Lately, It Occurs to Me What a Long, Strange Trip It's Been for the Farnesyltransferase Inhibitors

If you had told me 5 years ago that I would be writing a commentary on an exciting report on the prospects of incorporating the farnesyltransferase (FTase) inhibitors, the class of agents that ushered in the era of rationally designed, targeted-based anticancer therapeutics, into the neoadjuvant treatment of breast cancer with doxorubicin and cyclophosphamide (AC), I might have called you crazy, among other expletives. But, to borrow an expression from the contemporary prophets, the Grateful Dead, “what a long strange trip it’s been” for the FTase inhibitors! Nevertheless, after considering that long strange trips are actually more common than expected for anticancer therapeutics that are ultimately incorporated into our therapeutic armamentarium, an apology for any untoward remarks would have been forthcoming. Unfortunately, these long strange trips reflect the complexity of the cancer cell and our ignorance in predicting whether subtle changes induced by new therapeutics in the environment of a cancer cell, which is undergoing an inordinate amount of simultaneous biochemical processes in a primordial soup of chemicals and enzymes, will result in a relevant therapeutic response. And, if so, what will be the nature of the response? Yet, targeting FTase made so much sense to many of us who knew just enough biochemistry, cell biology, and pharmacology to be dangerous (and you know who you are!), and Big Pharma was vested in the prospects that FTase inhibitors had reasonable blockbuster potential based on somewhat teleologic presumptions, which, in hindsight, were incorrect. Like other rationally designed, target-based agents, the FTase inhibitors were designed to disrupt fundamental aberrations in malignant cells, whereas normal cells, whose growth and sustenance are not principally dependent on or driven by the target itself or aberrations of the target, were supposed to be spared. Preferential tumor growth inhibition, by perturbing the equilibrium between cancer cell proliferation and death, is tantamount to the Holy Grail in oncology therapeutics, and the optimal development of anticancer drugs is all about exploiting therapeutic indices, that is, maximizing activity-toxicity quotients. Even agents with modest anticancer activity with highly favorable therapeutic indices may be useful in the clinic if the favorable therapeutic indices increase breathing room or the potential to administer them in combination with relevant doses of other agents with much narrower therapeutic indices. A core determinant of the therapeutic index is the redundancy of critical cellular functions in living cells, with these redundant systems evolving over billions of years as a result of selection pressure. Because of the high genomic instability of malignant cells, billions of years of selection pressure are, in essence, collapsed into the lifetime of a tumor, which is much more adept at exploiting redundant cellular processes than normal tissues. For this reason, among many others, administering single agents with the intent of imparting a major therapeutic response is usually tantamount to spitting into the ocean. In other words, unless the tumor is driven by a dominant aberration of the target, its redundant cellular processes will eventually overcome the insult. We often marvel about the complexity of living systems, but, ironically, the complexity of life is all about its simplicity. This paradox is illustrated by considering how innumerable numbers of cellular constituents perform an unfathomable amount of widely disparate functions by using only a few nearly identical chemical processes catalyzed by a small number of highly conserved enzymes. Reflecting on this notion may help to explain why nonspecific cancer therapeutics, which are directed at target structure instead of target function, perturb a wide array of cellular functions, thereby conferring low therapeutic indices in the clinic. A case in point is the process by which proteins are modified after translation and thereby rendered functional. In particular, prenylation, a lipid-modification process with intermediates in the cholesterol biosynthesis pathway, such as the 15-carbon farnesyl and the 20-carbon geranylgeranyl, is of great importance because Ras and other prenylated proteins, which are integrally involved in proliferative signal transduction, gain function on mutation. The complexity of protein prenylation relates to its simplicity. That is, the existence of only three highly conserved prenyltransferases, namely FTase and geranylgeranyltransferase-I (GGTase-I) and GGTase-II, catalyze the covalent formation of a thioether bond between the prenylgroups and the thiol group of cysteines at the carboxyl terminal of a least 300 structurally and functionally disparate proteins in the human proteome. To complicate matters even further, or maybe to simplify them, although the prenyltransferases have varying preferences for substrate proteins, FTase and GGTase-I can cross-prenylate a long list of proteins with CAAX (C cysteine, A aliphatic amino acid, and X any amino acid) at their carboxy terminus. And this capacity for cross-prenylation or redundancy is the Achilles’ heel of the premise that targeting FTase alone will deleteriously affect critical cellular machinery if the intent of the therapeutic maneuver is to chiefly disrupt K-Ras and other proteins that are crossprenylated by GGTase-I. Herein lies the dilemma with the JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 24 NUMBER 19 JULY 1 2006