Trimethoprim/Sulfamethoxazole Versus Placebo: A ...

Trimethoprim/Sulfamethoxazole Versus Placebo: A Double-Blind Comparison of Infection Prophylaxis in Patients With Small Cell Carcinoma of the Lung By Carlos A. de Jongh, James C. Wade, Rebecca S. Finley, Jai H. Joshi, Joseph Aisner, Peter H. Wiernik, and Stephen C. Schimpff The suppression of pathogenic aerobes and the preservation of anaerobes provides a degree of infection prevention during granulocytopenia. Trimethoprim/ sulfamethoxazole (TMPISMZ) suppresses Enterobacteriaceae and probably maintains colonization resistance through sparing of anaerobes. TMP/SMZ (320/ 1600 mg/day) treatment was compared to placebo in a double-blind, randomized trial in patients with newly diagnosed small cell carcinoma of the lung during the initial courses of chemotherapy with cyclophosphamide, doxorubicin, and etoposide. Infections were evaluated as microbiologically documented, with or without bacteremia, and clinically documented and were correlated to granulocytopenia. Of the

S

MALL CELL LUNG CANCER is a distinct clinical and histopathologic entity accounting for 20%-25% of all lung cancers. The tumor typically produces hilar and mediastinal masses and metastasizes early, usually before the patient becomes symptomatic. Untreated patients with extensive disease have a median survival of less than 8 wk.' In recent years, this particular neoplasia has been shown to be highly responsive to chemotherapy; different regimens result in prolongation of life and some long-term, diseasefree survivors." The use of aggressive chemotherapy, however, is associated with certain complications, particularly myelosuppression,

From the University of Maryland CancerCenter, University of Maryland School of Medicine and Hospital. Baltimore, Md. Submitted October 22, 1982; accepted January 10, 1983. This paper was presented in part at the 18th Annual Meeting of the American Society of Clinical Oncology, St. Louis, Mo., April 25-27. 1982. A portion of the patients reportedin this paper will be included in a report by the EORTC InternationalAntimicrobial Therapy Project Group evaluation of infection prophylaxis among patients with cancer of various types. Address reprint requests to CarlosA. de Jongh, M.D., University of Maryland Cancer Center, University of Maryland Hospital, 22 South Greene Street, Baltimore, Md. 21201. V 1983 by American Societv of Clinical Oncology. 0732-183X/83/050302-06$0.1.00/0

302

61 patients evaluated, 32 were given TMP/SMZ and 29 were given placebo; both groups had similar characteristics with regard to disease extent, performance status, age, sex, chemotherapy, and days of granulocytopenia. Incidence of infection at less than 100 granulocytes//Cl was significantly reduced in the TMP/ SMZ group (2 infections/1 00 days) compared to placebo (11 infections/100 days, p = 0.005). Also reduced were the number of bacteremias and the mean proportion of study time on broad-spectrum antibiotics (p < 0.01). Compared to placebo, TMP/SMZ provided infection prophylaxis without an increase in marrow suppression among patients with small cell carcinoma of the lung receiving intensive chemotherapy.

with resultant granulocytopenia and hence an increased incidence of infection. Infections in patients with granulocytopenia are often life threatening. Most of these infectious episodes develop from organisms colonizing the patient's alimentary canal and are manifested as oral infections, esophagitis, perianal cellulitis, and pneumonias.5 Among the logical approaches for infection prevention, suppression of potential pathogens already colonizing the patient as well as reduction of organism acquisition have been the subject of intensive evaluations. The use of oral nonabsorbable antibiotics or absorbable, anaerobic-sparing agents has resulted in a substantial reduction of infection rates. However, considerable controversy still exists concerning their proper use.6 Reports that trimethoprim/sulfamethoxazole (TMP/SMZ) reduces the incidence of infection in patients with acute leukemia7" prompted us to compare its efficacy to that of placebo as infection prophylaxis in a prospective, randomized, double-blind trial among patients with small cell carcinoma of the lung who were receiving intensive combination chemotherapy. 0 MATERIALS AND METHODS Patients at the University of Maryland Cancer Center, Baltimore, with newly diagnosed small cell carcinoma of the lung, Journal of Clinical Oncology, Vol. 1, No. 5 (May) 1983

Information downloaded from jco.ascopubs.org and provided by at UCLA on July 30, 2011 from 164.67.83.5 Copyright © 1983 American Society of Clinical Oncology. All rights reserved.

