• whole tumor capsule

Background: With the dramatic increase of the incidence of thyroid cancer, follicular variant of papillary thyroid cancer (FVPTC) has been the thyroid cancer variant with the highest rate of increase in recent years. Despite its high incidence, the malignant biological behavior of FVPTC, and in particular its degree of aggressiveness, still remains controversial. Moreover, a subgroup of FVPTC, those with a whole tumor capsule without vascular and capsular invasion has been declassified; these tumors are now classified as non-cancerous and called noninvasive follicular thyroid neoplasm with papillary-like nuclear features with the acronym of (NIFTP) because of their very good outcome. The rationale of the project lies in the evidence that the presence of an intact whole tumoral capsule identifies a subgroup of cases with an indolent course particularly in FVPTC but no studies have verified if it was true also in classic variant of papillary thyroid cancer (CVPTC), so far. However, all previous data are mainly obtained in retrospective studies and, in many cases, in series of patients treated not only by surgery but also with radioiodine thyroid ablation therapy.
Aim of the research project: the first aim of this research project was to retrospectively evaluate the epidemiological, clinical and pathological data of FVPTC and CVPTC in order to shed some light on the most important prognostic factors for the outcome and to verify if the presence of a whole tumor capsule could have a similar pathological and clinical significance also in CVPTC. The second aim of this research project was to prospectively evaluate the clinical behavior of cases with either NIFTP or En-CVPTC to verify the possibility to have cases of En-CVPTC with an indolent course similar to the NIFTP. The other objective of the study was to verify a correlation between the presence of a whole tumor capsule and the molecular profile of these tumors.
Methods: in the first part, epidemiological, clinical, pathological and molecular data of 451 FVPTC and 439 CVPTC patients surgically treated from 1999 to 2004 were collected. Cases were all revised by two independent pathologists and an accurate analysis of the tumor capsule was performed in order to distinguish those cases with a whole tumor capsule those with capsular invasion. The presence of a whole tumor capsule was defined as the presence of a distinctive fibrous capsule around the lesion without evidence of capsular and/or vascular invasion. According to these criteria, cases with an intact whole tumor capsule (without invasion) were named En-FVPTC and En-CVPTC; cases with a tumor capsule with invasion were named Ei-FVPTC and Ei-CVPTC; cases without tumor capsule were named I-FVPTC and I-CVPTC. The patients were followed-up at the Endocrine Unit of the University of Pisa with periodic clinical and biochemical control for a long follow-up period. All patients underwent total thyroidectomy and, except the cases with unifocal microcarcinoma, performed the radioiodine thyroid remnant ablation. In order to increase the number of FVPTC and CVPTC with a whole tumor capsule, in the second part of the project we extended the selection period thus included a total of 1154 consecutive patients (600 FVPTC and 554 CVPTC), underwent total thyroidectomy between 1999 and 2006. In a sub-group of each histological variant, we also analyzed BRAF gene alterations using the direct Sanger sequencing. Simultaneously with the retrospective study, we prospectively collected data of patients treated only by surgery for En-CVPTC or NIFTP. In both cases the tumor was accurately analyzed by the pathologists according to the criteria used for the NIFTP. All patients performed at least a clinical control and neck ultrasound within 6 months from surgery and then every 6-12 months thereafter. To eliminate possible bias due to a different distribution of mutations according to age, sex and tumor size, we have analyzed the molecular profile of two matched subgroups and a total of 27 case-control pairs (En-CVPTC and NIFTP) were identified. The DNA of 27 case-control pairs was sequenced and run on the ION S5 platform.
Results: Several significant differences were found between FVPTC and CVPTC such as the mean age at diagnosis, the presence of tumor capsule, the presence of thyroid capsule invasion, the presence of peri-thyroid soft-tissue invasion, the lymph node metastases, the multifocality and bilaterality. At the end of follow-up only 9% (77/879) patients were not cured. However, a statistically significant lower percentage of persistent disease was found in the FVPTC than in the CVPTC group (3% vs 14.5%, respectively, p<0.0001). In multivariate analysis, the absence of the tumor capsule (OR=6.75) or its invasion (OR=7.89), the tumor size ≥4 cm (OR=4.29), the variant CVPTC (OR=3.35) and the presence of lymph node metastases (OR=3.16) were all independent risk factors for the persistence of the disease. A statistically significant higher prevalence of encapsulated cases in FVPTC respect to CVPTC (p<0.0001) was found. When we compared the epidemiological, clinical pathological features of En-FVPTC vs En-CVPTC and Ei-FVPTC vs Ei-CVPTC we did not find any statistically significant difference with the exception of a greater number of not cured cases in the Ei-CVPTC respect to Ei-FVPTC at the end of a rather long follow up (median 18 years; mean 17 ± 2 years, range 15-20). At variance, no difference was observed in the outcome of patients with En-FVPTC and En-CVPTC. Among the entire group of PTC, 614 cases (384 FVPTC and 230 CVPTC cases) were assessed for BRAF molecular status. 187 of 614 PTCs (30.4%) carried a BRAF alteration. We found BRAFV600E in 177 cases (28.8%) and BRAF rare alterations in 10 cases (1.6%). All these rare variants were found in the group of FVPTC. According to the histological variant, BRAFV600E mutation was present in 116/230 (50.4%) CVPTC and in 61/384 (15.9%) FVPTC. According to the presence or absence of the tumor capsule or its invasion we found that BRAFV600E mutation was present in 37/244 (15.2 %) En-PTC, in 7/42 (16.7 %) Ei-PTC cases and in 133/328 (40.5 %) I-PTC (p <0.0001). In particular, we found a significantly higher rate of mutated cases in En-CVPTC (22/49, 44.9%) than in En-FVPTC (15/195, 7.7%) (p<0.0001). Similarly, a significantly higher rate of mutated cases was found in Ei-CVPTC (4/4, 100%) than in Ei-FVPTC (3/38, 7.9%) (p=0.0003) as well as in I-CVPTC (90/177, 50.8%) respect to I-FVPTC (43/151, 28.5%) (p<0.0001). The multivariate clustering analysis confirmed that the En-FVPTC, the En-CVPTC and the Ei-FVPTC have similar clinical and pathological features which are significantly different from I-FVPTC and I-CVPTC and, to a lesser extent, also from Ei-CVPTC.
