Personalized medicine is a relatively new field that is being studied vigorously, because ultimately, all cancer patients should receive a personalized treatment regimen. In this section, some of the advancements in the clinic are discussed, including Trastuzumab in treating breast cancer, Imatinib in treating chronic myelogenous leukemia, Gefitinib/ Erlotinib in EGFR inhibition and finally, cancer immunotherapy.
Trastuzumab and breast cancer
Trastuzumab, also known as Herceptin, is a monoclonal antibody that targets and interferes with the HER-2 receptor, and it is the most commonly used therapy to treat patients with metastatic breast cancers (1). HER-2 is a transmembrane receptor that normally signals for cell proliferation and angiogenesis, and it is a predictive marker that provides information on whether or not Herceptin is a beneficial therapy for patients. There are many ways to detect the expression levels of HER-2 in breast cancer patients, and the two most common methods are fluorescent in situ hybradization and immunohistochemistry. Herceptin can only effectively target tumour growth with detected elevation of HER-2 receptor expression, which is found on both primary tumours as well as metastatic tumours. Herceptin disrupts cancer cells by inducing HER-2 receptor internalization and degradation, where a downregulation of the receptor disrupts dimerization and its downstream internal cellular signalling. Ultimately, this results in cell arrest at G1 and inhibits cell proliferation. As well, Herceptin can inhibit angiogenesis by upregulating anti-angiogenesis factors, and inhibiting cleavage of the HER-2 extracellular domain, the active domain of the receptor via different mechanisms (5). Herceptin can also be used in combination with chemotherapy, and it has been shown to increase survival rate of patients (5). However, relapse of the disease appears to take place within one year of treatment, so many new strategies revolving HER-2 targeting are still in preclinical and clinical development stages. Many side effects that exist include heart problems, such as a decline in heart function, irregular heartbeats, high blood pressure, and heart attacks; lung problems, such as shortness of breath or fluid in or around the lungs, if untreated, can lead to death of the patient (2).
Imatinib and chronic myelogenous leukemia
Imatinib is a competitive tyrosine-kinase inhibitor used to treat patients with chronic myelogenous leukemia (CML). Tyrosine kinases are important in regular cellular signalling by acting as switches to control other downstream proteins, and the two classes are: receptor and non-receptor tyrosine kinases. c-Abl is a non-receptor tyrosine kinase that is expressed in all cells. In the nucleus, c-Abl regulates cell differentiation, division, apoptosis, and stress response to DNA damage. In the cytoplasm, it has a role to mediate integrin binding. In patients with CML, c-Abl is deregulated from the result of a mutation creating the Bcr-Abl, and is constantly activating proteins for cell growth and proliferation enhancement (4). The kinase domain is the active portion of the protein, and it phosphorylates subsequent substrates. Imatinib targets and binds the kinase domain of the mutated Bcr-Abl, keeps the protein in its inactive form, and hence, reduces its activity and decreases the activation of proteins involved in cell proliferation or anti-apoptosis (4). However, resistance to Imatinib progressively develops as the protein tends to shift its equilibrium from closed to more open conformations. Some adverse effects of Imatinib include nausea, diarrhea, muscle cramps, and heart failures if treated with large doses due to toxicity.
Gefitinib/ Erlotinib in EGFR inhibition
Epidermal growth factor receptor (EGFR) is a receptor expressed on cell surfaces that activates upon binding of its ligands, predominantly epidermal growth factor and transforming growth factor α (TGFα), initiates an intracellular signal cascade leading to DNA synthesis and cell proliferation (4). Mutations in the EGFR gene result in upregulation of the receptor or increased activation, leading to uncontrollable cell growth and proliferation, and ultimately, cancer (8). Hyperactivity of EGFR is treated by gefitinib and erlotinib. Gefitinib and erlotinib, are both EGFR inhibitors that target the tyrosine kinase domain of the EFGR receptor. In order for epidermal growth factors to successfully transfer its signal to downstream messengers in the cell, the receptors need to dimerize and phosphorylate each other to generate phosphotyrosine residues for properly signalling of downstream proteins. These block the binding of phosphate groups and prevent the cross-transfer of phosphate groups within receptor homodimers, thus inhibiting recruitment of downstream signals for cell growth and proliferation. EGFR inhibitor targets the tyrosine kinase domain of EGFR, and it provides fewer side effects as compared to traditional therapy (6). However, some common side effects of both gefitinib and erlotinib include skin reactions, diarrhea, loss of appetite, and fatigue.
