Proteomic and transcriptomic analyses identify apo-transcobalamin-II as a biomarker of overall survival in osteosarcoma

Abstract

Background:

The large-scale proteomic platform known as the SomaScan® assay is capable of simultaneously measuring thousands of proteins in patient specimens through next-generation aptamer-based multiplexed technology. While previous studies have utilized patient peripheral blood to suggest serum biomarkers of prognostic or diagnostic value in osteosarcoma (OSA), the most common primary pediatric bone cancer, they have ultimately been limited in the robustness of their analyses. We propose utilizing this aptamer-based technology to describe the systemic proteomic milieu in patients diagnosed with this disease.

Methods:

To determine novel biomarkers associated with overall survival in OSA, we deployed the SomaLogic SomaScan® 7k assay to investigate the plasma proteomic profile of naive primary, recurrent, and metastatic OSA patients. Following identification of differentially expressed proteins (DEPs) between 2-year deceased and survivor cohorts, publicly available databases including Survival Genie, TIGER, and KM Plotter Immunotherapy, among others, were utilized to investigate the significance of our proteomic findings.

Results:

Apo-transcobalamin-II (APO-TCN2) was identified as the most DEP between 2-year deceased and survivor cohorts (Log2 fold change = 6.8, P-value = 0.0017). Survival analysis using the Survival Genie web-based platform indicated that increased intratumoral TCN2 expression was associated with better overall survival in both OSA (TARGET-OS) and sarcoma (TCGA-SARC) datasets. Cell-cell communication analysis using the TIGER database suggested that TCN2+ Myeloid cells likely interact with marginal zone and immunoglobin-producing B lymphocytes expressing the TCN2 receptor (CD320) to promote their proliferation and survival in both non-small cell lung cancer and melanoma tumors. Analysis of publicly available OSA scRNA-sequencing datasets identified similar populations in naive primary tumors. Furthermore, circulating APO-TCN2 levels in OSA were then associated with a plasma proteomic profile likely necessary for robust B lymphocyte proliferation, infiltration, and formation of intratumoral tertiary lymphoid structures for improved anti-tumor immunity.

Conclusions:

Overall, APO-TCN2, a circulatory protein previously described in various lymphoproliferative disorders, was associated with 2-year survival status in patients diagnosed with OSA. The relevance of this protein and apparent immunological function (anti-tumor B lymphocyte responses) was suggested using publicly available solid tumor RNA-sequencing datasets. Further studies characterizing the biological function of APO-TCN2 and its relevance in these diseases is warranted.

1 Introduction

The current mainstay for clinical management of solid tumors, including the determination of cancer type and staging, is predicated on the analysis of tumor specimens collected through invasive biopsy and/or surgical resection (1). Recent characterization of these specimens through next-generation genomic sequencing technologies has not only helped guide the selection of targeted therapeutics in the age of personalized medicine and precision oncology (2–4), but has also been a major driver of cancer biomarker discovery (5, 6). The collection of tumor specimens for these analyses, however, can often pose critical challenges including obtaining sufficient sample quantity (biopsy), adequately characterizing tumors which have metastasized to multiple, often unresectable locations, and repeatedly monitoring therapeutic response over time (1).

Conversely, liquid biopsies, including the collection of patient blood, saliva, or urine, are diagnostic modalities offering inherent advantages over surgically collected tumor specimens. These advantages include their minimal invasiveness in addition to the possibility of serial sampling longitudinally throughout the course of disease (1, 7). The most comprehensively evaluated liquid biopsy to date is patient peripheral blood and its liquid (plasma, serum) and cellular components. Numerous biomarkers isolated from blood specimens including circulating tumor cells (CTCs) (8, 9), tumor exosomes (10, 11), and circulating tumor DNA (ctDNA) (12, 13) have offered invaluable insight into disease processes, prognosis, and therapeutic response for a variety of solid tumors (14, 15), including pediatric malignancies such as sarcoma (16–18).

Large-scale proteomic technologies (19), capable of simultaneously characterizing hundreds to thousands of proteins in these specimens, have also gained significant traction in the field of cancer biomarker research (20). While numerous techniques have been established (21, 22), next-generation aptamer-based multiplexed proteomic technology, first published by Gold et al. in 2010, has shown extraordinary promise. These aptamer-based platforms alleviate some of the inherent limitations of previous proteomic techniques such as mass spectrometry (MS) by offering increased sample throughput, larger dynamic ranges of detection, and lower average coefficients of variation, all while necessitating minimal sample volumes (23). The basis of this technology lies in the development of chemically modified aptamers, which form complex three-dimensional matrices with robust specificity to their target proteins. This new class of aptamers, known as Slow Off-rate Modified Aptamers (SOMAmers), are an evolution of the previously described short single-stranded oligonucleotides identified by Systemic Evolution of Ligands by Exponential (SELEX) enrichment in the early 1990s (23–25).

SOMAmers and the SomaScan® platform (26, 27), developed commercially by SomaLogic, Inc. (Boulder, CO, USA), have now been used to characterize a variety of cancers, including but not limited to hepatocellular carcinoma (28), colorectal cancer (29), pancreatic cancer (30), glioma (31), lung cancer (32), oral squamous cell carcinoma (33), and ovarian cancer (34). The platform has expanded its characterization from approximately 800 proteins in 2010 to well over 7000 proteins today, offering extensive insight into nearly all biological pathways relevant to human disease (35). Due to the SomaScan’s wide ranging assessment of proteins involved in the immune system including cytokine signaling, signaling by interleukins, and immunoregulatory pathways, among others, the platform has even been used to identify circulating proteins associated with immunotherapy response in diseases such as melanoma (36). As immunotherapies begin to dominate the world of oncology, systemic assessments of the immune system, in real time, are likely critical to monitor immunotherapeutic response and better predict clinical success. To our knowledge, no study has yet utilized the SomaScan® platform for proteomic biodiscovery in osteosarcoma (OSA).

OSA is the most common primary pediatric bone malignancy (37, 38). Patients with localized disease, treated with surgical resection and neoadjuvant/adjuvant multi-drug chemotherapy (39), display five-year survival rates greater than 75% (40, 41). Unfortunately, this rate decreases to approximately 25% in those with advanced disease, most commonly in the form of metastases to the lung (42). Numerous studies have been conducted to determine biomarkers of prognostic or diagnostic value in OSA using liquid biopsies (43–47). Of utmost relevance, serum biomarkers (48) including tumor necrosis factor (TNF) (49) and other interleukins (50), vascular endothelial growth factor (VEGF) (51–53) and angiogenin (ANG) (54), macrophage migration inhibitory factor (MIF) and T-Cell Immunoglobulin and Mucin Domain-Containing Protein 3 (TIM-3) (55), as well as various chemokines including C-X-C motif chemokine ligand (CXCL)4, CXCL6, and CXCL12 (56) have shown preliminary promise as diagnostic and/or prognostic biomarkers in this disease. Additionally, analysis of plasma proteomic profiles using surface-enhanced laser desorption/ionization-time of flight (SELDI-TOF) MS identified two proteins (amyloid protein A and transthyretin) involved in the innate immune system associated with positive response to chemotherapy (57). While limited in the robustness of their analyses, these studies ultimately support that characterization of the OSA systemic proteome can offer unique insight into disease progression, therapeutic response, and prognosis in this disease.

To determine novel biomarkers associated with overall survival in OSA, we utilized the SomaScan® 7k assay to investigate the plasma proteomic profile of primary, recurrent, and metastatic OSA patients. Plasma samples, isolated from patients with confirmatory diagnosis of OSA and treated at the University of Pittsburgh, Department of Orthopaedic Surgery, were processed and analyzed for the simultaneous quantification of over 7000 circulatory proteins (Figure 1A). Following the identification of differentially expressed proteins (DEPs) between 2-year deceased and survivor patient cohorts, various analyses were then conducted to suggest their biological relevance in OSA including investigation of publicly available bulk and single cell (sc)RNA-seq datasets, further associations with overall and progression free survival in solid tumors, as well as correlations with previously published biomarkers and gene signatures (Figure 1B).

Figure 1

2 Materials and methods

2.1 Patients, sample collection, and processing

The human subject research protocol and subject informed consent were reviewed and approved by the University of Pittsburgh Medical Center Institutional Review Board (IRB STUDY 19060152). According to IRB guidelines, informed consent was obtained from all participants (Table 1) before study entry. Peripheral blood specimens were collected from primary, recurrent, and metastatic OSA patients diagnosed from 2014-2021 via venipuncture into Cell-Free DNA BCT tubes (Cat. #218996, Streck, La Vista, NE, USA). Blood components (buffy coat, plasma) were then separated for storage following centrifugation, with time from sample collection to processing averaging 17.4 ± 3.6 hours (ranging 0.04 to 44.06 hours).

2.2 SomaLogic SomaScan® 7k assay

Profiling of the OSA proteome was performed using the SomaScan® 7k assay, a highly multiplexed aptamer-based proteomic technology capable of determining over 7,000 protein measurements, at SomaLogic headquarters according to a previous publication (58). Briefly, frozen OSA Streck plasma samples were thawed and then buffer exchanged twice into SomaLogic Assay Buffer using Zeba 7kDa MWCO spin desalting columns (Cat. #89882, Thermo Scientific, Waltham, MA, USA) according to manufacturer’s instructions. These buffer-exchanged plasma samples (diluted to 20%, 0.5%, and 0.005%) were then placed in a 96-well plate with a mix of thousands of chemically modified aptamers called SOMAmer® Reagents. These SOMAmers®, consisting of a 5’ fluorophore, photocleavable linker, and biotin on streptavidin-coated beads, form complex three-dimensional shapes which bind to epitopes on their target protein with high affinity and specificity. Following the formation of SOMAmer-protein complexes, unbound proteins are washed away, captured proteins are labeled with biotin, and the SOMAmer-protein complexes are released by photocleavage with ultraviolet light in a buffer containing an unlabeled polyanionic competitor. The biotin-labeled SOMAmer-protein complexes are captured on a second set of streptavidin-coated beads before releasing SOMAmer reagents in a denaturing buffer. The amount of available protein epitope is read by hybridizing the SOMAmer reagents in the SomaScan Assay eluate to complementary sequences on a DNA microarray. Here, fluorescence is measured as relative fluorescence units (RFUs) and reflects overall epitope abundance (Figure 1A). The RFU readout is normalized by three non-consecutive steps including hybridization control normalization, intraplate median signal normalization, and median signal normalization to a reference and then delivered in a tab-delimited text file (.adat extension) for analysis using SomaLogic’s web-based tool known as DataDelve Statistics alongside the DataDelve Stats v1.0 User Guide. A T-test was used to compare the means of two groups (such as 2-year deceased and survivor cohorts). Considering the exploratory nature of our study and limited OSA patient samples analyzed, differentially expressed proteins (DEPs) between comparative groups were identified as having a Log2FC > 0.585 or < -0.585 and P-value < 0.05. Median normal patient plasma RFU measurements (provided by SomaLogic) for both APO- and HOLO-transcobalamin-II (TCN2) are graphed on the associated box plot and are representative of a healthy, control patient population derived from previous publication (58). Considering these median values were determined using normal patient ethylenediaminetetraacetic acid (EDTA)-plasma samples that had been centrifuged and stored 2-10 hours post-blood collection (as opposed to the Streck-collected specimens analyzed here), they should only serve as a general guideline to the major findings from our analyses. A Volcano plot depicting those DEPs between comparative groups was generated using GraphPad Prism 10 (GraphPad Software, Boston, MA, USA).

2.3 Reactome pathway analysis

A pathway analysis of those DEPs between 2-year deceased and survivor cohorts (all OSA and advanced OSA cohorts) was conducted using the Reactome Pathway Analysis Tool (https://reactome.org/PathwayBrowser/, version 87) according to the Reactome User Guide (Docs – Analysis Tools) and previous publications (59–61). Briefly, all proteins with Log2FC > 0.585 or < -0.585 and P-value < 0.05 were inputted into the identifier list alongside their associated Log2FC. For consistency, proteins identified by the same SOMAmer, each with a unique sequencing ID, however, sometimes detecting multiple proteins [example seq.3714-49 quantifying both Creatine kinase M-type and Creatine kinase B-type heterodimer (P12277|P06732 or CKB|CKM)], were inputted in the identifier list multiple times with the same corresponding Log2FC. Following data input and use of default analysis settings (projection to human without interactors), the pathway analysis results (.csv) and analysis report (pdf) were exported. A genome-wide overview plot is used to highlight the Reactome pathways over- (blue) or underrepresented (yellow) in the input dataset, with light grey signifying non-significant pathways. Significant Reactome pathways for the given data input were identified as having an FDR < 0.05.

2.4 Survival Genie overall survival and CIBERSORTx TIL correlation analysis

The Survival Genie web-interface (https://bhasinlab.bmi.emory.edu/SurvivalGenie/) was used for both overall survival and CIBERSORTx tumor-infiltrating lymphocyte (TIL) analyses according to the user guide and a previous publication (62). Briefly, single gene analyses [transcobalamin-II (TCN2), alpha 2-HS glycoprotein (AHSG)] were conducted using the National Cancer Institute (NCI) Therapeutically Applicable Research to Generate Effective Treatments-Osteosarcoma (TARGET-OS) and The Cancer Genome Atlas (TCGA)-Sarcoma (SARC) databases. Here, single gene normalized expression [fragments per kilobase of transcript per million mapped reads (FPKM)] from primary tumors is associated with overall survival through Cutp estimated martingale residuals (63) and patient stratification into low and high expressing groups using the survMisc package (64). Resulting Kaplan-Meier (KM) survival curves with high (red) and low (blue) group stratification are compared using the log-rank test, with log-rank P-value < 0.05 considered statistically significant (64). The provided Forest plot details the association between the high and low groups [stratified by cut point (CP)] based on the Cox Proportional Hazards regression model (survMisc package). Hazard ratio (HR) with 95% confidence interval as well as the associated Wald-test (HR) and log-rank (LR) P-values are reported (64). The relative fraction of TILs was estimated using the tumor-infiltrating immune cell type matrix LM22 gene signature from bulk tumor FPKM gene expression data and the CIBERSORTx deconvolution method (65). A Pearson correlation matrix of deconvoluted immune cell RNA-seq gene expression data and our genes of interest are reported. Shape (square or circle) denotes significance while color denotes positive (red) or negative (blue) correlation with the genes of interest (TCN2, AHSG).

