Thyroid Eye Disease: Therapy in the Active Phase

Whole-orbit radiomics: machine learning-based multi- and fused- region radiomics signatures for intravenous glucocorticoid response prediction in thyroid eye disease

BACKGROUND

Radiomics analysis of orbital magnetic resonance imaging (MRI) shows preliminary potential for intravenous glucocorticoid (IVGC) response prediction of thyroid eye disease (TED). The current region of interest segmentation contains only a single organ as extraocular muscles (EOMs). It would be of great value to consider all orbital soft tissues and construct a better prediction model.

METHODS

In this retrospective study, we enrolled 127 patients with TED that received 4·5 g IVGC therapy and had complete follow-up examinations. Pre-treatment orbital T2-weighted imaging (T2WI) was acquired for all subjects. Using multi-organ segmentation (MOS) strategy, we contoured the EOMs, lacrimal gland (LG), orbital fat (OF), and optic nerve (ON), respectively. By fused-organ segmentation (FOS), we contoured the aforementioned structures as a cohesive unit. Whole-orbit radiomics (WOR) models consisting of a multi-regional radiomics (MRR) model and a fused-regional radiomics (FRR) model were further constructed using six machine learning (ML) algorithms.

RESULTS

The support vector machine (SVM) classifier had the best performance on the MRR model (AUC = 0·961). The MRR model outperformed the single-regional radiomics (SRR) models (highest AUC = 0·766, XGBoost on EOMs, or LR on OF) and conventional semiquantitative imaging model (highest AUC = 0·760, NaiveBayes). The application of different ML algorithms for the comparison between the MRR model and the FRR model (highest AUC = 0·916, LR) led to different conclusions.

CONCLUSIONS

The WOR models achieved a satisfactory result in IVGC response prediction of TED. It would be beneficial to include more orbital structures and implement ML algorithms while constructing radiomics models. The selection of separate or overall segmentation of orbital soft tissues has not yet attained its final optimal result.

Using the Random Forest Algorithm to Detect the Activity of Graves Orbitopathy

PURPOSE

The aim of this study is to establish a random forest model to detect active and quiescent phases of patients with Graves Orbitopathy (GO).

METHODS

A total of 243 patients (486 eyes) diagnosed with GO in Beijing TongRen hospital were included in the study. The Clinical Activity Score of GO was regarded as the golden standard, whereas sex, age, smoking status, radioactive I131 treatment history, thyroid nodules, thyromegaly, thyroid hormone, and Thyroid-stimulating hormone receptor antibodies were chosen as predictive characteristic variables in the model. The random forest model was established and compared with logistic regression analysis, Naive Bayes, and Support vector machine metrics.

RESULTS

Our model has a sensitivity of 0.81, a specificity of 0.90, a positive predictive value of 0.87, a negative predictive value of 0.86, an F1 score of 0.85, and an out-of-bag error of 0.15. The random forest algorithm showed a more precise performance compared with 3 other models based on the area under receiver operating characteristic curve (0.92 versus 0.77 versus 0.76 versus 0.75) and accuracy (0.86 versus 0.71 versus 0.69 versus 0.66).

CONCLUSIONS

By integrating these high-risk factors, the random forest algorithm may be used as a complementary method to determine the activity of GO, with accurate and reliable performance.

T2 -Weighted MR Imaging-Derived Radiomics for Pretreatment Determination of Therapeutic Response to Glucocorticoid in Patients With Thyroid-Associated Ophthalmopathy: Comparison With Semiquantitative Evaluation

BACKGROUND

MR imaging has been applied to determine therapeutic response to glucocorticoid (GC) before treatment in thyroid-associated ophthalmopathy (TAO), while the performance was still poor.

PURPOSE

To investigate the value of T₂ -weighted imaging (T₂ WI)-derived radiomics for pretreatment determination of therapeutic response to GC in TAO patients, and compare its diagnostic performance with that of semiquantitative parameters.

STUDY TYPE

Retrospective.

POPULATION

A total of 110 patients (49 ± 12 years; male/female, n = 48/62; responsive/unresponsive, n = 62/48), divided into training (n = 78) and validation (n = 32) cohorts.

