# Longitudinal changes in plasma biomarkers of immune activation, neuronal inflammation and injury in persons with HIV initiating ART.

- **DOI:** [10.1111/hiv.70236](https://doi.org/10.1111/hiv.70236)
- **Rating:** Sound (3.1 out of 5)
- **Verdict:** Longitudinal biomarker data enrich the HIV treatment evidence base, but the brain-health conclusion overstates a null primary NfL result.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-10-08
- **Evaluation page:** https://www.nabu.science/paper/10.1111%2Fhiv.70236

## Key claims

1. Plasma biomarkers of brain health improved following 2 years of ART.
   - Support: Tentative
   - Type: Primary finding
   - Basis: headline scope exceeds the non-significant primary NfL result and the systemic nature of most measured markers
   - Location: Conclusions
2. A decline in biomarker concentrations was observed: GFAP -14.4%, sCD14 -6.8%, neopterin -47.4%, CXCL10 -58.8%, IL-6 -29.5% (all p < 0.001) and NfL -4.4% (p = 0.075).
   - Support: Supported
   - Type: Secondary
   - Basis: paired measurements support the reported direction, with selection concerns and no untreated comparator limiting interpretation
   - Location: Abstract, Results; Results, Table 2
3. Change in GFAP, CXCL10, sCD14, neopterin and NfL was negatively associated with change in CD4 (all p ≤ 0.002) but not change in NPZ (p > 0.05).
   - Support: Supported
   - Type: Secondary
   - Basis: reported regressions support the associations, although age-sex adjustment and incomplete cognitive follow-up limit inference
   - Location: Abstract; Results, Table 3
4. A greater decline in neopterin concentration was observed with dual- (-50.2%) versus triple-ART (-44.3%; p = 0.022).
   - Support: Tentative
   - Type: Secondary
   - Basis: randomization supports comparison, but the isolated exploratory result was underpowered and uncorrected for multiple biomarker tests
   - Location: Abstract; Results, Table 4
5. Change in cognitive function was not associated with change in biomarker concentrations.
   - Support: Supported
   - Type: Secondary
   - Basis: the reported models were null, but cognitive data were incomplete and differentially available by treatment arm
   - Location: Conclusions; Results, Table 3

## Strengths

- **Randomized regimen comparison within longitudinal follow-up** The biomarker analysis follows participants for 96 weeks within a randomized dual-versus-triple ART trial, allowing a stronger regimen comparison than an uncontrolled switching cohort.
  - Dimension: methodological rigour
  - Location: Methods, Study Design and Participants; Statistical Analysis
- **Prespecified hierarchy and transparent assay methods** NfL and GFAP were designated primary biomarkers, other objectives were identified as exploratory, and laboratory procedures include plate randomization, duplicate measurements, and Simoa quality-control coefficients.
  - Dimension: methodological rigour
  - Location: Laboratory Analyses; Statistical Analysis
- **Substantive engagement with contrasting evidence** The Discussion compares the results with treatment-initiation and switching studies, including findings of increased, unchanged, and reduced inflammatory markers across different dual regimens.
  - Dimension: positioning
  - Location: Discussion, paragraph 2

## Limitations

- **Brain-health headline exceeds primary evidence** The conclusion that brain-health biomarkers improved encompasses a non-significant NfL change and several systemic immune markers, while biomarker change was not associated with cognitive change.
  - Dimension: reporting
  - Location: Abstract; Results, Tables 2–3; Conclusions
- **Selection and missingness remain insufficiently tested** Only 238 of 805 trial participants had paired samples, included participants differed on CD4, and cognitive data availability differed by randomized arm despite an MCAR assumption and no reported sensitivity analysis.
  - Dimension: methodological rigour
  - Location: Statistical Analysis; Results, Participant Characteristics; Tables S1–S2
- **Exploratory arm signal lacks multiplicity control** The neopterin difference was one exploratory comparison across multiple biomarkers, the study was not powered for biomarker arm effects, and the result was nonetheless carried into the conclusion.
  - Dimension: contribution
  - Location: Statistical Analysis; Results, Table 4; Discussion, paragraph 2; Conclusions

## Reliability checks

No concerns (4 of 4 checks).

- **Results check:** Clean. The reported numbers and statistics are consistent with the stated results.
- **Reporting and transparency:** Clean. No existing reporting or transparency concern warrants retention.
  - Conflict of interest: declared · Ethics approval: declared · Data availability: declared · Code availability: not stated
- **Reference integrity:** Clean
  - 35 references in manuscript
  - 33 of 33 checkable references found in an index
  - 2 are books, websites or datasets — counted, but not index-checkable
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
