Broad CGP vs Targeted Hotspot Panels in Oncology

A broad comprehensive genomic profiling panel can find alterations that a targeted hotspot panel will miss, preserve limited tissue, and support therapy selection in complex disease. It can also produce a longer report, a higher bill, and a collection of biologically interesting findings that do not change treatment.

That is the operational reality behind comprehensive genomic profiling vs targeted panel decisions. The question is not which assay is more advanced. The question is whether the clinical problem justifies broader coverage—and whether the laboratory, tumor board, and treating team can convert that coverage into an actionable decision.

Diagnostic yield: more genomic territory, fewer blind spots

Targeted hotspot testing is designed around a narrow hypothesis. If the clinical team suspects a specific driver alteration, the assay interrogates the relevant hotspot or coding region and returns a focused answer. This is often exactly what is needed when the tumor type has a relatively well-defined molecular landscape and treatment decisions depend on a small number of established biomarkers.

Broad CGP starts from a different premise. Rather than asking whether one or two known variants are present, it surveys hundreds of genes in parallel and evaluates several classes of genomic alteration:

• Single nucleotide variants, or SNVs
• Insertions and deletions, including short indels
• Copy number alterations
• Gene fusions and rearrangements
• Tumor Mutational Burden, or TMB
• Microsatellite Instability, or MSI

That distinction matters because clinically relevant biology does not always stay inside a convenient hotspot. A driver may be represented by a fusion, a copy number change, a less common variant, or a genomic signature that a narrow assay was never designed to capture.

The yield difference can be substantial. In a comparative study of 57 metastatic solid tumor cases, hotspot testing identified genomic alterations in 77% of cases, or 44 out of 57. CGP found at least one alteration in 98%, or 56 out of 57. That is not a universal performance guarantee; it is a useful illustration of what happens when the assay’s search space expands.

The larger panel does not create biology. It creates more opportunities to observe it.

That sounds obvious, but the distinction becomes important at sign-out. Detection is not the same as clinical utility. A broad assay may identify a pathogenic alteration without a corresponding approved therapy, a trial the patient can access, or sufficient evidence in that tumor context. The report can be analytically correct and clinically inconclusive at the same time.

Broader sequencing improves the odds of finding a signal. It does not promise that the signal comes with a treatment attached.

A targeted assay, by contrast, has a smaller field of view but a cleaner interpretive burden. When the question is specific, a narrow test can be efficient. In colorectal cancer or melanoma, for example, a defined set of driver alterations may be central to immediate management. In those settings, a hotspot or focused panel may provide the required answer without making the laboratory process carry hundreds of additional genes.

The phrase broad CGP vs targeted hotspot testing therefore describes two different testing philosophies:

The exact gene count is not a magic legal boundary. There is no universal regulatory definition establishing one minimum number of genes that turns a panel into CGP. In practice, comprehensive assays commonly evaluate more than 60 biomarkers and often hundreds or more than 500 genes. But counting genes is a poor substitute for examining what the assay can actually detect.

A 500-gene panel that does not robustly assess fusions is not equivalent to a 300-gene panel with validated fusion detection. A report that lists TMB but does not establish how the metric was calculated or validated should not be treated as interchangeable with another report using a different genomic footprint. Breadth is multidimensional. The marketing slide usually is not.

Tissue efficiency: the hidden argument for going broad

Tissue is not an abstract laboratory input. In advanced cancer, it is a constrained clinical resource, and repeated testing can consume it faster than the treatment team expects.

Sequential single-biomarker testing looks economical when each assay is viewed in isolation. A single-gene test costs less than a broad panel. The arithmetic changes when the team orders one test, waits for the result, returns to the block, orders another test, and repeats the cycle after an inadequate result or a change in treatment strategy.

Up to almost 30% of patients with advanced lung cancer may lack sufficient tissue for sequential single-biomarker molecular testing. That is not a minor inconvenience. It can force a choice between exhausting the specimen and accepting an incomplete molecular profile. Neither outcome is especially compatible with precision oncology.

The practical advantage of CGP is not simply that it tests more genes. It can consolidate several molecular questions into one tissue event. A properly selected assay may evaluate SNVs, indels, copy number alterations, fusions, and genomic signatures from the same specimen. This is particularly relevant when the biopsy is small, necrotic, decalcified, or already heavily consumed by histology and immunohistochemistry.

