Keywords

13.1 Introduction

In order to provide optimal treatment options for each patient, genetic panel tests with genomic profiling, which can detect diverse genetic abnormalities found in tumors in a multiplex fashion and provide medical interpretation and meaningfulness, have been implemented in routine clinical practice. For standard treatment-resistant solid tumors, two panel tests were authorized in June 2019, in Japan. Although the primary outcome of these panel tests is to find a way for possible treatment options, in some cases secondary findings for germline mutation could be suggested or pointed out.

This article describes the system cancer genome panel test, the results obtained, and how they are interpreted, including secondary findings, and also mentions briefly about panel tests for germline mutations.

13.2 Cancer Genome Panel Tests

Two gene panel tests covered by Japanese national health insurance are FoundationOne Companion Diagnostics (hereinafter, “F1CDx”) and OncoGuide™ NCC Oncopanel System (hereinafter, “NCC Oncopanel Test”).

These cancer genome profiling tests were offered at specific institutions that have a system to conduct the test, such as “core center hospitals,” “center hospitals,” and “collaborative hospitals.”

The “Guidelines for the Development of Core Cancer Genome Center Hospitals” issued by the Health Bureau of the Ministry of Health, Labor and Welfare lists the following seven requirements for core center hospitals: (1) the hospital must have a specimen laboratory and pathology laboratory certified by a third party and must be able to properly conduct gene panel tests in accordance with the procedures, and must be able to medically interpret the results, and a panel of experts to meet at least once a month to interpret the results; (2) having a system to ensure that the treatment of secondary findings is clearly documented and that genetic counseling can be conducted appropriately; (3) having a system that can appropriately collect and manage genomic information and register it with the Center for Cancer Genomics and Advanced Therapeutics (C-CAT); (4) having a system that can appropriately store the biological materials; (5) having a department that oversees cancer genome medicine; (6) having a system that can provide information on cancer genome medicine to patients and their families; and (7) having a system as a core clinical research hospital or equivalent to it.

In order to implement genomic medicine, pathologists, specialists in molecular genetics, and genetic counselors are required to work for and have a proven track record in the field.

As of April 2020, 12 core center hospitals, 33 center hospitals, and 161 collaborating hospitals, altogether 206 facilities, have been designated.

Currently, only about 10–15% of patients who undergo cancer genome profiling are actually found to have the recommended treatment [1, 2].

13.3 Types of Panel Tests

With the development of genetic analysis equipment, we are now able to examine a large number of genes in a single test. The generic term for the genes to be searched is called panel. There are several types of panel tests, and the specimens used for each test are different depending on the purpose. Panel tests for detecting genetic abnormalities in cancer include those for tumor cells only, those for both tumor cells, and peripheral blood. The panel test to detect germline mutation is intended for peripheral blood only. In the current situation, there are many kinds of panel tests are available on the market for different purposes.

The F1CDx is provided by Foundation Medicine Inc. (Massachusetts, USA) and is the first FDA-approved tissue-based broad companion diagnostic for all solid tumors. The NCC Oncopanel Test is a gene panel testing designed specifically for Japanese solid tumor genome mutations, including childhood cancers, which was developed jointly by the National Cancer Center Japan and Sysmex Corporation (Kobe, Japan) [3, 4].

Next-generation sequencers are used in both gene panel tests: the F1CDx uses Illumina’s HiSeq4000 targeting in 324 genes and selects gene rearrangements (Table 13.1), as well as genomic signatures including microsatellite instability (MSI) and tumor mutational burden (TMB); on the other hand, the NCC Oncopanel uses Illumina’s NextSeq 550Dx to test 114 genes where Japanese people are prone to express cancer mutations (Table 13.2). Both target all exons and are examined using sequential synthetic sequencing on next-generation sequencers after library preparation using hybrid capture methods.

Table 13.1 A list of 324 target genes in the FoundationOne CDx [3]
Full size table
Table 13.2 A list of 114 target genes in the NCC Oncopanel Test [4]
Full size table

With the advent of next-generation sequencers, the number of nucleotide sequences and regions that can be analyzed at a time has been dramatically increased, enabling the analysis of multiple genes at a time and enabling cancer gene panel testing. Extraction of nucleic acids used for panel testing is obtained from formalin-fixed paraffin-embedded (FFPE) samples.

The percentage of tumor content on the section is important, and a tumor content of 20% or more is recommended for both testing. If the tumor content in the specimen is low, genetic analysis cannot be performed adequately, and the opportunity to obtain useful information from the expensive tests performed may be lost.

