Thyroid cancer : a novel insight into diagnostic and therapeutic approach

Abstract

There has been a substantial increase, worldwide, in the incidence of thyroid cancer over the past few decades. The diagnosis of thyroid cancer poses a clinical challenge to medical practitioners. Fine needle aspiration cytology (FNAC) is a current standard method to diagnose thyroid cancer. FNAC has a false-positive rate of 1% - 5% and a false-negative rate of 0% - 7.7%, whilst 20% - 30% of cases remain non-diagnostic. Grey-scale ultrasound is a useful imaging technique for the assessment of the thyroid gland, because it can characterize various features of thyroid nodules such as the number, site, size, shape, echogenicity, and internal architecture. However, grey-scale ultrasound has limited value in differentiating benign and malignant thyroid nodules because of its varied sensitivity (52% - 97%) and specificity (26% - 83%). The varied sensitivity and specificity of grey-scale ultrasound might be due to the qualitative analysis and subjective interpretation of the characteristics of thyroid nodules and renders it thus vulnerable to inter- and intra-observer variations. Nevertheless, certain features of thyroid nodules can be quantified, for example tissue stiffness and the vascularity index. Shear wave elastography is a novel ultrasound technique that can quantify tissue stiffness by tracking the propagation of shear waves through thyroid nodules. In general, malignant thyroid nodules tend to be stiffer than benign nodules. Depending upon the differences in stiffness values, thyroid nodules can be differentiated as benign or malignant. The first study of this thesis evaluated the feasibility of using shear wave elastography in predicting thyroid malignancy, and determined if any (and if so, which ones) of the shear wave elastography indices (Emaximum, Emean, and Eminimum) are potential predictors of thyroid malignancy. The study further evaluated the diagnostic accuracy in differentiating benign and malignant thyroid nodules when grey-scale ultrasound was combined with shear wave elastography. Vascularity is a dynamic feature usually detected on Doppler ultrasound. Central vascularity and hypervascularity are usually associated with malignant thyroid nodules. However, not all previous studies appreciate the usefulness of vascularity in predicting thyroid malignancy. Controversial results might be due to the fact that previous studies evaluated vascularity by visual assessment methods that involve subjective interpretation and thus result in high inter- and intra-observer variations. Moreover, in previous studies, the methods used to delineate a border between peripheral and central regions of a thyroid nodule were not standardized. In the second study in this thesis, we have thus developed a computer algorithm that can perform regional segmentation of thyroid nodules by using an 'offsetting' method, and quantify the overall vascularity as well as the vascularity in peripheral and central regions of thyroid nodules. Based on the differences in the vascularity indices, thyroid nodules can be differentiated into benign and malignant types. The study further evaluated the potential advantage of combining the vascularity index with grey-scale ultrasound to enhance the diagnostic accuracy of thyroid malignancy. In the first and second study, a total of 111 patients with solitary thyroid nodules were included. Each thyroid nodule was assessed with grey-scale ultrasound, shear wave elastography, and colour Doppler ultrasound. The diagnosis of the thyroid nodules was confirmed by FNAC and/or histological examination. Grey-scale ultrasound features including microcalcification (malignant: 77.8% versus benign: 7.1%), hypoechoic (92.6% versus 33.3%), irregular margins (55.6% versus 16.7%), and a height-to-width ratio > 1 (59.3% versus 13.1%) were found to be more frequently associated with malignant thyroid nodules than with benign thyroid nodules. The differences were statistically significant (all P < 0.05). Regarding shear wave elastography, the results suggested that Emaximum ≥ 67.3 kPa and Emean ≥ 23.1 kPa are independent predictors of thyroid malignancy. Emaximum was found to be the best adjunct to grey-scale ultrasound. The combination of Emaximum or Emean with grey-scale ultrasound enhanced the diagnostic accuracy of grey-scale ultrasound from 58.5% to 80.2% and 78.4%, respectively (P < 0.05). The results of the vascularity index (VI) quantification showed that a 22% offset was optimal for regional subdivision of thyroid nodules. At the optimum offset, the mean VI of peripheral, central, and overall regions of malignant nodules were significantly higher than those of benign nodules (26.5 ± 16.2%, 21.7 ± 19.6%, 23.8 ± 4.6% versus 18.2 ± 16.7%, 11.9 ± 15.1%, and 16.6 ± 1.8%, respectively, P < 0.05). The optimum cut-off points of peripheral, central, and overall VI were 19.7%, 9.1%, and 20.2%, respectively. When compared to grey-scale ultrasound alone, a combination of VI assessment and grey-scale ultrasound evaluation of thyroid nodules increased the overall diagnostic accuracy from 58.6% to 79.3% (P < 0.05). Studies 1 and 2 have clinical significance in establishing methods for accurate diagnosis of thyroid cancer. The results of Study 1 suggest that shear wave elastography has clinical importance in differentiating benign and malignant thyroid nodules. The combination of grey-scale