Tumor markers :towards improving the landscape of cancer biomarker research

Development and clinical uses of tumor markers

1. Tumor markers Towards Improving the Landscape of Cancer B iomarker R esearch Ekbal Mohamed Abohashem –MD Professor of Clinical Pathology Mansoura University -Egypt

2. U ltimate reliance in history requires written records. In medicine, the earliest written description of diseases and cancer, a breast cancer, is found in the Edwin Smith Papyrus that was written approximately 3000 BC. The writer concluded that bulging tumor of the breast was a grave disease and there was no treatment for it. The Ebers Papayrus , dated circa 1500 BC, contains the first reference to a soft-tissue tumor , a fatty tumor , and includes reference to possible cancers of the skin, uterus, stomach, and rectum. The Egyptians attempted to treat tumors and cancers with cautery, knives, and salts, and introduced arsenic paste that remained in use as ‘‘Egyptian ointment’’ until the 19th century. Historical background

3. The word cancer came from a Greek word karkinos to describe carcinoma tumors by Hippocrates (460-370 B.C). In 1998, the National Institutes of Health Biomarkers De fi nitions Working Group de fi ned a biomarker as “ a characteristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention.

4. T he first tumor marker in modern medicine was identified by Bence -Jones, who in 1846 , detected a heat precipitate in samples of acidified urine from patients suffering from " Mollities ossium ". History of cancer biomarker discovery From 1930 to 1960, scientists identified numerous hormones, enzymes and other proteins, the concentration of which was altered in biological fluids from patients with cancer. The modern era of monitoring malignant disease, however, began in the 1960s with the discovery of alfa -fetoprotein and carcinoembryonic antigen (CEA), which was facilitated by the introduction of immunological techniques such as the radioimmunoassay.

5. Since then numerous potential tumor markers have been reported in the literature. Tissue polypeptide antigen (TPA) is a chemically well defined substance identified by Bjorklund in 1957. In 1965, GOLD et al . isolated a glycoprotein molecule from specimens of human colonic cancer and thus discovered the first "tumor antigen", later identified as carcinoembryonic antigen (CEA). In the 1980s, the era of hybridoma technology enabled development of the ovarian epithelial cancer marker carbohydrate antigen (CA) 125. In 1980, prostate-specific antigen (PSA [KLK3]), considered one of the best cancer markers, was discovered.

6. Tumor markers have been used in oncology for about half a century . Since their discovery in the mid-1960s and 1970s, such molecules were used to combat cancer through screening, early diagnosis, monitoring of therapy, prognosis, and prediction of therapeutic response. S creening for early disease detection could change the course of cancer, thus more people would be cured by early interventions,. However , for some cancers, screening is clearly bene fi cial (such as colon and cervical cancer), while for other cancers, screening is not effective.

7. Some major cancers (such as breast and pancreatic) proliferate quickly and when the cancers are detected by screening, they have already spread. On the other hand, slow growing cancers are not usually lethal and their detection may lead to over-treatment, which has its own side effects. These caveats underline the need for fi nding tumor markers with outstanding analytical and clinical performance (high sensitivity, speci fi city, predictive value).

8. Table (1):The chemical nature of some tumor markers

9. Table (2):Commonly-used tumor markers

10. Figure (1):Organ-related tumor markers

11. There are 18 FDA-cleared protein cancer biomarkers, with several others being used clinically without FDA clearance. There is a clear need to identify additional biomarkers for optimal patient management. These serological tumor markers were discovered at least 30 years ago. With n o major serological tumor markers introduced to the clinic since then (although a few genomic markers were Food and Drug Administration (FDA)-approved for predicting therapeutic response). Numerous cancer researchers and organizations have strived to identify novel biomarkers that can fulfil these unmet clinical needs. So : The Need for New Cancer Biomarkers:

12. why are there so few new biomarkers entering the clinic ? * and can anything be done about it?

13. Table(3) Cancer biomarkers and their use

14. Table(3)( cont.)

15. Table (4) FDA approved protein cancer biomarkers

16. (a) Fake publications (very rare ). (b) Discovery of markers with weak performance characteristics such as low sensitivity, specificity or predictive value , precluding their clinical utility . (c) False discovery, i.e ., reports on tumor markers, which initially promise to revolutionize cancer management but which subsequently fail rigorous validation . REASONS FOR BIOMARKER FAILURES : There are three reasons of newly discovered biomarker failures :

17. In addition ,the discovery of new cancer biomarkers is still relatively slow because identification of the thousands of proteins that are dysregulated in cancer and shed into the body fluids poses many technological challenges . Advances in genomic and proteomic technologies e.g gene array , serial analysis of gene expression(SAGE),2D - PAGE ,mass spectrometry and the chip technology together with developments in bioinformatics will allow for discovery of new biomarkers that are both sensitive and specific .

