Low-Cellularity Lung Biopsy for EGFR Testing
A 62-year-old never-smoker with metastatic lung adenocarcinoma has a 4 mm CT-guided core biopsy of a right upper lobe mass. The FFPE block shows adenocarcinoma; the pathologist estimates tumor cellularity at 8% and notes crush artifact and anthracotic pigment. The laboratory extracts DNA from four 5-micrometer sections: yield 6.2 ng/microliter (about 310 ng total), A260/280 = 1.58, A260/230 = 1.10. The laboratory's EGFR assay is an allele-specific real-time PCR covering four hotspot codons, validated at a limit of detection of 1% variant allele fraction (VAF) with 10 ng input. A negative result is currently reported as "no mutation detected." The clinical team wants a result in three days to decide on first-line therapy.
1. Is the specimen adequate for the EGFR assay? Show the arithmetic that supports your answer.
Read the worked answer
Marginally adequate at best, and not adequate to support a confident negative.
If the tumor is 8% of nucleated cells and the mutation is heterozygous, the expected VAF in bulk DNA is about half of the tumor fraction:
- 8% cellularity x 0.5 = ~4% VAF.
The validated limit of detection is 1% VAF, so a clonal heterozygous mutation should in principle be detected. But that calculation assumes ideal DNA. Two things erode it:
- Effective amplifiable molecules: FFPE cross-linking and fragmentation, plus crush artifact, reduce amplifiable template well below the 10 ng input figure. At 4% VAF and ~3,000 haploid genome equivalents, only ~120 mutant molecules are present before fragmentation losses.
- Sampling error: an 8% estimate from a 4 mm core has wide confidence limits, and anthracotic pigment and stroma further dilute the tumor.
A260/230 of 1.10 suggests carryover of guanidinium or other extraction residues, and 260/280 ratios are unreliable for FFPE. Run an amplifiability/qPCR quality check rather than trusting the spectrophotometer.
2. What would you do before testing and reporting?
Read the worked answer
- Macrodissect: have the pathologist mark the tumor on the H&E slide and enrich to the highest achievable cellularity. Getting from 8% to 40% moves the expected VAF from 4% to 20% and converts a borderline call into a robust one.
- Optimize input: use the maximum validated input in the smallest possible elution volume, and confirm amplifiability with a size-matched qPCR control.
- Consider the platform: if enrichment cannot be achieved, a next-generation sequencing panel with a validated 1-2% VAF limit on adequate input may be more informative, but only if the amplifiable input requirement is met; a low input with a low-VAF assay is not automatically better.
- Reserve tissue: do not cut the entire block; retain material for PD-L1 and possible reflex testing.
- Communicate: tell the treating team the result may be indeterminate before the three-day clock is invoked.
3. The assay is negative after macrodissection to 35% tumor. How do you report this, and what is the safest next step?
Read the worked answer
Do not report "wild-type" or an unqualified "no mutation detected." Neither is supportable when the tested material is a small, fragmented, low-yield core from a never-smoker, a population in which EGFR mutation prevalence is high.
Suggested report language: "No EGFR mutation detected by allele-specific real-time PCR (assessed tumor cellularity 35%; input X ng; amplifiability control acceptable). The assay detects mutations at a variant allele fraction of 1% or greater; subclonal or low-level variants below this threshold cannot be excluded. Clinical correlation and consideration of repeat sampling or plasma cell-free DNA testing are advised."
Safest next step: plasma cfDNA genotyping, which is validated for EGFR in this setting and avoids a second procedure; if plasma is negative and tissue is exhausted, a repeat biopsy or a cytology cell block (pleural fluid, fine-needle aspirate) should be pursued. A negative result in a never-smoker should trigger re-evaluation of specimen adequacy, not a default to chemotherapy.