1
|
Borzooei S, Briganti G, Golparian M, Lechien JR, Tarokhian A. Machine learning for risk stratification of thyroid cancer patients: a 15-year cohort study. Eur Arch Otorhinolaryngol 2024; 281:2095-2104. [PMID: 37902840 DOI: 10.1007/s00405-023-08299-w] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/21/2023] [Accepted: 10/17/2023] [Indexed: 11/01/2023]
Abstract
PURPOSE The objective of this study was to train machine learning models for predicting the likelihood of recurrence in patients diagnosed with well-differentiated thyroid cancer. While thyroid cancer mortality remains low, the risk of recurrence is a significant concern. Identifying individual patient recurrence risk is crucial for guiding subsequent management and follow-ups. METHODS In this prospective study, a cohort of 383 patients was observed for a minimum duration of 10 years within a 15-year timeframe. Thirteen clinicopathologic features were assessed to predict recurrence potential. Classic (K-nearest neighbors, support vector machines (SVM), tree-based models) and artificial neural networks (ANN) were trained on three distinct combinations of features: a data set with all features excluding American Thyroid Association (ATA) risk score (12 features), another with ATA risk alone, and a third with all features combined (13 features). 283 patients were allocated for the training process, and 100 patients were reserved for the validation of stage. RESULTS The patients' mean age was 40.87 ± 15.13 years, with a majority being female (81%). When using the full data set for training, the models showed the following sensitivity, specificity and AUC, respectively: SVM (99.33%, 97.14%, 99.71), K-nearest neighbors (83%, 97.14%, 98.44), Decision Tree (87%, 100%, 99.35), Random Forest (99.66%, 94.28%, 99.38), ANN (96.6%, 95.71%, 99.64). Eliminating ATA risk data increased models specificity but decreased sensitivity. Conversely, training exclusively on ATA risk data had the opposite effect. CONCLUSIONS Machine learning models, including classical and neural networks, efficiently stratify the risk of recurrence in patients with well-differentiated thyroid cancer. This can aid in tailoring treatment intensity and determining appropriate follow-up intervals.
Collapse
Affiliation(s)
- Shiva Borzooei
- Department of Endocrinology, Faculty of Medicine, Hamadan University of Medical Sciences, Hamadan, Iran
| | - Giovanni Briganti
- Chair of AI and Digital Medicine, Faculty of Medicine, University of Mons, Mons, France
- Department of Clinical Sciences, Faculty of Medicine, Université de Liège, Liège, Belgium
| | - Mitra Golparian
- Hamadan University of Medical Sciences, Pajoohesh Blvd., Hamadan, Iran
| | - Jerome R Lechien
- Department of Otolaryngology-Head Neck Surgery, Elsan Hospital, Paris, France
| | - Aidin Tarokhian
- Hamadan University of Medical Sciences, Pajoohesh Blvd., Hamadan, Iran.
| |
Collapse
|
2
|
Lee-Saxton YJ, Egan CE, Bratton BA, Thiesmeyer JW, Greenberg JA, Marshall TE, Tumati A, Romero-Arenas M, Beninato T, Zarnegar R, Scognamiglio T, Fahey TJ, Finnerty BM. Low Mitotic Activity in Papillary Thyroid Cancer: A Marker for Aggressive Features and Recurrence. J Clin Endocrinol Metab 2024:dgae203. [PMID: 38554391 DOI: 10.1210/clinem/dgae203] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 11/09/2023] [Revised: 03/21/2024] [Accepted: 03/28/2024] [Indexed: 04/01/2024]
Abstract
CONTEXT The significance of low mitotic activity in papillary thyroid cancer (PTC) is largely undefined. OBJECTIVE We aimed to determine the behavioral landscape of PTC with low mitotic activity compared to that of no- and high-mitotic activity. METHODS A single-institution consecutive series of PTC patients from 2018-2022 was reviewed. Mitotic activity was defined as no mitoses, low (1-2 mitoses/2 mm2) or high (≥3 mitoses/2 mm2) per the World Health Organization. The 2015 American Thyroid Association risk stratification was applied to the cohort, and clinicopathologic features were compared between groups. For patients with ≥6 months follow-up, Cox regression analyses for recurrence were performed. RESULTS 640 PTCs were included - 515 (80.5%) no mitotic activity, 110 (17.2%) low mitotic activity, and 15 (2.3%) high mitotic activity. Overall, low mitotic activity exhibited rates of clinicopathologic features including vascular invasion, gross extrathyroidal extension, and lymph node metastases in between those of no- and high-mitotic activity. PTCs with low mitotic activity had higher rates of intermediate- and high-risk ATA risk stratification compared to those with no mitotic activity (p < 0.001). Low mitotic activity PTCs also had higher recurrence rates (15.5% vs. 4.5%, p < 0.001). Low mitotic activity was associated with recurrence, independent of the ATA risk stratification (HR 2.96; 95% CI 1.28-6.87, p = 0.01). CONCLUSIONS Low mitotic activity is relatively common in PTC and its behavior lies within a spectrum between no- and high-mitotic activity. Given its association with aggressive clinicopathologic features and recurrence, low mitotic activity should be considered when risk stratifying PTC patients for recurrence.
