By: David Warmflash
Cancer survival is better today, compared with 50 years ago, for a variety of reasons. First, certain cancers can be detected earlier than they could in the past. Second, patients can be assigned specific protocols of drug combinations and other treatments based on accurate disease staging. Third, when patients do respond to treatment and achieve remission, they can be monitored for months and years for signs that the cancer is coming back, and then given adjusted treatment.
We still have a very long way to go, but all of these strategies can be more effective if very tiny quantities of cancer tissue, or cells, could be detected in a person. Minimal-invasive detection is particularly desirable, especially tests requiring only a blood sample, since these can be performed over and over as opposed to surgical biopsy of deep, solid tissues. Furthermore, capability to detect just a few cancer cells would be an improvement over imaging modalities, such as computed tomography (CT) and magnetic resonance imaging (MRI), which have a certain minimum cancer volume that they can see, depending on the cancer type and the body parts where physicians are looking.
For decades, all of these factors have inspired researchers to identify chemical signatures of disease, called biomarkers. For screening and monitoring, various biomarkers are in routine use and are generally proteins. Prostate specific antigen (PSA) is a well-known example. PSA level can be checked in a man’s blood sample, where the level increases as the cancer grows and decreases when the cancer shrinks. This makes PSA very useful for tracking cancer progression and treatment, and also for monitoring for signs of returning cancer after treatment. However, PSA is not very good for screening the population, because it’s specificity for cancer is not high. In other words, one’s PSA can be moderately elevated for many reasons having nothing to do with prostate malignancy. Similar examples of modestly useful protein biomarkers exist not just in oncology, but throughout medicine, so researchers have always been looking for biomarkers more sensitive and more specific, yet that require only a blood draw.
Over the past couple of decades, that search led to the realization that cancer cells themselves can be a biomarker, because small numbers of such cells can leak from tumor tissue into the blood stream. These are called circulating tumor cells and they are used today for monitoring certain cancers. Living in an age of genetic technology, however, the search has also revealed that we need only detect DNA sequences from the circulating tumor cells, or DNA sequences that have leaked into the bloodstream from the tumor, with or without any other cell material.
For a growing number of cancers, circulating tumor DNA (ctDNA) has been identified. Use of ctDNA to infer the presence or progression of cancer is called liquid biopsy, but there are various strategies to specify such DNA. First, there are specific sequences within ctDNA that are altered compared with the wild type in cancer cells, meaning the corresponding genes from non-cancer cells. Examples are mutated sequences in genes for epidermal growth factor (EGFR) and KRAS genes, which can be useful for liquid biopsy of non small cell lung cancer.
Liquid biopsies based on sequence mutations are the furthest along in the pathway to becoming clinically routine, but liquid biopsy also can involve tests for fusion of genes in DNA as well as altered RNA, the molecules that are made from DNA sequences. RNA can have incorrect splicing associated with cancer, for instance.
Another promising area is in epigenetic modification patterns of DNA genes. Epigenenics refers to processes that do not change gene sequences but change how genes are expressed. Epigenetic changes in a gene can be inherited, if they occur in sperm or egg cells, but they can also occur specifically in tumor cells. One epigenetic phenomenon is methylation of one of the four DNA letters (addition of a CH3 group), where more methylation turns down the activity of a gene and less methylation makes the gene more active. There are certain patterns of methylation of certain genes associated with certain cancers. GSTP1, for instance, is a gene that is methylated in 90 percent of prostate cancers. STRATIFIN is a gene that is methylated in many breast cancers, and methylation of the HOXA9 is associated with ovarian cancer. A liquid biopsy for methylation has been developed specifically for prostate cancer.
Given the ease and lack of invasiveness of such tests, imagine the consequences for cancer monitoring, and eventually perhaps even screening, as the tests are developed further in the years to come.
David Warmflash, M.D., is an astrobiologist, science writer, and physician. He is principal investigator on a Planetary Society-sponsored investigation of the effects of the space environment on organisms.