SOLIDE TUMOREN

Boswellicofacids in the palliative therapy children with progressive

or relapsed brain tumors

nnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn 189 n

G. Janen1, U. Bode2, H. Breu3, B. Dohrn4, V. Engelbrecht5, U. Gbel1

1

Medizinische Einrichtungen der Universitt Dsseldorf, Klinik fr Pdiatrische Hmatologie und Onkologie 2 Zentrum fr Kinderheilkunde der Universitt Bonn, Abt. fr Hmatologie/Onkologie 3 Stdt. Kliniken Dortmund, Kinderklinik 4 Stdtische Kliniken Wuppertal 5 Medizinische Einrichtungen der Universitt Dsseldorf, Institut fr diagnostische Radiologie

Abstract: 19 children and adolescents with intracranial tumors received a palliative therapy with H 15 at a maximum dose of 126 mg/kg BW/day. All patients had previously been treated with conventional therapy. No side effects were observed during a median 9 months application. The recently reported antiedematous effect of H 15 was documented by MRI in one patient with a peritumoral edema, thus sparing steroid therapy with its typical side effects. Five/19 children reported an improvement of their general health status; this might be a psychological effect of hope for tumor response during palliative care. Three/17 patients with malignant tumors showed a mainly transient improvement of neurological symptoms such as pareses and ataxia. Three further patients showed an increased muscular strength and one cachectic patient achieved a weight gain. These improvements might be attributed to the antiedematous effect of H 15. Because of the palliative situation of these patients, H 15 application was performed without prior rebiopsy for histological evaluation. Overlapping effects with a previous radiotherapy or chemotherapy may have occurred. An antiproliferative effect cannot be stated. To prevent an uncritical use of H 15, further studies with prospective central documentation have to be initiated to evaluate the clinical indications for H 15 in palliative therapy, optimal dosage and duration of application.
Key words: Brain tumor boswellic acid palliative care H 15

rungen der neurologischen Symptome, wie z. B. Rckbildung von Paresen. 3/17 Patienten zeigten eine Steigerung der Muskelkraft und bei einer kachektischen Patientin wurde eine Gewichtszunahme verzeichnet. Diese transienten Verbesserungen knnen Folge eines antidematsen Effektes von H 15 sein. Eine antiproliferative Wirkung von H 15 kann nicht besttigt werden, da in der palliativen Situation dieser Patienten auf eine erneute histologische Sicherung des Tumors verzichtet wurde. Auch ein berlappender Effekt der vorher verabfolgten Chemotherapie bzw. Bestrahlung muss bercksichtigt werden. Anstelle eines unkritischen Einsatzes ist die prospektive Erfassung von Patienten mit rezidivierten oder progredienten Hirntumoren und die zentrale Dokumentation zu empfehlen, um die Indikation fr H 15 auch im Hinblick auf Dosierung und Applikationsdauer zweifelsfrei zu prfen. Schlsselwrter: Hirntumor Boswelliasure palliative Therapie H 15

Boswelliasuren in der palliativen Therapie bei Kindern mit progredienten oder rezidivierenden Hirntumoren: 19 Kinder

und Jugendliche mit intrakraniellen Tumoren erhielten in der Palliativsituation nach vorheriger konventioneller Therapie Boswellia-Suren (H 15) in einer Dosierung bis 126 mg/kg KG/d. Nebenwirkungen wurden bei einer mittleren Behandlungsdauer von 9 Monaten nicht festgestellt. Der bereits diskutierte antidematse Effekt konnte bei einem Patienten mit peritumoralem dem in unserer Gruppe kernspintomographisch belegt werden, ohne dass die typischen Steroidnebenwirkungen zu verzeichnen waren. 5/19 Kinder gaben eine Verbesserung des subjektiven Befindens an, was aber auch als psychologischer Effekt der Hoffnung auf ein Ansprechen in der palliativen Therapie interpretiert werden kann. 3/17 Patienten mit malignen Tumoren zeigten meist vorbergehend objektivierbare Vernde-