FIELD STRENGTH/SEQUENCE

3.0 T, T₂ -weighted fast spin echo.

ASSESSMENT

W.C. and H.H. (6 and 10 years of experience, respectively) performed the measurements. Maximum, mean, and minimum signal intensity ratios (SIRs) of extraocular muscle (EOM) bellies were collected to construct a semiquantitative imaging model. Radiomics features from volumes of interest covering EOM bellies were extracted and three machine learning-based (logistic regression [LR]; decision tree [DT]; support vector machine [SVM]) models were built.

STATISTICAL TESTS

The diagnostic performances of models were evaluated using receiver operating characteristic curve analyses, and compared using DeLong test. Two-sided P < 0.05 was considered statistically significant.

RESULTS

The responsive group showed higher minimum signal intensity ratio (SIR_min ) of EOMs than the unresponsive group (training: 1.46 ± 0.34 vs. 1.18 ± 0.39; validation: 1.44 ± 0.33 vs. 1.19 ± 0.20). In both cohorts, LR-based radiomics model demonstrated good diagnostic performance (area under the curve [AUC] = 0.968, 0.916), followed by DT-based (AUC = 0.933, 0.857) and SVM-based models (AUC = 0.919, 0.855). All three radiomics models outperformed semiquantitative imaging model (SIR_min : AUC = 0.805) in training cohort. In validation cohort, only LR-based radiomics model outperformed that of SIR_min (AUC = 0.745). The nomogram integrating LR-based radiomics signature and disease duration further elevated the diagnostic performance in validation cohort (AUC: 0.952 vs. 0.916, P = 0.063).

DATA CONCLUSION

T₂ WI-derived radiomics of EOMs, together with disease duration, provides a promising noninvasive approach for determining therapeutic response before GC administration in TAO patients.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY: Stage 4.

Current insights of applying MRI in Graves' ophthalmopathy

Graves' ophthalmopathy (GO) is an autoimmune disease related to Grave's disease (GD). The therapeutic strategies for GO patients are based on precise assessment of the activity and severity of the disease. However, the current assessment systems require development to accommodate updates in treatment protocols. As an important adjunct examination, magnetic resonance imaging (MRI) can help physicians evaluate GO more accurately. With the continuous updating of MRI technology and the deepening understanding of GO, the assessment of this disease by MRI has gone through a stage from qualitative to precise quantification, making it possible for clinicians to monitor the microstructural changes behind the eyeball and better integrate clinical manifestations with pathology. In this review, we use orbital structures as a classification to combine pathological changes with MRI features. We also review some MRI techniques applied to GO clinical practice, such as disease classification and regions of interest selection.

Prognostic factors of restrictive myopathy in thyroid eye disease

To investigate the prognostic factors of extraocular muscle restriction in patients with thyroid eye disease (TED), 65 patients with TED and restrictive myopathy were evaluated. Demographics, clinical activity score (CAS), smoking status, thyroid disease status, thyroid hormone status, thyroid autoantibody status, orbital computed tomography (CT) scan at initial presentation, and treatment regimens were assessed. The movements of the most severely affected extraocular muscles were categorized into five grades. The patients were divided into the improved and the not-improved group based on the improvement in the limitation of the extraocular muscle excursion (LOM) throughout the follow-up, and the groups were compared using clinical factors. The mean LOM significantly improved from 2.3 ± 1.1 to 1.7 ± 1.2 after 1 year of follow-up. The excursion of the most restricted muscle improved in 32 patients but not in 33 patients during the follow-up. The initial concentration of the thyroid-stimulating antibody (TSAb) was significantly lower in the improved (229.3 ± 114.1) than in the not-improved group (345.0 ± 178.6) (P = 0.02) Age, sex, smoking status, CAS, thyroid status, and muscle thickness on the CT scan did not significantly differ in the groups. This study showed that the initial concentration of TSAb is a factor affecting the recovery of restrictive myopathy.

