Ocular Microbiology and Immunology Group
Back to OMIG Main Page

< Previous | 2026 Agenda and Abstracts | Next >

2026 OMIG Abstract

Tokyovirus in Acanthamoeba Keratitis

Gerami D. Seitzman^1,2, Kelly Ngo¹, Danny Yu², Thomas Abraham², YuHeng Liu¹, Armin Hinterwirth¹,
and Thuy Doan^1,2, on behalf of the capriCORN Study Group

¹Francis I. Proctor Foundation, University of California, San Francisco, California; ²Department of Ophthalmology, University of California, San Francisco, California

Purpose: Tokyovirus, a giant virus of the family Marseilleviridae, is known to infect Acanthamoeba, though its clinical relevance to Acanthamoeba keratitis (AK) is unknown. Using metagenomic deep sequencing (MDS), we identified Tokyovirus in AK cases and characterized the associated clinical features and outcomes.

Methods: Corneal swabs from patients enrolled in capriCORN (comprehensive Analysis of Pathogens, Resistomes, and Inflammatory-markers in the CORNea), an international multi-center keratitis consortium, were analyzed for MDS. MDS identified both Acanthamoeba and Tokyovirus A1 in a subset of corneal specimens. We restricted our analysis to corneal specimens positive for Acanthamoeba and compared baseline demographics, clinical features, and visual acuity outcomes between cases with Tokyovirus co-detection and those with Acanthamoeba alone.

Results: 53 corneal samples were MDS-positive for Acanthamoeba, of which 8 (15%) were co-infected with Tokyovirus. Acanthamoeba-only cases and those co-infected with Tokyovirus did not differ significantly in contact lens wear (76% vs. 88%), sex, age (40 vs. 31 years), days of symptoms (21 vs. 15), central corneal involvement (89% vs. 100%), or baseline visual acuity (median logMAR 1.09 vs. 0.75; all P>0.05). No patient with Tokyovirus presented with hypopyon (0/8), compared with 25% (11/44) of Acanthamoeba-only patients. Tokyovirus cases, however, showed a trend toward greater improvement in visual acuity from baseline to follow-up (median Δ logMAR: -0.35 vs. 0.00, P=0.059).

Conclusions: In this preliminary series, Tokyovirus co-detection in Acanthamoeba keratitis was not associated with differences in presenting risk factors or baseline severity, but it trended toward an absence of hypopyon and greater visual improvement. Given the small number of co-detected cases, these findings should be considered as hypothesis-generating rather than confirmatory. Tokyovirus is known to exert cytopathic, growth-limiting effects on Acanthamoeba in vitro. Whether viral parasitism attenuates the amoeba’s pathogenicity in vivo is an interesting hypothesis raised by these descriptive findings.