Cryptosporidium parvum is an extremely common cause of diarrhea in immunocompromised patients with AIDS. While infection in immunocompetant patients is usually self-limited and spontaneously clears after several days (1), patients with compromised immune systems may develop severe diarrhea, leading to volume depletion, malabsorption, weight loss, and sometimes death.
Currently, there is no established treatment for diarrhea caused by C. parvum. Limited benefit may be derived in certain patients from paromomycin (2,3) or azithromycin (4). However, despite numerous clinical trials with various antibiotics, no therapy has been shown to be curative. Therefore, patients are often maintained on high-dose, nonspecific antidiarrheal agents, such as loperamide or octreotide.
Another approach to treatment is oral passive immunotherapy with bovine-derived antibodies to C. parvum. Since C. parvum is a common bovine infection, cows are naturally immunized against this parasite. Use of specific active immunization techniques can elevate these antibody titers. Thus, postpartum milk (colostrum) from cows immunized with C. parvum oocysts contains high antibody titers to C. parvum. In vitro and animal studies suggest that colostrum from cows immunized with C. parvum oocyts is highly effective in inhibiting C. parvum infection (5-8). Several prior small clinical studies suggest a possible role for passive immunotherapy with bovine colostrum (9-12). Therefore, we evaluated the safety and efficacy of oral passive immunotherapy with immunized bovine colostrum in 24 patients with AIDS and severe diarrhea.
MATERIALS AND METHODS
The study protocol was approved by the Institutional Review Board at the University of California, San Francisco. All patients had the opportunity to ask questions about the study and signed informed consent was obtained at the time of enrollment.
A total of 24 patients were recruited into three cohorts: (1) C. parvum infection alone (n = 16), (2) C. parvum and a second opportunistic infection (n = 4), and (3) idiopathic AIDS enteropathy with no identified source of infection (n = 3) or an untreatable opportunistic infection other than C. parvum (n = 1). All patients had long-standing and severe diarrhea with a minimum mean daily stool weight of 500 g.
Evaluation included stool specimens analyzed for ova and parasites (minimum of two), routine bacterial culture, stains for acid-fast organisms, and toxin to Clostridium difficile. All patients underwent upper endoscopy to the level of the second portion of the duodenum and complete colonoscopy. In all patients, biopsy specimens were taken from the duodenum and throughout the large intestine. Intestinal areas with endoscopic abnormalities were also biopsied. Terminal ileal biopsy specimens were taken when possible. Routine laboratory evaluations were performed at study enrollment, during each week of treatment, and at the end of this study.
Inclusion criteria for enrollment included documented HIV infection, age of ≥18 years, and severe diarrhea, as defined by symptoms for at least 2 weeks with a minimum mean daily stool weight of 500 g while on a stable dose of antidiarrheal agents. Patients were asked to continue their usual dose of antidiarrheal medicines throughout this study. Patients were recruited into three separate cohorts. Patients in cohort 1 had documented C. parvum by either stool microscopy or endoscopic biopsy or both, but no evidence for any other established diarrheal pathogen. In cohort 2, patients were infected with C. parvum and a second agent known to cause diarrhea in immunocompromised patients (microsporidia in one and cytomegalovirus inclusion bodies on biopsy without gross enteritis in three). Patients in cohort 3 had either no known source of diarrhea (three patients) or an untreatable pathogen (microsporidia in one patient). The criteria for exclusion included life expectancy of <3 months, cytomegalovirus enteritis with demonstrated gastrointestinal ulcers, diarrhea caused by any pathogen with well-established treatment, milk allergy or lactose intolerance, pregnancy, hepatic or renal failure, pancreatitis, or known biliary tract infection with C. parvum.
Treatment was administered for 21 consecutive days and patients were followed for a total of 7 weeks. Patients in cohort 1 were randomized to receive the medication in powder (n = 8) or enteric-coated capsules (n = 8). All patients in cohorts 2 and 3 received the medication in powder form. The powder form of the drug was administered in 10-g dosages, mixed with 4 oz (0.1 L) of water, and given 4 times a day after meals. Patients in cohort 1, who received the medication in capsule form, took 12 capsules, 4 times a day after meals. The medication in capsule form contained 2 g bovine immunoglobulin concentrate per dosage, which was determined to be bioequivalent to 10 g free powder, based on bioavailability studies performed in normal human volunteer studies.
