Treatment of Category III A Prostatitis With Zafirlukast: a Randomized Controlled Feasibility Study

Summary: The cause of category III A prostatitis, chronic prostatitis/chronic male pelvic pain syndrome category A (CP/CPPS A), is uncertain. Treatments for it are based on consensus opinion rather than on scientific data. Our aim was to examine the effect of zafirlukast, a leucotriene antagonist, on the symptoms of CP/CPPS A in our genitourinary (GU) medicine unit. CP/CPPS A was diagnosed by comparative white cell counts of split urine (Stamey) analysis or by finding an excess of polymorphs in expressed prostatic fluid. Symptom change was assessed by the National Institutes of Health Chronic Prostatitis Symptom Index (CPSI). Patients were given zafirlukast or placebo for four weeks in a random double-blind fashion. All patients also received doxycycline. In all, 31 patients were asked to participate and 17 entered the study. No difference in outcome could be shown between the active (10) and placebo (seven) patients. Zafirlukast cannot be demonstrated to be useful in the symptomatic treatment of CP/CPPS A. The problems of recruitment into this study (in spite of a large number of patients with prostatic type pain being seen in our unit) suggest that multicentre treatment trials using non-invasive diagnostic techniques such as the CPSI (rather than single GU medicine units diagnosing CP/CPPS A by uncomfortable direct prostatic testing) are likely to be the most effective and objective methods of undertaking treatment trials in the CP/ CPPS A field in the future.

Keywords: zafirlukast, prostatitis, treatment trial

Introduction

The prevalence of chronic prostatitis in the general population is estimated to be as high as 15%.1 The extent of the negative impact of the illness on patients is severe, and on a par with recent myocardial infarction, unstable angina or active Crohn’s disease. Pain is the main symptom of chronic prostatitis.3

Prostatitis has been classified by consensus opinion into various categories.4 These include acute prostatitis (category I), chronic bacterial prostatitis (II), chronic abacterial prostatitis/ chronic pelvic pain syndrome (III) and category IV (asymptomatic inflammatory prostatitis). Category III has been further subdivided into IUa (inflammatory) and IUb (non-inflammatory). The commonest type presenting to genitourinary (GU) medicine clinics in the UK is type III,5 which can be more specifically categorized as chronic prostatitis/chronic male pelvic pain syndrome category A (CP/CPPS A).

The cause of CP/CPPS A is uncertain, although there is some evidence that Chlamydia tmchomatis, mycoplasma and Trichomonas vaginalis may be implicated at least as initiating agents.6’7 CP/ CPPS A is commonly associated with non-gonococcal urethritis.8’9 Patients with urethritis and prostatitis have a higher than expected rate of HLA B27 histocompatibility antigen.9 Furthermore, persistence of chlamydial heat shock protein (HSI 60) is associated with chronic urogenital inflammation.1″ HSP60 contains antigenic sites that are specific to chlamydia, but also has shared sites with human tissue. The above data suggest that CP/CPPS A may have an autoimmune component, and this has also been hypothesized by other workers.11

The wide scope of treatments for CP/CPPS A bears testimony to how little is really understood about the condition. A recent Cochrane review found that treatment trials are few, methodologically weak and involve small sample sizes. Although the routine use of antibiotics and alpha blocking agents is advocated,4 their efficacy is not supported by good evidence.12

In spite of this, antibiotics such as doxycycline and ciprofloxacin are first-line treatment in 98% of GU medicine clinics in the LJK, mostly in four- to six-week courses.5

Because of the lack of an effective treatment for CP/CPPS A, we considered that a leucotriene antagonist might be useful for the treatment of CP/CPP for the following reasons.

Studies of patients with CP/CPPS A suggest that the prostate is infiltrated with chronic inflammatory cells,13 in particular macrophages.14 Leucotrienes are important metabolites of arachidonic acid and are produced by inflammatory leucocytes. They mediate and modulate local inflammatory responses. The leucotriene precursor 5- lipooxygenase (5-LO) is found in macrophages.15 Of relevance to CP/ CPPS A is the fact that leucotrienes have also been shown to sensitize prociceptors and may be a long-lasting hyperalgesic factor.16

Leucotriene antagonists, largely used for the control of mild to moderate asthma, possess distinct anti-inflammatory effects.17 It has also been suggested that they may contribute to decreased pain experienced in patients with acute appendicitis.18

Zafirlukast antagonizes the cysteinyl receptor, which inhibits leucotriene synthesis. It has a treatment licence in the UK for the management of asthma and appears to have a good safety record and low toxicity profile.19

We therefore set up a randomized controlled feasibility trial to compare treatment of CP/CPPS A with zafirlukast against placebo over a four-week period.

