Personalized oncology: genomic screening in ... - Wiley Online Library

APMIS 122: 723–733

© 2014 APMIS. Published by John Wiley & Sons Ltd. DOI 10.1111/apm.12293

Review Article

Personalized oncology: genomic screening in phase 1 IDA VILLER TUXEN,1 LARS JØNSON,2 ERIC SANTONI-RUGIU,3 JANE PREUSS HASSELBY,3 FINN CILIUS NIELSEN2 and ULRIK LASSEN1 Department of Oncology, Rigshospitalet, University of Copenhagen, Copenhagen; 2Center of Genomic Medicine, Rigshospitalet, University of Copenhagen, Copenhagen; and 3Department of Pathology, Rigshospitalet, University of Copenhagen, Copenhagen, Denmark

1

Tuxen IV, Jønson L, Santoni-Rugiu E, Hasselby JP, Nielsen FC, Lassen U. Personalized oncology: genomic screening in phase 1. APMIS 2014; 122: 723–733. Improvements in cancer genomics and tumor biology have reinforced the evidence of cancer development driven by numerous genomic alterations. Advanced genomics technology can be used to characterize genomic alterations that potentially drive tumor growth. With the possibility of screening thousands of genes simultaneously, personalized molecular medicine has become an option. New treatments are being investigated in phase 1 trials around the world. Traditionally, the goal of phase 1 studies was to determine the optimal dose and assess dose-limiting toxicity of a potential new experimental drug. Only a limited number of patients will benefit from the treatment. However, introducing genomic mapping to select patients for early clinical trials with targeted molecular therapy according to the genomic findings, may lead to a better outcome for the patient, an enrichment of phase 1 trials, and thereby accelerated drug development. The overall advantage is to determine which mutation profiles correlate with sensitivity or lack of resistance to specific targeted therapies. The utility and current limitations of genomic screening to guide selection to Phase 1 clinical trial will be discussed. Key words: Personalized medicine; targeted treatment; next generation sequencing; cancer; phase 1. Ida Viller Tuxen, Department of Oncology, Rigshospitalet, University of Copenhagen, Blegdamsvej 9, DK-2100 Copenhagen, Denmark. e-mail: [email protected]

The traditional approach to cancer treatment has been ‘one size fits all’, but advances in tumor biology and cancer genomics have begun to challenge the paradigm (1). In the recent years, a deeper understanding of oncogenes important for cancer cell growth has enabled the development of new targeted approaches to cancer treatment (2–4). The clinical utility of drugs designed to specifically inhibit the oncogenic kinases and signaling pathways is widely recognized and may prove to be effective in advanced or refractory disease. The treatments have often proven to be well tolerable by the patients with much less toxicity compared with conventional cytotoxic drugs. New targeted cancer treatments are being developed following the detection of specific molecular changes in tumor cells, and although recent results have shown that it is possible to inhibit cell growth in vitro, only limited knowledge about the use of targeted drugs in patients has been described. Increasing availability and decreasing cost of next Received 16 April 2013. Accepted 3 June 2014

generation sequencing technology allow profiling of each cancer patient for specific actionable genomic aberrations. Translating high-throughput sequencing for biomarker-driven trials for personalized treatment presents a unique opportunity in the future perspectives of oncology. Personalized oncology based on next generation sequencing techniques is introduced in the phase 1 setting in several institutions, including ours. Traditionally the goal of phase 1 studies was to determine the optimal dose and assess dose-limiting toxicity of a potential drug. In ‘first in human’ trials, only a small number of patients will benefit from the treatment. Introducing genomic selection in phase 1 clinical trials by molecular biomarker-driven analysis, we propose a better outcome for the patient, an enrichment of phase 1 trials, and thereby accelerated drug development. Recent studies have shown that patients entering in Phase 1 trials by genomic analyses have higher chances of efficacy of treatments than patients not being selected on the basis of genetic expression (5). 723

TUXEN et al.

Although the introduction of personalized oncology is showing a bright future, there are still challenges to overcome as it will be discussed.

CLINICAL SUCCESS STORIES OF TARGETED THERAPY BASED ON GENOMIC ALTERATIONS Although we are more than a decade into the postgenomic era, only a limited number of therapeutic agents targeting specific genetic alterations have been implemented in the clinical practice. The treatment of breast cancer (BC), the most common malignancy in women, was the first area in clinical oncology to introduce targeted therapy against specific genomic alterations, thanks to the discovery of ERBB2 (HER2) amplification (6). Indeed, 20–30% of BC patients harbor this gene amplification and represent a subgroup of patients with a more severe and aggressive disease. However, by targeting the overexpressed receptor tyrosine-protein kinase erbB-2 with monoclonal antibodies, such as trastuzumab (7) and lapatinib (8), great therapeutic improvement has been achieved in these patients. It is also necessary to examine tumor tissue samples from patients with colorectal cancer (CRC) for KRAS mutational status, as only patients with metastatic CRC harboring a wild-type KRAS gene can be offered treatment with monoclonal antibodies against epidermal growth factor receptor (EGFR), such as cetuximab or panitumumab (9). Another example of the importance of dividing oncological patients into subtypes according to genomic alterations is the case of non-small cell lung cancer (NSCLC). Despite the poor outcomes of advanced stage disease, prolonged survival can be seen in some subgroups of NSCLC patients with specific mutations in oncogenic driver genes. In particular, activating mutations of the EGFR gene were the first genetic predictive biomarker discovered to be present in approximately 10% of Caucasian NSCLC patients, especially non-smokers with the adenocarcinoma subtype (10). Today patients with advanced NSCLC harboring activating EGFR mutations are offered targeted treatment with EGFR tyrosine kinase inhibitors (TKIs) as first line therapy based on high-level evidence from randomized trials showing that administration of TKIs compared to standard platinum-based chemotherapy is associated with enhanced response rate, longer progression free survival (PFS), and less toxicity (11–13). In 2007, Soda et al. reported the finding of a chromosomal translocation in 3–7% of patients

