Stratification of follicular thyroid tumors using data-independent acquisition proteomics and a comprehensive thyroid tissue spectral library

Proteomics profile in encapsulated follicular patterned thyroid neoplasms

Diagnosing encapsulated follicular-patterned thyroid tumors like Invasive Encapsulated Follicular Variant of Papillary Thyroid Carcinoma (IEFVPTC), Non-invasive Follicular Thyroid Neoplasm with Papillary-like Nuclear Features (NIFTP), and Well-Differentiated Tumor of Uncertain Malignant Potential (WDT-UMP) remains challenging due to their morphological and molecular similarities. This study aimed to investigate the protein distinctions among these three thyroid tumors and discover biological tumorigenesis through proteomic analysis. We employed total shotgun proteome analysis allowing to discover the quantitative expression of over 1398 proteins from 12 normal thyroid tissues, 13 IEFVPTC, 11 NIFTP, and 10 WDT-UMP. Principal component analysis revealed a distinct separation of IEFVPTC and normal tissue samples, distinguishing them from the low-risk tumor group (NIFTP and WDT-UMP). IEFVPTC exhibited the highest number of differentially expressed proteins (DEPs) compared to the other tumors. No discriminatory proteins between NIFTP and WDT-UMP were identified. Moreover, DEPs in IEFVPTC were significantly associated with thyroid tumor progression pathways. Certain hub genes linked to the response of immune checkpoint inhibitor therapy, revealing the potential predictor of prognosis. In conclusion, the proteomic profile of IEFVPTC differs from that of low-risk tumors. These findings may provide valuable insights into tumor biology and offer a basis for developing novel therapeutic strategies for follicular-patterned thyroid neoplasms.

[Applications of high performance liquid chromatography-mass spectrometry in proteomics]

Proteomics profiling plays an important role in biomedical studies. Proteomics studies are much more complicated than genome research, mainly because of the complexity and diversity of proteomic samples. High performance liquid chromatography-mass spectrometry (HPLC-MS) is a fundamental tool in proteomics research owing to its high speed, resolution, and sensitivity. Proteomics research targets from the peptides and individual proteins to larger protein complexes, the molecular weight of which gradually increases, leading to sustained increases in structural and compositional complexity and alterations in molecular properties. Therefore, the selection of various separation strategies and stationary-phase parameters is crucial when dealing with the different targets in proteomics research for in-depth proteomics analysis. This article provides an overview of commonly used chromatographic-separation strategies in the laboratory, including reversed-phase liquid chromatography (RPLC), hydrophilic interaction liquid chromatography (HILIC), hydrophobic interaction chromatography (HIC), ion-exchange chromatography (IEC), and size-exclusion chromatography (SEC), as well as their applications and selectivity in the context of various biomacromolecules. At present, no single chromatographic or electrophoretic technology features the peak capacity required to resolve such complex mixtures into individual components. Multidimensional liquid chromatography (MDLC), which combines different orthogonal separation modes with MS, plays an important role in proteomics research. In the MDLC strategy, IEC, together with RPLC, remains the most widely used separation mode in proteomics analysis; other chromatographic methods are also frequently used for peptide/protein fractionation. MDLC technologies and their applications in a variety of proteomics analyses have undergone great development. Two strategies in MDLC separation systems are mainly used in proteomics profiling: the "bottom-up" approach and the "top-down" approach. The "shotgun" method is a typical "bottom-up" strategy that is based on the RPLC or MDLC separation of whole-protein-sample digests coupled with MS; it is an excellent technique for identifying a large number of proteins. "Top-down" analysis is based on the separation of intact proteins and provides their detailed molecular information; thus, this technique may be advantageous for analyzing the post-translational modifications (PTMs) of proteins. In this paper, the "bottom-up" "top-down" and protein-protein interaction (PPI) analyses of proteome samples are briefly reviewed. The diverse combinations of different chromatographic modes used to set up MDLC systems are described, and compatibility issues between mobile phases and analytes, between mobile phases and MS, and between mobile phases in different separation modes in multidimensional chromatography are analyzed. Novel developments in MDLC techniques, such as high-abundance protein depletion and chromatography arrays, are further discussed. In this review, the solutions proposed by researchers when encountering compatibility issues are emphasized. Moreover, the applications of HPLC-MS combined with various sample pretreatment methods in the study of exosomal and single-cell proteomics are examined. During exosome isolation, the combined use of ultracentrifugation and SEC can yield exosomes of higher purity. The use of SEC with ultra-large-pore-size packing materials (200 nm) enables the isolation of exosomal subgroups, and proteomics studies have revealed significant differences in protein composition and function between these subgroups. In the field of single-cell proteomics, researchers have addressed challenges related to reducing sample processing volumes, preventing sample loss, and avoiding contamination during sample preparation. Innovative methods and improvements, such as the utilization of capillaries for sample processing and microchips as platforms to minimize the contact area of the droplets, have been proposed. The integration of these techniques with HPLC-MS shows some progress. In summary, this article focuses on the recent advances in HPLC-MS technology for proteomics analysis and provides a comprehensive reference for future research in the field of proteomics.

