IV or Not IV? Just One of the Antibiotic Questions in Whipple's Disease
Article Outline
- Should all patients with Whipple's Disease be treated with an initial 2-week course of parenteral ceftriaxone or meropenem?
- Why did the clinical response rate in this trial exceed prior observations?
- Can the Study Design of Feurle et al3 Be Strengthened in Future Studies of the Therapy of Whipple's Disease?
- References
- Copyright
See “Efficacy of ceftriaxone or meropenem as initial therapies in Whipple's disease,” by Feurle GE, Junga NS, Marth T, on page 478.
In 1907, Dr. George H. Whipple, an Instructor of Pathology at The Johns Hopkins University, provided the initial description of the disease he named “intestinal lipodystrophy.”1 The patient, a 36-year-old male physician, presented 5 1/2 years into an insidious illness characterized by recurrent bouts of arthralgia and arthritis; some episodes lasting only 6–8 hours. Subsequent weight loss, cough, and low-grade fevers with occasional night sweats followed with diarrhea first occurring nearly 7 years into the patient's disease course. Beyond signs of arthritis, fatty stools owing to malabsorption and peripheral eosinophilia, the patient's physical examination was normal until the final weeks of his life when an abdominal mass was appreciated, leading to an exploratory laparotomy that defined the mass as enlarged mesenteric glands. The patient died abruptly 2 days postoperatively in respiratory distress with signs of acidosis.
The paper presents a tour-de-force of pathologic analysis leading the renowned internist Dr. William Thayer to comment, “with the remarkable observations at autopsy, it is difficult to resist the conclusion that we are here dealing with a definite and hitherto unrecognized clinical picture with which we shall meet again.” Most notably, Dr. Whipple comments in his paper on the presence of large numbers of “peculiar,” “pink-staining, frothy-looking” or “foamy” mononuclear cells in many tissues (but particularly in the stroma of the small intestinal villi and mesenteric glands) and notes finding great numbers of “rod-shaped organisms” within these cells. An inoculated rabbit died 7 weeks later with pathology demonstrating similar infiltrating cells.
Dr Whipple provided in this seminal paper a remarkably complete description of the illness now bearing his name, except that he did not describe central nervous system (CNS) involvement because this was not part of the autopsy he performed on the patient. Subsequent observations both enhanced the clinical descriptions of Whipple's disease with the important addition of CNS extension to the disease spectrum and elucidated that antibiotics (focusing mostly on doxycycline and trimethoprim-sulfamethoxazole [TMP-SMX]) could be life saving. However, progress in understanding Whipple's disease truly began to accelerate when, in the early 1990s, we came to understand that the rod-shaped intracellular bacteria were Tropheryma whipplei, the organism was cultured, the genome of T whipplei was sequenced and molecular diagnostics began to be applied to populations with and without Whipple's disease.2
As reported in this issue of Gastroenterology, Feurle et al,3 with the support of the European Commission began in January 1999, to conduct the first open-label, prospective, randomized trial, enrolling patients accrued throughout central Europe, to evaluate optimal therapy for Whipple's disease. The authors opted to compare a 2-week intravenous course of ceftriaxone or meropenem followed by 12 months of oral TMP-SMX. All patients were to undergo a lumbar puncture for cerebrospinal fluid (CSF) polymerase chain reaction (PCR) for T whipplei at study entry; the initial clinical follow-up occurred at 3 months with serial small bowel mucosal biopsies beginning 6 months after initiation of treatment. The study ended in July 2006, when the 42nd patient completed 36 months of follow-up. In 2 patients, the diagnosis of Whipple's disease was not confirmed. Thus, the trial contained 40 evaluable patients (20 per study arm): 38 presented with small intestinal involvement suggesting classic Whipple's disease, 1 with diagnostic mesenteric lymph nodes but a negative duodenal biopsy, and the remaining patient with endocarditis, a case of localized Whipple's disease. Three patients exhibited clinical CNS disease and 10 of 26 evaluated patients were PCR positive for T whipplei in the CSF. Strikingly, no patient was lost to follow-up and the overall clinical outcomes were excellent. The study demonstrates clearly that the therapy of rare diseases, such as Whipple's disease, can be addressed with an outstanding clinical trial framework, dedicated expert investigators, and appropriate funding support. However, the trial raises important questions and does not provide clear evidence that all patients with Whipple's disease require initial intravenous therapy with ceftriaxone or meropenem.
Should all patients with Whipple's Disease be treated with an initial 2-week course of parenteral ceftriaxone or meropenem?
Concern has been raised that Whipple's disease routinely involves the CNS based on limited autopsy data, a reported tendency for Whipple's disease to relapse with CNS manifestations, and isolation of T whipplei from the CSF of 2 patients without neurologic symptoms or signs. Historical data suggest doxycycline as well as TMP-SMX may be insufficient for Whipple's disease involving the CNS. CNS Whipple's disease may manifest as brain parenchymal, spinal cord, or peripheral nervous system disease with an abnormal CSF examination in 50%–65%. The pathogenesis of Whipple's disease is thought to be due to a potential genetic immune disorder involving macrophage biology. Tetracyclines are concentrated in macrophages and TMP-SMX is effective in treating CNS infections with Listeria monocytogenes (another intracellular bacterium). Nonetheless, it is possible that meningeal macrophages are abundant in Whipple's disease, thereby providing a sanctuary impenetrable to tetracyclines or TMP-SMX.
