|
| |||||||||||||||||||||||||||||||
Background Although methods of viral attenuation in plasma-derived clotting-factor concentrates have improved, there is still a possibility that such concentrates may transmit certain blood-borne viruses. For this reason, the use of recombinant DNA-derived factor VIII (which is virus-free) to treat hemophilia A has generated considerable interest.
Methods We conducted a multicenter trial in previously untreated children with hemophilia A. They received recombinant factor VIII for all treatment or for prophylaxis and were evaluated at their respective clinics at intervals of no more than three months.
Results Between January 1, 1989, and July 1, 1992, 95 patients who could be evaluated received recombinant factor VIII. By September 1, 1992, they had received the concentrate exclusively for 2.4 months to 3.5 years (median, 1.5 years). All responded well, with no treatment failures. A total of 3315 infusions were administered; there were three reports of minor adverse reactions. Inhibitor antibodies to factor VIII developed in 16 of 81 patients tested for them, after a median of nine days of exposure to factor VIII treatment. Inhibitor titers were or became low in 9 of the 16 patients despite continued episodic treatment with the concentrate. Inhibitors disappeared completely in 4 patients and remained at a low level (<10 Bethesda units) in 5 patients receiving episodic treatment.
Conclusions Transient or low levels of inhibitor, as observed in this study, may represent part of the natural history of hemophilia in infants. In view of the transient nature and lower concentration of the inhibitors detected and the generally satisfactory response to treatment, the benefits of recombinant factor VIII for the treatment of hemophilia seem to outweigh the risks.
After the safety and efficacy of recombinant DNA-derived factor VIII concentrate were demonstrated in patients previously treated for hemophilia,7,8 a multicenter trial in previously untreated patients was begun in January 1989. This report describes the study design and the observations made in the first 3.5 years of this ongoing study. The first 20 of these patients were described in the Journal in 19908; since then, an additional 75 patients have been treated, and the entire group has been followed for a much longer period. A noteworthy finding is the early appearance of factor VIII inhibitors in 16 previously untreated patients; in 9 of the 16, the inhibitors have been present only in low concentrations or have disappeared with time despite continued episodic treatment with recombinant factor VIII.
Methods
Patients
All the patients were children or infants with hemophilia A; none had previously received clotting-factor concentrates. Informed written consent was obtained from the parent or guardian of each patient before study. A family history of factor VIII inhibitors was not a reason for exclusion. Between January 1, 1989, and July 1, 1992, 125 patients were recruited for the study, of whom 98 were actually treated with recombinant factor VIII. Two patients who inadvertently received plasma-derived factor VIII concentrate and one who received a massive blood transfusion shortly after the study started were not considered suitable for evaluation and were thus withdrawn from the study. Of the 95 patients who could be evaluated, 93 had no previous exposure to any plasma-derived factor VIII and 2 had received minimal treatment (each had received one treatment with fresh-frozen plasma or cryoprecipitate at birth). The degree of hemophilia A was severe or moderately severe in 59 patients (factor VIII concentration, <0.02 U per milliliter, or <2 percent), moderate in 18 (factor VIII concentration, 0.02 to 0.05 U per milliliter, or 2 to 5 percent), and mild in 18 (factor VIII concentration, >0.05 U per milliliter, or >5 percent). The data reported here were collected before September 1, 1992.
Study Design
A base-line blood sample was drawn from each patient. If the patient had not been immunized with hepatitis B vaccine, immunization was begun (in case red cells or other human blood products were required later). Once a patient received recombinant factor VIII (day 0), he returned to his respective clinic at least once every three months for follow-up evaluation, which included factor VIII-inhibitor assays (Bethesda assay) and antibody screening by enzyme-linked immunosorbent assay (at Miles Inc., Berkeley, Calif.) as well as hematologic and chemical tests for safety. The patients received only recombinant factor VIII (Kogenate, Miles) for the treatment or prophylaxis of bleeding episodes. Details of the preparation of the concentrate have been summarized previously8; the cell line used in its manufacture has been extensively characterized9 and has not been found to contain endogenous infectious agents. In addition, the purification process includes a combination of steps for clearance and inactivation that have been shown to be very effective in removing model viruses10.
