news

In the past decade, gene sequencing technology has been widely used in cancer research and clinical practice, becoming an important tool to reveal the molecular characteristics of cancer. Advances in molecular diagnosis and targeted therapy have promoted the development of tumor precision therapy concepts and brought great changes to the entire field of tumor diagnosis and treatment. Genetic testing can be used to warn cancer risk, guide treatment decisions and evaluate prognosis, and is an important tool to improve patient clinical outcomes. Here, we summarize the recent articles published in CA Cancer J Clin, JCO, Ann Oncol and other journals to review the application of genetic testing in cancer diagnosis and treatment.

Somatic mutations and germline mutations. In general, cancer is caused by DNA mutations that can be inherited from parents (germline mutations) or acquired with age (somatic mutations). Germ line mutations are present from birth, and the mutator usually carries the mutation in the DNA of every cell in the body and can be passed on to offspring. Somatic mutations are acquired by individuals in non-gametic cells and are usually not passed on to offspring. Both germline and somatic mutations can destroy the normal functional activity of cells and lead to malignant transformation of cells. Somatic mutations are a key driver of malignancy and the most predictive biomarker in oncology; however, approximately 10 to 20 percent of tumor patients carry germline mutations that significantly increase their cancer risk, and some of these mutations are also therapeutic.
Driver mutation and passenger mutation. Not all DNA variants affect cell function; on average, it takes five to ten genomic events, known as “driver mutations,” to trigger normal cell degeneration. Driver mutations often occur in genes closely related to cell life activities, such as genes involved in cell growth regulation, DNA repair, cell cycle control and other life processes, and have the potential to be used as therapeutic targets. However, the total number of mutations in any cancer is quite large, ranging from a few thousand in some breast cancers to more than 100,000 in some highly variable colorectal and endometrial cancers. Most mutations have no or limited biological significance, even if the mutation occurs in the coding region, such insignificant mutational events are called “passenger mutations”. If a gene variant in a particular tumor type predicts its response to or resistance to treatment, the variant is considered clinically operable.
Oncogenes and tumor suppressor genes. Genes that are frequently mutated in cancer can be roughly divided into two categories, oncogenes and tumor suppressor genes. In normal cells, the protein encoded by oncogenes mainly plays the role of promoting cell proliferation and inhibiting cell apoptosis, while the protein encoded by oncosuppressor genes is mainly responsible for negatively regulating cell division to maintain normal cell function. In the malignant transformation process, genomic mutation leads to the enhancement of oncogene activity and the decrease or loss of oncosuppressor gene activity.
Small variation and structural variation. These are the two main types of mutations in the genome. Small variants alter DNA by changing, deleting, or adding a small number of bases, including base insertion, deletion, frameshift, start codon loss, stop codon loss mutations, etc. Structural variation is a large genome rearrangement, involving gene segments ranging in size from a few thousand bases to the majority of the chromosome, including gene copy number changes, chromosome deletion, duplication, inversion or translocation. These mutations may cause a reduction or enhancement of protein function. In addition to changes at the level of individual genes, genomic signatures are also part of clinical sequencing reports. Genomic signatures can be seen as complex patterns of small and/or structural variations, including tumor mutation load (TMB), microsatellite instability (MSI), and homologous recombination defects.

Clonal mutation and subclonal mutation. Clonal mutations are present in all tumor cells, are present at diagnosis, and remain present after treatment advances. Therefore, clonal mutations have the potential to be used as tumor therapeutic targets. Subclonal mutations are present in only a subset of cancer cells and may be detected at the beginning of diagnosis, but disappear with subsequent recurrence or appear only after treatment. Cancer heterogeneity refers to the presence of multiple subclonal mutations in a single cancer. Notably, the vast majority of clinically significant driver mutations in all common cancer species are clonal mutations and remain stable throughout cancer progression. Resistance, which is often mediated by subclones, may not be detected at the time of diagnosis but appears when it relapses after treatment.

