TPMT: Thiopurine S-methyltransferase; UGT1A1: Uridine diphosphate Glucuronosyltransferase 1A1; K-RAS (v-Kiras2 Kirsten Rat Sarcoma viral oncogene homologue), BRAF: serine/ threonine-protein kinase B-Raf; ABL: Abelson; BCR: Breakpoint Cluster Region; Ph1: Philadelphia chromosome; c-KIT: v-kit Hardy-Zuckerman 4 feline sarcoma viral oncogene homolog; DPD: Dihydropyrimidine Dehydrogenase; EGFR: Epidermal Growth Factor Receptor; 5-FU: 5-Fluorouracil; G6PDH: Glucose-6-Phosphate Dehydrogenase; HER: Human Epidermal Growth Factor Receptor; 6-MP: 6-Mercaptopurine; Ph1: Philadelphia; G6PD:Glucose-6- Phosphate Dehydrogenase.

Pharmacogenomics utilizes genomic technologies to identify patient’s genetic polymorphisms, making them more susceptible to developing certain diseases or impairing the pharmacologic function of, and therapeutic response to, specific drugs. The ultimate goal of pharmacogenomics is to focus therapy on specific receptors or targets in order to reduce adverse drug reactions (ADRs) and improve therapeutic outcomes by applying pharmacogenomics in selecting targeted pharmacotherapy for patients. Narrow therapeutic indices, low overall response rates, rapid and severe systemic toxicity, and unpredictable efficacy are all hallmarks of cancer therapies. Therefore, nowhere is pharmacogenomics research needed more than in cancer treatment to guide clinicians to better predict the differences in drug response, resistance, efficacy, and toxicity among chemotherapy and targeted-therapy patients, and to optimize the treatment regimens based on these differences [1]. For example, clinical evidence indicates that the steady-state levels of 6-mercaptoputine (6-MP, Purinethon) in acute lymphocytic leukemia (ALL) patients can range up to 10 fold or higher among cancer patients with the same administered drug dose because of the highly variable and polymorphic metabolic enzyme thiopurine methyltransferase (TPMT). It is critical for clinicians to identify those patients with TPMT polymorphisms and then adjust the dose of 6-MP accordingly [2]. Thus pharmacogenomics is the key to a simple question: what dose of chemotherapy is appropriate for a patient that may minimize side effects without compromising patient care? It is critical to integrate pharmacogenomics into the curriculum of nursing, pharmacy and medical education to fully prepare our future clinicians to embrace personalized medicine.

The field of Pharmacogenomics encompasses the interplay between the genome and the drug [3]. From its recent advances, we gain a deeper understanding of how the genetic variability within our population directs the pharmacokinetics and pharmacodynamics of drug therapy. To apply pharmacogenomics to cancer genetics, we need to understand how acquired (somatic) and inherited (germline) variations affect the efficacy and safety of drug therapy [4]. Acquired variations are those that apply to predicting drug efficacy and resistance (Pharmacodynamics), whereas inherited variation is the ability to identify the changes in drug metabolism (Pharmacokinetics). Pharmacogenomics is one of the key factors to take into consideration when designing and selecting a cancer therapeutic regimen. Figure 1 summarizes these factors that can potentially affect drug efficacy and toxicities of cancer treatments, including morphometric, demographic, physiologic and pathophysiologic, pharmacologic and pharmacogenomic factors [1]. In particular, there are multitudes of clinical evidence indicating that interracial and inter-individual polymorphisms in genes encoding for drug-metabolizing enzymes, drug transporters, and drug targets are linked to the toxic clinical presentation of cancer patients [1].

Figure 1: Factors Affect Drug Efficacy and Toxicities [1].

The application of pharmacogenomics in oncology is in the discovery of biomarkers that guide selective therapy, predict toxicities, and target the mechanisms of drug resistance. There are currently around 25 FDA approved targeted therapies that require genetic testing for biomarkers in order to determine appropriate patients who can receive them. Cutting edge pharmacogenetic research plays an essential role to identify these biomarkers that are critical for personalized patient treatments and enable clinicians to minimize the risk of one-size-fit-all or trial and failure patient care approaches. Since the FDA approved the first pharmacogenetic test (AmpliChip CYP450 Test) in 2004 to identify a patient’s CYP2D6 and CYP2C19 genotype by analyzing DNA extracted from a whole blood sample, many protein and DNA based in vitro pharmacogenetic tests have been developed and approved by the FDA. Table 1 lists some examples for the FDA approved phamracogenetic tests for cancer patients [1,2-6].

The study of pharmacogenomics in Oncology ushered in new therapies targeting susceptibilities in cancer cells, with the goal to spare normal cells and thereby create regimens that increase efficacy and decrease toxicities. Genetic testing of individuals for certain biomarkers allows the clinician to then tailor the treatment regimen to separate responders from non-responders, saving valuable time and limiting the toxicities associated with regimens not associated with this level of patient specificity. Examples of these biomarkers include EGFR, K-RAS, HER2, c-Kit, B-RAF, and BCR-ABL (Table 1).

