A New Routine Immunity Score (RIS2020) to Predict Severe Infection in Solid-Organ Transplant Recipients

Introduction

Severe infection, mainly bacterial infection, is a leading cause of morbidity and mortality in solid-organ recipients during the first year after transplantation [1–4]. Infections are also associated with prolonged hospital stays and high healthcare costs [5]. Considering that antibiotic-resistant bacterial infections could be the leading cause of death by the year 2050, the optimization of resources for prediction of infection will play a key role in management of immunosuppressed patients [6]. The increase in the number of patients with secondary immunodeficiencies (eg, transplant recipients) is expected to increase the frequency of antimicrobial resistance. Therefore, new solutions that help predict and treat these complications are urgently required.

The usefulness of immunological monitoring to predict infections and other transplant-related complications and their severity has been demonstrated in various studies. Furthermore, the COVID19 pandemic highlighted its possible role at a time when resources were scarce and the need to identify cases of severe infection was more urgent [7–9]. A similar example can be seen in the transplant setting, where guidelines recently included the functional activity of CD8 T cells as a predictor of CMV infection [10]. Given the complexity of the immune response against microbial antigens, it is unlikely that a single biomarker can predict all outcomes in immune monitoring [11].

The primary objective of this study was to evaluate a new routine immunity score, RIS2020, which is applied during the first month after transplantation. The score includes routine biomarkers of innate and acquired immunity to predict infections in solid-organ transplant recipients. We aimed to ascertain whether the score added predictive value to the determination of isolated IgG and other immunological parameters as risk factors for infection. The components of this score have been evaluated in previous studies. The link between IgG hypogammaglobulinemia and severe infection has been demonstrated in several single-center and multicenter studies and in meta-analyses [11–17]. Hypocomplementemia has been less investigated as a biomarker of severe infection in solid-organ transplant recipients [11,12,18], as has the role of quantification of lymphocyte subpopulations [11,12,18–20].

We previously reported an immunological score in a small number of heart transplant recipients. The combination of low levels of IgG, C3, C4, CD4, and NK cells was more strongly associated with risk of infection than IgG hypogammaglobinemia alone [11]. Based on a T cell-depleting protocol in kidney transplant recipients, Fernandez-Ruiz et al [12] demonstrated that the combination of IgG hypogammaglobulinemia, CD8 lymphopenia, and C3 hypocomplementemia was associated with the risk of infection. However, the combined use of biomarkers of innate and acquired immunity must be validated before it can be applied in clinical protocols. Moreover, we still lack studies that address the controversies associated with their predictive ability.

Material and Methods

The authors designed an international prospective study to test the primary objective of evaluating the effectiveness of an immunological score, RIS2020, in predicting infections within the first month after transplantation.

The COVID19 pandemic greatly limited the inclusion of centers, and, eventually, 2 centers in Madrid, Spain (n=353) and Lima, Peru (n=24) participated. We took advantage of available data from a prospective immune monitoring database at our center in Madrid. The recruitment period differed between the centers and was affected by the COVID19 pandemic at the end of recruitment (2020–2022). Recruitment in Lima ran from 2021 to 2022. In Madrid, the recruitment period was longer, running between 2008 and 2022 (heart), 2019 and 2022 (liver), and 2021 and 2022 (kidney). These difficulties had a greater impact on the number of kidney transplant patients included, which was very low.

We included patients who were on the waiting list for SOT and then underwent heart, kidney, or liver transplantation and received protocol-based immunosuppressive therapy. Other inclusion criteria were age ≥18 years and ≤70 years and written informed consent. We excluded patients with a previous diagnosis of primary immunodeficiency.

Severe infections included bacterial infections and CMV disease. Severe bacterial infection was defined as documented infection (positive culture and/or clinical definition) that required at least 1 dose of intravenous antimicrobial therapy at any time during the first 6 months after transplantation. Orally treated Clostridium difficile infections were also considered severe infections. Catheter-associated bloodstream and urinary tract infections, and superficial incisional surgical site infections were not included.

CMV disease was defined as CMV infections detected by polymerase chain reaction assay and treated with oral valganciclovir or IV ganciclovir. CMV disease included tissue biopsy confirmed invasive disease, or a clinical diagnosis of CMV syndrome with at least 2 of the following: Fever, malaise, fatigue, leukopenia or neutropenia, atypical lymphocytes, thrombocytopenia, or elevation of hepatic aminotransferases. Asymptomatic CMV infections were not included.

Biomarkers were tested before transplantation at inclusion on the waiting list and at days 7 and 30 after transplantation. These time points were selected as they were early evaluation points preceding the onset of infection according to previous studies [11]. Clinical follow-up was based on a project-specific questionnaire during the first 6 months after transplantation. The questionnaire included clinical and laboratory variables that were considered to be associated with the risk of severe post-transplant infection.