An individualized protein-based prognostic model to stratify pediatric patients with papillary thyroid carcinoma

Pediatric papillary thyroid carcinomas (PPTCs) exhibit high inter-tumor heterogeneity and currently lack widely adopted recurrence risk stratification criteria. Hence, we propose a machine learning-based objective method to individually predict their recurrence risk. We retrospectively collect and evaluate the clinical factors and proteomes of 83 pediatric benign (PB), 85 pediatric malignant (PM) and 66 adult malignant (AM) nodules, and quantify 10,426 proteins by mass spectrometry. We find 243 and 121 significantly dysregulated proteins from PM vs. PB and PM vs. AM, respectively. Function and pathway analyses show the enhanced activation of the inflammatory and immune system in PM patients compared with the others. Nineteen proteins are selected to predict recurrence using a machine learning model with an accuracy of 88.24%. Our study generates a protein-based personalized prognostic prediction model that can stratify PPTC patients into high- or low-recurrence risk groups, providing a reference for clinical decision-making and individualized treatment.

Risk stratification of papillary thyroid cancers using multidimensional machine learning

BACKGROUND

Papillary thyroid cancer (PTC) is one of the most common endocrine malignancies with different risk levels. However, preoperative risk assessment of PTC is still a challenge in the worldwide. Here, the authors first report a Preoperative Risk Assessment Classifier for PTC (PRAC-PTC) by multidimensional features including clinical indicators, immune indices, genetic feature, and proteomics.

MATERIALS AND METHODS

The 558 patients collected from June 2013 to November 2020 were allocated to three groups: the discovery set [274 patients, 274 formalin-fixed paraffin-embedded (FFPE)], the retrospective test set (166 patients, 166 FFPE), and the prospective test set (118 patients, 118 fine-needle aspiration). Proteomic profiling was conducted by FFPE and fine-needle aspiration tissues from the patients. Preoperative clinical information and blood immunological indices were collected. The BRAFV600E mutation were detected by the amplification refractory mutation system.

RESULTS

The authors developed a machine learning model of 17 variables based on the multidimensional features of 274 PTC patients from a retrospective cohort. The PRAC-PTC achieved areas under the curve (AUC) of 0.925 in the discovery set and was validated externally by blinded analyses in a retrospective cohort of 166 PTC patients (0.787 AUC) and a prospective cohort of 118 PTC patients (0.799 AUC) from two independent clinical centres. Meanwhile, the preoperative predictive risk effectiveness of clinicians was improved with the assistance of PRAC-PTC, and the accuracies reached at 84.4% (95% CI: 82.9-84.4) and 83.5% (95% CI: 82.2-84.2) in the retrospective and prospective test sets, respectively.

CONCLUSION

This study demonstrated that the PRAC-PTC that integrating clinical data, gene mutation information, immune indices, high-throughput proteomics and machine learning technology in multicentre retrospective and prospective clinical cohorts can effectively stratify the preoperative risk of PTC and may decrease unnecessary surgery or overtreatment.

A Proteomic Analysis of Human Follicular Fluid: Proteomic Profile Associated with Embryo Quality