Boswellic acids are extracted from the resin of the Boswellia tree. They were already used by Hippokrates and Dioskurides in various preparations for antiinflammatory therapy [16]. Salai Guggal is the Indian incense prepared from Boswellia serrata. It continues to be used by the traditional Indian folk medicine (Ajurveda) for the treatment of inflammatory and rheumatic diseases. On biochemical analysis, this Indian incense represents a standardized mixture of various Boswellic acids and a minor component of dienes and terpenes [11,16]. In vitro, the acetyl-11-keto-boswellic acid has been shown to be the most effective component of all Boswellic acids. [7]. The extract of incense became known to German physicians in 1997 by a publication in the Deutsche rzteblatt, where favorable effects were reported in adult patients with malignant gliomas [2]. Despite slashing criticism of inadequate CT follow up [17], parents of children with incurable brain tumors who had been informed about this study through the media, urgently requested for H 15. We report our experiences with Boswellic acid given to children with progressive or relapsed brain tumors known to have been treated previously by protocols of the German Society of Pediatric Oncology and Hematology. The data could be collected retrospectively from the participating four clinics.

Klin Pdiatr 2000; 212: 189 195  Georg Thieme Verlag Stuttgart New York ISSN 0300 8630

190 Klin Pdiatr 2000; 212

Patients
Between 1995 and 1998, 19 patients from the childrens hospitals of Bonn, Dortmund, Dsseldorf, and Wuppertal were treated for various brain tumors. Eleven boys and 8 girls aged 0.5 to 18 years, (median 8.4 years) received Boswellic acids (H 15) upon their parents urgent request (Table 1). Histological diagnoses were ependymoma WHO III (n = 7), astrocytoma WHO III/glioblastoma (n = 6), PNET/medulloblastoma (n = 3), malignant germ cell tumor (n = 1) and histologically benign tumors (n = 2; astrocytoma WHO I and dysembryoblastic neuroepithelial tumor). Tumor material of 13/19 patients at the time of diagnosis had been centrally reviewed by the German brain tumor reference center (Institute of Neuropathology of the Bonn University, Head: Professor Dr. Ottmar Wiestler). Primary tumor sites were hemispheres (n = 9), cerebellum (n = 6), brain stem (n = 3) and pineal gland (n = 1). All patients had had primary surgery. According to surgical reports, 9 tumors had been totally, 5 tumors subtotally and 2 tumors partially resected, while 3 patients had had biopsies only. Current therapy protocols of the German Society of Pediatric Oncology and Hematology had been administered postoperatively as adjuvant chemotherapy in 14 patients. Thirteen patients received radiotherapy. Thus, all patients received adequate pretreatment.

Janen G et al
WHO criteria were used for the evaluation of tumor response: complete response: no visible tumor partial response: tumor regression 50 % stable disease: no change, or regression < 50 % of the tumor volume progressive disease: increase of tumor volume > 25 % of the tumor volume

Results
No side effects were observed which were attributable to H 15 in 19 patients over 1 26 (median 9) months. Two/19 asymptomatic patients without any evidence of residual tumor received H 15 as adjuvant therapy. Therefore, the remaining 17 patients can be evaluated for a potential effect of H 15. Subjective relief, clinical and radiological findings and clinical course are shown in Tab. 2. Any change of subjective or clinical symptoms was documented for all patients during the first 3 months of H 15 application.

Subjective symptoms
Five/19 children stated a subjective relief of symptoms during treatment with H 15: An improvement of general condition was reported by patients 3, 4, 5, 9 and 12, an improvement was stated in 2/4 patients with headache before H therapy (patient 3 and 9).

Application of Boswellic acids (H 15)

Indication for the application of Boswellic acids were progressive (n = 9) or relapsed (n = 6) tumors. Two patients with benign tumors received Boswellic acids after partial tumor resection. In 2 further patients, whose malignant tumors had been completely resected, H 15 was administered as adjuvant therapy. The oral dose of H 15 was 40 126 (median 77) mg/kg BW/ day administered in 3 doses over 1 26 (median 9) months (Tab. 1). Fourteen patients received H 15 as mono-therapy while 5 patients received H 15 with additional therapy (chemotherapy 2 pts., radiotherapy 3 pts., dexamethasone 1 pt.). Patients were evaluated for clinical and radiological response. Subjective relief of symptoms like headache, fatigue, loss of appetite and nausea were monitored, in order to evaluate the palliative effect of H 15-application. Objective criteria like vomiting, body weight and the course of neurological deficits like pareses, ataxia, impaired coordination and sensory deficits were documented by repeated careful examinations. Side effects on hepatic- and renal function and on the hematopoietic system were monitored by monthly routine laboratory examinations. CT/MRI with or without contrast medium were performed prior to H 15-application and during therapy at 8 to 12 week intervals. The images were centrally reviewed.