Predictive Biomarkers for Personalized Medicine
Predictive biomarkers are a type of cancer biomarker which is used to predict the likelihood that a tumour will respond to an administered drug. This allows a certain level of “personalization” to be considered in the treatment regimen. Many cancer drugs that have been developed target specific molecular aberrations. The problem with this is that these are usually effective on a small subset of the patient population. Proper patient selection for administered drugs allows for an improvement of the cost effectiveness of the therapy together with safer drugs with increased clinical benefit. With predictive biomarkers, there is expected to be an overall decrease of unnecessary treatment and adverse effects (e.g. Adverse Drug Reactions). Research in this field is adding another layer of diagnostic tests which will allow an improved benefit economically as well as clinically. Below are a list of some predictive biomarkers that are given increasing amounts of attention in cancer research and clinical settings.
Trastuzumab and HER2 Associated Breast Cancer
Amplification and overexpression of the HER2 is observed in about a quarter of patients with breast cancer. Those with tumours which have this mutation have an overall poorer prognosis than other types of breast cancer. Trastuzumab is a recombinant humanized monoclonal antibody which targets HER2 protein. It has a major impact on patients with HER2-positive breast cancer and has shown to even reduce breast cancer relapse by 50%. The accurate detection of HER2 status as predictive biomarkers is still being improved upon. This is important as many therapeutic decisions are increasingly relying on the level of HER2 expression.
APL and All Trans Retinoic Acid (ATRA)
Acute Promyelocytic Leukemia (APL) constitutes about 10% of all acute myeloid leukemias. Almost all cases of APL can be observed with a translocation between chromosome 17 and 15 which leads to the fusion of the retinoic acid receptor RARα gene with the PML gene. The presence of this fusion is used as a biomarker for APL for the determination of an appropriate treatment regimen.
On a clinical level, the resulting protein fusion binds to retinoic acid response elements in the promoter region of target genes which recruit co-repressors such as N-CoR and histone deacetylase (HDAC) which leads to the repression of retinoic acid responsive target genes. This fusion protein binds co-repressors with more affinity than the RARα protein resulting in ATRA levels to not be as effective in physiological levels. The understanding of the mechanism of the fusion protein allowed an increase in disease remission in up to 90% of newly diagnosed patients. In a clinical setting, a higher dose of ATRA for selected patients overcame the problem faced with the fusion protein. Relapse is common, however combination therapy with HDAC inhibitors and arsenic trioxide which targets the PML-RARα fusion protein has proved effective. Overall, the detection of such biomarker allowed for a more specific diagnosis and treatment regimen.
PARP inhibitors and BRCA
BRCA1 and BRCA2 are both involved with the DNA damage response in the homologous recombination repair pathways. Mutations in this genes significantly increase the susceptibility of people to acquire breast cancer. It is interesting to note that mutation resulting in the loss of the protein counterparts of the BRCA genes result in the use of compensatory pathways. These pathways are more sensitive to therapeutic drugs such as PARP inhibitors for tumours lacking BRCA1 and BRCA2. Overall, PARP inhibitors provides little efficacy in terms of targeting tumours; however, these inhibitors prove effective in BRCA1- and BRCA2- tumours. The inhibition of PARP then results in cellular apoptosis. Thus, the detection of the double negative mutant allows a more targeted therapy with the use of PARP inhibitors.