Teprotumumab: The Dawn of Therapies in Moderate-to-Severe Thyroid-Associated Ophthalmopathy

Thyroid-associated ophthalmopathy (TAO) is a potentially sight-threatening ocular disease. About 3-5% of patients with TAO have severe disease with intense pain, inflammation, and sight-threatening corneal ulceration or compressive optic neuropathy. The current treatments of TAO are often suboptimal, mainly because the existing therapies do not target the pathogenesis of the disease. TAO mechanism is unclear. Ocular fibrocytes express relatively high levels of the functional TSH receptor (TSHR), and many indirect evidences support its participation. Over expression of insulin-like growth factor-1 receptor (IGF-IR) in fibroblasts, leading to inappropriate expression of inflammatory factors, production of hyaluronic acid and cell activation in orbital fibroblasts are also possible mechanisms. IGF-1R and TSHR form a physical and functional signaling complex. Inhibition of IGF-IR activity leads to the attenuation of signaling initiated at either receptor. Teprotumumab (TMB) is a human immunoglobulin G1 monoclonal antibody, binding to IGF-IR. Recently two TMB clinical trials had been implemented in TAO patients, indicating dramatic reductions in disease activity and severity, which approved its use for the treatment of TAO in the US. This review summarizes the treatments of TAO, focusing on the pathogenesis of IGF-1R in TAO and its application prospects.

Predicting the response to glucocorticoid therapy in thyroid-associated ophthalmopathy: mobilizing structural MRI-based quantitative measurements of orbital tissues

PURPOSE

We aimed to evaluate the performance of structural magnetic resonance imaging (MRI)-based quantitative measurements at extraocular muscle (EOM), orbital fat (OF), and especially lacrimal gland (LG) in predicting response to glucocorticoid (GC) in patients with active and moderate-severe thyroid-associated ophthalmopathy (TAO).

METHODS

Forty-seven active and moderate-severe TAOs (responsive group, 29 patients and 58 eyes; unresponsive group, 18 patients and 36 eyes) were enrolled. Pretreatment MRI-based parameters of EOM, OF, and LG, and clinical factors were retrospectively collected and compared between two groups. Logistic regression and receiver operating characteristic curve analyses were used to assess the predictive value of identified independent variables.

RESULTS

Responsive group showed significantly higher minimum signal intensity ratio of EOM (EOM-SIR_min) (p < 0.001), higher EOM-SIR_mean (p = 0.034), higher LG herniation (LGH) (p = 0.019), lower OF thickness (OFT) (p = 0.017), higher LGH/OFT ratio (p = 0.001), and shorter disease duration (p = 0.004) than unresponsive group. Multivariate analysis showed that EOM-SIR_min, LGH/OFT ratio, and disease duration were independent predictors for responsive TAOs (all p < 0.05). Integration of three independent predictors demonstrated optimal predictive efficiency (area under curve, 0.829). Combining EOM-SIR_min ≥1.43 and LGH/OFT ratio ≥1.65, optimal predictive specificity (94.4%) could be obtained, while optimal predictive sensitivity (82.8%) was achieved when integrating disease duration ≤3.5 and LGH/OFT ratio ≥1.65.

CONCLUSIONS

Structural MRI-based quantitative measurements at EOM, OF, and LG, specially EOM-SIR_min and LGH/OFT ratio, together with disease duration, may serve as promising markers to predict response to GC in patients with active and moderate-severe TAO.

T2 mapping histogram at extraocular muscles for predicting the response to glucocorticoid therapy in patients with thyroid-associated ophthalmopathy

AIM

To evaluate the performance of T2 mapping histograms at the extraocular muscles (EOMs) in predicting the response to glucocorticoid therapy in the patients with active and moderate-severe thyroid-associated ophthalmopathy (TAO).

MATERIALS AND METHODS

Thirty active and moderate-severe TAO patients (responsive group, n=20; unresponsive group, n=10) were enrolled, and evaluated using T2 mapping before treatment. Histogram parameters (mean, median, max, min, 10th, 90th percentiles, skewness, and kurtosis) of T2 relaxation time (T2RT) at the EOMs for each orbit, and clinical variables (age, sex, disease duration, anti-thyroid treatment, smoking habit, pre-treatment thyroid function, thyrotrophin receptor antibody, diplopia presence, activity and severity scores) were collected and compared between groups. Logistic regression and receiver operating characteristic (ROC) curve analyses were used to assess the predictive value of identified independent variables for treatment response.

