Recurrence of Papillary Thyroid Cancer: A Systematic Appraisal of Risk Factors

Background Thyroid cancer recurrence is associated with increased mortality and adverse outcomes. Recurrence risk is currently predicted using clinical tools, often restaging patients after treatment. Detailed understanding of recurrence risk at disease-onset could lead to personalised and improved patient care. Objective To perform a comprehensive bioinformatic and experimental analysis of 3 levels of genetic change (mRNA, microRNA, and somatic mutation) apparent in recurrent tumours and construct a new combinatorial prognostic risk model. Methods We analysed The Cancer Genome Atlas data (TCGA) to identify differentially expressed genes (mRNA/microRNA) in 46 recurrent versus 455 non-recurrent thyroid tumours. Two exonic mutational pipelines were used to identify somatic mutations. Functional gene analysis was performed in cell-based assays in multiple thyroid cell lines. The prognostic value of genes was evaluated with TCGA datasets. Results We identified a total of 128 new potential biomarkers associated with recurrence, including 40 mRNAs, 39 miRNAs and 59 genetic variants. Among differentially expressed genes, modulation of FN1, ITGα3 and MET had a significant impact on thyroid cancer cell migration. Similarly, ablation of miR-486 and miR-1179 significantly increased migration of TPC-1 and SW1736 cells. We further utilised genes with a validated functional role and identified a 5 gene risk score classifier as an independent predictor of thyroid cancer recurrence. Conclusions Our newly proposed risk model based on combinatorial mRNA and microRNA expression has potential clinical utility as a prognostic indicator of recurrence. These findings should facilitate earlier prediction of recurrence with implications for improving patient outcome by tailoring treatment to disease risk and increasing post-treatment surveillance.

Effectiveness of Different Thyroid Stimulating Hormone Thresholds for Thyroid Hormone Withdrawal - Aided Radioiodine Ablation in Patients With Differentiated Thyroid Cancer: A Systematic Review and Meta - Analysis

Background: Patients with differentiated thyroid cancer (DTC) may benefit of radioiodine ablation (RAI) to reduce the probability of thyroid cancer recurrence. Guidelines recommend that thyroid stimulating hormone (TSH) of >30mIU/L before RAI to optimize treatment response. However, evidence regarding this recommendation is conflicting. We conducted a systematic review and meta-analysis to compare outcomes (thyroid cancer recurrence and survival at 10 years of follow up) of stimulated TSH threshold before RAI with primary analysis focus on <30 mIU/L versus ≥30 mIU/L, and subgroup analysis on <90 mIU/L versus ≥90 mIU/L in patients with DTC after initial total thyroidectomy. Methods: The protocol for this study is registered and available online (CRD42020158354). Briefly, we searched several databases from their inception to April 2020. Reviewers, working independently and in duplicate selected studies for inclusion, extracted data, and evaluated each study’s risk of bias. We excluded studies that used recombinant human thyroid stimulating hormone before ablation. Results: We included five retrospective cohort studies, which enrolled a total of 2,514 patients. Risk of bias was low in four studies and high in one study. Mean age was 47 years old (ranged from 40.7 to 47.9) and most of them were female (69%). The most common DTC type was papillary thyroid cancer (78%). From those articles that reported tumor characteristics, 48% had a size ≤2cm (T1b) and 47% >2cm. Moreover, 73% of the patients had no regional lymph metastasis (N0). Two studies reported radioiodine mean dose given of 30 and 100 mci. A total of 301 patients were included in the TSH threshold <30 mIU/L group and 1788 patients in the TSH ≥30 mIU/L group. Comparing stimulated TSH threshold before RAI (<30 mIU/L versus ≥30 mIU/L), there was difference in recurrence at 1 year (RR 2.46 (C.I. 1.09-5.55) and at 20 years (RR 1.71 (C.I. 1.19 – 2.47). However, there was no difference in mortality at 20 years (RR 0.53 (Confidence Interval (C.I.) 0.12-2.23). In addition, 10-years recurrence was not different when we compared <90 mIU/L versus ≥90 mIU/L TSH (RR 1.06; 95%CI: 0.88 – 1.27). Conclusions: Mortality do not differ between recommended TSH goal (≥30 mIU/L) vs <30 mIU/L in thyroid hormone withdrawal-aided radioiodine ablation in DTC patients. However, the risk of recurrence is reduced when patients achieved a TSH level >30 UI/mL. These results suggest that patients may need to reach a stimulated TSH ≥30 mIU/L stimulated TSH threshold to be treated. Randomized trials are needed to confirm these findings.

