3.6 Diagnostic Testing Strategies

Diagnostic testing in the Domain 3B frame

Diagnostic genetic testing asks: Does this person have a genetic explanation for their clinical findings? Contrast that with screening (risk stratification in asymptomatic populations), predictive/pre-symptomatic testing (future disease risk in currently healthy people), carrier testing (reproductive risk), somatic tumor profiling (tumor-only alterations), and research testing (protocol-driven discovery). On the CGC exam, Domain 3B items often hinge on recognizing which of these goals the stem is describing before you pick a method or counseling stance.

Diagnostic testing is indicated when there is an existing phenotype, abnormal clinical or laboratory finding, or a confirmed clinical diagnosis that needs molecular confirmation for management, prognosis, or family counseling. A positive diagnostic result can reclassify a syndrome, open targeted therapies or surveillance, and create a family-specific variant for cascade testing.

Who to test first

Counseling pearl: Testing an unaffected person first when no familial variant is known is usually low-yield and can produce VUS that are hard to interpret. Prefer an informative affected relative whenever feasible.

Phenotype-driven vs genotype-first strategies

Phenotype-driven (classic diagnostic pathway)

Start with the differential diagnosis built from history, exam, imaging, and family history, then order the narrowest high-yield test:

1. Single-gene test when one gene explains the presentation (for example, CFTR for classic cystic fibrosis; FBN1 for highly specific Marfan features meeting clinical criteria).
2. Multigene panel when the differential includes several phenotypically overlapping genes (hereditary hearing loss panels; cardiomyopathy panels; epilepsy panels).
3. Chromosomal microarray (CMA) for developmental delay/ID, autism with additional features, or multiple congenital anomalies when copy-number variants are high on the differential.
4. Karyotype / FISH when aneuploidy, large rearrangements, or mosaicism visible at cytogenetic resolution is suspected (for example, Turner syndrome workup, balanced translocation evaluation in recurrent pregnancy loss couples after appropriate indication).
5. Exome or genome sequencing when the phenotype is genetically heterogeneous, prior targeted testing is negative, or a Mendelian diagnosis is still strongly suspected.

Genotype-first (broad sequencing early)

A genotype-first approach orders exome/genome (sometimes with concurrent CMA) early—especially in critically ill neonates, nonspecific neurodevelopmental phenotypes, or when sequential testing would delay care. Advantages include higher cumulative diagnostic yield and shorter time-to-diagnosis in selected cohorts. Trade-offs you must counsel:

Exam trap: Genotype-first is not “always better.” Stems that describe a classic, single-gene phenotype still favor targeted testing first.

Cascade testing

Cascade testing is systematic testing of relatives after a known pathogenic/likely pathogenic familial variant is identified. It is usually site-specific (familial variant analysis), cheaper and clearer than repeating a large panel, and directly answers whether relatives inherited the family’s disease-causing change.

Cascade workflow counselors should know

1. Confirm the proband’s variant classification and clinical correlation.
2. Provide written results and a family letter describing the variant, inheritance, and testing options.
3. Offer testing first to first-degree relatives at highest risk, then extend outward (true cascade).
4. For autosomal dominant cancer predisposition, a negative familial-variant test generally returns that relative to near-population risk for that gene’s cancer spectrum (with caveats for family history from other causes).
5. Document who was informed, who declined, and any barriers (geography, insurance, psychosocial readiness).

Duty-to-warn / family communication themes appear more fully in Domain 5, but Domain 3B expects you to know that cascade testing depends on relatives learning about the familial variant—and that genetic counselors facilitate, not coerce, that communication.

Choosing modality by clinical question

Method–indication mismatch is a high-yield trap: ordering NGS panels when a repeat expansion or methylation test is required will be “negative” for the wrong reason.

Timing, urgency, and parallel testing

• Urgent / inpatient neonatal ICU: Broad sequencing ± CMA may run in parallel with metabolic labs; turnaround and sample type (blood, buccal if transfusion issues) matter.
• Outpatient diagnostic odyssey: Stepwise testing may be appropriate if insurance requires, phenotype is evolving, or a single high-yield test remains untried.
• Prenatal diagnostic testing (CVS/amnio-based) is Domain 3C territory, but the same who/what logic applies: define the familial variant or phenotypic target before choosing the assay.

Integrating pretest counseling into diagnostic strategy

Even when the indication is diagnostic, counsel on:

• Expected yield for the phenotype and test
• Possibility of VUS, negative, or unexpected positive results
• Secondary findings if exome/genome is ordered
• Implications for relatives and reproductive planning
• Limits of a negative test (residual clinical diagnosis; genes not covered; mosaicism; deep intronic / structural variants)

Worked scenario A — phenotype-driven

A 4-year-old with hypotonia, developmental delay, and a clinical suspicion for Prader–Willi syndrome. Best first diagnostic strategy: methylation analysis for 15q11.2-q13 (with follow-up studies to distinguish deletion, UPD, or imprinting defect)—not a random “neurology NGS panel” as the sole first step.

Worked scenario B — cascade

A woman with ovarian cancer has a pathogenic BRCA1 variant. Her unaffected sister seeks testing. Best test: familial BRCA1 variant analysis, not a full hereditary cancer panel as the first step (panel may be considered later if family history suggests additional differentials and the familial variant is negative—or concurrently only when justified).

Worked scenario C — genotype-first

A neonate with unexplained encephalopathy and multiorgan involvement has a nondiagnostic metabolic workup. Rapid exome/genome with parental samples is appropriate; counsel secondary findings and VUS before ordering.

Exam scenarios and traps

• Trap — testing the healthy consultand first when an affected relative is available and no familial variant is known.
• Trap — equating “negative panel” with “not genetic.” Residual risk remains; consider missed variant types, locus heterogeneity, and non-Mendelian causes.
• Trap — using screening language for diagnostic testing. Diagnostic testing does not require Wilson–Jungner population criteria.
• Trap — repeating a large panel in every relative instead of cascade testing for a known P/LP variant.
• Scenario: Adult with unexplained ataxia—ask whether a repeat-expansion disorder is on the differential before relying on standard NGS alone.

Practice checkpoints

1. Define diagnostic vs screening vs predictive indications in one sentence each.
2. State who to test first in cancer, pediatric, and deceased-proband scenarios.
3. Contrast phenotype-driven escalation with genotype-first trade-offs.
4. Explain why cascade testing requires a known familial P/LP variant.