2.5 TIGER scRNA-seq analysis

The Tumor Immunotherapy Gene Expression Resource (TIGER) portal (http://tiger.canceromics.org/#/) was used to investigate the target genes of interest [TCN2, transcobalamin receptor (CD320)] according to the user guide and a previous publication (66). Briefly, after inputting the genes of interest, the “Single-cell Immunity” tab was used to investigate gene expression across various scRNA-seq datasets of the TIGER database. Here, the Cell Type Marker (Log2FC) diagram represents differential expression of various clusters and subclusters in each scRNA-seq dataset of the TIGER database. These cells, organized by cancer type, dataset ID, main lineage, cell type, and Log2FC, were then sorted for highest Log2FC expression of the target genes. The top 20 clusters with increased gene expression were reported. After identifying the various cell types (myeloid, plasma, and B cells) and cancers [non-small cell lung cancer (NSCLC) – dataset NSCLC (67), NSCLC1 (68), NSCLC5 (69), NSCLC6 (70)], skin cutaneous melanoma (SKCM) – dataset SKCM1 (71)] with increased expression of the target gene(s), a subsequent cell-cell communication analysis was performed. Briefly, cell-cell communication in the NSCLC (67) and SKCM1 (71) datasets was investigated due to expression of both genes of interest (TCN2, CD320) across different cell types. Following navigation to the “Single-cell Immunity” tab and selection of the NSCLC dataset, a cell-cell communication analysis was conducted for the Mye_C1_CCL18 cluster (with greatest expression of TCN2) and the B_C10_MZB1 cluster (with greatest expression of CD320). The resulting heatmap displays the expression of all receptor-ligand interactions between any cell type of interest and its location [example – tumor, peritumoral (if applicable)]. The top three cell-cell interactions with the greatest total effect score are reported. Upon toggling the effect score between cell types of interest (example Mye_C1_CCL18 | B_C10_MZB1), the gene expression of the corresponding receptor-ligand interactions between those cells is displayed and sorted from greatest to least expression. For the top three cell-cell interactions, the top 20 receptor-ligand gene interactions (present intratumorally) and their expression were reported. A similar analysis was conducted for both the Mye_C4_C1QA (with greatest expression of TCN2) and the B_C2_IGHG1 (with greatest expression of CD320) of the SKCM1 dataset without responder/non-responder stratification (71).

2.6 scRNA-seq analysis of OSA tumor specimens

Raw OSA scRNA-seq data of GSE162454 (naive primary tumors) and GSE152048 (chemotherapy treated primary, recurrent, and metastatic tumors) were downloaded from the Gene Expression Omnibus (GEO) under their accession number. The FASTQ files were mapped to the GRCh38/hg38 reference genome using Cell Ranger (10X Genomics, Pleasanton, CA, USA, version 6.1.2). Individual cloupe files for each patient sample were then analyzed using the Loupe Browser (10X Genomics, version 7.0.1) according to the 10X Genomics Support guidelines. Briefly, each cloupe file was reanalyzed to filter barcodes through thresholds by Unique Molecular Identifiers (UMIs), threshold by Features, and mitochondrial UMIs. Considering the sheer heterogeneity of the processed samples, this filtering was performed on a sample-to-sample basis. On average, for the GSE162454 dataset, each sample was filtered by UMIs with a min = 666.67 ± 105.41 and max = 74166.67 ± 16654.16, threshold by Features with a min = 500.00 ± 0.00 and max = 9500.00 ± 500.00, and mitochondrial UMIs (%) with a min = 0.00 ± 0.00% and max = 31.33 ± 0.88%. This filtering resulted in approximately 10.3 ± 1.1% of barcodes removed from each sample (range 7.4-14.7%), in accordance with guidelines. On average, for the GSE152048 dataset, each sample was filtered by UMIs with a min = 502.18 ± 1.00 and max = 35000.00 ± 5680.91, threshold by Features with a min = 500.00 ± 0.00 and max = 5681.82 ± 527.74, and mitochondrial UMIs (%) with a min = 0.00 ± 0.00% and max = 20.91 ± 1.63%. This filtering resulted in approximately 13.0 ± 1.5% of barcodes removed from each sample (range 4.6-20.7%), again, in accordance with guidelines. Following barcode filtering, clusters were identified in each sample of both the GSE162454 and GSE152048 datasets according to canonical marker expression of the major cell types identified by Liu et al. in a previous publication including Myeloid cells 1/2 (LYZ, CD68), NK/T cells (CD2, CD3D, CD3E, CD3G, GNLY, NKG7, KLRD1, KLRB1), CAFS (COL1A1, ACTA2, VIM), Plasmocytes (IGHG1, MZB1), Osteoblastic OSA cells (ALPL, RUNX2, IBSP), Osteoclasts (ACP5, CTSK), Endothelial cells (EGFL7, PLVAP), and B cells (MS4A1, CD79A) (72). The expression of TCN2 and CD320 was then assessed across all major cellular clusters via visual inspection of t-distributed Stochastic Neighbor Embedding (t-SNE) plots before further downstream analysis. The expression of TCN2 and CD320 on Myeloid cells and Plasmocytes/B cells clusters, respectively, was then investigated. For quantification of TCN2+ Myeloid cells and CD320+ Plasmocytes and B cells, the “Feature Min” function of the Loupe Browser (displaying the minimum expression value of the features in each cluster’s associated feature set when all features are expressed) was utilized. Here, for consistency across patient samples, only those cells expressing all features within a designated feature set (example: TCN2+ Myeloid cells with feature set LYZ, CD68, and TCN2) were quantified. Representative t-SNE plots with log normalized expression of the gene or feature gene set of interest are presented. In addition to total number of cells, the percent (%) of total cells and percent (%) of TCN2+ Myeloid cells (relative to all Myeloid cells) were calculated by dividing the cell population of interest by total number of filtered cells (barcodes) or by total number of Myeloid cells, respectively. Bar graphs quantifying abundance of various cell types across OSA specimens analyzed were generated using GraphPad Prism 10.

2.7 SomaLogic proteomic correlation analyses

Various correlations between plasma APO-TCN2 and previously described soluble markers of plasma cell maintenance (73) and B cell activation (74) [Interleukin (IL)-6, IL-6 receptor(R)a, IL-6Rb, CXCL12, TNF, IL2RA, TNF Superfamily Member 13b (TNFSF13B) or B-cell activating factor (BAFF), TNF Receptor Superfamily Member 13B (TNFRSF13B) or Transmembrane Activator and CAML Interactor (TACI), TNF Receptor Superfamily Member 17 (TNFRSF17) or B-cell maturation antigen (BCMA)], proteins critical for TLS formation [lymphotoxin alpha 1 beta 2 (LTa1b2), Fc Receptor Like 5 (FCRL5), Selectin L (SELL), TNF Superfamily Member 14(TNFSF14)] (75), circulating immunoglobulin (Ig) levels (IgA, IgD, IgE, IgM, J chain, and IgG), as well as proteins of the 12-chemokine (12-CK) tertiary lymphoid structure (TLS) signature (75) [exception – CCL4 (not measured)] were conducted. Briefly, using GraphPad Prism 10, scatter plots representing all naive OSA patient samples (n = 14) with APO-TCN2 (RFU) on the x-axis and values of various other circulatory proteins (RFU) on the y-axis were constructed. A Spearman correlation analysis was conducted using GraphPad Prism 10 with the correlation coefficient (R) and associated P-value (P) presented on each scatter plot.

2.8 KM Plotter Immunotherapy survival and correlation analysis

Using the Kaplan-Meier (KM) Plotter Immunotherapy online database (https://kmplot.com/analysis/), both TCN2 and CD320 were evaluated for overall and progression free survival in solid tumors (esophageal, gastric, head and neck, melanoma, lung, and urothelial cancer) treated with various immunotherapies (76). Specimens from all solid tumors, collected pre-treatment, irrespective of immunotherapy target [anti-programmed death cell protein 1 (PD-1), anti-programmed death ligand 1 (PD-L1), and anti-cytotoxic t-lymphocyte associated protein 4 (CTLA-4)], were included in both overall and progression free survival analysis using default KM Plotter settings (auto selection of best cutoff based on the calculation of all upper and lower quartiles with selection of the best performing threshold). For each gene of interest, a resulting KM plot was constructed with a reported HR, log-rank P-value, and FDR, with the y-axis representing probability of survival and the x-axis representing time (months). The total number of patients at risk for each time point is also reported. Similarly, an overall survival analysis was conducted for all chemokines of the 12-CK signature (75) using the same analysis parameters. Additionally, Spearman correlations between both TCN2 and CD320 with all chemokines of the 12-CK TLS signature including C-C Motif Chemokine Ligand (CCL)2, CCL3, CCL4, CCL5, CCL8, CCL18, CCL19, CCL21, CXCL9, CXCL10, CXCL11, CXCL13 were also conducted. Due to an inherent limitation of the KM Plotter Immunotherapy database, a maximum of three genes of the 12-CK signature (in addition to either TCN2 or CD320) were included within the same correlation analysis. Spearman correlation coefficients and P-values (* = P < 0.05, ** = P < 0.01, *** = P < 0.001, **** = P < 0.0001) were reported for all four-gene correlation matrices.

2.9 TCGA 12-CK signature correlation analysis

Using the Gene Expression Profiling Interactive Analysis 2 (GEPIA2) web server (http://gepia2.cancer-pku.cn/#index) (77), a multi-gene Spearman correlation analysis between either TCN2 or CD320 and the 12-CK TLS signature (75) was also performed. Two separate analyses using only the TCGA-SARC gene expression dataset or all tumor types of the TCGA were conducted. For each correlation analysis, the log2 of the transcript count per million [log2(TPM)] of the gene of interest (TCN2 or CD320) was plotted on the x-axis while the log2(TPM) of the 12-CK gene signature is plotted on the y-axis. Importantly, while a log scale is used for graphical representation, the Spearman correlation is calculated on a non-log scale. The resulting Spearman correlation coefficients (R) and P-values are reported.

3 Results

3.1 Study design and patient stratification

Using the SomaLogic SomaScan® 7k assay, the plasma proteomic profile of primary, recurrent, and metastatic OSA patients was investigated. In total, 25 OSA patient plasma samples at various stages of disease and treatment status were processed in this analysis. Considering our group’s inherent interest in tumor immunology, two radiation-induced OSA patient samples, five active chemotherapy patient samples, and four post-chemotherapy patient samples were excluded from downstream analysis to mitigate confounding influences to the OSA proteome. The remaining samples, representing 14 naive OSA patients collected at diagnosis (before standard of care chemotherapeutic and/or surgery regimens), were then stratified into 2-year deceased and survivor all comers cohorts (Figure 1C). Patient information including sex, vital status, treatment, disease, and 2-year survival status, in addition to Streck-plasma spin down time (hours) from blood collection to plasma isolation and frozen storage, are reported (Table 1).

3.2 Proteomic assessment of naive OSA patients reveals DEPs associated with 2-year survival status

All experimental patient plasma samples processed at SomaLogic’s Boulder, CO facility passed standardized quality assessment and control (Supplementary File 1 – Supplementary Figure 1). The plasma proteomic profiles of the 2-year deceased (n = 3) and survivor (n = 11) patients from an all comers OSA cohort (n = 14) were then compared (Supplementary File 2). Importantly, a total of 24 human DEPs were identified from this analysis, including 10 significantly upregulated and 14 significantly downregulated in the 2-year survivor cohort (Figure 2A and Table 2). These 24 DEPs were then inputted into the Reactome Pathway Analysis tool. The Reactome genome-wide overview map highlighted overrepresentation of these 24 DEPs in various pathways involved in the Immune System, Signal Transduction, and Metabolism (Supplementary File 1 – Supplementary Figure 2). The most significant pathways (FDR < 0.05) and their enriched entities included those associated with FOXO-mediated transcription [catalase (CAT), resistin (RETN)], MAPK activation (IL-6R), Interleukin-6 signaling (IL-6R), and creatine metabolism [creatine kinase (CK)M, CKB] (Figure 2B).

Figure 2

Of the 24 individual DEPs, APO-TCN2 was the most significantly (P = 0.0017) upregulated protein associated with 2-year survival, reporting a Log2FC = 6.8. For n = 5 of total 11 patients in the 2-year survivor cohort, APO-TCN2 levels vastly exceeded median normal patient plasma RFU measurements (provided by SomaLogic) (58), with an additional two patients reporting well above minimal detectable levels. This finding contrasts with all three deceased patients, which measured near minimal detectable levels (Figure 2C). Interestingly, while the vitamin-B12 unbound form of Transcobalmin-2 (APO-TCN2) differed vastly between 2-year deceased and survivor cohorts, no difference in the vitamin-B12 bound form (HOLO-TCN2) was measured (Figure 2D). Importantly, the lack of measurable differences in HOLO-TCN2 levels suggests that these OSA patients, likely, do not have a genetic TCN2 alteration or abnormality, as previously reported in various case reports (78–81). Of note, a significant (P = 0.0012) decrease in Alpha-2-HS-glycoprotein (AHSG) was also associated with 2-year survival status, reporting a log2FC = -6.1 (Table 2). Furthermore, a secondary analysis comparing the proteome of 2-year deceased (n = 2) and survivors (n = 3) of the Advanced Disease sub-cohort (n = 5) further supported these findings. A significant (P = 0.000034) increase in plasma APO-TCN2 (Figure 2E), the most differentially upregulated protein in 2-year survivors, and no difference in HOLO-TCN2 levels (Figure 2F), was again associated with 2-year survival status. Similarly, AHSG levels were significantly (P = 0.026) reduced in the 2-year survival cohort (Supplementary File 3). Overall, a comparative proteomic analysis of OSA patient peripheral plasma identified 24 DEPs associated with 2-year survival status. Of those proteins, both APO-TCN2 and AHSG measured vastly increased or decreased levels in the survivor cohort, respectively. To further elucidate both proteins’ relevance in OSA and other solid tumors, publicly available bulk and scRNA-seq tumor datasets were then examined.

