Cancer Genetics and Genomics for Personalized Medicine

Il-Jin Kim, Il-Jin Kim

• Publisher: Jenny Stanford Publishing
• Year: 2017
• ISBN:
  9781315341101
• Edition: 1
• Topic: Medizin
• Subtopic: Onkologie

This book covers almost all fields of cancer genetics and genomics for personalized medicine. Targeted therapy, or precision medicine, or personalized medicine is becoming a standard treatment for many diseases, including cancer. However, how much do we know about the personalized medicine approach? This lucid book helps undergraduate and graduate students, professional researchers, and clinicians to better understand the key concept of personalized medicine.

The most up-to-date topics on personalized medicine in this book cover the recent trends in and updates on lung, gastric, liver, breast, and other types of cancers. Circulating tumor cell, cell-free circulating DNA, and microRNAs are discussed as new diagnostic and prognostic markers for cancer. The avatar mouse model is also discussed for maximizing treatment efficacy and prognosis prediction, and so is microenvironment as a drug resistance mechanism. With classical and new pathological approaches, the book provides a systemic overview of personalized immunotherapies and hyperthermic intraperitoneal chemotherapy, followed by new emerging fields of hereditary cancer, thereby equipping readers to eventually contribute in developing more advanced tools and therapies for curing cancer.

Chapter 1

Personalized Medicine for Cancer: Introduction and Overview of the Book

Il-Jin Kim and David Jablons

Thoracic Oncology Laboratory, Department of Surgery, Comprehensive Cancer Center,
University of California San Francisco, San Francisco, CA, USA
[email protected], [email protected]

1.1 Changing the Treatment Paradigm for Cancer

We live in an age in which tremendous technological breakthroughs in many fields—the Internet, mobile computers and smart phones, new energy technologies, to name just a few—have dramatically changed the global economy and human life. However, treatments for human disease have not kept pace with advances in other fields. One of the very few examples is antiretroviral therapy for HIV, whereby multiple-drug treatment can dramatically delay disease progression and prolong survival so that AIDS becomes a chronic disease [1].

For cancer, the second-most lethal human disease in the United States [2], there has been no significant improvement for decades. The reasons include continued high smoking rates, air pollution, exposure to environmental toxic material (e.g., asbestos), and many unknown epidemiological and genetic causes. Cancer is known to be a genetic disease mainly caused by genetic alterations such as somatic mutations of tumor suppressor genes (e.g., TP53) and oncogenes (e.g., K-ras and EGFR) [3, 4]. However, cancer is notoriously complex and heterogeneous. In addition to genetic alterations, many other mechanisms, such as epigenetic, immune related, and environmental factors, contribute to cancer development, evolution, metastasis, and acquisition of drug resistance [3, 4]. Moreover, cancer is not static—its characteristics change to survive in different situations and environments in the human body. Thus, it is not easy to identify a clear target to attack and cure cancer.

Among all potential factors that cause a high incidence of cancer and inhibit effective treatment, the biggest culprit may be the “one-fits-all” approach. This classical approach has been used since the first chemotherapy treatment was given in the 1940s [5], but oncologists and cancer researchers have known for years that it is rarely successful. A new paradigm of cancer treatment has been needed to move into the modern era of “personalized” or “precision” medicine (PM), which can be defined as a medical treatment decision and action based on individual patient’s genetic, epigenetic, histopathological, or other health information.

The first interruption of the one-fits-all treatment approach was the identification of BCR-ABL fusion protein in patients with chronic myelogenous leukemia (CML) [6, 7]. Soon after this discovery, imatinib (Gleevec) was developed to cure CML with BCR-ABL fusion. This was to become the first successful PM attempt. After imatinib treatment for BCR-ABL positive CML patients, the 5-year survival rate has doubled from 31% in the early 1990s to 60% in 2010 [8]. The FDA later approved an imatinib treatment for KIT-positive gastrointestinal stromal tumors (GIST) patients [9, 10] and Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ ALL) patients [10]. The imatinib breakthrough has also shifted the research focus for cancer treatment in many academic institutes and pharmaceutical companies that now seek to identify molecular markers and drug targets. For example, hormone epidermal growth factor receptor 2 (HER-2), estrogen receptors, and progesterone receptors were identified in breast cancer patients [11], and trastuzumab (Herceptin) has proved effective for those with HER-2 positive tumors [11].

