The immune microenvironment in mantle cell lymphoma : Targeted liquid and spatial proteomic analyses

Sammanfattning: The complex interplay of the tumour and immune cells affects tumour growth, progression, and response to treatment. Restorationof effective immune response forms the basis of onco-immunology, which further enabled the development of immunotherapy. Inthe era of precision medicine, pin-pointing patient biological heterogeneity especially in relation to patient-specific immunemicroenvironment is a necessity for the discovery of novel biomarkers and for development of patient stratification tools for targetedtherapeutics. Mantle cell lymphoma (MCL) is a rare and aggressive subtype of B-cell lymphoma with poor survival and high relapserates. Previous investigations of MCL have largely focused on the tumour itself and explorations of the immune microenvironmenthave been limited. This thesis and the included five papers, investigates multiple aspects of the immune microenvironment withrespect to proteomic analysis performed on tissue and liquid biopsies of diagnostic and relapsed/refractory (R/R) MCL cohorts.Analyses based on liquid biopsies (serum) in particular are relevant for aggressive cases such as in relapse, where invasiveprocedures for extracting tissues is not recommended. Thus, paper I-II probes the possibility of using serum for treatment andoutcome-associated biomarker discovery in R/R MCL, using a targeted affinity-based protein microarray platform quantifyingimmune-regulatory and tumor-secretory proteins in sera. Analysis performed in paper I using pre-treatment samples, identifies 11-plex biomarker signature (RIS – relapsed immune signature) associated with overall survival. Further integration of RIS with mantlecell lymphoma international prognostic index (MIPI) led to the development of MIPIris index for the stratification of R/R MCL intothree risk groups. Moreover, longitudinal analysis can be important in understanding how patient respond to treatment and thiscan further guide therapeutic interventions. Thus, paper II is a follow-up study wherein longitudinal analyses was performed onpaired samples collected at pre-treatment (baseline) and after three months of chemo-immunotherapy (on-treatment). We showhow genetic aberrations can influence systemic profiles and thus integrating genetic information can be crucial for treatmentselection. Furthermore, we observe that the inter-patient heterogeneity associated with absolute values can be circumvented byusing velocity of change to capture general changes over time in groups of patients. Thus, using velocity of change in serumproteins between pre- and on-treatment samples identified response biomarkers associated with minimal residual disease andprogression. While exploratory analysis using high dimensional omics-based data can be important for accelerating discovery,translating such information for clinical utility is a necessity. Thus, in paper III, we show how serum quantification can be usedcomplementary tissue-identified prognostic biomarkers and this can enable faster clinical implementation. Presence of CD163+M2-like macrophages has shown to be associated with poor outcome in MCL tissues. We show that higher expression of sCD163levels in sera quantified using ELISA, is also associated with poor outcome in diagnostic and relapsed MCL. Furthermore, wesuggest a cut-off for sCD163 levels that can be used for clinical utility. Further exploration of the dynamic interplay of tumourimmunemicroenvironment is now possible using spatial resolved omics for tissue-based analysis. Thus, in paper IV and V, weanalyse cell-type specific proteomic data collected from tumour and immune cells using GeoMx™ digital spatial profiler. In paperIV, we show that presence as well as spatial localization of CD163+ macrophage with respect to tumour regions impactsmacrophage phenotypic profiles. Further modulation in the profile of surrounding tumour and T-cells is observed whenmacrophages are present in the vicinity. Based on this analysis, we suggest MAPK pathway as a potential therapeutic target intumours with CD163+ macrophages. Immune composition can be defined not just by the type of cells, but also with respect tofrequency and spatial localization and this is explored in paper V with respect to T-cell subtypes. Thus, in paper V, we optimizeda workflow of multiplexed immunofluorescence image segmentation that allowed us to extract cell metrics for four subtypes ofCD3+ T-cells. Using this data, we show that higher infiltration of T-cells is associated with a positive outcome in MCL. Moreover,by combining image derived metrics to cell specific spatial omics data, we were able to identify immunosuppressivemicroenvironment associated with highly infiltrated tumours and suggests new potential targets of immunotherapy with respect toIDO1, GITR and STING. In conclusion, this thesis explores systemic and tumor-associated immune microenvironment in MCL, fordefining patient heterogeneity, developing methods of patient stratification and for identifying novel and actionable biomarkers.