Collapse
Affiliation(s)
- Yeon J Lee-Saxton
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Caitlin E Egan
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Brenden A Bratton
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Jessica W Thiesmeyer
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Jacques A Greenberg
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Teagan E Marshall
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Abhinay Tumati
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | | | - Toni Beninato
- Department of Surgery, Rutgers-Robert Wood Johnson Medical School, Cancer Institute of New Jersey, New Brunswick, NJ
| | - Rasa Zarnegar
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Theresa Scognamiglio
- Department of Pathology, New York Presbyterian Weill Cornell Medicine, New York, NY USA
| | - Thomas J Fahey
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| | - Brendan M Finnerty
- Department of Surgery, Weill Cornell Medicine, 1300 York Avenue, New York, NY 10065
| |
Collapse
|
3
|
Tran A, Weigel RJ, Beck AC. ATA risk stratification in papillary thyroid microcarcinoma has low positive predictive value when identifying recurrence. Am J Surg 2024; 229:106-110. [PMID: 37968147 DOI: 10.1016/j.amjsurg.2023.11.003] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/27/2023] [Revised: 09/25/2023] [Accepted: 11/01/2023] [Indexed: 11/17/2023]
Abstract
BACKGROUND Rising incidence of papillary thyroid microcarcinomas (PTMC) has raised concerns for overdiagnosis. Utility of the American Thyroid Association Risk Stratification System (ATA-RSS) 2015 in predicting risk of disease recurrence in patients with PTMC was assessed. METHODS Electronic health records of patients who underwent total thyroidectomy were queried. ATA-RSS 2015 risk stratification was performed on those with PTMC, and validity for predicting disease recurrence was calculated. RESULTS With 10-year median follow up, recurrence was higher in PTMC patients with high/intermediate vs low ATA risk (33 % vs 4 %, p = 0.002). Sensitivity of ATA-RSS for detecting recurrence was 60 %, specificity 90 %, PPV 33.3 %, NPV 96.6 %, and accuracy 88 %. When microscopic extrathyroidal extension (ETE) was excluded as an intermediate risk criterion, PPV improved to 50 % and accuracy improved to 92.5 % CONCLUSIONS: ATA-RSS 2015 predicts recurrence in PTMC with high NPV but low PPV. Exclusion of microscopic ETE improved PPV, which may help prevent overtreatment.
Collapse
Affiliation(s)
- Andy Tran
- University of Iowa Carver College of Medicine, USA
| | - Ronald J Weigel
- University of Iowa Hospitals and Clinics, Department of Surgery, USA
| | - Anna C Beck
- University of Iowa Hospitals and Clinics, Department of Surgery, USA; University of Wisconsin School of Medicine and Public Health, Department of Surgery, USA.