Embryo selection is a key point of in vitro fertilization (IVF). The most commonly used method for embryo selection is morphological assessment. However, it is sometimes inaccurate. Follicular fluid (FF) contains a complex mixture of proteins that are essential for follicle development and oocyte maturation. Analyzing human FF proteomic profiles and identifying predictive biomarkers might be helpful for evaluating embryo quality. A total of 22 human FF samples were collected from 19 infertile women who underwent IVF/intracytoplasmic sperm injection (ICSI) treatment between October 2021 and November 2021. FFs were grouped into two categories on the basis of the day 3 embryo quality, grade I or II in the hqFF group and grade III in the nhqFF group. FF was analyzed by liquid chromatography-tandem mass spectrometry (LC/MS/MS). The key differentially expressed proteins (DEPs) were validated by parallel reaction monitoring (PRM) and enzyme-linked immunosorbent assay (ELISA). Differentially expressed proteins were further analyzed using DAVID software. A total of 558 proteins were identified, of which 50 proteins were differentially expressed in the hqFF versus nhqFF group, including 32 upregulated proteins (> 1.20-fold, P < 0.05) and 18 downregulated proteins (< 0.67-fold, P < 0.05). Bioinformatics analyses showed that the upregulated DEPs were enriched in components of the coagulation and complement systems and negative regulation of peptidase activity, while the downregulated DEPs were enriched in molecular function of extracellular matrix structural and constituent collagen binding. Our results suggested that a number of protein biomarkers in FF were associated with embryo quality. It may help develop an effective and noninvasive method for embryo selection.

Automated diagnosis and management of follicular thyroid nodules based on the devised small-dataset interpretable foreground optimization network deep learning: a multicenter diagnostic study

BACKGROUND

Currently, follicular thyroid carcinoma (FTC) has a relatively low incidence with a lack of effective preoperative diagnostic means. To reduce the need for invasive diagnostic procedures and to address information deficiencies inherent in a small dataset, we utilized interpretable foreground optimization network deep learning to develop a reliable preoperative FTC detection system.

METHODS

In this study, a deep learning model (FThyNet) was established using preoperative ultrasound images. Data on patients in the training and internal validation cohort ( n =432) were obtained from Ruijin Hospital, China. Data on patients in the external validation cohort ( n =71) were obtained from four other clinical centers. We evaluated the predictive performance of FThyNet and its ability to generalize across multiple external centers and compared the results yielded with assessments from physicians directly predicting FTC outcomes. In addition, the influence of texture information around the nodule edge on the prediction results was evaluated.

RESULTS

FThyNet had a consistently high accuracy in predicting FTC with an area under the receiver operating characteristic curve (AUC) of 89.0% [95% CI 87.0-90.9]. Particularly, the AUC for grossly invasive FTC reached 90.3%, which was significantly higher than that of the radiologists (56.1% [95% CI 51.8-60.3]). The parametric visualization study found that those nodules with blurred edges and relatively distorted surrounding textures were more likely to have FTC. Furthermore, edge texture information played an important role in FTC prediction with an AUC of 68.3% [95% CI 61.5-75.5], and highly invasive malignancies had the highest texture complexity.

CONCLUSION

FThyNet could effectively predict FTC, provide explanations consistent with pathological knowledge, and improve clinical understanding of the disease.

Identifying Patients With Rapid Progression From Hormone-Sensitive to Castration-Resistant Prostate Cancer: A Retrospective Study

Prostate cancer (PCa) is the second most prevalent malignancy and the fifth cause of cancer-related deaths in men. A crucial challenge is identifying the population at risk of rapid progression from hormone-sensitive prostate cancer (HSPC) to lethal castration-resistant prostate cancer (CRPC). We collected 78 HSPC biopsies and measured their proteomes using pressure cycling technology and a pulsed data-independent acquisition pipeline. We quantified 7355 proteins using these HSPC biopsies. A total of 251 proteins showed differential expression between patients with a long- or short-term progression to CRPC. Using a random forest model, we identified seven proteins that significantly discriminated long- from short-term progression patients, which were used to classify PCa patients with an area under the curve of 0.873. Next, one clinical feature (Gleason sum) and two proteins (BGN and MAPK11) were found to be significantly associated with rapid disease progression. A nomogram model using these three features was generated for stratifying patients into groups with significant progression differences (p-value = 1.3×10-4). To conclude, we identified proteins associated with a fast progression to CRPC and an unfavorable prognosis. Based on these proteins, our machine learning and nomogram models stratified HSPC into high- and low-risk groups and predicted their prognoses. These models may aid clinicians in predicting the progression of patients, guiding individualized clinical management and decisions.

Paving the path toward multi-omics approaches in the diagnostic challenges faced in thyroid pathology

INTRODUCTION

Despite advancements in diagnostic methods, the classification of indeterminate thyroid nodules still poses diagnostic challenges not only in pre-surgical evaluation but even after histological evaluation of surgical specimens. Proteomics, aided by mass spectrometry and integrated with artificial intelligence and machine learning algorithms, shows great promise in identifying diagnostic markers for thyroid lesions.

