Evidence Paper Review

 Long Explanation Copied

Critical Review: Boye et al., Molecular Therapy (2013)

This review by the University of Florida group (Boye, Boye, Lewin, Hauswirth) synthesizes ~208 references on ocular gene therapy, spanning in-human AAV2-RPE65 trials for LCA2, preclinical MERTK, Ush1b, Stargardt, choroideremia, GUCY2D, achromatopsia, XLRS, ADRP, LHON, and anti-VEGF strategies Boye et al. 2013 Retinal Gene Therapy Review. Its strength is the inside-track synthesis of the first successful in-human retinal gene replacement trials: the 3-year Jacobson et al. follow-up reported statistically significant full-field sensitivity improvements in all 15 treated patients, ranging from tenfold to 10,000-fold, with up to 63,000-fold local sensitivity gain within the treated retinal region LCA2 3-year trial efficacy data.

Critical weaknesses. The authors hold a disclosed financial interest in AAV therapies and equity in AGTC Inc. AGTC equity conflict-of-interest disclosure — a classic sponsor-adjacent bias risk for a review predicting strong clinical translation. It is also a narrative review with no systematic search protocol, and the supplied evidence records note likely publication bias toward successful trials Documented publication-bias limitation. Historical hindsight confirms over-optimism: durable central-vision gains lagged light-sensitivity gains — a 2025 study of voretigene neparvovec found no significant BCVA or foveal sensitivity improvement in 10 patients, with hemisphere-dependent cortical responses 2025 voretigene cortical dominance study. The review also correctly flags its own unknowns — foveal detachment risk, AAV large-gene packaging truncation for MYO7A/ABCA4, and the need for intravitreal-acting vectors for macular cone diseases AAV packaging and delivery challenges.

What Would Change the Assessment

Confidence in the review's clinical claims is high for LCA2 safety (multi-trial, 15-patient, 3-year data) but moderate for generalized efficacy across the other diseases, most of which rested on small animal cohorts (e.g., n=7 per group in later Nr2e3 work). Demonstrating that mouse model results do not predict human outcomes, or that dominant-form RP silencing strategies fail, would substantially revise the review's optimistic translational framing. Blind spots: ~208 references drawn largely from the authors' own vector-development lineage, and minimal discussion of cost, manufacturing variance, or negative trial results.

 BGPT Paper Review

Study Novelty

At publication (2013), consolidating the first in-human retinal gene replacement trials with preclinical landscape was timely and moderately novel; the underlying vector biology (AAV, lentivirus) was well established.

Scientific Quality

Well-organized synthesis of 208 references by field leaders, but a narrative (non-systematic) review with disclosed AGTC equity conflicts for the senior author, no formal bias assessment, and citation density skewed toward the authors' own lineage.

Study Generality

Covers ~10 distinct retinal disease classes and five therapeutic strategies (replacement, optogenetic, antiangiogenic, neuroprotective, antioxidative), giving broad field-level scope.

Study Usefulness

Served as a key roadmap pre-Luxturna approval; quantified trial outcomes (FST gains) and vector challenges (large-gene packaging) remain historically useful reference points.

Study Reproducibility

As a review, reproducibility rests on cited primary sources; trial identifiers (NCT00481546, etc.) are supplied, but no systematic methodology or search strategy is documented.

Explanatory Depth

Provides mechanistic detail per disease (RPE65 isomerase biology, GC1 cGMP feedback, MERTK phagocytosis, ABCA4 flippase) but depth varies by section and some supplementary content was not peer-reviewable in the main text.