3.3 Survival Genie analysis reveals association of elevated TCN2 with overall survival in sarcoma

The Survival Genie web-based platform was then used to associate expression of TCN2 and AHSG with overall survival in publicly available OSA (TARGET-OS) and sarcoma (TCGA-SARC) bulk tumor RNA-sequencing datasets (62). To begin, evaluation of TCN2 expression in the TARGET-OS database identified 47 patients with low and 39 patients with high TCN2 expression in OSA primary tumors. Using CIBERSORTx (65), a significant (P ≤ 0.05) positive correlation between TCN2 expression and T.cells.CD8, T.cells.follicular.helper, Macrophages.M1, T.cells.gamma.delta, T.cells.regulatory.Tregs, and Macrophages.M2 gene signatures was revealed. Additionally, a significant (P ≤ 0.05) negative correlation between TCN2 expression and Macrophages.M0, T.cells.CD4.naive, and NK.cells.resting gene signatures was also apparent. Importantly, Kaplan-Meier survival curve analysis associated an increase in intratumoral TCN2 expression with better overall survival (log-rank P= 0.00033), with a reported HR = 0.22 (0.089 – 0.54) (Figure 3A). Considering a shared sarcomatous background, our group expanded the analysis to assess TCN2 gene expression in the larger TCGA-SARC dataset to provide rigor and support to the experimental findings uncovered in TARGET-OS. Analysis of the TCGA-SARC dataset identified 177 patients with low and 82 patients with high expression of TCN2 in sarcoma primary tumors. Using CIBERSORTx (65), a significant (P ≤ 0.05) positive correlation between TCN2 expression and Macrophages.M2, T.cells.CD8, T.cells.CD4.memory.activated, and Macrophages.M1 was identified. Additionally, a significant (P ≤ 0.05) negative correlation between TCN2 expression and Macrophages.M0, T.cells.CD4.memory.resting, Eosinophils, and NK.cells.activated gene signatures was also apparent. Importantly, Kaplan-Meier curve survival analysis again associated an increase in TCN2 expression with better overall survival (log-rank P = 0.0047), with a reported HR = 0.5 (0.31 – 0.82) (Figure 3B). These results suggest that increased intratumoral TCN2 is associated with an activated immune signature and better overall survival. While a similar analysis revealed that low AHSG expression was associated with better overall survival in the TARGET-OS dataset (corroborating our proteomic findings), this trend was not apparent in the TCGA-SARC analysis, and neither finding was statistically significant. Therefore, AHSG was excluded from further downstream analysis (Supplementary File 1 – Supplementary Figure 3). Overall, Kaplan-Meier survival curve analysis associated elevated TCN2 expression with significantly improved outcomes in both the TARGET-OS and TCGA-SARC datasets. Importantly, the association of increased intratumoral TCN2 expression with better overall survival correlated with our previous plasma proteomics findings (Figures 2B, D). These results ultimately support both peripheral blood plasma APO-TCN2 and intratumoral TCN2 gene expression as possible biomarkers of overall survival in this disease.

Figure 3

3.4 TIGER reveals expression patterns of TCN2 in solid tumors

The TIGER portal was then used to investigate the expression of TCN2 across identified cellular clusters of various solid tumor scRNA-seq datasets (66). While previous analyses of bulk transcriptomic data through CIBERSORTx deconvolution had suggested the likely cellular source for TCN2 gene expression in sarcomatous tumors, our group sought to determine which cells differentially expressed TCN2 (for a variety of tumor types) using a method that provides single cell resolution to the complex and heterogenous tumor microenvironment (TME). Numerous cellular clusters across many solid tumors revealed differential expression of TCN2 including non-small cell lung cancer (NSCLC), intrahepatic cholangiocarcinoma (ICC), breast cancer (BC), nasopharyngeal carcinoma (NPC), and skin cutaneous melanoma (SKCM), among others (Figure 4A). The top 20 cellular clusters with the greatest increase in TCN2 expression were reported, highlighting apparent increases for B cell, Endothelial, and Myeloid cellular clusters, particularly in both NSCLC and SKCM datasets (Table 3). To visualize these expression changes, subsequent analysis of Uniform Manifold Approximation and Projections (UMAPs) and gene expression boxplots for the NSCLC dataset (67) confirmed isolated increases in TCN2 expression on both Myeloid and Endothelial cellular clusters (Figures 4B, C). Subcluster analysis of the Myeloid cells revealed near ubiquitous expression of TCN2 across myeloid populations, with the greatest increases apparent for the Mye_C1_CCL18 and Mye_C6_CCL18 subclusters (Figures 4D, E). Further visualization of the NSCLC6 (70) dataset supported increased TCN2 expression on Endothelial, Plasma, AT2 Epithelial, and Myeloid cell populations (Figures 4F, G). Additional subcluster analysis of B cells revealed increased expression across B_C3_JCHAIN, B_C1_MZB1, and B_C7_MZB1 cellular clusters (Figures 4H, I), in accordance with the findings highlighted in Table 3. Similar results were also apparent for the SKCM1 dataset (71), which revealed increased TCN2 expression on all Myeloid cells, various Myeloid cell subclusters including Mye_C4_C1QA, as well as the B_C2_IGHG1 B cell subcluster (Supplementary File 1 – Supplementary Figure 4). Overall, these data corroborated our Survival Genie CIBERSORTx data and previous literature (82) which suggested that TCN2 is mainly expressed (in immune cells) on Myeloid populations.

Figure 4

3.5 TIGER reveals expression patterns of CD320 in solid tumors

The TIGER portal was also used to investigate the expression of CD320 (83), the TCN2 receptor, across identified cellular clusters of solid tumor scRNA-seq datasets (66). Numerous cellular clusters across many solid tumors revealed differential expression of CD320 including stomach adenocarcinoma (STAD), colorectal cancer (CRC), Merkel cell carcinoma (MCC), NSCLC, ICC, and SKCM, among others (Figure 5A). The top 20 cellular clusters with the greatest increase in CD320 expression were reported, highlighting apparent increases for Endothelial, Malignant, B cell, CD4, and Myeloid cellular clusters across a variety of solid tumors (Table 4). To visualize these expression changes on immune cells of interest (B and Myeloid cells), subsequent analysis of UMAPs and gene expression boxplots for the NSCLC5 dataset confirmed increases in CD320 expression on Endothelial, Malignant, and Fibroblast cellular clusters (Figures 5B, C). Additional subcluster analysis of B cells supported an increase in CD320 expression on both B_C11_MZB1 and B_C4_IGHG1 (Figure 5D, E). Further visualization of UMAPs and gene expression boxplots for B cell subclusters of both the NSCLC (Figures 5F, G) and NSCLC1 (Figures 5H, I) datasets supported increased CD320 gene expression on various Marginal Zone B and B1 Cell Specific Protein (MZB1)+ and IGJ+ B cell subclusters, in accordance with findings from Table 4. Similar results were also apparent for the SKCM1 dataset, which revealed increased CD320 expression on all Plasma cells, Myeloid cell subclusters including Mye_C1_GZMB, as well as the B_C2_IGHG1 B cell subcluster (Supplementary File 1 – Supplementary Figure 5). Overall, these data corroborated previous publications which suggested that while CD320 is expressed ubiquitously across many cell types (84), its expression is increased on proliferating cells (85), such as malignant cells or B lymphocytes (86), in a variety of solid tumors (87).

Figure 5

3.6 TIGER cell-cell communication analysis reveals interaction between TCN2+ myeloid cells and CD320+ B cells in solid tumors

To investigate a possible interaction between TCN2+ Myeloid and CD320+ B cells in these solid tumors, the TIGER portal was then used to perform a cell-cell communication analysis (66). Further investigation of both the NSCLC (67) and SKCM1 (71) datasets was warranted considering the increased expression of both TCN2 (ligand) and CD320 (receptor) on various immune cell populations of these tumors. To begin, a cell-cell communication analysis for the Mye_C1_CCL18 subcluster, with the greatest increase in TCN2 expression of all myeloid cells of the NSCLC dataset (67), was performed. Analysis revealed that Mye_C1_CCL18 had greatest interaction with various other myeloid subclusters including Mye_C2_FBP1 (effect score = 97.684), Mye_C7_PPA1 (effect score = 97.64), and Mye_C6_CCL18 (effect score = 96.877). The top 20 receptor-ligand interactions driving these effect scores were reported (Table 5). Of note, this analysis consistently highlighted likely receptor-ligand interactions between CD74_MIF, CD74_COPA, CD74_APP, and HLA-DPB1_TNFSF13B as well as additional communication through HLA-DRB1_TNFSF9, CCR1_CCL18, and TNFRSF1A_GRN (Table 5). Similar cell-cell communication patterns were evident for the Mye_C4_C1QA subcluster of the SKCM1 dataset (71), which displayed the greatest increase in TCN2 expression of all SKCM1 myeloid subclusters. Here, Mye_C4_C1QA also reported the greatest cell-cell communication effects scores with other myeloid subclusters (Supplementary File 1 – Supplementary Table 1).

Additionally, a cell-cell communication analysis for the B_C10_MZB1 subcluster, with the greatest increase in CD320 expression of all B cells of the NSCLC dataset (67), was performed. Analysis revealed that B_C10_MZB1 had the greatest interaction with various myeloid subclusters including Mye_C1_CCL18 (effect score = 48.993), Mye_C2_FBP1 (effect score = 47.944), and Mye_C7_PPA1 (effect score = 46.079). The top 20 receptor-ligand interactions driving these elevated effect scores were reported. Of note, this analysis consistently highlighted likely receptor-ligand interactions between CD74_MIF, CD74_COPA, CD74_APP, and HLA-C_FAM3C as well as additional communication through C5AR1_RPS19, HLA-DPB1_TNFSF13B, CD94:NKG2A_HLA-E, SPP1_CD44, and various chemokine axes (Table 6). Similar cell-cell communication patterns were also evident for the B_C2_IGHG1 subcluster of the SKCM1 dataset (71), which displayed the greatest increase in CD320 expression of all SKCM1 B cell subclusters. Importantly, the B_C2_IGHG1 subcluster displayed the greatest cell-cell communication effect score with various myeloid subclusters including Mye_C4_C1QA (Supplementary File 1 – Supplementary Table 2). Overall, cell-cell communication analysis revealed that those B cell subclusters with greatest expression of CD320 [B_C10_MZB1 (NSCLC), B_C2_IGHG1 (SKCM1)] reported the greatest cell-cell communication effect scores with those Myeloid cell subclusters with greatest expression of TCN2 [Mye_C1_CCL18 (NSCLC), Mye_C4_C1QA (SKCM1)]. These results suggest a possible Myeloid (TCN2) to B/Plasma cell (CD320) communication network in the tumor microenvironment of immunogenic solid tumors. Considering TCN2’s previously described biological functions (88–90), it was hypothesized that this interaction in solid tumors would allow for enhanced B lymphocyte infiltration, proliferation, and germinal center/TLS formation (91).

3.7 scRNA-seq of OSA tumors supports presence of TCN2+ myeloid and CD320+ plasmacytes/B cells

To investigate whether TCN2+ Myeloid and CD320+ B Lymphocyte clusters existed in OSA tumors, and whether an increase in TCN2+ Myeloid cells supports B lymphocyte infiltration and proliferation, scRNA-seq data from naive primary (GSE162454) and chemotherapy treated primary, recurrent, and metastatic (GSE152048) OSA tumors were analyzed. First, scRNA-seq analysis of six naive OSA primary tumors (GSE162454) was conducted. Each major cellular cluster was identified according to expression of canonical markers previously detailed by Liu et al. (72) (Figure 6A). Visualization of t-SNE plots revealed ubiquitous expression of CD320 in OSA primary tumors, with greatest expression on Osteoblastic OSA cells, Endothelial cells, NK/T cells, and Plasmocytes (Figure 6B). For TCN2, t-SNE plots conveyed concentrated gene expression on Myeloid cells, with additional positivity on both Endothelial cells and Plasmocytes (Figure 6C). The number of Plasmocytes (Figure 6D), B cells (Figure 6E), and Myeloid cells (Figure 6F) expressing CD320 (Plasmocytes/B cells, Figures 6G, H) or TCN2 (Myeloid cells, Figure 6I) was then quantified. Unfortunately, only the OS-6 patient sample contained elevated levels of Plasmocyte or B cell populations (Supplementary File 1 – Supplementary Figure 6). This finding was unsurprising considering Li et al. noted that only one patient sample had identifiable TLSs during sample processing (72). Nevertheless, total CD320+ Plasmocytes, CD320+ B cells, and TCN2+ Myeloid cells, as well as the percent (%) of total cells, were quantified for each patient sample (Figures 6J, K). Importantly, while only OS-6 contained a substantial number of CD320+ Plasmocytes and B cells, this primary tumor also measured the greatest percentage (%) of TCN2+ Myeloid cells (relative to all Myeloid cells) at 38.68% (Figure 6L and Supplementary File 4).

Figure 6

A subsequent scRNA-seq analysis of eleven chemotherapy treated primary, recurrent, and metastatic OSA tumor specimens (GSE152048) was conducted using the same canonical markers and cellular clusters. Again, while CD320 was ubiquitously expressed across numerous cellular populations, TCN2 expression was concentrated on Myeloid cells. Unfortunately, subsequent quantification of Plasmocyte and B cell clusters revealed minimal B lymphocyte infiltration in the chemotherapy treated OSA tumors examined. Interestingly, these OSA samples also had substantially lower overall TCN2+ Myeloid cell populations in comparison to the previously analyzed naive primary tumors (GSE162454). Additionally, the percentage (%) of TCN2+ Myeloid cells (relative to all Myeloid cells) was no greater than 21.26% across all tumors examined (Supplementary File 1 – Supplementary Figure 7, Supplementary File 4). Overall, these analyses supported that TCN2+ Myeloid and CD320+ Plasmocyte and B cell clusters were ultimately present in OSA tumors. Considering the previous publication’s note regarding isolation of these B lymphocytes from confirmed TLSs in one naive primary tumor (72), we also associated the highest % of TCN2+ Myeloid cells (relative to all Myeloid cells) with intratumoral TLS formation. Ultimately, investigation of additional OSA tumor specimens (positive for B lymphocyte and/or Plasmocyte populations) and further cell-cell communication analysis is required to support the hypothesis that TCN2+ Myeloid cells drive infiltration and proliferation of B lymphocyte populations through increased TCN2 (ligand) and CD320 (receptor) interactions.