Assays for factor VIII-inhibitor antibody were performed every three months after the first treatment with the concentrate. If a patient was found to have an inhibitor antibody, he was recalled for repeat testing, which included an assay at the laboratory of the hemophilia center and the study's reference laboratory (University of California, Davis). If the level of the inhibitor was confirmed as being 0.6 Bethesda unit or greater, factor VIII recovery was determined by sampling before and 10 minutes after the administration of a challenge dose of the concentrate (50 IU per kilogram of body weight). Recovery was expressed as the increase in the plasma factor VIII level above base line after the challenge (the plasma factor VIII level 10 minutes after infusion, minus the base-line level, expressed as a percentage of the normal level) per unit of the challenge dose per kilogram. If an inhibitor was identified, subsequent assays were performed more frequently.
All data collected were reviewed every six months by an oversight committee comprising six clinical experts not employed by the sponsoring manufacturer. This committee was notified of any adverse events.
The management of inhibitor formation was left to the discretion of the treating physician. Thus, patients with inhibitors could continue to receive the concentrate (in the usual dose, an increased dose, or a high dose to induce immune tolerance) or could be withdrawn from study and treated with another product.
Laboratory Tests
Plasma samples were processed, stored, and assayed for factor VIII as previously described8. Assays for factor VIII inhibitor were carried out at the coagulation laboratory of each participating institution according to the Bethesda method11. Any plasma sample found to have an inhibitor level of 0.6 Bethesda unit or more was considered to be positive for factor VIII inhibitor. Such samples were also sent to the study's reference laboratory for repeat assay for antibodies to factor VIII derived from human plasma, as well as for inhibitor to recombinant factor VIII.
Plasma aliquots from selected patients with inhibitors were also sent to another laboratory (American Red Cross, Rockville, Md.) to determine the heavy-chain and light-chain specificity of the antibody by immunoblotting12,13.
Levels of antibodies to murine IgG, baby-hamster-kidney protein, and recombinant factor VIII were measured by enzyme-linked immunosorbent assay (Miles), and the results were reported as negative, borderline, or positive (expressed as 1+, 2+, or 3+).
Statistical Analysis
The method of Kaplan and Meier14 was used to calculate the cumulative risk that an inhibitor would develop after the first factor VIII treatment; the cumulative risk was plotted in relation to the number of days of exposure to factor VIII. The Wilcoxon rank-sum test15 was used to compare the number of days of exposure in the patients in whom inhibitor developed with the number in the patients in whom inhibitor did not develop.
Results
As of July 1, 1992, 125 previously untreated patients had been recruited for participation in this trial; 98 were treated with recombinant factor VIII at least once. Ninety-five of these 98 patients could be evaluated. The median age at the first treatment with the concentrate in the 59 patients with severe or moderately severe hemophilia was 9.1 months (range, 0 to 320); in the 18 with moderate hemophilia, 14.7 months (range, 0.1 to 551); and in the 18 with mild hemophilia, 47 months (range, 4.5 to 658). Figure 1 shows the duration of study in each patient; time 0 represents the day of initial treatment with the factor VIII concentrate. To date, the patients have received the concentrate exclusively for 2.4 months to 3.5 years (median, 1.5 years), and all have responded well (the concentrate was given for spontaneous or traumatic bleeding or invasive procedures). A total of 3315 infusions were administered to the 95 patients who could be evaluated. The concentrate was well tolerated; there were three reports of adverse reactions, all minor, including one report each of urticaria, flushing, and erythema at the infusion site. Follow-up observation revealed no elevations in alanine aminotransferase concentrations, no appearance of antibody to hamster or murine protein, and no evidence of any blood-borne viral infection or any other unusual infection or illness.
