RADICAL THERAPY OF THALASSAEMIAS

 

GLAUCO TORLONTANO (a,b,c),PAOLO DI BARTOLOMEO(b),GABRIELE DI GIROLAMO (b), PAOLA OLIOSO (b),GABRIELE PAPALINETTI (b), PASQUA BAVARO (b),DONATO NATALE(b), CESARE BOSMAN (c,d)

 

(a) Università «Gabriele D'Annunzio» di Chieti‑Pescara (Italy);(b) Dipartimento di Ematologia, Terapia Intensiva ‑ Centro Trapianti Midollo Osseo ‑ Ospedale Civile ‑ Pescara (Italy);(c) Ospedale «Casa Sollievo della Sofferenza» (IRCCS) ‑ San Giovanni Rotondo (Foggia)(Italy);(d) Cattedra di Anatomia Patologica ‑ Università «La Sapienza» ‑ Roma (Italy).

 

* Estratto da Atti Convegno Lincei “The  Thalassemic Syndromes: A Symposium in honour of Ezio Silvestroni and Ida Bianco.Roma 13.dicembre 1999.

                   

               At present, bone marrow transplantation (BMT) can be considered the treatment of choice for young homozygous thalassaemic patients who have as donor an unaffected HLA‑identical sibling. However, the possibility of cure regards only 30% or less of patients in myeloablative BMT with HLA‑identical sibling donor. The transplant procedure must be considered ethically acceptable only in presence of low risk of death. In fact, the conservative trans‑fusion and iron chelating therapy has improved progressively the quality of life and survival. In Seattle, Thomaset al. considered possible and acceptable in thalassaemia a 10% mortality risk with 90% post transplant survival and 75% recovery, on selected very young patients, scarcely or not yet transfused without evident signs of damaged organs. The first transplant was successfully realized in 1981 by Thomas, in «a good risk» patient, managed with few transfusions, an Italian thalassaemic 16 month old boy from a 6yr old HLA‑identical normal sister [1]. The Seattle protocol for graft included:

 

     1. High dose intravenous Methyl‑Busulphan (BU) and moderate doses of Cyclophosphamide (CY), for host immunosuppression and eradication of uderlying disease and creation of «marrow space»;

 

    2. Stem cell graft for rescue from regimen related myelosuppression and establishment of normal hemopoiesis;

 

    3. Post transplant Methotrexate (MTX) for induction of host‑versus‑graft tolerance and prevention of severe graft‑versus‑host disease (GvHD).

 

             The Seattle initial experiment was interrupted after 4 other transplants where 2 children   died. After Seattle, Lucarelli in Pesaro transplanted 13 pa­tients with homozygous thalassaemia from identical sibling donors. Accord­ing to Lucarelli a variety of preparative regimens was utlized involving high doses BU and/or CY and/or Total Body Irradiation (TBI). MTX  was given af­ter transplant. There were difficulties of eradication of Thalassaemia,relevant toxicity of treatment and high mortality (46%) [2]. Subsequently, Lucarelli encountered difficulties even with high doses of BU (16mg/K2) and without TBI. Only afterwards following the Pescara‑London criteria [3, 5,7] utilizing Cyclosporin (CyA) and moderate doses of BU he obtained better results [8].

 

      From 1976, year in which our BMT activity started in Pescara, our ma­jor aim has always been to reduce the transplant related mortality and enhance the clinical and supportive management of the transplanted patients[3]. The progressively decreasing risk related to the procedure encouraged us to evalu­ate BMT also in haematological diseases such as thalassaemias, characterized by several decades of acceptable quality of life with conservative therapy.

        Our BMT program in Thalassaemia Major was planned in 1981 with a selection of «good risk» children. Finally in 1983, Barrett in London [4] in collaboration with our team in Pescara [5,7], began to transplant some homozygous thalassaemic «good risk» young patients. We were the pioneers in utiliz­ing moderate doses of BU‑ CY‑ Cy and Acyclovir (Acv) protocol (Table 1):

 

•        CY with moderate doses of BU to reduce hepatic toxicity;

 

•       CyA because we considered it more effective than MTX in post transplant treatment

          for induction of host‑versus‑graft tolerance and prevention of se­vere GvHD; 

 

•       Acyclovir to reduce the risk of severe Herpes viruses including CMV in­fection.