3.8 Proteomic correlation analysis of APO-TCN2 reveals detectable signatures associated with plasma cell maintenance, B cell activation, TLS formation, and circulating immunoglobulins

B lymphocyte infiltration, proliferation, and formation of TLSs is well regarded as a clinical biomarker of long term survival and response to immunotherapy in many solid tumors (92–115), including soft-tissue sarcoma (116, 117). Previous literature has suggested the necessity of the CD320 receptor for germinal center B cell growth and proliferation (83). Considering the apparent association between myeloid-derived TCN2 and CD320+ B lymphocytes in various solid tumors, a subsequent correlation analysis between plasma APO-TCN2 and previously described soluble markers of plasma cell maintenance (73), B cell activation (74), proteins critical for TLS formation (75), chemokines of the 12-CK signature (75) [exception – CCL4 (not measured)], and circulating immunoglobulin levels was then conducted. To begin, for those markers of plasma cell maintenance (73) (Figures 7A–K), positive correlations between plasma APO-TCN2 and IL-6 (Figures 7A, B), IL-6Ra (Figures 7C–E), IL-6Rb (Figure 7F), and TNF (Figures 7J, K) were observed, with statistical significance apparent for IL-6Ra. Additionally, a negative correlation between APO-TCN2 and CXCL12 (Figures 7G–I) was evident. Furthermore, for the soluble markers of B cell activation (74) (Figures 7L–O), IL-2RA (Figure 7L), TNFSF13B (BAFF) (Figure 7M), and TNFRSF13B (TACI) (Figure 7N) all displayed positive yet nonsignificant correlations with plasma APO-TCN2, while a negative, nonsignificant correlation was reported for TNFRSF17 (BCMA) (Figure 7O). Overall, these data suggest that increases in plasma APO-TCN2 were associated with increases in previously described soluble markers of B lymphocyte activity, albeit nonsignificant for most associations.

Figure 7

Furthermore, analysis of various proteins critical for TLS formation (75) (Figures 7P–T) revealed similar trends. Positive yet nonsignificant correlations between APO-TCN2 and LTa1b2 (Figure 7P), FCRL5 (Figure 7Q), SELL (Figure 7R), and TNFSF14 (Figures 7S, T) were observed, suggesting a possible association of APO-TCN2 levels with those proteins necessary for TLS formation. Furthermore, negative, nonsignificant correlations between APO-TCN2 and most chemokines of the 12-CK signature (75) were evident (Figure 8). Overall, these data [much like the findings for CXCL12 (Figures 7G–I)] may suggest that increased plasma APO-TCN2 is associated with the development of exacerbated chemokine axes (lower blood plasma and higher tissue levels) for potent lymphocytic infiltration into patient solid tumor and/or other tissues. Additional tissue/tumor-level analyses are necessary to further support this finding. Of note, additional correlation analyses of circulating immunoglobulin levels suggested negative correlations between plasma APO-TCN2 and IgA, IgD, IgE, IgM, and IgG, with a positive correlation to J chain (Supplementary File 1 – Supplementary Figure 8). Limited research has correlated peripheral blood immunoglobulin levels with intratumoral TLSs, therefore, the significance of these findings is unknown. However, these data may indicate the presence of improved intratumoral humoral immune responses (increased J chain) in those patients with elevated APO-TCN2. Overall, these data suggested that elevated APO-TCN2 levels may be associated with a peripheral blood plasma signature necessary for robust B lymphocyte proliferation and infiltration in OSA tumors.

Figure 8

3.9 KM Plotter Immunotherapy suggests association of TCN2/CD320 with response to immunotherapy and formation of intratumoral TLSs

Our group then hypothesized that if increases in plasma APO-TCN2 (and as a result, intratumor TCN2) is associated with lymphocytic infiltration and possible formation of TLSs, then increased TCN2/CD320 expression in solid tumors, treated with immunotherapy, should correlate with better overall survival. To investigate this hypothesis, both TCN2 and CD320 were evaluated for overall and progression free survival in solid tumors treated with various immunotherapies (anti-PD-1, anti-PD-L1, anti-CTLA-4) using the KM Plotter Immunotherapy online database (76). Increased intratumoral expression of TCN2 was indeed associated with better overall (log-rank P = 8.6e-05, FDR = 0.03) (Figure 9A) and progression free (log-rank P = 1.6e-11, FDR = 0.01) (Figure 9B) survival. Similar findings were also evident for the TCN2 receptor, as increased CD320 expression was associated with better overall (log-rank P = 9.4e-08, FDR = 0.01) (Figure 9C) and progression free (log-rank P = 1.3e-08, FDR = 0.01) (Figure 9D) survival in the same dataset. These data suggest that increased pre-treatment TCN2 and CD320 expression is associated with better outcomes in a variety of solid tumors treated with immunotherapy.

Figure 9

To further examine whether intratumoral TCN2 and CD320 expression is associated with TLS formation, a correlation analysis to the 12-CK TLS signature (75) was performed using the KM Plotter Immunotherapy database (76). Unsurprisingly, increased expression of all 12-CK signature chemokines was significantly (FDR < 0.05) associated with better overall survival in the solid tumors examined (exception – CCL18, HR = 0.71, log-rank P = 0.00055, FDR = 0.20) (Supplementary File 1 – Supplementary Figure 9). A Spearman correlation analysis supported significant (P < 0.05) positive correlations between TCN2 and CD320 expression with each chemokine of the 12-CK TLS signature (exception – CD320 and CCL4) (Supplementary File 1 – Supplementary Table 3). A subsequent multi-gene Spearman correlation analysis between TCN2 and CD320 with the 12-CK TLS signature (75) using the GEPIA2 web server and solid tumors of the TCGA database (77) was also performed. Importantly, TCN2 expression displayed a significant positive correlation to the 12-CK gene signature for TCGA-SARC (P-value = 5.1e-13, R = 0.43). Similar results were evident for a combinatory analysis of all TCGA tumors (P-value = 3.4e-123, R = 0.24). In contrast, CD320 expression displayed an opposite trend, with a weak negative correlation in both datasets (Supplementary File 1 – Supplementary Figure 10). Overall, these data highlight a positive correlation between TCN2 and the 12-CK TLS signature in a variety of solid tumors. While preliminary, these data may suggest an association between elevated peripheral plasma APO-TCN2 (alongside probable increases in intratumoral TCN2 expression) with formation of TLSs. Further investigation of sarcoma tumor specimens is ultimately necessary to associate increased plasma APO-TCN2 with intratumoral B lymphocyte aggregation and/or TLS formation.

4 Discussion

This proteomic biodiscovery analysis of OSA patient plasma identified apo-transcobalamin-II (APO-TCN2) as a novel circulatory biomarker of survival in this disease. To our knowledge, the only known publication which previously characterized TCN2 in OSA was published by Rothzerg et al. in 2021. Here, TCN2 was identified as one of four upregulated genes associated with better overall survival in OSA tumors of the TARGET-OS dataset (118). To our knowledge, no further characterization of TCN2 in OSA has since been conducted. Ultimately, the work presented here provides the utmost support for its relevance in this disease, as an over 6-fold increase in circulating APO-TCN2 levels was associated with better overall survival in our patient cohort. The dual association of both increased TCN2 expression (in OSA tumors) and increased circulatory APO-TCN2 levels (in OSA peripheral blood plasma) with improved patient outcome provides credence to our findings and this biomarker’s future clinical utility in this disease. While additional studies to correlate intratumoral TCN2 expression with plasma APO-TCN2 levels are necessary, these data suggest that circulating protein measurements (through liquid biopsy) could act as a surrogate for tumoral assessments in both local and advanced OSA. Future studies to assess changes in peripheral APO-TCN2 throughout the course of disease through serial blood sampling could also further associate plasma protein levels with disease progression, therapeutic response, and/or clinical remission.

TCN2’s biological function of supporting cellular metabolic processes by transporting vitamin B12 (cobalamin, co-factor for various enzymes) through the blood stream, binding to the ubiquitously expressed CD320 surface receptor, and internalization of the receptor-ligand complex for B12’s intracellular release is well known (119–121). Interestingly, numerous reports of TCN2 deficiency due to loss-of-function mutations have been described. In addition to the common presentation of failure to thrive and megaloblastic anemia (due to elevation of homocysteine and methylmalonic acid), a number of these cases reported immunological deficiencies and abnormal immunity including measured pancytopenia, neutropenia, and hypogammaglobulinemia (78–81, 122–138). TCN2 deficiency was also previously associated with abnormal granulocyte function and limited antimicrobial response to Staphylococcus aureus infection (139). Of note, no difference in HOLO-TCN2 levels were measured between our 2-year deceased and survivor cohorts, ruling out the possibility of a genetic abnormality in those deceased patients. These reports, however, highlight TCN2’s integral role in proliferation and maintenance of cells of the immune system.

In this regard, increased TCN2 levels were previously reported in patients with a variety of inflammatory and lymphoproliferative disorders, including those diagnosed with multiple myeloma and Waldenstrom macroglobulinemia experiencing overproduction of immunoglobulins (hyperglobulinemia) (140). McLean et al. later determined that TCN2 could not only drive proliferation of human erythroleukemic and murine lymphoma cell lines in vitro, but antibody blockade of its receptor could also effectively inhibit cellular growth (89, 90). Furthermore, TCN2 levels in peripheral blood monocytes of patients with various inflammatory bowel diseases including shigellosis, ulcerative colitis, and Crohn’s were 3-4 times higher than those measured in healthy normal patients, with levels decreasing in the setting of clinical improvement (141). TCN2 was also elevated in the setting of various infectious diseases including malaria and typhus (142–144), supporting its role as a possible acute phase reactant (145).

Of utmost importance, increases in unsaturated TCN2 (APO-TCN2) have been reported in patients with active autoimmune diseases including systemic lupus erythematous (SLE), autoimmune hemolytic anemia, and dermatomyositis, with levels often correlating with disease activity (146–148). Haynes et al. in 2020 also identified a 93-gene signature for diagnosis of SLE through transcriptomic profiling of patient peripheral blood, of which TCN2 was included (149). Additionally, while total serum cobalamin was no different between patients with active rheumatoid arthritis (RA) and clinical remission, APO-TCN2 was significantly elevated in those with active disease (150). These data suggest that APO-TCN2 is elevated in the setting of pathogenic immune responses in various immune-mediated disorders, often correlating with disease activity. Considering these clinical observations, Liu et al. recently investigated TCN2’s pathogenic role in a murine model of lupus. Increased expression of TCN2 in both B and T lymphocytes of SLE patients and lupus-like mice were observed. Importantly, global genetic knockout of TCN2 in this murine model of lupus resulted in complete amelioration of lupus symptoms, as measured by lower dsDNA levels, reduced IgG deposition in the glomerulus, and decreased infiltration of both B and T follicular helper cells in the kidneys and spleen of these mice. The group suggested that TCN2 activity is associated with the abnormal germinal center responses in the setting of SLE and could act as a novel therapeutic target in this disease (91). Considering these findings, it appears that the elevated APO-TCN2 in our 2-year survival cohort, with no difference in HOLO-TCN2 levels, mirrors the heightened systemic immune activity seen in common autoimmune disorders. These in vivo results suggest the importance of TCN2 for germinal center-mediated immune activity, as seen in the kidneys of patients with lupus nephritis (151). While the majority of previous literature indicated that TCN2 was merely correlative in nature (152), these data ultimately suggest a causative role for this circulating protein in the development and progression of lymphoproliferative autoimmune disease.

To suggest a possible mechanism by which APO-TCN2 improves overall survival in solid tumors, various analyses were then conducted using publicly available scRNA-seq datasets. Analysis of the TIGER database supported previous literature which characterized TCN2 expression and active production from, among others, myeloid lineage (82, 153, 154) and endothelial (155, 156) cells. This analysis also confirmed the expression of CD320 ubiquitously across many cell types (84) including B lymphocytes (86). The cell-cell communication analyses highlighted the interaction between B lymphocytes (positive for CD320) with Myeloid cells (positive for TCN2) in both NSCLC and SKCM. These results suggest the presence of a coordinated communication network between these cellular populations which likely supports intratumoral B lymphocyte proliferation for enhanced anti-tumor immune responses. Considering the significant correlation between TCN2 and CD320 with the 12-CK Score (75) and previous literature’s characterization of the CD320 receptor as necessary for germinal center B cell growth (83), these interactions may help coordinate the formation of intratumoral TLSs in these patients. With macrophages having been previously described as potent lymphoid tissue inducer cells necessary for T and B cell recruitment in these structures (157), and knockout of TCN2 by Liu et al. inhibiting infiltration of B/T lymphocytes in kidneys and spleen in a murine model of lupus (91), further investigation of this intratumoral signaling network is therefore warranted.

Importantly, those B lymphocytes of the TIGER database with the greatest increased expression of CD320 included MZB1+ and immunoglobulin producing B cell clusters, commonly found in the marginal zone of germinal centers and tumor TLSs (158–160). Of the receptor-ligand interactions driving the cell-cell communication effect scores between these TCN2+ Myeloid and CD320+ B cell clusters, many are associated with promoting B cell activity. Of note, interactions between CD74 and MIF have been previously associated with both peripheral B cell survival (161) and chemotaxis (162, 163). Additionally, while interactions between CD74 and COPA (COPI coat complex subunit alpha) of malignant plasma and immune cells have been described in Waldenstrom macroglobulinemia (164), these interactions have also been shown to play a role in the maturation of B cells in the setting of antibody-mediated renal transplant rejection (165). Furthermore, scRNA-seq analysis of peripheral immune cells in patients with primary Sjogren’s syndrome highlighted not only enhanced CD74-COPA interactions between monocytes and naive B cells, but also CD74-amyloid beta precursor protein (APP) interactions between monocytes and memory B cells. Considering their enrichment in those with active disease, these receptor-ligand interactions were thought to be important drivers of autoimmunity (166). Further receptor-ligand interactions between HLA class II histocompatibility antigen, DP beta chain (HLA-DPB1) and TNFSF13B (BAFF) in addition to Major Histocompatibility Complex, Class II, DP Alpha 1 (HLA-DPA1) and TNF Superfamily Member 9 (TNFSF9) suggest myeloid-mediated B lymphocyte survival and maturation (167) in germinal centers of these solid tumors, as previously proposed (168). The association of high TCN2 and CD320 expression with better overall survival in patients treated with immunotherapy could likely reflect the increased propensity for B lymphocyte survival, proliferation, and maturation for TLS formation within the responding patient tumors (107).

