Signature · designed for one question

Signature Custom

When a catalogue tier cannot answer the question, the panel is designed around it instead: gene content, alteration classes, assay methods and reporting format, built backwards from the decision it has to inform.

Most cases are answered by a standard tier. We will tell you when yours is one of them.

Start a scoping conversation Compare the standard tiers

Signature Custom

When the catalogue does not fit the question

Most cases are answered by a standard tier. This page is about the ones that are not.

The Signature catalogue is designed around the questions that come up most often, and for the large majority of cases one of those tiers is the right answer. A custom panel costs more, takes longer to design and validate, and is worth it only when a standard tier genuinely cannot answer the question in front of you.

So the first thing this service does is check whether that is true. If a catalogue tier answers your question, we will say so and point you at it.

When it is warranted

Four situations where a custom design earns its place

These are the cases we see most often. If yours does not look like one of them, a catalogue tier is probably the better answer.

Situation 01

A rare disease the catalogue was not built around

Sarcoma, paediatric tumours, rare endocrine and neuroendocrine disease.

Why the catalogue struggles: Sarcoma alone covers more than a hundred distinct diagnoses, many defined by a specific fusion. A general solid-tumour panel reads the common drivers well and the defining fusion of a rare sarcoma not at all. Where the diagnosis itself depends on finding one structural event, the panel has to be built to look for it.
Typical examples: Ewing sarcoma, synovial sarcoma, infantile fibrosarcoma, rhabdoid tumours

Situation 02

A classification that is molecular by definition

Glioma, endometrial cancer, and other diseases where the diagnosis is the molecular result.

Why the catalogue struggles: Adult glioma cannot be classified without IDH status and 1p/19q codeletion, and MGMT promoter methylation is a separate epigenetic test that a mutation panel does not perform. Endometrial cancer is now sorted into four molecular groups. In diseases like these the panel is not an addition to the diagnosis, it is part of making it, and it has to carry every assay the classification requires.
Typical examples: Glioma (IDH, 1p/19q, MGMT), endometrial molecular classification, medulloblastoma subgroups

Situation 03

Exon-level detail that changes the answer

GIST, where the gene name alone is not enough to act on.

Why the catalogue struggles: In gastrointestinal stromal tumour, which exon of KIT is altered changes the discussion, and the PDGFRA D842V variant behaves differently from every other alteration in that gene. A report naming only the gene is not actionable. Where a disease turns on that level of resolution, the panel and the report format are designed around it.
Typical examples: GIST (KIT exon 9 and 11, PDGFRA D842V), RET in medullary thyroid carcinoma

Situation 04

A programme rather than a patient

Research cohorts, trial screening, and population or corporate screening programmes.

Why the catalogue struggles: A study, a trial pre-screen or a screening programme has fixed requirements: the same genes on every sample, a defined reporting format, and a governance framework agreed in advance. That is a design problem, not a catalogue choice, and it is scoped with the sponsor before a single sample is run.
Typical examples: Biomarker-matched trial pre-screening, cohort studies, structured screening programmes

Design inputs

What we need in order to design one

A custom panel is designed backwards from the decision it has to inform. These six things determine what is possible, and the first conversation covers all of them.

The clinical question

Stated as something that could be answered yes or no. If it cannot be stated that way, the panel cannot be designed around it.

The diagnosis, as precisely as it is known

Including where it is uncertain. Uncertainty is itself a design input, and often the reason a custom panel is being considered.

What has already been tested

So the design fills gaps rather than repeating work. This is frequently where we recommend Decipher instead.

The sample available

Blood, fresh tissue, an archival block, or bone marrow. Sample type constrains which assays are possible, not the other way round.

Which alteration classes matter

Point mutations, insertions and deletions, fusions, copy number, methylation, expression. Each needs a different method, and a panel that reads only one class will miss the others.

When the answer is needed

A custom design takes longer than a catalogue tier. If the decision is next week, the honest answer is usually a standard panel now.