 

This regimen demonstrated to be the most effective.

      After some successful transplants, Barrett transferred to Bethesda (USA) and the London (Westminster Hospital) experiment with Thalassaemic pa­tients was interrupted. After the Pescara/London experience (1983), Lucarel­li in Pesaro in 1985, utilized in Thalassaemia an analogous protocol [8].

       In Pescara, patient selection for transplantation has always been very im­portant. In the initial period (1983‑1987), patients were rigorously selected, according to: good general and

organic conditions, without severe iron over­load <10 mg/g liver dry); hemoglobin level generally maintained with trans­fusions at 9‑10 gr/dl; absence of severe chronic hepatitis; left ventricular ejection fraction >55% (6‑7,9).

 

 

TABLE 1 ‑ Marrow transplantation protocol for Thalassaemia.

 

 

 

 

 

 

 

 

 

 

 

        During this period we transplanted 30 patients (10-55 months of follow‑up) with the following results: Survival 96%; DSF 93%; Alive with Thalassaemia 4%; Deceased 4% (7,9). From 1983 till 1986 in Pescara we transplanted only patients up to 8 yrs, in 1987 up to 13 yrs old and since 1988 adult patients as well. These results encouraged us to transplant progressively older patients. Therefore on February 2, 1988, we transplanted the first adult patient, a 21 yr old woman. This was probably the first adult Thalassaemic patient in the world cured and at present, alive and well after a 12 yr follow up period. After this successful experience our BMT activity was extended also to adult patients and our policy was to exclude cirrhotic and cardiopathic patients with left ventricular ejection fraction <55%. However we cautiously admitted to the protocol patients with relevant iron overload, siderosis, fibrosis and chronic hepatitis. In these cases we offered the option of BMT only after pre‑transplant intensification of i.v. chelation therapy and/or following treatment with Interferon to enhance the chronic hepatitis, if present.

 

        The success of our transplants, other than the choice of the less toxic myeloablative regimen possible, was determined by the pre‑peri and post transplant careful management of patients with its progressive improvement.

 

        From May 1983 to June 1.998, we have performed from HLA identical sibling donors 115 transplants of which 4 second transplants in 111 Thalassaemic patients and with 3 patients off protocol (1 patient with insufficient doses of BU still living with Thalassaemia and 2 patients with cirrhotic evo­lution of the liver) (table 2) [10)

        Graft rejection occurred in 4 patients who underwent a second trans­plant. Two of these transplanted from the same donor, after immunosuppres­sive conditioning with CY and ALG are now alive and disease free 9 and 11 years respectively after the 2^nd BMT.

        In conclusion, as of December 1999 on the total of 108 Thalassaemic transplanted patients, the actuarial probabilities of survival and disease free survival from a minimal of post transplant follow up of 18 mths to 16 yrs, are 91% and 87% respectively with 4% alive with Thalassaemia and 9% mor­tality (Fig. 1).

        Of particular interest in our case group is also the reduced number of  rejection. This favourable result can be attributed to:

 

        I. hypertransfusional treatment during the two weeks pre‑transplant pe­riod, in order to reduce marrow mass and thus favor myeloablation;

 

        2. favorable take likely due to the always elevated number of marrow cells constantly infused.

 

        Moreover, the extreme rarity of rejection in our adult patients should be attributed to the noted reductions of the marrow mass in the adult thalassaemic with consequently easier myeloablation.

 

 

 

TABLE 2 ‑ Thalassaemic patients transplanted in Pescara.

 

 

 

 

 

 

Fig. 1 ‑ Thalassaemia major: survival and disease‑free survival.

 

 

 

          The following points of clinical management of BMT in thalassaemia are for us essential and are still applicable.

