What is the most promising and feasible proposed full treatment for HSV?

Gemini 3.1 Pro

Answer: The most promising and feasible proposed full treatment for curing Herpes Simplex Virus (HSV) remains AAV-delivered gene editing, with a specific emphasis on meganucleases to permanently excise latent viral DNA from sensory neurons.

Reviewing the responses from the other models confirms a strong consensus: a true virological cure for HSV requires eliminating the latent viral reservoir in sensory ganglia, a feat that therapeutic vaccines and standard antivirals cannot achieve. We all agree that gene editing is the only approach that has demonstrated the mechanistic ability to destroy latent viral DNA in vivo, transitioning this concept from science fiction to a feasible therapeutic goal. While my initial response grouped CRISPR-Cas9 and meganucleases together, Claude Opus 4.7 correctly highlights a crucial distinction that refines my argument: meganucleases currently possess a significant lead in preclinical efficacy. The pioneering work from Dr. Keith Jerome's lab at the Fred Hutchinson Cancer Center—specifically using dual-meganuclease combinations—has achieved unprecedented reductions of latent HSV in animal models. CRISPR approaches, while widely known, have not yet matched this specific level of in vivo success for targeting HSV latency. Furthermore, as noted across the panel, the feasibility of this approach is heavily bolstered by mature, FDA-approved Adeno-Associated Virus (AAV) delivery platforms. While engineering challenges certainly remain regarding broad delivery efficiency to all infected ganglia and minimizing off-target effects, these are measurable engineering hurdles rather than fundamental biological roadblocks. The combination of highly targeted meganucleases and proven neurotropic AAV vectors represents the most credible and advanced translational pathway toward a definitive cure.

GPT-5.5

Answer: I maintain that the most promising and feasible proposed full treatment for HSV is AAV-delivered targeted gene editing—especially meganuclease or related nuclease approaches—designed to disable or remove latent HSV DNA from sensory neurons.

I have not changed my position after reviewing the other models' answers; they largely reinforce the same central conclusion. A full HSV treatment has to solve the latency problem, because HSV persists as latent episomal DNA in sensory ganglia where current antivirals cannot reach it in a curative way. The strongest argument remains that gene editing directly targets the latent viral reservoir rather than merely suppressing outbreaks, reducing replication, or improving immune control. Therapeutic vaccines, helicase-primase inhibitors, and better antivirals may be clinically valuable and may arrive sooner, but they are more likely to be functional suppressive treatments than true cures. The other answers usefully sharpen the distinction between CRISPR and meganuclease approaches. While both fall under targeted gene editing, the most compelling published in vivo evidence appears to come from AAV-delivered engineered meganucleases, particularly work associated with Fred Hutch, showing large reductions of latent HSV DNA in animal models. That makes meganuclease-based editing arguably the leading version of the broader gene-editing cure strategy. Feasibility is still uncertain, not proven. The major barriers are safe and efficient delivery to the relevant ganglia, avoiding off-target damage, achieving enough coverage of latent virus to prevent reactivation, handling immune responses to vectors or nucleases, and demonstrating durable safety in humans. But compared with other proposed curative strategies, targeted gene editing is the clearest mechanistic match for HSV biology and has the strongest preclinical evidence for actual reservoir reduction.

Claude Opus 4.7

Answer: Gene-editing approaches using meganucleases or CRISPR/Cas9 delivered by AAV to excise latent HSV genomes from sensory neurons—anchored by Keith Jerome's Fred Hutch program—remain the most promising and feasible candidate for a full HSV cure.

All four models converged on essentially the same answer, which is itself meaningful: across independent reasoning chains, AAV-delivered gene editing emerged as the only modality that directly attacks the latent reservoir, which is the defining obstacle to a true cure. Nothing in the other responses contradicts my position; if anything, they strengthen it with complementary details I underweighted. Gemini 3.1 Pro's mention of BDgene's BD111 trial is the most substantive addition—this is a CRISPR/Cas9 therapy already in early human testing for HSV-1 herpetic stromal keratitis, with reported viral clearance and no severe adverse events. That moves gene editing for HSV from 'preclinical with a clear path' to 'already in first-in-human studies,' which reinforces feasibility rather than undermining it. Grok 4.20's reference to Excision Biotherapeutics and the analogy to approved gene-editing platforms (Casgevy for sickle cell) similarly strengthens the feasibility argument. No model proposed a serious alternative full cure. Therapeutic vaccines (Moderna mRNA-1608, GSK's candidates) and helicase-primase inhibitors (pritelivir) were uniformly characterized as suppressive or preventive rather than curative—consistent with my original framing. Since the consensus is broad and the new information (BD111 human trials, Excision's pipeline) reinforces rather than challenges my reasoning, I maintain my position without attribution to any single model.

Grok 4.20

Answer: CRISPR-based gene editing to eliminate latent HSV DNA in neurons is the most promising and feasible proposed full treatment.

My previous assessment focused on CRISPR-based gene editing (while noting meganucleases) because it directly attacks the latent viral episomes in sensory ganglia that no antiviral or vaccine can reach. The three other models reach nearly identical conclusions, citing the same Fred Hutchinson preclinical data showing 90%+ reservoir reduction in mice and guinea pigs, AAV delivery feasibility, and the mechanistic superiority of nuclease approaches over symptom-suppressing alternatives. This broad consensus reinforces rather than revises my stance. GPT-5.5, Gemini 3.1 Pro, and Claude Opus 4.7 all emphasize that only targeted nucleases have demonstrated actual clearance of latent HSV DNA in vivo; therapeutic vaccines and next-generation antivirals remain important but non-curative. The additional granularity around meganucleases and Keith Jerome's lab does not contradict my position, since my original argument already referenced those studies and the overlap between CRISPR and meganuclease platforms. Practical translational advantages further support this direction: approved AAV therapies, existing IND pathways, and active biotech efforts (Excision Biotherapeutics and others) suggest clinical trials are nearer than for more speculative modalities. Delivery efficiency, off-target risk, and dual HSV-1/HSV-2 coverage remain engineering hurdles, yet they are solvable within the current gene-editing paradigm. The alignment across independent AI analyses increases confidence that this class of therapy, with CRISPR as a core component, offers the clearest route to a genuine sterilizing cure.

Final Summary

The AI Roundtable identifies gene editing as the only strategy capable of destroying the latent viral reservoir in nerve cells. All four models highlighted the pioneering work at Fred Hutchinson Cancer Research Center as the vanguard for a definitive cure. Claude Opus 4.7 and Gemini 3.1 Pro noted that while various gene-editing tools exist, meganucleases currently show the strongest preclinical evidence for reservoir reduction.

All 4 models agreed