Is Poorly Differentiated Thyroid Cancer Poorly Characterized?

Poorly differentiated thyroid cancer (PDTC) has remained a controversial diagnosis. Some clinicians refer to PDTC as thyroid cancers that do not trap radioiodine on radioiodine scans (either initially or during monitoring), and others believe that it is appropriate to refer to rapidly growing thyroid cancers that have apparently not responded to radioactive iodine therapy (1). However, in many instances it is difficult to determine whether radioiodine therapy has been partially effective. We and others had noted that “poorly differentiated thyroid carcinoma is a concept proposed to include carcinomas of follicular thyroid epithelium that retain sufficient differentiation to produce scattered small follicular structures and some thyroglobulin, but generally lack the usual morphologic characteristics of papillary and follicular carcinoma” (2–4). In separate reports, Sakamoto et al (3) in 1983 and Carcangiu et al (5) in 1984 initially pathologically described the discrete category of PDTC. However, these authors used different characteristics to define this entity, resulting in a long-standing dispute regarding the appropriate pathological definition of PDTC. The World Health Organization tumor classification attempted to better define this entity, but this also has been considered controversial (6, 7). In an effort to quell the controversy, a group of established pathologists and investigators set forth what is referred to as the Turin proposal. This proposal set forth specific histological characteristics to attempt to universally define PDTC (7). These criteria include: 1) the presence of a solid/trabecular/insular pattern of growth; 2) the absence of the conventional nuclear features of papillary carcinoma; and 3) the presence of at least one of the following features: convoluted nuclei, mitotic activity ≥ 3 per 10 high-power fields, and tumor necrosis (7). However, PDTC remains a clinically and pathologically inchoate disorder, and it is not known whether it represents a single clinical or molecular entity. Furthermore, the optimal clinical approach for monitoring and treatment modalities is unknown. Recent articles on the topic of PDTC have assessed and commented upon the diagnostic criteria and have reviewed this entity (8–13). These articles emphasize the variations in the diagnostic criteria and that PDTC is generally considered in the middle of the spectrum between well-differentiated thyroid cancer and undifferentiated or anaplastic thyroid cancer. Asioli et al (14) reported that poorly differentiated carcinoma had a prevalence (in pathological thyroid malignancies) of 1.8% (56 of 3128) in the Unites States (at Mayo Clinic) and of 6.7% (96 of 1442) in northern Italy (University of Turin). Thyroglobulin immunostaining was positive in 92% (95 of 103) of PDTC cases, and thyroid transcription factor 1 immunostaining was positive in each of 103 PDTC cases. The authors concluded, using univariate analysis, that PDTC was associated with a greater risk of death in patients age 45 or greater and for tumors of 4 cm or larger; multivariate analysis confirmed the age-related finding. The present study by Ibrahimpasic et al (15) from the well-respected Memorial Sloan Kettering group attempts to better define the prognosis of PDTC patients. They retrospectively analyzed 91 patients with PDTC treated at their institution between the years 1986 and 2009. Over a median of 50 months of monitoring, the 5-year overall survival was 62%, and disease-specific survival was 66%. The vast majority of disease-specific deaths were related to the presence of distant, metastatic disease (23 of 27 deaths; 85%). Locoregional control was higher at 81% than was distant control at 59%. Multivariate analysis showed that initial pT4a stage distant metastases (M1) were independently associated with poorer disease-specific survival, whereas univariate analysis correlated age 45 or greater, tumor size larger than 4 cm, extrathyroidal extension, higher stage, positive margins, and distant metastases (M1) with worse disease-specific survival. The 5-year disease-specific survival for PDTC pT1/2 was 94%, for pT3 was 75%, and for pT4 was 45%; for MO disease it was 85% and for M1 disease, 34%. Gnemmi et al (16) observed approximately a 72% 5-year cause-specific survival for PDTC patients, whereas the cause-specific survival was approximately 98% for well-differentiated thyroid carcinoma. The study by Ibrahimpasic et al (15) does have several additional issues that require discussion; it was retrospective with the usual inherent problems. Patients were treated with a wide variety of therapies, including surgery, radioactive iodine, external beam radiotherapy, and sometimes chemotherapy. Therefore, it is difficult to determine which particular therapy is most appropriate. The authors recommend a total thyroidectomy and central node dissection, as appropriate, in patients with PDTC. If there is evidence of disease, a lateral neck dissection is also recommended. Empirically, the authors recommend the use of radioactive iodine therapy for patients with distant metastases that are radioactive iodine avid. Of course, this implies that a radioactive scan must be performed. Also, the authors recommend the use of external radiation therapy directed at local residual disease if the surgical team suspects residual disease or if there are positive margins on histological examination. These recommendations are empirically based but seem appropriate until more definitive studies are performed. None of the patients in this study were treated with a tyrosine kinase inhibitor, but presently, this treatment modality would be considered. Probably most interesting is that the authors (15) did not strictly follow the Turin proposal to make the histological diagnosis of PDTC. Instead, they