 

         1)Hypertransfusion during the 2 weeks prior BMT to reduce the mar­row mass and enhance myeloablation;

 

         2)Accurate skin and mucosal decontamination to reduce the risk of en­dogenous infection and severe GvHD and rigorous isolation in «laminar air flow» (LAF);

                                                                                                   

         3)High number of infused nucleated cells: possibly >5x10⁸ in order to favour a stable take. In order to assure adequate doses of marrow, many pa­tients were also transplanted with donor cryopreserved marrow, in addition to fresh cells;

 

        4)Careful microbiological monitoring of the patients with prophylaxis possibly with specific anti‑bacterial, anti‑mycotic and anti‑viral drugs;

 

        5)Enhancement of clinical and supportive treatment with progressive risk reduction for severe GvHD and fungal and CMV infections;

 

        6)Moderate dose BU (13‑14 mg/kg) to reduce the overall toxic effect of the liver; CyA, at least for one year, alone or initially associated with short course MTX to prevent rejection and severe acute and/or chronic GvHD;

 

        7) Very prolonged treatment with Acyclovir (9 months post‑BMT) to re­duce the risk of severe Herpes virus infections;

 

        8)Careful management in the first 3 months post BMT and general pro­tection during the first year post‑transplant;

 

        9)Exclusion of chelation, which increases the growth of siderofilic mi­crobes, possibly during the first year post BMT; and, if necessary, preferably use of early systematic phlebotomy for gradual reduction of iron overload. In effect the residual post‑transplant immunosupression, relevant especially dur­ing the first year, increases the risk of severe and life‑threatening infections, especially of siderofilic microbes as Yersinia, Pnemocystics Carinii and Picornomycosis:

 

        10) Psychological assistance.

 

         In 1990 Lucarelli et al. published the results of 222 young patients (age <l6 yrs) transplanted between 1983 up to 1989 from HLA identical sibling donors. After one year post transplantation, the survival and DFS were 82% and 75% respectively with 8% rejection and 18% mortality [8).Only in 1985, Lucarelli initiated a transplant regimen similar to the Pescara protocol with moderate doses of BU and CyA starting from day ‑1 to 365 days post trans­plant [8]. In the first 99 young patients transplanted from 1985, <16 yrs, the outcome of transplantation was examined retrospectively in an attempt to iden­tify eventual prognostic parameters. Three factors seemed to be associated with significantly reduced probability of survival and disease free survival: poor quality of chelation; presence of hepatomegaly more or less 2 cm; pres­ence of portal fibrosis. Therefore three classes were individualized: Class 1-­patients with no factors, characterized by survival and disease free survival 95% and 90% respectively; Class 2‑patients with only one risk factor and survival and disease free survival 85% and 81% respectively; Class 3‑pa­tients with hepatomegaly, liver fibrosis and survival and disease free survival 78% and 54% respectively.

        Subsequently, these results were substantially confirmed from wider cas­es reported, but showed less satisfactory results with this protocol and also its variations, in all adult patients and also in young patients belonging to the third class [11]. In effect, in 109 adult thalassaemic patients aged more than 16 yrs, transplanted in Pesaro from HLA‑identical sibling donor, from 1988 to 1997, the results were the following: Survival 67%, DFS 63%, Mortality 35%, with different protocols for 2^nd and 3^rd class patients.

 

 

         The above classification merits some considerations: 1) the degree of splenomegaly appears scarcely measurable on physical exam; 2) difficulty in quantifying the degrees of chelation and the compliance, based on the pa­tient's history; 3) easier to quantify are the fibrosis and siderosis studied by histochemical methods. In summary, only the iron overload siderosis and the fibrosis appear adapt to a quantitative type of evaluation.

        As published in the Annals New York Academy of Sciences (Lucarelli, [11]), Pescara Center came second only to Pesaro for the number of trans­plants but not for the quality of the results.

        As mentioned before, today in Pescara on the total of 108 transplanted thalassaemic patients, aged 11 months to 26 yrs, the actuarial probability of survival and disease free survival, after 16 yrs of experience are the follow­ing: Survival 91%; DFS 87% ; rejection 4%; mortality 9% (Fig. 1). In the 30 adult patients aged 16‑26 yrs, the results are similar to those obtained in the global casistics with: Survival 90%; DFS 87%; rejection 3% (1 patient donor was HLA identical uncle); mortality 10%.

        Therefore the age of the transplanted patients (11 months to 26 yrs) did not significantly influence the transplant's outcome. However, if we consider the age with a take off at 10 yrs old, the trend seems more favorable for pa­tients under 10 yrs (Fig. 2). This is easily understood since younger patients, in respect to older patients are less exposed to iron overload, anemia and infections, especially hepatitis. These damages are greater in Thalassaemics who

 

Fig. 2 ‑ BMT inThalassaemia ‑ Pescara experience 1983 ‑ 1998.

