Review papers with raw data transparency

Concise critical appraisal (visual-first then bullets)

• Strengths: largest collated CDH1 cohort at time of publication; rigorous penetrance modelling using MENDEL and many families; inclusion of panel sequencing expands the genetic view beyond CDH1 and identified CTNNA1 as a bona fide HDGC gene in at least two families — this supports a pathway‑centric genetic definition of HDGC rather than clinical‑only criteria (Hansford et al. (JAMA Oncol 2015)).
• Limitations / biases:
  • Ascertainment bias — many families are multicase and enriched for penetrant disease; penetrance estimates likely overestimate risks for unselected carriers (authors state this explicitly) (Best Practice & Research Clin Gastroenterol review 2014).
  • Genetic assay limits: targeted amplicon panels and MLPA miss some structural/CNV and regulatory/allele-specific expression defects; lack of RNA and tumor material limited follow-up for many candidate variants (authors note inability to validate somatic second‑hits broadly).
  • Panel positives require careful clinical interpretation — several identified variants (eg MSR1 R293X, SDHB S163P, STK11 F354L) have contested penetrance or syndrome overlap; family co-segregation and functional evidence remain incomplete for many calls.
• Clinical implication: penetrance estimates support strong counseling and consideration of risk‑reducing strategies (surveillance vs prophylactic gastrectomy) for CDH1 carriers; importantly, HDGC phenotype may reflect mutations in non‑CDH1 genes and hence management should be gene‑specific (e.g., BRCA2 pathways vs CDH1 pathway) (Hansford et al. (JAMA Oncol 2015)).

What would change the conclusions?

• Large population‑based ascertainment of CDH1 carriers (not enriched families) showing substantially lower penetrance would reduce clinical urgency for prophylactic surgery.
• Demonstration that many panel-identified variants do not co-segregate with disease nor show second‑hit/tumor evidence would weaken the claim that these genes explain a sizable fraction of CDH1‑negative HDGC.

Reproducibility and what is needed to improve it

1. Share de‑identified pedigree-level data and variant lists (with allele frequencies and segregation) to enable independent penetrance re‑modelling.
2. Provide raw panel sequencing calls, coverage and validation files; expand to include CNV and RNA/ASE analyses to capture regulatory/CDH1 locus events.
3. Functional assays and tumor somatic follow-up (IHC, LOH, methylation, second‑hit sequencing) for candidate non‑CDH1 genes to demonstrate pathogenic mechanism.