724

with NSCLC resulting in the production of EML4– ALK fusion protein (14). The EML4–ALK gene fusion leads to ALK protein stabilization and constitutive activation of ALK kinase activity and the downstream signaling pathways leading to cell proliferation, invasion, and inhibition of apoptosis (15). Patients with EML4–ALK translocation define a new molecular subset of NSCLC and tend to be younger and non-smokers/light-smokers with adenocarcinoma (16). The identification of the EML4– ALK fusion led to the development of clinical trials targeting this molecular alteration with the ALK-inhibitor crizotinib. Phase I–II trials of this ALK-inhibitor in advanced NSCLC patients harboring the EML4–ALK translocation showed impressive response rates around 60% and PFS at 6 month of 72% (17). These results led to the approval of crizotinib as second-line treatment of NSCLC patients carrying ALK rearrangement by the US Food and Drug Administration in August 2011 and by the European Medicine Agency in Autumn 2012. Finally, crizotinib has recently been shown to be superior to standard chemotherapy in patients with previously treated NSCLC with ALK rearrangement (18). Another important development in the genomic era of cancer therapy in recent years is the detection of activating mutations of the Serine/threonine-protein kinase B-raf (BRAF) in melanoma. Metastatic melanoma presents with aggressive behavior and has been refractory to current therapies with a poor prognosis. More than 50% of melanoma patients harbor mutations within the proto-oncogene BRAF, where the activating BRAF c.1799T>A (V600E) is seen in more than 80% of all BRAF mutations (19). BRAF V600E mutation contributes to constitutive activation of the MAPK signaling pathway leading to cell proliferation and escape from apoptosis (20). In 2010, Flaherty et al. reported remarkable results from a phase I study with a BRAF inhibitor, vemurafinib, with overall response of 81% in the extended group consisting of patients with melanoma harboring the BRAF V600E mutation (21). These results revolutionized the treatment of metastatic melanoma. Despite the initial sensitivity of most patients treated by single-targeted agents, the long-term response of such therapies is invariably restricted by the development of resistance, usually due to secondary mutations or amplification of the kinase target gene or activation of alternative signaling pathways (22). Thus, because of its dynamic and multifactorial nature, acquired resistance to targeted therapies increases the requirement for simultaneous investigation of multiple targets and signaling pathways that might be inhibited by new

© 2014 APMIS. Published by John Wiley & Sons Ltd

GENOMIC SCREENING

more effective targeted drugs or drug combinations (23–25).

PERSONALIZED ONCOLOGY USED TO SELECT PATIENTS FOR EARLY CLINICAL TRIALS Introducing genomic mapping to select patients for early clinical trials with targeted therapy according to the molecular findings is a tendency seen in several cancer institutions. The goal of these trials is to determine which mutation profiles correlate with sensitivity/lack of resistance to specific targeted therapies and whether outcomes are consistent among different histologies (1). Alternative strategies and methods are attempted to implement personalized oncology. One of the first centers to establish a personalized medicine program for patients in phase I setting was MD Anderson (5). From 2007 to 2011, they conducted a non-randomized observatory trial allocating 1144 patients to early trials selected by somatic mutations in DNA obtained from paraffin-embedded tumor samples. A minor panel of actionable and ‘drugable’ genes were tested and aberrations were found in 460 (40%) patients, of whom less than half were treated with a matched therapy. They reported an overall response rate in the 175 patient treated with matched therapy of 27%. Compared to traditionally response rate in phase 1 of 5–10%, this approach showed promising results. The first group conducting a prospective study with real-time molecular profiling was von Hoff et al. (26) in 2010, who undertook a multicenter study of 86 patients with different types of refractory metastatic cancer. Microarray gene expression assays of 51 genes were performed on frozen tissue from metastatic sites. A total of 66 patients (77%) were treated with matched therapy and an increased PFS was seen in 18 patients (27%) when measured as PFS ratio >1.3. The PFS ratio (also known as time-to-progression ratio or growth modulation index) is defined as the ratio between PFS of matched treatment/PFS of most recent treatment and if the PFS ratio is above 1.3, the matched therapy selected is defined as having benefit for the patient (27). This study revealed the feasibility of a personalized approach to clinical treatment. Around the world, several clinical trials using genomic alteration to select patients to phase I trials are currently ongoing. Preliminary data from the French MOSCATO 01 (Molecular Screening for Cancer Treatment Optimization) were presented at the annual meeting of the American Society of Clinical Oncology in 2013 (28, 29). Based on


