PDF
Citation
Print
INTRODUCTION
Oncocytic carcinoma (OCA), previously referred to as Hürthle cell carcinoma, is a rare subtype of thyroid cancer that was historically classified as a variant of follicular carcinoma. However, in the 4th edition of the World Health Organization (WHO) classification of endocrine tumors, it was reclassified as a distinct entity due to its unique molecular and clinicopathologic features [1]. This reclassification was further refined in the 2022 edition, where the term Hürthle cell carcinoma was officially replaced with OCA to more accurately reflect its specific characteristics [2]. These updates underscore the importance of differentiating OCA from other thyroid neoplasms to ensure appropriate diagnosis and management.
OCA is known for being more aggressive than differentiated thyroid cancers (DTCs), which often necessitates a more assertive therapeutic approach [34]. However, OCA can only be definitively diagnosed through histopathologic examination, as it requires confirmation of features such as capsular or vascular invasion [5]. Additionally, distinguishing OCA from oncocytic adenoma (OA) before surgery remains a significant challenge due to their overlapping clinical and cytologic features, making preoperative identification unreliable [67].
Adding to this complexity, OCA is a rare disease, accounting for only 3%–5% of DTCs [89]. Its low prevalence has limited the availability of large-scale studies investigating the differences between OCA and OA. As a result, appropriate guidelines for preoperative diagnosis and treatment planning remain insufficient, complicating clinical decision-making.
This study aims to differentiate OCA from OA by evaluating their preoperative characteristics, including clinical presentation, ultrasonography, and cytologic findings. By identifying distinguishing features, we seek to enhance preoperative diagnostic accuracy and contribute to more informed clinical management of thyroid neoplasms.
METHODS
Ethics statement
This study was approved by the Institutional Review Board of Kyungpook National University Chilgok Hospital (No. 2024-09-032). Informed consent from patients was not required due to the retrospective design of the study.
Patients
This retrospective study analyzed the medical records of patients who underwent thyroid surgery and were diagnosed with oncocytic neoplasm based on pathologic reports between January 2011 and July 2024 at Kyungpook National University Chilgok Hospital in Daegu, Korea. Patients with available medical records, including ultrasonographic and cytologic results, were included in the study. A total of 83 patients met the inclusion criteria, comprising 23 patients (27.7%) diagnosed with OCA and 60 patients (72.3%) diagnosed with OA.
Clinical characteristics, including age and sex, were compared between the 2 groups. Histopathologic findings of the tumors, such as size, presence of concurrent carcinoma, and Hashimoto thyroiditis, were reviewed. For OCA cases, additional factors, including extrathyroidal extension (ETE), angioinvasion, and N and M stages based on the American Joint Committee on Cancer 8th edition, were evaluated [10]. According to the WHO classification, OCA was subcategorized into minimally invasive, encapsulated angioinvasive, and widely invasive, reflecting the degree of invasiveness [2]. This factor was also evaluated in this study. Furthermore, the diagnostic accuracy of preoperative ultrasonography and cytology was assessed to differentiate OCA from OA.
Preoperative ultrasonography and cytology
All patients underwent preoperative ultrasonographic evaluations for thyroid nodules, and all the images were retrospectively reviewed by a single thyroid surgeon with over 20 years of experience in thyroid ultrasonography. The ultrasonographic findings were classified according to the Korean Thyroid Imaging Reporting and Data System (K-TIRADS) and the American College of Radiology Thyroid Imaging Reporting and Data System (ACR TI-RADS) [1112]. Additionally, other sonographic features, including the presence of a halo, vascularity, and trabecular formation, were analyzed to assess their diagnostic utility. This classification was applied because, in many cases, the original descriptions lacked specific sonographic features relevant to this study.
Ultrasonography-guided fine-needle aspiration (FNA) was performed for diagnosis, and cytologic results were reported according to the Bethesda System for Reporting Thyroid Cytopathology [13]. Bethesda category IV includes oncocytic (Hürthle cell) neoplasm. Therefore, cases diagnosed as follicular neoplasm and oncocytic neoplasm were analyzed separately. For FNA results obtained from external medical facilities, slides were re-evaluated by pathologists at our institution. In instances with multiple FNA results, the most suspicious cytologic findings were used for analysis.
In cases with multiple thyroid nodules, FNA was performed selectively on the most suspicious nodule based on ultrasonographic findings. If suspicious nodules were present in both lobes, FNA was performed bilaterally, as the results could impact the surgical extent. The final pathological diagnosis and the nodule subjected to FNA were confirmed on a nodule-to-nodule basis, ensuring accuracy in the correlation between cytologic and histopathologic findings.
