DNA Methylation Profiling Predicts Post-Surgical Regrowth in SF1-lineage Nonfunctioning Pituitary Neuroendocrine Tumors.

Summary

Genome-wide methylation profiling of 117 NFPitNETs revealed five subgroups with distinct recurrence risks, especially within SF1-lineage tumors. A classifier based on 562 DMPs achieved ~97% accuracy and retained prognostic separation across external cohorts, supporting epigenetic risk stratification for postoperative surveillance.

Key Findings

• Five methylation-based clusters identified; four SF1-predominant and one TPIT/PIT1-enriched subgroup.
• Clusters k3, k4, and k5 had significantly higher recurrence risk than k1–k2; SF1 k3 showed volume expansion starting ~6 years post-op.
• A DMP-based classifier achieved ~97% accuracy and maintained prognostic separation across three external cohorts.

Clinical Implications

Methylation subgrouping could inform imaging intervals, counseling, and trial enrichment for adjuvant therapies in high-risk SF1-lineage NFPitNETs, pending prospective validation.

Limitations

• Retrospective design; potential selection biases
• Clinical utility and cost-effectiveness require prospective trials

Future Directions

Prospective validation to guide surveillance intervals and adjuvant therapy; exploration of cell-cycle and immune pathway targets highlighted by DMPs.