303

SELECTIVE SUPPRESSION FOR PROPHYLAXIS who were afebrile, were not receiving therapeutic antibiotics, and were about to begin cytotoxic chemotherapy for remission induction as part of two successive studies 2 ' 0 were study candidates. After informed consent was obtained, patients were randomized in a double-blind fashion to receive TMP/SMZ tablets (Burroughs-Wellcome Co., Research Triangle Park, N.C.) 160/880 mg every 12 hr or an identically appearing placebo. Compliance was monitored through review of computerized drug administration records and patients were frequently reminded of the importance of following the prophylactic regimen. Some patients were monitored for suppression of aerobic microflora; the code was broken for these patients and correlated with the stool cultures only after completion of the oral prophylaxis. 2 In order to simplify the measures for outpatient usage, no other specific antiinfective measures, such as cooked-food diets, skin antisepsis, etc., were utilized other than to stress the importance of hand washing. Antifungal agents such as nystatin or amphotericin B were not utilized. The control chemotherapy regimen in each study was given at 3-wk intervals and consisted of cyclophosphamide 1000 mg/m2 IV on day 1; doxorubicin 45 mg/m 2 IV on day 1, and etoposide (VPI6-213). In the first study, an alternating regimen was initiated to patients randomized to alternating chemotherapy; however, the second combination began after the third course of cyclophosphamide, doxorubicin, etoposide chemotherapy and after the completion of the TMP/SMZ-placebo prophylaxis study. In the subsequent study, the experimental chemotherapy consisted of cyclophosphamide as before, doxorubicin as before, and etoposide given as 100 mg/m 2/day on days 1-5 via a continuous infusion. Thus, most patients received the cyclophosphamide, doxorubicin, etoposide regimen. On-study time began with the institution of the prophylactic antibiotics, usually the same day as the beginning of the chemotherapy, and continued until the patient had completed three courses of chemotherapy, the patient developed adverse effects felt to be secondary to the study drug, the study drug was discontinued due to adverse effects, the patient withdrew from the study, or the patient died. All febrile episodes were classified according to clinical findings, microbiologic data, and course as microbiologically documented (with or without bacteremia), clinically documented, possible infection, or not infected, as previously described." Only documented infections were analyzed for this evaluation. Patients with episodes of fever (temperature 101 F) were treated with empiric, broad-spectrum intravenous antibiotics. The prophylactic regimen was continued throughout these periods. All evaluations were completed before the code was broken.

RESULTS

Of the 61 patients randomized, 29 received placebo and 32 received TMP/SMZ. The characteristics of the patients in both groups (Table 1) were similar with regard to age, sex, extent of disease at presentation, performance status, cancer chemotherapy, and response to therapy. The mean number of days on study and duration of

granulocytopenia were equivalent.

Table 1. Patient Characteristics Prophylactic Regimen Characteristic Total no. of patients Age (yr) Mean Range Sex (no. of patients) Male Female Granulocytopenia: PMN/p/I* (mean number of days) S100 100-499 500-999 Total < 1000

Placebo

TMP/SMZ

29

32

58 29-79

61 40-76

20 9

23 9

4 5 5 14

3 7 6 16

*PMN: polymorphonuclear leukocytes.

Among patients who received placebo, 19 infections occurred, including 6 bacteremias (3 Klebsiella pneumoniae, 1 Streptococcus mitis, I S. pneumoniae, and 1 polymicrobial bacteremia: Pseudomonas aeruginosa,Escherichiacoli, and Lactobacillus amalonatica), 3 nonbacteremic microbiologically documented (1 Hemophilus influenzae, 1 S.faecalis, and 1 Candidasp.), and 10 clinically documented infections. Of the 19 infections, 16, including 5 bacteremias, occurred at granulocyte counts below 500/pd. There were 8 infections among the patients who received TMP/SMZ, including 1 bacteremia (K. pneumoniae resistant to TMP/SMZ), 4 nonbacteremic microbiologically documented (1 P. aeruginosa, 1 S. faecalis, 2 Candida sp.), and 3 clinically documented infections (Table 2); 4 of these 8 infections occurred at granulocyte counts below 500//pl. The overall number of infections was significantly lower in the TMP/SMZ group (8) than the placebo group (19) (p = 0.01). The incidence of infection per 100 days was found to be lower in the TMP/SMZ group at granulocyte levels of less than 500//l (p = 0.006) and less than 100/xl (p = 0.005) (Fig. 1).