on-purpose biopsy from metastatic or primary tumors in patients referred to phase 1 trial, DNA was extracted and analyzed by microarray (Agilent Comparative Genomic Hybridization) and by sequencing of 30 target genes. Data from 129 patients revealed an actionable target in 52 patients (40%) with a PFS ratio >1.3 in 47% of the treated patients. Furthermore, the WIN consortium (Worldwide Innovative Networking) in personalized cancer medicine have established a global clinical trial conducted in six academic cancer centers called WINTHER, which is a multicenter multinational prospective study introducing an extended approach including findings of alterations in both DNA and RNA from tumor tissue (30, 31). The Michigan Oncology Sequencing Project (MIONCOSEQ) is currently enrolling patients with advanced cancers across all histologies subjected to integrative sequencing, including whole-exome sequencing and transcriptome sequencing (32). Since April 2011, they have enrolled over 200 patients and recently they published elucidating results from a subgroup of patients with hormoneresistant metastatic BC suggesting that activating mutations in the estrogen receptor could be a key mechanism in acquired resistance to endocrine therapy (33).

COPENHAGEN PROSPECTIVE PERSONALIZED ONCOLOGY In May 2013, we introduced a personalized genomic screening program in the Phase 1 unit at Righospitalet called ‘Copenhagen prospective personalized oncology (CoPPO)’. The Phase unit at Rigshospitalet is a dedicated experimental clinical oncology department receiving patients with no other treatment options from all parts of Denmark (200 patients per year), are being offered enrollment in the program. The program will be including an estimated number of 500 patients over a period of 5 years. Description of the clinical setup

Patients with metastatic or refractory solid cancer with good performance status (ECOG score 0–1), for whom standard therapies are insufficient to sustain disease control and who are considering Phase 1 clinical trial are eligible for participation after signing the informed written consent. The tumor/ metastatic site has to be accessible for biopsy and evaluable or measurable according to Response Evaluation Criteria in Solid tumors (RECIST) (34). Fresh tumor biopsies (n = 3) are obtained under

725

TUXEN et al.

local anesthesia. Two are stored in RNAlaterâ (Life Technologies, Carlsbad, CA, USA) for RNA expression analyses and DNA gene mutation analyses, while one biopsy is formalin-fixed and paraffinembedded (FFPE) for histopathological analyses. The latter confirm the suitability and representativeness of the material, including the presence of minimum 100 tumor cells and the evaluation of the amount of necrosis, if any, in the tissue. The results from the genomic analysis are available after 4 weeks. A blood sample (7 mL) is taken, from which germline mutations can be subtracted in the tumor/normal analysis. Genomic analysis

We have established a sequencing- and array-based pipeline to provide a comprehensive landscape of actionable targets and informative findings (Fig. 1). Tumor and normal DNA is examined by exome sequencing and tumor mutations are organized according to the cellular function of the affected genes, for example protein kinases, in relation to the suspected drug candidates (Table 1). Tumor and germline DNA are subjected to whole-exome sequencing (WES) using SureSelect v5 sequence capture (Agilent Technologies, Santa Clara, CA, USA) and Illumina HiSeq2500. In addition to WES

Copenhagen Pr

of the DNA isolated from the tumor containing biopsy, targeted sequencing of 48 important cancerrelated genes (Fig. 2) is performed with The TruSeq Amplicon – Cancer Panel (Illumina MiSeq). Expression array is performed to molecularly classify the origin of the primary tumor and to evaluate the expression level of therapeutic targets. In addition RNA sequencing (Nugens Ovation RNA-seq system v2) is performed to verify the data from the expression array, verify the expression of a transcript harboring an activating mutation, and to investigate whether chromosomal translocations are the reason for tumor-specific expression of an oncogene. Data are analyzed using CLC Genomics Workbench (Qiagen, Venlo, Limborg, the Netherlands), where data from WES are mapped against hg19, aiming at an average coverage between 80 and 1009. Identification of somatic mutations is performed using a tumor-normal analysis in which the inherited variants are subtracted from the tumor variants. The identified somatic mutations are exported into Ingenuity Variant Analysis (Qiagen) to identify causal variants. Data from The TruSeq Amplicon – Cancer Panel are analyzed using the CLC Cancer Research Workbench (Qiagen) to directly call cancer driver mutations. Gene expression levels and copy number variants are called from expression array and SNP

ve Personalized Oncology (CoPPO)

Informed Consent

Blood sample

Tumor Biopsy

Pathology

RNAseq

Germline DNA

Tumor DNA

Tumor RNA

Expression array

SNP array

48 gene panel

ctDNA

Whole Exon sequencing

Personalized treatment sugges on

Fig. 1. Copenhagen Prospective Personalized Oncology (CoPPO) flowchart. After signing informed consent, on-purpose biopsy from metastatic site and a blood sample are taken. Next step is pathological validation, DNA and RNA purification. Extended genomic analyses are performed including RNA sequencing, Expression array, SNP array, targeted sequencing of a 48 gene-panel and whole-exome sequencing. The latter is performed on both tumor and germline DNA. Germline mutations are subtracted to detect tumor-specific alterations. CtDNA will be introduced in the program. Results from all analyses are reviewed by the multidisciplinary tumor board defining an actionable target and suggest a treatment based on the genomic findings.