| |
Collapse
|
4
|
Maino F, Botte M, Dalmiglio C, Valerio L, Brilli L, Trimarchi A, Mattii E, Cartocci A, Castagna MG. Prognostic Factors Improving ATA Risk System and Dynamic Risk Stratification in Low- and Intermediate-Risk DTC Patients. J Clin Endocrinol Metab 2024; 109:722-729. [PMID: 37804529 DOI: 10.1210/clinem/dgad591] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/21/2023] [Revised: 09/06/2023] [Accepted: 10/05/2023] [Indexed: 10/09/2023]
Abstract
CONTEXT American Thyroid Association (ATA) guidelines do not consider age at diagnosis as a prognostic factor on the estimation of the risk of persistent/recurrent disease in differentiated thyroid carcinoma (DTC) patients. While age at diagnosis has already been assessed in high-risk patients, it remains to be established in low- and intermediate-risk patients. OBJECTIVE The aim of our study was to investigate the role of age as a prognostic factor in the short- and long-term outcome of DTC patients classified at low and intermediate risk according to the ATA stratification risk system. METHODS We retrospectively evaluated 863 DTC patients (mean follow-up: 10 ± 6.2 years) 52% classified as low (449/863) and 48% as intermediate risk (414/863). For each ATA-risk class patients were divided into subgroups based on age at diagnosis (<55 or ≥55 years). RESULTS In the intermediate-risk group, patients aged 55 years or older had a higher rate of structural disease (11.6% vs 8.9%), recurrent disease (4.1% vs 0.7%), and death (4.1% vs 1%) when compared with younger patients (<55 years) (P = .007). Multivariate analysis confirmed that older age at diagnosis (odds ratio [OR] = 3.9; 95% CI, 1.9-8.6; P < .001) was an independent risk factor for worse long-term outcome together with response to initial therapy (OR = 13.0; 95% CI, 6.3-27.9; P < .001), and T (OR = 32; 95% CI, 1.4-7.1; P = .005) and N category (OR = 2.3; 95% CI, 1.1-5.0; P = .03). Nevertheless, a negative effect of older age was documented only in the subgroup of intermediate DTC patients with persistent structural disease after initial therapy. Indeed, the rate of worse long-term outcome rose from 13.3% in the whole population of intermediate DTC patients to 47.8% in patients with persistent structural disease after initial therapy (P < .001) and to 80% in patients older than 55 years and persistent structural disease after initial therapy (P = .02). CONCLUSION Our results suggest that age at diagnosis further predict individual outcomes in Intermediate-Risk DTC allowing ongoing management to be tailored accordingly.
Collapse
Affiliation(s)
- Fabio Maino
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| | - Monica Botte
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| | - Cristina Dalmiglio
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| | - Laura Valerio
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| | - Lucia Brilli
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| | - Andrea Trimarchi
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| | - Elisa Mattii
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| | - Alessandra Cartocci
- Department of Medical Biotechnologies, University of Siena, 53100 Siena, Italy
| | - Maria Grazia Castagna
- Department of Medical, Surgical and Neurological Sciences, University of Siena, 53100 Siena, Italy
| |
Collapse
|
5
|
Stewardson P, Eszlinger M, Wu J, Khalil M, Box A, Perizzolo M, Punjwani Z, Ziehr B, Sanyal R, Demetrick DJ, Paschke R. Prospective Validation of ThyroSPEC Molecular Testing of Indeterminate Thyroid Nodule Cytology Following Diagnostic Pathway Optimization. Thyroid 2023; 33:1423-1433. [PMID: 37742115 DOI: 10.1089/thy.2023.0255] [Citation(s) in RCA: 3] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 09/25/2023]
Abstract
Background: Molecular testing for cytologically indeterminate thyroid nodules (ITNs) is often reported with incomplete data on clinical assessment and ultrasound malignancy risk (USMR) stratification. This study aimed to clinically validate the diagnostic accuracy of a novel molecular test, assess the incremental preoperative malignancy risk of other clinical factors, and measure the impacts of introducing molecular testing at the population level. Methods: Comprehensive clinical data were collected prospectively for the first 615 consecutive patients with ITNs in a centralized health care system following implementation of a reflexive molecular test. Clinical data include patient history, method of nodule discovery, clinical assessment, USMR, cytology, molecular testing, and surgery or follow-up along with surgeon notes on surgical decision-making. Accuracy of molecular testing and the impact of the introduction of molecular testing were calculated. A multivariable regression model was developed to identify which clinical factors have the most diagnostic significance for ITNs. Results: A locally developed, low-cost molecular test achieved a negative predictive value (NPV) of 76-91% [confidence interval, CI 66-95%] and a positive predictive value (PPV) of 46-65% [CI 37-75%] in ITNs using only residual material from standard liquid cytology fine-needle aspiration (FNA). Sensitivity was highest (80%; [CI 63-92%]) in the American Thyroid Association (ATA) intermediate-suspicion ultrasound category, and lowest (46%; [CI 19-75%]) in the ATA high-suspicion ultrasound category. Following implementation of molecular testing, diagnostic yield increased by 14% (p = 0.2442) and repeat FNAs decreased by 24% (p = 0.05). Mutation was the primary reason for surgery in 76% of resected, mutation-positive patients. High-risk mutations were associated with a 58% (p = 0.0001) shorter wait for surgery. Twenty-six percent of patients with a negative molecular test result underwent surgery. Multivariable regression highlighted molecular testing and USMR as significantly associated with malignancy. Conclusions: Molecular testing improves preoperative risk stratification but requires further stratification for intermediate-risk mutations. Incorporation of clinical factors (especially USMR) with molecular testing may increase the sensitivity for detection of malignancy. Introduction of molecular testing offers some clinical benefits even in a low resection rate setting, and directly influences surgical decision-making. This study illustrates the importance of the local diagnostic pathway in ensuring appropriate integrated use of molecular testing for best outcomes.
