Tumor marker tests vs truly personalised medicine

Should you demand tumour marker tests early on and have them frequently?

Automated blood test - Image courtesy of Science 

Photo Library

The vast majority of cancer patients receive frontline chemotherapy; compounds or drugs that have been discovered over the past 50-60 years. Although these drugs are often effective (initially) they are far from perfect (in fact for some drugs a legitimate question would be to ask whether the perceived benefits outweigh the risks and horrendous side effects), including the possibility of tumours becoming resistant.

Numerous investigations into how genetic signatures impact on response to cancer therapies and prognosis have shown that drugs work best when they are selected on the basis of a tumour’s genetic makeup. Hence, analysing and obtaining this genetic information for each individual patient and their tumour is really a necessity in order to develop personalised treatments that have the potential to greatly improve cancer-treatment success rates.

Ion Proton Chip being prepared

However, the majority of current mainstream cancer treatment decisions are rarely made on the basis of an individual’s complete genetic information. A reasonable alternative to a comprehensive genetic analysis is perhaps found in the field of  next generation biomarkers.

Tumor markers (also known as biomarkers or tumour markers) are incredibly important because they can be used to screen high-risk individuals, confirm diagnosis, guide treatment decisions, assess the response of a tumour to treatment, monitor for recurrence and predict prognosis. However, in most instances any one biomarker on its own is unlikely to accurately reflect the status of a disease. Hence, biomarker tests should really be performed throughout your treatment (i.e. at multiple time points (which is called serial testing)), alongside other diagnostic tests such as MRI scans, tumour biopsies, etc.  Importantly, more than one biomarker for detecting and monitoring your cancer should be utilised.

why multiple biomarkers should be utilised in a clinical setting

For info about Biozantium by Paeon Laboratories:  http://www.biozantium.com

There are several reasons why multiple biomarkers should be utilised in a clinical setting, which I have listed below:

1. An elevated biomarker level may be caused by other diseases or conditions
2. Some biomarkers occur at a naturally high level in some people who do not have cancer
3. Unrelated to cancer, the concentration or quantity of some biomarkers in the blood / urine may vary over time
4. In some cases no significant increases in a biomarker level will occur until advanced cancer is diagnosed (hence some biomarkers may not be useful for early detection of cancer)

In other words it is the combined data (obtained from testing a range of different biomarkers on multiple occasions) that will be informative about your cancer and how well you are responding to therapy.

Current research into more accurate tumor markers / biomarkers

Hence the question, what kind of research is currently conducted in order to develop more accurate biomarkers?

Well, there is a tremendous amount of cutting edge research being carried out in anticipation of developing screening tests that can detect cancer at its earliest stages (before there are any symptoms). In order to achieve these objectives, a biomarker test should be both highly sensitive and specific (i.e. correctly identify people that have cancer as well as accurately identify those that do not have cancer respectively). In other words, you want a test that gives you no false negatives whilst simultaneously you require the test to present you with very few or no false positives.

In essence there are three areas of research in the biomarker space (in addition to the antibody based techniques (e.g. ELISA)). These can be classified on the basis of the type of molecules studied and / or the techniques that researchers employ to develop and investigate these next generation biomarkers. This includes genomics, transcriptomics, and proteomics. These techniques are generally aimed at analysing circulating cell free DNA, RNA, microRNA, or proteins respectively, either in the blood or urine (i.e. a liquid biopsy). These techniques are of course also employed in the case of tissue biopsies.

However, as alluded to earlier, most of the currently approved or licenced biomarker tests are based on detection of certain proteins. Often these proteins are glycoproteins (soluble molecules) in the blood, which can be detected by specialised kits that contain monoclonal and / or polyclonal antibodies (this is also how circulating tumour cells are identified). Detection of increased levels of biomarkers can provide your doctor with useful information. Many people will be aware of classic tests such as PSA (prostate cancer) and CA-125 (ovarian cancer) tests. Unfortunately, large randomised clinical trials have shown that these tests on their own are not very informative as they are neither sensitive nor specific enough to accurately predict the disease status during screening. In case  you have an interest in learning more about specific tests and when they might be used, I have included a list of the more commonly used biomarker tests at the bottom of this page. It is however, not a complete list of tumour marker tests. 