726


GENOMIC SCREENING

Table 1. Glossary regarding utilized genomic techniques Next generation sequencing (NGS) High-throughput parallel sequencing procedure unlike Sanger sequencing Whole genome sequencing (WGS) Full genome sequencing determines the complete DNA sequence of the genome Whole-exome sequencing (WES) Determines the sequence of all protein coding regions (exons) Targeted gene panel sequencing Sequencing a limited number of genes RNA sequencing (RNA-seq) Whole transcriptome sequencing Gene expression arrays Determines the expression levels of mRNA and is used for classification of primary tumor Single nucleotide polymorphism Array-based methods to detect variations in the genome. Determines DNA (SNP) array copy number variation (CNV)

Fig. 2. First line gene targets are sequenced to a coverage above 500–10009 (labeled in RED (n = 48)). Second line genes labeled in BLACK (n = 117) and whole exome (n = 21 500) are sequenced to an average coverage of 100x.

array by using Partek software (Partek Incorporated, St. Louis, MO, USA). Data will be paired with clinical and gene expression profile to identify potential causative association between mutations and clinical factors as well as gene expression, gene activation, and prognosis. Evaluating the results and possible therapy

Results are reviewed by a multidisciplinary team (tumor board) that verifies the clinical relevance of the findings and defines an ‘actionable’ target. The multidisciplinary tumor board consists of members with expertise in clinical oncology, molecular biology, genome biology, bioinformatics, clinical genetics, and clinical pathology. The treatment suggested by the tumor board can be either a drug under development (Table 2) or a marketed drug. If no actionable target can be revealed from the screening, the patient will be allocated to a non-matched, © 2014 APMIS. Published by John Wiley & Sons Ltd

random available slot in a phase 1 trial (Fig. 3). Specific targets, for which Phase 1 trials exist, but are based on the assessment and/or verification of the target as such or of other members of the target’s pathway at the protein level, can be further investigated in the parallel FFPE biopsy by using immunohistochemical techniques. If a germline mutation is suspected, the patient will be referred to genetic counseling. To assess the antitumor effects of treatments, tumor response will be evaluated using RECIST criteria (34). PFS from the treatment will be compared to PFS of the most recent standard treatment (PFS ratio).

CHALLENGES TO ADDRESS Despite the bright future of introducing personalized medicine in clinical oncology, there still remain challenges to overcome.

727

TUXEN et al.

Table 2. Currently available phase 1 studies for solid tumors in phase 1 unit. International cooperation allows access to additional 50 ongoing clinical trials Drug Target Tumor type Description RO6895882 Carcinoembryonic Advanced and/or Intravenously administrated variant of antigen (CEA) metastatic solid tumors interleukin-2 targeting CEA Advanced and/or Inhibits FGF-stimulated growth Dovitinib Fibroblast growth metastatic solid tumors (TKI258) factor (FGF) signaling pathway Advanced and/or PARP inhibitor, inhibits single-strand break Olaparib Poly ADP ribose metastatic solid tumors and base-excision repair DNA repair pathways, polymerase and induces ‘synthetic lethality’ in cells with (PARP) other defective DNA repair pathway such as homologous recombination GSK3052230 FGF signaling NSCLC, stage IV/ FGFR1 receptor antagonist pathway recurrent metastatic disease HuMaxâ; Tissue factor Advanced and/or Tissue factor specific antibody drug conjugate metastatic solid tumors Genmab, Copenhagen, Denmark TF ADC HER-3 antibody RO5479599 HER-3 HER-3 positive advanced and/or metastatic solid tumors KPT-330 Nuclear export Advanced and/or Selective inhibitor of nuclear export metastatic solid tumors LiPlaCis – Advanced and/or Liposomal cisplatin formulation metastatic solid tumors LY3039478 Notch pathway Advanced and/or Notch inhibitor metastatic solid tumors Dabrefenib/ BRAF Advanced and/or BRAF inhibitor/MEK inhibitor tremetinib metastatic solid tumors Foxy5 Wnt-5a pathway Advanced and/or Wnt-5a mimicking hexapeptide metastatic solid tumors

Pa

nt example:

Pa ent:

Biopsy :

Pathology:

Genomics:

Tumor board:

Treatment:

Cancer Mammae Female 6 prior lines

Liver biopsy No compli ons

Cancer Mammae Tumor > 60 %

Expression clasifica on: Cancer Mammae c DNA mu ons: TP53, PTEN So Gene expression: ROS1↑

Ac onable target: ROS1 Treatment sugges on: Inhibi on of ROS1

Matched treatment Off-label Crizo nib

Fig. 3. Illustration of a fictive patient enrolled in the CoPPO phase 1 genomic program.

Ethical aspects

Obtaining a wide genomic profile of tumor tissue requires a parallel investigation of germline DNA to detect tumor-specific mutations. In cases where the patient is young and has signed the letter of consent allowing the bioinformatician to look for diseasecausing variants in the germline material, there exists a minor risk that mutations that are not diseaserelated could be identified. Handling of incidental 728

knowledge has to be taken into account when introducing genomic screening in personalized medicine. Participants have to be aware of this possibility and have to be well informed of consequences. This is handled within the informed consent interview prior to enrollment in the study. The patients are informed about the possible results and limitations of targeted sequencing and exome sequencing. The implications and procedures involved in the analysis and handling © 2014 APMIS. Published by John Wiley & Sons Ltd