AREAS COVERED

This review provides in-depth exploration of how proteomics has contributed to the understanding of thyroid pathology. It discusses the technical advancements related to immunohistochemistry, genetic and proteomic techniques, such as mass spectrometry, which have greatly improved sensitivity and spatial resolution up to single-cell level. These improvements allowed the identification of specific protein signatures associated with different types of thyroid lesions.

EXPERT COMMENTARY

Among all the proteomics approaches, spatial proteomics stands out due to its unique ability to capture the spatial context of proteins in both cytological and tissue thyroid samples. The integration of multi-layers of molecular information combining spatial proteomics, genomics, immunohistochemistry or metabolomics and the implementation of artificial intelligence and machine learning approaches, represent hugely promising steps forward toward the possibility to uncover intricate relationships and interactions among various molecular components, providing a complete picture of the biological landscape whilst fostering thyroid nodule diagnosis.

Multi-Omics and Management of Follicular Carcinoma of the Thyroid

Follicular thyroid carcinoma (FTC) is the second most common cancer of the thyroid gland, accounting for up to 20% of all primary malignant tumors in iodine-replete areas. The diagnostic work-up, staging, risk stratification, management, and follow-up strategies in patients who have FTC are modeled after those of papillary thyroid carcinoma (PTC), even though FTC is more aggressive. FTC has a greater propensity for haematogenous metastasis than PTC. Furthermore, FTC is a phenotypically and genotypically heterogeneous disease. The diagnosis and identification of markers of an aggressive FTC depend on the expertise and thoroughness of pathologists during histopathological analysis. An untreated or metastatic FTC is likely to de-differentiate and become poorly differentiated or undifferentiated and resistant to standard treatment. While thyroid lobectomy is adequate for the treatment of selected patients who have low-risk FTC, it is not advisable for patients whose tumor is larger than 4 cm in diameter or has extensive extra-thyroidal extension. Lobectomy is also not adequate for tumors that have aggressive mutations. Although the prognosis for over 80% of PTC and FTC is good, nearly 20% of the tumors behave aggressively. The introduction of radiomics, pathomics, genomics, transcriptomics, metabolomics, and liquid biopsy have led to improvements in the understanding of tumorigenesis, progression, treatment response, and prognostication of thyroid cancer. The article reviews the challenges that are encountered during the diagnostic work-up, staging, risk stratification, management, and follow-up of patients who have FTC. How the application of multi-omics can strengthen decision-making during the management of follicular carcinoma is also discussed.

Curcumin Induces Ferroptosis in Follicular Thyroid Cancer by Upregulating HO-1 Expression

Follicular thyroid cancer (FTC) is a highly aggressive type of endocrine malignancy. It is necessary to investigate the mechanisms of tumorigenesis and therapeutic pathways in patients with FTC. Haem oxygenase-1 (HO-1) can regulate oxidative stress and the occurrence of tumors and diseases. In this study, we discovered that HO-1 was abnormally overexpressed in FTC compared with adjacent tissues. However, the HO-1 overexpression was demonstrated to decrease cell viability and to potentially activate the ferroptosis signalling pathway. Ferroptosis is a newly identified form of oxidative cell death and is currently being targeted as a new cancer treatment. Tumorigenesis is significantly inhibited by curcumin. The present study shows that curcumin inhibits the growth of FTC by increasing the HO-1 expression, further activating the ferroptosis pathway. This study demonstrates that the HO-1-ferroptosis signalling pathway might play an important role in FTC tumorigenesis, and that curcumin inhibits the growth of FTC cells by affecting this pathway.

The Plasma DIA-Based Quantitative Proteomics Reveals the Pathogenic Pathways and New Biomarkers in Cervical Cancer and High Grade Squamous Intraepithelial Lesion

OBJECTIVE

The process of normal cervix changing into high grade squamous intraepithelial lesion (HSIL) and invasive cervical cancer is long and the mechanisms are still not completely clear. This study aimed to reveal the protein profiles related to HSIL and cervical cancer and find the diagnostic and prognostic molecular changes.

METHODS

Data-independent acquisition (DIA) analysis was performed to identify 20 healthy female volunteers, 20 HSIL and 20 cervical patients in a cohort to screen differentially expressed proteins (DEPs) for the HSIL and cervical cancer. Subsequently, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were used for functional annotation of DEPs; the protein-protein interaction (PPI) and weighted gene co-expression network analysis (WGCNA) were performed for detection of key molecular modules and hub proteins. They were validated using the Enzyme-Linked Immunosorbent Assay (ELISA).