18. Why is it highly unlikely to identify a single marker which will be elevated in nearly all patients with a specific malignancy, with adequate sensitivity and speci fi city to be used in the clinic? * First, it is now known that each site-speci fi c cancer has histological sub-types with different origins and mutational spectra, such that the subtypes can be considered different diseases. *Second, the latest genomic advances in oncology are suggesting that tumors are highly heterogeneous, and no two tumors (with some exceptions) have the same mutational spectrum. Even within the same tumor , molecular heterogeneity is enormous and differences can be seen in primary vs. metastatic sites or as tumors evolve over time ., UNDISCOVERED, HIGHLY SENSITIVE BIOMARKERS A RE UNLIKELY TO EXIST :

19. How then are some currently used tumor markers elevated in most patients, especially at the advanced stages? Among the reasons are the following: 1) Some markers are tissue-speci fi c (such as prostate-speci fi c antigen (PSA) and their elevation in serum is due to leakage of the highly abundant PSA molecules in prostate cells (benign or malignant) into the adjacent blood vessels . T his molecule is not known to be involved with the initiation or progression of prostate cancer . These facts also explain why PSA is elevated in non-malignant diseases, such as prostatitis and benign prostatic hyperplasia . Circulating tumor DNA is a similar example of an elevation of a biomarker in nearly all patients with cancer, especially at late stages. This test is highly sensitive and speci fi c, since the DNA leaks from dying tumor cells.

20. 2) Some other clinically used tumor markers (e.g.: carcinoembryonic antigen; alpha-fetoprotein) are elevated in the majority of patients with cancer because they represent onco-fetal antigens : molecules which were expressed at high levels in fetal life and then re-expressed again in malignancy due to the immature nature (de-differentiation) of cancer cells. 3) A third class of biomarkers that are elevated in many patients include the carbohydrate antigens , highly glycated molecules involved in the ubiquitous processes of cell adhesion and barrier function.

21. Table (5) Biochemical properties and clinical applications of some tumor markers

22. REGULATORY ISSUES: The FDA has established the Voluntary Exploratory Data Submission , which encourages sponsors to share their data with the agency, even at early stages of the development, without being considered as part of the regulatory decision-making process. The agency leans toward establishing a collaborative relationship with the sponsors rather than serving strictly as a regulatory body. More details regarding regulatory processes for biomarker development in the US can be found on the FDA website (www.fda.gov) . Every country has its own regulatory body that governs the requirements for biomarker use within its own medical system, and approval by one country’s regulatory body does not translate into approval by another. How to start searching for new biomarkers :

23. To implement a proper structure in the biomarker discovery process ,the E arly D etection R esearch N etwork (EDRN) has been established by the National Cancer Institue,USA , to improve the coordinaton between biomarker research laboratories. It promotes collaboration among academic and industrial researchers and organizations, both nationally and internationally, for the development and testing of promising biomarkers or technologies for early detection of cancer (http://edrn.nci.nih.gov). A five-phase categorization is suggested to guide the development of a new biomarker .

24. Table (6) Phases of biomarker development

25. Figure (2): The biomarker development pipeline

26. C ancer biomarker researchers can submit their “ rare tumor markers ” as broadly de fi ned above (< 30% sensitivity at > 90% speci fi city) for inclusion into an open access database. To facilitate uniform submissions and inclusion of critical information, a “ submission form ” ( supplementary information ) is created. The provided information should be enough for independent data reproduction. 1) Through this database, researchers could identify candidate tumor markers, which are informative for one or a few patients; then use the biomarker to guide patient management 2) The data would facilitate further investigations (e.g., wholegenome sequencing), to delineate as to why a particular tumor produces the aforementioned biomarker. This information may enrich our knowledge about tumor biology and may pinpoint new therapeutic targets. REPOSITORY OF RARE TUMOR MAKERS

27. Figure(3):Steps for personalized tumor marker development

28. C entralized laboratories should be created to develop and validate highly robust assays for rare tumor markers. Such laboratories could seek clinical laboratory improvement amendments certi fi cation and FDA approval of their tests, to ensure high quality results. Patients/physicians could submit a serum sample from newly diagnosed cancer patients to be used for screening 100 – 1000 candidates , with a goal to identify 1 – 5 biomarkers that are most informative for these patients and thus use them for management . CENTRALIZED TESTING LABORATORY:

29. Preferably, submitted samples are collected before initiation of any treatment and after 4 – 6 weeks post treatment, to identify molecules that are altered by treatment. Although initially, this facility may have a limited assay menu, it is conceivable that over 1 – 3 years, enzyme-linked immunosorbent assay (ELISA) assays for at least 1000 or more markers could be developed.

