Abstract
Introduction
Allogeneic stem cell transplantation (HCT) survivors are at a relevant risk of developing long-term complications such as chronic GvHD (cGvHD), which importantly affects their quality of life and increases their morbidity and mortality. Being able to early identify high risk patients would enable us to tailor preventive strategies. Current approach on prophylaxis of GvHD is lacking of predictive biomarkers that could guide patient-tailoring of drugs choice, tapering and treatment schedules. Immune system is the cause of cGvHD but is also a target of it, and cGvHD patients are characterized by lymphoid hypocellularity. Moreover, immune reconstitution (IR) is a good candidate biomarker being it an easily-available and reproducible parameter. We investigated IR variables as predictive biomarker of cGvHD.
Methods
A standardized follow-up of HCT-survivors is applied at our center. We analyzed 307 adult patients consecutively undergoing first allogeneic HCT transplant between July 2012 and December 2016 at our Institution. A written consent was given for the use of medical records for research in accordance with the Declaration of Helsinki. Median follow-up for surviving patients was 2.8 years (range 1.1-5.5). We prospectively collected IR data of our entire cohort at specific time-points (+30, +60, +90, +180, +365 days) and followed patients up recording events. IR variables were CD3+, CD3+CD4+, CD3+CD8+, CD19+, CD56+ cell counts, measured by flow-citometry, and immunoglobulins IgG, IgA and IgM levels, measured by immunoturbidimetric assays. Time as a continuous parameter could not be studied since the number of events would have been too low for the analysis. For this reason, a series of landmark analyses were performed at 3, 6 and 12 months post-HCT in order to identify predictive factors of cGvHD, transplant-related mortality (TRM), progression-free survival (PFS) and overall survival (OS) for patients alive and in good conditions at the beginning of each time interval. Factors predicting cGvHD incidence and survival endpoints were studied using multivariate analysis by Cox regression model. Variables included in the model were patient and donor age and Sorror-Comorbidity Index (according to median values), disease-related index, type of donor, stem cell source, IR values at the timepoint according to landmark cut-off for cell counts and median values for immunoglobulin levels. A backward stepwise procedure was used for variable selection with a p-value <0.05. All statistical analyses were performed with R (R Development Core Team, Vienna, Austria) software package.
Results
Chronic-GvHD of any grade and severity was diagnosed in 111 patients. Immune recovery in our cohort was in line with the current knowledge: CD3+CD8+ and NK (CD56+) cells normalized first, followed by CD3+ and CD3+CD4+ cell. B cells (CD19+) took at least 1 year to normalize in terms of absolute counts. IgM levels were the first to rise among immunoglobulins, followed by IgG and then IgA which can also be subnormal for a long time after transplant. Results of multivariate analysis are shown in Table 1. Single lymphocyte subset counts did not prove to be associated to cGvHD onset significantly; conversely, immunoglobulins were strongly predictive of cGvHD in our multivariate model. Median time to GvHD onset was 198 days, thus the most important analysis was the one performed at +90 days as the majority of patients had still not developed cGvHD. IgG, IgA and IgM at +90 days from HCT below the median value were found before the onset of GvHD and could predict its onset. This data were confirmed on the analysis at later time-points in which low IgG levels predicted cGvHD diagnosis.
Conclusions
Day-90 low immunoglobulin levels predict cGvHD, confirming that subclinical immune dysregulation mechanisms could be already present before overt clinical onset of cGvHD symptoms. Early prediction of subsequent cGvHD will be operationally translated into patient-tailored preventive measures.
Bonini:Intellia Therapeutics: Research Funding.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal