Nucleic acid testing in oncology

The nucleic acid test (NAT) category includes a variety of diagnostic tests that analyze nucleic acids, either DNA or RNA, extracted from clinical samples such as blood, tissue or tumor samples. The laboratorian then uses molecular biology techniques to detect either the presence or absence of particular nucleic acid sequences, the relative frequency of various sequences, or changes (mutations) in sequences of interest. One of the fastest-growing segments of laboratory medicine, NAT has become central to the life sciences industry and improved healthcare, and is particularly important in personalized medicine. Tools based on NAT have been used to develop a large number of in vitro diagnostic tests.

Applications of NAT in oncology

NAT first came into wide use in testing for infectious disease but is now increasingly applied in oncology. Cancer can be thought of as the iconic genetic disease, since all cancers are due to genetic changes in normal tissue cells that convert them to cancer cells. These molecular genetic changes may consist of single or multiple sequence changes in cell DNA (mutations), rearrangements of chromosomes, or changes in the number of copies of particular DNA sequences in cancer cells.

NAT can either examine the changes in gene sequence or copy number. This information can help identify patients likely to have a recurrence and patients likely to respond to a particular treatment, as well as distinguish a cancer that began in one type of tissue from another type. Knowledge of these changes can often be used to predict the subsequent behavior of the cancer, including likelihood of spread and response to particular therapies. Oncologists and pathologists are increasingly turning to molecular technologies to improve diagnosis and guide treatment.

Cancer predisposition. Genetic factors, including inherited mutations (changes in the DNA sequence), can affect the tendency to develop cancer. For some patients in affected families a single inherited mutant gene may be enough to cause a high cancer risk. Just a few examples include mutation of the APC gene, causing early onset colon cancer; mutation of either the BRCA1 or BRCA2 genes, causing an increased risk of breast cancer; or mutation of the TP53 gene, causing various early onset cancers such as bone or soft tissue sarcoma. In appropriately selected patients, NAT can identify which patients have inherited these mutations and therefore are or are not at increased risk of developing the associated cancer.

Personalized therapeutic response. Advances in understanding cancer at the genetic level has led to development of targeted therapies, which, in eligible patients, have improved patient care due to fewer side effects and improved response rates to therapy as compared to standard chemotherapy. However, targeted therapies are only effective in subsets of patients who may or may not carry a somatic mutation, which has led to the American Society for Clinical Oncology (ASCO) and National Comprehensive Cancer Network (NCCN) Guidelines recommending mutation testing for determining response to certain targeted therapies.

One such example is the gene encoding for epidermal growth factor receptor (EGFR) in non-small cell lung cancer, the most common type of lung cancer. In adenocarcinomas with particular mutations in the DNA sequence encoding EGFR, this protein can be activated even in the absence of epidermal growth factor, stimulating the growth of the lung cancer. Approximately 10% of lung cancer patients have advanced cancers that are EGFR mutation positive, and are more likely to be sensitive to therapy with tyrosine kinase inhibitors (TKIs) which block the action of the mutated EGFR. Thus, guidelines recommend testing of lung cancer patients to ascertain the presence of EGFR mutations prior to selecting first-line therapy for metastatic disease.

Another test performed to aid in targeted therapy selection in advanced or metastatic non-small lung cancer is the ALK gene rearrangement test. If a patient’s tumor cells have the ALK gene rearrangement, which stimulates tumor growth, then the drug crizotinib, an ALK inhibitor, is used for treatment. ALK gene rearrangements have been reported in 2% to 7% of non-small cell lung cancers.

Given the relatively small patient population benefitting from these targeted therapies, research continues to further understand tumor pathways and develop drugs to benefit other subsets of patients. The number of tests and the number of tumor types requiring such testing to optimize treatment selection is steadily increasing.

Tissue of origin testing. A key step in cancer treatment is identifying exactly what kind of cancer the patient has developed. Cancers are best treated according to their respective sites of origin and tissue type, even when the disease is metastatic at the time of diagnosis. An accurate diagnosis of the primary tumor site helps the physician choose the best course of treatment for the patient, especially with the increased availability of targeted cancer therapies.

