Marcadores inmunológicos predictores de evolución del trasplante

Abstract

Antecedentes. La alorrespuesta humoral, particularmente la mediada por la presencia de anticuerpos frente a los antígenos leucocitarios humanos (HLA) del aloinjerto, se asocia a un incremento en la prevalencia de rechazo y a una reducción en la supervivencia del trasplante. La mayor sensibilidad de los ensayos de fase sólida, basados en el empleo de microesferas magnéticas recubiertas con un único antígeno HLA (single antigen beads o SAB), los consolidó como técnica gold-estándar para identificar estos anticuerpos anti- HLA, garantizando un mayor éxito en los programas de trasplante de órgano sólido. La intensidad de fluorescencia media (MFI) proporcionada por el ensayo estandarizado (SABpanIgG) es el valor que se usa habitualmente para estratificar el riesgo inmunológico, asumiendo que es una estimación semicuantitativa fiable del nivel de anticuerpos circulantes. Sin embargo, este valor se ve frecuentemente afectado por ciertos efectos inhibitorios, que enmascaran la concentración real a la que se encuentra el anticuerpo en sangre. Más allá del valor MFI, ciertas características intrínsecas, como su capacidad para activar la cascada del complemento (SAB-C1q) o el perfil de subclases de IgG1-4 que lo conforma (SAB-subclass), han sido estudiadas con el objetivo de elucidar sus propiedades funcionales y definir de forma más precisa la relevancia clínica de los anticuerpos, usados como biomarcadores de la evolución del trasplante. No obstante, todavía existen cuestiones por resolver. Objetivo. El objetivo de este trabajo de tesis doctoral fue esclarecer la relación entre las diferentes propiedades de los anticuerpos anti-HLA identificables por los métodos de detección disponibles (valor MFI, capacidad de activar la cascada del complemento y perfil de subclases de IgG1-4) y evaluar prospectivamente el impacto de dichas propiedades sobre la evolución del aloinjerto renal trasplantado. Métodos. Para estudiar la relación entre las diferentes propiedades de los anticuerpos anti-HLA utilizadas en la estratificación del riesgo inmunológico, se analizaron las muestras de suero de 20 pacientes HLA-sensibilizados en lista de espera de trasplante renal, utilizando los ensayos de fase sólida disponibles: SAB-panIgG (para detectar y definir las especificidades de anticuerpos anti-HLA panIgG), SAB-subclass (para identificar la presencia de subclases IgG1-4 que componen una determinada especificidad de anticuerpo anti-HLA) y SAB-C1q (para definir los anticuerpos capaces de fijar C1q, el primer componente de la cascada del complemento humano por la vía clásica). Los ensayos SAB-panIgG y SAB-subclass se realizaron sobre muestras de suero neto y previamente diluido 1:16. Además, en el ensayo SAB-panIgG, se evaluaron los pretratamientos con calor (56 ºC durante 30 minutos) y EDTA (25 mM) como métodos para eliminar el efecto inhibitorio. Por otra parte, para estudiar el impacto de las propiedades de los anticuerpos anti-HLA, se examinó la función, ocurrencia de eventos de rechazo y supervivencia temprana del aloinjerto renal en una cohorte de 14 pacientes hiperinmunizados trasplantados con DSA preformados no fijadores de C1q (DSA+), en relación a una cohorte control de 58 pacientes trasplantados con anticuerpos anti-HLA preformados no específicos de donante (DSA-). Asimismo, se exploraron los cambios producidos en el perfil del DSA, así como en sus propiedades intrínsecas (valor MFI, capacidad de activar la cascada del complemento y perfil de subclases IgG1-4). Resultados. Al analizar las muestras de suero de los 20 pacientes en lista de espera de trasplante, se identificaron 1.285 anticuerpos anti-HLA como positivos, 473 (36,8%) de los cuales resultaron ser fijadores de C1q. La dilución de las muestras de suero aumentó la correlación entre la fuerza de los anticuerpos, estimada como el valor MFI, y su capacidad de unir C1q (rneto = 0,248 vs. rdiluido = 0,817). Los pre-tratamientos con calor y EDTA también aumentaron dicha correlación (rcalor = 0,699 y rEDTA = 0,656), si bien, ésta no fue tan elevada como la obtenida tras la dilución. El ensayo SAB-subclass reveló al menos una subclase de IgG1-4 en 1.012 (78,8%) de los anticuerpos positivos predefinidos en el ensayo SAB-panIgG. Las subclases con fuerte capacidad de activar complemento, principalmente IgG1, fueron particularmente frecuentes (98,9%) y, de hecho, no se encontraron diferencias entre anticuerpos C1q-positivos y C1q-negativos en relación a su presencia (99,4 vs. 98,5 %; p = 0,193). Por el contrario, las subclases incapaces o débilmente capaces de activar el complemento (IgG4/IgG2) se detectaron con más frecuencia en anticuerpos fijadores de C1q (78,9 vs. 38,6%; p < 0,001). La correlación entre la capacidad de unir C1q y la fuerza de la subclase IgG1 resultó ser fuerte (rIgG1 = 0,796). Aunque menor, la correlación entre la fuerza de la subclase IgG2 y la capacidad de unir C1q también resultó ser fuerte (rIgG2 = 0,758), estando ambas subclases