Final outcomes from a phase 2 trial of posoleucel in allogeneic hematopoietic cell transplant recipients

Abstract

Allogeneic hematopoietic cell transplantation (allo-HCT) recipients are susceptible to viral infections. We conducted a phase 2 trial evaluating the safety and rate of clinically significant infections (CSIs; viremia requiring treatment or end-organ disease) after infusion of posoleucel, a partially HLA-matched, allogeneic, off-the-shelf, multivirus-specific T-cell investigational product for preventing CSIs with adenovirus, BK virus, cytomegalovirus, Epstein-Barr virus, human herpesvirus-6, or JC virus. This open-label trial enrolled allo-HCT recipients at high risk based on receiving grafts from umbilical cord blood, haploidentical, mismatched, or matched unrelated donors; post-HCT lymphocytes of <180/mm3; or use of T-cell depletion. Posoleucel dosing was initiated within 15 to 49 days of allo-HCT and subsequently every 14 days for up to 7 doses. The primary end point was the number of CSIs due to the 6 target viruses by week 14. Of the 26 patients enrolled, only 3 (12%) had a CSI by week 14, each with a single target virus. In vivo expansion of functional virus-specific T cells detected via interferon-γ enzyme-linked immunosorbent spot assay was associated with viral control. Persistence of posoleucel-derived T-cell clones for up to 14 weeks after the last infusion was confirmed by T-cell-receptor deep sequencing. Five patients (19%) had acute graft-versus-host disease grade 2 to 4. No patient experienced cytokine release syndrome. All 6 deaths were due to relapse or disease progression. allo-HCT recipients at high risk who received posoleucel had low rates of CSIs from 6 targeted viruses. Repeat posoleucel dosing was generally safe and well tolerated and associated with functional immune reconstitution. This trial was registered at www.ClinicalTrials.gov as #NCT04693637.

Graphical abstract

Figure 1.

Viremia and CSIs up to week 14. For each numbered patient in the study (N = 26), there are 6 rows, 1 per target virus. Bars represent the duration of measured viremia (viral load > lower limit of quantification [LLOQ]) per virus. The first panel shows all viremia including viremia that was categorized as CS or causing EOD, the second panel shows duration of viremia categorized as CS (2 patients with CMV), the third panel shows duration of viremia categorized as causing EOD (1 patient with EBV). Nine of 22 patients with viremia had detection of 1 target virus, 5 with 2 viruses, 6 with 3 viruses, and 2 with 4 viruses. Fifteen of 26 patients (58%) had detectable BKV by week 14, making it the most frequently detected virus, followed by EBV (13 of 26 patients, 50%), HHV-6 (6 of 26 patients, 23%), CMV (5 of 26 patients, 19%), AdV (3 of 26 patients, 12%), and JCV (3 of 26 patients, 12%). Notably, there were no CSls from BKV, HHV-6, or JCV, and a single EBV-associated CSI leading to EOD. CS, clinically significant.

Figure 2.

Viral load (VL), functional virus-specific immune reconstitution, and posoleucel persistence functional immune reconstitution evaluated by ELISpot assay. The frequency of virus- specific IFN-γ⁺–producing cells (posoleucel and endogenous derived) was evaluated after stimulation of patient peripheral blood mononuclear cells (PBMCs) with AdV, BKV, CMV, EBV, or HHV-6 antigens (spot-forming cells [SFCs] per 5 × 10⁵ PBMCs). (A) ELISpot responses are plotted for all patients (and target viruses) with evaluable data before infusion (Pre; N = 53; mean = 14 SFCs), who were aviremic for a target virus through week 14 (aviremic; n = 51; mean = 28 SFCs), and who were viremic for a target virus through week 14 (viremic, n = 39 of a total of 45 viremic events; mean = 171 SFCs). Box plots show the median with all values plotted. (B) ELISpot responses are plotted for all patients with evaluable data who were viremic for ≥1 viruses during the primary end point. Data shown represent the circulating frequency of IFN-γ⁺ T cells before posoleucel infusion (Pre) and the peak response through week 14 (Post). VLs are also shown per virus by plotting the peak VL through week 14 (peak VL) and at week 14/last time point available (week 14 VL). VLs from 3 patients with CSIs are excluded (CMV, n = 2; and EBV, n = 1). (C) TCRβ sequencing was used to track presence of TCRβ sequences unique to posoleucel during the infusion period and after (postinfusion). The percentage of patients with detectable posoleucel T cells during each indicated study period are shown.

Figure 3.

Detection of functional immune reconstitution and posoleucel clones over time coincident with viremia reduction. Patient examples of viral load plotted with unique posoleucel clones detected by TCRβ sequencing (left panels) and functional IFN-γ⁺ virus-specific T-cell responses detected by ELISpot (posoleucel and endogenous derived; right panels) through week 26 of the study for patients with viremia of ≥1 target virus(es). Three patients with viremia that did not progress to CSI are shown: patient 1: BKV, EBV, and HHV-6 viremia (A-B; received all 7 doses of posoleucel); patient 2: BKV and EBV viremia (C-D; received all 7 doses of posoleucel); and patient 3: BKV, CMV, EBV, and JCV viremia (E-F; received all 7 doses of posoleucel). TCRβ clones unique to posoleucel are shown as the log₂ fold change of the sum frequency of clones relative to first time point detected. Virus-specific IFN-γ⁺–producing cells were measured by ELISpot after stimulation of patient PBMCs with AdV, BKV, CMV, EBV, or HHV-6 antigens (SFCs per 5 × 10⁵ PBMCs). All detectable viremia (viremia > LLOQ) is shown in the left panels with TCRβ sequencing data. In right panels with ELISpot data, only viremia for which there was corresponding ELISpot data is shown.

Figure 4.

Overall survival (OS) through week 52.