For lung cancer, EGFR mutations and EML4-ALK fusion proteins have been identified and patients with EGFR oncogenic mutations were found to be highly responsive to tyrosine kinase inhibitor (TKI), gefitinib, or erlotinib. This entailed the new concept of routine screening of EGFR mutations in lung cancer patients to find the subpopulation suitable for TKI treatment. However, many of these patients eventually experience tumor recurrence because of drug resistance caused by EGFR T790M mutation, MET amplification, noncoding RNA-related mechanisms, or transformation of the tumor’s identity [12]. A recently identified fusion protein, EML4-ALK, in a lung cancer patient harboring no EGFR and K-ras mutation, may be a new target [13]. Interestingly, the ALK inhibitor crizotinib was not a successful cancer treatment until this new oncogenic fusion was identified. It took less than 5 years from identifying the new marker, EML4-ALK, to FDA approval for crizotinib for lung cancer patients. This may be the fastest approval on record for a cancer drug and illustrates a new paradigm of “drug repurposing or repositioning” after new molecular markers are identified.

The focus of this book is on the genetic and genomic aspects of PM for cancer. Dr. Daniela Morales-Espinosa et al. detail a PM approach for treating lung cancer patients in Chapter 2, and Drs. Jae-Jun Shim and Ju-Seog Lee describe PM for liver cancer in Chapter 3.

1.2 Companion Diagnostics and New Sequencing Technologies

As molecular markers like EGFR mutations and targeting drugs like gefitinib initiated a new era of PM, robust and reliable methods are needed to detect molecular alterations that identify patient subpopulations suitable for a new treatment. Companion diagnostics can be defined as “an assay or diagnostic system for selecting and identifying the best treatment decision for patients” [14]. One example is the fluorescence in situ hybridization (FISH) assay developed by Abbott Diagnostics for selecting lung cancer patients with EML4-ALK. Developing a companion diagnostic assay is tremendously expensive, but an accurate and cost-effective assay is key to the PM approach.

Next-generation sequencing (NGS) has become popular for genetic analysis of mutation (exome or whole genome sequencing [WGS]) and RNA expression (transcriptome or RNA-seq) analysis. NGS uses either a hybridized capturing technology (Illumina HiSeq and MiSeq) or a semiconductor sequencing technology that detects hydrogen ions released during DNA polymerization (e.g., ion torrent PGM and proton). These methods can scan a large region (e.g., exome or WGS) or screen a specific target region (e.g., targeted panel for companion diagnostics). The NGS method that we developed, called NextDay Seq (NDS), enables fast, targeted sequencing of clinically actionable genes such as EGFR and K-ras in lung, colorectal, and other cancers (these technologies are licensed from UCSF to CureSeq Inc. www.cureseq.com). Sequencing can be completed within 48 hours from a DNA extraction of formalin-fixed paraffin-embedded (FFPE) samples through NDS to final data analysis. A targeted NGS panel can be used to discover a companion diagnostic tool for efficiently selecting specific types of cancer for targeted therapy.

NGS still needs to be improved. Although exome or whole genome screening is powerful, it is not trivial to filter out many false positives and accurately select true mutations. A homopolymer or a repeat of the same sequence is one major problem for NGS. As many microsatellite and loss of heterozygosity (LOH) markers are just repeats of the same sequences, it can be an issue to use NGS for studying those markers. For example, around 10% of colorectal cancer samples have a microsatellite instability caused by a deletion or insertion of microsatellite sequences [16]. Thus, it is important to select the best NGS method for a particular type of genetic screening. This is also illustrated by the following example of EGFR. Currently, EGFR may be the most important gene for PM in lung cancer. Using different NGS technologies to screen more than 1000 lung cancer tissues, we found that some technologies have a problem detecting EGFR exon 19 deletion mutations. Unlike other common oncogenic mutations (i.e., EGFR L858R, K-ras codon 12 and 13 mutations, and BRAF V600E), EGFR exon 19 deletions are multiple bp deletions as long as 15–25 bp. This kind of long deletion can cause a wrong sequence alignment or be regarded as sequence noise by bioinformatics programs. Next on the horizon is third-generation sequencing or nanopore sequencing, which can sequence up to 50 kb [17] and might overcome several issues found in NGS.