Methods

Men aged between 18 and 60 with pain symptoms suggestive of CP/ CPPS A were included in the study. The locations of these symptoms was at one or more of the following sites: area between the rectum and testicle, testicles, tip of penis (not related to urination), suprapubic area; symptoms occurred with pain or burning during masturbation, pain or discomfort during or after ejaculation, or as urinary symptoms.20 Additionally, patients underwent a standard four- glass urine test,21 with microscopy of expressed prostatic fluid (EPS).22

The diagnosis of CP/CPPS A was made by excluding chronic bacterial prostatitis21 and sexually transmitted urethral pathogens. In particular, CP/CPPS A was diagnosed by finding at least 10 white cells per 400 microscopic field in EPS, or clumps of at least 10 such cells in one field, or by finding a 10-fold higher white cell count in post-prostatic massage urine compared with preprostatic massage urines. Although the four-glass test has been useful in diagnosing prostatitis for decades, its accuracy and reliability have not been clearly demonstrated.23 Other studies demonstrated that the specificity of the four-glass test and microscopic prostatic fluid examination to detect leucocytes is high.8’24 Additionally, urinary localization techniques/prostatic microscopy are used in diagnosing prostatitis in the majority of GU medicine clinics in the UK.5

Men who had symptoms of CP/CPPS A for longer than two years, those who had taken medication for CP/CPPS A in the past four weeks, those known to be HIV positive or at high risk of it, patients with prostatic carcinoma, those allergic to zafirkulast or doxycycline, men with known renal or hepatic impairment, those with a poor command of English and those with known eosinophilia or severe allergic rhinititis (contraindication for zafirlukast) were excluded from the study.

Transrectal ultrasound of the prostate was undertaken routinely on all patients to exclude abscesses and carcinomas. Serum prostate- specific antigen (PSA) levels were requested where appropriate.

The National Institutes of Health Chronic Prostatitis Symptom Index (CPSI) inventory 2″ was used in the study. This has an index of nine symptom items. The primary component is pain. The nine items have a high test rate reliability (r = 0.83-0.93) and internal consistency (α = 0.86-0.91). All but the urinary items discriminate well between prostatitis and controls.20

A physician screened patients for eligibility and, where applicable, asked for informed written consent before they were passed to a research nurse. The research nurse asked them to self- complete the CPSI inventory and allocated the next sequential number in the randomization list. This number was recorded on their notes and on an outpatient prescription form, which was presented to the pharmacy. The pharmacy dispensed the appropriate medication according to the trial number either zafirlukast 20 mg twice daily or placebo for four weeks. Patients were randomized using permuted blocks of four different block sizes. All patients additionally received doxycycline 100 mg twice daily for four weeks. We added doxycycline into both treatment arms because of its standard use for CP/CPPS A in 86% of GU medicine clinics in the UK rather than because of objective efficacy.23,25

After four weeks, the CPSI was re-administered (blind-to- treatment arm).

The study had the approval of the local ethics committee.

As the investigation was principally a feasibility study, no sample size calculations were undertaken. Our primary hypothesis was that pain or discomfort scores would be lower in the group receiving zafirlukast. Pain/discomfort scores were calculated from the responses to four questions on the CPSI inventory, namely:

In the last week, have you experienced any pain or discomfort in the following areas:

area between rectum and testicles (perineum)?

testicles?

tip of the penis (not related to urination)?

below your waist in your pubic or bladder area?

Patients score one for each area

In the last week have you experienced:

pain or discomfort during urination?

pain or discomfort during or after sexual climax (ejaculation)?

Patients score one for each

How often have you had pain or discomfort in any \of these areas over the last week?

Never, rarely, sometimes, often, usually, always

Scored 0 (Never) through 5 (always)

Which number best describes your average pain or discomfort on the days that you had it, over the last week?

Scale 0 (no pain) to 10 (pain)

The maximum score in response to these questions was 21.

Although there was little chance of detecting a clinically important effect with such a small sample size, an intention-to- treat analysis was undertaken. Statistical comparisons were made of follow-up scores (post-treatment scores), change score (subtracting the follow up score from the baseline score) and by analysis of covariance. Analysis and randomization were carried out using Stata 6. Given the small sample size, model adequacy/ assumptions were not tested. However non-parametric techniques were used to confirm the findings (Mann-Whitney and quantile (median) regression).