 

received less effective conservative therapies. In effect, beyond the frequent virus hepatitis from multiple transfusions, the most common damage in Tha­lassaemic patients is induced by iron loading responsible for lipid peroxida­tion of lysosome membranes, followed by release of acid hydrolase into cy­toplasm and finally by cell death [12]. The long term clinical consequences of this process are: heart failure, liver fibrosis, cirrhosis and endocrinopathies [13, 14]. Principal causes of death in our patients were: heart and liver fail­ure, encephalopaty, GvHD, pneumonia and VOD. According to our experi­ence, fibrosis and chronic hepatitis singularly do not seem so decisive as cause of mortality from transplant.

        The transplant risk in our patients seemed particularly relevant in ado­lescent and adult subjects with serious iron overload due to siderosis and fi­brosis, associated to severe chronic active hepatitis. In fact, with the passing of the years, this situation increases the damage of iron overload and that of chronic infections. lf we consider the patients who underwent a liver biopsy with histochemical studies: of 11 patients (10% of total cases) characterized by the above mentioned most unfavorable conditions, 4 of them deceased.

While in  the remainin73 patients, only 7 deceased. This difference of mor­tality is statistically significant (P= 0.03) (Fig. 3).

       Our study documents the feasibility of curing Thalassaemia Major with BMT from an HLA identical sibling, maintaining, also in adult patients, the transplant related mortality under 10% with disease‑free near 90%.

 

        It is of particular noteworthy that our patients, independently of age, re­ceived a1ways the same protocol for transplantation.

        The continuous enhancement in clinical and supportive treatment has al­lowed the progressive reduction of risks of severe GvHD, interstitial pneumonia

 

 

 

 

 

Fig. 3 ‑ Iron overload associated to severe chronic hepatitis and transplant risk.

 

 

 

and fungal and cytomegalovirus infections. In fact, in the last 6 years, from June 1992 to June 1998, our results have been particularly favorable [15]. Of the 47 transplanted patients (of which 22 adults) in our protocol, the survival and disease‑free survival have been 96% and 94% respectively, with rejection in only 1 patient (transplanted from HLA‑identical uncle) and with 2 patients deceased (Fig. 4). Of particular interest is also the reduced number of rejections [15].

 

        The problem of post‑transplant persistence of iron overload is very important. Our experience and that of other authors document the efficacy of phlebotomy in iron overload patients following sustained stable engraftment. Above all, in well iron depleted patients, after this treatment, it is also possible to increase the eventual response to interpheron therapy given to improve the chronic hepatitis. Instead early post‑transplant chelation can increase, according to our experience, the risk of life‑threatening infections by syderofil microbes, such as Yersinia, Pneumocystis Carinii, Picomomycetes [16].

 

 

 

NEW DIRECTIONS

 

1. Marrow Transplantation with matched unrelated donor or mismatched related donor

 

        Myeloablative BMT by compatible donor is a curative therapy only for 30% or less of Thalassaemic patients. Recently, relevant progress has been

 

 

Fig. 4 ‑ BMT in Thalassaemia ‑ Pescara experience 6/1992 – 6/1998.

 

 

made in alternative donor transplantation, thanks to improved methods of HLA typing for donor selection and for the control of GvHD and the pre­vention and cure of fungal and Cytomegalovirus infections [17]. This should be particularly important to realize BMT in Thalassaemics even in absence of an HLA‑identical related donor early in the onset of non compliance or oth­erwise in case of interruption of chelation therapy. However, further prudent studies are necessary and should be only within the context of a well defined research environment such as GITMO or similar groups. These studies could be realized with less difficulty within a genetically homogeneous population such as found in Sardinia, according to Cao [ 11 ] and Contu [ 11

 

2. Mini‑transplant non‑myeloablative low risk regimen

 

        In order to reduce the risk due to toxicity from the actual transplant pro­gram, Storb [ 18] and Slavin [ 19] have considered the possibility of obtaining the goal of a satisfying treatment with less toxic non myeloablative protocols in order to realize a mixed donor‑host chimerism with clinical control of Tha­lassaemic disease.

        Conditioning regimen of non myeloablative transplantation for its scarce toxicity would have the advantage of being administered to patients who, for reason of age limit, or of phases of disease, could not tolerate a BMT with myeloablative conditioning. Preliminary data of Slavin [19] seem to suggest that non myeloablative conditioning including Fludarabine, anti‑T‑lymphocyte globulin and low doses of Busulphan is extremely well tolerated. At present, besides several patients with malignant pathologies, also 6 patients with con­genital non malignant disorders, including 1 Thalassaemic, treated with non myeloablative regimen and with mixed chimerism seem to be alive and well.