|
|
|
|
10 Bethesda units), in vivo studies demonstrated that the recovery of factor VIII levels was reduced to 0.25 percent to 1 percent per international unit per kilogram (normal,
2 percent8). All five patients have continued to respond well to factor VIII administered for the treatment of bleeding episodes; most of these patients did not require an increased dose. Figure 3 shows the course of a patient (Patient 3) in whom the inhibitor disappeared despite continued episodic treatment with factor VIII, and the course of another (Patient 1) whose inhibitor titer initially peaked at 15 Bethesda units but has subsequently declined to less than 5 Bethesda units with continuing treatment. Eight patients had somewhat higher concentrations of factor VIII inhibitors that would meet the criteria for high titers (>10 Bethesda units). One of these eight (Patient 4) had an inhibitor concentration of more than 300 Bethesda units after three days of exposure to the concentrate and was withdrawn from the study. He is now being treated with an activated factor IX-complex concentrate (Feiba, Immuno, Rochester, Mich.). Four patients (Patients 5, 9, 11, and 12) were placed on regimens to induce immune tolerance (by their physicians' choice) with high doses of recombinant factor VIII16,17. Since a standardized protocol for immune tolerance was not available until early 1992, each patient was treated somewhat differently. Patients 5 and 12 underwent the induction of immune tolerance with a dose of 50 IU per kilogram per day, Patient 11 received 100 IU per kilogram per day, and Patient 9 received 50 IU per kilogram every other day. Patients 5, 9, and 11, who had peak inhibitor titers of 19, 19.5, and 34 Bethesda units, respectively, had excellent responses to the induction of immune tolerance, with complete disappearance of their inhibitors 13, 381, and 35 days, respectively, after the regimen began. The fourth, Patient 12, has had no serious bleeding episodes, despite a poor response to the immune-tolerance regimen. His inhibitor concentration was more than 300 Bethesda units when the immune-tolerance regimen began, and he had a strong family history of high titers of inhibitors to factor VIII (>3000 Bethesda units).
|
At present, 15 of the 16 patients with inhibitor formation are continuing to receive treatment with recombinant factor VIII; 11 receive it "on demand" (as necessary), and 4 according to immune-tolerance regimens.
To date, plasma samples from three of the previously untreated patients with higher titers of inhibitors have been analyzed for the heavy-chain and light-chain specificity of factor VIII antibody. All three inhibitors were reactive, showing both 44-kd and 72-kd bands on Western blotting that were consistent with reactivity with both heavy-chain and light-chain determinants of factor VIII (Hoyer L: personal communication). This pattern has been observed in inhibitors that develop after exposure to plasma-derived factor VIII products12,13,18,19. When studied with the Bethesda assay, the inhibitors had similar titers against both recombinant and plasma-derived factor VIII (data not shown).
Discussion
During the first 3.5 years of observation in this ongoing study, 95 patients with hemophilia A were treated with recombinant factor VIII. The majority have had excellent clinical responses. No patient had evidence of any unusual infection or illness. Factor VIII inhibitors developed in 16 patients, usually after relatively brief exposure to the concentrate (median, 9 days of exposure; range, 3 to 18). In nine of these patients the inhibitors were present in low titers or declined to low levels despite continued episodic treatment with recombinant factor VIII. Inhibitors disappeared completely in four patients during episodic ("on demand") treatment, while in five others the titers remained below 10 Bethesda units during this treatment.
Althoviii ugh most published reports state that inhibitor antibodies to factor VIII develop in 10 to 15 percent of persons with hemophilia A,20,21,22,23,24 these data and estimates have been derived from studies that differed from the current investigation in a number of important design considerations and risk factors. For example, the Cooperative Inhibitor Study sponsored by the National Heart, Lung, and Blood Institute reported an incidence of new inhibitor formation of eight cases per 1000 patient-years of observation, but this incidence was based on the study of previously treated patients, a group known to be at much lower risk for inhibitor formation than previously untreated patients24. Since that study did not include previously untreated patients, its results cannot be compared directly with ours.