Objective clinical findings

The physicians observed an increase of muscular strength (patient 11, 14, 16), improvement of a hemiparesis (patients 8 and 12) or cranial nerve palsy (patient 12) as well as ataxia (patient 4) and weight gain in a patient with kachexia (patient 14).

Radiological findings
Radiological examinations showed a regression of the peritumoral edema in patient 9, and a regression of the volume of a tumor cyst in patient 14. In patient 10, a complete remission was achieved by additional chemotherapy over 11 months. After completion of chemotherapy, a progression in tumor growth was seen again despite the ongoing application of H 15. One child (patient 2) showed progressive disease under chemotherapy. She was irradiated and received H 15 at the same time. Although a decrease of tumor size was documented by MRI 8 months after radiotherapy, she developed a cranial nerve palsy 2 months later and progressive disease was documented by MRI. A partial remission, which has been lasting for more than 26 months at the time of this report, was achieved in patient 4 by giving H 15 without any additional therapy. Patients 9, 12 and 14 were in stable disease for 3 6 months until further tumor progression occurred. The course of disease of patient 4 is demonstrated, since he showed some clinical improvement and a partial remission according to MRI: At the age of 4 years, a cerebellar anaplastic ependymoma had been diagnosed and completely resected

Tab. 1 Diagnoses and clinical data of 19 children treated with H 15 d1 = ependymoma; d2 = malignant glioma WHO8 III, glioblastoma; d3 = PNET/medulloblastoma; d4 = germ cell tumor; d5 = astrocytoma WHO8 I; d6 = dysembryoblastic neuroepithelial tumor patients characteristics # sex age/yr. dx pretreatment surgery irradiation (Gy) loc 54 csp 32/51 cr 54 csp 35/55 csp csp 35/54 cr ? loc 54 loc 48 csp 35/52+ seed csp 36/45 csp 34/55 loc 54 chemotherapy HIT 91 HIT 91 HIT 91 HIT SKK 92 HIT SKK 92 HIT SKK 92 VCR HIT 91 Cyc/VP 16 CPL/VP 16/ CCNU PVC/PEI IFO/CPL/ VP 16 HIT 91 SIOP CNS GCT 96 months after dx 31 7 42 14 26 5 3 3 19 26 1 4 20 35 8 12 6 52 39 Treatment with Boswellia acids reason H 15 mg/kg/day 48 85 54/71 126 92 90 80 45 52 51 114 66 100 76 96 114 50 80 104 duration in months 8 10 15 26 8 5 2 10 3 18 2 4 18 8 1 2 7 15 12 in combination with OP/irradiat. irradiation PEI irradiation EVAIA Dexamethason

Boswellic acids in the palliative therapy of children with progressive or relapsed brain tumors

01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19

m f m m f f f m m f m m f f m f m m f

13,6 8,11 7 4,1 3,1 1,11 0,9 18,1 14,11 14,8 13,9 12,7 0,5 12,9 5,6 3,5 11,10 4,5 4,10

d1 d1 d1 d1 d1 d1 d1 d2 d2 d2 d2 d2 d2 d3 d3 d3 d4 d5 d6

t st t t st t st b t t b b t st t st t pr pr

rel pg rel rel pg pg pg pg pg rel pg pg adj pg rel rel adj inop inop

f = female, m = male, t = total, st = subtotal, pr = partial, b = biopsy, loc = local, cr = cranial, csp = craniospinal with tumorboostpg = progress, rel = relapse, inop = inoperability, adj = adjuvant

Klin Pdiatr 2000; 212 191

192 Klin Pdiatr 2000; 212

Tab. 2 Subjective relief, clinical and radiological findings in 11/19 children with H 15-application patient 02 03 04 05 08 09 10 11 12 14 well-being muscular strength hypoglossal paresis hemiparesis bodyweight muscular strength is able to walk again muscular strength headache well-being headache well-being well-being well-being hemiparesis ataxia is able to walk again best subjective relief best clinical effect best radiological finding PR PD PR SD PD SD peritumoral edema CR PD SD SP tumor cyst volume PD clinical course

Janen G et al

PD after 10 months dead dead PR over 26 months, alive PD after 8 months dead dead PD after 6 months, dead PD after 11 months dead dead PD after 3 months dead