estrechamente relacionadas (rIgG1-IgG2 = 0,817). No se encontró ninguna correlación entre la capacidad de unir C1q y la fuerza de las subclases IgG3 e IgG4. Por otra parte, en relación al análisis de las muestras de suero de los 14 pacientes trasplantados con DSA preformados no fijadores de C1q (DSA+), se observó que el valor MFI del DSA disminuyó significativamente en el post-trasplante (4.467,28 ± 2.052,95 vs. 2.030,25 ± 1.979,36; p = 0,002) en 12 (85,7%) de los pacientes, mientras que en 2 (14,3%) de ellos aumentó. Ningún DSA modificó su estatus C1q. El valor MFI de las subclases del DSA también disminuyó significativamente, excepto el de la subclase IgG3, que permaneció invariable (IgG1: 5.443,36 ± 9.033,25 vs. 1.488,91 ± 2.035,98; p = 0,028; IgG2: 752,34 ± 1.958,23 vs. 17,24 ± 15,04; p = 0,022; IgG3: 88,50 ± 215,23 vs. 91,93 ± 140,22; p = 0,386 e IgG4: 257,89 ± 612,41 vs. 47,66 ± 63,39; p = 0,037). La proporción de pacientes que desarrollaron DSA de novo fue similar en el grupo DSA+ con respecto al grupo DSA- (7,1 vs. 5,2%; p = 1,000). Los valores de creatinina sérica (mg/dL), filtrado glomerular (mL/min/m2), proteínas en orina (mg/dL) y cociente creatinina/proteínas en orina (mg/mg) tampoco fueron significativamente diferentes (p > 0,05) en ninguno de los intervalos de tiempo analizados (15, 30, 90, 180 y 365 días post-trasplante). En el grupo DSA-, una proporción significativamente mayor de pacientes fue diagnosticada de rechazo celular (TCMR; 0 vs. 41,7%; p = 0,035), mientras que hubo una clara tendencia en la supervivencia renal similares unidos a la invariabilidad en las propiedades funcionales del DSA asociadas a su potencial patológico (principalmente su capacidad de unir C1q), durante el primer año de seguimiento, sugieren que el trasplante renal en presencia de DSA preformados no fijadores de C1q podría ser factible en aquellos pacientes altamente sensibilizados con pocas opciones de trasplante.

Background. Humoral alloresponse, particularly that mediated by the presence of antibodies against allograft human leukocyte antigens (HLA), is associated with an increased prevalence of rejection and a reduced transplant survival. The higher sensitivity of solid phase assays, based on magnetic microbeads coated with single HLA antigens (single antigen beads or SAB), consolidated them as the gold-standard method to identify these anti-HLA antibodies, ensuring a greater succeed in solid-organ allograft allocation programs. Mean fluorescence intensity (MFI) value provided by the standardized assay (SAB-panIgG) is regularly used to stratify the immunological risk, assuming it as a reliable semiquantitative estimation of the circulating antibody-level. However, this value is often affected by several inhibitory effects, which mask the real concentration of antibodies in blood. Beyond MFI, certain intrinsic characteristics, such as the complement-binding ability (SAB-C1q) or the IgG1-4 subclass profile comprising a particular specificity (SABsubclass), have been examined to clarify their functional properties and more accurately define the clinical relevance of antibodies, used as biomarkers of transplant outcome. However, there are still unresolved issues. Aim. The aim of this study was to elucidate the relationship between the different properties of anti-HLA antibodies detectable with the available assays (the MFI value, the ability to activate the complement cascade and the IgG1-4 subclass profile) and prospectively evaluate the impact of those properties on kidney transplanted allograft outcome. Methods. In order to study the relationship between different anti-HLA antibody properties use to stratify the immunological risk, serum-samples from 20 HLA-sensitized patients awaiting kidney transplantation were analyzed by the available solid-phase assays: SAB-panIgG (to detect and define the specificities of panIgG anti-HLA antibodies), SAB-subclass (to identify the presence of IgG1-4 subclasses comprising a particular anti HLA antibody specificity) and SAB-C1q (to define those antibodies capable of binding C1q, the first component of human complement classical pathway). SAB-panIgG and SABsubclass were performed on neat and 1:16 pre-diluted serum-samples. Additionally, in SAB-panIgG, heat (56 ºC for 30 minutes) and EDTA (25 mM) pre-treatments were evaluated as methods to avoid the inhibitory effect. Furthermore, in order to study the impact of anti-HLA antibody properties, allograft function, the occurrence of rejection events and early allograft survival rate were evaluated in a cohort of 14 highly-sensitized transplanted patients with preformed non-C1q-binding DSA (DSA+) and contrasted with a control cohort of 58 transplanted patients with preformed non-DSA anti-HLA antibodies (DSA-). Changes produced in the profile and in the intrinsic properties (the MFI value, the ability to activate the complement cascade and the IgG1-4 subclass profile) of the DSA were also explored. Results. When analyzing serum samples belonging to the 20 patients awaiting kidney