Health Articles

 

Journal.library.iisc.ernet.in

Journal of the Indian Institute of Science A Multidisciplinary Reviews Journal ISSN: 0970-4140 Coden-JIISAD Indian Institute of Science One Size does not Fit All—The Future of Cancer Therapy Sujaya Srinivasan and Kumaravel Somasundaram* Abstract Cancer is a complex disease where normal cells of the body are
transformed such that they begin to divide in an uncontrolled manner and can even invade other tissues in the body. Cancer can occur in many differ-ent tissues in the body, each requiring different forms of treatment. It is a dis-ease that is caused by changes or mutations in genes, leading to a cascade of other genetic changes in the body. There is a high degree of genetic het-erogeneity in tumors of a single type of cancer, which might explain why each patient responds to standard treatments dif erently. This makes it necessary to tailor treatments for cancer patients based on the molecular profiles of their tumors. This is the idea behind personalized medicine, where patients are treated based on their individual genetic changes or molecular profiles. In this paper, we look at some of the molecular profiles that are commonly used Mutation: It is a permanent
in dif erent types of cancers, and some personalized therapies that already change in the DNA sequence exist. In addition, we also attempt to predict how cancer treatment wil be of a gene resulting in a defec-tive gene and can alter the revolutionized by some of the new technologies that are emerging today.
amino acid sequence of the protein encoded by the gene Keywords: Cancer, oncology, personalized medicine, targeted therapy, gene signatures, molecular
resulting in the production of signatures, genomics, next generation sequencing a non functional protein.
Prognosis: It is a prediction
of the probable course and
1 Heterogeneity of Cancer
Due to this heterogeneity, even within a single outcome of a disease or the Cancer is a disease of the cells in the body, where type of cancer, for example breast cancer, mortal- likelihood of recovery from a disease.
the normal processes of cell division and cell death ity and prognosis of patients differ widely. Some go awry. Cells divide uncontrollably and live for- patients respond very well to standard surgery, Genetics: It is a branch of
ever. It is estimated that 1 in 3 individuals develop chemotherapy and radiotherapy treatments, while biology that deals with the molecular structure and cancer at some time in their life. According to the the same treatments are ineffective in others. On function of genes, heredity, medical journal Lancet, it is estimated that in India the other hand, patients who do not have very and variation in living alone, 555,000 people died of cancer in 2010.1 aggressive forms of cancer are overtreated with Cancers can develop in various tissues in the standard chemotherapy regimens, which may lead Epigenetics: It is the study of
body, and can manifest themselves in different to other undesired side-effects. Cancer research heritable alterations in gene ways. Cancer is caused by mutations in genes that over the last decade has uncovered that not one, expression or cellular pheno-type caused by mechanisms can further lead to other genetic and epigenetic but many genes can play a role in cancer devel- other than changes in DNA changes. In spite of a few genetic and epigenetic opment and progression. This explains why some changes that dominate a particular cancer, each patients require different forms of treatment, Gene: It is a molecular unit of
tumor differs from the rest. Study of tumors shows which are not based on the phenotype or mani- heredity of a living organism that there is a lot of genetic heterogeneity among festation of the disease, but on their underlying and is a segment of DNA on a them, leading to subtype classifications on the genetic or epigenetic changes.
specific site on a chromosome that is responsible for an basis of molecular profiles. We therefore need to inheritable trait.
understand cancer as an accumulation of acquired 2 The Impact of Genomics Research
Department of and inherited gene mutations, in addition to epi- on Cancer
Microbiology and Cell genetic changes that could be caused both by both The main goals of cancer therapy are focused on Biology, Indian Institute interactions in the tissue microenvironment as prevention, detection and treatment. In order to of Science, Bangalore 560012, India. well as the external environment.
