HSCT), so differences between arms may also be due to GvHD prophylaxis.