RESULTS

A total of 243 DEPs were identified in the study groups. GO and KEGG analysis showed that DEPs were mainly enriched in the complement and coagulation pathway, cholesterol metabolism pathway, the IL-17 signaling pathway as well as the viral protein interaction with cytokine and cytokine receptor pathway. Subsequently, the WGCNA analysis showed that the green module was highly correlated with the cervical cancer stage. Additionally, six interesting core DEPs were verified by ELISA, APOF and ORM1, showing nearly the same expression pattern with DIA. The area under the curve (AUC) of 0.978 was obtained by using ORM1 combined with APOF to predict CK and HSIL+CC, and in the diagnosis of HSIL and CC, the AUC can reach to 0.982. The high expression of ORM1 is related to lymph node metastasis and the clinical stage of cervical cancer patients as well as the poor prognosis.

CONCLUSION

DIA-ELSIA combined analysis screened and validated two previously unexplored but potentially useful biomarkers for early diagnosis of HSIL and cervical cancer, as well as possible new pathogenic pathways and therapeutic targets.

Artificial intelligence defines protein-based classification of thyroid nodules

Determination of malignancy in thyroid nodules remains a major diagnostic challenge. Here we report the feasibility and clinical utility of developing an AI-defined protein-based biomarker panel for diagnostic classification of thyroid nodules: based initially on formalin-fixed paraffin-embedded (FFPE), and further refined for fine-needle aspiration (FNA) tissue specimens of minute amounts which pose technical challenges for other methods. We first developed a neural network model of 19 protein biomarkers based on the proteomes of 1724 FFPE thyroid tissue samples from a retrospective cohort. This classifier achieved over 91% accuracy in the discovery set for classifying malignant thyroid nodules. The classifier was externally validated by blinded analyses in a retrospective cohort of 288 nodules (89% accuracy; FFPE) and a prospective cohort of 294 FNA biopsies (85% accuracy) from twelve independent clinical centers. This study shows that integrating high-throughput proteomics and AI technology in multi-center retrospective and prospective clinical cohorts facilitates precise disease diagnosis which is otherwise difficult to achieve by other methods.

Stratification of diabetic kidney diseases via data-independent acquisition proteomics-based analysis of human kidney tissue specimens

AIM

The aims of this study were to analyze the proteomic differences in renal tissues from patients with diabetes mellitus (DM) and diabetic kidney disease (DKD) and to select sensitive biomarkers for early identification of DKD progression.

METHODS

Pressure cycling technology-pulse data-independent acquisition mass spectrometry was employed to investigate protein alterations in 36 formalin-fixed paraffin-embedded specimens. Then, bioinformatics analysis was performed to identify important signaling pathways and key molecules. Finally, the target proteins were validated in 60 blood and 30 urine samples.

RESULTS

A total of 52 up- and 311 down-regulated differential proteins were identified as differing among the advanced DKD samples, early DKD samples, and DM controls (adjusted p<0.05). These differentially expressed proteins were mainly involved in ion transport, apoptosis regulation, and the inflammatory response. UniProt database analysis showed that these proteins were mostly enriched in signaling pathways related to metabolism, apoptosis, and inflammation. NBR1 was significantly up-regulated in both early and advanced DKD, with fold changes (FCs) of 175 and 184, respectively (both p<0.01). In addition, VPS37A and ATG4B were significantly down-regulated with DKD progression, with FCs of 0.140 and 0.088, respectively, in advanced DKD and 0.533 and 0.192, respectively, in early DKD compared with the DM control group (both p<0.01). Bioinformatics analysis showed that NBR1, VPS37A, and ATG4B are closely related to autophagy. We also found that serum levels of the three proteins and urine levels of NBR1 decreased with disease progression. Moreover, there was a significant difference in serum VPS37A and ATG4B levels between patients with early and advanced DKD (both p<0.05). The immunohistochemistry assaay exhibited that the three proteins were expressed in renal tubular cells, and NBR1 was also expressed in the cystic wall of renal glomeruli.

CONCLUSION

The increase in NBR1 expression and the decrease in ATG4B and VPS37 expression in renal tissue are closely related to inhibition of the autophagy pathway, which may contribute to DKD development or progression. These three proteins may serve as sensitive serum biomarkers for early identification of DKD progression.