Study End Points
The primary end point was change in stool weight. During specified 3-day periods at study entry and 7 weeks, patients weighed all bowel movements with the aid of a disposable container and scale. A third stool weight analysis was added midway through this study at week 3 (the end of treatment). Secondary end points included change in stool frequency and body weight, and clearance of C. parvum oocyts from stool specimens. Patients recorded all bowel movements in a diary during this 7-week study. Body weight at study entry was compared to that at study termination.
Bovine immunoglobulins (GalaGen, Arden Hills, MN, U.S.A.) were concentrated from colostrum obtained during the first six postpartum milkings of cows immunized with a preparation of purified C. parvum oocysts and stored at -20°C. Thawed colostrum was later pooled and skimmed. The skim was acidified to precipitate the casein, which was removed by centrifugation. The supernatant was concentrated, heat treated, and spray dried to produce the bulk powder product. The specific anti-C. parvum activity in the immune product was 1:6,400, which is at least 16- to 20-fold greater than activity in a nonimmune colostrum-derived product.
Absolute means were compared and, in addition, paired data were analyzed by t test before and after treatment. All p values of <0.05 were considered significant, and all tests were two-tailed.
Patients were well matched as to baseline characteristics (Table 1). Only mean serum cholesterol was significantly higher in patients taking capsules (cohort 1) than in the other cohorts. All patients were severely immunocompromised, as demonstrated by a mean (±SEM) CD4 count of 34.5 ± 7.0 cells/mm3 (range, 0-150 cells/mm3). Diarrhea was chronic and severe, occurring 6.9 ± 0.6 times daily, with a mean weight of 1,278 ± 143 g/day. Patients reported a mean body weight loss of 3.4 ± 1.3 kg prior to study entry. Of 20 patients with C. parvum (in cohorts 1 and 2), 16 (80%) were diagnosed with C. parvum by the presence of oocysts in at least one stool sample, whereas the remaining patients were diagnosed with C. parvum on the basis of endoscopic biopsies.
Patients with C. parvum infection alone (cohort 1), who were randomized to receive bovine immunoglobulin concentrate in powder form, had a decrease in stool weight (Fig. 1). Mean daily stool weight in this cohort was 1,125 ± 170 g at baseline, 518 ± 75 g (p = 0.30) at the end of treatment, and 723 ± 158 g at study termination (p = 0.07). Three of seven evaluable patients had at least a 50% decrease in stool weight at week 7, with improvements of 57%, 59%, and 67%, respectively, compared to baseline. One patient in this cohort had a 30% increase in stool weight. Mean stool frequency (Fig. 2) decreased from 6.3 ± 0.8 bowel movements per day at baseline, to 4.9 ± 0.7 during weeks 1-3 (p = 0.09) and 4.7 ± 0.9 during weeks 4-7 (p = 0.21).
In contrast, patients with C. parvum infection alone, who received bovine immunoglobulin concentrate in capsule form, had no benefit from treatment. Mean daily stool weights (Fig. 1) at baseline, week 3, and week 7 were 1,439 ± 315 g, 1,437 ± 683 g, and 1,525 ± 552 g, respectively [p = NS (not significant) at weeks 3 and 7, compared to baseline]. However, two of seven evaluable patients had a reduction of >50% in stool weight at week 7. Three patients had an increase in stool weight. Mean stool frequency (Fig. 2) decreased from 8.1 ± 1.4 bowel movements daily at baseline, to 6.9 ± 2.6 at weeks 1-3 (p = 0.28) and 6.0 ± 1.2 during weeks 4-7 (p = 0.12).
For the 12 patients in cohorts 1 and 2 with C. parvum who received bovine immunoglobulin concentrate in powder form, mean daily stool weight at study entry was 1,158 ± 114 g. As shown in Fig. 1, mean stool weight decreased significantly to 595 ± 63 g/day (p = 0.04) by week 3 and to 749 ± 123 g by week 7 (p = 0.03). However, only three of nine evaluable patients had a reduction of at least 50% in stool weight at week 7, while 5 others had smaller reductions in stool weight. One patient (mentioned above) had an increase in stool weight. Mean stool frequency (Fig. 2) decreased from 6.6 ± 0.6 per day at study entry, to 5.4 ± 0.7 during weeks 1-3 (p = 0.04) and 5.4 ± 0.9 during weeks 4-7 (p = 0.12).