From January 2018 to December 2019, a total of 167 En-CVPTC and 236 NIFTP were prospectively selected. Eighty-eight/236 (37.3%) NIFTP and 67/167 (40.1%) En-CVPTC were eliminated due to the presence of exclusion criteria. The remaining 148/236 (62.7%) NIFTP cases and 100/167 (59.8%) En-CVPTC cases were included in the study. No clinical, epidemiological and pathological features between NIFTP and En-CVPTC case was found. A significantly higher rate of NIFTP patients was submitted to lobectomy respect to the En-CVPTC patients, more frequently submitted to total thyroidectomy (p=0.0003). From a pathological point of view, no cases of vascular invasion, intra-thyroidal and extra-thyroidal extension were found in both groups. The only relevant difference between the two groups was the finding that NIFTP showed a significantly bigger tumor size than En-CVPTC (23.5±16 mm vs 8.5±11 mm, p<0.0001). After a median follow-up of 15 months all NIFTP and En-CVPTC patients (only surgically treated) had an excellent response. Twelve/27 (44.4%) NIFTP and 9/27 (33.3%) En-CVPTC harbored one or more genetic alteration. Nine/27 (33.3%) NIFTP were mutated in the RAS genes, particularly NRAS and HRAS genes; one/27 (3.7%) NIFTP harbored a rare BRAF gene alteration (BRAFK601E substitution). No NIFTP case harbored BRAFV600E substitution. Two/27 (7.4%) NIFTP cases presented mutations in other genes (i.e., TSHR M453T and PTEN P190fs, c.566_567insG). Two/27 (7.4%) En-CVPTC harbored RAS gene mutations, particularly NRAS and KRAS genes. Five/27 (18.5%) were mutated in BRAF gene: 3 harbored a classical BRAFV600E mutation; the other two cases harbored rare BRAF mutations such as an in-frame deletion of a triplet c.1799_1801delTGA (p.V600_K601delinsE). Two/27 (7.4%) En-CVPTC cases had mutations in other genes (i.e., TSHR Ile568Phe and STK11 p.Ser19Leu). No gene fusions targeted in our panel were detected in both groups. The three En-CVPTC BRAFV600E mutated were further tested with the screening protocol to rule out the classical BRAF mutation commonly used in our Hospital for the histological diagnosis of NIFTP: all the three cases showed a strong positivity at the immunohistochemistry for BRAF.
Conclusions: Despite an overall excellent prognosis of both variants, a higher percentage of CVPTC than FVPTC patients had a persistent disease. The absence of tumor capsule or its invasion, the tumor size ≥4 cm and the presence of lymph node metastases are other prognostic factors for the persistence of the disease. In contrast, the presence of a whole tumor capsule is the only good prognostic factor for their outcome. En-CVPTC are less prevalent respect to the En-FVPTC, in particular if we exclude the BRAFV600E mutated cases. Nevertheless, they have a clinical and pathological behavior similar to that of En-FVPTC, that nowadays is NIFTP per new nomenclature. This finding confirms our hypothesis about the good prognostic role of the presence of an intact whole tumor capsule also in CVPTCs. Even if after a short follow-up period, we can conclude that En-CVPTC patients, prospectively collected and treated conservatively without radioiodine ablation therapy, have an indolent behavior and favorable outcomes. Although further studies are warranted on NIFTPs and En-CVPTC, the molecular landscape that we have found seems to indicate that the indolent histopathologic appearance of these neoplasms corresponds to a low-risk molecular profile. If a prospective multicentric study with a longer follow-up will confirm these findings, probably in the future we could consider En-CVPTC as a “pre-malignant lesion” as like as NIFTP.