While only preliminary, both TCN2+ Myeloid and CD320+ Plasmocytes and B cells were identified by scRNA-seq analyses of OSA tumor specimens. Most B lymphocytes characterized in the original analysis were confirmed to have been isolated from a single primary tumor containing TLSs (72). The identification of intratumoral B cells associated with TLSs in one of six naive tumors reflects the rates previously described in the amended PEMBROSARC trial of STS (approx. 20% of patients on initial screening) (169). These preliminary results, however, suggest the necessity for elevated TCN2+ Myeloid cells for B lymphocyte proliferation in the OSA tumor microenvironment. Additional analyses of known TLS-positive and -negative OSA or sarcoma tumor specimens are necessary to further support this interactive mechanism in sarcomas disease. Importantly, various correlation analyses presented here may indicate that elevated APO-TCN2 was associated with a favorable B lymphocyte (73, 74) proteomic profile likely necessary for potent immune cell recruitment and TLS-formation (75) to drive antibody-mediated anti-tumor immune responses in the OSA patients examined. Ultimately, while the mechanism by which TCN2 is associated with better overall survival in OSA was extensively postulated in these analyses, further confirmatory studies are indeed necessary.

Of note, a distinct biological function of APO-TCN2 has, to our knowledge, not yet been described. The specific elevation of APO-TCN2 (as opposed to HOLO-TCN2) was previously associated with various lymphoproliferative autoimmune disorders but not further investigated. Our group, however, can postulate several mechanisms by which the increased circulatory APO-TCN2, measured here in OSA, is associated with better overall survival. First and foremost, our group hypothesizes that increased circulatory APO-TCN2 likely reflects the development of systemic and local proliferative immune responses, driven by activated myeloid lineage cells (82), as a result of immune system detection of cancer. Previous literature suggests that not only do activated lymphocytes have a preference for TCN2 over other carriers of cobalamin (170), but TCN2 is also preferentially absorbed by normal tissues and organs, even in the setting of proliferative tumorigenesis (171). More than likely, increased APO-TCN2 improves the probability of delivering vitamin B12 (through HOLO-TCN2) to lymphoid organs to drive robust immune proliferation and activated immune responses. The elevated APO-TCN2 levels within the survivor patient cohort would then reflect an increased demand for cobalamin in the setting of enhanced lymphocytic proliferation. We provided evidence that this proliferation is occurring within B lymphocytes and associated with intratumoral germinal center formation for enhanced anti-tumor immune responses.

Secondarily, previous literature suggests that APO-TCN2 can likely act as a competitive inhibitor of the CD320 receptor. In this scenario, APO-TCN2 could block the binding and delivery of vitamin B12 (HOLO-TCN2) on CD320+ cells (172), such as malignant tumor cells, thereby limiting their proliferation. While possible, our group believes APO-TCN2’s anti-tumor effect is more than likely through direct lymphocyte activation and proliferation as opposed to activity on tumor cells as a competitive inhibitor. Additionally, APO-TCN2 could possibly drive an undescribed signaling mechanism in these cells, associated with immune activation or pro-inflammatory signal transduction, through binding the CD320 receptor, binding the previously described megalin [thought to be essential in the accumulation of TCN2 in the kidney cortex and other absorptive epithelia (173)], binding an undefined receptor, or forming an additional receptor-ligand complex. Efforts to characterize the activity (if any) of APO-TCN2 are necessary for better understanding its role in both OSA and autoimmune disease. Nevertheless, these results suggest that surviving OSA patients are mounting a robust, systemic immune response as reflected through elevated levels of a biomarker (APO-TCN2) previously associated with active lymphoproliferative autoimmune disease.

While robust in our analyses, this study has inherent limitations. To begin, our plasma proteomic investigation analyzed only 14 naive OSA patients, with n = 3 in the 2-year deceased and n = 11 in the 2-year survival cohort. Considering the limited number of patient samples and the exploratory nature of the study, DEPs between comparative groups were determined based on having a Log2FC > 0.585 or < -0.585 and P-value < 0.05, leading to the possible discovery of false positive DEPs. Future investigations with larger sample sizes will utilize more stringent statistical measures [false discovery rate (FDR), corrected P-values (Bonferroni)] to determine DEPs. Importantly, however, the major finding from this study was supported by gene expression analysis of the largest known OSA RNA-seq dataset (TARGET-OS) in addition to other solid tumors (i.e. TCGA-SARC). Furthermore, several of the analyses presented here, including APO-TCN2’s relationship to other plasma proteins and scRNA-seq cell-cell communication analysis, are mostly correlative and associative in nature. These findings ultimately require further validation in confirmatory studies. Additionally, as to be expected with clinical specimens, the Streck-collected plasma samples analyzed here were processed and stored over varying periods of time. It is well understood that delays in sample processing can affect the measured proteome of patient samples. Fortunately, a previous analysis by Savage et al. using the O-link proteomic platform reported no change in TCN2 levels (in comparison to fresh samples) even after 18 hours of room temperature storage prior to blood specimen processing (log2FC = -0.0087, P-value = 0.860127) (174). These results suggest that the measured APO-TCN2 levels reported here were likely minimally influenced by differential processing times. Ultimately, future studies which plan to expand our OSA patient cohort will be more cognizant of sample collection and processing time to limit confounding influences during comparative analyses.

In conclusion, this study identified apo-transcobalamin-II (APO-TCN2) as a novel plasma proteomic biomarker of survival in OSA. We provide evidence that increases in APO-TCN2 likely reflect a systemic and local inflammatory myeloid response which drives proliferation and intratumoral infiltration of B lymphocytes for improved anti-tumor immunity in these patients. This finding correlated with the only known documented investigation of TCN2 in this disease, which previously suggested TCN2 as one of four upregulated survival genes in OSA tumors of the TARGET-OS dataset (118). Further expansion of our patient cohort for validation of APO-TCN2’s clinical utility as a biomarker of improved overall survival is warranted. Future studies to investigate the distinct biological function of APO-TCN2 in both OSA and lymphoproliferative autoimmune disorders are necessary. Considering the availability of recombinant forms of this globulin protein, studies which investigate therapeutic injection of APO-TCN2 in murine models of metastatic OSA could ultimately suggest a future therapeutic role for this protein. Exogenous TCN2 may be critical to drive the robust expansion and lymphocytic proliferation necessary for enhanced anti-tumor immune responses in these patients.

References

• 1

  DiazPMLeehansARavishankarPDailyA. Multiomic approaches for cancer biomarker discovery in liquid biopsies: advances and challenges. biomark Insights. (2023) 18:11772719231204508. doi: 10.1177/11772719231204508

  • CrossRef
  • Google Scholar

• 2

  PruisMAGroenendijkFHBadloeKSvan PuffelenARobbrechtDDinjensWNMet al. Personalised selection of experimental treatment in patients with advanced solid cancer is feasible using whole-genome sequencing. Br J Cancer. (2022) 127:776–83. doi: 10.1038/s41416-022-01841-3

  • CrossRef
  • Google Scholar

• 3

  Jiménez-SantosMJGarcía-MartínSFustero-TorreCDi DomenicoTGómez-LópezGAl-ShahrourF. Bioinformatics roadmap for therapy selection in cancer genomics. Mol Oncol. (2022) 16:3881–908. doi: 10.1002/1878-0261.13286

  • CrossRef
  • Google Scholar

• 4

  AvilaMMeric-BernstamF. Next-generation sequencing for the general cancer patient. Clin Adv Hematol Oncol. (2019) 17:447–54.

  • Google Scholar

• 5

  YangXKuiLTangMLiDWeiKChenWet al. High-throughput transcriptome profiling in drug and biomarker discovery. Front Genet. (2020) 11:19. doi: 10.3389/fgene.2020.00019

  • CrossRef
  • Google Scholar

• 6

  HristovaVAChanDW. Cancer biomarker discovery and translation: proteomics and beyond. Expert Rev Proteomics. (2019) 16:93–103. doi: 10.1080/14789450.2019.1559062

  • CrossRef
  • Google Scholar

• 7

  SisodiyaSKasherwalVKhanARoyBGoelAKumarSet al. Liquid Biopsies: Emerging role and clinical applications in solid tumours. Transl Oncol. (2023) 35:101716. doi: 10.1016/j.tranon.2023.101716

  • CrossRef
  • Google Scholar

• 8

  LinDShenLLuoMZhangKLiJYangQet al. Circulating tumor cells: biology and clinical significance. Signal Transduct Target Ther. (2021) 6:404. doi: 10.1038/s41392-021-00817-8

  • CrossRef
  • Google Scholar

• 9

  Cortés-HernándezLEEslami-SZPantelKAlix-PanabièresC. Circulating tumor cells: from basic to translational research. Clin Chem. (2024) 70:81–9. doi: 10.1093/clinchem/hvad142

  • CrossRef
  • Google Scholar

• 10

  ZhouBXuKZhengXChenTWangJSongYet al. Application of exosomes as liquid biopsy in clinical diagnosis. Signal Transduct Target Ther. (2020) 5:144. doi: 10.1038/s41392-020-00258-9

  • CrossRef
  • Google Scholar

• 11

  YuDLiYWangMGuJXuWCaiHet al. Exosomes as a new frontier of cancer liquid biopsy. Mol Cancer. (2022) 21:56. doi: 10.1186/s12943-022-01509-9

  • CrossRef
  • Google Scholar

• 12

  CesconDWBratmanSVChanSMSiuLL. Circulating tumor DNA and liquid biopsy in oncology. Nat Cancer. (2020) 1:276–90. doi: 10.1038/s43018-020-0043-5

  • CrossRef
  • Google Scholar

• 13

  MaMZhuHZhangCSunXGaoXChenG. "Liquid biopsy"-ctDNA detection with great potential and challenges. Ann Transl Med. (2015) 3:235. doi: 10.3978/j.issn.2305-5839.2015.09.29

  • CrossRef
  • Google Scholar

• 14

  MathaiRAVidyaRVSReddyBSThomasLUdupaKKolesarJet al. Potential utility of liquid biopsy as a diagnostic and prognostic tool for the assessment of solid tumors: implications in the precision oncology. J Clin Med. (2019) 8. doi: 10.3390/jcm8030373

  • CrossRef
  • Google Scholar

• 15

  PinzaniPD'ArgenioVDel ReMPellegriniCCucchiaraFSalviantiFet al. Updates on liquid biopsy: current trends and future perspectives for clinical application in solid tumors. Clin Chem Lab Med. (2021) 59:1181–200. doi: 10.1515/cclm-2020-1685

  • CrossRef
  • Google Scholar

• 16

  WeiserDAWest-SzymanskiDCFraintEWeinerSRivasMAZhaoCWTet al. Progress toward liquid biopsies in pediatric solid tumors. Cancer Metastasis Rev. (2019) 38:553–71. doi: 10.1007/s10555-019-09825-1

  • CrossRef
  • Google Scholar

• 17

  van der LaanPvan HoudtWJvan den BroekDSteeghsNvan der GraafWTA. Liquid biopsies in sarcoma clinical practice: where do we stand? Biomedicines. (2021) 9. doi: 10.3390/biomedicines9101315

  • CrossRef
  • Google Scholar

• 18

  LiXSeebacherNAHornicekFJXiaoTDuanZ. Application of liquid biopsy in bone and soft tissue sarcomas: Present and future. Cancer Lett. (2018) 439:66–77. doi: 10.1016/j.canlet.2018.09.012

  • CrossRef
  • Google Scholar

• 19

  DingZWangNJiNChenZS. Proteomics technologies for cancer liquid biopsies. Mol Cancer. (2022) 21:53. doi: 10.1186/s12943-022-01526-8

  • CrossRef
  • Google Scholar

• 20

  SallamRM. Proteomics in cancer biomarkers discovery: challenges and applications. Dis Markers. (2015) 2015:321370. doi: 10.1155/2015/321370

  • CrossRef
  • Google Scholar

• 21

  MesriM. Advances in proteomic technologies and its contribution to the field of cancer. Adv Med. (2014) 2014:238045. doi: 10.1155/2014/238045

  • CrossRef
  • Google Scholar

• 22

  KhalilpourAKilicTKhalilpourSÁlvarezMMYazdiIK. Proteomic-based biomarker discovery for development of next generation diagnostics. Appl Microbiol Biotechnol. (2017) 101:475–91. doi: 10.1007/s00253-016-8029-z

  • CrossRef
  • Google Scholar

• 23

  GoldLAyersDBertinoJBockCBockABrodyENet al. Aptamer-based multiplexed proteomic technology for biomarker discovery. PLoS One. (2010) 5:e15004. doi: 10.1371/journal.pone.0015004

  • CrossRef
  • Google Scholar

• 24

  TuerkCGoldL. Systematic evolution of ligands by exponential enrichment: RNA ligands to bacteriophage T4 DNA polymerase. Science. (1990) 249:505–10. doi: 10.1126/science.2200121

  • CrossRef
  • Google Scholar

• 25

  EllingtonADSzostakJW. In vitro selection of RNA molecules that bind specific ligands. Nature. (1990) 346:818–22. doi: 10.1038/346818a0

  • CrossRef
  • Google Scholar

• 26

  KraemerSVaughtJDBockCGoldLKatiliusEKeeneyTRet al. From SOMAmer-based biomarker discovery to diagnostic and clinical applications: a SOMAmer-based, streamlined multiplex proteomic assay. PLoS One. (2011) 6:e26332. doi: 10.1371/journal.pone.0026332

  • CrossRef
  • Google Scholar

• 27

  GoldLWalkerJJWilcoxSKWilliamsS. Advances in human proteomics at high scale with the SOMAscan proteomics platform. N Biotechnol. (2012) 29:543–9. doi: 10.1016/j.nbt.2011.11.016

  • CrossRef
  • Google Scholar

• 28

  QiaoZPanXParlayanCOjimaHKondoT. Proteomic study of hepatocellular carcinoma using a novel modified aptamer-based array (SOMAscan™) platform. Biochim Biophys Acta Proteins Proteom. (2017) 1865:434–43. doi: 10.1016/j.bbapap.2016.09.011

  • CrossRef
  • Google Scholar

• 29

  LiHVanarsaKZhangTSoomroSCicalesePADuranVet al. Comprehensive aptamer-based screen of 1317 proteins uncovers improved stool protein markers of colorectal cancer. J Gastroenterol. (2021) 56:659–72. doi: 10.1007/s00535-021-01795-y