The three patients with idiopathic HIV-related diarrhea and the one patient with microsporidial disease (cohort 3) did not benefit from treatment with bovine immunoglobulin concentrate. At study entry, week 3, and week 7, mean daily stool weights (Fig. 1) were 1,318 ± 553 g, 1,484 g (only one measurement), and 1,257 ± 710 g, respectively (p = NS, baseline compared with week 7). Stool frequency (Fig. 2) was 5.2 ± 1.4 at baseline, 5.3 ± 1.3 during treatment, and 5.4 ± 2.0 during weeks 4-7 (p = NS, baseline compared to weeks 1-3 and to weeks 4-7).
In the 20 patients with documented C. parvum (cohorts 1 and 2 combined), a total of 66 stool samples were submitted for microscopy prior to treatment. In 35 (53%) of 66 specimens, C. parvum was demonstrated by standard microscopic analysis. Analysis of stool specimens collected from these patients during and after treatment with bovine immunoglobulin concentrate demonstrated C. parvum in 25 (37%) of 68 (p = 0.09).
Despite reported weight loss prior to the start of this study, body weight stabilized over the 7-week course of this study. Mean body weight increased by 1.6 kg compared to baseline in patients with C. parvum taking bovine immunoglobulins in powder form. In contrast, patients with C. parvum who received the medication in capsule form lost a mean of 0.3 kg (p = 0.23). Patients in cohort 3 (without C. parvum) gained a mean of 0.1 kg over baseline measurements (p = 0.59, compared to cohort-1 patients taking powder).
Overall, bovine immunoglobulin concentrate was well tolerated. However, one patient treated with bovine immunoglobulin concentrate in powder form developed severe nausea and vomiting, necessitating early study termination after only 1 week of treatment. No other patient developed symptoms that could be related to the medication. One patient died of lymphoma 3 months after starting the study.
Treatment was associated with a statistically significant increase in liver function tests, as shown in Fig. 3. Mean serum alanine aminotransaminase (p = 0.007) and aspartate aminotransaminase (p = 0.06) were higher during the study as compared to baseline values. This effect appeared to be greater in patients taking capsules than in those on powder (p = NS). Mean hematocrit was also significantly lower during treatment than at study entry (p = 0.02; Fig. 3). However, there was no significant change in other routine laboratory measurements.
Overall, compliance with the study medication was excellent. All but two patients completed their stool diary during the 3 weeks of active treatment. However, 10 of 24 patients did not fully complete the diary during the observation period, after active treatment had been completed. Stool weights were generally completed as required by this study, except when intercurrent illness interfered with the measurements.
All 24 patients enrolled in this study were severely immunocompromised and had chronic diarrhea. Patients with C. parvum, who were treated with bovine immunoglobulin concentrate in powder form, experienced a significant reduction in stool weight and frequency. In contrast, there was no apparent benefit from bovine immunoglobulin concentrate in capsule form. The lack of efficacy in patients receiving capsules may have been due to the smaller dosage of bovine immunoglobulin concentrate. Alternatively, decreased bioavailability from the capsules may have occurred, as several patients noted intact capsules in emesis or stool.
In patients with C. parvum who received bovine immunoglobulins in powder form, mean body weight stabilized during treatment. In contrast, patients receiving capsules lost body weight, although this difference was not statistically significant. In addition, C. parvum was detected in fewer stool samples after treatment, as compared to baseline (p = NS). Although these data are intriguing, the lack of statistical significance precludes firm conclusions.
Patients with idiopathic- and non-C. parvum-related diarrhea (cohort 3) experienced no benefit from bovine immunoglobulin concentrate. The lack of improvement in this control cohort strongly suggests that the mechanism of bovine immunoglobulin concentrate is not a nonspecific antidiarrheal effect. Instead, because improvement was restricted to patients with C. parvum, bovine immunoglobulin concentrate appears to have specific activity against C. parvum. The identification of the active component will be important to understanding the pathophysiology and mechanism of diarrhea in patients infected with C. parvum.
Prior published studies of passive immunotherapy with bovine colostrum have been conducted with limited numbers of patients (12). Two case reports illustrated dramatic improvement with hyperimmune bovine colostrum (9,11). A double-blind study of five patients, however, demonstrated greater improvement in patients treated with placebo than in those treated with active hyperimmunized bovine colostrum. However, it appeared that patients treated with active medication had a reduction in oocyst load that was not experienced by those receiving placebo (10). Another study demonstrated no improvement in three patients with C. parvum(13). However, because colostrum preparations are produced differently, comparisons between studies are difficult. In addition, in the present study, we treated patients for 21 days, which is substantially longer than treatment in prior studies.