Table 1 Baseline characteristics of patients in the study

Table 2 Mean and median scores (in brackets) on the National Institutes of Health Chronic Prostatitis Symptoms Index inventory at baseline and follow-up for the two treatment arms -placebo and zafirlukast

Results

In all, 31 patients were asked to participate in the study over a period of 16 months. Twenty agreed to participate and entered the initial assessment and randomization session. Seventeen patients completed the study (10 in the zafirlukast arm and seven in the placebo arm). Three patients were excluded from the analysis because they failed to return for the follow-up visit (two in the zafirlukast arm and one in the placebo arm) Baseline characteristics are presented in Table 1.

At initial assessment, eight of 10 (80%) in the zafirlukast arm had an excess of leucocytes in prostatic fluid compared with five of seven (71 %) in the placebo arm. For a significant leucocytosis in the post-prostatic massage urine, the figures were three of 10 (30%) and three of seven (43%) for zafirlukast and placebo, respectively. The number of days (mean, SD and range) between initial and follow- up visits was 32.3 (2.5) (29-36) for the actively treated group and 35.7 (15.3) (228-270) for the placebo arm.

The pre- and post-treatment scores on the NIH CPSI are shown in Table 2. Table 3 gives the results of the comparison of follow-up scores, change scores and analysis of covariance for pam or discomfort. There was no statistical difference in mean pain scores using the three approaches. These findings were confirmed in the non- parametric analysis (results not shown).

Three of the zafirlukast patients and four in the placebo group had gastrointestinal side-effects. One patient in the active arm complained of a dry mouth and one patient in the placebo group had a mild hypersensitivity reaction. Three patients in each arm took concomitant medications.

Table 3 Pain scores by treatment group

Discussion

This study could not show that there was any benefit from adding zafirlukast to doxycycline for the treatment of CP/CPPS A. However, there was little chance of detecting a clinically important effect with such a small sample size. This is not to say that there was no benefit, only that we could not demonstrate one. In spite of quite large numbers of patients seen in routine clinics with CP/CPPS A, we found difficulty in recruiting patients into this study. This may be for a number of reasons, including the necessity for cessation of medication for CP/CPPS A for four weeks prior to study entry, a relatively invasive and uncomfortable diagnostic test involving prostatic massage and inclusion/exclusion criteria being too strict – in particular, the necessity to demonstrate cellular inflammatory change suggestive of CP/CPPS A.

The American Multicentre Chronic Prostatitis Collaborative Research Network sponsored by the National Institutes of Health has suggested that leucocytosis findings in prostatic fluid or postprostatic urine have never provided a clear distinction that is important to patients, regarding either an aetiology or treatment. 6 Indeed, in terms of inflammatory change, where this is carefully assessed by prostatic biopsy, there may be as great a degree of chronic inflammatory infiltrate in group UIb as in group Ilia.27 Furthermore, quite a degree of overlap exists between leucocyte counts in men with and without prostatitis.28 Data from the NIH report showed no association between leucocyte counts and the NIH- CPSI symptom scores.29 Because of all of the above, we recommend that entry and outcome criteria for studies into CP/ CPPS A should be rationalized and eased to, for example, a certain score on the NIH-CPSI, e.g. 15 or more out of a potential 43.26 This, we feel, would encourage more patients to enter into studies of chronic prostatitis, thus giving more power to detect differences in treatment arms, as well as not requiring unpleasant, undignified and painful tests to be taken by patients for diagnostic purposes.

As stated above, it is possible that zafirlukast does have a beneficial effect on symptoms of CP/ CPPS A but that this is masked by a positive therapeutic effect of doxycycline. The pain/ discomfort scores fell in both active and placebo arms after treatment compared with these before treatment. This may be because of a ‘placebo’ effect in both arms, or it may represent a true positive therapeutic effect – albeit one that does not reach statistical significance. As this present study did not set out to compare the effect of doxycycline on CP/CPPS A in a placebo controlled fashion, the cause of the fall in pain scores must remain speculative. However, there is some subjective, clinical non- placebo-controlled evidence that doxycycline has a beneficial outcome in the treatment of prostatitis from at least two studies.25’30 Furthermore, a number of in vitro studies show that tetracyclines, particularly doxycycline, have significant effects on impairing polymorph function,31″35 macrophage /monocy te cell function36’37 suppressing delayed hypersensitivity reaction38 and generally inhibiting mammalian cell proliferation and migration.39 Additionally, many patients with chronic prostatitis/chronic pelvic pain syndrome have a wide variety of bacterial DNA encoding sequences, 75% of which are tetracycline sensitive, despite extensive negative microbiological investigations. Patients with these encoded sequences are significantly more likely to have expressed prostate secretion leucocytosis.6 However, as stated above, we felt that even though doxycycline has not been shown objectively to improve symptoms of CP/CPPS A, it might be unethical to withhold it. Our ethics committee concurred with these thoughts. With the benefit of hindsight, we feel we should have, and could ethically have, dispensed with the doxycycline arms altogether.