The most common site of infection was the alimentary canal: 12 of 19 infections in the placebo group and 2 of 8 in the TMP/SMZ group occurred there (Table 2). These infections involved predominantly the upper alimentary tract (mouth, pharynx, esophagus). Pneumonias were more frequent among patients receiving placebo


304

DE JONGH ET AL. Table 2.

Infection Sites and Pathogens

Documentation, Sites, and Pathogens Documentation Microbiologically documented With bacteremia Without bacteremia Clinically documented Total infections Site of Infection Alimentary canal* Upper Lower Lung Skin Urine Bacteremia of unknown origin Infecting organisms Gram-negative organisms Gram-positive cocci Yeasts Total pathogens

Prophylactic Regimen Placebo TMP/SMZ

6 3 10 19

11 (1)t 1 5 (3) 1 (1) 1 (1) 8 (4) 3 (2) 1 12 (6)

1 4 3 8

2 2 3 (1) 1 2 (1) 1 2 5 (1)

*The upper alimentary canal was defined as the mouth, pharynx, and esophagus; the lower alimentary canal as the small intestine, colon, and rectum. tThe number with bacteremia is given in parentheses. $One patient in the placebo group had a polymicrobial bacteremia.

(5, of which 3 had positive blood cultures) than among patients receiving TMP/SMZ (2, both clinically documented). Skin infections, urinary tract infections, and bacteremias of unknown origin accounted for the other documented sites. Those patients randomized to placebo spent significantly more on-study time receiving parenteral broad-spectrum antibiotics than did patients who received TMP/SMZ: 16.5% _ 3.8% versus 7.2% - 2.5% (p < 0.01). Of the 29 patients who received placebo, 4 died as a result of an infection: 3 had gramnegative bacteremia (2 K. pneumoniae and 1 polymicrobial P. aeruginosaplus E. coli) and 1 had S. faecalis peritonitis. There were 2 infection-related deaths among the 32 patients who received TMP/SMZ: 1 had a P. aeruginosa urinary tract infection and developed a superinfection after prolonged therapy with systemic broad-spectrum antibiotics, and the other had a clinically documented pneumonia. Enterobacteriaceae were usually, but not al-

ways, eliminated from rectal cultures. Detailed evaluation of rectal flora, presented elsewhere,12 indicated that anaerobes persisted unless the patient received systemic empiric antibiotics for febrile episodes at which time the anaerobic flora was also markedly suppressed. Two patients in the TMP/SMZ and one in the placebo group had their prophylactic regimen stopped because of a skin rash attributed to the study drug at 14, 19, and 29 days. No statistically significant differences in the mean duration of granulocytopenia were found between the TMP/ SMZ and placebo groups. In addition, the dosage of chemotherapy was not significantly reduced among those patients who received TMP/SMZ. DISCUSSION Most combination chemotherapy regimens for remission induction in small cell carcinoma of the lung utilize drugs such as cyclophosphamide, doxorubicin, epipodophyllotoxin (VP16-213), or methotrexate. 2'"- 5 The utilization of these agents in high doses increases the rate of complete response and median survival; most longterm, disease-free survivors have received a high-dose regimen. 3 ,"I8 The use of this intensive approach results in various degrees of myelosuppression, alimentary canal mucositis, and disturbances of the respiratory epithelium ciliary function. As a result, the risk of infectious complications increases and a significant proportion of the patients require hospitalization and administration of parenteral broad-spectrum antibiotics. Oral nonabsorbable antibiotics that are used in an attempt to suppress all bacteria and fungi along the alimentary canal have been found to reduce infections among profoundly granulocytopenic patients.6,"9,20 However, these agents suppress anaerobes (rarely pathogenic among neutropenic leukemia patients) in addition to the potentially pathogenic aerobes.'2 Utilizing the concept of colonization resistance, i.e., that the anaerobic flora will prevent overgrowth of resistant aerobes and acquisition of new potential pathogens, one can suppress the aerobes selectively and hope to obtain similar degrees of prophylaxis.21,22 The ability of TMP/SMZ to suppress the alimentary tract aerobic flora while preserving the anaerobes, plus its low cost, ease of administra-