transplantation, a total of 1,285 anti-HLA antibodies were identified as positive, 473 (36.8%) of which were C1q-binding. Serum-dilution enhanced the correlation between the antibody-strength, measured as the MFI and the C1q-binding ability (rneat = 0.248 vs. rdiluted = 0.817). Even though the heat and EDTA pre-treatments also improved that correlation (rheat = 0,699 y rEDTA = 0,656), it was not as high as that obtained after serum-dilution. SAB-subclass assay revealed at least one IgG1-4 subclass in the 1,012 (78.8%) positive antibody-specificities predefined by SAB-panIgG. Strong complement-binding subclasses, mainly IgG1, were particularly frequent (98.9%) and no differences were found between C1q-positive and C1q-negative antibodies regarding their presence (99.4% vs. 98.5%; p=0.193). In contrast, non- or weak C1q-binding subclasses (IgG4/IgG2) were more commonly detected in C1q-binding antibodies (78.9% vs. 38.6%; p < 0.001). Interestingly, a strong association was found between the C1q-binding ability and the IgG1 strength (rIgG1dil = 0.796). Though lower, the correlation between the IgG2 strength and the C1q binding ability was also strong (rIgG2dil = 0.758), being both subclasses closely related (rIgG1- IgG2 = 0.817). No correlation was found with the C1q-binding ability and the strength of IgG3/IgG4 subclasses. Furthermore, regarding the analysis of serum samples belonging to the 14 patients transplanted with preformed non-C1q-binding DSA (DSA+), it was observed that in 12 (85.7%) patients, the DSA MFI value significantly decreased after transplantation (4,467.28 ± 2,052.95 vs. 2,030.25 ± 1,979.36; p = 0.002), whereas in 2 (14.3%) of them it increased. No DSA changed their C1q-binding status. The MFI value of IgG-DSA subclases significantly decreased as well, except that of the IgG3, which remained invariable (IgG1: 5,443.36 ± 9,033.25 vs. 1,488.91 ± 2,035.98; p = 0.028; IgG2: 752.34 ± 1,958.23 vs. 17.24 ± 15.04; p = 0.022; IgG3: 88.50 ± 215.23 vs. 91.93 ± 140.22; p = 0.386 and IgG4: 257.89 ± 612.41 vs. 47.66 ± 63.39; p = 0.037). The proportion of patients who developed de novo DSA was similar in the DSA+ with regard to the DSA- group (7.1 vs. 5.2%; p = 1.000). Neither the creatinine serum level (mg/dL), glomerular filtration rate (mL/min/m2), urine protein level (mg/dL) nor the urine creatinine/protein ratio (mg/mg) were significantly different (p < 0.05) between groups at any point throughout the follow-up time (15, 30, 90, 180 and 365 days after transplantation). While a significantly higher proportion of patients in the DSA- group were diagnosed with T-cell mediated rejection (TCMR; 0 vs. 41.7%; p = 0.035), a clear trend was found in the proportion of biopsies with C4d staining in the DSA+ group (75.0 vs. 33.3%; p =0.096). No significant differences were found in the occurrence of TCMR with humoral component (62.5 vs. 33.3%; p = 0.219) nor in the occurrence of antibody-mediated rejection without cellular component (ABMR; 12.5 vs. 8.3; p = 1.000). Early kidney allograft survival until the end of the follow-up time was similar between both groups (85.7 vs. 93.2%; p = 0,329). Conclusions. The different properties of anti-HLA antibodies used to define their pathological potential (the MFI value, the ability to bind C1q and the circulating IgG1-4 subclass profile) seem to be related, albeit in a distinct manner as it initially could be expected. Thus, a particular profile of IgG subclasses (IgG1/IgG3 or IgG2/IgG4) itself does not determine at all the ability to bind complement of anti-HLA antibodies, given that the first (IgG1/IgG3), which are potent complement activators, are present both in C1qpositive and C1q-negative antibodies and the second (IgG2/IgG4) are, paradoxically, more frequent in C1q-positive antibodies. It is the subclass strength (MFI), mainly that of the ubiquitous IgG1, which usually appears combined with IgG2, that best correlates with the ability to bind C1q. That real strength (MFI) is more exactly estimated when analyzing diluted serum-samples, which allows to avoid the prozone effect, a common artifact in highly HLA-sensitized patients. Under an immunosuppressant context, the MFI value of the DSA and the particular IgG subclasses comprising it decreases after transplantation, while its C1q-binding ability, a proven worse prognosis biomarker, uses to remain invariable. Despite the undeniable interaction between the DSA and the kidney endothelium, evinced by the presence of C4d staining, early allograft function and survival rate of patients transplanted with preformed non-C1q-binding DSA is similar with respect to those of patients transplanted without DSA. The similar functionality and renal survival together with the invariability of DSA functional properties associated with their pathological potential (mainly the ability to bind C1q), during the first year of follow-up, suggest that kidney transplantation with preformed non-C1q-binding DSA could be feasible in those highly-sensitized patients with scarce transplantation recourses.