Prevention of cGvHD by naive T-cell depletion

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) can potentially cure patients with advanced hematologic cancers; however, it can also lead to morbidity and mortality due to graft-versus-host disease (GvHD). αβT cells in grafts can identify recipient alloantigens and promote engraftment by killing host hematopoietic and immune cells, and reduce relapse rates by destroying neoplastic blood cells, thus mediating the graft-versus-leukemia effect. However, alloreactive T cells can also damage host tissues and thus cause GvHD, therefore warranting pharmacologic immunosuppression or T-cell-depleted grafts.

αβT cells include naive (T_N), effector (T_E), and memory (T_M) subsets, which have a distinct surface phenotype. In murine models, it has been demonstrated that T_N cause severe GvHD, whereas T_M cause mild-to-no GvHD and maintain graft-versus-tumor capability.

Marie Bleakley et al.¹ recently published an article in the Journal of Clinical Oncology that analyzed the effects of T_N-depleted allo-HSCT on the occurrence of chronic GvHD (cGvHD) in patients with acute leukemia or myelodysplastic syndromes treated in three prospective phase II clinical trials, which are summarized below.

The GvHD Hub has recently published an editorial theme article on the T-cell depletion strategies to reduce incidence of GvHD in pediatric HSCT recipients here.

Study designs and patient characteristics

In total, data from 138 patients from three phase II trials (NCT00914940 [n = 41], NCT01858740 [n = 20], and NCT02220985 [n = 77]) were analyzed. Study designs can be seen in Table 1.

• Seven patients from NCT02220985 were not included in this analysis because the donor apheresis products did not meet the protocol-specified criteria for cell selection and therefore they did not receive T_N-depleted grafts.
• Differences across the trials included the age groups eligible, and that there were two treatment arms for NCT02220985, whereas NCT00914940 and NCT01858740 only had one treatment arm.

Table 1. Study design and participants*

HLA, human leukocyte antigen; MMF, mycophenolate mofetil; MRD, matched-related donors; MTX, methotrexate; MUD, matched-unrelated donors; TBI, total body irradiation. *Adapted from Bleakley, et al.1
	NCT00914940 (N = 41)	NCT01858740 (N = 20)	NCT02220985 (N = 77)
Eligible age, years	14–55	0–21	0–60
HLA-matched donor type	MRD	MRD (n = 2) MUD (n = 18)	MRD (n = 41) MUD (n = 36)
Conditioning + prophylaxis	Fludarabine, thiotepa, TBI 13.2 Gy + tacrolimus	Fludarabine, thiotepa, TBI 13.2 Gy + tacrolimus, MTX	Fludarabine, thiotepa, TBI 13.2 Gy + tacrolimus, MTX (n = 39)
Fludarabine, thiotepa, cyclophosphamide, TBI 4 Gy + tacrolimus MMF (n = 38)

• Primary outcome: cGvHD (diagnosed by the 2014 National Institutes of Health criteria).

• Secondary outcomes: graft failure, Grade III–IV acute (a)GvHD within the first year of allo-HSCT, overall survival, relapse, non-relapse mortality (NRM), and survival free of moderate or severe cGvHD or relapse (cGvHD-free, relapse-free survival).
• Patient characteristics for all patients who received T_N-depleted peripheral blood stem cells (PBSCs) can be seen in Table 2. Of note
  • the median age was 37 years;
  • there were more females than males;
  • and most patients received T_N-depleted PBSCs for the treatment of leukemia or lymphoma.

Table 2. Patient characteristics*

ALL, acute lymphoid leukemia; AML, acute myeloid leukemia; CML, chronic myelogenous leukemia; CR1, first complete remission; CR2, second complete remission; GvHD, graft-versus-host disease; HLA, human leukocyte antigen; MDS, myelodysplastic syndrome; MMF, mycophenolate mofetil; MRD, matched-related donors; MUD, matched-unrelated donors; PBSC, peripheral blood stem cell; TN, naive T cells. *Adapted from Bleakley, et al.1 †Measured by the Karnofsky (age, ≥16 years) or Lansky (age, <16 years) performance status scales. A lower score represents more disability.
Characteristic, % (unless otherwise stated)	TN-depleted PBSC recipients (N = 138)
Median age (range), years	37 (1–60)
Sex
Male	41
Female	59
Performance status score†
≥90	75
<90	25
Diagnosis
Myeloid	52
AML	43
MDS with excess blasts	4.4
Blastic plasmacytoid dendritic cell neoplasm	0.7
CML with a history of myeloid blast crisis	1.5
Mixed phenotype acute leukemia	2.9
Lymphoid	48
ALL	46
CML with a history of lymphoid blast crisis	1.5
Disease risk
Standard-risk (CR1, no residual disease)	62
High-risk (beyond CR1 and/or residual disease)	38
Donor
HLA-MRD	61
HLA-MUD	39
Conditioning regimen
High intensity	72
Intermediate intensity	28
GvHD pharmacologic prophylaxis
Tacrolimus monotherapy	29.7
Tacrolimus and methotrexate	43
Tacrolimus and MMF	27.5

Results

Incidence of GvHD

• The cumulative incidence of Grade II aGvHD cases was 71% (95% confidence interval [CI], 64–79), which were predominantly Stage I (upper) gastrointestinal GvHD.
• The cumulative incidence of Grade III aGvHD was low (4%; 95% CI, 1–8) and did not differ according to graft source (matched-related donor, 5%; matched-unrelated donor, 4%) or conditioning intensity (high intensity, 5%; intermediate intensity, 3%).
• Only two patients with aGvHD required systemic treatment with agents other than corticosteroids.
• In total, 7% of patients developed cGvHD, which was commonly mild and responsive to steroids.