Collapse
Affiliation(s)
- Paul Stewardson
- Arnie Charbonneau Cancer Institute, Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Markus Eszlinger
- Arnie Charbonneau Cancer Institute, Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
- Department of Oncology, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Jiahui Wu
- Arnie Charbonneau Cancer Institute, Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Moosa Khalil
- Alberta Precision Laboratories, Department of Pathology and Laboratory Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Adrian Box
- Alberta Precision Laboratories, Department of Pathology and Laboratory Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Marco Perizzolo
- Alberta Precision Laboratories, Department of Pathology and Laboratory Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Zoya Punjwani
- Arnie Charbonneau Cancer Institute, Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Bjoern Ziehr
- Arnie Charbonneau Cancer Institute, Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Ratna Sanyal
- Arnie Charbonneau Cancer Institute, Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Douglas J Demetrick
- Alberta Precision Laboratories, Department of Pathology and Laboratory Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
| | - Ralf Paschke
- Arnie Charbonneau Cancer Institute, Department of Medicine, Cumming School of Medicine, University of Calgary, Calgary, Canada
- Department of Oncology, Cumming School of Medicine, University of Calgary, Calgary, Canada
| |
Collapse
|
6
|
Mukhtar N, Alhamoudi K, Alswailem M, Alhindi H, Murugan AK, Alghamdi B, Alzahrani AS. How do BRAFV600E and TERT promoter mutations interact with the ATA and TNM staging systems in thyroid cancer? Front Endocrinol (Lausanne) 2023; 14:1270796. [PMID: 37859987 PMCID: PMC10582750 DOI: 10.3389/fendo.2023.1270796] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 08/01/2023] [Accepted: 09/14/2023] [Indexed: 10/21/2023] Open
Abstract
Context The American Thyroid Association risk stratification (ATA) and the American Joint Committee on Cancer Tumor Node Metastases (TNM) predict recurrence and mortality of differentiated thyroid cancer (DTC). BRAFV600E and TERT promoter mutations have been shown to correlate with the histopathological features and outcome of DTC. Our objectives were to study the correlation of these molecular markers with these clinicopathological-staging systems. Patients and methods We studied 296 unselected patients, 214 females and 82 males with a median age of 36 years (IQR 23.3-49.0). BRAFV600E and TERT promoter mutations were tested by PCR-based Sanger sequencing. Data were extracted from medical records and analysed using Chi-Square and Fisher Exact tests and Kaplan Meier analysis. Results Of 296 patients tested, 137 (46.3%) had BRAFV600E-positive tumors and 72 (24.3%) were positive for TERT promoter mutations. The BRAFV600E mutation did not correlate with the ATA and TNM staging, being non-significantly different in various stages of these systems and did not predict the development of persistent disease (PD) (P 0.12). Unlike BRAFV600E, TERT promoter mutations were more frequent in the ATA high-risk than in intermediate- or low-risk tumors (P 0.006) and in TNM stages III and IV than lower stages (P <0.0001). TERT promoter mutations also predicted the outcome, being present in 37.2% of patients with PD compared to only 15.4% in those without evidence of disease (P <0.0001). The same pattern was also seen when BRAFV600E and TERT promoter mutations were combined. Conclusion TERT promoter mutations alone or in combination with BRAFV600E mutation, but not BRAFV600E mutation alone, correlated well with the ATA and TNM staging and predicted development of PD, especially in higher stages of these systems.
Collapse
Affiliation(s)
- Noha Mukhtar
- Department of Medicine, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
| | - Kheloud Alhamoudi
- Department of Molecular Oncology, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
| | - Meshael Alswailem
- Department of Molecular Oncology, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
| | - Hindi Alhindi
- Department of Pathology and Laboratory Medicine, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
| | | | - Balgees Alghamdi
- Department of Molecular Oncology, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
| | - Ali S. Alzahrani
- Department of Medicine, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
- Department of Molecular Oncology, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia
| |
Collapse
|