Table 1: Examples of Selected Clinically Valid Pharmacogenomic Tests Approved by the FDA for Cancer Patients [1,5,6].

The American Society of Clinical Oncology (ASCO) and National Comprehensive Cancer Network (NCCN) guidelines recommend treatment regimens based on selective biomarkers for common cancers such as colorectal, lung, breast, melanoma, and certain leukemias. Common biomarkers screened for Non-Small Cell Lung Cancer (NSCLC) are EGFR and ALK mutations [1,6]. Based on amplification and/or mutation of the receptor, certain targeted agents recommended in the guidelines are cetuximab or panitumumab for EGFR amplification positive, erlotinib or afatinib for EGFR mutation positive patients, dabrafenib, vemurafenib, or tramenitinib for BRAF mutations, and crizotinib and ceritinib for ALK positive patients [6].

Another example is the Human Epidermal Growth Factor Receptor-2 (HER2) in breast cancer. Similar to EGFR, HER2 is tested for amplification by either immunohistochemistry (IHC) or the fluorescence in situ hybridization (FISH). A positive test indicates treatment with the biological agent trastuzumab, pertuzumab, adotrastuzumab and lapatinib. Some other examples of biomarkers that are relevant to clinical practice are K-RAS in metastatic colorectal cancer, BRAF for metastatic melanoma, and the Philadelphia Chromosome translocation causing the formation of the BCR-ABL fusion gene in Chronic Myelogenous Leukemia (CML) [1].

The Panitumumab Randomized Trial in Combination with Chemotherapy for Metastatic Colorectal Cancer to Determine Efficacy (PRIME) study established the predictive power of pharmacogenetic testing of RAS mutations (KRAS or NRAS) before utilizing Panitumumab in addition to FOLFOX regimen in patients with metastatic colorectal cancer. Progression-free survival (PFS) and overall survival (OS) were both increased in patients with non-mutated KRAS mutations with Panitumumab-FOLFOX compared with FOLFOX treatment alone, increasing the clinician’s armamentarium against metastatic colorectal cancer [7-15].

Another example of how pharmacogenomics revolutionized the colon cancer field comes from the Crystal trial which evaluated the efficacy of cetuximab + FOLFIRI for 1^st line treatment of metastatic colorectal cancer. The study compared FOLFIRI alone against FOLFIRI + cetuximab. The trial concluded in patients with KRAS wild type allele, the addition of cetuximab to FOLFIRI improved PFS and is an acceptable 1st line therapy for patients with KRAS wild type allele with metastatic colorectal cancer according to NCCN guidelines Table 2 [16].

Table 2: Pharmacogenomic Biomarkers for Selection of Cancer Therapy [1,5,6].

Serious adverse drug reactions (ADRs) and dose-limiting toxicities (DLTs) are common with cytotoxic agents utilized in oncology treatment regimens. Toxicities such as severe myelosuppression, diarrhea, and nephrotoxicity can lead to treatment delays and/or dose reductions; subsequently decreasing the efficacy of treatment in a critical patient population. The genetic variability in drug-metabolizing enzymes (DMEs) between individuals is the major cause of drug toxicity [7]. Consequently, the majority of studies and standards of practice in Oncology focuses on DMEs. These enzymes can be broken into four categories of metabolizers: extensive, poor, intermediate, and ultra-rapid. Due to the nature of many chemotherapeutic agents having narrow therapeutic indices, the placing of individuals into one of these four categories through testing of pharmacogenetic biomarkers can direct selection of treatment that decreases toxicities and maintains maximal efficacy.

A prime example of this importance is in the metabolism of 6-Mercaptopurine (6-MP) by Thiopurine Methyltransferase (TPMT) in childhood patients with Acute Lymphoblastic Leukemia (ALL). The therapeutic effect of 6-MP relies on its activation by TPMT to 6-thioguanine. TPMT deficiency leads to severe myelosuppression, hepatotoxicity and increases the risk of secondary cancer formation [7]. Although partial deficiency is rare with only around 10% of Caucasians and total deficiency around 0.3%, the importance of genetic testing for TPMT deficiency is reflected in the FDA recommendation of a dosage reduction in patients with heterozygous or homozygous mutations. An example of the dose reduction for 6-MP based on TPMT phenotype variation is found on Table 3.

Table 3: Clinical recommendation of 6MP based on TPMT phenotype.

Other biomarkers of importance in reducing toxicity include UGT1A1 deficiency in Irinotecan and nilotinib, DPD deficiency in 5-Fluoruracil or Capecitabine, CYP3A4 variability in activation of Cyclophosphamide, and CYP2D6 variability in Tamoxifen therapies [7]. Dosing of 5-FU DPD phenotype can be found in Table 4 below.

