Copy Number Analysis in Congenital Nevi: Concordance and Diagnostic Limitations of aCGH, sWGS, and Methylation Profiling

This study demonstrates that while high-resolution platforms like shallow whole-genome sequencing and methylation arrays detect more focal copy number alterations than traditional aCGH in congenital nevi nodules, their discordance with established aCGH-based criteria necessitates the continued use of aCGH as the most conservative and clinically validated method for risk stratification until platform-specific diagnostic thresholds are developed.

The Gist

The Big Picture: The "Mole Mystery"

Imagine you have a giant, birthmark-sized mole (a Congenital Melanocytic Nevus) that you've had since you were born. Sometimes, little lumps or bumps grow inside these big moles. Doctors call these Proliferative Nodules (PNs).

Here is the scary part: Sometimes, these lumps are just harmless overgrowths (like a pimple). But other times, they are actually early-stage skin cancer (melanoma).

The problem? To the naked eye and even under a microscope, these harmless lumps and the dangerous cancer lumps look almost identical. It's like trying to tell the difference between a harmless cloud and a storm cloud just by looking at a single fluffy white puff.

The Old Tool vs. The New Tools

To solve this mystery, doctors use a "genetic magnifying glass" to look at the DNA inside the cells. They are looking for Copy Number Alterations (CNAs).

Think of DNA as a library of books (chromosomes).

• Benign (Harmless) Lumps: Usually, the library just has a few extra whole books on the shelf. It's messy, but the books are intact.
• Cancerous Lumps: The library is a disaster. Pages are ripped out, chapters are missing, and random pages are glued together in weird ways. These are "segmental" changes.

For years, doctors used a tool called aCGH (array Comparative Genomic Hybridization) to check the library. It's a reliable, trusted tool, but it's a bit like using a wide-angle camera. It sees the whole books clearly, but it might miss a single ripped page.

Recently, two newer, super-high-tech tools have become popular:

1. sWGS (Shallow Whole-Genome Sequencing): Like a drone that flies over the library.
2. Methylation Profiling: Like a forensic expert who checks the ink and paper quality to find hidden clues.

The Experiment: Do the New Tools Agree?

The researchers in this paper wanted to know: If we use all three tools on the same mole, do they tell the same story?

They took 16 samples from 14 patients and ran them through all three machines. Here is what they found:

1. The "Over-Enthusiastic Detective" Problem

The old tool (aCGH) was the conservative detective. It said, "I see some whole books missing, but no torn pages. This is likely harmless."
The new tools (sWGS and Methylation) were the hyper-vigilant detectives. They said, "Wait! I see a tiny tear on page 42 of Book 16! And a smudge on page 10 of Book 8! This looks dangerous!"

The Result: The new tools found way more tiny genetic errors than the old tool. In fact, they found so many tiny errors that they often labeled harmless lumps as "cancerous" just because they were looking for things the old tool couldn't even see.

2. The "Resolution" Difference

Imagine looking at a painting.

• aCGH sees the painting from 10 feet away. It sees the big shapes and colors.
• sWGS and Methylation are looking through a microscope. They see the individual brushstrokes and tiny cracks in the canvas.

When the "painting" (the mole) was a total disaster (advanced cancer), all three tools agreed: "This is a mess!"
But when the "painting" was mostly fine with just a few small smudges, the new tools got confused. They saw tiny smudges that the old tool missed and panicked, saying, "It's a disaster!"

3. The Diagnosis Clash

Because the new tools found so many tiny errors, they changed the diagnosis in 6 out of 16 cases.

• The old tool said: "Safe."
• The new tools said: "Dangerous."

This is a huge problem. If a doctor tells a parent their child's mole is cancerous based on a tiny smudge that might not matter, the child might undergo unnecessary, scary surgery.

The Verdict: Stick to the Old Tool (For Now)

The researchers concluded that while the new tools are technically amazing and find more details, we don't have the rulebook for them yet.

The current rules for deciding if a mole is cancer were written based on the "wide-angle camera" (aCGH). If you use the "microscope" (new tools) but apply the "wide-angle" rules, you will get false alarms.

The Analogy:
It's like having a speed limit sign that says "50 mph."

• The old tool is a speedometer that only shows whole numbers (50, 51, 52).
• The new tools are digital speedometers that show decimals (50.001, 50.002).

If the law says "Speeding is anything over 50," the digital speedometer will catch you for going 50.001 mph. But maybe the law was written assuming you were using the old speedometer. Until the law is updated to account for the digital precision, the old speedometer is the safest, most reliable way to avoid getting a ticket you didn't deserve.

The Takeaway for Patients and Families

1. Don't Panic: If a new, high-tech test says your child's mole has "genetic errors," it doesn't automatically mean it's cancer. It might just mean the test is looking too closely.
2. The Gold Standard: For now, the "old school" genetic test (aCGH) is still the most trusted way to decide if a lump in a giant mole is dangerous or just a harmless bump.
3. Future Hope: Scientists need to do more research to write new rules for the high-tech tools so they can be used safely without causing unnecessary fear.

In short: The new technology is powerful, but it's currently "over-reading" the situation. Until we learn how to interpret its extra details correctly, the old, proven method remains the best guide for making life-or-death decisions.

Technical Summary

1. Problem Statement

The distinction between benign proliferative nodules (PNs) and melanoma arising within congenital melanocytic nevi (CMN) is a critical diagnostic challenge in pediatric dermatology and oncology.