In our patients with C. parvum, treated with bovine immunoglobulin concentrate in powder form, we demonstrated a 49% reduction in mean stool weight after 21 continuous days of therapy. The response was somewhat attenuated at the end of the study, after patients had been observed off treatment for 4 weeks. Therefore, longer or maintenance treatment may be needed to optimize the response, eradicate C. parvum, and prevent autoinfection (14).
Bovine immunoglobulin concentrate was well tolerated with the exception of one patient who experienced self-limited nausea and vomiting. While elevations in liver function tests and mild decrease in serum hematocrit were noted after drug administration, the significance of these observations is unclear. There was no known morbidity related to the increase in liver function tests.
This pilot study is the largest series published to date on the efficacy of bovine immunoglobulins in the treatment of C. parvum. This study was designed as an open-label study to test the efficacy of bovine immunoglobulin concentrate in powder and capsule forms in patients with and without C. parvum. Attempts were made to minimize the potential for bias by using objective end points. Unfortunately, follow-up, which was nearly universal during the 3 weeks of active treatment, was incomplete during the 4-week observation period after completion of treatment.
Because there is currently no known curative treatment for C. parvum, any safe agent that is at least partly effective may have a role in the treatment of this parasite. Bovine immunoglobulin concentrate appears to be moderately effective when given in powder form to patients with C. parvum. Thus, further study of bovine immunoglobulin concentrate is warranted and should be evaluated in a double-blind, placebo-controlled trial.
Acknowledgment: This work was funded in part by GalaGen, Incorporated (Arden Hills, MN, U.S.A.).
1. DuPont HL, Chappell CL, Sterling CR, Okhuysen PC, Rose JB, Jakuboswki W. The infectivity of Cryptosporidium parvum
in healthy volunteers. N Engl J Med
2. Bissuel F, Cotte L, Rabodonirina M, Rougier P, Piens MA, Trepo C. Paromomycin: an effective treatment for cryptosporidial diarrhea
in patients with AIDS. Clin Infect Dis
3. White AC, Chappell CL, Hayat CS, Kimball KT, Flanigan TP, Goodgame RW. Paromomycin treatment for cryptosporidiosis in AIDS: a prospective, double-blind trial. J Infect Dis
4. Ritchie DJ, Becker ES. Update on the management of intestinal cryptosporidiosis in AIDS. Ann Pharmacother
5. Flanigan T, Marshall R, Redman D, Kaetzel C, Ungar B. In vitro screening of therapeutic agents against Cryptosporidium:
hyperimmune cow colostrum is highly inhibitory. J Protozool
6. Doyle PS, Crabb J, Petersen C. Anti-Cryptosporidium parvum
antibodies inhibit infectivity In vitro and in vivo. Infect Immun
7. Hoskins D, Chrisp CE, Suckow MA, Fayer R. Effect of hyperimmune bovine colostrum raised against Cryptosporidium parvum
on infection of guinea pigs by Cryptosporidium wrairi. J Protozool
8. Riggs MW, Cama VA, Leary HL, Sterling CR. Bovine antibody against Cryptosporidium parvum
elicits a circumsporozoite precipitate-like reaction and has immunotherapeutic effect against persistent cryptosporidiosis in SCID mice. Infect Immun
9. Ungar BL, Ward DJ, Fayer R, Quinn CA. Cessation of Cryptosporidium
in an acquired immunodeficiency syndrome patient after treatment with hyperimmune bovine colostrum. Gastroenterology
10. Nord J, Ma P, DiJohn D, Tzipori S, Tacket CO. Treatment with bovine hyperimmune colostrum of cryptosporidial diarrhea
in AIDS patients. AIDS
11. Tzipori S, Roberton D, Chapman C. Remission of diarrhoea due to cryptosporidiosis in an immunodeficient child treated with hyperimmune bovine colostrum. BMJ
12. Shield J, Melville C, Novelli V, et al. Bovine colostrum immunoglobulin concentrate for cryptosporidiosis in AIDS. Arch Dis Child
13. Saxon A, Weinstein W. Oral administration of bovine colostrum anti-cryptosporidia antibody fails to alter the course of human cryptosporidiosis. J Parasitol
14. Petersen C, Gut J, Doyle PS, Crabb JH, Nelson RG, Leech JH. Characterization of a >900,000-MrCryptosporidium parvum
sporozoite glycoprotein recognized by protective hyperimmune bovine colostral immunoglobulin. Infect Immun