PD after 6 months

dead dead

16

(Fig. 1 a + b). After craniospinal irradiation in combination with vincristine and consecutive chemotherapy with cisplatin, vincristine and CCNU according to the HIT 91 protocol, the clinical condition of this boy had improved very little. At the end of the therapy, the child suffered from severe ataxia and was unable to walk. Furthermore, he showed a severe dysarthria. Three months after completion of therapy, MRI showed evidence of a local relapse (Fig. 2 a). In view of this patients reduced clinical condition, prolonged bone marrow depression and the unfavorable prognosis of an early relapse, he received no further surgery or chemotherapy. According to

the parents wish, application of H 15 3 800 mg/day was started. Follow-up examinations showed an improvement of the clinical condition. The ataxia decreased, the boy consecutively relearned walking and his speech became more fluent. At the time of this report he is able to visit a kindergarten. Two years after start of H 15, recent MRI controls documented a marked regression of the disease (Fig. 2 b). Patients 18 and 19 with benign tumors remain in stable disease with more than 12 and 15 months, resp., of H 15 treatment as the sole therapy.

Fig. 1 a, b Cerebellar ependymoma, preand post-surgical MRI: a coronar T1 weighted image after internal Gd-DTPA application: large tumor of the 4.th ventricle with GdDPTA enhancement, internal hydrocephalus b coronar T1-weighted image: defect filled with CSF, reduction of hydrocephalus (patient 4).

Boswellic acids in the palliative therapy of children with progressive or relapsed brain tumors

Klin Pdiatr 2000; 212 193

Fig. 2a, b Tumor relapse: coronar T1weighted images: a two nodular tumors with Gd-DPTA uptake at the primary tumor localization. b follow-up examination after 2 years of H 15-application: Subtotal regression of the centrally located tumor relapse. Marked reduction of the left parasagittal tumor (patient 4).

Seven tumors (patients 3, 6, 7, 8, 11, 15, 16) progressed under H. In these patients, the short clinical improvement which was observed might have been caused by the antiedematous effect of H 15. No assessment of the anti-tumor effect of H 15 can be given in patients 13 and 17, since these patients received Boswellic acids in first remission without evidence of tumor, the patients remain free of relapse after a follow up of 38 and 53 months.

of H 15 (3600 10 000 mg/day). Well aware of the fact that the study presented was a retrospective analysis, we did not observe these side effects nor other clinical or laboratory changes. The maximal applied doses were 3600 mg/day in patients with more than 50 kg body weight. Various pharmacological effects of Boswellic acids, which might account for the observed clinical effects, have been described in vitro and partially in vivo: Prostaglandines and leukotrienes are mediators of inflammatory response. They are produced in the arachidonic acid cascade. Leukotriene synthesis is regulated by 5-hydroxylipoxygenase. Glucocorticosteroids inhibit prostaglandine and leukotriene synthesis. Incense selectively inhibits leukotriene synthesis by reducing 5-hydroxylipoxygenase activity. This effect has been demonstrated in vitro in neutrophilic granulocytes of the rat [1,10], as well as in vivo in patients with astrocytoma. In these patients, a normalization of previously increased leukotriene urine concentration was observed [6,12,14]. Furthermore, incense blocks C3-convertase. This is an additional antiphlogistic mechanism which inhibits the complement activation in vitro [8]. In vitro, Boswellic acids interfere with the activity of plasmin and human leukocyte elastase. However, an in vivo influence on plasmatic coagulation has not been observed [13]. There is some evidence that Boswellic acids may inhibit topoisomerase I and II [7, 9]. This might explain the antiproliferative effects which were observed in patients with brain tumors or melanomas. However, the Boswellia concentrations, which are necessary to mediate an antiproliferative effect in vitro, have to be 10 12 fold higher than the ones required for the inhibition of leukotriene synthesis. There are no pharmacokinetic data on the Boswellia concentrations in