Preventative cancer vaccines
The main goals of cancer immunotherapy are to reject cancer cells by stimulating self protective antitumor response, as well as, to mitigate the effects of immune signals that stimulate tumor growth (9). Methods of therapy include immunization of cancer vaccines or therapeutic drug administration. Viruses and bacteria have been known to cause different types of cancer, and certainly, cancer vaccine is an effective therapy against cervical cancer and stomach cancer. The most commonly known vaccine is the human papilloma virus (HPV) vaccine, and it aims to preventing viral acquisition which results in a high efficacy against HPV (9).
Therapeutic antibody vaccines and adoptive T cell therapy
Many therapeutic vaccines available for cancer treatment are based on monoclonal antibodies that target specific antigens on tumour cells. Two examples include trastuzumab, targeting the HER-2 receptor in breast cancer patients, and panitumumab, which targets EGFR in colorectal cancer (9). Adoptive T cell therapy uses tumour reactive CD8+ cytotoxic T cells extracted from a patient's tumour after surgical removal of the tumour bulk. This T cell population is scaled up in culture and then injected back into the patient in order to mount an immune response against any remaining tumour tissue. Adoptive T cell therapy may cause serious immune overreactions in patients, which has so far limited its effectiveness (9).
Limitations of Personalized Cancer Therapies
The goal of personalized cancer therapies is to increasing effectiveness while reducing toxicity to the patient, based on molecular characteristics of the tumor, its microenvironment and the patient's physiology. Research is driven by the above examples of success, rapidly increasing speed and decreasing cost of sequencing technologies, and an array of available molecular therapeutic targets and modes of delivery. However, newspaper headlines are not yet announcing the end of cancer. Below are some of the current limitations to the development and efficacy of personalized cancer therapies.
Tumor composition and characteristics in terms of biomarkers vary between patients and individual tumors. Within the tumor, cells maybe also exhibit functional heterogeneity, differing in ability to self-renew, invade new tissues or develop metastases. This can also be affected by interactions with different microenvironments, epigenetics, or the establishment of a balance between two steady states (10). Tumors that share a common initiation point may have acquired additional mutations or changes in copy number (10). Mutations may simply be 'passengers' rather than 'drivers' of tumor growth, and may be impacted by tumor lineage or presence of concurrent mutations (10). Thus the 'functional relevance' of a biomarker may differ, especially between cancer types and subtypes (10). Importantly, heterogeneity of mutations/populations of subclones may be spatially defined, even within the same tumor or metastasis. A biopsy that accesses only part of the tumor may miss a valuable therapeutic target, or may falsely represent a mutation as more ubiquitous than it is (10). Importantly, it should be noted that not all identified mutations are currently targetable (10). Some overexpressed gene products may be easily subdued with inhibitors or monoclonal antibodies. However, some gene proteins do not present this potential. Loss of function mutations can be critical to cancer cell survival, but cannot be thusly treated.
One approach to overcoming tumoral heterogeneity is to make use of deep sequencing (at a depth of 500X to 1000X) to distinguish rare and dominant clones. Because very rare mutants can be detected, a timeline of genetic aberrations can be established, indicating early genetic targets that will therefore be present in the majority of cells. Since the tumor is continually evolving, future biopsies would need to be completed to establish if any new mutations could limit therapeutic potential (10). This approach would clearly be associated with elevated costs, need for increased capacity for bioinformatic data analysis, and possible discomfort to the patient. Alternatively, multiple rounds of lower depth sequencing could be used to identify and target dominant clones at each stage of treatment (10). New mutations leading to resistance would thus become the next round of targets. This approach is less likely to lead to success because the evolution of a untreatable resistant clonal population is likely.