Ion Proton next generation sequencer being operated by technician

As mentioned earlier, the more interesting type of biomarker tests currently being developed at universities and research institutes around the world utilise qPCR, digital PCR and “omics” or next generation techniques, with which genes (DNA) or gene products such as RNA or proteins are studied. Because of relatively recent advances and technological breakthroughs scientists are now able to measure and accurately delineate genome wide mutations as well as expression levels of specific genes or microRNAs. These advances have already resulted in gene / microRNA expression patterns being identified that are associated with specific cancers. In fact some of these associated patterns are very good predictors of disease progression and / or whether or not a given therapy will work. Depending on the outcome of clinical trials and licencing by regulatory agencies, such as the EMA and FDA, of newly developed in vitro diagnostics (IVD), hospitals should be in a position to offer next generation biomarker testing in the not too distant future.

Some of these next generation techniques are already being rolled out for use in clinical research settings. Such as the new South Glasgow Hospital in a collaborative effort with a programme called Stratified Medicine Scotland Innovation Centre or SMS-IC (once they will finish building the new facility, it will be one of the largest acute hospitals in Europe). Interestingly, SMS-IC selected Life-Technologies' Ion-Proton System as their next generation sequencing platform (see video on the left). The Ion Proton platform is based on a semiconductor chip and pH measurements instead of optics. As a result patient genomes can be read in a couple of hours and for less than 1000 €. As such, personalised medicine should be within reach for the majority of cancer patients. Especially when one considers the price tag of most cancer therapies.  

FDA approvals for new in vitro diagnostics (IVD)

In relation to licencing by the FDA, a recent development with regards to the issued Final Guidance on “for research use only” (RUO) and / or “for investigational use only” (IUO) in vitro diagnostic (IVD) products, may allow manufacturers to supply clinical laboratories with such products (as long as they state very clearly that the equipment is for research use only). Also, the FDA recently cleared the Illumina MiSeqDx Platform for clinical use (targeted high-throughput DNA sequencing).

New Tumor marker access for current terminal cancer patients 

Biomarkers paired with specific cancer drugs

Given these recent technological advances as well as regulatory developments in the in vitro diagnostics market space, the prospect of current end stage patients gaining access to these new technologies is not outside the realm of possibilities. Keeping in mind that for people that go through the scary process of being diagnosed with a life threatening or chronic illness, the trial and error process that often follows the diagnosis in order to find a therapy that works best for the patient can be equally terrifying or worse.  So wouldn’t it be ideal if, at the time of diagnosis, or shortly thereafter, a doctor could with a high degree of confidence identify what the best treatment was for that patient and avoid this trial and error process that is currently practiced in mainstream medicine (even if it involves technologies that are not necessarily licensed for clinical use)? 

See below the table for some classical biomarkers.