In the current investigation, previously untreated patients were prospectively screened at three-month intervals by means of inhibitor assays, whether or not clinically indicated; inhibitors that were transient or present in low concentrations might have been missed entirely had the patients not been under such close surveillance. Although several relatively small cohorts of previously untreated patients have been enrolled in trials testing the safety of new factor VIII preparations with regard to hepatitis, the majority of these studies have been short-term trials to determine the safety of a particular product25 for hepatitis viruses (and HIV), rather than prospective studies of inhibitor development. The majority of these reports do not even mention inhibitor assays, and in many of their previously untreated patients the exposure days were few; follow-up generally lasted only six months, the period required for trials examining safety in relation to hepatitis. If the current investigation had been limited to six months, only 8 of the 16 patients with inhibitors would have been detected among the 95 patients who could be evaluated.
Thus, previous studies may have underestimated the incidence of inhibitor formation and the number of exposure days before inhibitor formation in previously untreated patients. In three of four recent studies of previously untreated patients who received plasma-derived factor VIII concentrates,26,27,28,29,30 the incidence of inhibitor formation was high, ranging from 18.4 to 52 percent. Two recent retrospective studies of patients with hemophilia who were followed from birth and treated with plasma-derived factor VIII products of intermediate purity have also reported cumulative lifelong incidences of inhibitor development of 22 percent31 and 24 percent,32 respectively. These more recent observations suggest that inhibitor antibodies may develop more frequently in patients with hemophilia A than has previously been recognized. In a 15-year study (1976 to 1991) by Ehrenforth et al.,29 inhibitors were found in 15 of 46 previously untreated patients with hemophilia A (33 percent) treated with plasma-derived factor VIII concentrates. Among those with base-line factor VIII values below 1 percent, the incidence of inhibitor formation was even greater -- 14 of 27 patients, or 52 percent. Inhibitors were detected early, between the ages of 1 month and 5 years; 11 of the 15 patients with inhibitors were identified during the first 2.6 years of life. The median number of exposure days before inhibitor formation was 11, an interval similar to the 9 exposure days in the present study. It thus appears that in previously untreated patients in whom an inhibitor will develop, inhibitor will form quite early after treatment.
There may be a "natural history" of transient, low-level inhibitor formation in infants and children with hemophilia who are treated with factor VIII. It is recognized that some inhibitors with low titers disappear,24 others remain present in low titers, and a few ultimately have high titers20. Even high titers may decline with episodic factor VIII treatment33. The pathogenesis of inhibitor formation and the reason why some inhibitors disappear are poorly understood at present.
Investigators in Belgium and the Netherlands recently reported a sudden increase in the incidence of factor VIII inhibitors among previously treated patients with hemophilia who received a new plasma-derived factor VIII concentrate produced in the Netherlands34. There is no evidence that the inhibitors observed in the current study were related to the concentrate used, but the possibility that it has an unusual antigenicity cannot be completely excluded35. However, no such antigenicity was observed during extensive preclinical investigation of the preparation,36 and only 1 of 102 previously treated patients had inhibitor formation for the first time after being treated with recombinant factor VIII8. If recombinant factor VIII were immunogenic, one would expect a much greater incidence of inhibitor formation. In our study, the three inhibitors with high titers recognized both the light and heavy chains of factor VIII, as do inhibitors in patients with hemophilia treated only with plasma-derived factor VIII12,13,18,19. These data do not suggest that recombinant factor VIII has a unique antigenicity.
In our study, factor VIII was well tolerated, and the response to treatment was excellent. We observed a relatively high prevalence of inhibitors with low titers among our patients. Surveys of populations with hemophilia for antiidiotypic antibodies, studies of the T-cell response during the first several exposures to recombinant factor VIII, and similar studies of plasma-derived factor VIII products may elucidate the mechanisms involved. Despite the generally low-titer inhibitors detected in our prospective study, the benefits of recombinant factor VIII are thought to outweigh its risks.