The selected immunological biomarkers included serum levels of IgG, IgA, IgM, complement C3 and C4, and the absolute number and percentage of T CD3, T CD4, T CD8, B CD19 lymphocytes, and NK-cell subpopulations. Immunochemical parameters were quantified using nephelometry or turbidimetry. Lymphocyte subpopulations were studied using flow cytometry. Based on the criteria of the participating laboratories or data from previous reports on immunological scores, we applied commonly used definitions of IgG hypogammaglobulinemia (moderate, IgG <600 mg/dL, evaluated at day 7 or day 30 after transplantation; severe, IgG <400 mg/dL evaluated at day 7 or day 30 after transplantation), hypocomplementemia (C3 <80 mg/dL, C4 <20 mg/dL), low T cell counts (CD4 <400 cells/uL, CD8 <200 cells/uL), and high C-reactive protein values (CRP >3 mg/dL) [11–13]. Receiver operating characteristic (ROC) analysis was used to define the RIS2020 cut-off that indicated risk of severe infection.

Demographic donor and recipient variables were recorded in a database. Selected clinical complications were recorded during follow-up, as were infection episodes. Cellular rejection in heart recipients was defined as International Society for Heart and Lung Transplantation grades 2–4 necessitating treatment. Acute cellular rejection of the liver was defined as T cell-mediated damage to the liver allograft characterized by cellular infiltrates. In kidney recipients, rejection was clinically defined as a >0.3 mg/dL elevation in serum creatinine and diagnosis based on kidney biopsy. Blood samples were collected according to the protocol at baseline and at follow-up visits to determine biochemistry and hematological values.

The study sample was calculated based on a type-1 error (α-value) taken as 0.05, power (1-β) taken above 80%, effect size as the smallest clinically relevant difference in the outcome from previous studies that evaluated immunological scores [11,12], variance/standard deviation observed in previous studies of immunological scores [11,12], and the anticipated dropout rate based on the experience of researchers in previous studies.

The risk of developing severe infection was assessed using logistic regression analysis. First, univariate models were constructed to identify the variables significantly associated with the outcome variable. The immunological variables entered into the univariate models are those described above. Multivariate models were then constructed to adjust for possible confounding clinical variables. Variables with a p value of <0.5 from univariate analysis were entered into the multivariate logistic regression models. RIS2020 was set up according to the hazard ratio (HR) of each biomarker in univariate logistic regression models. The sum of the HRs was the score. The assignment of points based on the HR values was pre-specified according to our previous publication [11]. We rounded up to the nearest whole number based on the decimal of the HRs (>0.5). ROC analysis was used to choose the most appropriate cut-off for the RIS2020 score. The area under the curve (AUC) was used to determine the quality of this selection. The selected cut-off point was the value whose sensitivity and specificity were closest to that of the AUC. SPSS 15.0 was used for all statistical analyses.

Our study complied with the Declaration of Helsinki and was approved by the Institutional Ethics Review Board (Comité de Ética de la Investigación en Medicamentos, Ceim), Hospital General Universitario Gregorio Marañon, Madrid, Spain: Title of project: Estudio multicéntrico internacional de escalas de riesgo de infección en trasplante de órgano sólido, code FIS 1501472). Informed consent was obtained.

Results

IMMUNOLOGICAL RISK FACTORS FOR CMV DISEASE:

Moderate and severe IgG hypogammaglobulinemia at day 7 or 30 after transplantation were risk factors for development of CMV disease (HR 16.43, 95% CI 3.87–69.7, p<0.001 and HR 8.84, 95% CI 4.14–18.9, p<0.001, respectively). C3 hypocomplementemia and low CD8 count at day 30 after transplantation were risk factors for CMV disease (HR 2.76, 95% CI 1.08–7.06, p=0.034 and HR 2.97, 95% CI 1.080–8.09, p=0.035, respectively). Low CD4 counts at day 30 tended to be a risk factor for CMV disease (HR 2.18, 95% CI 0.76–6.28, p=0.15). The combination of moderate IgG hypogammaglobulinemia or severe IgG hypogammaglobulinemia at day 7 or 30 after transplantation, C3 hypocomplementemia at day 30, and low CD8 counts at day 30 were significantly associated with higher rates of CMV disease (HR 5.16, 95% CI 1.52–17.52, p=0.009 and HR 9.78, 95% CI 2.57–37.26, p=0.001, respectively).

