Kessel 2025 — sNfL in DOA: boundary-defining negative for B-Re-4

One-paragraph summary

Case-control study comparing serum NfL (Simoa-quantified) in 22 OPA1-genetic-confirmed dominant optic atrophy (DOA) patients vs 22 matched controls. The headline result is negative: sNfL trended higher in DOA (median 7.39 IQR 5.25-11.26) than controls (median 5.86 IQR 4.50-9.88) but was not significantly different (p = 0.405). sNfL did not correlate with peripapillary RNFL thickness, visual acuity, or disease duration within DOA patients. Age was a confounding correlate of sNfL in both groups. This is a boundary-defining negative anchor for the project's B-Re-4 bridge: hereditary RGC axon loss via OPA1 mitochondrial dysfunction does not produce serum NfL elevation detectable at this sample size. The implication is that the B-Re-4 sNfL↔RNFL coupling probably operates in sporadic immune-mediated injury (MS, FM, SFN) but NOT in chronic hereditary axon attrition — narrowing the bridge's mechanistic scope.

Claims as triples

serum_NfL_level — fails_to_differ_in → dominant_optic_atrophy [evidence: between-group p = 0.405; confidence: emerging — n=22 per arm limits power]
serum_NfL_level — fails_to_correlate_with → OCT_RNFL_thickness [evidence: within-DOA correlation analysis; confidence: emerging — single-cohort, hereditary substrate only]
serum_NfL_level — fails_to_correlate_with → visual_acuity [evidence: within-DOA correlation; confidence: emerging]
serum_NfL_level — fails_to_correlate_with → disease_duration [evidence: within-DOA correlation; confidence: emerging]
serum_NfL_level — correlates_with → age [evidence: within-group correlation, both arms; confidence: established — replicates known age-confounding finding]

Methods note

Case-control design; OPA1-genetic-confirmed DOA patients (n=22) vs age-matched controls (n=22). sNfL via Simoa SR-X analyzer (gold-standard ultra-sensitive immunoassay). OCT for peripapillary RNFL thickness. Statistical approach: between-group comparison (Mann-Whitney implied by reported medians and IQR), within-group Spearman correlation. The Simoa platform is the same used in MS sNfL studies; sensitivity is not the limitation.

Limitations

Small sample (n=22 per arm) — power is limited; a true effect of 30% NfL elevation in DOA might be missed at this n
Single time point — chronic axon attrition in hereditary disease may have plateaued; sNfL kinetics may differ from acute/subacute injury
DOA disease duration not stratified — early vs late DOA may behave differently
No comparison to MS or other immune-mediated optic-nerve injury at the within-paper level (would need to be done as cross-study comparison)

Open questions raised

Q-Re-19 (new): Does sNfL elevation require an active/ongoing inflammatory or immune-mediated injury process, vs being insensitive to chronic structural axon attrition? If so, FM patients with active autoimmune-subtype phenotype (anti-SGC IgG+, anti-GFAP IgG+) should show sNfL elevation, while burned-out central-only FM may not.
Q-Re-20 (new): What is the within-FM correlation of sNfL with RNFL? Cano-Cano 2026 shows the panel is feasible; Fundaun 2026 shows sNfL is elevated in FM at SFN-comparable levels; Garcia-Martin 2016 shows RNFL is thinned in FM — but the within-FM coupling of the two has not been directly tested. This paper's negative DOA finding makes the within-FM question more interesting, not less: if FM shows the coupling that DOA lacks, the FM phenotype is mechanistically distinct from chronic hereditary axon attrition.

Triangulation notes

Boundary-defines B-Re-4 (Serum NfL ↔ retinal RNFL thinning): forces the bridge to be reframed. The coupling operates in sporadic immune-mediated injury but probably NOT in hereditary mitochondrial axon attrition. This is mechanistically interpretable — sNfL is released from acutely injured axons; chronic gradual axon loss may not produce detectable serum spillover.
Strengthens FM specificity of the sNfL signal: Fundaun 2026 found sNfL elevated in FM at SFN-comparable levels. If FM sNfL elevation reflects active immune-mediated injury (per the boundary this paper draws), then FM sNfL is biomarking the autoimmune-subtype mechanism, not just generic neural-tissue attrition.
Implications for retinal-biomarker priority cohort design: include both anti-SGC-IgG+ FM and anti-SGC-IgG- FM in the biomarker-mapping cohort; predict sNfL elevation tracks IgG status; predict RNFL thinning tracks IgG status. The DOA-negative anchor sharpens this prediction by ruling out one alternative (chronic structural attrition explanation).
Reinforces Bandinelli 2025 + PASC-SFN framing: PASC-SFN is acute immune-mediated; FM may share that acute/recent-immune-onset character. DOA-negative shows the sNfL signal is not just "any axonal injury" — it's selectively responsive to immune-mediated injury.

Bridges

B-Re-4 boundary added: the bridge now includes a negative anchor at the hereditary-axon-attrition endpoint, sharpening its mechanistic claim. Update synthesis/bridges.md to note this boundary explicitly so the bridge framing doesn't over-claim.
No new bridges.

Confidence-tier framing

The negative finding itself: emerging tier (n=22/arm needs replication; could be power-limited)
The implication that B-Re-4 is mechanism-restricted to immune-mediated injury: inferred tier (one interpretation of the negative; alternative explanations including power limitation are also viable)

← All papers