There's something almost perverse about the idea. Parasites—organisms we've spent millennia trying to eliminate—might actually hold blueprints for treating some of medicine's most intractable conditions.
That's the wager being made by Ditto Biosciences, a three-person startup that emerged from Y Combinator's Winter 2026 batch with a pitch that sounds equal parts science fiction and evolutionary logic. The San Francisco company, founded just last year, is sifting through the molecular arsenals of parasites, ticks, and viruses—organisms that have spent eons perfecting the art of hijacking human immunity without destroying their meal ticket.
Ditto has already analyzed over 1 million proteins from these unlikely sources. Thousands, the company claims, appear promising: computational models predict they'll bind to clinically validated human targets, with early experimental data showing affinity in the 1-2 nanomolar range. If that sounds audacious, there's precedent. Exenatide, a diabetes drug, came from Gila monster venom. Captopril, the blood pressure medication, was inspired by peptides from Brazilian vipers. But applying this nature-inspired approach to autoimmune disease? That's aiming considerably higher.
A Market Hungry for Something Different
The numbers tell a story of both enormous opportunity and persistent inadequacy. Estimates for the autoimmune therapeutics market vary wildly depending on who's counting and what they're including. SNS Insider pegs it at $72.3 billion in 2023, climbing to $116.8 billion by 2032. Zion Market Research forecasts an even steeper trajectory: $249.71 billion by 2034.
Yet for all that projected growth, the treatment playbook has barely evolved. The field remains dominated by broad immunosuppression—sledgehammers where scalpels are needed. AbbVie's Humira, despite losing patent exclusivity and facing a swarm of biosimilars, still looms large. The company has pivoted toward newer assets like Skyrizi and Rinvoq (the latter protected by patent settlements through April 2037), but the fundamental problem hasn't changed. These drugs work by dampening the entire immune system rather than correcting whatever specific malfunction is driving disease.
That blunt-force strategy is showing its age, and not gracefully. In 2021, the FDA slapped boxed warnings on JAK inhibitors—tofacitinib, baricitinib, upadacitinib—citing risks of major cardiovascular events, cancer, blood clots, and death. Europe's drug regulator followed with restrictions for high-risk patients. Meanwhile, formulary managers are getting aggressive: CVS Caremark booted Humira from major commercial formularies on April 1, 2024, accelerating the biosimilar shift.
Payers and providers aren't just looking for cheaper versions of the same drugs. They want different mechanisms. Better safety profiles. Something that doesn't require trading one set of problems for another.
When Biology Meets Silicon
Three trends, converging roughly in parallel, have created space for Ditto's unconventional thesis.
First: the science of host-pathogen interactions has matured in ways that would've seemed fantastical two decades ago. Researchers have characterized helminth secreted proteins—ES-62, hookworm AIP-2, the recently described Trichuris p43—that demonstrate robust anti-inflammatory effects in preclinical models of asthma, arthritis, colitis. Even more telling, tick-derived proteins have advanced into late-stage human trials. Nomacopan, developed by Akari Therapeutics from the saliva of Ornithodoros moubata ticks, targets both complement C5 and leukotriene B4. It's made it through Phase 2/3 trials for complement-mediated diseases. A January 2026 paper in ACS Accounts detailed the engineering of tick salivary evasins as multi-chemokine inhibitors, validating that parasite proteins offer what researchers call "privileged scaffolds" for hitting complex inflammatory networks.
Translation: these proteins aren't just biochemical curiosities. They're drug scaffolds that evolution has already debugged.
Second: AI has collapsed timelines that used to span years into months, or weeks. AlphaFold and protein language models like ESM-2 enable structure prediction and design-build-test-learn cycles at speeds unthinkable even half a decade ago. The 2024 Nobel Prize to Jumper, Hassabis, and Baker recognized this shift. Ditto's platform leverages these tools to screen millions of parasite proteins and engineer them for therapeutic use—computational heavy lifting that would've been prohibitively expensive in an earlier era.
Third, perhaps less obvious but no less important: the policy winds are shifting. The NIH released its first agency-wide Strategic Plan for Autoimmune Disease Research covering fiscal years 2026-2030, prioritizing mechanisms, data standards, cross-disease approaches. In February 2026, the NIH Office of Autoimmune Disease Research elevated autoimmune disease as a "Highlighted Topic" for grants. That's a tailwind for mechanistically novel platforms, especially those targeting tolerance and precision immunomodulation rather than broad suppression. Maybe not a guarantee of success, but an environment more receptive to moonshots.
The Team and the Bet
Ditto's founding team brings the kind of credentials that look good on pitch decks. Adair Borges holds a PhD from UCSF and was a Miller Fellow at UC Berkeley, with publications in Cell on anti-CRISPR biology and patents on anti-CRISPR inhibitor technology. Dennis Sun comes from Harvard and UC Berkeley with product and partnerships experience. Emily Weiss earned a PhD in computational and molecular biology, with stints at Illumina and DuPont. They incorporated in California on September 23, 2025, and launched publicly via Y Combinator about 18 days into February 2026.