  • CrossRef
  • Google Scholar

• 30

  NarasimhanAShahdaSKaysJKPerkinsSMChengLSchlossKNHet al. Identification of potential serum protein biomarkers and pathways for pancreatic cancer cachexia using an aptamer-based discovery platform. Cancers (Basel). (2020) 12. doi: 10.3390/cancers12123787

  • CrossRef
  • Google Scholar

• 31

  ThanasupawatTGlogowskaAPascoeCKrishnanSNMunirMBegumFet al. Slow off-rate modified aptamer (SOMAmer) proteomic analysis of patient-derived Malignant glioma identifies distinct cellular proteomes. Int J Mol Sci. (2021) 22. doi: 10.3390/ijms22179566

  • CrossRef
  • Google Scholar

• 32

  OstroffRMBigbeeWLFranklinWGoldLMehanMMillerYEet al. Unlocking biomarker discovery: large scale application of aptamer proteomic technology for early detection of lung cancer. PLoS One. (2010) 5:e15003. doi: 10.1371/journal.pone.0015003

  • CrossRef
  • Google Scholar

• 33

  BlattSKämmererPWKrügerMSurabattulaRThiemDGEDillonSTet al. High-multiplex aptamer-based serum proteomics to identify candidate serum biomarkers of oral squamous cell carcinoma. Cancers (Basel). (2023) 15. doi: 10.3390/cancers15072071

  • CrossRef
  • Google Scholar

• 34

  FinkernagelFReinartzSSchuldnerMMalzAJansenJMWagnerUet al. Dual-platform affinity proteomics identifies links between the recurrence of ovarian carcinoma and proteins released into the tumor microenvironment. Theranostics. (2019) 9:6601–17. doi: 10.7150/thno.37549

  • CrossRef
  • Google Scholar

• 35

  CandiaJDayaGNTanakaTFerrucciLWalkerKA. Assessment of variability in the plasma 7k SomaScan proteomics assay. Sci Rep. (2022) 12:17147. doi: 10.1038/s41598-022-22116-0

  • CrossRef
  • Google Scholar

• 36

  LimSYLeeJHWelshSJAhnSBBreenEKhanAet al. Evaluation of two high-throughput proteomic technologies for plasma biomarker discovery in immunotherapy-treated melanoma patients. biomark Res. (2017) 5:32. doi: 10.1186/s40364-017-0112-9

  • CrossRef
  • Google Scholar

• 37

  GillJGorlickR. Advancing therapy for osteosarcoma. Nat Rev Clin Oncol. (2021) 18:609–24. doi: 10.1038/s41571-021-00519-8

  • CrossRef
  • Google Scholar

• 38

  YangCTianYZhaoFChenZSuPLiYet al. Bone microenvironment and osteosarcoma metastasis. Int J Mol Sci. (2020) 21:6985. doi: 10.3390/ijms21196985

  • CrossRef
  • Google Scholar

• 39

  SmrkeAAndersonPMGuliaAGennatasSHuangPHJonesRL. Future directions in the treatment of osteosarcoma. Cells. (2021) 10. doi: 10.3390/cells10010172

  • CrossRef
  • Google Scholar

• 40

  CorreIVerrecchiaFCrennVRediniFTrichetV. The osteosarcoma microenvironment: A complex but targetable ecosystem. Cells. (2020) 9. doi: 10.3390/cells9040976

  • CrossRef
  • Google Scholar

• 41

  GasparNMarques Da CostaMEFromigueODroitRBerlangaPMarchaisA. Recent advances in understanding osteosarcoma and emerging therapies. Faculty Rev. (2020) 9. doi: 10.12703/r/9-18

  • CrossRef
  • Google Scholar

• 42

  CasciniCChiodoniC. The immune landscape of osteosarcoma: implications for prognosis and treatment response. Cells. (2021) 10. doi: 10.3390/cells10071668

  • CrossRef
  • Google Scholar

• 43

  AranVDevalleSMeohasWHeringerMCunha CarusoAPinheiro AguiarDet al. Osteosarcoma, chondrosarcoma and Ewing sarcoma: Clinical aspects, biomarker discovery and liquid biopsy. Crit Rev Oncol Hematol. (2021) 162:103340. doi: 10.1016/j.critrevonc.2021.103340

  • CrossRef
  • Google Scholar

• 44

  UcciARucciNPonzettiM. Liquid biopsies in primary and secondary bone cancers. Cancer Drug Resist. (2022) 5:541–59. doi: 10.20517/cdr.2022.17

  • CrossRef
  • Google Scholar

• 45

  RaimondiLDe LucaACostaVAmodioNCarinaVBellaviaDet al. Circulating biomarkers in osteosarcoma: new translational tools for diagnosis and treatment. Oncotarget. (2017) 8:100831–51. doi: 10.18632/oncotarget.19852

  • CrossRef
  • Google Scholar

• 46

  EvolaFRCostarellaLPavoneVCaffGCannavòLSessaAet al. Biomarkers of osteosarcoma, chondrosarcoma, and ewing sarcoma. Front Pharmacol. (2017) 8:150. doi: 10.3389/fphar.2017.00150

  • CrossRef
  • Google Scholar

• 47

  Wan-IbrahimWISinghVAHashimOHAbdul-RahmanPS. Biomarkers for bone tumors: discovery from genomics and proteomics studies and their challenges. Mol Med. (2016) 21:861–72. doi: 10.2119/molmed.2015.00183

  • CrossRef
  • Google Scholar

• 48

  SavitskayaYARico-MartínezGLinares-GonzálezLMDelgado-CedilloEATéllez-GastelumRAlfaro-RodríguezABet al. Serum tumor markers in pediatric osteosarcoma: a summary review. Clin Sarcoma Res. (2012) 2:9. doi: 10.1186/2045-3329-2-9

  • CrossRef
  • Google Scholar

• 49

  HolzerGPfandlsteinerTBlahovecHTriebKKotzR. Serum levels of TNF-beta and sTNF-R in patients with Malignant bone tumours. Anticancer Res. (2003) 23:3057–9.

  • Google Scholar

• 50

  MarkiewiczKZemanKKozarAGołebiowska-WawrzyniakM. Evaluation of selected cytokines in children and adolescents with osteosarcoma at diagnosis - preliminary report. Med Wieku Rozwoj. (2011) 15:25–31.

  • Google Scholar

• 51

  ZhangCWangLXiongCZhaoRLiangHLuoX. The role of vascular endothelial growth factor as a prognostic and clinicopathological marker in osteosarcoma: a systematic review and meta-analysis. J Orthop Surg Res. (2021) 16:738. doi: 10.1186/s13018-021-02888-3

  • CrossRef
  • Google Scholar

• 52

  ChenDZhangYJZhuKWWangWC. A systematic review of vascular endothelial growth factor expression as a biomarker of prognosis in patients with osteosarcoma. Tumour Biol. (2013) 34:1895–9. doi: 10.1007/s13277-013-0733-z

  • CrossRef
  • Google Scholar

• 53

  QuJTWangMHeHLTangYYeXJ. The prognostic value of elevated vascular endothelial growth factor in patients with osteosarcoma: a meta-analysis and systemic review. J Cancer Res Clin Oncol. (2012) 138:819–25. doi: 10.1007/s00432-012-1149-7

  • CrossRef
  • Google Scholar

• 54

  KushlinskiiNEBabkinaIVSolov'evYNTrapeznikovNN. Vascular endothelium growth factor and angiogenin in the serum of patients with osteosarcoma and Ewing's tumor. Bull Exp Biol Med. (2000) 130:691–3. doi: 10.1007/BF02682107

  • CrossRef
  • Google Scholar

• 55

  LiuGLiuBTangLLiuZDaiH. Cytokines as prognostic biomarkers in osteosarcoma patients: A systematic review and meta-analysis. J Interferon Cytokine Res. (2023) 43:335–43. doi: 10.1089/jir.2023.0083

  • CrossRef
  • Google Scholar

• 56

  LiYFloresRYuAOkcuMFMurrayJChintagumpalaMet al. Elevated expression of CXC chemokines in pediatric osteosarcoma patients. Cancer. (2011) 117:207–17. doi: 10.1002/cncr.v117.1

  • CrossRef
  • Google Scholar

• 57

  LiYDangTAShenJHicksJChintagumpalaMLauCCet al. Plasma proteome predicts chemotherapy response in osteosarcoma patients. Oncol Rep. (2011) 25:303–14. doi: 10.3892/or.2010.1111

  • CrossRef
  • Google Scholar

• 58

  WilliamsSAKivimakiMLangenbergCHingoraniADCasasJPBouchardCet al. Plasma protein patterns as comprehensive indicators of health. Nat Med. (2019) 25:1851–7. doi: 10.1038/s41591-019-0665-2

  • CrossRef
  • Google Scholar

• 59

  FabregatASidiropoulosKGarapatiPGillespieMHausmannKHawRet al. The reactome pathway knowledgebase. Nucleic Acids Res. (2016) 44:D481–7. doi: 10.1093/nar/gkv1351

  • CrossRef
  • Google Scholar

• 60

  FabregatAJupeSMatthewsLSidiropoulosKGillespieMGarapatiPet al. The reactome pathway knowledgebase. Nucleic Acids Res. (2018) 46:D649–D55. doi: 10.1093/nar/gkx1132

  • CrossRef
  • Google Scholar

• 61

  FabregatASidiropoulosKViteriGFornerOMarin-GarciaPArnauVet al. Reactome pathway analysis: a high-performance in-memory approach. BMC Bioinf. (2017) 18:142. doi: 10.1186/s12859-017-1559-2

  • CrossRef
  • Google Scholar

• 62

  DwivediBMummeHSatpathySBhasinSSBhasinM. Survival Genie, a web platform for survival analysis across pediatric and adult cancers. Sci Rep. (2022) 12:3069. doi: 10.1038/s41598-022-06841-0

  • CrossRef
  • Google Scholar

• 63

  ContalCO'QuigleyJ. An application of changepoint methods in studying the effect of age on survival in breast cancer. Comput Stat Data Anal (1999) 30:253–70. doi: 10.1016/S0167-9473(98)00096-6

  • CrossRef
  • Google Scholar

• 64

  DardisC. survMisc: miscellaneous functions for survival data. R-project. (2022) 0.5:6.

  • Google Scholar

• 65

  ChenBKhodadoustMSLiuCLNewmanAMAlizadehAA. Profiling tumor infiltrating immune cells with CIBERSORT. Methods Mol Biol. (2018) 1711:243–59. doi: 10.1007/978-1-4939-7493-1_12

  • CrossRef
  • Google Scholar

• 66

  ChenZLuoZZhangDLiHLiuXZhuKet al. TIGER: A web portal of tumor immunotherapy gene expression resource. Genomics Proteomics Bioinf. (2023) 21:337–48. doi: 10.1016/j.gpb.2022.08.004

  • CrossRef
  • Google Scholar

• 67

  LambrechtsDWautersEBoeckxBAibarSNittnerDBurtonOet al. Phenotype molding of stromal cells in the lung tumor microenvironment. Nat Med. (2018) 24:1277–89. doi: 10.1038/s41591-018-0096-5

  • CrossRef
  • Google Scholar

• 68

  WuTDMadireddiSde AlmeidaPEBanchereauRChenYJChitreASet al. Peripheral T cell expansion predicts tumour infiltration and clinical response. Nature. (2020) 579:274–8. doi: 10.1038/s41586-020-2056-8

  • CrossRef
  • Google Scholar

• 69

  KimNKimHKLeeKHongYChoJHChoiJWet al. Single-cell RNA sequencing demonstrates the molecular and cellular reprogramming of metastatic lung adenocarcinoma. Nat Commun. (2020) 11:2285. doi: 10.1038/s41467-020-16164-1

  • CrossRef
  • Google Scholar

• 70

  MaynardAMcCoachCERotowJKHarrisLHaderkFKerrDLet al. Therapy-induced evolution of human lung cancer revealed by single-cell RNA sequencing. Cell. (2020) 182:1232–51.e22. doi: 10.1016/j.cell.2020.07.017

  • CrossRef
  • Google Scholar

• 71

  Sade-FeldmanMYizhakKBjorgaardSLRayJPde BoerCGJenkinsRWet al. Defining T cell states associated with response to checkpoint immunotherapy in melanoma. Cell. (2019) 176:404. doi: 10.1016/j.cell.2018.12.034

  • CrossRef
  • Google Scholar

• 72

  LiuYFengWDaiYBaoMYuanZHeMet al. Single-cell transcriptomics reveals the complexity of the tumor microenvironment of treatment-naive osteosarcoma. Front Oncol. (2021) 11:709210. doi: 10.3389/fonc.2021.709210

  • CrossRef
  • Google Scholar

• 73

  KhodadadiLChengQRadbruchAHiepeF. The maintenance of memory plasma cells. Front Immunol. (2019) 10:721. doi: 10.3389/fimmu.2019.00721

  • CrossRef
  • Google Scholar

• 74

  SangesSGuerrierTDuhamelAGuilbertLHauspieCLargyAet al. Soluble markers of B cell activation suggest a role of B cells in the pathogenesis of systemic sclerosis-associated pulmonary arterial hypertension. Front Immunol. (2022) 13:954007. doi: 10.3389/fimmu.2022.954007

  • CrossRef
  • Google Scholar

• 75

  LiRBerglundAZempLDhillonJPutneyRKimYet al. The 12-CK score: global measurement of tertiary lymphoid structures. Front Immunol. (2021) 12:694079. doi: 10.3389/fimmu.2021.694079

  • CrossRef
  • Google Scholar

• 76

  KovácsSAFeketeJTGyőrffyB. Predictive biomarkers of immunotherapy response with pharmacological applications in solid tumors. Acta Pharmacol Sin. (2023) 44:1879–89. doi: 10.1038/s41401-023-01079-6

  • CrossRef
  • Google Scholar

• 77

  TangZKangBLiCChenTZhangZ. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Res. (2019) 47:W556–W60. doi: 10.1093/nar/gkz430

  • CrossRef
  • Google Scholar

• 78

  KoseEBesciOGudelogluESuncakSOymakYOzenSet al. Transcobalamin II deficiency in twins with a novel variant in the TCN2 gene: case report and review of literature. J Pediatr Endocrinol Metab. (2020) 33:1487–99. doi: 10.1515/jpem-2020-0096

