DeVita. Cancer

187

Chapter 15  Precision Medicine in Oncology

Figure 15.1  Molecular monitoring of patients treated with “precision” therapeutics to demonstrate proof-of-mechanism in vivo. The current approach to the development of targeted therapeutic agents focuses on the use of preclinical models to validate biomarkers that can subsequently be used with tissues obtained from patients for molecular tumor characterization and proof-of-mechanism pharmacodynamic studies. Repeated tissue acquisition during therapy and at the time of disease progression, correlated with clinical assessment of efficacy and toxicity, facilitates understanding of the spe- cific mechanisms of drug sensitivity and resistance operating in each individual patient. FISH, fluorescence in situ hybridization; IHC, immunohistochem- istry; PK, pharmacokinetics; PD, pharmacodynamics; CTCs, circulating tumor cells; CECs, circulating endothelial cells. (Reproduced from Doroshow JH, Kummar S.Translational research in oncology—10 years of progress and future prospects. Nature Rev Clin Oncology 2014;11:649–662, with permission.)

may underlie the efficacy of specific, known targeted therapeutic agents, so-called “actionable” mutations of interest. 21 Furthermore, the level of evidence used to support the potential clinical utility of a specific mutation determined by NGS that is assessed in a par- ticular gene panel often varies widely, not only from gene to gene but also by disease context. It is also clear that an ongoing, major effort is required to assure the accuracy and reproducibility of any mutational profile, including the concordance of results from lab- oratory to laboratory, 24 as well as to provide the decision support necessary to facilitate the use of this information by busy clini- cians. 25 Another important issue that can affect the utility of NGS panels is that tumor tissues are often sequenced without concom- itant normal tissue controls. Unfortunately, even using the most sophisticated algorithms, not correcting for germline variants can increase the false-positive variant call rate. 26 Germline sequencing may produce findings that can have a direct impact on the choice of a specific treatment (e.g., discovery of alterations in BRCA1/2 ) or unexpected results that are incidental to cancer treatment but could provoke the need for timely genetic counseling for the pa- tient’s family. 20,27 As genomic testing becomes more common and more broadly based, there is a growing need to clarify how and in what context incidental findings should be disclosed. 28 Although NGS panels provide broad mutational coverage and can be performed and interpreted on time scales consistent with on- cologic practice, whole exome sequencing (WES;  20,000 genes) provides greater breadth (at less depth) of coverage than NGS pan- els with the opportunity for analysis of signaling pathways, in addi- tion to specific gene mutations, of importance for understanding drug resistance mechanisms and the potential role of tumor het- erogeneity in treatment response. 29 However, the added informa- tion obtained from WES requires a longer time frame and greater cost for both the genetic testing and bioinformatic analysis—a time frame and cost that is not currently consistent with routine clinical requirements. In addition, the use of WES has led to the discovery of large numbers of mutational variants of unknown bio- logic or clinical significance (VUS). 30 Substantial effort to develop international, open-access databases to catalog the presence and

functional significance of mutational variants across disease histol- ogies is currently under way to address this burgeoning issue. 31

CANCERTHERAPEUTICS

Broadening the Spectrum of Molecular Characterization Precision oncology has now become substantially more than an isolated NGS panel or WES testing of tumor or germline DNA. Evaluating the mutational spectrum of the entire genome rather than individual genes has recently been demonstrated to provide important clues to the underlying homologous recombination status of a tumor (which is important for determining sensitivity to inhibitors of poly [ADP-ribose] polymerase); DNA sequencing profiles can also be used to assess microsatellite instability (and enhanced presentation of neoantigens) by a patient’s tumor, a predictive biomarker for response to anti–PD-1 antibodies. 32,33 Combining simultaneous RNA and DNA sequencing enhances molecular characterization by allowing comparison of expression profiling with mutational analysis and permitting an understand- ing of the importance of the expression of specific alleles as well as the ability to better define the presence of gene fusions, an in- creasingly important source of validated tumor targets. 23 It seems likely that in the near future clinical molecular characterization will employ a multidimensional approach that combines genetic with epigenetic and proteomic analyses of malignant disease, espe- cially important to further our understanding of single nucleotide variants. 13,34 Such multidimensional studies have begun to show that gene expression at the mRNA level may not fully describe the expression of critical proteins in some tumors, suggesting that an integrated genomic and proteomic perspective may be required for the most effective development of targeted therapeutics. 13,35 Because obtaining longitudinal tumor biopsy specimens for molecular characterization may entail substantial difficulty, 36 and because of the heterogeneity of the mutational spectra that occur across multiple sites of disease, 37 new technologies for molecular analysis of CTCs and circulating tumor DNA (ctDNA) (Fig. 15.2)

Copyright © 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of the article is prohibited.