Endocrine Service, Division of General Medicine, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, New York 10021
Address all correspondence and requests for reprints to: Richard J. Robbins, M.D., Endocrine Service, Box 296, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, New York 10021. E-mail: robbinsr{at}mskcc.org.
Anaplastic thyroid carcinoma (ATC) is one of the most lethal of all human cancers (1, 2). Numerous in vitro and mice xenograft studies have shown that a wide variety of chemotherapeutic agents and factors that target specific intracellular pathway mediators can reduce proliferation and lead to the death of anaplastic thyroid cancer cell lines. The most widely used ATC cell line is named ARO and was derived from a patient with ATC. Unfortunately, only rarely have any of these observations led to useful clinical benefit for ARO patients (3). Like many cell lines, ARO cells have been subcloned and passaged so many times that their relationship to ATC in a patient is highly unlikely (4).
The incidence of ATC seems to be decreasing, which may be due to increased detection of small differentiated thyroid carcinomas that, with time, could acquire additional mutations leading to the ATC phenotype. However, only a few studies support the concept that well differentiated thyroid cancers can actually change into ATC (5). It is equally likely that ATC arises de novo because of the gain or loss of a small number of unique genes, such as c-abl/p73 (6, 7), leading to tremendous proliferation.
The class of agents known as HMG CoA reductase inhibitors (also known as statins), first introduced in 1987, has been shown to safely and effectively lower serum cholesterol levels in patients with a wide variety of lipoprotein disorders. Their ability to reduce atherosclerotic diseases such as myocardial infarction and stroke, in fact, seems to be greater than their ability to lower serum cholesterol by blocking mevalonate synthesis, suggesting that they may function in other ways. Studies over the past 10 yr have found that statins can have direct antioxidant, anti-inflammatory, and anticoagulant activities. They may act on endothelial cells, smooth muscle cells, neurons, osteoclasts, and on newly forming blood vessels (8) to exert unexpected benefits for macrovascular diseases, osteoporosis, and Alzheimers disease.
This apparently wide spectrum of activity may have a common theme in the ability of statins to inhibit isoprenoid production, including lipids such as farnesyl-pyrophospate and geranylgeranyl-pyrophospate. These isoprenoids are lipid moieties that are bonded to small GTP-binding proteins, such as Ras and Rho, and serve as attachment anchors for the proper membrane orientation of these important intracellular signaling agents.
In this issue of JCEM, Wang et al. (9) report on their studies of the effects of one HMG CoA reductase inhibitor, lovastatin, on the differentiation and survival of ATC cells in vitro. They postulated that redifferentiation of ATC cells may enable reexpression of proteins such as the sodium-iodide symporter, which could render the cells susceptible to radioiodine therapy. Using the ARO cell line, they found that 25 µM lovastatin caused redifferentiation of the cells based on reappearance of microvilli and dense core secretory granules, and on increased secretion of thyroglobulin into the medium. At 5075 µM, lovastatin induced apoptosis and reduced ARO cell survival. Furthermore, they found that levels of Rho, but not Ras, were markedly reduced by lovastatin. Exposure of lovastatin-treated cells to mevalonate or geranylgeranoil (but not with farnesol) was able to restore Rho levels and reverse the apoptosis but, interestingly, had no effect on the morphological differentiation. It, therefore, seems that geranylgeranylated Rho is necessary for the full survival and proliferation of ARO cells.
The Rho family of proteins has many structural similarities with Ras-related proteins and has been implicated in cancer cell invasiveness, the ability to suppress apoptosis while metastasizing (i.e. anoikis), and angiogenesis, all of which are necessary for malignant cells to survive in a new tissue environment (10). Ras and Rho are GTPases that are active when associated with GTP, and both generally require attachment to the inner plasma membrane by isoprenoid lipid anchors (11). When activated, Rho proteins interact with a wide variety of effector proteins that mediate cytoskeletal remodeling, disruption of intercellular adhesions, and stimulation of cell motility (12). In transformed cells, Rho is frequently overexpressed. Rho proteins can stimulate cell cycle progression, in part, by increasing the levels of cyclin D1 and lowering p27, both known to be associated with thyroid cancer aggressiveness (13). Finally, Rho family members are involved in apoptosis. Rac1, a Rho-like molecule, has been shown to produce resistance to apoptosis in a variety of cell lines (14). Drugs such as lovastatin can block farnesylation or geranylgeranylation, resulting in abrogation of proper Rho or Rac membrane binding and function with loss of the transformed phenotype.
As early as 1993, lovastatin was shown to disrupt intercellular adhesions among FRTL-5 cells and to produce long cytoplasmic extensions (15). In 1999, Vitale et al. (16) found that lovastatin induced differentiation in several immortalized or transformed thyroid cell lines and reduced apoptosis that could be reversed by geranylgeranyl transferase inhibition but not by a farnesyl transferase inhibitor. This was basically the same result that Wang et al. (9) found with the ARO cell line, although that work was not cited in the their article. Li et al. (17) in 2002 also demonstrated that lovastatin-induced apoptosis was blocked by geranylgeranyl transferase inhibition but not farnesyl transferase inhibitor in human umbilical vein endothelial cells. Lovastatin has recently been shown to induce apoptosis in melanoma (18), colon cancer (19), squamous cell carcinoma (20), and myeloma cells (21) in vitro. A common theme in these studies was the action of lovastatin to reduce Rho or Ras membrane binding. In addition to its putative action of blocking the membrane anchorage of key GTPase proteins, lovastatin may also reduce proliferation of transformed cells by elevating cell cycle inhibitors p21 and p27, resulting in G1 cell cycle arrest (22).
Do the observations by Wang et al. (9) have any value in the treatment of patients with ATC? Like many investigators before them, they have shown that blockade of Rho reduces proliferation and induces apoptosis in a transformed cell line, and that it can be reversed by restoring geranylgeranylation. It would have been helpful if these investigators had followed the rationale that they developed in their Introduction, to see whether they had restored the sodium iodide symporter and the ability to concentrate radioiodine.
The history of cancer research is littered with thousands of articles documenting interesting in vitro effects that are seldom translated into the control or cure of human cancers. However, a small number of targeted therapies that have made it into clinical trails are beginning to reap rewards. One example is the ability of STI-571 to inhibit BCR-ABL kinase and cause remission of chronic myelogenous leukemia. Interestingly, STI-571 can also inhibit proliferation of ARO cells (7). Very aggressive cancers such as ATC likely have multiple gene defects and complex dysregulation of intracellular signaling pathways. Blockage of a single biochemical event, such as isoprenylation of small GTP binding proteins, could induce apoptosis in a clone of ATC cells, but is unlikely to prolong the lives of patients with ATC. Because of the small number of ATC cases, only a concerted effort to create multicentered cooperative group trials to translate novel molecular findings to patients with ATC will impact on the devastating prognosis that this disease carries.
Footnotes
Abbreviation: ATC, Anaplastic thyroid carcinoma.
Received May 12, 2003.
Accepted May 13, 2003.
References