Clonal populations can change according to selective pressure, such as radiation treatment or even a targeted therapy. Resistance to personalized therapies could be the result of existing multiple mutations, selection of resistant subclones, or generation of new mutations. Imatinib resistance developed in some patients as a result of novel mutations in the targeted Bcr-Abl gene that decreased binding affinity with the drug (11). Such resistant protein would be difficult to target. Thus, targeted therapies may be best suited to use in combination with more traditional therapies, or as part of a cocktail (11).
The microenvironment's impact on cancer cell survival can be referred to as the impact of non-cell autonomous determinants (12). Microenvironmental effects may change the context for a given cell, allowing its function to change. It may also contribute to drug resistance. Several studies have implicated stromal cell secretions in acquired drug resistance (12). For instance, stromal cell secretions of hepatocyte growth factor were linked to poor response to BRAF mutant inhibition targeted therapy. When HGF was itself inhibited, drug resistance to BRAF inhibition was reversed (12). While the microenvironment complicates the delivery and effectiveness of targeted therapies, it is possible to predict and respond to some factors. However, it will be a great challenge to elucidate and minimize all the effects of the microenvironment can have on tumor evolution.
There are some concerns that the current design of clinical drug trials is not amiable to determining efficacy of personalized cancer therapies. For instance, if an inhibitor is tested only on a population of patients with a particular biomarker, it is impossible to determine whether the drug is broadly or specifically active (13). The biomarker itself may be an indicator of greater or lesser survival chances, making it difficult to compare this subpopulation with the total population's survival rates under traditional therapies (10). It could be informative and effective to incorporate analyses of biomarkers into clinical trials so that they can be used to inform dosage schedules and provide additional information to researchers (14).
Biopsies before, during and after personalized treatment trials are essential to understanding the effectiveness of the drug being tested. What if the patient did not have the associated biomarker to begin with? How does resistance develop? However, biopsies are associated with risks, and when given the option most patients opt out (13). Ethical concerns lead to biopsies typically being optional in such clinical trials. This results in a deficit of critical, informative molecular data. It can be difficult to draw meaningful conclusions on treatment efficacy from such trials.
One possibility for the study of personalized cancer therapies is the idea of "n of 1" trials, wherein an individual patient is the only participant. This study design could be used in the early stages to demonstrate effectiveness and ability to reduce progression before larger trials, and to obtain funding or investment in novel treatments (10). These studies would obviously require cooperation from pharmaceutical companies and government agencies, such as the FDA, that oversee clinical trials.
As with all drugs, determining effectiveness is complicated by a host of factors: determining optimal dosages, validating specificity, pharmacodynamics, measuring toxicities (13). Since many of the personalized therapies described above may be administered in combination with a more traditional therapy, issues of additive toxicities, schedule of treatments, and drug interactions further obscure clinical results.
The amount of data from deep sequencing is enormous and requires experts in bioinformatics to efficiency interpret. This adds to the difficulties of translational medicine, or how to get treatments from the research bench to the bedside, and feedback in the opposite direction. This can involve new roles, new clinical teams and relationships, and could require new regulations (15). Making this data comprehensible to doctors and useful for prescribing appropriate treatments is key.
To this end, a discipline known as systems biology has arisen. Systems biology combines physics, mathematics and engineering to generate visual representations of what is happening inside a cell based on such factors as environmental inputs and genetic abnormalities. The idea is that in addition to aiding in the recognition of biomarkers driving cancerous cells and their therapeutic potential, systems biologists will also be able to use signaling and regulatory pathways to predict unintended effects of a drug and determine ideal drug combinations (16). The development of system biology would be key to individualized drug trials and investigation of tumors with atypical response to treatment. Systems biology would address the challenges of personalized cancer therapies by revamping cellular biology to match the technological possibilities of high throughput sequencing. It could ideally incorporate microRNA, gene expression profiles, proteomics and epigenetics to provide the most complete picture of cancerous tissues possible (16).