treat cancer effectively, it is necessary to look at the Journal of the Indian Institute of Science VOL 92:3 Jul.–Sept. 2012 journal.library.iisc.ernet.in Sujaya Srinivasan and Kumaravel Somasundaram Genomics:
entire genome as a whole, rather than individual for pathologists to distinguish between Grades It is a discipline in genetics genetic changes. In a disease like cancer, due to 3 and 4, which are malignant forms of glioma. concerned with the study of the genomes of organisms. the heterogeneity of genetic changes, the taxon- Grade 3 glioma, also known as anaplastic astro- The field includes efforts to omy and classification of tumors is increasingly cytoma (AA), is less malignant and aggres- determine the entire DNA depending on molecular profiles, rather than mor- sive, whereas Grade 4 glioma, also known as sequence of organisms and fine-scale genetic mapping.
phologic characteristics. This means that genomic glioblastoma (GBM), is extremely aggressive and Proteomic:
and proteomic profiles of patient tumors can help has very poor prognosis. We have developed a It is the study of all the pro- decide what type of treatment they should receive, molecular signature based on microRNA expres- teins expressed by a given cell, and predict the prognosis of patients.
sion that can accurately classify Grade 3 and 4 tissue, or organism at a given time. Given time and under Personalized medicine is a form of therapy tumors.2 The expression profiles of 23 microR- that uses the information about a person's genetic NAs is used to accurately classify anaplastic astro- MicroRNA:
and protein profile to diagnose and treat a disease. cytoma (Grade 3) and glioblastoma (Grade 4) It is a small RNA molecule This form of medicine is especially promising for samples (see Figure 1). We have also developed encoded by the DNA, highly a disease like cancer, which exhibits a large degree a signature based on microRNA expression that conserved and can regulate the expression of genes.
of heterogeneity. Personalized medicine aims to can sub-divide GBM patients into high-risk Protein:
end the idea that one treatment can work for all and low-risk groups, and predict survival.3 Sev- It is a large molecule composed patients, and attempts to find therapies that fit the eral mRNA signatures have also been developed of one or more chains of amino individual needs of a patient.
for GBM4,5 that can help in prognostication and acids in a specific order, which is determined by the RNA The idea behind personalized medicine is that derived from a gene. Each pro- a patient's genomic (genetic changes in DNA), Over the last decade, several research groups tein like a hormone, enzyme, or transcriptomic (RNA levels) and epigenetic (her- have developed molecular signatures for various antibody has unique functions.
itable changes in gene expression without changes cancers, but very few of these have been used in the Deoxyribonucleic acid (DNA):
It is a nucleic acid containing in the underlying DNA) profiles are characterized clinic. A signature that is developed in one patient the genetic instructions used in from which target biomarkers that are known to cohort, very frequently needs to be validated in the development and function- be involved in cancer development and progres- another patient cohort, due to the underlying het- ing of all known living organ- isms and it consists of two long sion are identified. Identification of these criti- erogeneity of tumors and use of different platforms polymers of simple units called cal genes that play a key role in carcinogenesis for the validation process. Unless these signatures nucleotides, with backbones involves many scientific challenges. Once these are applicable in widely varying cohorts of patients made of sugars and phosphate groups joined by ester bonds.
gene targets are identified, drugs that target these and tested in multiple cancer centers, clinicians are genes can then be developed.
unable and reluctant to use them on patients.
Ribonucleic acid (RNA):
It is very similar to DNA, but Several examples of personalized medicine Despite the difficulties in translating molecular differs in a few important exist in clinics today. Most of them involve some signatures from research labs to the clinic, there are structural details and is usually form of molecular profiling to decide the type a few success stories. The Oncotype DX™ Breast single-stranded, while DNA is usually double-stranded in the of treatment that should be given. In some cases, Cancer Assay and the Oncotype DX™ Colon Can- cell. RNA is transcribed from a group of genes that is known to be deregulated in cer Assay (Genomic Health, Redwood CA), Mam- DNA by enzymes called RNA the specific cancer, or is a "signature" of a certain maPrint® for breast cancer (Agendia, Amsterdam, polymerases and it carries the information from a gene to subtype of the cancer is profiled to better classify The Netherlands) and the Breast Cancer Gene form a protein.