In spite of this, we still had problems recruiting patients to the study, as alluded to above. We feel the way forward is to perform multicentre studies in the UK, as has been proposed and now been undertaken in the USA.26

Our final conclusions are that although the efficacy of zafirkulast in the treatment of CP/CPPS A remains unproven, we recommend that multicentre studies be undertaken to assess therapeutic agents for efficacy, with arms that include a single active agent, and placebo, and yet other arms with more than one agent.

Acknowledgements: This study was undertaken with the use of generous grants from The Prostate Research Campaign UK and AstraZeneca, UK.

References

1 Moon TD, Hagen L, Heisey DM. Urinary symptomatology in younger men. Urology 1997;50:700-3

2 Wenninger K, Heimon JR, Rothman I, cf nl. Sickness impact of non-bacterial prostatitis and its correlates. J Urol 1996;155:965-8

3 National Institute of Health Summary Statement. National Institute of Health/National Institutes of Diabetes and Digestive and Kidney Diseases Workshop on Chronic Prostatitis. Executive Summary. Bethesda, MD; December 1995

4 Nickel JC, Nyberg LM, Hennenfent M. Research guidelines for chronic prostatitis: consensus report from the first national institute of health international prostatitis collaborative network. Urology 1999;54:229-33

5 Luzzi GA, Bignell C, Mandai D, Maw RD. Chronic prostatitis/ chronic pelvic pain syndrome. National Study of genito-urinary medicine clinics, hit ] STD AIDS 2002;13:416-19

6 Kreiger JN, Riley DE. Prostatitis, what is the role of infection? lut ] Agents 2002;19:475-9

7 Weidner W, Schiefer HG, Krauss H. Role of Chlamydia trnchomatis and mycoplasmas in chronic prostatitis. A review. UmI Int 1988;43:167-73

8 Kreiger JN, Jacobs R, Ross SO. Detecting urethral and prostatic inflammation in patients with chronic prostatitis. Urology 2000;55:186-91

9 Taylor Robinson D, Furr PM, Coufalik ED. A detailed examination of some patients with recurrent non gonococcal urethritis. Eur J Sex Transm Dis 1985;3:25-8

10 Herner PJ, Cain D, McClure M, t’f at. Association of antibodies to Chlamydia trnchomatis heat shock protein 60 kD with chronic non gonococcal urethritis. Clin Infect Dis 1997;24:653-60

11 Batstone GR, Doble A, Gaston JS. Autoimmune T cell responses to seminal plasma in chronic pelvic pain syndrome (CPPS). Clin Exp linmnnol 2002;128:302-7

12 McNaughton C, MacDonald R, Wilt T. Intervention for chronic abacterial prostatitis. Cochrane Database Syst Rev 2001;1:CD002080

13 Doble A, Thomas BJ, Furr PM, et al. A search for infectious agents in chronic abacterial prostatitis using ultrasound guided biopsy. Br] Urology 1989;64:297-301

14 Nishimara T, Kanamori S, Akimoto M, Kawai H. Macrophages in prostatic fluid. Br I UwI 1980;52:381-5

15 Coffey MJ, Phare SM, Peters-Golden M, Haffnagle GB. Regulation of 5-Iipoxygenase metabolism in mononuclear phagocytes by CD4 lymphocytes. Exp Lung Res 1999;25:617-29

16 Madison S, Whitesel EA, Suarez-Roca H, Maixner W. Sensitizing effects of leukotriene B4 on intradental primary afferents. Pain 1992;49:99-104

17 Aprile A, Lucarelli S, Vagnucci B, Frediani T. The useof antileukotrienes in paediatrics. Eur Rev Med Pharmacol Sd 2001:5:53- 7