SELECTIVE SUPPRESSION FOR PROPHYLAXIS

305

M, I

I

anted docu imented

O U)

ions

Z

0 I-o LL

Z

Fig. 1. Incidence of infection per 100 days was lower in the TMP/SMZ group at granulocyte levels of under 100 and under 500 polymorphonuclear leukocytes//AL.

(4)

B

500 GRANULOCYTE COUNT (PMN/pL)

tion, and minimal toxicity prompted us to evaluate its efficacy in a prospective, randomized, double-blind, and placebo-controlled trial of its use as infection prophylaxis for patients with newly diagnosed small cell carcinoma of the lung who were to receive intensive combination chemotherapy. No other specific prophylactic measures were utilized other than to instruct the patients with regard to simple hygiene measures such as hand washing. Antifungal agents were not utilized as we suspected that the relatively brief periods of granulocytopenia to be encountered should not markedly increase the risk of candida infections, although Hughes et a17 were concerned about possible increased fungal infections among children who received TMP/ SMZ as pneumocystis prophylaxis during remission maintenance. We have previously presented data indicating the efficacy of TMP/SMZ in suppressing the aerobic flora of these patients and those with acute leukemia.' 2 Anaerobes persist and few organisms are acquired unless the patient must also receive empiric therapeutic antibiotics for a febrile episode. In that situation, the empiric therapeutic antibiotics (e.g., ticarcillin) suppress the anaerobes. 2 As expected, the incidence of infection was higher at granulocyte levels of less than 500/p1 (20 of 27 episodes, or 74% occurred at this level)

and increased sharply at granulocyte counts of less than 100/p.l (14 episodes occurred at this level, including 5 of 7 bacteremias). The incidence of infection was significantly reduced for patients receiving TMP/SMZ: there were 16 infectious episodes in the placebo group compared to 4 infections among patients who received TMP/SMZ at granulocyte counts less than 500/ A1.This reduction was especially noticeable when the number of bacteremias in the two groups was compared at this granulocyte level: 5 in the placebo group versus 1 in the TMP/SMZ group. With regard to the site of documented infections, a reduction in the number of alimentary canal infections was observed with TMP/SMZ (12 versus 2) as well as reduction in the incidence of pneumonias: 5 occurred in the placebo group (4 of which were microbiologically documented, 3 of them with positive blood cultures), compared to 2 in the TMP/SMZ group (both clinically documented). This reduction in infection along the upper alimentary canal and the respiratory tract might be due to the suppression of oral flora or may be related to the systemic potential of TMP/SMZ. As a result of this decreased incidence of documented infectious episodes, patients receiving TMP/SMZ spent less time in the hospital receiving broad-spectrum antibiotics. It must be noted that there is not uniformity of


306

DE JONGH ET AL.

opinion that the use of TMP/SMZ or other similar agents is effective as infection prophylaxis.23 Some studies that have utilized a control receiving either placebo or nothing have shown excellent reductions in infection frequency. 7-9,24 In general, this has been observed only among patients with an extended episode of profound (< 100/p1) granulocytopenia, i.e., those patients at substantial risk of infection (Fig. 1). One evaluation, which also included erythromycin, noted efficacy only if patient compliance was excellent. 25 Others have observed little or no efficacy. 26 Of note is the EORTC prophylactic trial (a portion of our patients were included) which, except among these patients with lung cancer, could not demonstrate a convincing reduction in infection incidence. 27 In many of these patients, however, chemotherapy was designed to produce only a very transient leukopenia. Two additional studies from our institution compared TMP/SMZ to oral nonabsorbable antibiotics or to nalidixic acid among patients with leukemia. 28,29 In those trials, TMP/SMZ fared well, although there was no placebo control group. Some reports are available of possible hematologic toxicity related to TMP/SMZ among patients with acute leukemia, resulting in prolongation in the duration of granulocytopenia. 25,26. 30 No evidence of such an effect was present among our patients either in prolonged granulocytopenia or in a need for greater dosage reduction of chemotherapy. This may be related to a lower dose of the TMP/SMZ combination than that used in other studies or to the fact that these patients do not have a primary bone marrow disease. The chemotherapy of these patients, although very intensive, was not designed to lead to marrow aplasia as in patients with acute leukemia. One patient receiving TMP/SMZ developed a microbiologically documented cellulitis and bac-