RESULTS

The responsive group showed significantly shorter disease duration (p=0.003), while higher T2RT_min than unresponsive group (p<0.001). Multivariate analysis showed that T2RT_min and disease duration were independent predictors for responsive TAOs. ROC curve analyses indicated that setting a cut-off value of ≥54.3 for T2RT_min demonstrated the optimal predicting specificity for responsive TAOs (100%), while a combination of T2RT_min ≥54.3 and disease duration ≤4.5 showed optimal predicting efficiency and sensitivity (area under the curve, 0.820; sensitivity, 65%).

CONCLUSIONS

Histogram analysis can help to exhibit the heterogeneity of T2RT at the EOMs. T2RT_min, together with disease duration may be the promising marker for predicting response to glucocorticoid therapy in the patients with active and moderate-severe TAO.

Orbital apex disorders: Imaging findings and management

Orbital apex disorders include orbital apex syndrome, superior orbital fissure syndrome and cavernous sinus syndrome. These disorders result from various etiologies, including trauma, neoplastic, developmental, infectious, inflammatory as well as vascular causes. In the past, these have been described separately based on anatomical locations of disease process; however, these three disorders share similar causes, diagnostic evaluation and management strategies. The etiology is diverse and management is directed to the causative process. This imaging review summarizes the pertinent anatomy of the orbital apex and illustrates representative pathological processes that may affect this region. The purpose of this review is to provide an update on the current status of diagnostic imaging and management of patients with orbital apex disorders.

Seeing is believing: A review of apheresis therapy in the treatment of ophthalmologic disease

Apheresis procedures have a role in treatment of disparate diseases involving many different organ systems. Often the disease processes where apheresis plays a role in treatment are considered "orphan diseases"-relatively rare disease processes that lack specific pharmaceutical agents or established treatment protocols. Many of these disease processes can affect the eye with devastating results for the eyesight of these patients. The unique ability of apheresis to affect disease by modifying blood plasma and modulating disease-causing agents therein renders apheresis procedures valuable tools in the treatment of certain ophthalmologic diseases. This review comprehensively evaluates the role of apheresis in the treatment of ophthalmologic diseases of the eye and surrounding orbit including age-related macular degeneration, bilateral diffuse uveal melanocytic proliferation, paraneoplastic retinopathy, atopic keratoconjunctivitis, sympathetic ophthalmia, and endocrine-associated ophthalmopathy. Apheresis procedure parameters are provided for the apheresis practitioner based on review of the relevant literature.

Dysthyroid optic neuropathy: update on pathogenesis, diagnosis, and management

INTRODUCTION

Dysthyroid optic neuropathy (DON) is a severe manifestation of thyroid eye disease (TED) that can result in permanent vision loss. Management is complex, multidisciplinary, and involves medical and/or surgical therapies. This review describes current concepts in the epidemiology, pathophysiology, diagnosis, and treatment of DON.

AREAS COVERED

An extensive review of the literature was performed to detail current concepts on the diagnosis and management of DON. This includes utilization of various medical and surgical modalities for disease management.

EXPERT COMMENTARY

DON can result in permanent blindness and often requires the use of corticosteroids and surgical decompression. We favor the use of intravenous corticosteroids and a transcaruncular approach when surgical decompression is indicated. The use of orbital radiation for DON is often reserved for patients that are poor surgical candidates and/or patients with refractory disease.

Thyroid eye disease: a review

Thyroid eye disease is a multifactorial autoimmune disease with a spectrum of signs and symptoms. Oftentimes, the diagnosis of thyroid eye disease is straightforward, based upon history and physical examination. The purpose of this review is to assist the eye-care practitioner in staging the severity of thyroid eye disease (mild, moderate-to-severe and sight-threatening) and correlating available treatment modalities. Eye-care practitioners play an important role in the multidisciplinary team by assessing functional vision while also managing ocular health.