       Recently, Nagler et al. [20], on the basis of their clinical experiences, believe that non‑myeloablative conditioning precells of peripheral blood stem cell transplantation (PBSCT) in connection with donor lymphocyte infusion (DLI) post allo‑PBSCT, is a promising approach towards increasing the suc­cess rate and optimizing BMT especially in children with genetic diseases even at an advanced stage of disease when conventional myeloablative BMT is too risky.

        Therefore, pre‑clinical experiments and clinical observations have led to the development of radically new concepts in which the myeloablative toxic pre‑transplant conditioning regimen may be replaced with minimally toxic non‑myeloablative regimens. Intense temporary immunosuppression before transplantation may be accomplished also by biological agents, e.g. monoclonal antibodies to surface determinants on T‑cells, complemented by post­grafting immunosuppression directed preventing HvG and GvH reactions.

 

3. Cord blood stem cell transplantation

 

       After the first experience of Gluckman et al. [21], the possibility and the utility of transplantation with Cord Blood Cells (CBCT) are well known to­day and evident that this method has less tendency to develop severe GvHD. But at present, the actual methods of collection have allowed to obtain only in 10‑20% of cases, according to our experience, a sufficient number of stem cells for the «take» in subjects who weigh over 50 Kg. Therefore, it seems necessary to optimize the methods of cord blood cells collection and / or to utilize «ex‑vivo» expansion of these stem cells.

The GITMO‑GRACE data have shown success in non neoplastic dis­eases superior to those in neoplastic ones; moreover, American data document good results also in transplants from non siblings HLA‑mismatched up to 3 loci [22].

At present, the Eurocord data regarding CBCT in beta Thalassaemia are not enthusiastic. In effect, of 12 beta Thalassaemic patients transplanted, all are alive with the exception of 5 who rejected [23]. Therefore it is necessary to prudently collect with more extensive experience and overall increase the quantity of injected stem cells.

 

 

4. In utero transplantation in absence of matched sibling donor

 

 

 

        In absence of a matched sibling donor, the «in utero» transplantation could be an alternative to post‑natal transplantation for the treatment of con­genital haematological disorders, such as Thalassaemia, that can be diagnosed early in gestation. But until now, according to Make and Zanjani [24], the treatment of haemoglobinopathies and thus also Thalassaemia, has been un­successful because this approach has not been adequately tested.

        According to Flake and Zanjani [24], a winning strategy could include: early in utero transplant also with minimal chimerism from parent donor with consequently donor specific tolerance induction; post natal non myeloablative transplantation from same haploidentical donor.

        Of interest is our case of a Thalassaemic child who underwent an «in utero» transplantation in Palermo with blood from her normal biovular twin sister. However, this attempt failed and the child was born with Thalassaemia. Subsequently, at two years of age, at the request of the Palermo Center, the

 

                patient was re‑transplanted in our Center from the same donor, according to our standard 

                protocol.

        The take was very rapid and complete, with perfect tolerance and with­out minimal disturbances as typically observed in transplants between true monovular twins. At present after a two year follow up the patient is not tha­lassaemic and is well.

        Although it was not possible to document the preceding microchimerism after the in utero transplant, the speed and the ease of the post‑natal trans­plantation take, leads one to think of the immunological tolerance attributed to the preceding attempt of the in utero transplantation.

 

5. Gene Therapy

 

        According to Luzzatto [25, 26], if it were easy to transfer genes even to a few stem cells, the genetically transmitted diseases today curable by allo­geneic bone marrow transplants, would be equally curable, using genetic cor­rection of the stem cells, followed by an autologous transplant. Included among these pathologies would be Thalassaemia.

        Of interest are the pre‑clinical studies of Luzzatto's proposal on glucose­6‑phosphate‑dehydrogenase (G‑6‑PD), transduced in the bone marrow cells of the mouse.

        But unfortunately, attempts of curing Thalassaemia utilizing gene ther­apy have failed. In order to achieve this aim it would be necessary more ad­equate gene vectors to reach the levels of gene expression; a better knowl­edge of the complex system of regulation of the locus of human beta‑globin gene chain.