Relapse and survival

The median duration of follow-up for surviving patients was 1,485 days (range, 262–1,826 days). The 3-year cumulative incidence of relapse was 23% (95% CI, 16–30). The median time to relapse was 206 days (interquartile range, 111–343 days).

Relapse rates according to patient subgroups can be seen in Table 3.  

Table 3. Relapse rates according to patient subgroups*

CI, confidence interval; MRD, matched-related donor; MUD, matched-unrelated donor. *Data from Bleakley, et al.1
Patient group	Relapse rate, % (95% CI)
Graft type
Received MRD grafts	27 (17–37)
Received MUR grafts	17 (7–27)
Condition treatment
High intensity	19 (11–27)
Intermediate intensity	35 (18–52)
Diagnosis
Myeloid malignancy	27 (17–38)
Lymphoid leukemia	19 (9–28)
Age, years
<30	18 (7–29)
≥30	26 (17–36)

Of the 31 patients who relapsed, 90% were not receiving prednisone at the time and 32.3% had discontinued or were quickly weaned off immunosuppression without developing GvHD. Of these relapsing patients, the median survival from relapse was 273 days (interquartile range, 85–590 days). A total of 21 relapsed patients were given donor lymphocyte infusions or other immunotherapies: 11 reached complete remission for at least 6 months. The survival outcomes of all T_N-depleted PBSC recipients are displayed in Table 4.

Table 4. Survival data of T_N-depleted PBSC recipients*

CI, confidence interval; CRFS, chronic graft-versus-host disease-free relapse-free survival; GRFS, graft-versus-host disease-free relapse-free survival; MRD, matched-related donor; MUD, matched-unrelated donor; OS, overall survival; RFS, relapse-free survival. *Adapted from Bleakley, et al.1
	3-year estimate, % (95% CI)
	OS	RFS	CRFS	GRFS
Total group (N = 138)	77 (71–85)	69 (61–77)	68 (61–76)	64 (56–72)
Graft type
MRD (n = 84)	77 (68–86)	67 (58–78)	67 (58–78)	62 (52–74)
MUD (n = 54)	79 (69–91)	72 (61–85)	70 (59–83)	66 (55–80)
Condition treatment
High intensity               (n = 100)	77 (69–86)	72 (63–81)	72 (63–81)	67 (58–77)
Intermediate intensity            (n = 38)	78 (66–93)	60 (45–80)	57 (41–78)	54 (38–75)
Diagnosis
Myeloid (n = 72)	74 (64–85)	66 (55–78)	64 (54–77)	60 (49–73)
Lymphoid (n = 66)	81 (72–91)	72 (62–84)	72 (62–84)	68 (57–80)
Age, years
<30 (n = 49)	84 (74–95)	76 (64–89)	76 (64–89)	67 (55–82)
≥30 (n = 89)	74 (65–84)	65 (55–76)	64 (54–75)	61 (52–73)

NRM, causes of death, adverse events, and infectious complications

• Of the patients who received T_N-depleted grafts, the cumulative incidences of NRM at 100 days and 3 years were 4% (95% CI, 1–8) and 8% (95% CI, 3–13), respectively.
• The 3-year NRM rates were estimated to be
  • 6% (95% CI, 1–11) and 11% (95% CI, 3–20) for patients who received MRD and MUD grafts;
  • 9% (95% CI, 3–15) and 5% (95% CI, 0–13) for patients who had high-intensity and intermediate-intensity conditioning;
  • and 6% (95% CI, 0–13) and 9% (95% CI, 3–15) in patients aged <30 years and ≥30 years, respectively.
• The frequency and type of Grade III–V non-hematologic adverse events were typical of patients who receive myeloablative HSCT.
• Epstein-Barr virus reactivation was uncommon with only one patient requiring treatment. Cytomegalovirus disease occurred in 4.3% of patients.

Conclusion

The authors demonstrated that the depletion of T_N from PBSC allografts results in low incidences of severe aGvHD and cGvHD, without increasing the risks of relapse or NRM.

Limitations to the trial included the possibility that the lower total T-cell dose administered with T_N-depleted grafts may be correlated with the cGvHD reduction, and the risk of type 1 errors due to the nature of the single arm study. Therefore, randomized trials will be important to confirm these findings. Indeed, two phase II trials have already been initiated (NCT03970096 and NCT03779854).