Search

Radioiodine (RAI) therapy has been used to treat thyroid diseases for around 80 years, and yet it is only relatively recently that we are beginning to manipulate its use, as we understand more of the cellular complexities which govern its success. From the benign nature of hyperthyroidism to malignant thyroid carcinomas and their metastases, RAI has profoundly changed the management of thyroid disorders. However, the complex journey which has elicited this simple therapy is worth exploring.

The sodium iodide symporter (NIS) functions to transport iodide and is critical for successful radioiodide ablation of cancer cells. Approaches to bolster NIS function and diminish recurrence post-radioiodide therapy are impeded by oncogenic pathways that suppress NIS, as well as the inherent complexity of NIS regulation. Here, we utilize NIS in high-throughput drug screening and undertake rigorous evaluation of lead compounds to identify and target key processes underpinning NIS function. We find that multiple proteostasis pathways, including proteasomal degradation and autophagy, are central to the cellular processing of NIS. Utilizing inhibitors targeting distinct molecular processes, we pinpoint combinatorial drug strategies giving robust >5-fold increases in radioiodide uptake. We also reveal significant dysregulation of core proteostasis genes in human tumors, identifying a 13-gene risk score classifier as an independent predictor of recurrence in radioiodide-treated patients. We thus propose and discuss a model for targetable steps of intracellular processing of NIS function.

New combinatorial drug strategies are urgently needed to improve radioiodide (RAI) uptake and efficiently ablate thyroid cancer cells, thereby addressing recurrent and metastatic disease. Cellular iodide uptake is accomplished solely by the sodium iodide symporter (NIS), but the complexity of NIS functional regulation and a lack of amenable high-throughput screening assays has impeded progress. We utilised mutated yellow fluorescent protein (YFP) as a surrogate biosensor of intracellular iodide for ∼1200 FDA-approved drugs, allowing us to appraise the impact of 73 leading compounds at 10 doses on 125I uptake in thyroid cancer cell lines. Subsequent mechanistic analysis suggests three predominant modes of drug action: Firstly, a number of drugs inhibited specific regulation of NIS function by the protein VCP. Secondly, some drugs enhanced transcriptional or post-transcriptional regulation of NIS expression. Thirdly, several drugs strongly implicated proteasomal degradation and the unfolded protein response in the cellular processing of NIS. Exploiting these mechanistic insights, multiple compounds gave striking increases in radioiodide uptake when combined with the drug SAHA. Importantly, our new drug combination strategies were also effective in human primary thyrocytes, suggesting they target endogenous NIS physiology. In patients with papillary thyroid cancer, genes involved in proteostasis were remarkably altered and predicted significantly worse outcome, but only in those patients who received RAI therapy. Collectively, we therefore propose a new model of intracellular NIS processing, and identify key nodes which may now be druggable in patients with aggressive thyroid cancer.SUMMARYOur data identify FDA-approved drugs that enhance radioiodide uptake outside of the canonical pathways of NIS processing, leading to a new mechanistic understanding of endogenous NIS function which is subverted in cancer.

