At AAO, an OMIG-cosponsored symposium followed viruses from dormancy to disease, showing how genomics, trials and global surveillance are reshaping day-to-day care.
An early Sunday morning session in Orlando came with sunshine, strong coffee and five fast climbs up the viral learning curve. Under the guidance of chairs Dr. Bhupesh Bagga (India) and Dr. Rachel Wozniak (USA), the speakers traced a path from neuron to tear film, from bench devices to randomized trials, and from local lesions to global outbreaks.
The themes felt familiar but newly sharpened: latency is messy, sequence data now informs prognosis, the eye often flags systemic threats, trials can recalibrate practice and some “tumors” start with bugs. Equal parts lab savvy and clinic grit, this session turned molecular puzzles into practical playbooks.
Cracking HSV’s wake-up call
Dr. Todd Margolis (USA) opened with candor and comic timing, then got clinical. Epithelial dendrites, stromal keratitis driven by inflammation and anterior segment involvement are still the day job of herpes simplex virus (HSV) in the eye. The night shift is latency and reactivation, which refuse to behave.
“We know a little about latency but not a lot about reactivation,” he said, after four decades in the trenches.
Why the struggle? Latent HSV hides across many neuron types in the trigeminal ganglion, each with its own gene program and triggers. Reactivation events are rare and asynchronous, which makes them hard to catch in vivo and fickle in vitro.
Model choices matter too, like viral strain, cell line, passaging history and growth conditions that can tilt outcomes.
READ MORE: Retina Roundup at AAO 2025: Fresh data, Familiar Challenges and the Future of Eye Care
One myth was happily retired. Dendritic patterns on the cornea are not roadmaps of nerves. Citing work that infected human and pig corneal epithelium in culture, Dr. Margolis noted that “in the absence of immune cells and innervation, you get dendrites.”
Corneal epithelial resistance forces cell-to-cell spread that sketches the dendrite, a property of the tissue rather than the wiring beneath.
The take-home felt both humble and helpful: keep treating the disease in front of you while the lab keeps chasing the spark that flips HSV back on.
Genomes that forecast outcomes
If Dr. Margolis served up the problem, then Dr. Russell Van Gelder (USA) brought the power tools. Handheld nanopore sequencers and deep sequencing pipelines now turn swabs into genomes within two days, and those genomes tell stories that standard tests miss.
Case in point: a large epidemic keratoconjunctivitis (EKC) drug trial with banked specimens became a natural experiment. Sequencing adenovirus D8 from placebo eyes revealed about 660 single nucleotide polymorphisms that clustered into three clades. Same serotype, different genomes.
And the outcomes were not random. In the trial dataset, one clade correlated with more subepithelial infiltrates. A student then trained a machine model on the polymorphism pattern to predict both geography and infiltrate risk. The holdout set stunned the room.
READ MORE: Seeing Patterns, Preserving Corneas: Subspecialty Day Named Lectures at AAO 2025
“It was 16 of 16 predicting country of origin and 16 of 16 predicting who got subepithelial infiltrates (SEIs), which suggests all the information about whether someone will get an SEI is actually embedded in the virus and not so much in the host,” Dr. Margolis noted.
Story two flipped to endophthalmitis. Culture-negative cases often stayed pathogen-negative by 16S PCR, yet deep sequencing surfaced torque teno virus in many samples. Prognosis shifted with its presence.
“None of the torque teno-negative, culture-negative samples ultimately needed vitrectomy,” Dr. Margolis shared. “Every one of the torque teno-positive, culture-positive samples ultimately needed vitrectomy.”
Sequencing, in other words, can sort who needs the OR from who can ride out a tap and inject.
Windows into pandemics
Dr. Steven Yeh (USA) mapped the ocular footprints of emerging viral threats, from chikungunya and dengue to Zika, mpox and COVID-19.
“We feel the eye represents a unique window to these emerging infectious diseases,” he said, framing the cornea, retina and aqueous as surveillance sites for systemic outbreaks.
Zika led the tour. Drawing on work from Brazil, Colombia and Venezuela, he recapped congenital Zika syndrome with retinal pigment mottling, macular atrophy, optic nerve changes and even congenital glaucoma in some infants.