Biomarker	Which cancer	Other diseases or cancers that can potentially be detected	Remarks
Alpha-fetoprotein (AFP)	Hepatocellular carcinoma, nonseminomatous germ cell tumours.	AFP can also be elevated in Gastric, biliary and pancreatic cancers, cirrhosis, viral hepatitis, and pregnancy.	AFP can help diagnose and guide the treatment of liver cancer. Normal levels of AFP are usually less than 10 ng/mL. AFP levels are higher in most patients with liver cancer. AFP is also high in acute and chronic hepatitis, where values of 100 ng/mL or less are usually observed. In someone with a liver tumour, an AFP level of 400 ng/mL or above is likely to be seen. of liver cancer. If the cancer is completely removed with surgery / therapy, the AFP level should go down to normal. If the level goes up again, it suggests that the cancer has come back.
Carcinoma Antigen 15-3 (CA 15-3)	Breast cancer	Levels of this marker can also be higher in other cancers, like lung, colon, pancreas, and ovarian, and in some non-cancerous conditions, like benign breast conditions, ovarian disease, endometriosis, and hepatitis.	High blood levels found in less than 10% of patients with early disease and in about 70% of patients with advanced disease. Levels usually drop if treatment is working, but they may go up in the first few weeks after treatment is started. Normal level is usually less than 30 U/mL (units/milliliter). But levels as high as 100 U/mL can be seen in women who do not have cancer.
Carbohydrate  / Cancer Antigen 19-9 (CA 19-9)	Pancreatic and biliary tract cancers (although first developed to detect colorectal cancer).	CA 19-9 can be high in other forms of digestive tract cancer, such as the stomach, Colonic, oesophageal and hepatic cancers. It can also be high in some non-cancerous conditions such as thyroid disease, rheumatoid arthritis, inflammatory bowel disease, and pancreatitis, biliary disease, and cirrhosis.	Normal blood levels of CA 19-9 are below 37 U/mL. A high CA 19-9 level in a newly diagnosed patient is suggestive of  advanced disease.   CA 19-9 can be used to monitor bladder cancer and its aggressiveness. It can also be used to monitor colorectal cancer, but a CEA test is likely to be better suited for this type of cancer.
Cancer Antigen 27.29  (CA 27.29)	Breast cancer.	This marker can be elevated in other cancers, such as colon, stomach, kidney, lung, ovary, pancreas, uterus, gastric, hepatic, lung, pancreatic, ovarian and prostate cancers. It may also be elevated in some non-cancerous conditions, such as in women in the first trimester of pregnancy; and in patients with endometriosis, ovarian cysts, non-cancerous breast disease, kidney disorders, and liver disease.	This test measures the same marker as the CA 15-3 test, but in a different way. Although it’s a more recent test than CA 15-3, it’s not better. The normal level is 40 U/mL or less. The CA 27.29 biomarker used in monitoring breast cancer may be superseded by the estimation of circulating tumour cells (CTCs)
Carcino Embryonic Antigen (CEA)	Colorectal cancer.	This marker can be high in some other cancers, such as lung, gastric, prostate, ovary, cervix, pancreatic, breast, bladder cancers, medullary thyroid and other head and neck, cervical and hepatic cancers, lymphoma, and melanoma. CEA levels can also be high in some non-cancerous diseases, like hepatitis, peptic ulcers, inflammatory bowel disease, pancreatitis, hypothyroidism, cirrhosis, biliary obstruction. chronic obstructive pulmonary disease (COPD), colitis, rheumatoid arthritis, and pancreatitis, and in smokers who are otherwise healthy.	CEA is not used to diagnose or screen for colorectal cancer, but it’s the preferred tumor marker to help predict outlook in patients with colorectal cancer. The normal concentration (including smokers) should not exceed 5.5 ng/mL. High CEA levels at the time colorectal cancer is detected, usually indicate that the cancer is advanced. CEA is used to monitor patients with colorectal cancer during and after treatment to check if the cancer is responding to treatment or if it has recurred after treatment.
b-hCG	Nonseminomatous germ cell tumours, gestational trophoblastic disease.	Rarely elevated in gastrointestinal cancers, hypogonadal states and marijuana use.
CA-125	Ovarian cancer.	CA-125 is often higher in women with menstruation, pregnancy, fibroids, ovarian cysts, pelvic inflammation, cirrhosis, ascites, pleural and pericardial effusions, uterine fibroids or endometriosis. It can also be higher in men and women with lung, pancreatic, breast, liver, oesophageal, uterine and colon cancer, and in people who have had cancer before. Because ovarian cancer is not a common disease, a high level of CA-125 is more likely to be caused by something other than ovarian cancer.	CA 125 is used to monitor women during or after treatment for epithelial ovarian cancer, as well as fallopian tube cancer and primary peritoneal cancer. Normal blood levels of CE 125 are less than 35 U/mL. More than 90% of women with advanced ovarian cancer have high levels of CA 125. Levels are also high in about half of women whose cancer has not spread outside the ovary. Because of this, CA 125 has been studied as a screening test. However, many early cancers would go undiagnosed. In addition, problems other than ovarian cancer can cause a high CA-125 level.
PSA	Prostate cancer.	Prostatitis, benign prostatic hypertrophy, prostatic trauma, after ejaculation.	Positive predictive value of PSA levels in prostate cancer greater than 4 ng/mL is 20-30%. This rises to 50% when PSA levels exceed 10 ng/mL. 20-30% of men with prostate cancer have PSA levels within normal ranges

Please note: Any medical or scientific information published on this website is not intended as a substitute for informed medical advice from a physician and you should not take any action before consulting with a health care professional. For more information, please read my terms & conditions.