Antibiotic Therapy
Early acute Lyme borreliosis is easily cured by conventional antibiotic therapy. However, some patients who have been correctly diagnosed initially as having Lyme disease, manifest various neurological and musculoskeletal symptoms several months after receiving what appeared to have been successful antibiotic therapy. Since it is unclear whether such symptoms are due to a long-term persistent infections or other causes, the term post-treatment chronic Lyme disease (PTCLD) is often used to describe this condition, so as not to imply any judgment on the actual mechanism(s) that may be involved.
New England Medical Center (NEMC) Clinical Study
A clinical study on the efficacy of antibiotic therapy for treating PTCLD was completed in late 2000. It was funded through a contract awarded to the New England Medical Center (NEMC) in Boston and involved randomized, double-blind, placebo-controlled multicenter studies to examine the safety and efficacy of ceftriaxone and doxycycline for treating patients with either seropositive or seronegative chronic Lyme disease. The clinical protocols for these studies were developed with extensive review and input from Lyme disease research experts as well as from an NIAID Lyme Disease Advisory Panel composed of patients with Lyme disease, members of patient advocate groups, practicing physicians who treat patients with Lyme disease, and basic research scientists with expertise in either infectious diseases or Lyme disease. The panel provided input on implementing the protocols selected for use in this as well as in intramural clinical studies. Since these clinical studies include an extensive research component, they represent the most comprehensive and thorough characterization of Lyme disease extant. They have provided—and will continue to generate—a wealth of new information that can be used to devise more effective therapeutic approaches.
On November 14, 2000, the Data and Safety Monitoring Board (DSMB) for the NEMC clinical trials on the efficacy of antibiotic therapy for the treatment of chronic Lyme disease reviewed a planned interim analysis of the data obtained to date. These trials involved testing the safety and efficacy of intensive antibiotic treatment in people with Lyme disease who had developed chronic symptoms, despite earlier treatments with antibiotics. The trials compared treatment with 30 days of intravenous ceftriaxone followed by 60 days of oral doxycycline to treatment with intravenous placebo followed by oral placebo for the same duration, using patients who were either seropositive or seronegative at the time of enrollment. Both trials enrolled patients with a well-documented history of prior Lyme disease and used the same regimen.
After its review, the DSMB unanimously recommended that NIAID terminate the treatment component of these studies since it had accomplished its objectives. Their preliminary analysis showed that after 90 days of continuous antibiotic therapy there were no significant differences in the percentage of patients who felt that their symptoms had improved, gotten worse, or stayed the same between the antibiotic treatment and placebo groups in either trial.
The DSMB’s review suggested there was only a slight chance that a difference between the placebo- and antibiotic-treated groups would be found even with continued accrual of another 131 patients, the number needed to reach full enrollment. Therefore, the Board recommended that the treatment component of the studies stop enrolling new patients and that those currently receiving treatment discontinue that therapy. It further recommended that the investigators continue to follow the study patients to monitor their longer-term safety and to obtain additional information that might have value in determining the underlying basis of chronic Lyme disease and in suggesting more effective therapeutic approaches. The results of the NEMC clinical trials have been published (New Eng J Med 345:85, 2001). They also indicate that patients with PTCLD do not show objective evidence of cognitive impairment, and that 90 days of continuous antibiotic therapy is not more beneficial for these patients than administering placebo (Neurology 60: 1916, 2003).
State University of New York (SUNY) Clinical Study
In another placebo-controlled study conducted at the State University of New York (SUNY) at Stony Brook, 55 patients with PTCLD were given either ceftriaxone or placebo (intravenous) for 28 days. They were then evaluated to determine whether there was significant improvement with respect to fatigue, cognitive function, and the clearance of OspA antigen that was present in the spinal fluid of only 16 percent of all enrolled patients. Ceftriaxone therapy was associated with improvement in fatigue but not with other primary outcome markers considered (Neurology 60:1923, 2003). Because fatigue, which is a non-specific symptom, was the only primary outcome measure affected and because the treatment examined was associated with adverse events, the results of the SUNY study do not support the use of additional antibiotic therapy with parenteral ceftriaxone in post-treatment, persistently fatigued PTCLD patients. We do not know whether the benefit observed for fatigue is due to antibacterial effects or some other mechanism. Additional follow-up studies on patients enrolled in both the NIAID-supported NEMC and SUNY studies revealed that 28 to 90 days of antibiotic therapy had no beneficial effect with respect to several measures of cognitive function (Neurology 60:1916, 2003).