the tumor. This subtyping not only helps in decid- Expression ratio (HOXB13:IL17BR) test (Quest Molecular signature:
ing treatment, but in some cases is also able to Diagnostics, Lyndhurst, NJ) are used by oncolo- It is a set of genes, RNA tran- predict the prognosis of the patient. In other cases gists to help decide treatment. In glioblastoma, scripts, proteins, genetic vari- ants or other variables that can of targeted therapy, a certain gene or protein is the methylation status of the MGMT gene is used be used as markers for a par- overexpressed (present at a much higher level than to determine if temozolomide (a chemotherapy ticular cell or tissue phenotype, normal) in tumors, and a drug that targets this drug) treatment will be effective.
such as a presence of a cancer.
specific gene can be used to inhibit tumor progres- The Oncotype DX® Breast cancer test is a diag- Polymerase chain
reaction (PCR):
sion. Some examples of molecular signatures and nostic test that helps identify the subset of women It is a molecular biology targeted therapy are discussed in the next section.
with early-stage, estrogen-receptor positive and technique to amplify a single lymph-node-negative stage I and II breast can- or a few copies of a piece of DNA across several orders of 3 Molecular Signatures for Cancer
cer that are likely to benefit from adding chemo- magnitude, producing thou- Diagnosis and Prognosis
therapy to regimen, which includes Tamoxifen, an sands to millions of copies of Molecular signatures of tumors aim to clas- anti-estrogen agent. This test also helps assess the a particular DNA sequence.
sify tumors into sub-classes, predict prognosis probability of cancer recurrence, which is also a Real-time polymerase chain
and decide treatment. They can provide a more factor in deciding treatment. The assay quantifies reaction (Real-time PCR):
It is a molecular biology accurate means of classification, where histopa- the expression of 21 genes in breast cancer tissues technique based on the PCR, thology-based grading is ambiguous or difficult. by Real-time PCR.6 which is used to amplify and For example, in glioma, which is a cancer of the Oncotype DX™ Colon Cancer Assay is a test simultaneously quantify a targeted DNA molecule.
glial cells in the brain, it is particularly challenging that provides information to clinicians on the Journal of the Indian Institute of Science VOL 92:3 Jul.–Sept. 2012 journal.library.iisc.ernet.in The Future of Cancer Therapy Figure 1: Heat map showing expression profiles of 23 microRNAs that can be used to classify anaplastic
astrocytoma (grade 3) and glioblastoma (grade 4) samples (red indicates that the miRNA is over expressed,
and green that it is under expressed).
Source: Modern Pathology (2010) 23, 1404–1417; doi:10.1038/modpathol.2010.135.
likelihood of disease recurrence in Stage II/III substances that block the growth and spread of colon cancer and the likelihood of tumor response cancer by interfering with specific molecules to standard chemotherapy regimens. It consists of involved in tumor growth and progression". Tar- a 761 gene assay based on RT-PCR.7 geted cancer drugs that have been approved by the The MammaPrint® Test for breast cancer is FDA for use in various cancers include those that based on microarray technology, and uses the inhibit molecules involved in cell growth, prolif- expression of 70 genes to predict which early-stage eration and angiogenesis, and those that promote breast cancer patients aged 61 years or younger are death of cancer cells or stimulate the immune sys- at high-risk for recurrence or metastasis following tem to destroy cancer cells.
surgery, irrespective of estrogen receptor status or The aim of targeted therapy is to target cancer prior treatment. The outcome of the test is to pre- cells, without affecting normal cells. For example, dict if a patient is at a high or low risk of distant if a protein is known to increase cell prolifera- recurrence. If the patient is classified as low risk tion in a certain type of cancer where it is overex- by MammaPrint®, they are treated with tamoxifen pressed, but is not expressed at all or at a very low or other hormones. High-risk patients are treated level in normal cells, a targeted drug can be devel- with aggressive chemotherapy in addition to hor- oped against that protein which would inhibit uncontrolled cancer cell growth, but would have The Breast Cancer Gene Expression Ratio no effect on normal cells.