GENOMIC SCREENING

Table 3. Possible outcomes of germline exome sequencing. When the patient signs the informed consent, a decision of the level of information must be made and noted 1 Finding of gene alterations considered being the cause of the hereditary disease. In this case, the patient will be offered genetic counseling 2 Finding of gene alterations that cannot be determined accurately. Thus, it remains unclear whether these are associated with the disease. In some cases, it may be necessary to examine several members of the family. The patients can decide the level of information 3 No specific gene alterations associated to disease are found. However, in the future, some of the genes may be associated with disease. In case, the patient or the family will be informed if additional information is encountered 4 When studying all genes with exome sequencing, there is a risk that gene alterations not associated with heritable disease are found, so called incidental findings. This could be genes associated with an increased risk of a disease of the nervous system. Such a finding may have implications for the patient or other family members

of patient data are discussed prior to consent. There are four possible outcomes of exome sequencing, as illustrated in Table 3. When the patient signs the informed consent, a choice of the level of information must be made and noted. The MI-ONCOSEQ program concentrated on developing an adequate informed consent process that addresses incidental findings and has demonstrated the need of genetic counselors in multidisciplinary tumor boards reviewing the results of genomic tumor profiling (35). However, one has to take into account that patients considering phase 1 trials represent a unique population of patients, who are strongly motivated by hope of therapeutic benefit (36) and thereby a vulnerable group. Therefore, the weight of risk and benefit of enrollment may be different from other patient populations. Patients referred to phase 1 unit trials have no further available standard treatment options. This population of patients has disseminated disease with expected limited survival and risk of rapid disease progression. The waiting time due to the time-consuming analysis prior to treatment can be a devastating experience for the patient and may result in the unfortunate situation where the patient no longer fulfills inclusion criteria at the time of receiving the results of the analysis. Extended genomic analysis is time-consuming because of the heavy amount of data gained. One way of dealing with this challenge could be either to prepare for the enrollment by implementing the analysis during the last line of treatment or by introducing a preliminary data review based on targeted DNA sequencing that can be performed in a more rapid time schedule than the other analyses. The latter would allow the clinical oncologist to prepare the enrollment in the early trial before the tumor board reviews the extended data. Tissue selection and preservation

A questionable issue is whether to use archival tissue from the primary tumor or fresh tissue from metastatic site when detecting genomic alterations © 2014 APMIS. Published by John Wiley & Sons Ltd

that can guide therapy. Most cancer patients have archival tissue readily available for profiling, which in principle is time-saving, whereas on-purpose biopsies are time-consuming and represent a risk for complications. However, it is acknowledged that new mutations can arise during the metastatic process and during development of resistance to antineoplastic treatment, consistent with the instability of cancer (1, 37). Molecular discordance between primary and metastatic site differs among cancer types. Data available from both KRAS and EGFR mutations in CRC and NSCLC have revealed high concordance between primary tumor and metastatic site (38, 39). Recently, Tran et al. demonstrated a 88% concordance between a panel of DNA mutations in the original archival tumor tissue and a new biopsy obtained a median of 33 months apart (40). In contrast, discordance between primary and metastatic mutations in the PIK3CA gene was reported in 32% of 102 cases of metastatic BC (41). A genomic profile of primary tumor from archive tissue probably does not exactly reflect the actual genomic profile of the metastatic disease. Attainment of on-purpose biopsies is preferred to get an overview of the metastasized tumor that has evolved since the primary tumor and gained additional cancer driver mutations. On-purpose biopsies in early-phase trials have been reported to be usually safe with serious complications in less than 1.5% of the patients (42) and they seem to be well accepted by the patients and clinicians. Whether to use fresh-frozen tissue, RNAlaterstored tissue, or FFPE material is another issue of discussion. Most hospitals collect and archive tissue as FFPE material suitable for histological assessment. Isolating adequate amount of DNA/ RNA from FFPE tissue is challenging as formalin fixation causes crosslinking and degradation into smaller fragments, although acceptable results in detecting DNA mutations have been shown by several research groups (1). Regarding on-purpose biopsies for prospective trials in phase 1, most institutions choose fresh-frozen tissue or tissue

729

TUXEN et al.

stored in RNAlater, especially when RNA expression profiles are being obtained (26, 32). Driver vs passenger

Explaining the complex nature of cancer still remains to be revealed, but numerous models have been proposed (37, 43). One accepted model is the assumption that cancer is a genetic disease attained as a result of somatic changes of the DNA in cancer cells. Somatic abnormalities are abundant in all cancer cells, although not all the detected somatic mutations are involved in the development of the cancer. To address this concept, the terms ‘passenger’ and ‘driver’ mutations have been hosted. Driver mutations are defined as mutations that are causally implicated in oncogenesis and confer growth advantage for the cells harboring them, while being positively selected during cancer development (3). Most mutations in cancers, however, do not seem to have a role in tumor development and are described as ‘passengers’. These mutations are gained as the result of the high mutation rate of tumor cells due to DNA instability or have happened to be present in the cancer cell acquiring a driver mutation (3). Studies including high-throughput RNA-seq have revealed that only a minor number of mutations detected are expressed significantly supporting the importance of performing both DNA mutation analysis as well as RNA expression (44). Defining the driver mutations represents a challenging subject in the era of personalized oncology. It requires a multidisciplinary approach gathering experts from multiple areas such as clinical oncology, molecular biology, genomics, clinical genetics, and pathology along with computational approaches (45). Intratumor heterogeneity