30. SAMPLE REQUIREMENTS The population used in the validation, including both cases and controls, should be selected to closely match the target population, and the selection process must have clearly defined inclusion/exclusion criteria. Sample size requirements must be calculated to ensure adequate statistical power for each study . Patient-related factors, including fasting, posture, circadian rhythms, age, and sex should be documented for every sample and carefully investigated to explore their relation to the analyte of interest. Additionally, biomarker stability should be studied under different conditions to identify the most efficient protocol for collection, handling, and storage. Finally, extra caution should be given to devices and reagents used during sample collection and storage.

31. A clinically useful biomarker should be measured reliably, so ease of adoption for use by routine clinical laboratories is key. Immunoassays such as ELISA remain the gold standard for validation and clinical use of protein biomarkers . Mass spectrometry– based approaches for clinical applications remain an appealing prospect for numerous reasons including , high analytical specificity and sensitivity and multiplexing capabilities. However, assay complexity and expertise requirements are currently preventing the adoption of mass spectrometry into routine use in clinical laboratories . ASSAY:

32. Basic analytical characteristics to be examined include assay dose–response curve, measuring range, limits of detection and quantification, accuracy,imprecision , and analytical specificity . Adhering to federal regulations that govern human diagnostic testing, known as the Clinical Laboratory Improvement Amendments , during analytical assay validation could increase the reliability and quality of the data . Before assay development, numerous preanalytical variables must be assessed, such as choice of a biological matrix and sample collection, handling, and storage. Depending on the availability of critical reagents (antibodies , calibrators), assay development can take from weeks to years.

33. T he amount of serum necessary to screen 500 or 1000 biomarkers is a challenge. One solution includes multi-parametric assays, such as the Luminex platform ( www.abcam.com ). Recently, other options have emerged. A few companies have recently developed ultrasensitive ELISA assays for many analytes . For example, MesoScale ’ s fi fth generation complexed PSA assay has a sensitivity of 6 fg /ml, which allows quanti fi cation of serum PSA in all women. HIGHLY SENSITIVE ASSAYS:

34. The technology allows for significant sample dilution before the final testing. For example, serum from a normal male can be diluted 1000–10,000 times and is still easily measurable for PSA with such assays. Since most of the known and newly reported cancer biomarkers exist in the circulation at levels of 1 pg /ml or higher, these technologies would allow a 100-fold dilution of the sample before analysis. Consequently, 2–5ml of serum would be enough to screen 1000 or more analytes .

35. Facilities or technologies which can provide quantitative information on thousands of proteins are becoming available now. For example, some companies already offer quantitative assays for thousands of proteins in micro-ELISA array formats using small volumes (< 2 ml) (e.g., www.raybiotech.com ). Also, mass spectrometric selected reaction monitoring assays for entire human and other proteomes have been published.

36. Robust statistical analysis of validation data of potential biomarkers is essential. Possible biases, including small sample size, inappropriate controls, non-independent training and validation cohorts, and multiple hypothesis testing should be critically examined.. Researchers should closely collaborate with biostatisticians that are experienced in the biomarker field to help plan and perform their study. With the advent of - omics technologies such as RNA microarrays and the development of multigene signatures, statistical analysis of validation studies became more complex,necessitating the development of newer and more appropriate methodologies, based on multisignal readings . STATISTICS :

37. As a biomarker moves from one phase of development to another, the financial burden increases. Soon after discovery and early validation studies, usually taking place in an academic setting, industry may enter the scene and provide financial support for the remaining development phases. Being able to clearly demonstrate that a biomarker addresses an unmet clinical need by carefully planning and executing early validation studies is critical to acquiring an industry partner. A potential limitation at that stage may be issues related to intellectual property. FUNDING:

38. A key element of biomarker commercialization is ownership, and industrial partners will not undertake the development of a marker that does not have potential for investment return, regardless of its clinical utility. Therefore investigators should apply for intellectual property rights as early as possible in the biomarker discovery process.