Specimens provided for panel testing include not only the tumor but also surrounding normal cells. The range of the targeted sequence region and the detection accuracy, that is, the read depth, is set in each test. Unlike whole genome sequencing, which reads a large amount of sequence information, the cancer gene panel test is designed to read only specific genes related to cancer with high accuracy. Therefore, the types of gene abnormalities, base substitutions, insertions, deletions, amplifications, gene fusions, etc. that can be analyzed differ depending on the type of test.

Based on the data detected by the sequencer, sequenced data are generated as FASTQ files. Based on this data, mapping is performed by referring to reference sequences of human genes in open public databases, and BAM files are generated. Then, based on the BAM files, each gene mutation location was mapped on the chromosomes, and VCF files are generated to show the results.

The F1CDx targets only tissue sample for genetic analysis, while the NCC Oncopanel Test also collects peripheral blood sample for the analysis to detect genetic changes in tumors. Thus, there are two types of panel tests used in cancer genome medicine, and the specimens handled by the panel tests are different from test to test. From normal cells obtained from blood samples, information on patient-specific genetic polymorphisms can be obtained, which can be used as a control for higher test accuracy. Both panel tests can determine or detect the possibility of germline genetic variants, and especially in cases with germline genetic variants, clinical genetic considerations are required.

13.4 Purpose of the Gene Panel Test

The purpose of an oncogene panel test is to detect cancer-derived somatic mutations and to realize personalized medicine that can be used to select more specific effective cancer drugs based on the genetic mutation information (Fig. 13.1). F1CDx has also been used as a companion diagnostic for some genes (lung cancer, EGFR mutation, ALK fusion, ROS1 fusion; malignant melanoma, BRAF mutation; breast cancer, ERBB2 amplification; colorectal cancer, KRAS, NRAS wild type; solid tumors, NTRK1/2/3 fusion; ovarian cancer, BRCA1/2 mutation).

Fig. 13.1
figure 1

Flow of genomic medicine

Full size image

The European Society for Medical Oncology (ESMO) Precision Medicine Working Group published recommendations for the use of next-generation sequencing (NGS) for patients with metastatic cancers. Based on the current evidence, ESMO recommends routine use of NGS on tumor samples in advanced non-squamous non-small-cell lung cancer (NSCLC), prostate cancers, ovarian cancers, and cholangiocarcinoma. In these tumors, large multigene panels could be used if they add acceptable extra cost compared with small panels. In colon cancers, NGS could be an alternative to PCR. In addition, based on the KN158 trial and considering that patients with endometrial and small-cell lung cancers should have broad access to anti-programmed cell death 1 (anti-PD1) antibodies, it is recommended to test tumor mutational burden (TMB) in cervical cancers, well- and moderately differentiated neuroendocrine tumors, salivary cancers, thyroid cancers, and vulvar cancers, as TMB-high predicted response to pembrolizumab in these cancers [5].

There is no drug approved in Japan for selective treatment of breast cancer for PIK3CA hotspot mutation, the relatively frequent detection of HER2 can be determined by immunohistochemical staining, and no other genetic mutation information is routinely useful for treatment selection for breast cancer. According to this guideline, it is not recommended that cancer gene panel testing be performed in routine clinical practice for breast cancer and limited to circumstances where patients can find opportunity to participate in clinical trials targeting some genetic mutations, such as AKT1E17K, PTEN, ERBB2, ESR1, and NF1, at certain hospitals, such as the aforementioned center hospitals.

13.5 Report Structure of the F1CDx

F1CDx consists of two sets of report for a result. First, CDx Associated Findings and Other Alterations & Biomarkers Identified report includes detected genetic variants and corresponding drug names related to the companion diagnosis and detected genetic variants which are not related to the companion diagnosis, as well as other biomarkers such as MSI status and TMB score. Japanese version of this report also includes “Approved Therapeutic Options in Japan.”

On the other hand, Professional Services report includes information of detected biomarkers and genetic mutations, as well as information of the corresponding therapeutic agents and ongoing clinical trials, including indications, recommendations based on NCCN guideline, and drug resistance with some available reference information. The report has not been approved by the FDA or the Ministry of Health, Labor and Welfare and is a supporting document.

13.6 Report Structure of the NCC Oncopanel Test

The NCC Oncopanel Test consists of three reports (summary report, sequencing report, and QC report). The summary report contains (1) gene abnormality information, (2) somatic mutation numbers and mutation occurrence rates, and (3) annotation information. First, the mutated genes and mutant allele frequencies, amplified genes, and fusion genes detected will be listed in the gene abnormality information section. Mutations suspected to be pathogenic variants among the germline mutation information will also be described here. Second, the number and frequency of single nucleotide variants (SNVs) and insertion/deletion mutations (InDel), as well as the total number and frequency of each mutation for the exon and non-exon regions, will be described for each somatic mutation. The total mutation rate is used to determine the tumor mutation burden (TMB). Third, annotation information obtained by referring to databases such as Expert Panel Data Base (EPDB), COSMIC, and ClinVar will be described.