Whether that pedigree translates to therapeutic success is another question entirely.
Their pitch centers on a data moat and a de-risking strategy designed to address the field's translational landmines. Early clinical trials with Trichuris suis ova—pig whipworm eggs, for those keeping score at home—showed safety but mixed efficacy across inflammatory bowel disease indications. It was a proof-of-concept that whole-organism approaches face serious hurdles. Ditto's bet is that defined, engineered proteins will translate better. Clean molecules, predictable pharmacology, none of the biological messiness of living parasites.
The company is building what it calls a tissue biobank to map human immune memory against parasite proteins. The goal: predict and mitigate immunogenicity before hitting clinical trials. Smart in theory. Immunogenicity remains the core development risk for non-human proteins—FDA guidance requires risk-based assessments, and computational de-immunization techniques (epitope prediction, MHC-II modeling, TCR-epitope deep learning predictors) are advancing but hardly foolproof. Nomacopan's clinical progress suggests the path is navigable. But it's a narrow path with drop-offs on either side.
The competitive landscape is getting crowded with mechanistically distinct approaches. Anokion is advancing KAN-101, a liver-targeted tolerance platform for celiac disease, through Phase 2 with FDA Fast Track designation. COUR Pharmaceuticals has INDs across primary biliary cholangitis, type 1 diabetes, myasthenia gravis, and celiac disease using tolerizing nanoparticles, with new mechanistic data published in Science Advances in January 2026. Engineered regulatory T-cell therapies are progressing, too: Sonoma Bio reported positive interim Phase 1 data in refractory rheumatoid arthritis in October 2025. Quell is pushing its preclinical CAR-Treg QEL-005 toward a 2026 Clinical Trial Application.
Ditto positions itself within a lineage of nature-to-drug successes—aspirin from willow bark, penicillin from mold, botulinum toxin, GLP-1 agonists—but applied specifically to the immune system's regulatory machinery. Fair enough. The question is whether evolutionary validation survives the gauntlet of AI-accelerated engineering, preclinical models, and human physiology.
The 18-Month Clock
The next year and a half will reveal whether Ditto's thesis can survive contact with experimental reality. Not in the business-speak sense of "de-risking" or "validating the platform," but in the grittier, more consequential sense of whether these molecules actually do what they're supposed to do.
The company needs to demonstrate that its computationally predicted binders modulate immune function in relevant assays. That engineered proteins retain activity while shedding immunogenic epitopes. That its tissue biobank approach accurately predicts clinical immunogenicity risk. These aren't trivial boxes to check.
Funding should follow scientific milestones—though "should" is doing heavy lifting there. Seed-stage biotech investors increasingly favor platforms with built-in diversification, multiple shots on goal from a single discovery engine. Ditto's million-plus protein library fits that model. But protein therapeutics development is capital-intensive, and the company will likely need compelling preclinical data to attract a meaningful Series A. Compelling, in this context, means more than "interesting." It means data that convince investors the platform can generate clinical candidates faster and cheaper than traditional approaches.
The regulatory pathway offers both challenges and opportunities. If Ditto can package its parasite-derived proteins as novel biologics with differentiated mechanisms and favorable safety profiles, the FDA's recent emphasis on precision medicine and antigen-specific tolerance could work in the company's favor. The JAK inhibitor safety concerns have created space for alternatives. PBM formulary dynamics—which drove rapid biosimilar uptake after CVS Caremark's April 2024 Humira exclusion—suggest payers will shift aggressively if safety or cost advantages are clear.
Broader scientific trends support the approach, at least in principle. Academic labs continue publishing mechanistic insights into helminth secretomes and host-parasite interactomes. Reviews in Trends in Parasitology and Frontiers in Parasitology highlight the potential of "big data" mining of these systems. The January 2026 ACS Accounts paper on engineered evasins suggests the field is accelerating. Expect more biomimetic parasite protein scaffolds entering preclinical pipelines across inflammation-driven diseases beyond autoimmunity.
The market opportunity remains enormous, though prevalence statistics vary wildly. The NIH Office of Autoimmune Disease Research cites a range of 23.5 to 50 million Americans affected. Johns Hopkins references around 3 percent of the population—roughly 10 million. What's indisputable is that current therapies leave substantial unmet need, and patients are still cycling through drugs that don't work well enough or carry unacceptable risks.
Evolution Versus the Clock
For investors and strategists tracking this space, Ditto represents a test case for whether AI-driven protein engineering can unlock previously undruggable biology. The company's success or failure won't hinge on scientific creativity—parasites genuinely are masters of immune manipulation, honed over evolutionary timescales that make human drug development look like a weekend project. The question is execution. Can they engineer proteins fast enough? De-immunize effectively enough? Generate data compelling enough to survive the valley of death that kills most platform biotechs?
The broader industry is watching, if not holding its breath. If Ditto or similar approaches demonstrate clinical proof of concept, it could open a new source class for biologics—not monoclonal antibodies or engineered cytokines, but evolution-tested immune modulators refined through computational design.
That's the promise, anyway. The parasites have had millions of years to get it right. Ditto has considerably less time.