1.3 Early Detection of Cancer and Tumor Recurrence Monitoring: Circulating Tumor Cell (CTC) and Circulating Tumor DNA (ctDNA)

Although new drug targets and molecular markers for novel, efficient therapy are needed to ultimately reduce the high mortality rates for cancer, a more practical approach will be to detect cancer cells at an early stage and provide the best treatment option, such as a surgical resection or targeted therapy if available.

Cancer can be detected early by imaging methods, such as X-ray, computerized tomography, or magnetic resonance imaging, and by less invasive or non-invasive screening of blood, sputum, urine, stool, and exhaled breath condensate [19, 20]. Currently, imaging methods may be a golden standard, but they may entail low sensitivity and specificity, high cost, and inconvenient access to an up-to-date facility.

Molecular screening could be a promising approach for detecting cancers early, but is in the preliminary stage and needs considerable clinical validation. To use human body fluids or samples for early cancer diagnosis, two issues should be solved. The first is to identify reliable molecular markers. Despite many studies that tested early detection of cancer by mutation, methylation, and noncoding RNA using blood, sputum, and other biological samples [19, 20], only a few methods were approved for early cancer screening. One example is prostate cancer antigen 3 (PCA3), which is a long noncoding RNA expressed mainly in prostate cancer [20, 21]. PCA3 is FDA-approved for a urine test for prostate cancer screening using realtime quantitative PCR [20, 21]. The second issue is to use a reliable and highly sensitive technology to detect cancer early. The ability to detect cancer-specific mutations such as TP53, K-ras, and EGFR in plasma or sera came about when very sensitive technologies such as NGS became available [19–22]. Circulating tumor DNA (ctDNA) and circulating tumor cells (CTC) are promising molecular markers for the early detection of cancer [22, 23]. Cell-free fetal DNA is clinically used for non-invasive prenatal testing (NIPT) by a mass spectrometry-based method, digital PCR, or NGS [24]. While it is relatively common to use cfDNA for NIPT, using ctDNA analysis to detect cancer is in the very early stage, though results are encouraging [22, 23]. Unlike ctDNA, CTC was FDA-approved for metastatic breast, colorectal, or prostate cancer screening with CellSearch, an epithelial cell adhesion model (EpCAM)-based method [25, 26]. Although several issues have been raised for CTC screening, such as a lack of metastasis detection and non-epithelial cell originated cancer detection [25, 26], it is encouraging that a less invasive blood-based screening tool is currently available for cancer screening in the clinic.

In Chapter 4, Drs. Dana Tsui and Muhammed Murtaza relate how ctDNA has been used for PM in cancer research and treatment. In a Chapter 5, Dr. Jin Sun Lee et al. describe the use of CTC in breast cancer research and treatment.

1.4 Cancer Animal (Mouse) Models and Microenvironment for Personalized Medicine

Many promising drug candidates showing striking tumor-inhibiting effects in in vitro ultimately fail in clinical trials. Although cancer cell lines with mutations (i.e., K-ras, EGFR, or BRAF) and human cancer tissues are wonderful resources for research, they are clearly limited in their ability to reflect real human immune, angiogenesis, microenvironment, and other physiological conditions. Thus, any new drug target or biological marker requires testing in an in vivo system. In Chapter 6, Dr. Laura Soucek and her colleagues comprehensively review an “avatar” mouse model of human cancer. A fast and reliable human cancer avatar model would be very helpful for selecting the best treatment regimen for cancer patients.

Drug resistance is a major obstacle for current and future PM therapy. The mechanisms of drug resistance can be genetic alterations such as EGFR T790M mutat...