The sequencing report contains detailed sequencing analysis information of tumor and non-tumor cells. The types of mutation detected and their locations will be described, and variant of unknown significance (VUS) will also be included here. Details of the detected germline gene mutations are also described in the column of germline mutation information.

The NCC Oncopanel Test includes 13 of the genes recommended by the American College of Medical Genetics and Genomics (ACMG) guidance to be informed to their personnel. If pathogenic variants are detected for APC, BRCA1/2, MLH1, MSH2, PTEN, RB1, RET, STK11, SMAD4, TP53, TSC1, and VHL, the results will be reported. Currently, results from other germlines obtained from peripheral blood samples are not returned; results other than the ACMG59 gene are reported as a difference between tumor sample and peripheral blood sample, which means that the germline variant may be present but masked and in most cases not be able to even be suspected. However, the format of the report may change as necessary in the future. The expert panel then should fully consider how and whether to disclose the information.

13.7 C-CAT Report Structure

C-CAT is a new center for cancer genome medicine established in accordance with the Cancer Control Act Law and provides a mechanism for collecting and storing information on cancer genome medicine from all over Japan. C-CAT supports cancer genome medicine in Japan not only by returning the “C-CAT reports” with the annotated genetic alterations to the expert panels at the Cancer Genome Center Hospitals but also by understanding or using the genome and medical information of cancer patients for secondary purpose such as the development of policies for cancer control. For patients who have not consented to secondary use, C-CAT will accept the information if they agree to provide it to C-CAT and will not delete the patient’s information after the patient’s death and will retain the patient’s information [6].

Not only NCC Oncopanel Test but also for F1CDx, if consent is obtained from the patient, the test data are sent to C-CAT, and the C-CAT report is generated. The report format for the Japanese version of the CDx Associated Findings and Other Alterations & Biomarkers Identified report, “Approved Therapeutic Options in Japan,” is based on the information on approved drugs at the time of F1CDx approval in Japan.

The C-CAT report is based on the Cancer Knowledge Data Base (CKDB), which was developed by C-CAT, and consists of survey results, candidate drugs and clinical trials, detailed information on mutated genes and references, and evidence levels. In the matched pair test, the following items are also reported for germline mutations: the names of mutant genes and mutation information, allele frequency, evidence type (predictive, predisposing), clinical significance and disease name, evidence level, and corresponding drug and availability.

13.8 Expert Panel Configuration

The reports need to be reviewed by a multidisciplinary panel of experts in order to provide patients with the appropriate treatment, which is the purpose of the panel tests. To this end, an expert panel meets to discuss and finalize all of the test reports before returning the results to the attending physician. The expert panel consists of the following eight items:

  1. 1.

    Include several full-time physicians in different fields of practice who have specialized knowledge and skills in anticancer drug treatment.

  2. 2.

    Include at least one physician with specialized knowledge and skills in genetic medicine.

  3. 3.

    Include at least one person with specialized genetic counseling skills in genetic medicine.

  4. 4.

    Include more than one physician with specialized knowledge and skills in pathology.

  5. 5.

    One or more experts with sufficient knowledge of molecular genetics and genomic medicine should be included. The expert should have written a peer-reviewed paper in English on cancer genome medicine or cancer genome research within the past 3 years prior to the time of application.

  6. 6.

    If the sequencing will be carried out at the institution, at least one expert with sufficient knowledge of bioinformatics for electronic analysis using next-generation sequencers should be included. The expert should have written a peer-reviewed English-language paper on cancer genome medicine or genome research within 3 years prior to the time of application.

  7. 7.

    In an institution that handles pediatric oncology cases, at least one pediatric oncology physician with some experience of participating in expert panels must be included.

  8. 8.

    Include the attending physician or alternate physician of the subject patient to be reviewed by the expert panel.

The final report consists of not only the reviewed summary of the presence and content of recommended treatments, or other treatment options, but also the presence and content of secondary findings that are recommended to be explained to patients. A secondary finding is defined as a genetic mutation in the germline that can be confirmed as a pathogenic variant in the cancer gene panel test.

