PTCy: Increased infection frequency and delayed CD4+ cell reconstitution, but faster B-cell reconstitution vs conventional GvHD prophylaxis

Historically, prophylaxis for graft-versus-host disease (GvHD) following allogeneic hematopoietic stem cell transplantation (allo-HSCT) with a human leukocyte antigen (HLA)-matched sibling donor (MSD) or matched unrelated donor (MUD) has consisted of a calcineurin inhibitor-based regimen, with tacrolimus/methotrexate (TAC/MTX) and tacrolimus/sirolimus (TAC/SIR) being the most common regimens.¹

More recently, posttransplantation cyclophosphamide (PTCy)—which has been used successfully in GvHD prophylaxis regimens following transplant with a haploidentical donor—has been used as the basis for prophylaxis regimens in MSD, MUD, and HLA-mismatched donor settings. Delayed immune reconstitution and risk of infection have been associated with both calcineurin inhibitor- and PTCy-based regimens, though it is not clear how these regimens compare to one another regarding these risks. Farhad Khimani and colleagues conducted a retrospective analysis to compare immune recovery parameters and frequency of infections after haploidentical hematopoietic cell transplantation (haplo-HCT) and MUD HCT based on the type of GvHD prophylaxis regimen, which we report on here.¹

Study design

This was a retrospective review of patients at a single center between January 2012 and December 2018. Patients who were included had a hematologic malignancy and underwent allo-HSCT from a haploidentical or MUD following either a myeloablative conditioning (MAC) or reduced-intensity conditioning (RIC) regimen. None of the patients received antithymocyte globulin.

Patients were excluded if they:

• Had a MUD allo-HCT and received TAC/SIR with interleukin-2 (IL-2) or TAC/SIR with pacritinib, panobinostat, or ustekinumab on prospective clinical trials
• Received TAC/MTX and bortezomib, maraviroc, or ruxolitinib
• Received rituximab as part of the conditioning regimen

The primary endpoints were:

• The kinetics of immune reconstitution
• The frequency of infections after allo-HSCT

Secondary endpoints included:

• Nonrelapse mortality (NRM), defined as death from any cause without relapse of hematologic malignancy
• Relapse incidence
• Relapse-free survival (RFS), defined as being alive and in remission at last contact
• Overall survival (OS), defined as time from transplantation to death from any cause

Baseline characteristics

The primary analysis compared four groups of patients who received a MUD allo-HCT based on the posttransplantation GvHD prophylaxis regimen, as shown in Table 1.

Table 1. Patient characteristics by prophylaxis regimen*

Results

Immune reconstitution

• Through the first year posttransplant, the recovery of absolute total CD4⁺ T cell count was significantly lower in the haplo-PTCy and MUD-PTCy groups compared with the TAC/MTX and TAC/SIR groups (p = 0.025 at Day 365).
• Recovery of CD19⁺ B cells at Day 180 and thereafter was faster in the haplo-PTCy and MUD-PTCy groups compared to the non-PTCy groups (p < 0.001 at Day 180 and Day 365).
• Recovery of absolute total CD8⁺ T cell count and total natural killer (NK) cell count were not significantly different among the four treatment groups.

Infection density after allo-HSCT

• There was a significantly higher frequency of total infections in the haplo-PTCy and MUD-PTCy groups (5.0 per 1,000 person-days for each) compared with the TAC/MTX (1.8 per 1,000 person-days) and TAC/SIR (2.6 per 1,000 person-days) groups in the year following allo-HSCT (p < 0.01).
  • This frequency was considerably higher in the PTCy-based regimen groups in the first 90 days and remained significantly higher until 180 days after transplantation.
• Analysis of infection frequency by infection type—viral, bacterial, or fungal—is shown in Table 2.
  • Cytomegalovirus (CMV) reactivation was the most common infection among all treatment groups, and the rates of CMV, BK virus, human herpesvirus-6, and other viruses were significantly higher in the PTCy-based groups (p < 0.01).

Table 2. Infection density per 1,000 person-days*

OS, relapse, RFS, NRM, and engraftment

Regarding OS:

• There were no significant differences in the 2-year OS between the groups (haplo-PTCy, 61.9%; MUD-PTCy, 56.2%; TAC/MTX, 69.4%; TAC/SIR, 62%).
  • Karnofsky performance status (KPS) <90, HCT comorbidity index (HCT-CI) ≥3, primary disease, and delayed CD4 recovery were identified as factors associated with poor OS on univariate analysis.
  • On multivariate analysis, only HCT-CI was an independent risk factor for OS.

Regarding RFS and NRM:

• The four groups had similar RFS at 2 years (haplo-PTCy, 55.9%; MUD-PTCy, 36.8%; TAC/MTX, 58%; TAC/SIR, 57%; p = 0.19), though the cumulative incidence of relapse was significantly higher in the MUD-PTCy group (p = 0.02).
• Among the groups, there was no significant difference in NRM. Reduced KPS and higher HCT-CI score were significant predictors of NRM.

There was a significant delay in platelet engraftment in the PTCy-based groups (p < 0.001); time to engraftment for platelets and neutrophils is shown in Table 3.

Table 3. Time to engraftment*

Within 6 months of allo-HSCT, 96% of patients in the PTCy-based groups vs 86% in the conventional prophylaxis groups achieved complete CD3/CD33 donor chimerism. The most common causes of death in all groups were relapse (37.7%) and infection (19.3%). Rates of relapse as cause of death were significantly different between the groups (haplo-PTCy, 42.9%; MUD-PTCy, 57.9%; TAC/MTX, 20.6%; TAC/SIR, 38%; p = 0.044).

Conclusion

In this comparison of differences in immune recovery and infection rates between PTCy-based GvHD prophylaxis and conventional calcineurin inhibitor-based regimens, slower recovery of absolute total CD4⁺ T cell counts were seen within the first year following allo-HSCT in patients who received PTCy-based regimens. Frequency of infection was also higher in patients receiving PTCy-based regimens; viral infections—CMV in particular—were significantly higher in this group in the 6 months following transplant. B cell recovery, however, was considerably faster in the PTCy-based groups compared with the conventional groups and should be investigated further. It should be noted that there were several limitations to this study: the exact stop dates on immunosuppressive medications were not included, and as such, the investigators were not able to comment on whether there were differences in immune suppression between the groups; in addition, not all patients achieved complete donor chimerism at 6 months, the MUD-PTCy group was small, and a diverse range of diseases and donor graft sources were included. Overall, however, these findings suggest that early infection surveillance and prophylaxis should be considered in patients receiving PTCy-based regimens.