Page 209 - 2019_11 Resto del Mondo-web

IFNγ in immune-mediated graft failure
donor-specific antibodies (DSA) in the recipient; iii) T-cell depletion (TCD) of the graft; iv) ABO-blood group mis- match; v) use of RIC; vi) a diagnosis of non-malignant disorders (in particular thalassemia, severe aplastic ane- mia, SAA, and hemophagocytic lymphohistiocytosis, HLH); vii) viral infections; viii) low nucleated cell dose in the graft; and ix) the use of myelotoxic drugs in the post- transplant period.1-4
In the last two decades, several groups have investigated immune-mediated GF. In particular, it has been shown that immune-mediated GF is mainly caused by host T and nat- ural killer (NK) cells surviving the conditioning regimen, through a classical alloreactive immune response against non-shared, major (in case of HLA-partially-matched HSCT) or minor (in case of fully HLA-matched HSCT) his- tocompatibility antigens.2,5,6 However, to date the molecu- lar pathways involved in immune-mediated GF have not yet been completely clarified. Indeed, since the inhibition of different pathways (including perforin-, FasL–, TNFR-1–, and TRAIL-dependent cytotoxicity) did not prove to be efficient in preventing GF, the pathophysiological mecha- nisms responsible for GF seem to be multiple and likely to be redundant.7 Nonetheless, consistently over the years, different groups have suggested a pivotal pathogenic role of IFNγ in GF pathophysiology,8-14 through both direct [e.g. inhibition of hematopoietic stem cell (HSC) self-renewal, proliferative capacity, and multilineage differentiation]10,11 and indirect (e.g. induction of FAS expression on HSC, with increased apoptosis in the presence of activated cyto- toxic T cells)8,12 effects.
Despite these experimental data, there has still not been any in vivo characterization of GF in humans. Indeed, although the expansion of host CD8+ T cells in patients experiencing GF has been previously demon- strated in vivo,15,16 a more detailed characterization of this cell population is lacking. Thus, we started a prospective study aimed at better characterizing the pathophysiology of GF, focusing on the identification of biological markers that: (i) could predict early the occurrence of GF in the clinical setting; and (ii) could be used as a therapeutic tar- get with clinically available biological agents. For this pur- pose, we broadly investigated cytokine and chemokine levels in peripheral blood (PB), as well as the cellular fea- tures in bone marrow (BM) biopsies of patients experi- encing this complication. After confirming in vivo a role of IFNγ-pathway in the development of GF, we also investi- gated in an animal model of GF whether the sole inhibi- tion of IFNγ would be able to prevent/treat GF. Finally, in view of these findings and the similarity between immune-mediated GF and HLH, we treated, in compas- sionate use (CU), with emapalumab, an anti-IFNγ mono- clonal antibody recently approved for the treatment of HLH,17 three patients with primary HLH, who, after hav- ing experienced GF, underwent a second HSCT.
Methods
Patients
Patients aged from 0.3 to 21 years, who received an allograft from any type of donor/stem cell source between January 1st 2016 and August 31st 2017 at the IRCCS Bambino Gesù Children’s Hospital in Rome, Italy, were considered eligible for the study. All patients or legal guardians provided written informed consent, and the entire research was conducted under
institutional review board approved protocols and in accordance with the Declaration of Helsinki. The Bambino Gesù Children’s Hospital Institutional Review Board approved the study.
Cytokine profile
In order to identify a cytokine/chemokine profile predictive of GF, PB samples were collected at different time points after HSCT: day 0, +3±2, +7±2, +10±2, +14±2, +30±2 after transplan- tation. Validated MesoScale Discovery (MSD, Rockville, MD, USA) platform-based immunoassay was used for the quantifica- tion of IFNγ, sIL2Rα, CXCL9, CXCL10, TNFα, IL6, IL10, and sCD163 serum levels.
Bone marrow biopsy: histopathology analysis and immunofluorescence
Bone marrow biopsies were obtained when GF was suspect- ed. (Since BM characterization was a secondary end point of this study and BM aspiration is not routinely performed in this con- dition, parents/legal guardians could refuse the procedure.) Details on BM specimen preparation, histopathology analysis and immunofluorescence are reported in the Online Supplementary Appendix.
Immune-phenotypic analysis
The following monoclonal antibodies (mAbs) were used: anti- CD3, CD4, CD8, CD25, CD27, CD28, CD45RA, CD45RO, CD56, CD57, CD62L, CD95, CD127, CD137, CD197, CD223 (Lag3), CD279 (PD1), and CD366 (TIM3) (BD Biosciences, NJ, Biolegend, CA and Affymetrix, CA, USA).
In vivo murine model of hematopoietic stem cell transplantation rejection
C57BL/6 Ifngr1-/- mice were used as recipient, while C57BL/6 Ifngr1+/+ were used as donor. All animal experiments were per- formed in accordance with the Swiss animal protection law. Details on experiments are reported in the Online Supplementary Appendix.
Emapalumab administration in compassionate use to hemophagocytic lymphohistiocytosis patients experiencing graft failure
Emapalumab (previously known as NI-0501), a fully human anti-IFNγ monoclonal antibody, was administered on a CU basis (after local ethical committee approval) to three patients affect- ed by HLH who experienced GF after a first TCD HSCT from a partially-matched family donor (PMFD) with the aim of pre- venting flares of HLH and a second GF. The drug was adminis- tered by 1-hour intravenous infusion twice a week until sus- tained donor engraftment or GF. The dose varied between 1 and 6 mg/kg, based on pharmacokinetic data.
Additional methods are presented in the Online Supplementary Appendix.
Statistical analysis
Unless otherwise specified, quantitative variables were reported as Mean±Standard Error of Mean (SEM); categorical variables were expressed as absolute value and percentage. Clinical characteristics of patients were compared using the χ2 test or Fisher exact test for categorical variables, while the Mann-Whitney rank sum test or the Student t-test (two-sided) was used for continuous variables, as appropriate. For multiple comparison analyses, statistical significance was evaluated by a repeated measure ANOVA test, followed by a Log-rank (Mantel-Cox) test for multiple comparisons.
haematologica | 2019; 104(11)
2315