• Clinical Context: Giant CMNs often contain multiple, rapidly enlarging nodules. Histopathology alone is frequently insufficient to differentiate benign PNs from malignant melanoma.
• Molecular Standard: Copy Number Alteration (CNA) analysis is the gold standard for classification. Benign PNs typically exhibit whole-chromosome gains/losses (aneuploidy), while melanomas display complex segmental aberrations.
• The Gap: Current diagnostic criteria were established using array Comparative Genomic Hybridization (aCGH). However, newer, higher-resolution platforms like shallow whole-genome sequencing (sWGS) and DNA methylation profiling (which can infer CNAs) are increasingly used. It is unknown whether applying aCGH-derived diagnostic thresholds to these higher-resolution platforms leads to over-diagnosis (false positives) or discordant classifications.

2. Methodology

The study employed a retrospective, multi-platform comparative analysis on a cohort of 14 patients (16 lesions) with atypical melanocytic proliferations in CMN.

• Cohort: 14 patients (ages 8 days to 26 years), including cases with giant CMN, medium/small CMN, and congenital blue nevi. Samples were formalin-fixed, paraffin-embedded (FFPE).
• Three Analytical Platforms:
  1. aCGH: Performed using Agilent microarrays (4x180K and 105K kits). This served as the historical reference standard.
  2. sWGS: Low-coverage whole-genome sequencing (average ~6.78x depth) on NovaSeq 6000.
  3. Methylation Profiling: Targeted methylome sequencing (EM-seq) on NovaSeq 6000, with CNAs inferred computationally (using tools like ClinCNV).
• Data Analysis:
  • CNAs were called using a tumor-only approach (assuming diploidy) with the ClinCNV algorithm.
  • Concordance: Measured using the Jaccard similarity index to quantify overlap in detected aberrations between platforms.
  • Classification: Lesions were classified as Benign (no CNAs or whole-chromosome changes only), Borderline (single segmental aberration), or Malignant (≥2 segmental/arm-level aberrations) based on established aCGH criteria.
  • Clinical Correlation: Molecular classifications were compared against clinical follow-up (disease-free vs. deceased).

3. Key Contributions

• First Systematic Comparison: This is the first study to directly compare aCGH, sWGS, and methylation-derived CNA profiles specifically within the context of proliferative nodules in congenital nevi.
• Quantification of Discordance: The study quantifies how higher-resolution platforms detect significantly more focal aberrations than aCGH, leading to potential reclassification of benign lesions as malignant.
• Diagnostic Framework Critique: It highlights the risk of applying legacy aCGH thresholds to modern high-resolution data, suggesting that current criteria may lead to "overcalling" malignancy in benign proliferative nodules.

4. Key Results

A. Detection Sensitivity and Concordance

• Total CNAs Detected:
  • aCGH: 39 CNAs.
  • sWGS: 60 CNAs.
  • Methylation: 66 CNAs.
• Concordance (Jaccard Index):
  • sWGS vs. Methylation: Highest concordance (Mean 0.67).
  • aCGH vs. sWGS: Moderate concordance (0.64).
  • aCGH vs. Methylation: Lowest concordance (0.49).
• Pattern of Discrepancy:
  • Cases with high aneuploidy (broad chromosomal changes) showed strong agreement across all platforms.
  • Cases with low CNA burden showed significant variability. Methylation profiling, in particular, detected recurrent terminal and focal deletions (e.g., del(20q11.21–qter), del(8q)) that were missed by both aCGH and sWGS.

B. Impact on Molecular Classification

Applying standard aCGH-derived criteria to the data from all three platforms resulted in divergent diagnoses for 6 out of 16 cases:

• aCGH: Classified 13/16 lesions as Benign.
• sWGS: Classified 5/16 lesions as Malignant.
• Methylation: Classified 7/16 lesions as Malignant.
• Major Discrepancies: In cases 1, 5, 6, 7a, 7b, and 13, the higher-resolution platforms detected additional segmental aberrations that shifted the diagnosis from Benign/Borderline to Malignant.

C. Clinical Correlation

• Outcome Data: Three patients died; seven remained disease-free.
• Complexity: In patients with multiple nodules (e.g., Patient 7), the sampled nodule did not always represent the precursor to the metastatic melanoma.
• Observation: Patients who remained disease-free (e.g., Cases 5 and 6) had "malignant" CNA profiles when analyzed by sWGS or methylation, but "benign" profiles by aCGH. This suggests that the additional focal aberrations detected by high-resolution methods may not always correlate with aggressive clinical behavior in this specific context.

5. Significance and Conclusion

• Platform-Specific Thresholds Needed: The study concludes that diagnostic criteria for CNAs in congenital nevi cannot be directly transferred from aCGH to sWGS or methylation arrays without adjustment. Higher resolution leads to the detection of focal/subclonal events that may be biologically insignificant or artifacts of FFPE processing.
• Risk of Over-diagnosis: Using sWGS or methylation data with aCGH thresholds risks upstaging benign proliferative nodules to melanoma, potentially leading to unnecessary aggressive treatment.
• Recommendation: Until platform-specific thresholds are validated in larger, prospective cohorts, aCGH remains the most conservative and clinically validated approach for risk stratification in proliferative nodules arising within congenital nevi.
• Future Directions: The authors call for the establishment of new diagnostic frameworks tailored to high-resolution platforms and the use of orthogonal validation methods (e.g., digital PCR, FISH) to determine the clinical relevance of focal CNAs.