Research Article: Incorporating Molecular Tools into Early-Stage Clinical Trials

Date Published: January 22, 2008

Publisher: Public Library of Science

Author(s): Robert J Weil

Abstract: The author discusses the implications of a new phase I trial investigating the role of rapamycin in patients with glioblastoma.

Partial Text: During the past several decades, and with an accelerating pace in the past several years, a primary focus of cancer research and treatment has been the development and refinement of specific, biologically directed therapies [1,2]. A number of attractive targets have been identified, dissected, and validated molecularly and biochemically, including multiple members of the family of receptor tyrosine kinases [1,2]. These potent enzymes, frequently concentrated or overexpressed on the surface of cancer cells, phosphorylate target proteins, with varied and manifold effects on numerous downstream, intracellular signaling pathways, leading to profound alterations in transcription and translation, cell growth, differentiation, apoptosis, angiogenesis, and invasion and metastatic potential [1,2]. A number of small molecular inhibitors of these tyrosine kinases (TKs) have been developed in recent years. Imatinib, for example, has shown impressive activity in many patients with chronic myelogenous leukemia [3,4].

GBM is an aggressive, primary tumor of the central nervous system [8]. Because of their intrinsic, infiltrative nature, GBMs follow a malignant clinical course. Classified as World Health Organization grade IV astrocytic tumors, GBMs have a pronounced mitotic activity, substantial tendency toward neoangiogenesis (microvascular proliferation), necrosis, and proliferative rates three to five times higher than grade III tumors, the anaplastic astrocytomas. The clinical behavior of GBMs is often mimicked by unusual pathological presentations, which gave rise to the old moniker of “glioblastoma multiforme” (Figure 1). Even with the survival advantage provided by the recently developed protocol of concurrent chemoradiation followed by adjuvant alkylating chemotherapy with temozolomide (the Stupp regimen), the prognosis of patients with GBM remains poor, with median overall survival in the range of 9–15 months and two-year survival rates of 26% in the most favorable subgroup [9].

These findings spurred the UCLA group to design an important, molecularly focused clinical study, published in this issue of PLoS Medicine [32], to analyze the effect of rapamycin in a subset of patients with recurrent GBM in whom activity of the tumor suppressor PTEN was absent. The study design, which is outlined in their Figure 1 [32], is a “treat-biopsy-treat” paradigm, in which only patients with the appropriate molecular features are selected to receive a targeted biological agent. In this case, patients who were known to have PTEN loss at the time of initial resection were chosen, after recurrence of tumor following standard treatment (surgery, radiation, and temozolomide), for inclusion in the study. Patients (n = 15) were treated for approximately one week with single-agent rapamycin, underwent resection, then resumed therapy and continued it until it was determined that tumor had recurred (time-to-progression).

The ultimate goal of most oncologists is to tailor therapy that takes into account—and exploits—the individual tumor’s unique biological features. While individualized therapy may be some years in the future, the work of Cloughesy et al. [32], and that of many others, is pointing a way toward rational design of therapy for stratified groups of patients who share common molecular features [2,35–37]. Therefore, one strategy in designing the next generation of clinical trials in oncology must be to address both the known interactions, and, as Cloughesy et al. have done here, interrogate clinical studies and tissues at an early stage to identify genetic and biochemical features that distinguish responders and non-responders so that both types of patients receive optimal therapy (Figure 3).



Leave a Reply

Your email address will not be published.