For >75 years radioiodide (RAI) has been used to target metastases and safely and specifically destroy remaining thyroid cancer cells post-surgery. Despite this, activity of the sodium iodide symporter (NIS) – the sole conduit for cellular iodide uptake – is diminished in 25-50% of thyroid cancers, limiting adequate RAI uptake for effective therapeutic ablation. Thus, identifying targetable processes that govern NIS function is urgently needed to enhance RAI uptake and diminish recurrent thyroid cancer. Here we utilized the mutated yellow fluorescent protein (YFP) as a surrogate biosensor of intracellular iodide and screened 1200 drugs (95% FDA approved), allowing us to identify putative candidate drugs which increased iodide uptake. Categorization revealed a high proportion of drugs that modulate the proteostasis network (20/50 top candidate drugs; 40%), including key processes in protein homeostasis such as endoplasmic reticulum-associated protein degradation (ERAD) and autophagy. Secondary screening validated the activity of proteostasis modulators in enhancing iodide uptake after ranking 73 compounds based on their pharmacologic properties (AUC, EC50) and specificity of response (NIS+ve vs NIS-ve YFP-thyroid cells) at ten different drug doses (0.1-50 uM). Dose-dependent increases in 125I uptake were apparent across multiple cancer cell models, as well as human primary thyrocytes, implying that proteostatic and related pathways are central to the innate control of NIS function. Subsequent mechanistic insight allowed us to evolve entirely novel combinatorial drug strategies, routinely leading to robust and significant > 5-fold increases in RAI uptake (all p < 0.001). Appraisal of TCGA to study the clinical relevance of proteostasis modulators identified significant dysregulation of 13 core proteostasis genes linked to recurrence in RAI-treated papillary thyroid cancer (PTC) compared to non-recurrent controls (all p < 0.05). Critically, a predictive risk model based on these 13 genes showed that RAI-treated PTC patients at high risk had a significantly worse prognosis than those at low risk [Hazard Ratio (HR) = 35.87, 95% CI 4.81-267.41; p < 0.001; n =137]. By comparison, there was no difference in the prognosis of non-RAI treated PTC patients stratified into risk groups. Moreover, after controlling for age, gender, disease stage, tumor stage and node status, multivariate analysis showed that the 13 proteostasis gene risk score classifier was the sole independent predictive factor for thyroid cancer recurrence (HR = 4.55, 95% CI 2.38-8.70; p < 0.001) in the entire TCGA cohort (n = 438). Whilst oncogene activation can suppress NIS expression and function, our study has identified new non-canonical pathways that govern radioiodide uptake. Collectively, we therefore propose a new model for the targetable steps of intracellular processing of NIS, with translatable potential to address the current lack of clinical options for patients treated with RAI who typically have poorer clinical outcomes. Citation Format: Martin L. Read, Katie Brookes, Caitlin E.M. Thornton, Hannah R. Nieto, Alice Fletcher, Patricia Borges de Souza, Mohammed Alshahrani, Ling Zha, Jamie R.M. Webster, Luke J. Alderwick, Kristien Boelaert, Vicki E. Smith, Christopher J. McCabe. Exploiting non-canonical pathways of NIS regulation to enhance radioiodide uptake and identify predictive markers of recurrence in thyroid cancer [abstract]. In: Proceedings of the AACR Virtual Special Conference on Radiation Science and Medicine; 2021 Mar 2-3. Philadelphia (PA): AACR; Clin Cancer Res 2021;27(8_Suppl):Abstract nr PO-006.

Extracellular fragments derived from plasma membrane receptors can play relevant roles in the development/progression of tumor pathologies, thereby offering novel diagnostic or therapeutic opportunities. The truncated variant of somatostatin receptor subtype-5, SST5TMD4, is an aberrantly spliced receptor with 4 transmembrane domains, highly overexpressed in several tumor types, whose C-terminal tail is exposed towards the extracellular matrix, and could therefore be the substrate for proteolytic enzymes. In silico analysis implemented herein predicted 2 possible cleavage sites for metalloproteases MMP2, 9, 14, and 16 in its sequence, which could generate 3 releasable peptides. Of note, expression of those MMPs was directly correlated with SST5TMD4 in several cancer-derived cell lines (ie neuroendocrine tumors and prostate, breast, and liver cancers). Moreover, incubation with SST5TMD4-derived peptides enhanced malignancy features in all cancer cell types tested (ie proliferation, migration, etc.) and blunted the antiproliferative response to somatostatin in QGP-1 cells, acting probably through PI3K/AKT and/or MEK/ERK signaling pathways and the modulation of key cancer-associated genes (eg MMPs, MKI67, ACTR2/3, CD24/44). These results suggest that SST5TMD4-derived peptides could contribute to the strong oncogenic role of SST5TMD4 observed in multiple tumor pathologies, and, therefore, represent potential candidates to identify novel diagnostic, prognostic, or therapeutic targets in cancer.