Surgery
The type of surgical procedure was determined based on various factors, including tumor size, location, the presence of concurrent carcinoma, and cytologic results. Decisions regarding central neck lymph node dissection were made using similar criteria.
Statistical analysis
All statistical analyses were conducted using IBM SPSS Statistics for Windows, ver. 27 (IBM Corp.). Categorical variables are presented as counts and percentages, while continuous variables are presented as means and standard deviations. The chi-square test was primarily used to compare categorical variables. However, if more than 20% of cells had an expected frequency of less than 5, the Fisher exact test was applied. The Student t-test was used to analyze continuous variables. A P-value <0.05 was considered statistically significant.
RESULTS
Comparison of clinical and pathological characteristics
The clinical and histopathologic characteristics of the 2 groups are summarized in Table 1. While patients with OCA tended to be older than those with OA, the difference was not statistically significant (59.4 ± 10.8 years vs. 53.2 ± 14.2 years, P = 0.063). The majority of patients in both groups were women, with similar proportions (73.9% vs. 73.3%, P = 0.957).
Tumor size, based on the final pathologic report, was significantly larger in OCA compared to OA (3.8 ± 1.7 cm vs. 2.3 ± 1.4 cm, P < 0.001). When tumors were categorized by size using cutoff values of 2.0 cm and 4.0 cm, nearly half of OCA cases fell into the >4.0 cm category, whereas OA cases were more frequently categorized as ≤2.0 cm (P = 0.001). Similarly, ultrasonographic tumor size measurements were consistent, with OCA demonstrating significantly larger sizes compared to OA (4.2 ± 1.7 cm vs. 2.7 ± 1.4 cm, P < 0.001). Notably, in both groups, tumor sizes measured using ultrasonography tended to be larger than their corresponding pathologic sizes.
Concurrent carcinoma was observed in 21.7% of OCA cases and 16.7% of OA cases, but the difference was not statistically significant (P = 0.654). Similarly, Hashimoto thyroiditis was found in 13.0% of OCA patients and 20.0% of OA patients, without a significant difference between the groups (P = 0.542).
The surgical extent was similar between the OCA and OA groups. One patient in each group underwent isthmectomy, while the majority of patients underwent lobectomy (65.2% vs. 78.3%, P = 0.355).
Comparison of ultrasonographic features
The sonographic features of OCA and OA, based on K-TIRADS criteria, are summarized in Table 2. Most nodules in both groups exhibited solid composition (65.2% vs. 85.0%, P = 0.056), parallel orientation (91.3% vs. 100%, P = 0.074), and smooth margins (73.9% vs. 88.3%, P = 0.196), with no echogenic foci present (87.0% vs. 91.7%, P = 0.728). Isoechoic and hypoechoic nodules were similarly distributed between the groups (P = 0.206). The majority of nodules in both groups were categorized as K-TIRADS category 3 (56.5% vs. 48.3%, P = 0.314). Although the difference was not statistically significant, OA tended to fall into a higher K-TIRADS category.
Table 3 compares the ultrasonographic features using ACR TI-RADS criteria. The results for composition, echogenicity, shape, and echogenic foci were similar to those observed with K-TIRADS criteria. However, OCA cases were more likely to exhibit lobulated or irregular features (17.4% vs. 3.3%, P = 0.047) and were more frequently categorized as category 5 (17.4% vs. 1.7%, P = 0.016).
Analysis of additional sonographic features (Table 4) demonstrated significant differences in vascularity and trabecular formation between the 2 groups. OCA cases showed a higher prevalence of intranodular vascularity (65.2% vs. 33.3%, P = 0.049) and trabecular formation (17.4% vs. 1.7%, P = 0.019) compared to OA cases.
Preoperative cytopathology
Table 5 presents the cytopathological results, revealing a significant difference between the groups (P < 0.001). The majority of oncocytic neoplasms were classified as Bethesda category IV, with 95.7% of OCA and 75.0% of OA falling into this category. Among these, most patients with OCA were diagnosed with follicular neoplasm, with the exception of 2 cases identified as oncocytic neoplasm. In contrast, the majority of OA cases were diagnosed as oncocytic neoplasm. Additionally, 6 patients (10.0%) with OA were diagnosed as benign, while 7 patients (11.7%) were classified as atypia of undetermined significance.
Oncocytic carcinoma characteristics
The histopathologic characteristics of OCA are presented in Table 6. OCA cases predominantly exhibited minimally invasive patterns (91.3%), except for 2 patients who demonstrated widely invasive disease. These 2 patients also exhibited gross ETE and angioinvasion involving more than 4 foci. Although prophylactic central neck lymph node dissection was not routinely performed, no patients were found to have lymph node metastasis. Distant metastasis was identified in only one patient (4.3%). During a mean follow-up period of 50 months, 2 patients (8.7%) experienced recurrence. One patient developed recurrence at the thyroidectomy bed, presenting as differentiated high-grade thyroid carcinoma 2 years after the initial surgery. The other patient experienced lung metastasis 8 years postoperatively. Importantly, no cancer-related deaths were observed during the follow-up period.