defined PDTC “by histological and/or immunohistochemical evidence of follicular cell differentiation and presence of tumor necrosis and/or ≥ five mitoses per 10 high-power fields (×400),” as reported in their previous publication (17). Gnemmi et al (16) did compare the Memorial Sloan Kettering diagnostic criteria to that of the Turin proposal and suggested that they were comparable with regard to classifying patients into two different prognosis groups, and each set of diagnostic criteria resulted in comparable prognostic information. Gnemmi et al (16) note that the Memorial Sloan Kettering diagnostic criteria represent a simple and reliable method allowing detection of tumors with worrisome features. The overall conclusion was that both methods provided comparable ability to predict prognosis, and they were complementary in defining a group of thyroid cancer patients with intermediate prognosis. Gnemmi et al (16) did suggest that carcinomas recognized by the Memorial Sloan Kettering definition as PDTC only on the basis of high-grade features should be acknowledged as carcinomas “with high-grade features,” rather than PDTC and, also, because vascular invasion is considered an important prognostic characteristic, this classification system should be used cautiously in patients with encapsulated carcinoma. It is unknown how reproducible any of these diagnostic criteria for PDTC would be among a large group of experienced pathologists, and further, how many of those tumors would be classified as PDTC using the Turin criteria. The two experienced pathologists involved in the present study (15) identified a total of 98 cases of PDTC with a concordance rate of 78%. Indeed, the authors themselves did not confirm the histological diagnosis at their own institution in 12 of 69 patients, with seven patients being reclassified as differentiated thyroid cancer, three as anaplastic thyroid cancer, and two as metastatic lung and adenoid cystic carcinoma to the thyroid gland. The authors astutely also reviewed the histological samples of all patients who died from well-differentiated thyroid cancer (n = 71) during the 1986–2009 time period and found an additional 36 patients who actually had PDTC. Of note, the authors found only 91 cases of PDTC out of a review of 3493 papillary thyroid cancers reviewed. This comprehensive histological review emphasizes the relatively high mortality rate of PDTC in a group of patients with papillary thyroid cancer. The Ibrahimpasic et al study (15) advances our understanding of PDTC and suggests that its prognosis stands, as expected, between well-differentiated and anaplastic thyroid cancer. Previous studies have suggested the presence of similar somatic mutations in PDTC as in both undifferentiated and well-differentiated thyroid cancer, although the frequencies of these mutations may vary. Santoro et al (18) identified RET/PTC rearrangement in eight of 62 (12.9%) PDTCs. Soares et al (19) studied 19 thyroid samples with PDTC (defined as insular or insular-like tumors) and did not identify any BRAF mutations. Nikiforova et al (20) identified BRAF mutations in 38% of well-differentiated papillary thyroid cancer and 13% of PDTC, and Donghi et al (21) found a p53 mutation in 25% of PDTC. Landa et al (22) identified somatic TERT (telomerase reverse transcriptase) promoter mutations in 40 of 78 (51%) advanced thyroid cancers (including anaplastic thyroid cancer and PDTC); these mutations frequently coexisted with a BRAF or RAS mutation. Future studies should be performed to assess the somatic molecular changes and molecular pathways involved in PDTC. The question arises as well whether medications such as selumetinib may enhance the avidity of PDTC tumors for radioactive therapy (23). Ho et al (23) analyzed 20 patients, seven of whom had pathologically defined PDTC (35%). Eight patients (of the 20) were treated with radioiodine, five had partial tumor responses, and three had stable disease; every treated patient had a decrease in serum thyroglobulin level. The American Thyroid Association Guidelines note the poorer prognosis of patients with PDTC and recommend consideration of radiological studies including a computed tomography/positron emission tomography scan for initial staging and monitoring of patients with PDTC because they typically trap radioactive iodine ineffectively (1). Furthermore, their ability to synthesize and secrete thyroglobulin is impaired, and typically the amount of disease noted on cross-sectional imaging is discordant to the serum thyroglobulin level. In summary, the article by Ibrahimpasic et al (15) is an important contribution because it assesses the disease-specific survival and evaluates prognostic features in a large group of patients with PDTC. The authors assiduously reviewed their institution's experience with PDTC. This clinically relevant study lays the foundation for future prospective, controlled studies assessing various treatment modalities, including the effectiveness of tyrosine kinase inhibitors. The histological diagnosis of PDTC varies between institutions, and it would be reasonable to attempt to reformulate a pathological consensus so patient care and clinical studies could be more uniformly interpreted. As noted in the title of this editorial, the question has been raised whether PDTC is poorly characterized. It would seem that the pathological criteria need further consensus, and there should be additional analysis of the clinical characteristics and molecular mechanisms involved in the pathogenesis of PDTC. Nevertheless, the study by Ibrahimpasic et al (15) is a welcome addition that advances our understanding and characterization of PDTC. The author thanks Dr Yuri Nikiforov for his review of the manuscript and helpful comments. Disclosure Summary: The author has nothing to declare. poorly differentiated thyroid cancer.