But tissue conservation is not automatic. A broad panel may require more nucleic acid than a focused assay, and low tumor fraction can compromise performance. A laboratory still needs to assess specimen type, tumor content, fixation quality, DNA and RNA requirements, and whether the assay’s validated input range fits the sample in front of it.

The friction appears at several points:

1. Specimen triage. The block must be reviewed before testing, not after the assay fails. Tumor percentage, viable tumor area, and prior tissue use shape what is feasible.

2. Nucleic acid strategy. DNA-only testing may not adequately address fusion detection when RNA-based analysis is required or when DNA breakpoints are difficult to capture.

3. Failure interpretation. An assay that returns no reportable result is not the same as a negative assay. Low input, low tumor fraction, or degraded material can produce a technically limited outcome.

4. Reflex planning. The laboratory needs a predefined path for orthogonal testing, liquid biopsy, or additional tissue—not an improvised rescue operation after the sample is gone.

5. Clinical timing. Tissue efficiency has value only if the result arrives while treatment decisions are still open.

This is where workflow integration becomes more important than platform specifications. A broad panel dropped into a fragmented process does not conserve tissue very well. It may simply move the bottleneck from assay ordering to accessioning, pre-analytic review, variant interpretation, or tumor board scheduling.

Liquid biopsy can provide another route when tissue is insufficient, but it is not a universal replacement. Circulating tumor DNA may be useful for detecting genomic alterations in selected settings, yet a negative plasma result can reflect low shedding rather than the absence of a tumor alteration. The decision to use a liquid-biopsy CGP assay should follow the clinical question and the limitations of the specimen—not the availability of a fashionable platform.

Clinical utility: the panel is only as useful as the decision around it

The strongest argument for broad CGP appears in advanced, refractory, and rare malignancies, where the likely genomic landscape is less predictable and the cost of missing an actionable alteration is high.

A retrospective study of 676 patients reviewed by a Molecular Tumour Board found that comprehensive NGS panels evaluating more than 60 biomarkers improved eligibility for personalised therapies compared with smaller panels evaluating 60 or fewer biomarkers. The differences were especially notable in several tumor types:

• Cholangiocarcinoma: 43% eligibility with the comprehensive approach versus 17% with the small-panel approach
• Pancreatic carcinoma: 35% versus 3%
• Gastro-oesophageal carcinoma: 40% versus 0%

These figures describe the study population and its testing and interpretation framework. They do not mean that every patient with one of these cancers will benefit from CGP, or that every detected alteration leads to an approved therapy. They do show why a narrow panel can be a poor fit when the clinical question is broad and the disease has few standard treatment paths.

Molecular tumor boards are often where the difference between detection and action becomes visible. A broad result may be categorized as:

• Directly actionable under an established biomarker-treatment relationship
• Potentially actionable through an off-label strategy
• Relevant to a clinical trial
• Prognostic or diagnostic rather than treatment-directing
• Biologically notable but without a current management consequence
• Technically uncertain or limited by assay performance

That classification requires expertise across pathology, oncology, molecular genetics, clinical trials, and laboratory medicine. The algorithmic output is not the tumor board. It is an input to the tumor board.

This is also where algorithmic bias can enter the workflow. Variant interpretation systems are trained and curated using available evidence, which is not evenly distributed across tumor types, populations, or rare alterations. A familiar hotspot with abundant literature may receive a straightforward classification, while a rare fusion or unusual variant may be underrepresented in databases and require manual review. A laboratory that treats automated annotation as the final clinical answer is outsourcing judgment to a software layer.

The report should make its limits legible. Clinicians need to know the assay’s coverage, analytical sensitivity, variant classes assessed, quality metrics, and whether a finding is supported by tumor-specific evidence or only by a broader biological analogy. Otherwise, a long report creates an illusion of certainty. More pages are not the same as more information.

Cost and turnaround: the cheap test can become an expensive sequence

In the United States, targeted single-gene or hotspot panels typically cost about $300–$3,000 per sample. Tissue-based CGP assays run roughly $1,300–$5,000, while liquid-biopsy CGP assays are commonly around $4,700–$5,000. These are US pricing ranges, not a universal tariff, and payer coverage, laboratory contracts, institutional purchasing, and test indication can change the actual financial exposure.

The obvious conclusion is that targeted testing costs less. The more useful conclusion is that the price comparison has to reflect the testing pathway.