(HOXB13:IL17BR) developed by Aviara DX and One of the first success stories for targeted licensed by Quest Diagnostics is a test is used in therapy for breast cancer was Herceptin (Trastu- patients that are lymph node-negative and estro- zumab), which targets the HER2 protein. HER2 is gen receptor-positive to predict the risk of disease a gene that is highly expressed in some breast can- It refers to the method of detecting antigens like recurrence. It measures the ratio of expression of cers, and is known to promote cell proliferation proteins in cells of a tissue two genes, HOXB13 (overexpressed in tumors) to and migration, making the cancer highly malig- section by exploiting the IL17BR (under expressed in tumors) to determine nant and aggressive. Trastuzumab is a monoclonal principle of antibodies binding specifically to anti- the risk factor of a patient, and helps clinicians in antibody that interferes with the HER2 protein, gens in biological tissues.
deciding the course of treatment.
preventing cells from proliferating in an uncon-trolled manner. In a routine clinical lab, patient Fluorescence in situ
hybridization:
4 Targeted Therapies for Cancer
tumors are tested for the overexpression of the It is diagnostic method The National Cancer Institute (NCI) of the HER2 through immunohistochemistry (IHC) or utilizing fluorescently labeled National Institutes of Health (NIH) in the United fluorescent in situ hybridization (FISH). Those DNA probes to detect a specific genetic material or States defines targeted therapy as "drugs or other patients who show an overexpression of HER2 are nucleic acid in a cell.
Journal of the Indian Institute of Science VOL 92:3 Jul.–Sept. 2012 journal.library.iisc.ernet.in Sujaya Srinivasan and Kumaravel Somasundaram Table 1: Drugs currently approved for targeted therapy for cancer.
Gastrointestinal stromal tumor, chronic myeloid leukemia, etc.
BCR-ABL and Src family BCR-ABL and other Monoclonal antibody NSCLC, pancreatic Monoclonal antibody HNSCC, colorectal Monoclonal antibody renal cell carcinoma kidney, pancreatic binding protein-12 medullary thyroid cutaneous T-cell retinoid X receptors retinoid X receptors retinoid acid receptors acute promyelocytic Proteosome inhibitor multiple myeloma, mantle cell lymphoma peripheral T-cell colorectal, kidney VEGFR, PDGFR, Raf kinases Small molecule renal cell carcinoma, tyrosine kinases renal cell carcinoma, VEGFR, c-kit, PDGFR renal cell carcinoma Monoclonal antibody B-cell non-Hodgkin Monoclonal antibody Tyrosine kinase: It is
Monoclonal antibody an enzyme that can Monoclonal antibody transfer a phosphate group from Adenosine Source: Fact sheet on Targeted Cancer Therapies—NCI.
5-triphospahte (ATP) to a protein by a process called phosphorylation, which forms treated with Herceptin in combination with other ABL that is formed when a portion of chromo- a important mechanism of chemotherapy or hormonal agents.
some 9 and a portion of chromosome 22 break communication of signals within the cell.
Gleevec is another cancer targeted therapy off and switch places. ABL is a tyrosine kinase drug that has been around since the 1990s. It was that is involved in signalling pathways control- Chromosome: It is an
originally developed as a drug against the tyrosine ling cell proliferation. In the BCR-ABL fused gene, organized structure of DNA and protein found in cells kinase ABL that is activated in chronic myeloid ABL signalling is constitutively active, promoting and is a single piece of coiled leukemia (CML). The Philadelphia chromosome proliferation. Gleevec works by targeting the ABL DNA containing many genes, or Philadelphia translocation, which is commonly protein and inactivating the signals that increase regulatory elements and other seen in CML, is a result of the fusion gene BCR- cell proliferation.