Supported by a grant from Miles Inc.
Source Information
From Children's Hospital of Michigan, Detroit (J.M.L.); Mount Sinai Medical School, New York (S.A.); the University of California, Davis (C.F.A.); and the Department of Clinical Research, Miles Inc., Berkeley, Calif. (R.S.S.). Presented in part as an abstract at the 33rd Annual Meeting of the American Society of Hematology, Denver, December 7-10, 1991.The members of the Kogenate Previously Untreated Patient Study Group are listed in the Appendix.
Address reprint requests to Dr. Lusher at Children's Hospital of Michigan, 3901 Beaubien Blvd., Detroit, MI 48201.
References
-propriolactone-UV-inactivated clotting factor. Thromb Haemost 1990;64:336-337. [Medline]
The following people and institutions were participants in this trial by the Kogenate Previously Untreated Patient Study Group. Children's Hospital of Michigan, Detroit: J. Lusher, I. Warrier, and L. Pfaffman; Mount Sinai School of Medicine, New York: S. Arkin, L.M. Aledort, and A. Forster; University of California, Davis: C.F. Abildgaard and J. Harrison; Albany Medical College, Albany, N.Y.: J.M. Pearce; University of Pittsburgh School of Medicine, Pittsburgh: M. Ragni, B. Miller, and E. Carfagna; Tulane Medical School, New Orleans: W.A. Andes; College of Human Medicine, Michigan State University, East Lansing: R. Kulkarni; Rehabilitation Hospital and Hemophilia Center, Rehabilitation Foundation, Heidelberg, Germany: R. Zimmerman and A. Huth-Kuhne; University of Milan, Milan, Italy: P.M. Mannucci, A. Gringeri, and L. Simoni; University of Rome "La Sapienza," Rome: F. Mandelli; University of Frankfurt, Frankfurt, Germany: I. Scharrer and E. Aygoren (Department of Internal Medicine), and W. Kreuz and R. Linde (Department of Pediatrics); Children's Hospital, Oakland, Calif.: J.E. Addiego, Jr., and J. Test; Cornell University Medical Center, New York: M.W. Hilgartner; St. Joseph's Health Centre, London, Ont., Canada: M.J. Inwood, E. Inwood, and E. Clegg; Children's Hospital, Akron, Ohio: C.E. Krill; Hershey Medical Center, Hershey, Pa.: E. Eyster and S. Neagley; State University of New York, Syracuse: T. Coyle; Yale University Medical School, New Haven, Conn.: D. Beardsley and A. Conroy; Hospital Nuovo Pellegrini, Naples, Italy: R. De Biasi; Malmo General Hospital, Malmo, Sweden: R. Ljung; Sahlgren's Hospital, Goteborg, Sweden: L. Tengborn; Children's Hospital, Barcelona, Spain: J. Tusell; Hospital "Marques de Valdecilla," Santander, Spain: A. Zubizarreta; Children's Medical Center, Dayton, Ohio: L.P. Valdez; Juan Canalejo Hospital, La Coruna, Spain: F. Batlle; University of Munich, Munich, Germany: W. Schramm; and Children's Hospital of Buffalo, Buffalo, N.Y.: G. Despande.
Data collection, clinical study coordination, and analysis: North America -- Miles Inc., Berkeley, Calif.: R.S. Schwartz, R.U. Allred, M.A. MacKenzie, P. Bailey, S. Dyla, and J.E. Pennington; Europe -- Bayer AG, Wuppertal, Germany: D. Maruhn; Tropon, Cologne, Germany: R. Neumann.
Recombinant Factor VIII Clinical Planning Committee: J. Lusher (chairperson), Detroit; P. Mannucci, Milan; M. Inwood, London, Ont., Canada; A. Bloom, Cardiff, Wales; H. Brackmann, Bonn, Germany; and H. Fukui, Nara, Japan.
| |||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |