Scientific context

Epstein-Barr virus has a causal role in the pathogenesis of multiple distinct lymphomas. Post-transplant lymphoproliferative diseases (PTLD) are the most frequent EBV-induced proliferations. PTLD after allogeneic stem cell transplantation has an incidence lower than 5% but may increase up to 10-20% in patients with established risk factors. EBV-DNAemia is predictive of EBV-PTLD and is routinely performed using qPCR on whole blood. Preemptive therapeutic strategies with anti-CD20 antibody are used when patients are above a defined EBV-DNAemia threshold. This approach remains limited since it does not discriminate between an EBV-induced lymphoproliferation (latent cycle) and/or a replicating virus (replicative cycle).

Epigenetic modifications plays a central role in regulating the switch from latent to lytic gene expression. Specific DNA modifications can be regarded as molecular signatures for EBV genomes associated with the status of the viral infection (latent vs lytic). Accordingly, these signatures may be envisioned as a potent tool to characterize the state of the viral infection in vivo.

Description of the project Our primary objective is to estimate the respective percentages of EBV-lytic and EBV-latent genomes (proliferating cells) in patients presenting with a high EBV-DNAemia after allogeneic stem cell transplantation HSCT by analysing the epigenetic modifications of EBV genome on specific sites. Our secondary objectives are i) to determine risk factors associated with each "latent versus lytic EBV" profiles and ii) to correlate the "latent versus lytic EBV" profiles with response to rituximab infusion and patient outcomes.

For this purpose, a retrospective study (n=80) and a prospective study (estimation n=58) will be established. The different steps of this project are:

1. To study epigenetic modifications. The laboratory is developing a new approach to distinguish between latent and lytic genomes.
2. To realize quantitative analysis by RT-PCR of different EBV transcripts specific of the latent or of the lytic phase of the virus This method will be applied on RNA extracted from patient blood samples with elevated EBV viral load, under condition preserving RNA integrity. The results will be validated on a prospective cohort of HSCT patients (n=58) (Saint-Antoine Hospital and La Pitié-Salpêtrière Hospital).
3. To perform quantitative analysis of EBV genomes in plasma, saliva and total blood samples by current routine procedures In addition to total blood samples, plasma and saliva will be collected since free viral particles are known to accumulate in these biological fluids upon EBV reactivation. These samples will be treated by normalized procedures that are routinely used in the medical virology laboratory to quantify EBV in human samples.

Expected results By establishing a simple method for studying epigenetic modifications of EBV genomes, we expect to understand the significance of high EBV viral load and the pathophysiology of post-HSCT PTLD. We aim to distinguish between the latent / lytic profiles of HSCT patients and to correlate their respective risks for developing PTLD. Establishing the epigenetic EBV profile in the post-HSCT setting when facing increase viral load and PTLD will improve our understanding of the biological mechanisms determining EBV-status in post-HSCT. This should improve major medical and economical issues. These results could have a major therapeutic

Study duration :

Total study duration: 37 months Duration of recruitment: 24 months Duration of participation for each patient: 13 months

Inclusion criteria:

Retrospective study: Patient who underwent HSCT, in Saint-Antoine hospital, between 2010-2015, treated by rituximab for high level EBV-DNAemia (above 3.3 log copies/mL).

Prospective study: Patients who underwent HSCT, in Saint-Antoine hospital and la Pitié-Salpêtrière, in 2016-2017, treated by rituximab for high level EBV-DNAemia (above 10 000c/mL), And/or having post-transplant lymphoproliferative diseases(PTLD) Concerned population Allogeneic hematopoietic stem cell transplantation (allo-HSCT) patients with a high Epstein-Barr virus (EBV) viral load

Study endpoints:

Primary endpoint (linked to the primary objective) Respective percentages of latent versus replicative EBV profiles, defined by the methylation level of specific sites of EBV genome, in a cohort of HSCT patients with a high EBV viral load (above 3.3 log copies/ml) before the initiation of treatment per rituximab. A methylation index will be calculated from the respective differences of the "Cycle thresholds" of the methyl-qPCR performed on patients blood samples Secondary endpoints (linked to the secondary objectives)

1. Characteristics of the patients according to the EBV q-PCR results

2. Patients' outcome according to the EBV profile defined by the methyl-qPCR :

   • Response to Rituximab infusion (EBV DNA-emia < 3.3 log copies/mL) after a maximum of 4 infusions)
   • Disease-free survival (DFS) at 12 months defined as survival without relapse
   • Overall survival (OS) at 12 months
   • Relapse incidence (RI) at 12 months
   • Non-relapse mortality (NRM) at 12 months
   • Number of post-HSCT infections within 12 months

Statistical Analysis :

1. Comparison between characteristics of latent and lytic genomes will be performed using parametric (Student t-test) or non-parametric (Mann Whitney test) when appropriate for continuous variables, Chi-Square test for qualitative variables.

   Multivariate analysis for determination of risk factors associated with "lytic EBV" profiles will be performed using logistic regression.

2. Correlation between "latent versus lytic EBV" profiles" and response to rituximab infusion will be performed using logistic regression including in the model all other factors potentially associated to the response to rituximab. OS and DFS will be estimated by using the Kaplan-Meier method. Cumulative incidences of RI and NRM will be calculated from the date of inclusion to the date of relapse or death in remission, respectively, with the other event being the competing risk. Death or progression will be considered as a competing events for estimating the incidence of acute GVHD. Multivariate analyses for survival endpoints will be performed using the Cox proportional hazards model. The type I error rate is fixed at 0.05 for determination of factors associated with time to event outcomes. All statistical analyses will be performed with R software packages (R Foundation for Statistical Computing, Vienna, Austria)

Inclusion Criteria:

Retrospective study:

• Patient who underwent HSCT, in Saint-Antoine hospital,
• between 2010-2015,
• treated by rituximab for high level EBV-DNAemia (above 3.3 log copies/mL).

Prospective study:

• Patients who underwent HSCT, in Saint-Antoine hospital and la Pitié-Salpêtrière,
• in 2017-2018,
• treated by rituximab for high level EBV-DNAemia (above 10 000c/mL),
• And/or having post-transplant lymphoproliferative diseases(PTLD)

Exclusion Criteria:

• Refusal to sign the informed consent
• Patient non-beneficiary of the Social Security system

1 population divided in 2 groups: Retrospective study (n=80): Patient who underwent HSCT, in Saint-Antoine hospital, between 2010-2015, treated by rituximab for high level EBV-DNAemia (above 3.3 log copies/mL).

Prospective study (n=58): Patients who underwent HSCT, in Saint-Antoine hospital and la Pitié-Salpêtrière, in 2016-2017, treated by rituximab for high level EBV-DNAemia (above 10 000c/mL), And/or having post-transplant lymphoproliferative diseases(PTLD) Concerned population Allogeneic hematopoietic stem cell transplantation (allo-HSCT) patients with a high Epstein-Barr virus (EBV) viral load