  • CrossRef
  • Google Scholar

• 79

  RatschmannRMinkovMKisAHungCRuparTMühlAet al. Transcobalamin II deficiency at birth. Mol Genet Metab. (2009) 98:285–8. doi: 10.1016/j.ymgme.2009.06.003

  • CrossRef
  • Google Scholar

• 80

  UnalSTezolOOztasY. A novel mutation of the transcobalamin II gene in an infant presenting with hemophagocytic lymphohistiocytosis. Int J Hematol. (2014) 99:659–62. doi: 10.1007/s12185-014-1545-7

  • CrossRef
  • Google Scholar

• 81

  ZhanSChengFHeHHuSFengX. Identification of transcobalamin deficiency with two novel mutations in the TCN2 gene in a Chinese girl with abnormal immunity: a case report. BMC Pediatr. (2020) 20:460. doi: 10.1186/s12887-020-02357-6

  • CrossRef
  • Google Scholar

• 82

  FehrJDe VecchiP. Transcobalamin II: a marker for macrophage/histiocyte proliferation. Am J Clin Pathol. (1985) 84:291–6. doi: 10.1093/ajcp/84.3.291

  • CrossRef
  • Google Scholar

• 83

  LiLZhangXKovacicSLongAJBourqueKWoodCRet al. Identification of a human follicular dendritic cell molecule that stimulates germinal center B cell growth. J Exp Med. (2000) 191:1077–84. doi: 10.1084/jem.191.6.1077

  • CrossRef
  • Google Scholar

• 84

  QuadrosEVSequeiraJM. Cellular uptake of cobalamin: transcobalamin and the TCblR/CD320 receptor. Biochimie. (2013) 95:1008–18. doi: 10.1016/j.biochi.2013.02.004

  • CrossRef
  • Google Scholar

• 85

  LaiSCNakayamaYSequeiraJMQuadrosEV. Down-regulation of transcobalamin receptor TCblR/CD320 by siRNA inhibits cobalamin uptake and proliferation of cells in culture. Exp Cell Res. (2011) 317:1603–7. doi: 10.1016/j.yexcr.2011.02.016

  • CrossRef
  • Google Scholar

• 86

  ChoWChoiJParkCHYoonSOJeoungDIKimYMet al. Expression of CD320 in human B cells in addition to follicular dendritic cells. BMB Rep. (2008) 41:863–7. doi: 10.5483/BMBRep.2008.41.12.863

  • CrossRef
  • Google Scholar

• 87

  SyselAMValliVENagleRBBauerJA. Immunohistochemical quantification of the vitamin B12 transport protein (TCII), cell surface receptor (TCII-R) and Ki-67 in human tumor xenografts. Anticancer Res. (2013) 33:4203–12.

  • Google Scholar

• 88

  OreshkinAEGudkovaMVMiasishchevaNV. The expression of plasmalemma transcobalamin-II receptors on human blood lymphocytes stimulated by mitogens. Biull Eksp Biol Med. (1992) 114:185–7. doi: 10.1007/BF00800082

  • CrossRef
  • Google Scholar

• 89

  McLeanGRQuadrosEVRothenbergSPMorganACSchraderJWZiltenerHJ. Antibodies to transcobalamin II block in vitro proliferation of leukemic cells. Blood. (1997) 89:235–42. doi: 10.1182/blood.V89.1.235

  • CrossRef
  • Google Scholar

• 90

  McLeanGRWilliamsMJWoodhouseCSZiltenerHJ. Transcobalamin II and in vitro proliferation of leukemic cells. Leuk Lymphoma. (1998) 30:101–9. doi: 10.3109/10428199809050933

  • CrossRef
  • Google Scholar

• 91

  LiuBXuKLiAXuJCuiY. Inhibition of transcolbalamin2 overexpression ameliorated renal injury in systemic lupus erythema via reducing pathogenic Tfh and B cell infiltration. Lupus Sci Med. (2023) 10. doi: 10.1136/lupus-2023-KCR.37

  • CrossRef
  • Google Scholar

• 92

  CalderaroJPetitprezFBechtELaurentAHirschTZRousseauBet al. Intra-tumoral tertiary lymphoid structures are associated with a low risk of early recurrence of hepatocellular carcinoma. J Hepatol. (2019) 70:58–65. doi: 10.1016/j.jhep.2018.09.003

  • CrossRef
  • Google Scholar

• 93

  DingGYMaJQYunJPChenXLingYZhangSet al. Distribution and density of tertiary lymphoid structures predict clinical outcome in intrahepatic cholangiocarcinoma. J Hepatol. (2022) 76:608–18. doi: 10.1016/j.jhep.2021.10.030

  • CrossRef
  • Google Scholar

• 94

  DomblidesCRochefortJRiffardCPanouillotMLescailleGTeillaudJLet al. Tumor-associated tertiary lymphoid structures: from basic and clinical knowledge to therapeutic manipulation. Front Immunol. (2021) 12:698604. doi: 10.3389/fimmu.2021.698604

  • CrossRef
  • Google Scholar

• 95

  GermainCGnjaticSTamzalitFKnockaertSRemarkRGocJet al. Presence of B cells in tertiary lymphoid structures is associated with a protective immunity in patients with lung cancer. Am J Respir Crit Care Med. (2014) 189:832–44. doi: 10.1164/rccm.201309-1611OC

  • CrossRef
  • Google Scholar

• 96

  HeWZhangDLiuHChenTXieJPengLet al. The high level of tertiary lymphoid structure is correlated with superior survival in patients with advanced gastric cancer. Front Oncol. (2020) 10:980. doi: 10.3389/fonc.2020.00980

  • CrossRef
  • Google Scholar

• 97

  JiangQTianCWuHMinLChenHChenLet al. Tertiary lymphoid structure patterns predicted anti-PD1 therapeutic responses in gastric cancer. Chin J Cancer Res. (2022) 34:365–82. doi: 10.21147/j.issn.1000-9604.2022.04.05

  • CrossRef
  • Google Scholar

• 98

  KemiNYlitaloOVäyrynenJPHelminenOJunttilaAMrenaJet al. Tertiary lymphoid structures and gastric cancer prognosis. APMIS. (2023) 131:19–25. doi: 10.1111/apm.v131.1

  • CrossRef
  • Google Scholar

• 99

  KroegerDRMilneKNelsonBH. Tumor-infiltrating plasma cells are associated with tertiary lymphoid structures, cytolytic T-cell responses, and superior prognosis in ovarian cancer. Clin Cancer Res. (2016) 22:3005–15. doi: 10.1158/1078-0432.CCR-15-2762

  • CrossRef
  • Google Scholar

• 100

  LiQLiuXWangDWangYLuHWenSet al. Prognostic value of tertiary lymphoid structure and tumour infiltrating lymphocytes in oral squamous cell carcinoma. Int J Oral Sci. (2020) 12:24. doi: 10.1038/s41368-020-00092-3

  • CrossRef
  • Google Scholar

• 101

  LiuZMengXTangXZouWHeY. Intratumoral tertiary lymphoid structures promote patient survival and immunotherapy response in head neck squamous cell carcinoma. Cancer Immunol Immunother. (2022) 72(6):1505–21. doi: 10.21203/rs.3.rs-1987850/v1

  • CrossRef
  • Google Scholar

• 102

  MoriTTanakaHDeguchiSYamakoshiYMikiYYoshiiMet al. Clinical efficacy of nivolumab is associated with tertiary lymphoid structures in surgically resected primary tumors of recurrent gastric cancer. PloS One. (2022) 17:e0262455. doi: 10.1371/journal.pone.0262455

  • CrossRef
  • Google Scholar

• 103

  JNTJNSlBGt. Tertiary lymphoid structures and B lymphocytes in cancer prognosis and response to immunotherapies. Oncoimmunology. (2021) 10:1900508. doi: 10.1080/2162402X.2021.1900508

  • CrossRef
  • Google Scholar

• 104

  Sautès-FridmanCLawandMGiraldoNAKaplonHGermainCFridmanWHet al. Tertiary lymphoid structures in cancers: prognostic value, regulation, and manipulation for therapeutic intervention. Front Immunol. (2016) 7:407. doi: 10.3389/fimmu.2016.00407

  • CrossRef
  • Google Scholar

• 105

  SchumacherTNThommenDS. Tertiary lymphoid structures in cancer. Science. (2022) 375:eabf9419. doi: 10.1126/science.abf9419

  • CrossRef
  • Google Scholar

• 106

  TrübMZippeliusA. Tertiary lymphoid structures as a predictive biomarker of response to cancer immunotherapies. Front Immunol. (2021) 12:674565. doi: 10.3389/fimmu.2021.674565

  • CrossRef
  • Google Scholar

• 107

  VaghjianiRGSkitzkiJJ. Tertiary lymphoid structures as mediators of immunotherapy response. Cancers (Basel). (2022) 14. doi: 10.3390/cancers14153748

  • CrossRef
  • Google Scholar

• 108

  VanherseckeLBrunetMGuéganJPReyCBougouinACousinSet al. Mature tertiary lymphoid structures predict immune checkpoint inhibitor efficacy in solid tumors independently of PD-L1 expression. Nat Cancer. (2021) 2:794–802. doi: 10.1038/s43018-021-00232-6

  • CrossRef
  • Google Scholar

• 109

  WakasuSTagawaTHaratakeNKinoshitaFOkuYOnoYet al. Preventive effect of tertiary lymphoid structures on lymph node metastasis of lung adenocarcinoma. Cancer Immunol Immunother. (2023) 72(6):1823–34. doi: 10.21203/rs.3.rs-2055351/v1

  • CrossRef
  • Google Scholar

• 110

  WangBLiuJHanYDengYLiJJiangY. The presence of tertiary lymphoid structures provides new insight into the clinicopathological features and prognosis of patients with breast cancer. Front Immunol. (2022) 13:868155. doi: 10.3389/fimmu.2022.868155

  • CrossRef
  • Google Scholar

• 111

  YuJSHuangWBZhangYHChenJLiJFuHFet al. The association of immune cell infiltration and prognostic value of tertiary lymphoid structures in gastric cancer. Neoplasma. (2022) 69:886–98. doi: 10.4149/neo_2022_220128N123

  • CrossRef
  • Google Scholar

• 112

  ZhangQWuS. Tertiary lymphoid structures are critical for cancer prognosis and therapeutic response. Front Immunol. (2022) 13:1063711. doi: 10.3389/fimmu.2022.1063711

  • CrossRef
  • Google Scholar

• 113

  ZhangKXieXZouLHGuoSQ. Tertiary lymphoid structures are associated with a favorable prognosis in high-grade serous ovarian cancer patients. Reprod Sci. (2023) 30(8):2468–80. doi: 10.1007/s43032-023-01188-x

  • CrossRef
  • Google Scholar

• 114

  ZhouLXuBLiuYWangZ. Tertiary lymphoid structure signatures are associated with survival and immunotherapy response in muscle-invasive bladder cancer. Oncoimmunology. (2021) 10:1915574. doi: 10.1080/2162402X.2021.1915574

  • CrossRef
  • Google Scholar

• 115

  ZouJZhangYZengYPengYLiuJXiaoCet al. Tertiary lymphoid structures: A potential biomarker for anti-cancer therapy. Cancers (Basel). (2022) 14. doi: 10.3390/cancers14235968

  • CrossRef
  • Google Scholar

• 116

  ItalianoABessedeAPulidoMBompasEPiperno-NeumannSChevreauCet al. Pembrolizumab in soft-tissue sarcomas with tertiary lymphoid structures: a phase 2 PEMBROSARC trial cohort. Nat Med. (2022) 28:1199–206. doi: 10.1038/s41591-022-01821-3

  • CrossRef
  • Google Scholar

• 117

  PetitprezFde ReynièsAKeungEZChenTWSunCMCalderaroJet al. B cells are associated with survival and immunotherapy response in sarcoma. Nature. (2020) 577:556–60. doi: 10.1038/s41586-019-1906-8

  • CrossRef
  • Google Scholar

• 118

  RothzergEXuJWoodDKõksS. 12 Survival-related differentially expressed genes based on the TARGET-osteosarcoma database. Exp Biol Med (Maywood). (2021) 246:2072–81. doi: 10.1177/15353702211007410

  • CrossRef
  • Google Scholar

• 119

  SeetharamBBoseSLiN. Cellular import of cobalamin (Vitamin B-12). J Nutr. (1999) 129:1761–4. doi: 10.1093/jn/129.10.1761

  • CrossRef
  • Google Scholar

• 120

  SeetharamB. Receptor-mediated endocytosis of cobalamin (vitamin B12). Annu Rev Nutr. (1999) 19:173–95. doi: 10.1146/annurev.nutr.19.1.173

  • CrossRef
  • Google Scholar

• 121

  SeetharamBLiN. Transcobalamin II and its cell surface receptor. Vitam Horm. (2000) 59:337–66. doi: 10.1016/S0083-6729(00)59012-8

  • CrossRef
  • Google Scholar

• 122

  BarshopBAWolffJNyhanWLYuAProdanosCJonesGet al. Transcobalamin II deficiency presenting with methylmalonic aciduria and homocystinuria and abnormal absorption of cobalamin. Am J Med Genet. (1990) 35:222–8. doi: 10.1002/ajmg.1320350216

  • CrossRef
  • Google Scholar

• 123

  BibiHGelman-KohanZBaumgartnerERRosenblattDS. Transcobalamin II deficiency with methylmalonic aciduria in three sisters. J Inherit Metab Dis. (1999) 22:765–72. doi: 10.1023/A:1005507204491

  • CrossRef
  • Google Scholar

• 124

  ChaoMMIllsingerSYoshimiADasAMKratzCP. Congenital transcobalamin II deficiency: A rare entity with a broad differential. Klin Padiatr. (2017) 229:355–7. doi: 10.1055/s-0043-120266

  • CrossRef
  • Google Scholar

• 125

  HitzigWHDohmannUPlussHJVischerD. Hereditary transcobalamin II deficiency: clinical findings in a new family. J Pediatr. (1974) 85:622–8. doi: 10.1016/S0022-3476(74)80503-2

  • CrossRef
  • Google Scholar

• 126

  HitzigWHFràter-SchröderMSegerR. Immunodeficiency due to transcobalamin II deficiency. Ciba Found Symp. (1978) 68:77–91. doi: 10.1002/9780470720516

  • CrossRef
  • Google Scholar

• 127

  KaikovYWadsworthLDHallCARogersPC. Transcobalamin II deficiency: case report and review of the literature. Eur J Pediatr. (1991) 150:841–3. doi: 10.1007/BF01955004

  • CrossRef
  • Google Scholar

• 128

  MeyersPACarmelR. Hereditary transcobalamin II deficiency with subnormal serum cobalamin levels. Pediatrics. (1984) 74:866–71. doi: 10.1542/peds.74.5.866

  • CrossRef
  • Google Scholar

• 129

  NiebruggeDJBenjaminDRChristieDScottCR. Hereditary transcobalamin II deficiency presenting as red cell hypoplasia. J Pediatr. (1982) 101:732–5. doi: 10.1016/S0022-3476(82)80304-1

  • CrossRef
  • Google Scholar

• 130

  PrasadCRosenblattDSCorleyKCairneyAERuparCA. Transcobalamin (TC) deficiency–potential cause of bone marrow failure in childhood. J Inherit Metab Dis. (2008) 31 Suppl 2:S287–92. doi: 10.1007/s10545-008-0864-3

  • CrossRef
  • Google Scholar

• 131

  RanaSRColmanNGohKOHerbertVKlempererMR. Transcobalamin II deficiency associated with unusual bone marrow findings and chromosomal abnormalities. Am J Hematol. (1983) 14:89–96. doi: 10.1002/ajh.2830140111

  • CrossRef
  • Google Scholar

• 132

  SakiyamaY. Hereditary transcobalamin II deficiency. Ryoikibetsu Shokogun Shirizu. (2000) 32:346–8.