Diagnosis and stage of Graves' ophthalmopathy: Efficacy of quantitative measurements of the lacrimal gland based on 3-T magnetic resonance imaging

The aim of the present study was to determine the efficacy of quantitative measurements of the lacrimal gland based on 3-T magnetic resonance (MR) imaging in the diagnosis and staging of Graves' ophthalmopathy (GO). The study retrospectively enrolled 33 patients with GO (the GO group) and 24 healthy volunteers [the healthy control (HC) group] with orbit MR imaging performed using a 3-T MR scanner. Quantitative parameters of the lacrimal gland, including axial length, axial width, axial area, coronal length, coronal width, coronal area, volume and signal intensity ratio (SIR) of the lacrimal gland to the ipsilateral temporal muscle were measured. The difference of quantitative parameters between the GO and HC groups, or between active and inactive GO groups were evaluated using a Student's t-test. Receiver operating characteristic analyses were used to evaluate the diagnostic value of the significant parameters in discriminating patients with GO from healthy controls, or discriminating active from inactive GO. All the quantitative measurements of the GO patients were significantly larger than those of the healthy controls (P<0.05), with the exception of the coronal length (P=0.150). Axial and coronal width had the best efficacy in discriminating patients with GO from healthy controls. Only SIR was found to be different between the active and inactive GO groups (P=0.001). Setting a SIR of 2.57 as the threshold value, the optimal efficacy was achieved (area under the curve, 0.711; sensitivity, 57.7%; specificity, 77.5%) in discriminating between active and inactive GO. Quantitative measurements of the lacrimal gland based on 3-T MR imaging may assist in the diagnosis and stage of GO.

Compressive Optic Neuropathy and Repeat Orbital Decompression: A Case Series

PURPOSE

To characterize the presenting characteristics, preoperative clinical activity score (CAS), surgical approach, and visual outcomes in patients with thyroid eye disease undergoing repeat orbital decompression for recurrent or recalcitrant compressive optic neuropathy (CON).

METHODS

The medical records of patients with recurrent or recalcitrant CON undergoing repeat orbital decompressions were retrospectively reviewed. The primary outcome measures included pre- and postoperative Humphrey visual field mean deviation, visual acuity (VA) measured in logarithm of the minimal angle of resolution, color vision measured by Ishihara plates, and presence of relative afferent pupillary defect. Details of the surgical procedure and each patient's CAS at presentation were also recorded.

RESULTS

Six patients, 9 orbits, with a mean preoperative CAS of 3.8 were included in this review. The mean time between initial decompression and presentation to our center for recurrent or persistent CON symptoms was 8.6 years (range, 1 to 15 years). At presentation, the average Humphrey visual field mean deviation was -16.5 (standard deviation: 8.8), improving to -3.8 (2.4) postoperatively with a mean of 9.3 months follow up (mean improvement of 75%). Preoperative VA was 0.34 (0.23) LogMAR, improving to 0.05 (0.10) LogMAR with a mean follow up of 10.4 months. Pre- to postoperative comparisons of clinical measures all showed statistically significant improvement (p < 0.05). Eight eyes presented with decreased VA (any VA < 20/20), 4 with decreased color vision (any color vision < 11), and 1 with a relative afferent pupillary defect, and all these patients demonstrated improvement following repeat orbital decompression.

CONCLUSIONS

In patients with thyroid eye disease, symptoms of recurrent CON occurred up to 15 years following initial orbital decompression underscoring the smoldering, progressive nature of the disease. Repeat decompression that focused on the orbital apex resulted in visual improvement in all 6 patients. Despite clinical evidence of CON, the mean CAS of these patients at presentation was only 3.8, highlighting the importance of close monitoring of patients with thyroid eye disease following decompression regardless of the external manifestations of disease activity.

Orbital apex disorders: a case series

Orbital apex syndrome is an uncommon disorder characterized by ophthalmoplegia, proptosis, ptosis, hypoesthesia of the forehead, and vision loss. It may be classified as part of a group of orbital apex disorders that includes superior orbital fissure syndrome and cavernous sinus syndrome. Superior orbital fissure syndrome presents similarly to orbital apex syndrome without optic nerve impairment. Cavernous sinus syndrome includes hypoesthesia of the cheek and lower eyelid in addition to the signs seen in orbital apex syndrome. While historically described separately, these three disorders share similar causes, diagnostic course, and management strategies. The purpose of this study was to report three cases of orbital apex disorders treated recently and to review the literature related to these conditions. Inflammatory and vascular disorders, neoplasm, infection, and trauma are potential causes of orbital apex disorders. Management is directed at the causative process. The cases described represent a rare but important group of conditions seen by the maxillofacial surgeon. A review of the clinical presentation, etiology, and management of these conditions may prompt timely recognition and treatment.