DISCUSSION
In this study, we comprehensively evaluated factors associated with malignancy risk in pathologically confirmed oncocytic neoplasms. Consistent with previous reports [141516], our study demonstrated that OCA cases had significantly larger tumor sizes compared to OA cases in both sonographic and pathologic measurements. Although male sex and older age have also been reported as indicators of malignancy [151718], our findings showed no significant differences in sex or age between the groups. Female patients predominated in both groups, and while the mean age tended to be higher in the OCA group, the difference was not statistically significant.
Regarding sonographic findings, K-TIRADS did not effectively distinguish OCA from OA, consistent with findings from previous studies [1416]. Although composition and margin have been considered key indicators of malignancy in prior research, our study did not find a significant difference in these features [1419]. However, when using ACR TI-RADS, OCA cases were more frequently categorized into a higher TIRADS category. The scoring criteria in ACR-TIRADS, which include distinctions between hypoechoic and very hypoechoic nodules, as well as assign equal points to lobulated and irregular margins, may have influenced these results. Additionally, trabecular formation, resembling a spoke wheel-like appearance, has been associated with increased malignancy risk in follicular neoplasms [20]. In our study, trabecular formation was more frequently observed in OCA cases, suggesting its potential as an additional diagnostic feature, although interpretation remains limited due to the small sample size.
Previous studies have reported that most oncocytic neoplasms were diagnosed as Bethesda category IV preoperatively, with no significant difference between OCAs and OAs [1416]. A study by Santana et al. [16] (similar to our study) subcategorized oncocytic neoplasms within Bethesda category IV but did not identify a clear distinction between benign and malignant groups. However, our findings revealed that patients diagnosed with follicular neoplasm on FNA had a higher probability of malignancy, whereas those diagnosed with oncocytic neoplasm were more likely to have benign lesions. These findings suggest that cytologic subcategorization within Bethesda category IV may provide supplementary diagnostic value in differentiating OCA from OA. However, as our study was restricted to OCA and OA cases, broader investigations including a wider spectrum of Bethesda IV nodules are needed to validate these observations. Additional research on this subject is currently in progress.
OCA has often been regarded as more aggressive than other DTCs and associated with a poorer prognosis [34]. However, our study suggests less aggressive behavior of OCAs compared to previous reports [2122]. This may be explained by the low proportion of widely invasive carcinoma in our cohort, which has been identified in prior studies as a key indicator of aggressive behavior [2123]. Jin et al. [23], in a multicenter study conducted in South Korea, similarly reported a favorable prognosis for OCAs. In their analysis, 3.1% of cases showed gross ETE, 3.1% demonstrated extensive vascular invasion, and 3.1% exhibited distant metastasis, findings that closely align with our results. Due to the small sample size, we were unable to evaluate the prognostic factors of OCAs in this study. Nevertheless, our findings add to the growing evidence suggesting that minimally invasive OCAs may exhibit a relatively favorable prognosis when compared to widely invasive cases.
This study has several limitations. First, it was conducted as a single-center study on a disease with a low incidence rate, resulting in a relatively small sample size, particularly for OCA cases. Additionally, data collection over a decade may have contributed to heterogeneity in diagnosis and treatment approaches. Second, our study design, which included only patients with confirmed OCA or OA, deviates from the standard diagnostic process. This approach may introduce a selection bias, as it excludes cases where OCA may have been misdiagnosed preoperatively. Also, our findings could either overestimate or underestimate the actual diagnostic performance of ultrasonography and FNA in distinguishing OCA from OA. Third, this study did not include a comparison of OCA with other similar neoplasms, such as follicular neoplasms, which could provide further insight into differential diagnosis.
In summary, predicting malignancy before surgery remains challenging for oncocytic neoplasms. However, in our study, larger tumor size was associated with malignancy, and higher ACR TI-RADS scores and the presence of trabecular formation may provide additional diagnostic clues although ultrasonography alone is insufficient for definitive differentiation. Additionally, nodules diagnosed as oncocytic neoplasm on FNA were more likely to be benign, while those diagnosed as follicular neoplasm had a higher probability of malignancy. Our findings highlight the favorable prognosis of minimally invasive OCAs while underscoring the importance of identifying widely invasive features, which are linked to poorer outcomes. Large-scale, multicenter studies are necessary to further clarify the differences between OCA and OA and to improve preoperative diagnostic accuracy.
XML Download