A Framework (SOCRATex) for Hierarchical Annotation of Unstructured Electronic Health Records and Integration Into a Standardized Medical Database: Development and Usability Study

Background Although electronic health records (EHRs) have been widely used in secondary assessments, clinical documents are relatively less utilized owing to the lack of standardized clinical text frameworks across different institutions. Objective This study aimed to develop a framework for processing unstructured clinical documents of EHRs and integration with standardized structured data. Methods We developed a framework known as Staged Optimization of Curation, Regularization, and Annotation of clinical text (SOCRATex). SOCRATex has the following four aspects: (1) extracting clinical notes for the target population and preprocessing the data, (2) defining the annotation schema with a hierarchical structure, (3) performing document-level hierarchical annotation using the annotation schema, and (4) indexing annotations for a search engine system. To test the usability of the proposed framework, proof-of-concept studies were performed on EHRs. We defined three distinctive patient groups and extracted their clinical documents (ie, pathology reports, radiology reports, and admission notes). The documents were annotated and integrated into the Observational Medical Outcomes Partnership (OMOP)-common data model (CDM) database. The annotations were used for creating Cox proportional hazard models with different settings of clinical analyses to measure (1) all-cause mortality, (2) thyroid cancer recurrence, and (3) 30-day hospital readmission. Results Overall, 1055 clinical documents of 953 patients were extracted and annotated using the defined annotation schemas. The generated annotations were indexed into an unstructured textual data repository. Using the annotations of pathology reports, we identified that node metastasis and lymphovascular tumor invasion were associated with all-cause mortality among colon and rectum cancer patients (both P=.02). The other analyses involving measuring thyroid cancer recurrence using radiology reports and 30-day hospital readmission using admission notes in depressive disorder patients also showed results consistent with previous findings. Conclusions We propose a framework for hierarchical annotation of textual data and integration into a standardized OMOP-CDM medical database. The proof-of-concept studies demonstrated that our framework can effectively process and integrate diverse clinical documents with standardized structured data for clinical research.

A Framework (SOCRATex) for Hierarchical Annotation of Unstructured Electronic Health Records and Integration Into a Standardized Medical Database: Development and Usability Study (Preprint)

BACKGROUND Although electronic health records (EHRs) have been widely used in secondary assessments, clinical documents are relatively less utilized owing to the lack of standardized clinical text frameworks across different institutions. OBJECTIVE This study aimed to develop a framework for processing unstructured clinical documents of EHRs and integration with standardized structured data. METHODS We developed a framework known as Staged Optimization of Curation, Regularization, and Annotation of clinical text (SOCRATex). SOCRATex has the following four aspects: (1) extracting clinical notes for the target population and preprocessing the data, (2) defining the annotation schema with a hierarchical structure, (3) performing document-level hierarchical annotation using the annotation schema, and (4) indexing annotations for a search engine system. To test the usability of the proposed framework, proof-of-concept studies were performed on EHRs. We defined three distinctive patient groups and extracted their clinical documents (ie, pathology reports, radiology reports, and admission notes). The documents were annotated and integrated into the Observational Medical Outcomes Partnership (OMOP)-common data model (CDM) database. The annotations were used for creating Cox proportional hazard models with different settings of clinical analyses to measure (1) all-cause mortality, (2) thyroid cancer recurrence, and (3) 30-day hospital readmission. RESULTS Overall, 1055 clinical documents of 953 patients were extracted and annotated using the defined annotation schemas. The generated annotations were indexed into an unstructured textual data repository. Using the annotations of pathology reports, we identified that node metastasis and lymphovascular tumor invasion were associated with all-cause mortality among colon and rectum cancer patients (both P=.02). The other analyses involving measuring thyroid cancer recurrence using radiology reports and 30-day hospital readmission using admission notes in depressive disorder patients also showed results consistent with previous findings. CONCLUSIONS We propose a framework for hierarchical annotation of textual data and integration into a standardized OMOP-CDM medical database. The proof-of-concept studies demonstrated that our framework can effectively process and integrate diverse clinical documents with standardized structured data for clinical research.