18 Lawyer C. Montelukast for children with asthma. JAMA 1998;280:967

19 Grossman J, Smith LJ, Wilson AM, Thyrum PT. Long term safety and efficacy of zafirkulast in the treatment of asthma; interim results of an open label extension trial. Ann Allergy Asthma linmuiiol 1999;82:361-9

20 Litwin MS, McNaughlan-Collins MM, Fowler FJ, et at. The National Institutes of Health Chronic Prostatitis Symptom Index: development and validation of a new outcome measure. J UmI 1999;162:369-75

21 Meares EM, Stamey TA. Bacteriological localisation patterns in bacterial prostatitis and urethritis. Invest Urol 1968;5:492-518

22 Kreiger JN. Prostatic syndromes In: Holmes KK, Sparling PR, March P-A, eds. Sexually Transmitted Diseases. 3rd edn. New York: McGraw-Hill, 1999;859-71

23 McNaughton-Collins M, MacDonald R, Timothy J. Diagnosis and treatment of chronic abacterial prostatitis: a systematic review. Ann Intern Meil 2000;133:367-8I

24 Muller CH, Berger RE, Mohr LE, Kreiger JN. Comparison of microscopic methods for detecting inflammation in expressed prostatic secretions. J UwI 2001 ;166:2518-24

25 Simons PD, Thin RN. Minocycline in chronic abacterial prostatitis: a double blind prospective trial. Br J Urol 1985;57:43- 5

26 Propert KJ, Alexander RB, Nickel JC, et al. Design of multicentre randomised clinical trial for chronic prostatitis/ chronic pelvic pain syndrome. Urology 2002;59: 870-6

27 John H, Barghorn A, Funke G, et al. Non-inflammatory chronic pelvic pain syndrome: immunological study in blood, ejaculate and prostate tissue. Eur Urol 2001;39: 72-8

28 Schaeffer AJ, Wendel EF, Dunn JK, et al. Prevalence and significance of prostatic inflammation. J Urol 1981;125: 215-19

29 Schaeffer AJ, Knauss JS, Landis JR, et al. Leukocyte and bacterial counts do not correlate with severity of symptoms in men with chronic prostatitis: the NlH chronic prostatitis cohort (CPC) study. J Urol (in press)

30 Magoha GA. Ten years experience with chronic prostatitis in Africans. East Afr Meet J 1996;73:176-8

31 Glette J, Sandberg S, Hopen G, Solberg CO. Influence of tetracycline on human polymorphonuclear leucocyte function. Antimicrob Agents Clieniother 1984;25:354-7

32 Grec V, Frei PC. Effect of amoxycillin and doxycycline on function of human granulocyte tested in vitro and on chemotaxis of granulocytes from rabbits given the two antibiotics. Inflammation 1984;8:417-27

33 Hoeben D, Bervenich C, Heyneman R. Antibiotics commonly used to treat mastitis and respiratory burst of bovine polymorphonuclear leukocytes. J Dairy Sd 1998;81: 403-10

34 Forsgren A, Banck G, Beckman H, Bellahsene A. Antibiotic host defence interactions in vitro and in vivo. Stand J Infect Dis Snppl 1980;24:195-203

35 Belsheim J, Blomqvist C, Lofberg J, Gnarpe H. Tetracyline influence on leukocyte function. Ada Otorhinolaryngol BeIg 1983;7:635-48

36 Ganguly R, Pennock D, Kluge RM. Suppression of the production of migration inhibitory factor in humans by tetracycline. J Infect Dis 1983;148:611

37 Nielson H. Antibiotics and human monocyte function 1. Chemotaxis. Ada Pnthol Microbiol Scand 1987;95:293-6

38 Thong YH, Ferrante A. Effect of tetracycline on immunological responses in mice. CHn Exp Imnninol 1980;39: 728-32

39 Bendeck MP, Conte M, Zhong M, ct at. Doxycycline modulates smooth muscle cell growth, migration and matrix remodelling after arterial injury. Am J Pathol 2002;160:1089-95

(Accepted 3 January 2003)

David Goldmeier MD FRCP1-, Peter Madden MSc2′, Michael McKenna MB1′ and Norbert Tamm RN M Res1′

1 Jefferiss Wing, St Mary’s Hospital, London; 2 lmperial College, London, UK

Correspondence to: Dr David Goldmeier Email: david.goldmeie@st- marys.nhs.uk

Copyright Royal Society of Medicine Press Ltd. Mar 2005