The process

How a custom panel is built

Step 1Scoping conversation
The question, the diagnosis, what has already been tested and when the answer is needed. This is where we say if a catalogue tier would do.
Step 2Design
Gene content, alteration classes, assay methods and the reporting format, drafted against the question rather than against a wish list.
Step 3Feasibility and validation
Whether the sample supports it, and whether the assay performs. A design that cannot be validated does not run.
Step 4Run and interpret
Accredited sequencing, then the OnKommon engine, with the same evidence tiering as any other Signature test.
Step 5KPCIRC sign-out
A molecular tumour board authors and signs, exactly as for a catalogue tier.

The timing question

When molecular information matters

Design and validation take time that a catalogue tier does not, so the decision to go custom is partly a decision about when the answer is needed.

Before anything elseA scoping conversation, to establish whether a standard tier answers the question.
If a standard tier fitsOrder it. A custom panel that arrives later with the same answer has cost you weeks.
If the diagnosis itself depends on molecular resultsDesign happens alongside pathology rather than after it, because the classification cannot be completed without both.
Where a sample is limitedDesign has to account for the material available. Sequential requests on a small biopsy exhaust it, which is a common and avoidable failure.
For a programmeScoping runs months ahead of the first sample, with governance agreed before recruitment starts.

Take these to your team

Questions worth raising

Before commissioning a custom panel

Questions worth raising with a treating team, or with us.

Would a standard Signature tier answer this question, and if not, precisely why not?
Which alteration classes does the question require, and does the design read all of them?
Is there enough sample for the design, or will it need to be prioritised?
How long will design and validation take, and is the decision able to wait that long?
Has prior testing been reviewed, so the design fills gaps rather than repeating work?

After the result

A custom panel is designed around a hypothesis. Sometimes the hypothesis is wrong.

If the panel returns nothing actionable, what is the next step, and was that agreed beforehand?
Has the design been revisited in light of what the first run showed?

Being clear about our limits

What we do and do not do

What we DO

Tell you when a catalogue tier answers your question, before you commission anything
Design backwards from the decision, not from a list of interesting genes
Validate the design on the sample type it will actually be run on
State which alteration classes the design can and cannot detect
Apply the same evidence tiering and the same KPCIRC sign-out as any other Signature test

What we DON’T do

Run a design that cannot be validated to our laboratory partner’s standard
Promise that a larger or bespoke panel will find something
Substitute a custom design for a standard panel that would have worked
Commit to a turnaround before the design and the sample are known

Held to the same standard as every other Signature test

A custom design does not run to a lower bar. The same laboratory accreditation, the same evidence frameworks and the same human sign-out apply.

NABL

ISO 15189

National Accreditation Board for Testing and Calibration Laboratories, the Indian accreditation for medical testing laboratories, assessed against ISO 15189.

CAP

College of American Pathologists

Accreditation covering laboratory quality management, proficiency testing and inspection on a two-year cycle.

CLIA

Clinical Laboratory Improvement Amendments

The United States standard for laboratory testing on human samples.

ISO 15189

Medical laboratory quality

The international standard for quality and competence in medical laboratories.

ISO 9001

Quality management

Quality management systems across the wider operation, not only the bench.

CE-IVD

Certified variant database

Variant classification and reporting run against a CE-marked in-vitro diagnostic database.

Sequencing and variant calling are performed by our accredited laboratory partner. Accreditation covers the laboratory work: the OnKommon interpretation engine is separately provided for research and decision-support use, and clinical authority rests with KPCIRC.

Plain-language glossary (5 terms)

Term	What it means
Custom panel	A gene panel and report designed for one specific question rather than chosen from a catalogue.
Alteration class	The kind of genetic change: point mutation, insertion or deletion, fusion, copy number, methylation.
Analytical validation	Demonstrating that the assay itself measures what it claims to, reproducibly.
Feasibility	Whether the available sample can actually support the assays the design requires.
Pre-screening	Testing a population against trial eligibility before a site visit.

Take the next step

Three ways forward. Pick the one that fits today.

Book a free consultation

A no-obligation conversation with our care team, arranged through KPCIRC.

Book a consultation

Begin Blueprint Care

Commission your decision report and a dedicated clinical team.

Explore Blueprint Care

Ask on WhatsApp, free

Signal replies to your first question within four hours, at no cost.

Start with Signal