Journal of the Indian Institute of Science VOL 92:3 Jul.–Sept. 2012 journal.library.iisc.ernet.in The Future of Cancer Therapy In the last decade, there have been several tar- as in the 1000 genomes project to identify com- geted therapies that have been approved for vari- mon SNPs, the scientist could opt for a low cover- ous cancers. Table 1 has a list of these therapies age of 5x for each genome and look for common and the cancers they have been approved for. variants across multiple samples.
Patient tumor samples are tested for specific tar- Next generation sequencing (NGS) technolo- gets, to decide if the patient would be a good can- gies promise to change the landscape of personal- didate for the targeted drug in combination with ized medicine and take it to a new level. The cost of other therapies.
sequencing the human genome in the early 2000's was around 100 million dollars, whereas the cost 5 Applications of Next Generation
of sequencing a single individual's genome has Technologies to Cancer
now come down to a few thousand dollars (See Figure 3). With the cost of genome sequencing When the human genome was sequenced in coming down rapidly, it is possible to envision a the early 2000s, the technology used was Sanger day not too far away when every individual can sequencing, which was costly and time-consuming. have their genome sequenced as part of routine Sanger sequencing is frequently referred to as "first clinical lab tests, and the results are used by doc- generation sequencing". "Second generation" or tors to tailor treatments for them. This is espe- Next generation sequencing (NGS) technologies cially true for a disease like cancer that exhibits a overcome the scalability limitations of Sanger high degree of heterogeneity. Doctors can decide sequencing by allowing millions of sequencing the combination of drugs that are likely to work reactions to happen in parallel. Next generation for the specific patient based on his genomic, epi- sequencing methods generate millions of short genomic and transcriptome profile.
reads, which then need to be aligned to generate It refers to a set of all the entire sequence (see Figure 2). There is a lot 6 Future Path for Personalized
RNAs expressed in a given organism or tissue or cell at of flexibility to fine-tune the degree of resolution Medicine in Cancer
a given a time.
for an experiment. For example, in order to With the advent of new technologies that are bring- detect variants with a high confidence, a scientist ing down the cost of sequencing human genomes, would design an experiment to have the genome it is no longer a fantasy to imagine that every indi- sequenced at 50x coverage, which means, that vidual can have their genome sequenced. In the every base, on average, was covered by 50 short future, it is possible to imagine that everyone car- reads. On the other hand, if the experiment ries a smart card with their sequenced genome on involves sequencing a large number of genomes, it, and doctors can then use this information to Figure 2: An overview of next generation sequencing technologies.
Source: www.illumina.com
Journal of the Indian Institute of Science VOL 92:3 Jul.–Sept. 2012 journal.library.iisc.ernet.in Sujaya Srinivasan and Kumaravel Somasundaram Figure 3: Cost of sequencing human genomes over the past decade. In 2001, the cost of sequencing a
whole human genome was around $100 million. The cost has steadily decreased and the current cost of
sequencing a human genome is now less than $10,000.
Source: NHGRI, NIH.
Figure 4: A future that uses genome smart card which helps in personalized treatment.
Journal of the Indian Institute of Science VOL 92:3 Jul.–Sept. 2012 journal.library.iisc.ernet.in The Future of Cancer Therapy determine possible risk factors and tailor thera- are predisposed to certain diseases. Despite the pies that would be effective for them. In the con- challenges, the rewards are enormous, and if the text of oncology, it is possible that in addition to promise of personalized medicine is fulfilled, we using information from the genome, the doctor can all look forward to better and healthier lives.
would also order a transcriptome profile of the tumor sample. This would help the doctor decide which targeted therapies, if any would be effective Authors thank DBT, Government of India for to combat the disease, and what combination of financial support. Infrastructural support by drugs to use (Figure 4).