Table 4: Clinical recommendation of 5-FU dosing based on DPD phenotype.

Applying dose-finding studies directed by genotyping provide clinicians a tool for patients predicted to have DLTs an opportunity to avoid delays in therapy by tailoring individual doses. In a study by Satoh et al, patients with gastrointestinal cancer and homozygous UGT1A1×28 or UGT1A1×6 alleles –polymorphisms associated with severe myelosuppression from irinotecan administration-were given gradually increasing doses of irinotecan from 75 to 150 mg/m². Only 25% of patients completed two cycles of 150 mg/m², a dose used in FOLFIRI regimens, without treatment delays and dose reductions due to DLTs. In contrast, no DLTs occurred at doses of 100 to 125 mg/m², allowing the patients to receive subsequent treatment safely [17-20]. Incorporation of routine testing of DME variants in Oncology can lead to predictions in patient’s response to therapy and direct treatment regimens that increase patient tolerability Table 5 [1,5-7].

Table 5: Examples of Pharmacogenomic Biomarkers for Prevention of Toxicities in Cancer Therapy [1,5,6,7].

Multi-drug resistance (MDR) in Oncology can be assessed through genetic testing of biomarkers and guide the clinician in patient-specific therapies. Key contributors to MDR in cancerous cells are the drug transporters of the Adenosine Triphosphate (ATP)-Binding Cassette (ABC), most notably ABCB1 also referred to as P-glycoprotein (P-gp) and ABCG2. Commonly seen in MDR tumor cells are over-expression of P-gp or ABCG2 causing efflux of many structurally ubiquitous cytotoxic agents such as the taxane and anthracycline classes of chemotherapeutics [7-9]. Other contributors to drug resistance relate to DNA repair enzymes and enzymes responsible for detoxification. The endonuclease excision repair cross-complementing group 1 (ERCC1) is a biomarker related to oxaliplatin resistance through its important role in the mechanism of nucleotide excision repair (NER). Variants of this enzyme such as ERCC-118 T/T lead to increased production of ERCC1 that counteracts oxaliplatin’s ability to induce apoptosis through crosslinking with DNA [10]. Glutathione-S-Transferase (GST) enzymes play a role in detoxification of cytotoxic agents such as cyclophosphamide and platinum-based agents. One example is with patients receiving the FOLFOX chemotherapy regimen, where the GSTP1-105×G allele lead to an increase in development of neurotoxicity, a complication of the oxaliplatin component [10-11]. Pharmacogenomics, specifically biomarker screening, is a powerful tool for the clinician to understand the basis behind drug resistance in Oncology and create patient-specific drug regimens that maximize efficacy and decrease toxicity.

The application of Pharmacogenomics in Oncology to predict patient responsiveness is well established in both the ASCO and NCCN guidelines to serve clinicians in evidence-based treatment options. The goal of pharmacogenomics is to provide patient-specific therapy options that both increase efficacy and decrease toxicity. To meet this goal, numerous biomarkers have been discovered, leading the way to targeted drug therapies. The success of selective biomarkers in predicting response is currently not witnessed in the application of biomarkers that limit drug toxicity and resistance. An issue with wide acceptance of biomarkers in national guidelines and clinical practice is the scarcity of randomized controlled trials for every biomarker discovered [1,7]. Small sample sizes and the multitude of biomarkers make it difficult to conduct studies in pharmacogenomics at a scale the represents the population. Another area of caution is the wide acceptance of utilizing genetic tests without the proper recommendations and randomized controlled trials. This is echoed in the study by Peppercorn et al. where oncologists were evaluated in regards to genetic testing of CYP2D6 variants and tamoxifen therapy while the current understanding of the implications in pharmacotherapy were still evolving. What the authors found were 31% of oncologists used the commercially available test for CYP2D6 variants and 56% stated they would order the test outside of a clinical trial if requested by a patient [13].

One of the biggest hurdles for the widespread use of pharmacogenetic testing is the economical impact of routine commercial testing on the healthcare system. A recent review by Frank et al. [20] on the costeffectiveness of pharmcogenomic profiling in metastatic colorectal cancer concluded that due to the increasing complexity of treatment choices, analyses of cost effectiveness and reimbursement decisions, a consensus guideline for health economic evaluations is desperately needed to create a standard approach missing from current economic studies. Nevertheless, genetic testing of biomarkers in predicting patient response is crucial to developing patient specific treatment plans in a patient population where time is of the essence. The success of applying pharmacogenomics in predicting response to treatment and its wide acceptance in national guidelines reinforces the importance of further studies in larger populations on drug toxicity and resistance. There are still many economic, ethical, legal and clinical issues needing to be addressed before pharmacogenomics is fully integrated in the care of cancer patients [1,5,7].