39. Because of their high financial risks, diagnostics have been regarded by industry as a less attractive investment opportunity than therapeutics, and consequently the development of diagnostic methods is not as well funded by industry as the development of therapeutic drugs . This reluctance to invest in diagnostics poses a challenge for biomarker development and further highlights the importance of carefully defining the clinical question to be addressed by the biomarker before initiating any biomarker discovery efforts.

40. Table (7): Causes of biomarker failure

41. The Example of Urokinase -Type Plasminogen Activator and Its Inhibitor Plasminogen Activator Inhibitor Type 1 Taking everything into account, more than 20 years of intense research and collaboration between academia, industry , physicians, and patients were required for demonstrating the clinical utility of these 2 protein cancer markers. Of note, uPA and PAI-1 are found in the list of recommended prognostic tumor markers for breast cancer published by the American Society of Clinical Oncology ; yet they are still not FDA cleared. On the basis of this example, one could conclude that , even in the case of biomarkers with clear clinical utility that are reproducible in independent studies, have analytically verified assays, and have achieved high levels of evidence for clinical validation, the process from bench to the clinic could last 2 or more decades .

42. In general, the greater the clinical impact the marker has, the less time it will take to be adopted for use in the clinic. Human epididymis protein 4 , for example, was first reported as a potential ovarian cancer biomarker in the late 1990s , and obtained FDA approval for monitoring recurrence and progression of ovarian cancer in 2010. PSA took approximately 6 years from its first detection in serum to FDA clearance for monitoring and another 8 years for its clearance for screening .

43.   Cancer researchers are turning to proteomics (the study of protein structure, function, and patterns of expression) and proteogenomics (the integration of proteomics with genomics and gene expression analysis, or transcriptomics ) with the hope of developing novel biomarkers that can be used to identify cancer in its early stages, to predict the effectiveness of treatment, and to predict the chance of cancer recurrence . What research is under way to develop more accurate tumor markers?

44. The most important markers that have merged from microarray analysis include :estrogen-progesterone receptor protein expression, HER2 gene protein alterations,17q23 genomic amplifications and cyclooxygenase-2 protein expression (all for breast cancer), IGFBP-2 protein expression for prostate cancer , vimentin protein expression for kidney cancer and Myc and A1B1 protein exression for HCC .However ,they are still lacking a unifying bioinformatics resource . The ONCOMINE is a cancer microarray data-base and web-based data- mining platform that aims at facilitating discovery from genome-wide expression analysis is now created.

45. Liquid biopsies , a new approach to studying tumors in which bits of tumor material—including DNA and other molecules as well as whole cells—that are released from tumors are analyzed in bodily fluids such as blood or urine, may yield additional new biomarkers. As of May 2019, FDA has approved one liquid biopsy test: the cobas ® EGFR Mutation Test for the detection of EGFR gene mutations in circulating tumor DNA of patients with lung cancer. FDA is prioritizing its review of several new liquid biopsy tests (e.g., Foundation One® Liquid, Guardant360®, Signatera ™), for which it has granted breakthrough and expedited access pathway designations .

46. Cancer biomarker development encompasses multiple contiguous phases, requires collaboration among different stakeholders, carries a major financial burden and faces many potential challenges. Even when all of these challenges are overcome, the history of clinically useful biomarkers suggests that at least a decade is required for the transition of a marker from bench to the bedside . Therefore, it may be too early to expect that the new technological advances will catalyse the anticipated biomarker revolution any time soon . MESSAGE :

47. F uture cancer biomarkers should fulfill the following additional predictors : Treatment or no treatment Surgery or no surgery Surgery and or radiation Extent of surgery ( e.g mastectomy or lumpectomy) When to employ systemic drug therapy

48. Key points :

49. Vathany Kulasingam , Ioannis Prassas and Eleftherios P. Diamandis :Towards personalized tumor markers ,Precision Oncology (2017) 1:17 Sabarni K Chatterjee and Bruce R Zetter : Cancer biomarkers : knowing the present and predicting the future ,Future Oncology ( 2005 ) 1 (1 ),37-50 Maria P. Pavlou , Eleftherios P. Diamandis , and Ivan M. Blasutig :The long journey of cancer biomarkers from the bench to the clinic ,Clinical Chemistry 59:1,147-157 (2013). Literature Sources :

50. THANK YOU