The authors indicate that they chose “relatively high” doses of ceftriaxone and meropenem because these antibiotics penetrate the blood brain barrier and to overcome any putative intermediate susceptibility of T whipplei. At the time of initiation of the study in 1999, limited data suggested that ceftriaxone was helpful in CNS disease whereas meropenem was a theoretical choice based on minimum inhibitory concentrations of related actinomycete organisms. The antibiotic doses in the trial were standard dosing for serious systemic infections but are not considered adequate for CNS infections where, to maximize CNS drug levels, ceftriaxone is administered at 2 g every 12 hours and meropenem at 2 g every 8 hours in patients with normal renal function. Adequate antibiotic dosing for therapy of CNS infections is even more critical when evidence of meningeal inflammation is limited, as occurs not infrequently in chronic Whipple's disease. Only 1.5% and trace amounts of ceftriaxone and meropenem, respectively, penetrate the noninflamed meninges compared with approximately 16%–32% and 12% penetration, respectively, for inflamed meninges. By contrast, the CSF penetration of doxycycline is estimated between 10% (noninflamed meninges) to 26% (in neurosyphilis) and 40% for TMP-SMX with noninflamed meninges.4 Failure of lower dose ceftriaxone and relapses after even protracted ceftriaxone treatment in Whipple's disease are reported.
Although it is possible that CNS spread of T whipplei is typical during the protracted course of Whipple's disease, it is not clear that such putative infection routinely results in clinical sequelae requiring directed antibiotic therapy. Syphilis and histoplasmosis provide examples of 2 infections that commonly invade the CNS, yet therapy is not directed to cure CNS involvement with either infection unless clinical or laboratory manifestations of CNS disease are apparent. The typically slow pace of Whipple's disease combined with the now available CSF PCR for detection of T whipplei suggests that timely diagnosis and institution of intravenous antibiotic therapy is feasible in individual patients without exposing all Whipple's disease patients to the inconvenience, potential complications, and expense of initial parenteral antibiotic therapy. The authors report that all patients were in “incipient” remission by 3 months in either arm of the trial with small bowel histologic remission in all by 12 months; the data do not allow the reader to discern whether this fairly prompt response resulted from the use of an initial 2 weeks of parenteral therapy or would have been observed with use of TMP-SMX alone. Other reports suggest that the response of Whipple's disease to oral antibiotic therapy alone may be prompt. Perhaps an appropriate analogy is tuberculosis, where oral antimicrobial therapy leads to frequent, prompt clinical and microbiologic responses despite the need for protracted therapy to induce cures.
Why did the clinical response rate in this trial exceed prior observations?
In the trial reported by Feurle et al,3 only 1 of 40 evaluable patients required institution of an alternative antibiotic regimen compared with prior reported failure rates up to 30%. It is certainly possible that the initial parenteral antibiotics—delivered with certainty—were key to the trial's success in which no clinical relapses of Whipple's disease were detected (over 71–126 months) and all CSF PCR examinations were negative at 6 months. However, our knowledge of T whipplei and its antibiotic sensitivity have been refined since initial cultivation of the organism. Subsequent investigations revealed that 2 pools of T whipplei—intracellular and extracellular—exhibit differential antibiotic sensitivities; that tetracycline therapy may be enhanced with chloroquine compounds that elevate macrophage vacuolar pH reducing T whipplei viability in vitro, that carbapenem sensitivity is variable, and that TMP is inactive as T whipplei lacks sequences for the target of TMP, dihydrofolate reductase.5, 6 Additional pharmacokinetic data suggest that sulfadiazine may be better than SMX to treat Whipple's disease with CNS involvement. Sulfadiazine is 20% protein bound with 50% penetration of normal CSF (50%–80% penetration of inflamed meninges), whereas SMX is 65% protein bound, resulting in lower levels in normal CSF (25%–30%, inflamed meninges). Taken together, these data suggest that TMP-SMX may never have been the optimal oral antibiotic choice and that doxycycline therapy may be enhanced by addition of chloroquine compounds. The data lead us to consider whether oral therapy can now be optimized to treat Whipple's disease, limiting the need for parenteral antibiotics. This latter concept is supported by the 1 therapeutic failure in the trial, a clinically asymptomatic patient with a persistently positive CSF PCR for T whipplei despite treatment in the ceftriaxone arm followed by treatment in the meropenem arm, each accompanied by 1 year of oral TMP-SMX therapy. The CNS infection of this patient ultimately cleared with a combination of minocycline and chloroquine.