In up to 10% of cancer cases, tumors are not readily classifiable in the course of the initial diagnostic workup. Diagnosing these difficult cases, including metastatic tumors and poorly differentiated tumors, is subjective, time-consuming, and often does not result in a definitive diagnosis. Traditionally, patients with uncertain tumors undergo extensive imaging, such as CT scans and PET scans, to assist in pinpointing a tumor’s source, and the tumor tissue itself is put through a battery of stains to help identify the tissue of origin. Even after exhaustive testing, it may still be difficult for the pathologist to determine where the cancer started.

Tissue of origin testing through gene expression analysis uses RNA extracted from a standard formalin-fixed, paraffin-embedded (FFPE) pathology sample to assist in identifying a tumor’s origin. In one FDA-cleared method RNA from the tumor sample is measured for gene expression of >2,000 genes in order to predict the most likely primary site of difficult-to-diagnose tumors from among a panel of 15 different cancer types.

The algorithm for this tissue of origin test uses a set of 2,000 probesets, commonly referred to as a “signature,” and a set of coefficients. Based on these parameters, it computes degree of likelihood of a match between the expression profile of a tumor sample and expression profiles of each of the 15 cancer types on the test panel.

A recently published study showed that after receiving tissue of origin test results for 107 patients with poorly differentiated cancers, oncologists’ working diagnosis of the most likely primary site was changed in the majority of patients, and cancer-specific treatment recommendations were changed for two-thirds of the patients. A majority of the oncologists identified the test results as at least moderately influencing the decision to make a change in therapy.¹

Cancer recurrence. Diagnostics companies have developed and commercialized genetic or genomic tests to determine risk of cancer recurrence. These tests often focus on relatively small numbers of genes that are the most prognostic for cancer recurrence, which were identified based on comparisons of the gene profiles of patients with disease recurrence versus those with no recurrence. A breast cancer recurrence test comprised of 21 genes was also shown by retrospective analysis to predict patient responsiveness to drug treatment. Prospective studies are being conducted to further assess the clinical utility of this technique and others for cancer recurrence and treatment selection.

Methods used in nucleic acid testing

Origins. Several different methods are available to detect changes in DNA sequence, arrangement, or number of copies of genes in cancer samples. In most cases these changes are measured directly through NAT, although in the past indirect methods such as histology, immunohistology, and enzyme assays have been used.

Polymerase chain reaction (PCR). PCR is a technique in molecular genetics that permits amplification of short sequences of nucleic acids—DNA or RNA—even in samples containing only minute quantities of nucleic acids. These amplified sequences can then be examined for changes in sequence or to quantitate the relative or absolute frequency of the sequence in the original sample.

PCR has been essential to the study of molecular origins of disease and a highly successful application of NAT, providing the ability to greatly amplify DNA sequences that come from minute samples or present at a low frequency in samples. PCR offers the advantages of providing highly specific amplification of the desired sequence regions quickly and cost effectively.

Sanger sequencing. Developed by Frederick Sanger in the 1970s, the Sanger sequencing technique utilizes chemically altered “dideoxy” bases to terminate newly synthesized DNA fragments at specific bases; these fragments are then size-separated, and the DNA sequence can be read directly.²

The Sanger method is currently the gold standard for clinical diagnostics and is increasingly used for clinical research as the cost of sequencing continues to drop. Recently, one-step Sanger sequencing methods combining amplification and sequencing have been developed that allow rapid sequencing of target genes.³

Next generation sequencing. Because a cancer may be linked to multiple genes and mutations, and mutations may be present in only a small percentage of the cells in tissue samples, high-throughput or “next generation” sequencing (NGS) has the potential to be used for many of the oncology NAT applications previously mentioned. NGS has the great advantage of being able to analyze many different sequences and mutations simultaneously from the same sample. This has positioned NGS as the method of choice for targeted re-sequencing of regions of the human genome identified by linkage analyses and genome-wide association studies.⁴

Since NGS technology is currently largely restricted to research or investigational use, the first applications of NGS in diagnostics are laboratory-developed tests (LDTs). For example, Children’s Hospital Los Angeles is developing an LDT for retinoblastoma gene testing—covering 250,000+ bases of the human genome—to determine likelihood of recurrence in retinoblastoma patients. NGS technology may be able to cover the entire process at a fraction of the cost of conventional sequencing. Similar LDT testing technologies are also in development at Thomas Jefferson University, Fox Chase Cancer Center, and the Mayo Clinic.⁵

Microarrays. Until recently, microarray technology was used primarily as a research tool, but it is now used increasingly in clinical diagnostics. Microarray technology consists of probes that are arranged or “arrayed” on a glass chip. These probes are small fragments of DNA that represent a section of a specific gene’s entire DNA sequence. Microarray technology is a robust tool for gene expression diagnostics because it measures expression levels of large numbers of genes simultaneously and can answer multiple diagnostic questions with one array.