The term “secondary findings” is used because they are considered important incidental findings for patients and their families, even though they are not the primary purpose of the test. Recently, it has been proposed that these findings be called germline findings. A pathogenic variant of a germline mutation may not be considered an incidental finding when a patient undergoes a paired-specimen cancer gene panel test, bearing in mind that a patient’s medical history and family history indicate that he or she may have a hereditary tumor. Furthermore, it is possible that a genetic variant such as BRCA1 or BRCA2, for example, that is called by the cancer gene panel test, may be a genetic variant that also defines a treatment approach.

13.9 Handling of Germline Findings in Panel Testing

The evaluation of germline findings is based on five levels of ACMG/AMP criteria: pathogenic, likely pathogenic, variant of uncertain significance, likely benign, and benign. The classification is based on information such as variant frequency information in the general population; functional prediction; functional analysis data; databases such as ClinVar, HGMD, and MGeND; and reported articles.

The possibility of germline findings should be explained to the patient during the explanation of the panel test, and consent should be obtained before having the test. Among the germline findings, genetic mutation findings will only be disclosed to patients for those variants that are determined to be pathogenic or likely pathogenic and that are related to hereditary tumors for which management methods have been established. Disclosing findings related to a disease for which there is no established treatment or surveillance would only cause anxiety and confusion for patients and families. It should be confirmed that the gene is an eligible gene for disclosure in the guidelines developed by each society. In Japan, a proposal on the information transfer process in genomic medicine has been published, and which describes which secondary findings should be picked up and how they should be disclosed [7].

Among the target genes included in the NCC Oncopanel Test, the following 13 genes (APC, BRCA1/2, MLH1, MSH2, PTEN, RB1, RET, STK11, SMAD4, TP53, TSC1, and VHL) correspond to the gene groups to be reported by ACMG.

Although the actual test results will be somewhat scattered with results that would be assessed as mutations of unknown significance, we should be cautious in disclosing such results. Therefore, a genetic medicine specialist or a genetic counselor is required for the expert panel. Cases that should be carefully evaluated for pathogenicity based on a detailed family history and other information should be handled in an outpatient genetic counseling clinic.

The frequency of detection of somatic or germline pathogenic variants using paired specimens varies by gene [8]. For instance, TP53 somatic variants were much more common: 337 patients had somatic and 10 had germline TP53 variants. The same trend was observed for RB1 and PTEN. In TSC2 and MSH2, approximately 80% of the patient had somatic variants. In comparison, BRCA1/2 was commonly seen as germline variants: there were 3 patients with somatic and 11 with germline BRCA1 variants and 3 with somatic and 10 with germline BRCA2 variants. In PALB2 variants, the frequency of germline and somatic variants was even, and in MSH6 variants, less than 60% of the patients had germline variants.

Among several panel testing, the percentage of secondary findings found in the ACMG gene has been reported to range from 3.3 to 10.7% [8,9,10,11].

Pathogenic variants of germline mutation may be suspected in panel tests such as F1CDx, which only targets tumor tissue. For example, BRCA1/2 is most likely pathogenic variant of the germline origin, as mentioned above, and germline confirmation testing should be performed regardless of its allele frequency. When dealing with genes other than BRCA1/2, it is important to consider the primary organ, duplicated or multiple cancers, family history, as well as tumor content of the specimen, copy number alterations, and variant allele frequency (VAF). It is known that VAF is often high when the detected gene mutation is a germline mutation and VAF is often low when the mutation is of tumor origin for genes such as CHEK2, ATM, and PALB2. ESMO Precision Medicine Working Group addressed a guideline for management of tumor-detected pathogenic variants of potential germline origin [12]. They found that crude “pan-tumor” VAF thresholds (20% for small insertions/deletions, 30% for SNVs) enabled reduction by 54% (9222/17075) the number of tumor-detected variants requiring follow-up while losing only 3.5% (52/1494) proportion of true germline variants. After excluding variants from germline-focused tumor analysis of gene/context/age scenarios in which the germline conversion rate is <10%, 27 genes remained. As a result, these 27 genes (at any age, any tumor type, BRCA1, BRCA2, BRIP1, MLH1, MSH2, MSH6, PALB2, PMS2, VHL, RAD51C, RAD51D, RET, SDHA, SDHAF2, SDHB, SDHC, SDHD, TSC2, MUTYH; at any age with associated tumor type only, FLCN, FH, BAP1, POLE; tumor arising age < 30 only with any tumor type, RB1, APC; tumor arising age < 30 with associated tumor type only, TP53, NF1) are recommended to be included for germline-focused analysis and triggering of germline sample laboratory confirmation.