The immense complexity of the genome in solid cancers and especially the reported inter- and intratumor heterogeneity represents a striking concern in the implementation of personalized treatment based on the genomic analysis of a single biopsy. Emerging evidence of spatially separated differences in somatic mutations and DNA copy number variations has been revealed, confirming that a neoplasm consist of multiple clonal subpopulations of tumor cells (37, 46). Intratumor heterogeneity of solid tumors has been described as a growing tree, where the trunk of the tree harbors the founding ubiquitary driver mutations present in every tumor subclone and region. The sprouting ‘branches’ represent different spatially separated subclones of the tumor, each harboring different alterations that distinguish their biological behavior and furthermore have the potential to become driver mutations 730

under distinct selection pressure (46, 47). Intratumor heterogeneity can contribute to treatment failure and development of resistance to therapy and represents a challenge in the personalized treatment based on a single biopsy. Multiple biopsies from different sites and regions are too invasive, often risky and not feasible for the patient and might not even bring the full picture of the cancer. Nevertheless, one could argue that a biopsy taken from a metastatic site in progression would represent an area of activity and thereby harbor the relevant driver alteration at a specific time point. FUTURE PERSPECTIVES AND CONCLUSION The dynamic behavior of solid tumors requires a dynamic approach of the clinician. Development of resistance to targeted therapies is abundant and can accrue due to secondary mutations or positive selection of rare cell populations already present in the primary tumor (48) as a result of treatment. For instance, in EGFR-mutated NSCLC developing resistance to EGFR TKIs, secondary mutations have been revealed in 50% of the cases (49). Tumor tissue biopsies thereby represent a snapshot in time. Re-biopsies at progression are mandatory to convert the treatment according to new alterations. However, this progressive treatment strategy can be devastating for the patient as solid cancers cannot be repeatedly sampled without invasive procedures. An alternative strategy is obtainment of ‘liquid biopsies’. Blood samples from peripheral blood contain circulating tumor cells (CTC) and cell-free circulating DNA (ctDNA), which can be obtained repeatedly during treatment (50, 51). Especially the analysis of ctDNA represents a promising approach to detect tumor-specific genomic alteration (51). Recently, Dawson et al. identified somatic alterations in ctDNA by targeted and whole-genome sequencing of repeatedly collected blood samples during treatment of metastatic BC in patients with known somatic mutations. They were able to detect alterations in ctDNA in 29 of 30 patients (97%). They anticipated that ctDNA levels showed superior correlation with tumor burden than CTC (52), noting that CTC were detected using EpCAMbased CellSearch©. EpCAM-based CellSearch is produced by Janssen Diagnostics, Raritan, NJ, USA, and is one of now many methods for detecting CTC. This method is based on capturing CTC in blood using immunomagnetic ferrofluids covered by antibodies against the cell adhesion protein EpCAM present on many epithelial cell types. It is currently the only FDA-approved technique for the detection and prognostic impact of CTC in patients with breast, colon, and prostate cancer (53). © 2014 APMIS. Published by John Wiley & Sons Ltd

GENOMIC SCREENING

Furthermore, measurement of ctDNA has revealed KRAS-mutated ctDNA in serum from KRAS wildtype CRC patient developed during monotherapy with panitumumab, a therapeutic anti-EGFR antibody (54). Obtainment of CTC and/or especially ctDNA might allow real-time monitoring of individualized treatment in terms of identification of genomic targets, response to therapy, and possible appearance of resistant clones. Implementation of ‘liquid biopsies’ during treatment seems favorable in terms of monitoring treatment in personalized cancer programs. In addition, challenges regarding tumor heterogeneity may be improved by the supplement of liquid biopsies as such biopsies represent a pool of CTC and ctDNA from different metastatic sites. Unexpected findings as, for instance, the recent reported sensitivity to everolimus in bladder, cancer patients with loss-of-function mutations in TSC1 (55) can possibly be revealed by personalized cancer programs by analyzing genomic profile of unusual responders. Registration of unusual responders and corresponding genomic profile should be mandatory in all personalized cancer programs. Future will presumably attain improved molecular understanding gained from this type of data. Genomic analysis and treatment based on targeting specific alterations are beginning to revolutionize cancer treatment development and requires extended molecular understanding by all the specialists involved to attain optimal benefit from these new techniques. Genomic-driven selection of patients to early clinical trials is now an option in a cost-effective way providing a new paradigm for clinical investigation. Individualized treatment according to genomic profile will benefit the patient, enrich early clinical trials, and hopefully accelerate drug development. Nevertheless, the immense complexity of solid tumors including tumor heterogeneity mandates the use of a wide range of ‘omic’ platforms and specialists to reveal the full picture sufficient to select personalized treatment. Furthermore worldwide networking and sharing of data are required to gain optimal knowledge from genomic studies to optimize cancer treatment. REFERENCES 1. Dancey JE, Bedard PL, Onetto N, Hudson TJ. The genetic basis for cancer treatment decisions. Cell 2012;148:409–20. 2. Garnett MJ, Edelman EJ, Heidorn SJ, Greenman CD, Dastur A, Lau KW, et al. Systematic identification of genomic markers of drug sensitivity in cancer cells. Nature 2012;483:570–5. 3. Stratton MR, Campbell PJ, Futreal PA. The cancer genome. Nature 2009;458:719–24.