funding from ICMR (Center for Advanced studies While this reality is possibly not far away, we in Molecular Medicine), DBT, DST (FIST) and should also be cautious to recognize the challenges UGC to department of MCB, Indian Institute we would need to face on the way. While many of Science is acknowledged. We also thank Irene gene signatures and biomarkers have been identi- Rosita Pia Patric for the technical assistance. K.S. is fied in cancer research labs around the world, most a J. C. Bose Fellow of the Department of Science of them have not been translated into practice in and Technology.
the clinic. The reasons for this are manifold—lack of validation in multiple patient cohorts, variabil- Received 7 September 2012.
ity in results, lack of standardization in clinical lab testing and the underlying nature of the disease that makes it difficult to find common features 1. Dikshit, R., et al., Cancer mortality in India: A nationally across large patient cohorts. In addition, trans- representative survey. Lancet. 379(9828): p. 1807–16.
lation of research into the clinic requires a con- 2. Rao, S.A., V. Santosh, and K. Somasundaram, Genome- certed effort and collaboration between bench wide expression profiling identifies deregulated miR- scientists, computer scientists, bioinformaticians, NAs in malignant astrocytoma. Mod. Pathol. 23(10):
biostatisticians and clinicians, so that meaningful p. 1404–17.
results are obtained. The initial costs of setting up 3. Srinivasan, S., I.R. Patric, and K. Somasundaram, A ten- the infrastructure required to support personal- microRNA expression signature predicts survival in ized medicine are also a consideration, especially glioblastoma. PLoS One. 6(3): p. e17438.
in developing countries like India. There are also 4. de Tayrac, M. et al., A 4-gene signature associated with clin- ethical issues to be considered, such as, who would ical outcome in high-grade gliomas. Clin. Cancer Res. 2011. have access to information about an individual's 17(2): p. 317–27.
predisposition to a certain disease or condition, 5. Colman, H. et al., A multigene predictor of outcome in especially with respect to insurance companies glioblastoma. Neuro. Oncol, 2010. 12(1): p. 49–57.
and employers.
6. Cobleigh, M.A. et al., Tumor gene expression and progno- It is clear that personalized medicine, if it sis in breast cancer patients with 10 or more positive lymph becomes a reality, has a lot of benefits for patients, nodes. Clin Cancer Res, 2005. 11(24 Pt 1): p. 8623–31.
with better treatment and fewer side-effects. It is 7. Clark-Langone, K.M. et al., Biomarker discovery for colon also a boon for care-givers, who can customize cancer using a 761 gene RT-PCR assay. BMC Genomics, treatments for their patients, based on their indi- 2007. 8: p. 279.
vidual conditions. In addition to treatment, doctors 8. van ‘t Veer, L.J. et al., Gene expression profiling predicts can also advise their patients on lifestyle changes clinical outcome of breast cancer. Nature, 2002. 415(6871):
and preventive measures that they can take, if they Ms. Sujaya Srinivasan received her B.E in
Dr. Kumaravel Somasundaram is a Professor
Electronics and Communications Engineer- at Department of Microbiology Cell Biology, ing from the University Visveswaraya College Indian Institute of Science, Bangalore, India. of Engineering in Bangalore, and subse- He did his Veterinary Medicine degree (1985) quently a MS in Computer engineering from from Madras Veterinary College, Masters the University of Maryland, College Park. After working in in Biotechnology (1987) and PhD in bacterial genetics the computer industry at Intel Corporation in California (1993) from Madurai Kamaraj University, Madurai, India. for 9 years in various capacities, she obtained her MS in His post-doctoral training was between 1993 to 1999 at Bioinformatics from Johns Hopkins University in 2010. Northwestern University and University of Pennsylvania. She is currently a Senior Research Fellow at the Indian The major focus of his laboratory is genetics of brain Institute of Science. Her interests are in using computa- tional methods to analyse and interpret high-throughput "-omics" data from biological experiments, especially in cancer.
Journal of the Indian Institute of Science VOL 92:3 Jul.–Sept. 2012 journal.library.iisc.ernet.in

Source: http://journal.library.iisc.ernet.in/index.php/iisc/article/download/11/10