Discussion
To our knowledge, the application of Boswellic acids in children with malignant diseases has not been studied. Therapeutic effects of this preparation remain to be proven. Unlike in the Swiss Canton Appenzell-Auenrohden, H 15 is not licensed in Germany. However, Boswellic acids are listed in the Deutsche Arzneimittelbuch, DAB [4, 5]. Therefore, H 15 can only be used for individual treatment. The first aim of this evaluation was to determine possible side effects in palliative treatment. Our data on H 15 mostly represent experiences on single cases, in which therapy was initiated upon parental request. Due to the heterogeneity of tumor histologies, the additional anti-tumor therapy and differences in the application of H 15 regarding dose and duration of application our data do not allow a conclusion on general anti-tumor effects of H 15 compared to conventional therapy. Regarding side effects of H 15, the literature reports occasional fatigue or unspecific gastrointestinal symptoms like vomiting or diarrhea [2, 3]. Some patients showed a transient unspecific skin rash. One patient had reportedly suffered from azoospermia, and 2 patients from neutropenia after overdose

194 Klin Pdiatr 2000; 212

patient plasma or tumors, so that the authors can but speculate on the role of this anti-proliferative effect for the clinical responses of our patients. Boswellic acids are lipophilic molecules which are able to pass the blood-brain barrier. Therefore, intracerebral effects of Boswellic acids would be conceivable. Simmet and Winking treated 29 adult patients with glioblastoma preoperatively with Boswellia serrata (H 15) in order to reduce peritumoral edema. Regression of edema could be demonstrated after administration of a dose of 3 1200 mg/H 15/day over 7 days. Clinical symptoms of intracranial pressure improved and CT scans showed regression of edema [2]. In addition, tumor necrosis was demonstrated on histological examination in half of the patients on H 15, versus 7 % of patients treated with steroids. [15]. These as of yet uncontrolled findings reached many patients via media in a desperate situation due to an incurable brain tumor. According to the urgent wish of the parents we administered various doses of H 15 to 19 children with brain tumors. Nearly all children suffered from incurable brain tumors despite conventional chemo- and/or radiotherapy. It is obvious that, concerning tumor growth in childhood, there are phases of progression and stability in the natural course of the disease. We cannot evaluate possible effects in 2/19 patients receiving H 15 because of the adjuvant situation. Six of the remaining 17 patients reported an improvement of their clinical condition. The subjective feeling of a better general condition can be interpreted as psychological effect of the oral treatment with H 15, since expectation and hope for improvement had been the reasons for starting application. In 2 of these children and 2 additional patients with no subjective statements regarding the improvement, regression of prior neurological symptoms like pareses and ataxia were documented. These effects might be attributed by an anti-inflammatory or anti-edematous effect of the Boswellic acids. Regression of an extensive peritumoral edema or reduction of a tumor cyst were documented by MRI in 2 of these patients. An anti-tumoral effect cannot be verified in two patients because of a combination with radiation or chemotherapy. Under Boswellic acids, 3 children with malignant brain tumors in progression remained in stable disease over 3 to 6 months. In one case, a partial remission over 26 months with administration of H 15 as monotherapy was documented by MRI. Considering the absence of a histological reevaluation of the relapsed tumor (patient 4), a temporary Gadolinium enhancement of a glial scar must be discussed. In the second patient with partial remission (patient 2) the ependymoma WHO III of the brain stem was only partially resected and progressed under chemotherapy according to HIT 91 protocol; tumor regression could be documented 2,5 and 8 months after finishing radiotherapy. During this time the patient received H 15 alone. But two months later under H 15 application, tumor progression was documented so that in this case the response to radiotherapy is more likely than an antitumoral effect of H 15.

Janen G et al
Palliative therapy should fulfill the following criteria: drugs should be easily applicable in an outpatient setting, should be free of severe side effects, and should induce at least subjective alleviation of the reported complaints, thus reducing the need of analgesics. In addition to the improvement of the clinical condition and of the quality of life, a prolongation of life would be most desirable. From our observations we can confirm a positive palliative effect of H 15. Five patients reported a relief of general symptoms, and 4 patients showed a transient or sometimes longer lasting regression of severe neurological symptoms like pareses or ataxia. Therefore, H 15 seems to be appropriate for the palliative care of children with incurable brain tumors. The observed clinical improvement may be attributed to a reduction of the peritumoral edema. The long time efficacy of H 15 can be compared to that of dexamethasone without the specific side effects observed under steroid therapy. Thus, H 15 might be considered as an alternative treatment in patients who experience severe side effects of dexamethasone therapy. Since some brain tumors show heterogeneous growth patterns with phases of slow tumor growth and others with rapid tumor progression we cannot clearly distinguish between the therapeutic impact of H 15 and the natural course of the illness in our patients with stable disease. With the histological verification of the relapse in patient 4 lacking, the partial remission cannot surely be attributed to an anti-tumor effect of H 15. Instead of an uncritical use of H 15, a prospective evaluation of patients with relapsed or progressive brain tumors of comparable histology is warranted, aiming to assess the application of H 15 with regards to optimal dose, time of application, and therapeutic impact.

