The Immunobiogram, a Novel In Vitro Assay to Evaluate Treatment Resistance in Patients Receiving Immunosuppressive Therapy

Abstract

Immunosuppressive drugs are widely used to treat several autoimmune disorders and prevent rejection after organ transplantation. However, intra-individual variations in the pharmacological response to immunosuppressive therapy critically influence its efficacy, often resulting in poor treatment responses and serious side effects. Effective diagnostic tools that help clinicians to tailor immunosuppressive therapy to the needs and immunological profile of the individual patient thus constitute a major unmet clinical need. In vitro assays that measure immune cell responses to immunosuppressive drugs constitute a promising approach to individualized immunosuppressive therapy. Here, we present the Immunobiogram, a functional pharmacodynamic immune cell-based assay for simultaneous quantitative measurement of a patient's immune response to a battery of immunosuppressive drugs. Peripheral blood mononuclear cells collected from patients are immunologically stimulated to induce activation and proliferation and embedded in a hydrogel mixture in which they are exposed to a concentration gradient of the immunosuppressants of interest. Analysis of samples from kidney transplant patients using this procedure revealed an association between the sensitivity of individual patients to the immunosuppressive regimen and their immunological risk of transplant rejection. Incorporation of the Immunobiogram assay into clinical settings could greatly facilitate personalized optimization and monitoring of immunosuppressive therapy, and study of the mechanisms underlying resistance to immunosuppressants.

Keywords: cellular pharmacodynamics; immune cell assay; immunosuppressive therapy monitoring; personalized medicine; transplant rejection.

Conflict of interest statement

This research involved the use of proprietary technologies belonging to BIOHOPE and covered by European Patent: EP 17 382 399.8 “METHOD FOR PREDICTING AND MONITORING CLINICAL RESPONSE TO IMMUNOMODULATORY THERAPY”. Inventors (current or former employees at Biohope): Javier Dotor de las Herrerias, Marianna di Scala, Veronica Sanchez, Isabel Portero Sanchez. AO, TD, and IP are employees at Biohope. LA and BD were employed by Adamas Engineering Consulting. The authors declare that this study was sponsored by Biohope. Biohope was beneficiary of a full funding from a competitive grant from the European Commission to conduct the study, and coordinated the study and the assignation of the adjudicated (and by EU audited) funding to the entities (research institutions, CRO and statistical company) that took part in the study. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2021 Portoles, Jimenez, Janeiro, Lopez-Oliva, Ortega-Carrion, Blanquez, Arribas, Gomez, Diez, Pascual and Portero.

Figures

Figure 1

Configuration of the Immunobiogram plate. Cartoon illustrating the location of the immunosuppressant (IMS) discs in the IMBG plate. No discs were placed in channels 3 and 8 (positive and negative controls, respectively). In channels 4–7, activated PBMCs were exposed to concentration gradients of the immunosuppressants indicated on the right-hand side of the cartoon. In the magnified image of the channel on the left-hand side, the horizontal lines indicate the 15 points at which fluorescence readings were taken along the concentration gradients of mycophenolic acid and sirolimus, respectively. Arrows indicate the direction of the concentration gradient, from maximum (closest to each disc) to minimum (at the mid-point of the channel, indicated by dashed red line).

Figure 2

Evaluation of PBMC activation: comparison of the Immunobiogram assay with flow cytometry analysis. Cell activation measured by flow cytometry (black bars) and with resazurin (gray bars), the redox indicator used in the IMBG assay in samples from five healthy blood donors. Cell activation is expressed as normalized fluorescence units and represented in the y axis. Fluorescence values obtained for each of the conditions tested are normalized to those obtained for conditions 1 (100% unstimulated cells) and 5 (100% stimulated cells). The x axis represents the different cells conditions tested: Condition 1, 0% stimulated cells + 100% unstimulated cells; Condition 2, 25% stimulated cells + 75% unstimulated cells; Condition 3, 50% stimulated cells + 50% unstimulated cells; Condition 4, 75% stimulated cells + 25% unstimulated cells; Condition 5, 100% stimulated cells + 0% unstimulated cells. Bar graphs represent mean (bar) and standard deviation (error bars).

Figure 3

Dose-response curve and key curve parameters. Example of dose-response curve generated based on the 15 fluorescence readings acquired for a given immunosuppressant. The x axis represents distance, normalized to a scale of 0–100, from the point of maximum immunosuppressant (IMS) concentration (0) to the point of minimum IMS concentration (100). The y axis represents cell activity, expressed as RFUs normalized to a scale of 0 (negative control [C-] value) to 100 (positive control [C+]). The following key curve parameters are indicated on the graph: maximal inhibitory response; minimal inhibitory response; ID50 (half maximal inhibitory response); and area under the curve (AUC).

Figure 4

Quadrant analysis to categorize patients based on immunobiogram results. (A) For a given immunosuppressant, a resistance map is generated after plotting the maximal inhibitory response (IRMAX) against the area under the curve (AUC) for the entire study population. The graph as then divided into 4 primary quadrants, based on the median IRMAX and AUC values, corresponding to sensitive (green), resistant (red) or partially sensitive (blue and salmon) profiles. The central area of the graph, defined by the 37.5–62.5% percentiles for IRMAX and AUC values, corresponds to a normal response (purple). (B) Each of the 4 primary quadrants was further subdivided into 4 subquadrants based on the respective Q1 and Q3 values for AUC and IRMAX, resulting in a resistance map consisting of 16 subquadrants, to which a specific numerical weight was assigned. Based on these values, the patient treatment score was calculated using the formula provided in the Materials and Methods.

Figure 5

Immunobiogram assay: experimental procedure. PBMCs are extracted from the patient’s blood sample and are immunologically stimulated to induce their activation and proliferation. These activated PBMCs are embedded in a hydrogel substrate, which is then loaded into segregated channels in the IMBG plate. PBMCs in each channel are then exposed to a concentration gradient of a different immunosuppressant, after which PBMCs activation and proliferation along the concentration gradient is measured using a resazurin-based assay, providing a fluorescence read-out of the immune cell response to each immunosuppressant. For each immunosuppressant, dose-response curves are generated based on the 15 immunofluorescence readings taken at sequential points along the concentration gradient in the IMBG channel.

Figure 6

Dose-response curves for each of the seven immunosuppressants tested in healthy blood donor samples. (A–G) IMBG curves (in gray) obtained for healthy blood donor samples normalized to positive (C+) and negative controls (C-), as well as the mean curve (in black), obtained for each of the immunosuppressants tested. The x axis represents distance, normalized to a scale of 0–100, from the point of maximum (0) to minimum (100) IMS concentration. The y axis represents cell activity, expressed as RFUs normalized to a scale of 0–100, where 0 and 100 represent the values obtained for the negative (C-) and positive (C+) control, respectively.

Figure 7

Boxplot depicting patient treatment score according to immunological risk of transplant rejection (low, intermediate, and high risk). Data are expressed as the median and interquartile range. Bars depict the maximum and minimum values of the data series. Patient treatment scores for the low-risk group were significantly lower than those for the high-risk group. *p < 0.05.