Another area of exciting research is the one pertaining to oncolytic viruses. As the name suggests, oncolytic viruses are viruses which have been engineered to lyse and destroy specifically cancer cells.(18) In other cases, viruses employed may not be engineered but simply asymptomatic when infected. Candidate oncolytic viruses are based on their ability to selectively infect and lyse cancer cell lines. Oncolytic viruses can attack the cancer in two ways. First is to induce cancer death by physically lysing the cell by means of viral replication. The other mean is through reactivation of the patient’s immune system to recognize and kill cancer cells.(19) Viral replication results in expression of viral surface proteins on the cancerous cells which may activate immune cells to recognize previously unrecognized cancer cells. There is much promise in the field of oncolytic viruses. Many clinical trials are in progress with many successes. OncovexGM-CSF, an HSV-1 derivative, combines both lytic and immune activation methods and is close to becoming the first approved OV in the western world.(20) In China, Oncorine is already approved, but efficiacy is less than ideal. Oncolytic viruses can also be employed as detection and diagnostic tools as well. Through engineering of special payloads that add channels to allow radioactive label entry, viruses that only infect cancer cells allows them to be imaged through radioactive imaging as in PET scanning. This allows for locating and determining cancer mass and metastasis.(21)
10. Meric-Bernstam, F., Mills, GB. 2012. Overcoming implementation challenges of personalized cancer therapy. Nature Reviews Clinical Oncology, 9(9): 542-8.
11. National Cancer Institute. Targeted Cancer Therapies Fact Sheet. Last modified: 12/05/2012. http://www.cancer.gov/cancertopics/factsheet/Therapy/targeted
12. Masuda, S, Izpisua Belmonte JC. 2013. The microenvironment and resistance to personalized cancer therapy. Nature Reviews Clinical Oncology. 10(2):79.
13. Schilsky, RL. 2011. Drug Approval Challenges in the Age of Personalized Cancer Treatment. Personalized Medicine, 8(6): 633-40.
14. Wistuba, II, Gelovani JG, Jacoby JJ, Davis SE, Herbst, RS. 2011. Methodological and practical challenges for personalized cancer therapies. Nature Reviews Clinical Oncology. 8(3): 135-41.
15. Norris Cotton Cancer Center News: The Challenge and Opportunity in Developing Personalized Cancer Therapy. Written by Steve Bjerklie, 01/03/11. http://cancer.dartmouth.edu/about_us/newsdetail/58573/
16. Gonzalez-Angulo, AM, Hennessy BT, Mills GB. 2010. Future of personalized medicine in oncology: a systems biology approach. Journal of Clinical Oncology, 28(16): 2777-83.
17. Bedard, PL, Hansen Ar, Ratain, MJ, Siu, LL. 2013. Tumour heterogeneity in the clinic. Nature. 501: 355-64.
18. Bell J, McFadden G (2014) Viruses for Tumor Therapy. Cell Host & Microbe 15:260-265
19. Dave, R.A., Jebar, A.H.S., Jennings, V.A., Adair, R.A., West, E.J., Errington-Mais, F., Toogood, G.J., Melcher, A.A. (2014) Viral Warfare! Front-line defence and arming the immune system against cancer using oncolytic vaccinia and other viruses. The Surgeon 12:210-220
20. Senzer, N.N., Kaufman, H.L., Amatruda, T., Nemunaitis, M., Reid, T., Daniels, G., Gonzalez, R., Glaspy, J., Whitman, E., Harrington, K., Goldsweig, H., Marshall, T., Love, C., Coffin, R., Nemunaitis, J.J. (2009) Phase II Clinical Trial of a Granulocyte-Macrophage Colony-Stimulating Factor–Encoding, Second-Generation Oncolytic Herpesvirus in Patients With Unresectable Metastatic Melanoma. J Clin Oncol 27:5763-5771
21. Brader, P., Wong, R.J., Horowitz, G., Gil, Z. (2012) Combination of pet imaging with viral vectors for identification of cancer metastases. Advanced Drug Delivery Reviews 64:749-755