Dissertation committee:
Γρηγόριος Καλτσάς, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ, Επιβλέπων
Μιχαήλ Βουλγαρέλης, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Νικόλαος Σύψας, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Ευανθία Κασσή, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Ευσταθία Καψογεώργου, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Αθανάσιος Πρωτογέρου, Αναπληρωτής Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Σταύρος Σουγιουλτζής, Αναπληρωτής Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Summary:
Thyroid cancer is the most common endocrine malignancy, and it can be categorized into distinct histologic variants: differentiated, medullary and anaplastic, each of which confers a different prognosis. Anaplastic thyroid cancer has the worst prognosis, while differentiated has (usually) an excellent prognosis. Thyroid cancer incidence is constantly rising, but so are the more aggressive cancers. Hence there is a clear need for discovering new treatment modalities.
Metformin is the most commonly prescribed anti-diabetic medication, and there have been multiple studies published in the last decade regarding its’ anticancer activity. Several retrospective studies have reported that diabetic patients on Metformin have better cancer outcomes than diabetic patients not on Metformin. In vitro studies have shown that Metformin’s anticancer activity is mainly because of the increased experession of AMPK, which is the master energy regulator of the cell, as well as the inhibition of mTOR. Preliminary studies in our lab showed that Metformin was also effective against thyroid cancer. However, as with all in vitro Metfromin studies, the doses used were very high (in the mM range) and not clinically achievable in humans (where the max concentration is 50-100 μΜ). As a result, the objective of the current study was to identify methods to increase Metformin’s efficacy in vitro.
Initially, we attempted to alter the expression of the transporter of Metformin OCT1 to increase the intracellular drug delivery. Despite detecting the transporter, no significant changes were observed in the pathologic characteristics of the patients that were examined and for that reason we did not proceed to the pharmacologic regulation of OCT1. Next step was targeting cell metabolism, and more specifically the micro-environment and the glucose concentration in the medium. The initial experiments demonstrated that Metformin in low glucose conditions 5mM (which corresponds to normoglycemia in humans – 100 mg/dl) did not only cause inhibition of cell proliferation, but also cell death. Activation of autophagy could not fully explain the phenomenon, and we went on to discover a rare type of cell death called oncosis. Oncosis is a multi-level process of cell death characterized by cellular swelling. If the process is not reverted by adding glucose to the medium, the cellular membrane disintegrates and necrosis occurs. Oncosis is caused by ATP depletion, which in turn causes activation of AMPK as shown in our study.
We next showed that glucose deprivation can augment Metformin’s efficacy significantly. We then turned to the molecular mechanism of Metformin, which is a mitochondrial complex I inhibitor. We combined Metformin with a glycolytic inhibitor (2-deoxy-glucose/2-DG) and demonstrated that the therapeutic combination can increase Metformin’s efficacy dramatically. In low glucose conditions, the therapeutic combination of 25 μΜ Metformin and 5mM 2DG caused an apoptotic cell death in thyroid cancer cells, proving that Metformin can be effective in concentrations that are clinically achievable in humans.
We then performed a screening via qRT-PCR and immunochemistry to identify the metabolic enzymes that are more heavily implicated in thyroid cancer. Pyruvate Kinase (PKM) was selected and further experiments showed that the expression of PKM2 is significantly reduced by Metformin in low glucose conditions. These results, in conjunction with immunohistochemistry results showing increased expression of PKM2 in aggressive thyroid cancers, suggest that PKM2 can potentially be used as a marker for response to the treatment with Metformin in patients with thyroid cancer.
In our Metformin experiments, we identified that the mitochondria, and more specifically cytochrome c oxidase plays an important role in the progression of thyroid cancer. Metformin can affect the mitochondria as a metabolic agent, but we also tried a more specific inhibitor of cytochrome c oxidase. Mitotane is a therapeutic agent mainly used for the treatment of adrenal carcinoma. Its’ mechanism of action is not fully understood, but recent studies have shown that its’ molecular target is cytochrome c oxidase subunit 4 (COX4). Hence, we peroceeded with Mitotane experiments.
Initially we showed that Mitotane is cytotoxic for all thyroid cancer cell lines that were examined and that derive from all histologic types. Mitotane was more effective against cells harboring the BRAFV600E mutation (BCPAP και SW1736) and cells that derive from medullary thyroid cancer (TT). Flow cytometry demonstrated that Mitotane causes apoptosis in thyroid cancer cells. The mechanism of action is probably induction of ER stress, that in turn causes DNA damage.
We then examined the mitochondrial respiratory chain, and showed that Mitotane inhibits the mitochondrial membrane potential as well as alters the expression of the enzymes of the mitochondrial chain, mainly down regulates ATP5B. Finally, through immunohistochemistry experiments we showed that ATP5B is expressed in all human thyroid cancers, but mainly in medullary cancer. These data suggest that medullary thyroid cancer relies more heavily on mitochondria, which corroborates our previous findings that TT cells were more sensitive to Mitotane. Certainly Mitotane’s toxicity can limit its’ use, but we showed that Mitotane was more toxic and specific for thyroid cancer cells than for human fibroblasts.
In conclusion, the current study demonstrates that Metformin can be used against thyroid cancer, and that it is more effective in normo- or hypo-glycemic conditions. The fact that Metformin causes an oncotic cell death and that it can be combined with other metabolic agents is certainly very promising and the next step should be clinical trials in thyroid cancer patients. Similarly, Mitotane was shown in our preclinical study to be effective against thyroid cancer cell lines, and these results could form the basis for clinical trials in patients with advanced thyroid cancer.