Findings
It is noteworthy that in both the NEMC and SUNY clinical trials cited above, 40 percent and 19 percent, respectively, of patients given placebo alone reported improvement in their symptoms (placebo effect). This underscores the need to conduct placebo-controlled trials to obtain a true and accurate assessment of the beneficial effects of any given therapeutic regimen being proposed. Otherwise, one would be unable to rule out improvements attributable to the placebo effect. Also, aside from their ability to eliminate persisting infection, several antibiotics have beneficial effects that can alleviate some of the musculoskeletal and neurological symptoms generally associated with chronic Lyme disease. In addition to the well known anti-inflammatory properties of doxycycline and tetracyclines, these as well as several beta lactam and cephalosporin antibiotics—including ceftriaxone which is commonly used to treat chronic Lyme disease—have neuroprotective properties that can positively influence various neurological disorders (Trends Pharmacol Sci 25:609, 2004; Nature 433: 73, 2005; Neurobiol Dis 25: 514, 2007). In view of these considerations, it is clear that in order to draw valid conclusions on the efficacy of any proposed therapy for treating chronic Lyme disease, one must conduct carefully designed, placebo-controlled clinical trials in which sufficient numbers of patients are enrolled to permit rigorous statistical analysis.
Using another therapeutic approach, a non-NIH funded pilot study was conducted on the efficacy of gabapentin for the symptomatic treatment of chronic neuropathic pain in patients with late-stage Lyme borreliosis. Although only 10 patients were enrolled in this small study, in which gabapentin was administered orally for 1-2 years, improvement was noted in both pain quality and pain quantity (9 of 10 patients), as well as a positive effect on mood, general feeling of health, and quality of sleep (5 of 10 patients). These promising initial findings (Dermatology 211, 123, 2005) suggest that gabapentin monotherapy may be efficacious in treating pain associated with neuroborreliosis, thereby improving the quality of life for patients with late-stage Lyme borreliosis. Other studies showed that gabapentin (1.2-2.4 grams/day for 12 weeks) is safe and efficacious for the treatment of the pain and other symptoms associated with fibromyalgia, a condition with symptoms similar to those associated with chronic Lyme disease (Arthritis and Rheumatism 56: 1336-1344, 2007). On June 21, 2007, the FDA approved the use of Lyrica - an analog of gabapentin - for the treatment of fibromyalgia.
The NEMC studies mentioned above involve data collection as well as the maintenance of specimen repositories. Such specimens have been made available to other investigators working on Lyme disease and thus have contributed significantly to the development of improved and/or novel diagnostic procedures.
Animal Models
Appropriate animal models also have provided considerable information on the transmission and pathogenesis of Lyme borreliosis, as well as on the mechanisms involved in the development of protective immunity. NIAID, in collaboration with NINDS, have broadened these efforts to include comprehensive studies on non-human primate (NHP) animal models for experimental research on the neuropathology associated with chronic Lyme borreliosis (Annals Neurology 56: 361, 2004). A major goal of these studies is to optimize the Rhesus model of Lyme borreliosis as well as to determine the pathogenesis of the disease with a focus on the neurological manifestations.
The results of these studies have expanded our knowledge of those factors that contribute to the pathology associated with infection of the central nervous system (CNS) by B. burgdorferi. They have revealed the following:
- Non-human primates (NHP) are readily infected by intradermal inoculation with low number of B. burgdorferi. In immunocompetent NHPs, the infection is transient and rapidly cleared coincident with the development of high levels of IgG antibody specific for B. burgdorferi; however, in NHPs that have been immunosuppressed by treatment with corticosteroids, infection is persistent and involves the central and peripheral nervous system, as well as extra-neural target organs such as the heart, bladder, skin, and skeletal muscle (Clin. Diagnostic Lab. Immunol. 8:225,2001; Ann. Neurol. 50:330,2001; Infect. Immun. 71:7087,2003).
- The epitope specificity of IgM and IgG antibodies developed in NHPs during infection follows a predictable course. The major Borrelia-specific proteins identified by the early IgM antibody response were P39, P41, BmpA, and OspC, whereas the major Borrelia-specific proteins identified by the late IgM antibody response were P39, P41, P18, P60, P66, BmpA, and DbpA. There was no significant response in NHPs to OspA or to Arp, a 37-kDA protein that elicits an antibody response in mice infected with B. burgdorferi (Clin. Diagnostic Lab. Immunol. 9:1348,2002).
- Damage to the heart with considerable myocardial cell necrosis and subsequent fibrosis was a consistent feature of Borrelia-induced carditis in immunosuppressed – but not immunocompetent – NHPs (Lab Invest. 84: 1439,2004).
- The expression of 73 Borrelial chromosomal genes and 33 Borrelial plasmid-encoded genes was increased in the CNS of infected NHPs relative to that of other tissues (Proc. Natl Acad. Sci. 100: 1953,2003;Microbial Path. 39;27,2005).