Discussion

We evaluated RIS2020 during the first month after transplantation and found that it was associated with risk of severe infections in SOT. The effect size of the RIS2020 in the logistic regression model to predict severe infections was larger than any of the other biomarkers evaluated individually, including moderate IgG hypogammaglobulinemia alone (HR 5 vs HR 2 in other biomarkers).

From the mechanistic point of view, RIS2020 integrated several relevant mechanisms that are necessary to control post-transplant infections and reinfections, including low immunoglobulin levels (which predispose transplant recipients to bacterial, viral, and fungal infections [11–17]), C3 hypocomplementemia (which has been shown to predispose heart and kidney recipients to severe infection [11,12,21,22] and is associated with pronounced inhibition of opsonization, phagocytosis and oxidative burst [23]), CD4 lymphocytopenia (which has been described as a risk factor for infection in heart, kidney, and lung recipients [11,20,24]), and high CRP levels (which precede clinical suspicion of bloodstream infections in patients undergoing hematopoietic cell transplantation or have been described in kidney recipients at risk of death [25,26]).

In translational terms, RIS2020 is readily available and inexpensive and can be obtained shortly after blood extraction. It could be taken into account in future clinical trials that evaluate the impact of interventions such as new antimicrobial therapies, advanced therapies, and IVIG replacement therapy in affected transplant recipients to prevent or treat severe infection. Scores such as RIS2020, if properly validated, could also be useful for making decisions to prolong antimicrobial therapies in at-risk patients. Other studies have evaluated the role of immune monitoring in transplant patients before and after transplantation [7–9].

Another interesting aspect of our work was the IgG hypogammaglobulinemia cut-off point that was best associated with the risk of severe infection. Severe hypogammaglobulinemia (IgG <400 mg/dL) during the first year after transplantation significantly increased the risk of CMV and fungal and respiratory infections and was associated with higher 1-year all-cause mortality in a previous meta-analysis performed in SOT [17]. Indeed, since 2019, the presence of severe IgG hypogammaglobulinemia is an indication for IVIG replacement therapy in secondary antibody deficiency after the latest update from the European Medicines Agency. In our study, severe hypogammaglobulinemia was observed in 10% of patients. This may have limited the evaluation of its role as a risk factor for infections, since we only observed a statistical trend. Various studies have evaluated the role of post-SOT hypogammaglobulinemia by considering higher cut-off points. We previously showed that higher cut-off points are closely associated with infection in heart recipients (IgG <600 mg/dL), lung recipients (IgG <600 mg/dL), liver recipients (IgG <600 mg/dL), and kidney recipients (IgG <700 mg/dL) [11–16]. These observations further highlight the potential role of RIS2020, as it added predictive value to IgG hypogammaglobulinemia alone.

Moreover, given that replacement therapy with IVIG or subcutaneous immunoglobulins uses a very scarce resource, only the candidates who can benefit most should be selected. Our research team designed an expanded protocol for indicating replacement therapy with IVIG based on the score previously published in heart recipients and that includes post-transplant moderate IgG hypogammaglobulinemia, along with other immunological abnormalities included in RIS2020 [11]. This protocol was approved by the Pharmacy Commission and is in use.

In line with previous studies, we found that both moderate and severe IgG hypogammaglobulinemia were risk factors for CMV disease [13–15,17]. The combination of moderate IgG hypogammaglobulinemia with C3 hypocomplementemia and low CD8 counts early after transplantation also warrants further evaluation in future multicenter studies as a predictor of CMV disease.

The limitations of our study included the low number of centers owing to the COVID19 pandemic during the final period of recruitment, the low number of kidney recipients, and the low number of patients with severe IgG hypogammaglobulinemia. In addition, the recruitment period was excessively long in one center compared with the other. The small number of kidney transplants does not enable extrapolation of the results for this type of transplant. However, other studies have described similar associations with immunological scores in large cohorts of kidney transplants [12]. Another limitation is that our score is original and has not been previously validated in multicenter studies. The strengths of our study are its prospective design and the inclusion of highly reproducible biomarkers that are readily available in transplant centers.

Based on our results, we suggest that an immune signature integrating hypogammaglobulinemia, hypocomplementemia, low CD4+ T cell counts, and high CRP levels in peripheral blood during days 7 to 30 after SOT may be a better strategy than IgG hypogammaglobulinemia alone for identifying patients at high risk of severe infections who require a targeted effort for prevention and control of infection.

Conclusions

Our new immunological score combining moderate IgG hypogammaglobulinemia and other parameters of innate and acquired immunity (RIS2020) could better identify the risk for severe infection in SOT than IgG hypogammaglobulinemia alone. We believe that immunological monitoring after SOT should include the use of routine immune profiles that are readily available at transplant centers.