  • Google Scholar

• 133

  ScottCRHakamiNTengCCSagersonRN. Hereditary transcobalamin II deficiency: the role of transcobalamin II in vitamin B 12 -mediated reactions. J Pediatr. (1972) 81:1106–11. doi: 10.1016/S0022-3476(72)80239-7

  • CrossRef
  • Google Scholar

• 134

  SourialNA. Transcobalamin II deficiency in infancy and the diagnostic value of serum B12 binders. Scand J Haematol. (1984) 33:327–8. doi: 10.1111/j.1600-0609.1984.tb02237.x

  • CrossRef
  • Google Scholar

• 135

  ThomasPKHoffbrandAV. Hereditary transcobalamin II deficiency: a 22 year follow up. J Neurol Neurosurg Psychiatry. (1997) 62:197. doi: 10.1136/jnnp.62.2.197

  • CrossRef
  • Google Scholar

• 136

  ToretEÖzdemirZCBorÖ. Transcobalamin II deficiency in an infant with a novel mutation. Turk Arch Pediatr. (2022) 57:670–2. doi: 10.5152/TurkArchPediatr.2022.22100

  • CrossRef
  • Google Scholar

• 137

  ÜnalŞRuparTYetginSYaralıNDursunAGürselTet al. Transcobalamin II deficiency in four cases with novel mutations. Turk J Haematol. (2015) 32:317–22. doi: 10.4274/Tjh.2014.0154

  • CrossRef
  • Google Scholar

• 138

  ÜnalSKarahanFArıkoğluTAkarAKuyucuS. Different presentations of patients with transcobalamin II deficiency: A single-center experience from Turkey. Turk J Haematol. (2019) 36:37–42. doi: 10.4274/tjh.galenos.2018.2018.0230

  • CrossRef
  • Google Scholar

• 139

  SegerRFràter-SchröderMHitzigWHWildfeuerALinnellJC. Granulocyte dysfunction in transcobalamin II deficiency responding to leucovorin or hydroxocobalamin-plasma transfusion. J Inherit Metab Dis. (1980) 3:3–9. doi: 10.1007/BF02312515

  • CrossRef
  • Google Scholar

• 140

  CarmelRHollanderD. Extreme elevation of transcobalamin II levels in multiple myeloma and other disorders. Blood. (1978) 51:1057–63. doi: 10.1182/blood.V51.6.1057.1057

  • CrossRef
  • Google Scholar

• 141

  RachmilewitzDLigumskyMRachmilewitzBRachmilewitzMTarcicNSchlesingerM. Transcobalamin II level in peripheral blood monocytes–a biochemical marker in inflammatory diseases of the bowel. Gastroenterology. (1980) 78:43–6. doi: 10.1016/0016-5085(80)90190-0

  • CrossRef
  • Google Scholar

• 142

  AreekulSChurdchuKCheeramakaraCWilairatanaPCharoenlarpP. Persistently elevated serum transcobalamin II in a patient with cerebral malaria and typhus infections. J Med Assoc Thai. (1995) 78:48–52.

  • Google Scholar

• 143

  CheeramakaraCThanomsakWSongmeangKNontprasertASanghirunCSuthisaiNet al. Elevation of serum transcobalamin II in patients with scrub typhus. Southeast Asian J Trop Med Public Health. (2005) 36:113–7.

  • Google Scholar

• 144

  CheeramakaraCSongmeangKNakosiriWSuthisaiNNontprasertAAreekulS. Study on serum transcobalamin II in patients with murine typhus. Southeast Asian J Trop Med Public Health. (2006) 37 Suppl 3:145–8.

  • Google Scholar

• 145

  BauerAPachlEHellmuthJCKneidingerNHeydarianMFrankenbergerMet al. Proteomics reveals antiviral host response and NETosis during acute COVID-19 in high-risk patients. Biochim Biophys Acta Mol Basis Dis. (2023) 1869:166592. doi: 10.1016/j.bbadis.2022.166592

  • CrossRef
  • Google Scholar

• 146

  Fráter-SchröderMHitzigWHGrobPJKennyAB. Increase of unsaturated transcobalamine II in autoimmune diseases; effect of immunosuppressive therapy (proceedings). Schweiz Med Wochenschr. (1978) 108:1604.

  • Google Scholar

• 147

  Fràter-SchröderMHitzigWHGrobPJKennyAB. Increased unsaturated transcobalamin II in active autoimmune disease. Lancet. (1978) 2:238–9. doi: 10.1016/S0140-6736(78)91747-6

  • CrossRef
  • Google Scholar

• 148

  LässerUKieratLGrobPJHitzigWHFráter-SchröderM. Transcobalamin II, a serum protein reflecting autoimmune disease activity, its plasma dynamics, and the relationship to established serum parameters in systemic lupus erythematosus. Clin Immunol Immunopathol. (1985) 36:345–57. doi: 10.1016/0090-1229(85)90055-8

  • CrossRef
  • Google Scholar

• 149

  HaynesWAHaddonDJDiepVKKhatriABongenEYiuGet al. Integrated, multicohort analysis reveals unified signature of systemic lupus erythematosus. JCI Insight. (2020) 5. doi: 10.1172/jci.insight.122312

  • CrossRef
  • Google Scholar

• 150

  ArnalichFZamoranoAFBenito-UrbinaSGijón-BañosJDe MiguelEPeñaJMet al. Increased apotranscobalamin II levels in rheumatoid arthritis. Br J Rheumatol. (1990) 29:171–3. doi: 10.1093/rheumatology/29.3.171

  • CrossRef
  • Google Scholar

• 151

  DorrajiSEKanapathippillaiPHovdAKStenersrødMRHorveiKDUrsvikAet al. Kidney tertiary lymphoid structures in lupus nephritis develop into large interconnected networks and resemble lymph nodes in gene signature. Am J Pathol. (2020) 190:2203–25. doi: 10.1016/j.ajpath.2020.07.015

  • CrossRef
  • Google Scholar

• 152

  VreugdenhilGLindemansJvan EijkHGSwaakAJ. Elevated serum transcobalamin levels in anaemia of rheumatoid arthritis: correlation with disease activity but not with serum tumour necrosis factor alpha and interleukin 6. J Intern Med. (1992) 231:547–50. doi: 10.1111/j.1365-2796.1992.tb00972.x

  • CrossRef
  • Google Scholar

• 153

  RachmilewitzBRachmilewitzMChaouatMSchlesingerM. The synthesis of transcobalamin II, a vitamin B12 transport protein, by stimulated mouse peritoneal macrophages. Biomedicine. (1977) 27:213–4.

  • Google Scholar

• 154

  RabinowitzRRachmilewitzBRachmilewitzMSchlesingerM. Production of transcobalamin II by various murine and human cells in culture. Isr J Med Sci. (1982) 18:740–5.

  • Google Scholar

• 155

  CarmelRNeelySMFrancisRB. Human umbilical vein endothelial cells secrete transcobalamin II. Blood. (1990) 75:251–4. doi: 10.1182/blood.V75.1.251.251

  • CrossRef
  • Google Scholar

• 156

  QuadrosEVRothenbergSPJaffeEA. Endothelial cells from human umbilical vein secrete functional transcobalamin II. Am J Physiol. (1989) 256:C296–303. doi: 10.1152/ajpcell.1989.256.2.C296

  • CrossRef
  • Google Scholar

• 157

  Johansson-PercivalAGanssR. Therapeutic induction of tertiary lymphoid structures in cancer through stromal remodeling. Front Immunol. (2021) 12:674375. doi: 10.3389/fimmu.2021.674375

  • CrossRef
  • Google Scholar

• 158

  Miyagawa-HayashinoAYoshifujiHKitagoriKItoSOkuTHirayamaYet al. Increase of MZB1 in B cells in systemic lupus erythematosus: proteomic analysis of biopsied lymph nodes. Arthritis Res Ther. (2018) 20:13. doi: 10.1186/s13075-018-1511-5

  • CrossRef
  • Google Scholar

• 159

  SiuJHYPitcherMJTullTJVelouniasRLGuesdonWMontorsiLet al. Two subsets of human marginal zone B cells resolved by global analysis of lymphoid tissues and blood. Sci Immunol. (2022) 7:eabm9060. doi: 10.1126/sciimmunol.abm9060

  • CrossRef
  • Google Scholar

• 160

  MeylanMPetitprezFBechtEBougoüinAPupierGCalvezAet al. Tertiary lymphoid structures generate and propagate anti-tumor antibody-producing plasma cells in renal cell cancer. Immunity. (2022) 55:527–41.e5. doi: 10.1016/j.immuni.2022.02.001

  • CrossRef
  • Google Scholar

• 161

  ShacharIHaranM. The secret second life of an innocent chaperone: the story of CD74 and B cell/chronic lymphocytic leukemia cell survival. Leuk Lymphoma. (2011) 52:1446–54. doi: 10.3109/10428194.2011.565437

  • CrossRef
  • Google Scholar

• 162

  IvesALe RoyDThéroudeCBernhagenJRogerTCalandraT. Macrophage migration inhibitory factor promotes the migration of dendritic cells through CD74 and the activation of the Src/PI3K/myosin II pathway. FASEB J. (2021) 35:e21418. doi: 10.1096/fj.202001605R

  • CrossRef
  • Google Scholar

• 163

  KlasenCOhlKSternkopfMShacharISchmitzCHeussenNet al. MIF promotes B cell chemotaxis through the receptors CXCR4 and CD74 and ZAP-70 signaling. J Immunol. (2014) 192:5273–84. doi: 10.4049/jimmunol.1302209

  • CrossRef
  • Google Scholar

• 164

  SunHFangTWangTYuZGongLWeiXet al. Single-cell profiles reveal tumor cell heterogeneity and immunosuppressive microenvironment in Waldenström macroglobulinemia. J Transl Med. (2022) 20:576. doi: 10.1186/s12967-022-03798-6

  • CrossRef
  • Google Scholar

• 165

  KongFYeSZhongZZhouXZhouWLiuZet al. Single-cell transcriptome analysis of chronic antibody-mediated rejection after renal transplantation. Front Immunol. (2021) 12:767618. doi: 10.3389/fimmu.2021.767618

  • CrossRef
  • Google Scholar

• 166

  LiuJGaoHLiCZhuFWangMXuYet al. Expression and regulatory characteristics of peripheral blood immune cells in primary Sjögren's syndrome patients using single-cell transcriptomic. iScience. (2022) 25:105509. doi: 10.1016/j.isci.2022.105509

  • CrossRef
  • Google Scholar

• 167

  MackayFBrowningJL. BAFF: a fundamental survival factor for B cells. Nat Rev Immunol. (2002) 2:465–75. doi: 10.1038/nri844

  • CrossRef
  • Google Scholar

• 168

  ZhaoSZhangHXingYNatkunamY. CD137 ligand is expressed in primary and secondary lymphoid follicles and in B-cell lymphomas: diagnostic and therapeutic implications. Am J Surg Pathol. (2013) 37:250–8. doi: 10.1097/PAS.0b013e318268c6ea

  • CrossRef
  • Google Scholar

• 169

  KendalJKShehataMSLofftusSYCromptonJG. Cancer-associated B cells in sarcoma. Cancers (Basel). (2023) 15. doi: 10.3390/cancers15030622

  • CrossRef
  • Google Scholar

• 170

  MeyerLMMillerIFGizisETrippEHoffbrandAV. Delivery of vitamin B12 to human lymphocytes by transcobalamins I, II and 3. Proc Soc Exp Biol Med. (1974) 146:747–50. doi: 10.3181/00379727-146-38185

  • CrossRef
  • Google Scholar

• 171

  WaibelRTreichlerHSchaeferNGvan StaverenDRMundwilerSKunzeSet al. New derivatives of vitamin B12 show preferential targeting of tumors. Cancer Res. (2008) 68:2904–11. doi: 10.1158/0008-5472.CAN-07-6771

  • CrossRef
  • Google Scholar

• 172

  HallCAGreenPD. Competition between apo and holo transcobalamin II (TC-II) for the TC-II mediated uptake process. Proc Soc Exp Biol Med. (1978) 158:206–9. doi: 10.3181/00379727-158-40172

  • CrossRef
  • Google Scholar

• 173

  BirnHWillnowTENielsenRNordenAGBönschCMoestrupSKet al. Megalin is essential for renal proximal tubule reabsorption and accumulation of transcobalamin-B(12). Am J Physiol Renal Physiol. (2002) 282:F408–16. doi: 10.1152/ajprenal.00206.2000

  • CrossRef
  • Google Scholar

• 174

  SavageAKGutschowMVChiangTHendersonKGreenRChaudhariMet al. Multimodal analysis for human. iScience. (2021) 24:102404. doi: 10.1016/j.isci.2021.102404

  • CrossRef
  • Google Scholar