4. Vogelstein B, Kinzler KW. Cancer genes and the pathways they control. Nat Med 2004;10:789–99. 5. Tsimberidou A-M, Iskander NG, Hong DS, Wheler JJ, Falchook GS, Fu S, et al. Personalized medicine in a phase I clinical trials program: the MD Anderson Cancer Center initiative. Clin Cancer Res 2012;18: 6373–83. 6. King CR, Kraus MH, Aaronson SA. Amplification of a novel v-erbB-related gene in a human mammary carcinoma. Science 1985;229:974–6. 7. Slamon DJ, Leyland-Jones B, Shak S, Fuchs H, Paton V, Bajamonde A, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 2001;344:783–92. 8. Thomson A. Proof of concept review Lapatinib : the evidence for its therapeutic value in metastatic breast cancer. Core Evid 2005;1:77–87. 9. Bokemeyer C, Van Cutsem E, Rougier P, Ciardiello F, Heeger S, Schlichting M, et al. Addition of cetuximab to chemotherapy as first-line treatment for KRAS wild-type metastatic colorectal cancer: pooled analysis of the CRYSTAL and OPUS randomised clinical trials. Eur J Cancer 2012;48:1466–75. 10. Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 2004;350:2129–39. 11. Paz-Ares L, Soulieres D, Melezınek I, Moecks J, Keil L, Mok T, et al. Clinical outcomes in non-small-cell lung cancer patients with EGFR mutations: pooled analysis. J Cell Mol Med 2010;14:51–69. 12. Mitsudomi T, Morita S, Yatabe Y, Negoro S, Okamoto I, Tsurutani J, et al. Gefitinib versus cisplatin plus docetaxel in patients with non-small-cell lung cancer harbouring mutations of the epidermal growth factor receptor (WJTOG3405): an open label, randomised phase 3 trial. Lancet Oncol 2010;11:121–8. 13. Maemondo M, Inoue A, Kobayashi K, Sugawara S, Oizumi S, Isobe H, et al. Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N Engl J Med 2010;362:2380–8. 14. Soda M, Choi YL, Enomoto M, Takada S, Yamashita Y, Ishikawa S, et al. Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer. Nature 2007;448:561–6. 15. Amin HM, Lai R. Pathobiology of ALK+ anaplastic large-cell lymphoma. Blood 2007;110:2259–67. 16. Shaw AT, Yeap BY, Mino-Kenudson M, Digumarthy SR, Costa DB, Heist RS, et al. Clinical features and outcome of patients with non-small-cell lung cancer who harbor EML4-ALK. J Clin Oncol 2009;27:4247–53. 17. Kwak EL, Bang Y-J, Camidge DR, Shaw AT, Solomon B, Maki RG, et al. Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer. N Engl J Med 2010;363:1693–703. 18. Shaw AT, Kim D-W, Nakagawa K, Seto T, Crin o L, Ahn M-J, et al. Crizotinib versus chemotherapy in advanced ALK-positive lung cancer. N Engl J Med 2013;368:2385–94. 19. Davies H, Bignell GR, Cox C, Stephens P, Edkins S, Clegg S, et al. Mutations of the BRAF gene in human cancer. Nature 2002;417:949–54.

731

TUXEN et al.

20. Gray-Schopfer V, Wellbrock C, Marais R. Melanoma biology and new targeted therapy. Nature 2007;445:851–7. 21. Flaherty KT, Puzanov I, Kim KB, Ribas A, McArthur GA, Sosman JA, et al. Inhibition of mutated, activated BRAF in metastatic melanoma. N Engl J Med 2010;363:809–19. 22. Gainor JF, Shaw AT. Emerging paradigms in the development of resistance to tyrosine kinase inhibitors in lung cancer. J Clin Oncol 2013;31:3987–96. 23. Friboulet L, Li N, Katayama R, Lee CC, Gainor JF, Crystal AS, et al. The ALK inhibitor ceritinib overcomes crizotinib resistance in non-small cell lung cancer. Cancer Discov 2014;4:662–73 24. Shaw AT, Kim D-W, Mehra R, Tan DSW, Felip E, Chow LQM, et al. Ceritinib in ALK-rearranged nonsmall-cell lung cancer. N Engl J Med 2014;370: 1189–97. 25. Menzies AM, Long GV. Dabrafenib and Trametinib, alone and in combination for BRAF-mutant metastatic melanoma. Clin Cancer Res 2014;20: 2035–43 26. Von Hoff DD, Stephenson JJ, Rosen P, Loesch DM, Borad MJ, Anthony S, et al. Pilot study using molecular profiling of patients’ tumors to find potential targets and select treatments for their refractory cancers. J Clin Oncol 2010;28:4877–83. 27. Mick R, Crowley JJ, Carroll RJ. Phase II clinical trial design for noncytotoxic anticancer agents for which time to disease progression is the primary endpoint. Control Clin Trials 2000;21:343–59. 28. Soria JC. Molecular Screening for Cancer Treatment Optimization (MOSCATO 01). http://clinicaltrials. gov/show/NCT01566019. 29. Hollebecque A, Massard C, De Baere T, Auger N, Lacroix L, Koubi-Pick V, et al. Molecular screening for cancer treatment optimization (MOSCATO 01): a prospective molecular triage trial—interim results. J Clin Oncol 2013; 31(Suppl; abstr 2512). 30. Soria JC, Tsimberidou A, Kurzrock R, Tabernero J, Rodon J, Berger R, et al. L04.04 * Winther: a study to select rational therapeutics based on the analysis of tumor and matched normal tissue biopsies in subjects with advanced malignancies. Ann Oncol 2013;24: i10–1. 31. Soria JC. A study to select rational therapeutics based on the analysis of matched tumor and normal biopsies in subjects with advanced malignancies (WINTHER). http://clinicaltrials.gov/ct2/show/NCT01856296. 32. Roychowdhury S, Iyer MK, Robinson DR, Lonigro RJ, Wu Y-M, Cao X, et al. Personalized oncology through integrative high-throughput sequencing: a pilot study. Sci Transl Med 2011;3:111ra121. 33. Robinson DR, Wu Y-M, Vats P, Su F, Lonigro RJ, Cao X, et al. Activating ESR1 mutations in hormoneresistant metastatic breast cancer. Nat Genet 2013;45:1446–51. 34. Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45:228–47. 35. Everett JN, Gustafson SL, Raymond VM. Traditional roles in a non-traditional setting: genetic counseling in precision oncology. J Genet Couns 2014 [Epub ahead of print].