 

        In conclusion, BMT is at present the only radical therapy for Thalassaemia. But, according to Weatherall: «Given the present limitations of bone marrow transplantation (BMT) and the low speed of progress towards defin­itive gene therapy, we may be left with transfusion and iron chelation as the mainstays of treatment for the foreseeable future» [27].

 

ACKNOWLEDGEMENTS ‑ Supported in part by grants from «Istituto Superiore della Sanità» ‑ Roma (Italia). Thanks to Mrs. Licia Timperio for reviewing the ma­nuscript.

 

                                                   

                                            REFERENCES 

 

 

1-E.D. THOMAS ‑ C.D. BUCKNER – J.E. SANDERS et al., Marrow Transplantation for Thalassaemia. Lancet, 8292, 1982, 227‑229.

 

2-G. LUCARELLI ‑ P. POLCHI ‑ T. IZZI et al., Allogenic Marrow Transplantation for Thalassaemia. Exp. Hematol., 12, 1984, 676‑681.

 

3-G. TORLONTANO ‑ A. IACONE, Transplantation of Bone Marrow in Aplastic Ane­mia and Leukemia. Haematologica, 64 (suppl.), 1979, 239‑248.

 

4-R. JOSHI ‑ A.J. BARRETT ‑ L. INGRAM et al., Bone Marrow Transplantation for B Thalassaemia Major. Exp. Hematol., 12 (15), 1984, 97‑98.

 

5-P. DI BARTOLOMEO ‑ G. Di GIROLAMO ‑ G. TORLONTANO et al., Allogenic Bone Marrow Transplantation for Thalassemia. Proc. XX Congr; Int. Soc. Hematol., Buenos Aires, Sept. 1984, p. 911.

 

6-G. TORLONTANO ‑ P. Di BARTOLOMEO ‑ G. DI GIROLAMO et al., Allogenic Bone Marrow Transplantation for Thalassaemia Major by Busulphan‑Cytoxan and Cyclosporin‑Acyclovir Prophylaxis Protocol. 2nd Int. Congr. Fetal Liver Trans­plantation, Pesaro, Sept. 1984. Haematologica, 69, 1985, 549‑552; Exp. Hema­tol., 13 (17), 1985, 37.

 

7-G. TORLONTANO ‑ A. IACONE ‑ G. Di GIROLAMO et al., Bone Marrow Trans­plantation with low risk in advanced and non advanced Thalassaemia. Bone Marrow. Transplant, 3 (1), 1988, 277‑278.

 

8-G. LUCARELLI ‑ M. GALIMBERTI ‑ P. POLCHI et al., Bone Marrow Transplanta­tion  in Patients with Thalassemia. N. Engl. J. Med., 322, 1990, 417‑421.

 

9-P. DI BARTOLOMEO ‑ G. DI GIROLAMO ‑ F. ANGRILLI et al., Bone Marrow Trans­Plantation for Thalassemia in Pescara. Prog. Clin. Med. Res., 39, 1989, 193‑199.

 

10-P. DI BARTOLOMEO ‑ G. DI GIROLAMO ‑ R OLIOSO et al., The Pescara Experi­ence of Allogeneic Bone Transplantation in Thalassemia. Bone Marrow Trans­plant, 19 (2), 1997, 48‑53.

 

11-G. LUCARELLI ‑ M. GALIMBERTI ‑ C. GIARDINI et al., Bone Marrow Trans­plantation in Thalassaemia. N.Y Acad. Sei., 1998, 270‑275.

 

12-C. BOSMAN ‑ G. TORLONTANO ‑ S. FUSILLI et al., Quantitative Assay by Image Analysis of Iron Loading. In: Hepatocytes of Thalassemic Patients in the Per­spective of Bone Marrow Homograft, 6th Int. Symp. on Diagnostic Quantita­tive Pathology, Basel (Switzerland) 1990, p. 73.

 

13-K. RELLER ‑ B. DRESOW ‑ M. COLLELL et al., Iron Overload and Antioxidant Status in Patients with B‑Thalassemia Major. Ann. N.Y Acad. Sci., 1998, 463‑465.

 

14-A.V. HOFFBRAND ‑ B. WONKE, Iron Chelation Therapy. J. Intern; Med., 242(740), 1997, 37‑41.