Impact of clinicopathological and molecular features of radioiodine-refractory thyroid cancer patients treated with multikinase inhibitors on their outcome

Introduction. Differentiated thyroid carcinoma (DTC) generally has an excellent prognosis after surgery and radioiodine treatment, with a survival rate of up to 96.4% at 35 years. However, a small percentage of DTC patients (5-10%) develop metastatic disease, with a 10-year survival rate of 50%. The prognosis worsens further when metastatic disease becomes refractory to radioactive iodine therapy. Although most radioiodine-refractory metastatic tumors (RAIR-TC) progress slowly, a subset of patients experiences rapid disease progression. For these patients, the primary challenge is identifying the ideal characteristics and the right time to start systemic therapy with multikinase inhibitors (MKI). Aims. The first aim of the study was to identify any epidemiological, clinical, and pathological factors in patients with RAIR-TC that could predict the need for systemic therapy with MKI. Specifically, this study aimed to evaluate whether there are prognostic factors for disease progression, either at the initial diagnosis or during follow-up, particularly at the time of the diagnosis of radioiodine-refractoriness, that can predict the need to initiate MKI treatment. The second aim was to identify a molecular signature of RAIR-TCaddressed to system therapy with MKI and evaluate whether the molecular profile could be associated with different responses to MKI. Patients and methods. For the first part of the study, clinical data about 279 RAIR-TC patients were evaluated. For the second part, 49 of these cases that were surgically treated and followed up at the University Hospital of Pisa and that performed lenvatinib as first-line therapy were selected. Nucleic acids were extracted from formaline-fixed paraffin-embedded tissues and analyzed using Next Generation Sequecing approach. TERT promoter mutations were assessed via digital PCR. Response to lenvatinib was assessed by CT scans according to RECIST. Results. Part 1.Ninety-nine patients received indication to MKI (Group A), while 180 remained under active surveillance (Group B). Group A had greater tumor size, more aggressive histotype, more frequent macroscopic extrathyroidal extension, distant metastases, advanced AJCC stage, and higher ATA risk of recurrence. After radioiodine-refractoriness (RAIR) diagnosis, 93.9% of Group A had progression of disease (PD) after which MKI therapy was started. The remaining 6.1% of patients had a so severe disease at the time of RAIR diagnosis that MKI therapy was immediately started. Among patients of Group B, 42.7% had PD, but the majority underwent local treatments. The mean time from RAIR diagnosis to the first PD was shorter in Group A, and the evidence of PD within 25 months from RAIR diagnosis was associated with the decision to start MKI. Part 2.Molecular alterations, one or more than one, were detected in most of the cases (n=39/49; 79.6%), while in the remaining 10/49 (20.4%) patients, no known mutations were identified according to our custom panel. Among all patients, 10/49 (20.4%) cases harbored a single driver mutation being BRAF (n=3; 6.1%), RAS (n=6; 12.2%) and RET/PTC rearrangement (n=1; 2%). In 22/49 (44.9%) cases a driver mutation coexisted with an additional mutation (TERT or TP53). Seven cases (14.3%) were negative for driver mutation but carried TERT alone in 6/49 (12.2%) cases and TP53 alone in 1/49 (2%) case.The median overall survival (OS) was 9.5 years(IQR 6.2-15.2; range 0.6-21.5), while median progression-free survival (PFS) was 9.5 months(IQR 3.75-21; range 1-62). Patients with mutations showed significantly longer PFS (HR 0.34; CI95% 0.16-0.75; p=0.007) but no difference in OS (p=0.16) compared to those without any mutations. Dividing the cohort by presence or absence of BRAF mutation and presence or absence of RAS mutations showed no significant differences in PFS or OS, though there was a trend favoring BRAF-mutated cases. When dividing the cohort in 3 groups (Group 1: cases with a single driver mutation, either BRAF or RAS or RET/PTC rearrangements; Group 2: cases with a driver associated with a second mutation, either TERT or P53; Group 3:absence of driver mutations with or without additional mutations, either TERT or P53), patients of group 1 showed a longer PFS compared to patients of group 3 [HR 3.095 (CI95% 1.06-8.99); p=0.04], while no significant differences were found between group 1 and 2 (p=0.13). Regarding the OS, we observed that patients of group 1 had a significantly longer OS compared to patients of group 2 [HR 3.552 (CI95% 1.17-10.79); p=0.0026] and compared to group 3 [HR 3.516 (CI95% 1.10-11.21); p=0.0034]. At univariate analysis only histotype and frequence of lymph node metastases were different among the three groups, while there were no significant differences for age, sex, distant metastases and stage. Multivariate analysis, including histotype (PTC vs not-PTC) and molecular profile (Group 1, 2, and 3), confirmed that lacking a driver mutation (Group 3) was a prognostic factor for a shorter PFS independently from the histotype [OR 3.11 (95CI% 1.06-9.13); p=0.039], while both absence of a driver mutation and presence of TERT or P53 with a driver mutation (Groups 2 and 3) were independently associated with worse OS [Group 2: OR 3.53 (95CI% 1.16-10.73) p=0.026; Group 3: OR 3.41 (95CI% 1.06-10.98) p=0.04]. Lastly, categorizing mutations as "ras-like" or "braf-like" showed no significant differences in PFS (p=0.077) or OS (p=0.713), although there was a slight trend favoring "braf-like" cases for PFS. Conclusions. We identified several clinical features at initial diagnosis that were more frequent in patients who started MKI therapy compared to those with RAIR-TC under active surveillance: larger tumor size, more aggressive histology, more frequent macroscopic extrathyroidal extension, distant metastases, advanced AJCC stage, and a higher ATA risk of recurrence. Additionally, we suggest that in RAIR-TC patients, the rate of disease progression should guide follow-up. Less-aggressive cases with slow progression may not require frequent imaging or strict monitoring. In contrast, patients diagnosed at an advanced stage or who experience their first progression within 25 months of RAIR-TC diagnosis should have closer follow-up to ensure timely initiation of MKI. Furthermore, the study highlights the complex genetic landscape of advanced RAIR-TC and its implications for lenvatinib treatment. BRAF and TERT promoter mutations were prevalent in our cohort, and their coexistence was associated with poorer prognosis. The presence of driver mutations may enhance the efficacy of lenvatinib, possibly due to hyperactivation of the MAPK pathway. Although tumor heterogeneity presents some challenges, these findings underline the importance of comprehensive molecular profiling to better inform treatment strategies for patients with advanced thyroid cancer.