732

36. Catt S, Langridge C, Fallowfield L, Talbot DC, Jenkins V. Reasons given by patients for participating, or not, in Phase 1 cancer trials. Eur J Cancer 2011;47:1490–7. 37. Vogelstein B, Papadopoulos N, Velculescu VE, Zhou S, Diaz LA, Kinzler KW. Cancer genome landscapes. Science 2013;339:1546–58. 38. Vignot S, Frampton GM, Soria J-C, Yelensky R, Commo F, Brambilla C, et al. Next-generation sequencing reveals high concordance of recurrent somatic alterations between primary tumor and metastases from patients with non-small-cell lung cancer. J Clin Oncol 2013;31:2167–72. 39. Knijn N, Mekenkamp LJM, Klomp M, Vink-B€ orger ME, Tol J, Teerenstra S, et al. KRAS mutation analysis: a comparison between primary tumours and matched liver metastases in 305 colorectal cancer patients. Br J Cancer 2011;104:1020–6. 40. Tran B, Brown AMK, Bedard PL, Winquist E, Goss GD, Hotte SJ, et al. Feasibility of real time next generation sequencing of cancer genes linked to drug response: results from a clinical trial. Int J Cancer 2013;132:1547–55. 41. Dupont Jensen J, Laenkholm A-V, Knoop A, Ewertz M, Bandaru R, Liu W, et al. PIK3CA mutations may be discordant between primary and corresponding metastatic disease in breast cancer. Clin Cancer Res 2011;17:667–77. 42. El-Osta H, Hong D, Wheler J, Fu S, Naing A, Falchook G, et al. Outcomes of research biopsies in phase I clinical trials: the MD anderson cancer center experience. Oncologist 2011;16:1292–8. 43. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646–74. 44. Shah SP, Roth A, Goya R, Oloumi A, Ha G, Zhao Y, et al. The clonal and mutational evolution spectrum of primary triple-negative breast cancers. Nature 2012;486:395–9. 45. Dienstmann R, Rodon J, Barretina J, Tabernero J. Genomic medicine frontier in human solid tumors: prospects and challenges. J Clin Oncol 2013;31: 1874–84. 46. Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med 2012;366: 883–92. 47. Yap TA, Gerlinger M, Futreal PA, Pusztai L, Swanton C. Intratumor heterogeneity: seeing the wood for the trees. Sci Transl Med 2012;4:127 ps10. 48. Garraway LA, J€ anne PA. Circumventing cancer drug resistance in the era of personalized medicine. Cancer Discov 2012;2:214–26. 49. Sequist LV, Waltman BA, Dias-Santagata D, Digumarthy S, Turke AB, Fidias P, et al. Genotypic and histological evolution of lung cancers acquiring resistance to EGFR inhibitors. Sci Transl Med 2011;3: 75ra26. 50. Pantel K, Alix-Panabieres C. Real-time liquid biopsy in cancer patients: fact or fiction? Cancer Res 2013;73:6384–8. 51. Diaz LA, Bardelli A. Liquid biopsies: genotyping circulating tumor DNA. J Clin Oncol 2014; 32:579–86.


GENOMIC SCREENING

52. Dawson S-J, Tsui DWY, Murtaza M, Biggs H, Rueda OM, Chin S-F, et al. Analysis of circulating tumor DNA to monitor metastatic breast cancer. N Engl J Med 2013;368:1199–209. 53. Miller MC, Doyle GV, Terstappen LWMM. Significance of circulating tumor cells detected by the cell search system in patients with metastatic breast colorectal and prostate cancer. J Oncol 2010;2010:617421.


54. Diaz LA, Williams RT, Wu J, Kinde I, Hecht JR, Berlin J, et al. The molecular evolution of acquired resistance to targeted EGFR blockade in colorectal cancers. Nature 2012;486:537–40. 55. Iyer G, Hanrahan AJ, Milowsky MI, Al-Ahmadie H, Scott SN, Janakiraman M, et al. Genome sequencing identifies a basis for everolimus sensitivity. Science 2012;338:221.

733