Fluorescence in situ hybridization (FISH). FISH is used to detect and localize the specific DNA sequences on chromosomes. It can be applied to detect gene rearrangements, such as ALK rearrangements in non-small cell lung cancer, as described earlier in this article.⁶ FISH uses fluorescent probes that bind to parts of the chromosome with which they show a high degree of sequence complementarity. Fluorescence microscopy is used to find out where the fluorescent probe has bound to the chromosomes.

Chromogenic in situ hybridization (CISH). CISH allows detection of gene amplification, chromosome translocations, and chromosome number using conventional enzymatic reactions under the brightfield microscope on FFPE samples.⁷ In June 2011 the FDA approved a new test to determine a breast cancer’s HER2 status, which affects treatment selection. CISH is an alternative to FISH that can be easily integrated into routine testing in laboratories, as it doesn’t require fluorescence microscopy.

Bioinformatics’ impact on nucleic acid testing

The evolution of bioinformatics will be essential in realizing the potential of advanced sequencing and gene expression techniques in nucleic acid testing such as NGS. For example, cancer molecular diagnostics that quantitatively measures the presence of thousands of mRNA transcripts using a DNA microarray requires an additional step. The mRNA transcript values must be normalized and analyzed appropriately by computer algorithms.

The development and validation of these types of algorithms is essential for a tissue of origin test. The algorithms compare the mRNA in the tumor specimen to the mRNA in a database consisting of gene expression measurement for thousands of different samples from tumors in the 15-tissue panel and then assist in making the diagnosis of tissue of origin.

The informatics employed consist of multiple subsystems, which perform various specific tasks: verification that expression data are acceptable; transformation of expression data into a format that can be normalized; normalization of the data; statistical analysis of the expression data; generation of similarity scores that identify the degree of similarity for each database tissue type; and generation of reports. Validation of the algorithm for one tissue of origin test involved a blind study with 462 FFPE specimens which was published in the January 2011 Journal of Molecular Diagnostics.⁸

With bioinformatics a rapidly developing area for pathologists and laboratorians, The American Board of Medical Specialties recently introduced clinical informatics as a new board-certified specialty.

Integration into the clinical laboratory

With the evolution of nucleic acid testing for oncology, tests once offered by specialty labs will continue to integrate into the mainstream clinical lab.

In addition to understanding the test method being used, laboratorians will be called upon to provide guidance to physicians to select the appropriate clinical specimen and test method, especially as improved cancer detection often results in smaller specimens, which must be carefully used for determining diagnosis, prognosis, and selection of targeted therapy.

Because the test methodologies remain fairly complex, implementation of robust control systems are essential for both laboratory-developed and in vitro diagnostic tests to assure high quality test results generated for physician interpretation, and ultimately, to provide the best patient care.

References

1. Nystrom et al. Clinical utility of gene-expression profiling for tumor-site origin in patients with metastatic or poorly differentiated cancer: impact on diagnosis, treatment, and survival. Oncotarget Advance Publications. June 9, 2012.
2. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA. 1977;74(12):5463—5467.
3. Sengupta D, Cookson B. SeqSharp: a general approach for improving cycle-sequencing that facilitates a robust one-step combined amplification and sequencing method. J Mol Diag. 2010;12(3):272—277.
4. Hall N. Advanced sequencing technologies and their wider impact in microbiology. J Exp Biol. 2007;(210):1518—1525.
5. Steenhuysen J. Doctors try to make sense of cancer’s genetic jumble. Reuters; June 5, 2012.
6. Langer-Safer PR et al. Immunological method for mapping genes on Drosophila polytene chromosomes. Proc Natl Acad Sci USA. 1982;79(14):4381—4385.
7. Madrid MA, Raymundo WL. Chromogenic in situ hybridization (CISH): a novel alternative in screening archival breast cancer tissue samples for HER-2/neu status. Breast Cancer Res. 2004;6(5):R593—R600.
8. Pillai et al. Validation and reproducibility of a microarray-based gene expression test for tumor identification in FFPE Specimens. J Mol Diag. 2011;13:48-56.