Alternatively, the excellent clinical outcomes in the current trial may relate to enhanced, long-term antibiotic adherence promoted by the context of a clinical trial in which regular physician and investigator contact contributed to treatment success. Adherence was monitored by the investigators, but no details are provided. The combined contributions of inadequate adherence and shorter courses of therapy to prior therapeutic failures in Whipple's diseases are unclear. Antimicrobial adherence and approaches to promoting adherence have evolved as paramount to the successful management of infectious diseases over the last 15 years with the advent of antiretroviral therapy for HIV infection and combined antimicrobial and antisecretory therapies for treatment of Helicobacter pylori infections.
Can the Study Design of Feurle et al3 Be Strengthened in Future Studies of the Therapy of Whipple's Disease?
The authors of the current trial of Whipple's disease therapy are to be applauded for tackling a randomized treatment trial for Whipple's disease. Nonetheless, this trial does not provide clear answers on how to approach the management and therapy of patients with Whipple's disease. We suggest (Figure 1) that future studies of Whipple's disease could be enhanced by application of molecular diagnostics to identify the organism in body fluids and evaluate its clearance over time. Further, cultures for T whipplei should be performed on samples in which the organism is detected by PCR and, if isolated, antibiotic resistance of the patient's isolate directly determined to guide antibiotic therapy. This is particularly important given the recent reports of T whipplei sulfonamide resistance. Given the relatively slow pace of Whipple's disease in most patients and the likelihood that enhanced use of molecular tools will lead us to make the diagnosis of Whipple's disease more briskly, we predict that patients will be diagnosed at earlier stages of the disease and can be stratified for evidence of microbiologic or clinical CNS involvement leading to enhanced cure of Whipple's disease, possibly with oral antibiotics. Although details are not available, an additional trial is reported to be in progress to assess clinical responses to a combination of doxycycline and hydroxychloroquine with the addition of a sulfonamide in those with detection of T whipplei in the CSF or with clinical CNS Whipple's disease. We suggest reserving use of broad-spectrum, high-dose, parenteral antibiotics for patients with Whipple's disease judged as advanced or life threatening by the physician, for Whipple's disease patients with an apparent accelerated course of disease, and for immunocompromised patients where Whipple's disease may also exhibit a more aggressive clinical course. For these limited circumstances, we favor use of ceftriaxone given the rising tide of antimicrobial resistance in which carbapenems should be reserved for therapy of resistant gram negative rods.
Figure 1.
Clinical trial design considerations in Whipple's disease. The figure summarizes potential considerations in designing future clinical therapeutic trials in Whipple's disease. Addition of hydroxychloroquine or chloroquine to doxycycline therapy may enhance intracellular killing of T whipplei. The relative anti-T whipplei potency of the 2 chloroquine formulations is unknown. Therapy of Whipple's disease with CNS extension should continue until clinical resolution and CSF PCR for T whipplei is negative at least twice. Additional details are in the text.
A limitation of the trial design as conducted by Feurle et al3 is that clinical data were not obtained and microbiologic samples not banked at frequent enough intervals to discern, in retrospect, the impact of the initial parenteral antibiotic therapy on the clinical course and/or elimination of T whipplei. The authors obviously could not have known that detection methods would soon appear making such a sample bank useful. However, this experience serves to alert other investigators to the importance of banking samples, particularly in rare diseases, with an eye to the possibilities of future testing. Additional detailed studies to better characterize, using combined molecular, microbiologic, genetic and clinical tools, which patients may do poorly with oral antibiotic therapy will sharpen the rational use of antibiotic therapy in Whipple's disease while providing the best care possible for individual patients. Life-long follow-up for disease relapse seems prudent.
References
- . A hitherto undescribed disease characterized anatomically by deposits of fat and fatty acids in the intestinal and mesenteric lymphatic tissues. Johns Hopkins Medical Bulletin. 1907;198:382–391
- Identification of the uncultured bacillus of Whipple's disease. N Engl J Med. 1992;327:293–301
- . Efficacy of ceftriaxone or meropenem as initial therapies in Whipple's disease. Gastroenterology. 2010;138:478–486
- In: Kucers A, Crowe SM, Grayson ML, et al. editor. The use of antibiotics: A clinical review of antibacterial, antifungal and antiviral drugs. 5th ed.. Woburn, MA: Butterworth–Heinemann; 1997;
- . Antibiotic susceptibility of Tropherma whipplei in MRC5 cells. Antimicrob Agents Chemother. 2004;48:747–752
- . Molecular evaluation of antibiotic susceptibility of Tropheryma whipplei in axenic medium. J Antimicrob Chemother. 2005;55:178–181
Conflicts of interest The authors disclose the following: Dr Sara E. Cosgrove serves on the advisory board of Forest Laboratories, as a consultant to Merck, and receives grant support from Cubist and Astellas. Dr Cynthia L. Sears discloses no conflicts.
PII: S0016-5085(09)02153-2
doi:10.1053/j.gastro.2009.12.011
© 2010 AGA Institute. Published by Elsevier Inc. All rights reserved.
Refers to article:
-
Efficacy of Ceftriaxone or Meropenem as Initial Therapies in Whipple's Disease
, 30 October 2009
