Treatment-Resistant Lyme in 2026: Where the Evidence Actually Stands

Some people treated for Lyme disease with a standard antibiotic course do not get better. Fatigue, cognitive difficulty, and pain persist for months or years. That much is not in dispute by anyone. Almost everything after it is.

The two positions

The 2020 IDSA, AAN, and ACR guidelines describe the condition as post-treatment Lyme disease syndrome: a post-infectious, likely immune-mediated process, not ongoing active infection. They advise against prolonged or repeated antibiotic courses, citing four randomised trials that failed to show durable benefit against real harms from long-term intravenous access.

ILADS, in its 2024 statement, argues that persistence of Borrelia in some patients is biologically plausible — supported by persister-cell work in vitro and in animal models — and supports individualised extended therapy. The disagreement is not primarily about ethics or empathy. It is about whether the persister findings from the bench describe what is happening in a symptomatic person years later.

Disulfiram: the correction

Disulfiram generated real enthusiasm after in vitro screens and a handful of striking case reports. A pilot randomised trial published in Frontiers in Medicine in 2025 found no significant benefit over placebo. That result deserves the same prominence the early anecdotes received. It does not prove disulfiram is useless in every context, but it removes the basis for treating it as an answer.

Dapsone combination therapy

Richard Horowitz's pulsed high-dose dapsone protocols have been published as case series in Antibiotics and Microorganisms between 2022 and 2023, reporting symptom improvement in patients who had failed prior regimens. These are uncontrolled cohorts with self-reported outcomes and no blinding. They are hypothesis-generating, and dapsone carries meaningful haematological risk — methaemoglobinaemia and haemolysis — requiring monitoring.

A case series tells you something happened. It cannot tell you the treatment is why.

What is genuinely new

The most interesting recent work is preclinical. A 2024 Science Translational Medicine screen identified azlocillin and piperacillin — beta-lactams — as outperforming doxycycline against Borrelia in mouse models, including against persister forms. Human trials have not followed yet, and mouse clearance has repeatedly failed to translate in this disease.

On prevention, Pfizer and Valneva's VLA15 vaccine reported over 70 percent efficacy in the Phase 3 VALOR trial in 2026 but missed its statistical primary endpoint, largely because case counts came in low. Regulatory filing is still planned. A vaccine would change the incidence curve; it does nothing for people already ill.

The fringe end

Whole-body hyperthermia, offered by several European and Mexican clinics, rests on the temperature sensitivity of Borrelia in culture. There are no controlled human trials, the procedure requires anaesthesia and intensive monitoring, and the cost is substantial. Methylene blue is in a similar position: plausible in vitro, unstudied clinically for this indication.

The unsatisfying summary is that in 2026 there is no evidence-backed treatment for persistent Lyme symptoms, and the most defensible care is symptom-directed: sleep, pain, autonomic dysfunction, and the deconditioning that follows years of illness. That is not a reason for despair, but it is a reason to be sceptical of anyone selling certainty at this particular moment.