If a patient needs one known biomarker and the result will determine the next treatment step, a targeted assay may be the rational choice. If the patient may require several sequential tests, or if the tumor is rare, refractory, or genomically heterogeneous, the cost of repeated narrow testing can become harder to defend—especially when tissue is consumed without producing a complete profile.

Turnaround time also complicates the comparison. Focused assays can be fast, which matters when an established treatment decision is waiting on one result. Yet many modern CGP assays report in under 10–14 days. That puts broad testing within a clinically relevant window in many workflows, provided the laboratory has stable logistics and the specimen does not require extensive rescue work.

The phrase provided the workflow is stable is doing considerable labor here.

A vendor may quote an assay turnaround time from sample receipt to report release. The treating team experiences a longer pathway: order entry, insurance authorization where applicable, specimen retrieval, pathology review, shipment, accessioning, quality assessment, analysis, sign-out, and clinical review. The operational metric that matters is not the instrument’s run time. It is time from decision to usable clinical action.

For laboratory managers, the relevant questions are less glamorous and more consequential:

• Can the assay be ordered through the existing electronic workflow?
• Does the pathology team review the specimen before submission?
• Are failed or insufficient samples tracked by root cause?
• Does the report clearly separate detected alterations from actionable alterations?
• Is there a mechanism for rapid molecular tumor board review?
• Can the laboratory maintain validation and version control as the bioinformatics pipeline changes?
• Are clinicians equipped to interpret negative results, especially in liquid biopsy?

This is deployment, not demonstration. The assay may perform impressively in a conference presentation and still create friction in the ordering queue, accessioning bench, or clinical interpretation process.

The relevant turnaround time is not when the pipeline finishes. It is when the result can still change the treatment decision.

Choosing the assay by disease context

There is no universal winner in the comprehensive genomic profiling vs targeted panel debate. The right choice depends on the uncertainty the assay is meant to reduce.

A targeted panel is usually defensible when:

• The tumor has a well-established, limited set of clinically relevant alterations.
• The treatment decision depends on one or a few known biomarkers.
• The specimen is adequate for the selected test.
• A rapid answer is more valuable than a broad exploratory profile.
• The result will be interpreted within a clear clinical pathway.

Broad CGP becomes more compelling when:

• The malignancy is advanced, metastatic, refractory, or rare.
• Several alteration classes could influence treatment.
• Tissue is limited and sequential testing risks exhausting the specimen.
• The initial diagnosis is uncertain or a molecular finding could refine classification.
• The patient may be eligible for a biomarker-selected clinical trial.
• A Molecular Tumour Board or equivalent multidisciplinary process is available.
• The assay’s coverage includes the variant classes relevant to the tumor and treatment question.

The selection process should begin with the clinical decision, not the panel brochure. Ask what must be known now, what may become relevant after progression, and whether the specimen can support a single broad analysis. Then examine the assay’s actual coverage and the laboratory’s ability to interpret it.

For example, a broad DNA panel may identify SNVs and CNVs well but have limited capability for RNA-dependent fusion detection. A focused test may be more appropriate if a specific rearrangement is the immediate question. Conversely, ordering only a hotspot assay in a heavily pretreated cancer may create a false sense of completeness. The assay answered the question it was built to answer. The problem is that the clinical team may have needed a different question answered.

Precision oncology panel selection is therefore a clinical-laboratory design decision. It includes analytical validation, specimen stewardship, bioinformatics, reporting, reimbursement, and treatment access. Ignoring any one of these produces a technically sophisticated but clinically awkward service.

The practical verdict

Broad CGP is not automatically superior to targeted hotspot testing. It is superior when the disease biology, specimen constraints, and clinical decision-making justify a wider search.

For a simple, urgent, well-defined biomarker question, a targeted panel remains efficient and clinically appropriate. It is cheaper, focused, and often easier to operationalize. Calling it obsolete would be a category error.

For advanced or rare disease, uncertain biology, multiple possible therapeutic pathways, or tissue at risk from sequential testing, broad CGP often offers the better diagnostic strategy. Its value comes from combining alteration classes, conserving a limited specimen, and increasing the chance that the laboratory finds something the original hypothesis did not include.

But the broad panel must clear a higher bar. It needs validated coverage, transparent limitations, controlled interpretation, and a workflow that connects the result to a real treatment decision. Without that integration, the laboratory has purchased genomic breadth but not clinical utility.

The binary answer is straightforward: targeted panels are ready when the clinical question is narrow; broad CGP is ready when the question is not.