Invansive aspergillosis (IA) represents a leading cause of morbidity and mortality in patients with hematological malignancies and hematopoietic stem cell (HSC)/solid organ transplant recipients. The limitations of the contemporary antifungal treatment include the lack of efficacy in several cases, the drug-associated toxicity, the emergence of uncommon or drug-resistant molds and the added financial burden in transplant care. Adoptive immunotherapy with Aspergillus-specific T cells (Asp-STs) constitutes an alternative and promising therapeutic approach against IA, however, the complex and costly production limits its broader application. Consequently, and in contrast to the remarkable clinical progress with virus-specific T cells, clinical development of fungus-specific T cell therapy is still in its infancy; only one clinical trial of adoptive immunotherapy for IA has been conducted up to date, showing that recipients of Asp-ST clones after haploidentical HSC transplantation cleared IA more often than conventionally treated patients.

We developed and validated a novel, rapid, simplified and minimally laborious process of generating and scaling up functionally active Asp-STs, from a single blood draw, in only 10 days. A total of 40x106 peripheral blood mononuclear cells (PBMCs) derived from 30-40ml of peripheral blood of normal donors were exposed to either an Aspergillus fumigatus lysate (n=10) or a pool of Aspergillus fumigatus overlapping peptides of the glycosidase Crf1, the 1,3-beta-glucanosyltransferase gel1 and the serine hydroxymethyltransferase SHMT, all of which are considered potential immunogenic targets (n=5). To determine the most potent stimulus for the generation of Asp-STs, dual, lysate-derived and peptide-derived, Asp-ST cell lines were produced from four donor samples and compared in terms of antigen specificity. Pulsed PBMCs were cultured in the presence of interleukin 4 (IL-4) and IL-7 for 10 days in G-rex bioreactors.

Cell lines produced after stimulation with either Aspergillus lysate or peptide pools had a similar fold-expansion reaching a mean absolute number of 230±81x106 cells (range 131-363x106) and 223±19x106 cells (range 197-244x106), respectively. The cell lines derived from either condition were polyclonal, comprised predominantly of CD4+ cells (75±10% and 74±10%, respectively) and CD8+ cells as well (17±9% and 15±2%, respectively), and expressed central (CD45RA-/CD62L+: 44±20% and 60±5%, respectively) and effector memory markers (CD45RA-/CD62L-: 36±20% and 21±2%, respectively). To address whether these highly expanded T cells are functional and specific against the targeted mold, each cell line was pulsed with its initial stimuli, either Asp lysate or peptide pool, and the secretion of interferon-gamma (IFN-γ) was measured by Elispot. For peptide-derived Asp-STs, specificity was assessed both, collectively for all and individually for each peptide. All lines (5/5) pulsed with peptide pools but only three of 10 cell clones pulsed with Asp lysate presented activity against Asp [lysate: 72±39 Spot forming cells (SFC)/5x105 cells; all 3 peptides: 297±81 SFC/5x105 cells]. Interestingly, not only all donor cell lines derived from pooled peptides elicited strong IFN-γ response, but 4/5 were inducible to stimulation by each peptide separately and 1/5 by Crf1, at comparable levels (Crf1: 109±59 SFC/5x105 cells, Gel1: 113±58 SFC/5x105 cells, SHMT: 121±34 SFC/5x105 cells), indicating that T cells specific for these antigens are commonly present in healthy subjects.

To directly compare the stimulatory capacity of Asp lysate versus peptide pools, 4 donor samples were tested for their ability to produce specific clones against both conditions. Interestingly, all 4 donor clones showed response against peptides but only 1 of 4 against the lysate, suggesting that the Asp proteins Crf1, Gel1 and SHMT induce stronger Th1 responses than Asp lysate.

In conclusion, we established and validated an optimized, rapid and simple process - which can be easily GMP-adapted- of scaling up Asp-STs, at clinically relevant numbers. Since the effective management of IA in immunocompromised patients still remains a desirable target, our proposal provides a promising therapeutic approach for the management of IA.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.