ultrasound with Emaximum or Emean significantly improved the diagnostic accuracy in predicting thyroid cancer. Study 2 has devised a new method to perform regional segmentation of thyroid nodules and to quantify vascularity in each segment. This approach is objective and standardized to quantify thyroid vascularity and helps in differentiating benign and malignant thyroid nodules. Colour Doppler ultrasound (CDU) and power Doppler ultrasound (PDU) are widely used for detecting the vasculature of tissues or organs. In the thyroid gland, CDU and PDU have been used for the differential diagnosis of, amongst others, Hashimoto's disease, Graves' disease, and Reidel's thyroiditis. Doppler ultrasound has also been used in monitoring treatment responses during therapy of thyroid disorders. However, the detection of blood flow in minute blood vessels or vessels with low blood flow is challenging due to technical limitations of previously developed Doppler ultrasound modalities. Most recently, a new ultrasound technology, namely AngioPLUS (Planewave UltraSensitive™ imaging), provides superb sensitivity in the detection of tissue vascularity. The beauty of this technique is that all colour pixels of the tissue can be reconstructed in a single image. AngioPLUS provides high resolution and 3D wall filtering that allow efficient discrimination between blood flow and other soft tissues by analysing space, time, and amplitude information. Study 3 aimed to evaluate the feasibility of using AngioPLUS imaging in assessing thyroid vascularity when combined with CDU or PDU. The study further evaluated whether the addition of AngioPLUS to CDU or PDU enhances the sensitivity in detecting vasculature of thyroid parenchyma. It also investigated whether there is any asymmetry of vascularity between the right and left thyroid lobes. A total of 45 healthy volunteers underwent grey-scale ultrasound, CDU, CDU+AngioPLUS, PDU, and PDU+AngioPLUS evaluations of both lobes of the thyroid gland. Thyroid vascularity was evaluated using our in-house computer algorithm. The results showed that the combination of CDU+AngioPLUS (14.7 ± 9.4%) and the combination of PDU+AngioPLUS (13.4 ± 9%) had significantly higher thyroid VI than CDU (8.8 ± 7.3%) and PDU (4.7 ± 5.4%) alone (all P < 0.05). No asymmetry was found between the VI of the right and left thyroid lobes (P > 0.05). Study 3 highlights the differences in sensitivity of detecting thyroid vasculature assessed by different Doppler ultrasound modalities. The results suggest that AngioPLUS enhances the detection of vascularity when added to PDU or CDU. The clinical significance of the study lies in the detection of small blood vessels and vessels with low blood flow that may help disease diagnosis and treatment monitoring. Study 4 highlights the scope of chemotherapeutic agents in treating papillary thyroid cancer. Current strategies to treat papillary thyroid cancer are largely based on surgery, where recurrence rate is high (up to 33%). Drug development is a costly and time-consuming process. Moreover, the identification of novel therapeutic targets is challenging. Study 4 introduces the concept of 'repurposing of drugs' that evaluates the therapeutic potential of already approved drugs beyond the scope of their primary clinical usage. Cannabinoids are derivatives of the marijuana plant. They have been used for recreation and to relieve pain. Two cannabinoid receptors (CB1 and CB2) are known to be distributed over different body organs and systems in humans. Cannabinoids receptor expression has been noted in many cancers including breast cancer, prostate cancer, hepatic cancer, lung cancer, and colorectal cancer. Anti-cancer actions, including antiproliferation, antimigration, antiangiogenesis, and apoptosis, have been validated in in vitro experiments in various cancer cell lines. In a recent study, immunohistochemistry analysis of surgical specimens of 87 thyroid nodules demonstrated that CB1 and CB2 receptor expression was more significantly associated with papillary thyroid cancer than with benign thyroid nodules. It was further found that CB2 expression was significantly higher than CB1 receptor expression. In Study 4, the therapeutic potential of CB2 receptor agonist (JWH-133) in treating papillary thyroid cancer was evaluated. A normal thyroid follicular cell line (N-thy-ori-3) was used as the control, and BCPAP was the papillary thyroid cancer cell line used in this study. Both cell lines were treated with JWH-133 at 0, 5, 10, 15, 20, 25, and 30-µM concentrations for 24, 48, and 72 hours. Cellular metabolic activity and cell viability were evaluated using an MTT assay. The results demonstrated that 25 µM was the lethal dose concentration for BCPAP cells at which cell viability was reduced to 50% after the optimal 48 hours of incubation. The results also suggested that BCPAP cells were more sensitive to the JWH-133 as compared with N-thy-ori-3. The cytotoxic effect mediated by JWH-133 was not significantly inhibited by CB2 receptor antagonist SR144528 in both cell lines. The results of the study demonstrated that JWH-133 has a potent cytotoxic effect, more pronounced in a papillary thyroid cancer cell line (BCPAP) than in a normal follicular thyroid cell line (N-thy-ori-3). Study 4 suggests that JWH-133 induces cell death in papillary thyroid cancer cells, whilst the survival of normal thyroid follicular cells can be maintained at an acceptable level.