teremia due to K. pneumoniae resistant to the prophylactic antibiotics. This episode plus others reported in the literature24',28- 30 underline a most important potential side effect of the use of TMP/ SMZ as prophylaxis, i.e., the finding of resistant strains among normally susceptible bacteria. A recent report3" indicates that resistance in some cases may be due to a plasmid that also codes for resistance to agents often used to systemic therapy of febrile episodes. It is therefore evident that close microbiologic evaluation with surveillance cultures should be maintained when this or other agents are to be used in a prophylactic role. In conclusion, the use of oral TMP/SMZ as a prophylactic antibiotic regimen among patients with small cell carcinoma of the lung receiving aggressive induction chemotherapy reduced the incidence of infection. As a consequence, the need for prolonged hospitalization and treatment with parenteral broad-spectrum antibiotics was lowered, improving the quality of life of these patients. The regimen was simple, was administered easily, had minimal side effects, and had no significant toxicities. Although these results are encouraging, only relatively few patients have been evaluated thus far; a number of evaluations in other patient populations, especially those with leukemia, do not substantiate consistently beneficial results, and resistance to TMP/SMZ has been noted to develop. Therefore, despite our positive results, more studies are needed before this or any other agent can be recommended as a definitive approach to infection prophylaxis during neutropenia.

ACKNOWLEDGMENT The authors thank Ms. Paula Salvatore for preparation of the manuscript, Margaret Whitacre, R.N., for technical assistance, and Margaret Wesley, Ph.D., for the statistical analysis.

REFERENCES 1. Greco FA, Oldham RK: Current concepts in cancer: Small cell lung cancer. N Engl J Med 301:355-358, 1979 2. Aisner J, Whitacre M, VanEcho DA, et al: Combination chemotherapy for small cell lung cancer: Continuous versus alternating noncrossresistant combination. Cancer Treat Rep 66: 221-230, 1982 3. Einhorn LH, Bond WH, Hornback M, et al: Long term results in combined modality treatment of small cell carcinoma of the lung. Semin Oncol 5:309-313, 1978 4. Ginsberg SJ, Couris RL, Gottlieb AJ, et al: Longterm

survivorship in small anaplastic lung carcinoma. Cancer Treat Rep 63:1347-1349, 1979 5. Schimpff SC, Young VM, Greene WH, et al: Origin of infection in acute nonlymphocytic leukemia: Significance of hospital acquisition of potential pathogens. Ann Intern Med 77:707-714, 1972 6. Schimpff SC: Infection prevention during profound granulocytopenia: New approaches to alimentary canal microbial suppression. Ann Intern Med 93:358-361, 1980 7. Hughes WT, Kuhn S, Chaudhary S, et al: Successful