COVID-19 literature pointed to vascular signals in sicker patients, including cotton wool spots, hemorrhages and reduced vessel density on OCT angiography.
Mpox demanded special attention for persistence and severity. Dr. Yeh noted reports of “replicating mpox virus” persisting in tear film for up to 150 days across multiple samples.
Finally, he widened the lens to Disease X and a One Health strategy that links human, animal and environmental data. With Ebola uveitis, Rift Valley fever retinitis and Nipah case reports on the radar, Dr. Yeh’s take-home was pragmatic truth serum for clinicians and public health teams alike: eyes do not just see disease, they signal it.
READ MORE: The Effect of the COVID-19 Pandemic on Babies and Other Updates in ROP
Valacyclovir for the long game
The Zoster Eye Disease Study (ZEDS)* brought old-school rigor to a modern zoster problem.
With herpes zoster ophthalmicus (HZO) incidence rising and age at presentation sliding lower, Dr. Bennie Jeng’s (USA) team asked a simple question with big clinic impact: does one year of suppressive valacyclovir reduce new or worsening keratitis or iritis in immunocompetent adults who already had ocular disease in the prior year?
The 95-center, double-masked randomized trial enrolled 527 participants to valacyclovir 1 gram daily or placebo, followed for 12 months on treatment and 6 months off. The frequentist primary analysis at 12 months missed significance, but the 18-month analysis did not.
Bayesian modeling clarified the signal clinicians care about. “If you look overall, it seems that the potential protective effect of valacyclovir is 92% at 12 months and 97% at 18 months.”
READ MORE: Rewiring Sight with Nanowires: Vision Restored and Infrared Detected in Primate Models
Recent-onset disease and younger patients appeared to gain more. Importantly, participants stayed in the study after hitting an endpoint, and those on valacyclovir accrued fewer subsequent events at both time points.
He recommended suppressive valacyclovir 1 g daily for one year in immunocompetent, non-pregnant adults with good renal function, citing a 92% probability of benefit at 12 months and 97% at 18 months, with the strongest signal in recent-onset HZO.
When microbes spark masses
Dr. Carol Karp (USA) closed with a tour of tumors that owe their start to microbes. “About 16% of new cancers are due to infections, and this number rises to 25% in low- to middle-income countries.”
On the ocular surface, that often means ultraviolet, HPV and immunosuppression conspiring toward ocular surface squamous neoplasia (OSSN) or low-risk HPV driving papillomas in younger patients.
Her chalk-talk on HPV oncology made the room smile and nod. “Here we have p53 and RB as controlling the classroom and the kids. Then comes E6 and E7 knocking them out, and then the room goes crazy and cancer starts.”
READ MORE: Oral Lamivudine Shows Vision Gains in DME Patients in Early Clinical Trial
The clinic responses are increasingly medical. Interferon, mitomycin C and 5-FU turn big, juicy lesions into quiet epithelium, with OCT serving as an optical biopsy to track response. Off-label surprises also pop up.
In a refractory papilloma, two Gardasil injections cleared the lesion when cimetidine and topical 5-FU failed.
However, the infectious web reaches even further. Kaposi sarcoma reflects human herpesvirus-8 (HHV-8) and immune status, with high active antiretroviral therapy (HAART) and local therapies as tools.
Conjunctival extranodal marginal zone lymphoma (EMZL) can link to chronic infection and sometimes melts away with doxycycline alone, a reminder to biopsy, treat and then reassess before radiating. Dr. Karp’s punch line was to keep an infectious mindset when evaluating conjunctival tumors, because the right antimicrobial can be the right oncologic move.
*Cohen EJ, Troxel AB, Liu M, et al; ZEDS Trial Research Group. Low-Dose Valacyclovir in Herpes Zoster Ophthalmicus: The Zoster Eye Disease Randomized Clinical Trial. JAMA Ophthalmol. 2025;143(4):269-276.
Editor’s Note: The American Academy of Ophthalmology Annual Meeting 2025 (AAO 2025) is being held on 17-20 October 2025, in Orlando, Florida. Reporting for this story took place during the event. This content is intended exclusively for healthcare professionals. It is not intended for the general public. Products or therapies discussed may not be registered or approved in all jurisdictions, including Singapore.