The traditional technique FISH or cell karyotype is used to detect changes at the chromosomal level. FISH can be used to detect gene fusions, deletions, and amplifications, and is considered the “gold standard” for detecting such variants, with high accuracy and sensitivity but limited throughput. In some hematologic malignancies, especially acute leukemia, karyotyping is still used to guide diagnosis and prognosis, but this technique is gradually being replaced by targeted molecular assays such as FISH, WGS, and NGS.
Changes in individual genes can be detected by PCR, both real-time PCR and digital drop PCR. These techniques have high sensitivity, are particularly suitable for the detection and monitoring of small residual lesions, and can obtain results in a relatively short time, the disadvantage is that the detection range is limited (usually only detect mutations in one or a few genes), and the ability to multiple tests is limited.
Immunohistochemistry (IHC) is a protein-based monitoring tool commonly used to detect the expression of biomarkers such as ERBB2 (HER2) and estrogen receptors. IHC can also be used to detect specific mutated proteins (such as BRAF V600E) and specific gene fusions (such as ALK fusions). The advantage of IHC is that it can be easily integrated into the routine tissue analysis process, so it can be combined with other tests. In addition, IHC can provide information on subcellular protein localization. The disadvantages are limited scalability and high organizational demands.
Second-generation sequencing (NGS) NGS uses high-throughput parallel sequencing techniques to detect variations at the DNA and/or RNA level. This technique can be used to sequence both the whole genome (WGS) and the gene regions of interest. WGS provides the most comprehensive genomic mutation information, but there are many obstacles to its clinical application, including the need for fresh tumor tissue samples (WGS is not yet suitable for analyzing formalin-immobilized samples) and the high cost.
Targeted NGS sequencing includes whole exon sequencing and target gene panel. These tests enrich regions of interest by DNA probes or PCR amplification, thereby limiting the amount of sequencing required (the whole exome makes up 1 to 2 percent of the genome, and even large panels containing 500 genes make up only 0.1 percent of the genome). Although whole exon sequencing performs well in formalin-fixed tissues, its cost remains high. Target gene combinations are relatively economical and allow flexibility in selecting genes to be tested. In addition, circulating free DNA (cfDNA) is emerging as a new option for genomic analysis of cancer patients, known as liquid biopsies. Both cancer cells and normal cells can release DNA into the bloodstream, and the DNA shed from cancer cells is called circulating tumor DNA (ctDNA), which can be analyzed to detect potential mutations in tumor cells.
The choice of test depends on the specific clinical problem to be addressed. Most of the biomarkers associated with approved therapies can be detected by FISH, IHC, and PCR techniques. These methods are reasonable for the detection of small amounts of biomarkers, but they do not improve the efficiency of detection with increasing throughput, and if too many biomarkers are detected, there may not be enough tissue for detection. In some specific cancers, such as lung cancer, where tissue samples are difficult to obtain and there are multiple biomarkers to test for, using NGS is a better choice. In conclusion, the choice of assay depends on the number of biomarkers to be tested for each patient and the number of patients to be tested for the biomarker. In some cases, the use of IHC/FISH is sufficient, especially when the target has been identified, such as the detection of estrogen receptors, progesterone receptors, and ERBB2 in breast cancer patients. If more comprehensive exploration of genomic mutations and the search for potential therapeutic targets is required, NGS is more organized and cost-effective. In addition, NGS may be considered in cases where IHC/FISH results are ambiguous or inconclusive..

Different guidelines give guidance on which patients should be eligible for genetic testing. In 2020, the ESMO Precision Medicine Working Group issued the first NGS testing recommendations for patients with advanced cancer, recommending routine NGS testing for advanced non-squamous non-small cell lung cancer, prostate cancer, colorectal cancer, bile duct cancer, and ovarian cancer tumor samples, and in 2024, ESMO updated on this basis, recommending the inclusion of breast cancer and rare tumors. Such as gastrointestinal stromal tumors, sarcomas, thyroid cancers and cancers of unknown origin.
In 2022, ASCO’s Clinical Opinion on somatic genome testing in patients with metastatic or advanced cancer states that if a biomarker related therapy is approved in patients with metastatic or advanced solid tumors, genetic testing is recommended for these patients. For example, genomic testing should be performed in patients with metastatic melanoma to screen for BRAF V600E mutations, as RAF and MEK inhibitors are approved for this indication. In addition, genetic testing should also be performed if there is a clear marker of resistance for the drug to be administered to the patient. Egfrmab, for example, is ineffective in KRAS mutant colorectal cancer. When considering a patient’s suitability for gene sequencing, the patient’s physical status, comorbidities, and tumor stage should be integrated, because the series of steps required for genome sequencing, including patient consent, laboratory processing, and analysis of sequencing results, require the patient to have adequate physical capacity and life expectancy.
In addition to somatic mutations, some cancers should also be tested for germline genes. Testing for germ line mutations may influence treatment decisions for cancers such as BRCA1 and BRCA2 mutations in breast, ovarian, prostate, and pancreatic cancers. Germline mutations may also have implications for future cancer screening and prevention in patients. Patients who are potentially suitable for testing for germline mutations need to meet certain conditions, which involve factors such as family history of cancer, age at diagnosis, and type of cancer. However, many patients (up to 50%) carrying pathogenic mutations in the germ line do not meet traditional criteria for testing for germ line mutations based on family history. Therefore, to maximize the identification of mutation carriers, the National Comprehensive Cancer Network (NCCN) recommends that all or most patients with breast, ovarian, endometrial, pancreatic, colorectal, or prostate cancer be tested for germ line mutations.
With regard to the timing of genetic testing, because the vast majority of clinically significant driver mutations are clonal and relatively stable over the course of cancer progression, it is reasonable to perform genetic testing on patients at the time of diagnosis of advanced cancer. For subsequent genetic testing, especially after molecular targeted therapy, ctDNA testing is more advantageous than tumor tissue DNA, because blood DNA can contain DNA from all tumor lesions, which is more conducive to obtaining information about tumor heterogeneity.
Analysis of ctDNA after treatment may be able to predict tumor response to treatment and identify disease progression earlier than standard imaging methods. However, protocols for using these data to guide treatment decisions have not been established, and ctDNA analysis is not recommended unless in clinical trials. ctDNA can also be used to assess small residual lesions after radical tumor surgery. ctDNA testing after surgery is a strong predictor of subsequent disease progression and may help determine whether a patient will benefit from adjuvant chemotherapy, but it is still not recommended to use ctDNA outside of clinical trials to guide adjuvant chemotherapy decisions.

Data processing The first step in genome sequencing is to extract DNA from patient samples, prepare libraries, and generate raw sequencing data. The raw data requires further processing, including filtering low-quality data, comparing it with the reference genome, identifying different types of mutations through different analytical algorithms, determining the effect of these mutations on protein translation, and filtering germ line mutations.
Driver gene annotation is designed to distinguish driver and passenger mutations. Driver mutations lead to loss or enhancement of tumor suppressor gene activity. Small variants that lead to the inactivation of tumor suppressor genes include nonsense mutations, frameshift mutations, and key splicing site mutations, as well as less frequent start codon deletion, stop codon deletion, and a wide range of intron insertion/deletion mutations. In addition, missense mutations and small intron insertion/deletion mutations can also lead to loss of tumor suppressor gene activity when affecting important functional domains. Structural variants that lead to loss of tumor suppressor gene activity include partial or complete gene deletion and other genomic variants that lead to destruction of the gene reading frame. Small variants that lead to enhanced function of oncogenes include missense mutations and occasional intron insertions/deletions that target important protein functional domains. In rare cases, protein truncation or splicing site mutations can lead to the activation of oncogenes. Structural variations that lead to oncogene activation include gene fusion, gene deletion, and gene duplication.
Clinical interpretation of genomic variation assesses the clinical significance of identified mutations, i.e. their potential diagnostic, prognostic, or therapeutic value. There are several evidence-based grading systems that can be used to guide the clinical interpretation of genomic variation.
The Memorial Sloan-Kettering Cancer Center’s Precision Medicine Oncology Database (OncoKB) classifies gene variants into four levels based on their predictive value for drug use: Level 1/2, FDA-approved, or clinically-standard biomarkers that predict the response of a specific indication to an approved drug; Level 3, FDA-approved or non-approved biomarkers that predict response to novel targeted drugs that have shown promise in clinical trials, and Level 4, non-FDA-approved biomarkers that predict response to novel targeted drugs that have shown convincing biological evidence in clinical trials. A fifth subgroup associated with treatment resistance was added.
The American Society for Molecular Pathology (AMP)/American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guidelines for the interpretation of somatic variation divide somatic variation into four categories: Grade I, with strong clinical significance; Grade II, with potential clinical significance; Grade III, clinical significance unknown; Grade IV, not known to be clinically significant. Only grade I and II variants are valuable for treatment decisions.
ESMO’s Molecular Target Clinical Operability Scale (ESCAT) classifies gene variants into six levels: Level I, targets suitable for routine use; Phase II, a target that is still being studied, is likely to be used to screen the patient population that could benefit from the target drug, but more data are needed to support it. Grade III, targeted gene variants that have demonstrated clinical benefit in other cancer species; Grade IV, only targeted gene variants supported by preclinical evidence; In grade V, there is evidence to support the clinical significance of targeting the mutation, but single-drug therapy against the target does not extend survival, or a combination treatment strategy can be adopted; Grade X, lack of clinical value.