Comparison
Biological comparison
This compares mechanisms and evidence profiles. It is not a ranking, a recommendation, or a safety assessment.
All-trans retinoic acid only
- Nuclear receptor transcription pathway
- Signalling by retinoic acid
Shared pathways
- No shared curated pathways
L-Ascorbic acid only
- Collagen biosynthesis and modifying enzymes
- Detoxification of reactive oxygen species
- Melanin biosynthesis
All-trans retinoic acid
C20H28O2 · Retinoid
All-trans retinoic acid is the receptor-active retinoid metabolite. It is a regulated drug substance in many countries; it is included here for biological context only.
Targets
- Retinoic acid receptor alpha (RARA) — activates (high confidence)
- Retinoic acid receptor beta (RARB) — activates (high confidence)
- Retinoid X receptor alpha (RXRA) — modulates (medium confidence)
L-Ascorbic acid
C6H8O6 · Antioxidant
L-ascorbic acid is vitamin C: a water-soluble reducing agent and an essential cofactor for the prolyl and lysyl hydroxylases that mature collagen.
Targets
- Prolyl 4-hydroxylase subunit alpha-1 (P4HA1) — activates (high confidence)
- Procollagen-lysine 5-dioxygenase (PLOD1) — activates (high confidence)
- Tyrosinase (TYR) — inhibits (low confidence)
- Sodium-dependent vitamin C transporter 2 (SLC23A2) — targets (medium confidence)
Evidence engine
Strong
Multiple independent human studies point in the same direction.
Skinceptor Evidence Index
78
experimental · of 100
This index is a research-navigation aid describing how much evidence exists and how it was produced. It is not a medical safety score, an efficacy guarantee, or a clinical recommendation.
Study composition
- Human1
- Randomised controlled trial2
- Review2
- Total studies
- 5
- Human studies
- 3
- Largest sample
- 297
- Range
- 1988–2015
Index breakdown
- Study design quality68% × 28%
Mean design weight across curated studies (meta-analysis 1.0, RCT 0.9, other human 0.7, review 0.45, animal 0.35, in-vitro 0.25).
- Human evidence share100% × 22%
Proportion of curated evidence measured in people rather than cells or animals.
- Evidence volume82% × 14%
Log-scaled count of curated studies, so volume alone cannot dominate the score.
- Largest sample size82% × 12%
Log-scaled largest reported sample size across curated human studies.
- Replication100% × 12%
Independent human studies, capped at three; one study is never treated as replicated.
- Recency45% × 7%
Linear decay over a twenty-year window from the most recent curated publication.
- Source traceability20% × 5%
Share of curated studies with a verified PMID or DOI in this dataset.
Insufficient comparable evidence for quantitative synthesis.
Evidence engine
Moderate
Some human evidence exists, but replication or sample sizes are limited.
Skinceptor Evidence Index
58
experimental · of 100
This index is a research-navigation aid describing how much evidence exists and how it was produced. It is not a medical safety score, an efficacy guarantee, or a clinical recommendation.
Study composition
- Human1
- Randomised controlled trial1
- In-vitro3
- Review2
- Total studies
- 7
- Human studies
- 2
- Largest sample
- 19
- Range
- 1991–2020
Index breakdown
- Study design quality46% × 28%
Mean design weight across curated studies (meta-analysis 1.0, RCT 0.9, other human 0.7, review 0.45, animal 0.35, in-vitro 0.25).
- Human evidence share57% × 22%
Proportion of curated evidence measured in people rather than cells or animals.
- Evidence volume95% × 14%
Log-scaled count of curated studies, so volume alone cannot dominate the score.
- Largest sample size43% × 12%
Log-scaled largest reported sample size across curated human studies.
- Replication67% × 12%
Independent human studies, capped at three; one study is never treated as replicated.
- Recency70% × 7%
Linear decay over a twenty-year window from the most recent curated publication.
- Source traceability14% × 5%
Share of curated studies with a verified PMID or DOI in this dataset.
Insufficient comparable evidence for quantitative synthesis.