Acknowledgement
The authors wish to thank W. Philippen for excellent secreterial work.

References
1

Ammon HPT, Mack, Singh GB, Safayhi H. Inhibition of leukotriene B4 formation in rat peritoneal neutrophils by an ethanolic extract of the gum resin exudate of Boswellia serrata. Planta Med 1991; 57: 203 207 2 Bker D-K, Winking M. Die Rolle von Boswellia-Suren in der Therapie maligner Gliome. Deutsch rztebl 1997; 94: A1197 A1199 3 Brckle W. Zusammenfassendes Gutachten zur klinischen Dokumentation 1988; zitiert nach Produktinformation: H15, Spezifische Weihrauchtherapie von Ebi Pharm, Kirchlindach 4 DAB 1 im Original. Verlag der kniglichen geheimen OberHofbuchdruckerei (R. V. Decker) 1887 5 DAB 6 im Ergnzungsband. Deutscher Apothekerverlag 1944 6 Heldt RM, Winking M, Simmet Th. Cysteinylleukotrienes as potential mediators of the peritumoral brain oedema in astrocytoma patients. Naunyn-S Arch Pharmacol 1996; 353/4S, R 142: Abstract 538

Boswellic acids in the palliative therapy of children with progressive or relapsed brain tumors
Hoernlein RF et al. Acetyl-11-keto-b-boswellic acid induces apoptosis in HL60 and CCRF-CEM cells and inhibits topoisomerase I. Proc Amer Assoc Cancer Res 1997; 38: 192: Abstract 1291 8 Kapil A, Moza N. Anti-complementary activity of boswellic acids an inhibitor of C3-convertase of the classical complement pathway. Int J Immunopharmacol 1992; 14: 1139 1143 9 Lee Y-W et al. Pentacyclic triterpenoid compounds as topoisomerase inhibitors or cell differentiation inducers. Patentschrift WO 90/01937, PCT 1990 10 Mack T. Hemmung der 5-Lipoxygenaseprodukte stimulierter neutrophiler Granulozyten der Ratte durch Extrakte aus dem Harz von Boswellia Serrata Roxb. und durch Boswelliasuren. Dissertation, Tbingen 1990 11 Pardhy RS, Bhattacharryya SC. b-Boswellic acid, acetyl-b-boswellic acid, acetyl-11-keto-b-boswellic acid -boswellic acid, four pentacyclic tripene acids from the resin of Boswellia serrata Roxb. Ind. J Chem 1978; 16B: 176 178 12 Safayhi H, Ammon PT. Pharmakologische Aspekte von Weihrauch und Boswelliasuren. PZ 1997; 39: 11 20 13 Safayhi H, Rall B, Ammon HPT. Inhibition by boswellic acids of human leukocyte elastase. J Pharmacol Exp Ther 1997; 281: 460 463 14 Sailer E-R et al. Acetyl-11-keto-boswellic-acid (AKBA): Structure requirements for binding and 5-lipoxygenase inhibitory activity. Br J Pharmacol 1996; 117: 615 618 15 Simmet Th. Weihrauch gegen Hirntumoren? Persnliche Mitteilung. Deutsche Apotheker Zeitung 1997; 137: 3 16 Tschirch A. Handbuch der Pharmakognosie. C. H. Tauchnitz, Leipzig 1909 17 Warnke P et al. Die Rolle von Boswellia Suren in der Therapie maligner Gliome methodische Mngel. Letter to the Editor. Deutsches rzteblatt 1998; 95: 192 193 18 Winterstein A, Stein G. Untersuchungen in der Saponinreihe. Zur Kenntnis der Monooxytriterpensuren. Hoppe-SZ Physiol Chem 1932; 208: 9 25
7

Klin Pdiatr 2000; 212 195

Dr. Gisela Janen

Heinrich-Heine-Universitt Klinik fr PdiatrischeHmatologie und Onkologie Moorenstr. 5 40225 Dsseldorf