Somatic Genetic Testing: Tumor Mutations, Cancer Treatment, and Results

Somatic genetic testing examines acquired changes in cancer cells. These changes develop during a person’s lifetime and help distinguish tumor cells from most normal cells; they are not automatically inherited or present throughout the body. Testing can clarify a cancer diagnosis, define a molecular subtype, estimate prognosis, identify treatments that target a tumor’s biology, and reveal resistance mechanisms. It may use a tissue biopsy, surgical specimen, bone marrow, blood from a blood cancer, or circulating tumor DNA in plasma. The report can include single-gene variants, copy-number changes, fusions, genomic signatures, and biomarkers such as microsatellite instability or tumor mutational burden. A detected alteration is useful only when it is linked to the person’s cancer type, disease stage, treatment history, assay quality, and current evidence. Some tumor findings can also suggest an inherited cancer variant, so somatic testing sometimes leads to separate germline testing with genetic counseling.

• Somatic variants arise in body cells and are usually not inherited.
• Tumor testing can support diagnosis, prognosis, treatment selection, and trial matching.
• Tissue and liquid biopsy results are complementary, not interchangeable in every situation.
• A mutation may be actionable, biologically relevant but not actionable, uncertain, or benign.
• Tumor purity, heterogeneity, prior therapy, and assay design can affect detection.
• A possible inherited finding requires confirmation in a non-tumor sample.

Table of Contents

• What Somatic Testing Looks For
• Why Oncologists Order It
• Tissue, Blood, and Liquid Biopsy Samples
• Tests and Biomarkers on the Report
• How to Interpret Somatic Results
• Limitations and Misleading Findings
• Using Results for Treatment and Follow-Up

What Somatic Testing Looks For

Cancer develops through genetic and epigenetic changes that alter how cells grow, divide, repair damage, avoid immune control, or die. Somatic genetic testing searches for those acquired changes in tumor cells. It may also be called tumor testing, molecular profiling, tumor biomarker testing, or cancer genomic testing.

The defining idea is the source of the variant. A somatic variant is present in some cells of the body—often the cancer and its descendants—rather than in every cell from conception. A germline variant is present in the egg or sperm that formed the person and can therefore be found in most tissues and potentially passed to children. The same gene can contain either type. A BRCA1 variant, for example, may be inherited or may arise only in a tumor.

Somatic testing is not a single assay. A narrow test may look for one common actionable mutation, such as a specific change in a cancer-associated gene. A panel may evaluate hundreds of genes and several types of alteration. More comprehensive profiling may analyze tumor DNA and RNA, compare tumor with normal DNA, or calculate genome-wide signatures.

The changes assessed can include:

• Single-nucleotide variants, in which one DNA letter changes
• Small insertions and deletions
• Copy-number gains or losses, including gene amplification
• Structural rearrangements and gene fusions
• Loss of heterozygosity
• Repeat or length changes in selected regions
• RNA expression or splice changes
• Epigenetic markers, such as methylation patterns
• Composite signatures reflecting DNA repair or mutation burden

Not every change drives cancer. Tumors accumulate passenger variants that are present but do not meaningfully support growth. Driver alterations contribute to cancer development or survival. Even a driver is not necessarily a treatment target: no approved drug may exist, the drug may not work in that tumor type, or resistance changes may undermine the expected benefit.

A somatic report describes the sampled cancer at a particular place and time. Tumors evolve. Different regions of one mass can contain distinct subclones, metastatic sites can differ from the original tumor, and therapy can select resistant cells. The result is therefore a molecular snapshot rather than an immutable profile.

Somatic testing also differs from germline genetic testing. Tumor profiling can incidentally raise hereditary concerns, but it cannot reliably determine inheritance by itself when no matched normal sample is analyzed.

Why Oncologists Order It

The purpose should be defined before the sample is sent because the appropriate test depends on the clinical question.

Diagnosis and classification. Some alterations help confirm that a tumor belongs to a particular disease entity or molecular subtype. A fusion, methylation pattern, or characteristic mutation can distinguish cancers that look similar under a microscope. In hematologic malignancies and some brain, soft-tissue, thyroid, lung, and pediatric tumors, molecular classification may be integrated into the diagnosis.

Treatment selection. A predictive biomarker indicates that a treatment is more or less likely to work. An activating alteration may make a tumor dependent on a signaling pathway that a targeted drug can inhibit. A DNA-repair defect may increase sensitivity to a particular therapy. Conversely, a resistance mutation can show that a drug is unlikely to help.

The strength of a treatment association varies. The most compelling situation is an alteration linked to a regulator-approved therapy for that cancer type, with a validated companion diagnostic when required. Evidence may be weaker when the drug is approved for another tumor type, supported only by a small study, or still experimental. “Targetable” in a laboratory sense is not the same as clinically recommended.

Prognosis. Some variants or signatures are associated with more aggressive or more favorable disease. Prognostic information describes the expected course independent of a particular treatment. It can help with risk stratification, but population-level associations do not predict an individual outcome with certainty.

Therapy resistance and relapse. Testing after progression can identify a newly selected mutation, pathway bypass, gene amplification, or loss of a prior target. This may guide the next treatment or explain why a once-effective drug stopped working. The optimal sample and timing depend on the cancer and therapy.

Clinical-trial matching. Basket trials enroll people with different cancers that share a biomarker, while umbrella trials assign treatments within one cancer type according to molecular findings. A report may list trials, but eligibility also depends on location, stage, prior treatment, organ function, performance status, and enrollment availability.

Monitoring and minimal residual disease. In selected cancers, patient-specific or tumor-informed assays can track circulating tumor DNA after treatment. A molecular signal may indicate residual disease or recurrence risk earlier than imaging, but clinical utility and recommended responses vary by cancer. A monitoring assay is not interchangeable with broad profiling.

Somatic testing may be routine at diagnosis for certain malignancies, particularly where biomarkers are essential to standard treatment. In other cancers it is considered at recurrence, metastatic disease, or when standard options are limited. Testing everyone with the broadest possible panel does not guarantee benefit. The likely yield, tissue availability, turnaround time, cost, and whether an actionable result would change care all matter.

Tissue, Blood, and Liquid Biopsy Samples

A test can only detect what is represented in the sample. Selecting and preparing the specimen is therefore a major part of accuracy.

Tumor tissue remains a central source for solid-cancer testing. The specimen may come from a core biopsy, fine-needle procedure, surgical resection, or archived pathology block. A pathologist reviews the slide to confirm the diagnosis, estimates tumor-cell percentage, marks an area for extraction, and determines whether enough material is present.

Formalin-fixed, paraffin-embedded tissue is widely used because it preserves specimens, but fixation can damage DNA and RNA. Small biopsies may be exhausted by diagnostic staining before molecular testing is ordered. Bone specimens that undergo harsh decalcification can yield degraded nucleic acid. Older archived samples may no longer represent the current resistant disease.

The most recent accessible metastatic site may provide relevant information after progression, but another biopsy is not always safe or feasible. The choice should balance clinical relevance against procedural risk. A low-tumor specimen may produce a false-negative result even when the assay itself performs well.

Blood and bone marrow can directly sample leukemias, lymphomas, myeloma, and related disorders. Interpretation requires awareness that nonmalignant blood cells and age-related clones can also carry somatic variants. Remission samples are not always equivalent to germline controls because residual disease or clonal hematopoiesis may persist.

Liquid biopsy usually refers to analysis of cell-free DNA in plasma, a fraction of which may be circulating tumor DNA. It requires a blood draw and can sometimes capture alterations shed from multiple tumor sites. It is useful when tissue is unavailable, biopsy risk is high, rapid results are needed, or serial sampling could identify emerging resistance.

A liquid biopsy has important limits. Some tumors shed little DNA, particularly when disease volume is low, confined to certain sites, or biologically less vascular. The fraction of tumor-derived DNA can be too small to detect an existing alteration. A negative plasma result may therefore need tissue testing when clinically possible.

Liquid biopsy can also detect DNA from clonal hematopoiesis—age-related blood-cell clones carrying somatic variants. Changes in genes such as DNMT3A, TET2, ASXL1, and others may originate from white blood cells rather than the tumor. Without paired white-cell analysis or expert interpretation, a blood-derived variant can be misassigned to the cancer.

Matched tumor-normal testing compares tumor DNA with DNA from a non-tumor sample, usually blood or saliva. It helps separate likely somatic from germline variants and can improve some analyses. However, blood may be unsuitable as the “normal” comparator for hematologic cancers, after allogeneic stem-cell transplantation, or when clonal hematopoiesis is relevant. Buccal or skin-derived samples may sometimes be preferred.

The report should state the specimen source, collection date, tumor percentage when applicable, nucleic-acid quality, and whether a matched normal sample was tested.

Tests and Biomarkers on the Report

Somatic reports combine different measurement types, and each requires its own interpretation.

DNA sequence variants. These include substitutions and small insertions or deletions. The report commonly lists the gene, DNA and protein notation, variant allele fraction, classification, and possible clinical associations. Variant allele fraction is the proportion of sequencing reads carrying the change. It is influenced by tumor purity, copy number, clonality, contamination, and assay noise; it is not a direct percentage of tumor cells and does not by itself establish germline status.

Copy-number alterations. A tumor may amplify an oncogene or delete a tumor-suppressor region. The clinical meaning depends on assay thresholds and cancer context. A report may describe copy number, fold change, gain, high-level amplification, or loss. Different platforms do not necessarily define these terms identically.

Gene fusions and rearrangements. Rearrangements can create an abnormal fusion protein or alter gene regulation. DNA-based tests may miss some fusions because introns are large or repetitive. RNA-based testing can demonstrate that a fusion transcript is expressed, although poor RNA quality can cause failure. Fluorescence in situ hybridization and immunohistochemistry may provide complementary evidence.

Microsatellite instability and mismatch-repair deficiency. MSI-high status reflects impaired repair of repetitive DNA sequences. It can be assessed by molecular assays, while mismatch-repair protein loss can be evaluated by immunohistochemistry. These biomarkers can inform immunotherapy in appropriate settings and may also raise concern for Lynch syndrome, requiring germline assessment when indicated.

Tumor mutational burden. TMB estimates the number of qualifying somatic mutations per amount of DNA analyzed. A “high” result depends on the platform, panel size, filtering, cutoff, tumor type, and companion-diagnostic context. TMB is an imperfect predictor; a high score does not guarantee immunotherapy response, and a low score does not prove that response is impossible.

Homologous recombination deficiency. HRD assays may combine genomic scars, loss of heterozygosity, or selected gene variants. The clinical usefulness and cutoff depend on cancer type and treatment. A genomic scar reflects past biology and may persist even after a tumor acquires resistance.

Methylation and expression signatures. These can refine tumor classification or predict biological behavior. Their interpretation depends on the reference dataset and technical platform. A classifier score near a threshold may be less definitive than a strong match supported by histology.

The report may rank findings by clinical significance. Older consensus frameworks use tiers such as strong clinical significance, potential clinical significance, unknown significance, and benign/likely benign. Other laboratories use evidence levels or actionability scales. The reader should examine the definitions rather than assume that “Tier II” means the same everywhere.

How to Interpret Somatic Results

A somatic result is interpreted through five linked questions: Is the alteration real? Is it biologically relevant? Is it clinically relevant in this cancer? Is there a treatment or trial? Does the patient’s situation allow that option?

Actionable alteration detected. This means evidence connects the biomarker with a diagnostic, prognostic, or treatment decision. The report should indicate the cancer context and evidence level. An alteration tied to an approved drug for a different tumor type may not justify off-label use, and a drug listed in a knowledge base may have failed in this cancer.

Actionability can be positive or negative. A biomarker may support a treatment, predict resistance, exclude a drug, or establish that another standard test is needed. The oncologist must consider prior exposure, drug availability, comorbidities, interactions, and current guidelines.

Biologically significant but not currently actionable. A driver mutation may confirm the cancer’s biology without changing available treatment. It can still have diagnostic or prognostic value and may become relevant as trials or therapies evolve.

Variant of unknown or uncertain significance. Somatic VUS findings lack enough evidence for clinical use. They should not be matched to treatment merely because they occur in a famous cancer gene. Functional impact, variant location, frequency, cancer type, and published data all matter. A VUS can later be reclassified, but treatment decisions generally should not wait on speculation.

No actionable alteration detected. This does not mean the tumor has no mutations or that targeted therapy can never work. The test may have assessed a limited gene list, missed a variant type, lacked sufficient tumor DNA, or found changes without established therapies. Standard treatment can still be effective and may remain the best-supported option.

Possible germline finding. A variant in BRCA1, BRCA2, mismatch-repair genes, TP53, APC, RET, or another cancer-predisposition gene may be present in the tumor because it was inherited. Variant allele fraction, tumor type, age, and family history can raise suspicion, but none proves inheritance. Separate clinical genetic testing of an appropriate non-tumor sample is needed. Confirming a germline variant can affect treatment, future cancer surveillance, and relatives.

Multiple findings. Tumors often have several drivers, resistance variants, and subclonal alterations. Their interactions can change expected drug sensitivity. A target can be present while a downstream resistance pathway makes inhibition ineffective. Molecular tumor boards can help with complex profiles by integrating pathology, oncology, genetics, pharmacology, and evidence review.

The report date matters. Drug approvals, guideline recommendations, trial availability, and variant databases change rapidly. Reinterpretation may be valuable when disease progresses or substantial time has passed.

Limitations and Misleading Findings

Somatic testing can fail before sequencing begins. Insufficient tissue, low tumor content, necrosis, degraded nucleic acid, decalcification, or exhausted blocks can produce cancellation or limited analysis. A “quantity not sufficient” result is not negative.

Analytical sensitivity creates another boundary. Every assay has a minimum variant allele fraction it can detect. Low-level subclones may fall below that threshold. Copy-number calling becomes difficult when tumor purity is low. Fusions may be missed if the panel lacks the partner or relevant introns. RNA degradation can prevent transcript detection.

Tumor heterogeneity means one biopsy may not represent all cancer cells. A primary tumor and metastasis can differ, as can separate metastatic sites. A treatment-resistant clone may be rare before therapy and dominant afterward. Timing and site should be considered when using an older result to make a new decision.

Liquid biopsy adds biological uncertainty. Low shedding can cause false-negative plasma results. Clonal hematopoiesis can cause false attribution of blood-cell variants to the tumor. Germline variants may also appear in cell-free DNA. Paired analysis and clinical correlation reduce but do not eliminate ambiguity.

Some biomarkers lack platform harmonization. TMB, HRD, copy-number amplification, and circulating tumor fraction may be calculated differently across assays. A threshold validated with one companion diagnostic should not automatically be transferred to another platform.

Annotation can overstate certainty. Automated reports may list drugs associated with a gene rather than the exact variant, disease, or evidence level. Preclinical sensitivity in cells or animals is not the same as proven patient benefit. A report-generated therapy list should be treated as a starting point for oncology review, not a prescription.

A result can also be outdated. Cancer knowledge changes more quickly than most static reports. A variant classified as uncertain may later gain significance; a once-promising treatment may fail in trials; a new resistance mechanism may be recognized. The laboratory’s update policy and the oncology team’s access to current resources matter.

Finally, testing can reveal information the patient did not expect. A likely inherited variant, unexpected biological relationship in paired analysis, or risk of a hematologic clone may require additional evaluation. Consent should address these possibilities where relevant.

Using Results for Treatment and Follow-Up

The oncology team should interpret the report alongside pathology, stage, treatment history, performance status, organ function, patient goals, and standard-of-care guidelines. A molecular match is one part of the decision, not a replacement for clinical judgment.

For each potentially actionable result, ask:

1. Is the alteration classified as pathogenic or otherwise validated as a predictive biomarker?
2. Is the association proven in this cancer type and disease stage?
3. Is there an approved therapy, guideline recommendation, or suitable clinical trial?
4. Was the biomarker measured by an accepted or required companion diagnostic?
5. Are resistance findings present?
6. Can the patient safely receive the treatment?

When evidence is strong, testing can lead to a targeted therapy, immunotherapy, altered chemotherapy choice, different surgical or radiation strategy, or enrollment in a biomarker-selected trial. When evidence is weak, the safest course may be standard therapy while a molecular tumor board considers research options.

A negative or nonactionable report should trigger a quality check rather than disappointment alone. Confirm that the sample was adequate and the relevant genes, fusions, signatures, and variant types were covered. If plasma testing was negative, tissue may be appropriate. If old tissue was used after several therapies, a new biopsy or repeat liquid biopsy may better capture resistance—provided the result would change care.

Possible germline findings should be referred for genetic counseling and confirmatory testing. Tumor testing does not replace hereditary cancer evaluation based on age, tumor type, personal history, and family history. A person may meet germline-testing criteria even when the somatic report does not flag an inherited variant.

Preserve the full report, specimen details, and pathology records. Future clinicians need the exact assay, test date, detection limits, and variant notation. A screenshot of a summary page is not enough.

Repeat testing is not automatically beneficial. It is most useful when the tumor has evolved, a new specimen is more representative, new therapies require biomarkers not previously assessed, or the original assay was limited. Serial testing should have a defined clinical question.

Somatic testing is most powerful when it connects three levels of evidence: the alteration’s biology, the test’s technical reliability, and the patient’s real treatment context. That integration turns a list of tumor mutations into a medically meaningful plan.

References

1. National Cancer Institute: Biomarker Testing for Cancer Treatment
2. ESMO Recommendations for the Use of Next-Generation Sequencing in Advanced Cancer in 2024
3. National Cancer Institute: The Genetics of Cancer
4. Lexicon for Clonal Hematopoiesis in Liquid Biopsy
5. Liquid Biopsy in Cancer: Current Status, Challenges, and Future Prospects
6. Standards and Guidelines for the Interpretation and Reporting of Sequence Variants in Cancer

Disclaimer

This article provides general education and does not replace oncology care, pathology review, genetic counseling, or interpretation by a qualified molecular laboratory professional. Cancer treatment should not be started, stopped, or changed solely from a tumor report without considering the cancer type, specimen quality, evidence level, prior therapy, and the patient’s clinical circumstances. A suspected inherited finding requires separate confirmatory germline testing.