307

SELECTIVE SUPPRESSION FOR PROPHYLAXIS chemoprophylaxis of pneumocystis carinii pneumonitis. N Engl J Med 297:1419-1426, 1977 8. Gurwith MJ, Brunton JC, Lank BA, et al: A prospective controlled investigation of prophylactic trimethoprim/sulfamethoxazole in hospitalized granulocytopenic patients. Am J Med 66:248-255, 1979 9. Sleijfer DT, Mulder NH, deVries-Hospers HG, et al: Infection prevention in granulocytopenic patients by selective decontamination of the digestive tract. Eur J Cancer 16:859-869, 1980 10. Aisner J, Whitacre M, VanEcho DA, et al: Doxorubicin, cyclophosphamide and VP16-213 (ACE) in the treatment of small cell lung cancer. Cancer Chemother Pharmacol 7:187193, 1982 11. Schimpff SC, Satterlee W, Young VM, et al: Empiric therapy with carbenicillin and gentamicin for febrile patients with cancer and granulocytopenia. N Engl J Med 284:10611065, 1971 12. Hargadon MT, Young VM, Schimpff SC, et al: Selective suppression of alimentary tract microbial flora as prophylaxis during granulocytopenia. Antimicrob Agents Chemother 20: 620-624, 1981 13. Cohen MH, Ihde DC, Bunn PA, et al: Cyclic alternating combination chemotherapy of small cell carcinoma. Cancer Treat Rep 63:163-170, 1979 14. Greco FA, Richardson RC, Snull JD, et al: Small cell cancer: Complete remission and improved survival. Am J Med 66:625-630, 1979 15. Hansen HH, Dombernowsky P, Hansen M, et al: Chemotherapy of advanced small cell anaplastic carcinoma: Superiority of a four drug combination to a three drug combination. Ann Intern Med 89:177-181, 1978 16. Hansen HH, Selawry OS, Simon R, et al. Combination chemotherapy of advanced lung cancer: A randomized trial. Cancer 38:2201-2207, 1976 17. Cohen MH, Creaven PJ, Fossieck BE, et al: Intensive chemotherapy of small cell bronchogenic carcinoma. Cancer Treat Rep 61:349-354, 1977 18. Aisner J, Alberto P, Bitran J, et al: Role of chemotherapy in small cell lung cancer: A concensus report of the International Association for the Study of Lung Cancer Workshop. Cancer Treat Rep 67:1-7, 1983 19. Schimpff SC, Greene WH, Young VM, et al: Infection prevention in acute nonlymphocytic leukemia: Laminar air flow room reverse isolation with oral nonabsorbable antibiotic prophylaxis. Ann Intern Med 82:351-358, 1975

20. Storring RA, Jameson B, McElwain TJ, et al: Oral nonabsorbed antibiotics prevent infection in acute nonlymphoblastic leukemia. Lancet 2:837-840, 1977 21. van der Waaij D, Berghuis J, Lekkerkerk JEC: Colonization resistance of the digestive tract of mice during systemic antibiotic treatment. J Hyg (Camb) 70:605-610, 1972 22. van der Waaij D, Berghuis J: Selective elimination of Enterobacteriaceae species from the digestive tract in mice and monkeys. J Hyg (Camb) 72:205, 1974 23. Pizzo PA, Schimpff SC. Strategies for the prevention of infection in the myelosuppressed or immunosuppressed cancer patient. Cancer Treat Rep (in press) 24. Dekker AW, Rozenberg-Arska M, Sixma JJ, et al: Prevention of infection by trimethoprin-sulfamethoxazole plus amphotericin B in patients with acute nonlymphocytic leukemia. Ann Intern Med 95:555-559, 1981 25. Pizzo PA, Lange M, Fialk MA: Oral antibiotic prophylaxis in cancer patients: A double blind randomized placebo controlled trial. J Pediatr (in press) 26. Weiser B, et al: Prophylactic trimethoprim/sulfamethoxazole during consolidation chemotherapy for acute leukemia: A controlled trial. Ann Intern Med 95:436-508, 1981 27. Zinner S, Gaya H, Glauser M: Co-trimoxazole and reduction of risk of infection in neutropenic patients. A progress report for the EORTC Antimicrobial Therapy Project Group. 21st Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, November 4-6, 1981, abstract 795 28. Wade JC, Schimpff SC, Hargadon MT, et al: A comparison of trimethoprim/sulfamethoxazole plus nystatin with gentamicin plus nystatin in the prevention of infection in acute leukemia. N Engl J Med 304:1057-1062, 1981 29. Wade J, de Jongh C, Newman K, et al: A comparison of trimethoprim/sulfamethoxazole to nalidixic acid: Selective decontamination as infection prophylaxis during granulocytopenia. 21st Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, November 4-6, 1981, abstract 796 30. de Jongh CA, Schimpff SC, Wiemik PH: Antibiotic prophylaxis in acute leukemia (letter). Ann Intern Med 95:783784, 1981 31. Wilson JM, Guiney DG: Failure of oral trimethoprim/ sulfamethoxazole prophylaxis in acute leukemia. Isolation of resistant plasmids from strains of Enterobacteriaceae causing bacteremia. N Engl J Med 306:16-20, 1982