The similar operational guideline is also used in Japan (Fig. 13.2, Table 13.3) [7, 13]. Within the guideline, BRCA1, BRCA2, MLH1, MSH2, MSH6, PMS2, APC, MEN1, RET, RB1, and VHL are the genes recommended to be disclosed to patients. The criteria for this are the existence of Japanese guidelines for surveillance of unaffected patients, the ability to outsource single-site tests to a registered laboratory from any center or affiliated hospital, and variants that are included in several gene panel tests. However, it is possible that other genes may be found in addition to these genes that are associated with hereditary tumors and should be thoroughly examined by expert panels.

Fig. 13.2
figure 2

Operational guidelines for germline tests to confirm secondary findings from tumor profiling test of tumor cells [7]

Full size image
Table 13.3 Disclosure recommendations list for secondary findings of cancer genetic panel test [13]
Full size table

If the possibility of pathogenic variants is considered, it is necessary to confirm the results with another single-site genetic test using normal tissues, such as blood, and it is recommended to disclose this fact in the report [14]. In clinical practice, it is essential to refer cases that are considered to require genetic counseling by the expert panel to a clinical geneticist and certifiedgenetic counselors as appropriate.

13.10 Gene Panel Testing for Germline Variants

In Japan, there are no insured genetic panel tests for the detection of germline mutations, so we must pay for them at our own expense. There are already many multigene panel tests for cancer predisposition mutations available overseas, for example, those provided by Ambry Genetics such as “CancerNext, BreastNext, and BRCAplus”; “Comprehensive Cancer Panel, Breast Cancer High/Moderate Risk, and BreastOvarian Cancer” from GeneDx; and “MyRisk, Breast and Ovarian, Breast Cancer” from Myriad, which are developed by commercial companies or by medical centers such as MSK-IMPACT, Color, Counsyl Reliant Cancer Screen, and University of Washington BROCA Cancer. The number of genes included in a test is so varied that it is difficult to decide which test to choose, even for breast and ovarian cancer. As a pretest probability model, Myriad II [15], BRCAPRO [16], IBIS [17], and others have been used, but these models only cover BRCA1/2; predictive model for other less frequent pathogenic mutations is limited to BOADACEA [18], now including ATM, CHEK2, and PALB2; all of these panel tests were developed on the basis of a small number of data or a biased cohort. Recently, a model has been developed by the Mayo Clinic in collaboration with Ambry Genetics that can predict larger numbers of genetic variations based on a larger number of data [19].

NGS can detect many genetic variants at one time, and the challenge that always accompanies such tests is that some variants of unknown clinical significance may be detected which must be treated with caution. In a study of 1085 BRCA1/2-negative breast cancer patients, O’Leary et al. [20] found that the higher the number of genes included in a panel test, the more the frequency of finding genetic mutations which are considered to be pathogenic or likely pathogenic would increase. Buys et al. [21] reported that pathogenic variants were found in about 10% of the total and about half were BRCA1/2. In addition, one or more VUS was detected in about 37% of all cases.

In light of this current situation, guidelines have been released. For example, American Society of Breast Surgeons (ASBrS) presented consensus guideline on genetic testing for hereditary breast cancer. Although BRCA1/2 is the most frequent pathogenic variant associated with the development of breast and ovarian cancer, more comprehensive panel tests, including other less common syndromes, have become widely available. The most frequently reported variants other than BRCA1/2 are PALB2, CHEK2, and ATM [22,23,24]. The addition of MRI with contrast to annual surveillance is supported when there is a lifetime breast cancer risk of 20% or more, including these genetic variants. In this way, panel testing can contribute to more efficient and cost-effective risk assessment and recommended management of patients for whom hereditary breast cancer testing is recommended than conventional sequential gene testing. There is a report examining the performance of the NCCN genetic testing criteria for BRCA-related breast and/or ovarian cancer syndrome and Lynch syndrome (version 1.2018) in 165,000 patients who underwent hereditary cancer predisposition testing [25]. Within the report, among the female BRCA1/2 pathogenic variant carriers not meeting BRCA1/2 testing criteria, 59.1% (143/242) had a personal history of breast cancer. Meanwhile, of patients with PVs in Lynch syndrome genes failing to meet Lynch criteria, 41.5% had a personal history of a Lynch syndrome-related cancer. Therefore, genetic testing is recommended that “should be made available to all patients with a personal history of breast cancer.”

Furthermore, for VUS, the ASBrS states in its recommendations that variants of uncertain significance are DNA sequences that are not clinically actionable in its recommendations. It is said that a VUS take several years to reclassify its uncertainty [26]. Until then, the variant should be considered as inconclusive and should be managed based on the patient’s own risk factors.

Although there are issues that need to be resolved, it is clear that multigene panel tests will be used more frequently in clinical practice for cases of potential hereditary diseases, and by accumulating evidences flexibly, the development of high-quality international guidelines is warranted in the near future.