- Cellular necrosis within the nervous system was not a feature of B. burgdorferi infection in either immunocompetent or immunosuppressed NHPs for at least 3 months after infection.
- The pattern of infection is dependent upon the strain of Borrelia used to induce infection; B. garinii induces a different type of disease in mice and NHPs than does B. burgdorferi (Ann. Neurol. 54;S49, 2004).
- In the NHP animal model of neuroborreliosis, the dorsal root ganglion is a preferred site of CNS infection, whereas the epineurium is a preferred site of peripheral nervous system (PNS) infection.
- Erp genes, which help B. burgdorferi avoid complement-mediated destruction, were detectably transcribed during NHP infection. Differences in expression levels were noted among various Erp loci. No significant differences in Erp expression were apparent between examined tissues, including CNS, PNS, skeletal muscle, bladder, skin, and heart tissues. The data strongly suggest that Erp proteins are expressed by B. burgdorferi throughout infection of their vertebrate hosts (Annal. Neurol. 54:S49,2004).
- The chemokine CXCL13 is preferentially up-regulated in NHP neuroborreliosis and is a sensitive biomarker for the presence of infection.
Inflammation of skeletal muscle is a consistent feature of Lyme borreliosis, both in humans and in experimental animal models of infection. Although several cytokines are expressed in muscle tissue, proinflammatory cytokines commonly associated with inflammation are not upregulated in Borrelia-infected muscle. However, the expression of B-lymphocyte chemoattractant (BLC), a chemokine implicated in the trafficking of B-cells to tissues, is increased in Borrelia-infected muscles of non-human primates (Cytokin 19, 297, 2002). Using protein expression profiling, it has been shown that BLC is upregulated in the spinal fluid of patients with neuroborreliosis, but not in patients with noninflammatory and various other inflammatory neurological diseases (Neulorogy 65, 448, 2005). Since the upregulation of BLC was found in every neuroborreliosis patient examined, it may be a valuable diagnostic marker for neuroborreliosis.
Other studies have shown that B. burgdorferi can be detected in mice for at least three months after treatment with therapeutic doses of various antibiotics (ceftriaxone, doxycycline, or azithromycin). These surviving spirochetes could not be transmitted to healthy mice and some lacked plasmid genes associated with infectivity. By six months, antibiotic-treated mice no longer tested positive for the presence of B. burgdorferi, and even cortisone immunosuppression failed to alter this result; that is, it failed to activate infection. Nine months after antibiotic treatment, low levels of Borrelia DNA still could be detected in some—but not all—of the mice. These findings (J Infect Dis 186: 1430, 2002) indicate that noninfectious B. burgdorferi can persist for a limited period of time after antibiotic therapy. The implications of these findings to persistent infection and the nature of chronic Lyme disease in humans remain to be assessed.
Alzheimer's Disease
Since various published reports suggested the possibility that B. burgdorferi may play a role in the etiology of Alzheimer’s disease, NIAID intramural scientists examined this issue in greater detail. The results of a recent study, using a very sensitive PCR assay capable of amplifying a Borrelia-specific DNA target sequence from all strains of B. burgdorferi sensu lato species known to cause disease in humans, provided no evidence to indicate the presence of Borrelia in the brains of deceased patients with Alzheimer’s disease (J Inf Dis 182: 1006, 2000). Although the Lyme Urinary Antigen Test (LUAT) was one of several diagnostic tests used in NIAID’s clinical studies on chronic Lyme disease, the results of independent quality control assessments of tests conducted by extramural and intramural scientists showed the LUAT to be unreliable because it yields an unacceptably high percentage of false positive reactions ( Amer J Med 110: 217, 2001). By contrast, the similar assessments confirmed a high degree of reproducibility and concordance (virtually 100 percent) for the results obtained using ELISA and Western blot assays (Amer J Med 110: 217, 2001).
Future Research
NIAID expects to sustain a high level of activity in research on Lyme disease in the coming years. NIAID investigators conducting controlled clinical studies on therapeutic approaches for treating chronic Lyme disease participated in an NINDS-sponsored conference on neuroborreliosis to facilitate the exchange and application of new knowledge as well as to foster collaborative interactions. Since the complete genome of B. burgdorferi has been sequenced by scientists at The Institute for Genomic Research (Nature 390: 580, 1997), the widespread application of this information will play a significant role in increasing our understanding of the pathogenesis of Lyme disease at the molecular and cellular levels, as well as to accelerate the development of improved diagnostic tests. Researchers already are applying microarray technology to this end.
back to top