LONDON, SATURDAY 23 NOVEMBER 1991
`
`Oral iron chelation is here
`
`New oral chelation agents are challenging desferrioxamine
`
`Iron overload is a potentially fatal disorder, damaging the
`heart, liver, and other organs. It may be due to repeated blood
`transfusions or increased gastrointestinal absorption of iron,
`or both-as occurs in the~ thalassaemias. In patients given
`regular transfusions signs of organ damage (secondary haemo(cid:173)
`chromatosis) become apparent when around 50-100 units of
`red blood cells have been given: at that time about 10-20 g of
`extra storage iron has been introduced into the body, six to 13
`times the amount stored in normal people. Patients with
`primary haemochromatosis absorb excess iron from the diet.
`This amounts to far less iron than those having regular
`transfusions, but in time they still accumulate a lethal
`overload unless treated by venesection.
`Patients with transfusional iron overload need treatment by
`chelation. Most such patients have a haemoglobinopathy.
`Around 250 million people are heterozygotes with one of
`the haemoglobinopathies, and about 200 000 are born each
`year with potentially lethal homozygous thalassaemia or sickle
`cell anaemia. Most of the 100 000 with thalassaemia have one
`of the~ variants,' and these patients can be maintained in good
`health by repeated transfusion and chelation from the first
`year of life. Even without chelation regular transfusion can
`prolong life to around 20 years-as opposed to death within
`three years of birth without treatment.
`Many other conditions are- or should be- treated by
`chelation to prevent overload with iron from repeated trans(cid:173)
`fusions. These include not only sickle cell anaemia but also
`aplastic anaemia, myelodysplasia/myelofibrosis, and, in some
`cases, chronic renal failure. 1 Desferrioxamine has been the
`mainstay of iron (and aluminium) chelation for the past I 5
`years. It is generally effective and non-toxic, but because of its
`high cost, the cumbersome subcutaneous method of adminis(cid:173)
`tration, and toxic side effects only a small fraction of patients
`requiring chelation worldwide receive it. In India, for
`example, less than 5% of patients with f) thalassaemia major
`could afford treatment with desferrioxamine.
`What, then, about other chelators? Those such as EDT A
`and diethylenetriaminepenta-acetic acid, which are not selec(cid:173)
`tive for iron because of their carboxylic acid binding site,
`increase the excretion of zinc and magnesium, making them
`toxic. A new group of chelators, the a-ketohydroxypyridines,
`has many advantages, including a high affinity for iron; a high
`selectivity for iron over other biologically important metals;
`stability in acidic and physiological conditions because of its
`heteroaromatic structure; ability to cross the cell membrane
`because of its neutral charge and the formation of neutral iron
`
`complexes; and ability to remove iron from transferrin,
`ferritin, and haemosiderin. ~ Several of this group of chelators
`have been shown to be effective in removing iron from mice,
`rats, rabbits, and monkeys, but only a few have been shown to
`be safe when given long term to animals. One of these, 1,2
`dirnethyl-3-hydroxypyrid-4-one (LI), has been shown in the
`past five years to be of comparable efficacy to desferrioxamine
`in increasing the urinary excretion of iron in patients with
`thalassaemia and myelodysplasia.' The effectiveness of LI in
`increasing iron excretion has been confirmed in centres
`around the world in trials of daily administration for up to 15
`months.s-s
`Pharmacokinetic studies have shown that orally adminis(cid:173)
`tered LI is ahsorhed from the stomach and enters the systemic
`circulation through the liver with a halflife of O· 7-32 minutes,
`is metabolised in the liver to a glucuronide conjugate which is
`unable to bind iron, cleared through the kidneys with a half
`life of 47-134 minutes, and excreted almost completely in the
`urine in the form of mainly a glucuronide conjugate,
`unchanged LI, and LI bound to iron.9 Variations among
`patients in the clearance and extent of metabolism of LI have
`been noted. Removal of iron by Ll is thought to take place
`mainly from the serum and the liver.' LI does not increase
`faecal iron or absorption from the gut.
`The amount of iron excreted by L 1 depends on the dose and
`frequency ofadministration of the drug and the iron load of the
`patient. Single doses of Ll of 45-62 mg/kg in patients loaded
`with iron resulted in the urinary excretion of 10-70 mg iron
`(compared with less than I mg in normal people).n Results so
`far indicate that doses of 50-100 mg of L l/kg/day seem to
`cause a rate of excretion of iron sufficient to reduce the iron
`load of patients and maintain in most cases serum ferritin
`concentrations of 1000-2000 µg/I. Studies of intensive
`chelation have also been carried out and proved very effective;
`in one case 325 mg of iron was excreted in the urine after the
`oral administration of 16 g of LI in divided doses. ~ Overall,
`the results of iron removal by L l are highly encouraging, and
`there is no reason to suspect that with safe doses this treatment
`will not be as effective as subcutaneous desferrioxamine.
`Major toxic side effects of LI in animals during the
`administration of subacute doses of 200 mg/kg/day for up to
`12 months include a lowering of the white cell count,
`hypocellularity of the bone marrow affecting mainly progeni·
`tors of white cells, enlargement of the adrenals, and hyper(cid:173)
`salivation. 3 Histopathological examination of rats treated with
`acute intragastric doses of LI of greater tfian 2 g/kg have
`
`BMJ VOLUME 303
`
`23 NOVEMBER 1991
`
`1279
`
`
`1 of 2
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1019
`
`

`

`shown congestion in various organs, suggesting death from
`congestive heart failure . LI and its iron complex are not
`mutagenic and do not increase the growth of yersinia in
`vitro. l
`Toxic side effects observed during multicentre clinical
`trials with L I in over 200 patients known to have mainly
`thalassaemia and other volunteers in the United Kingdom,
`India, Switzerland, Canada, the Netherlands, Italy, and
`Czechoslovakia include transient episodes of agranulocytosis
`in two patients; musculoskeletal and joint pains in 19 patients;
`one fatal incident with a patient who developed a lupus-like
`syndrome; gastric intolerance in 12; and an increased require(cid:173)
`ment for transfusions of red cells in a patient who had not had
`a splenectomy. 3611
`' A more detailed report on the efficacy and
`toxicity of LI and other chelators in humans is given
`elsewhere."
`The prospect of identifying the few people who may be
`susceptible to LI requires further attention. Such screening
`has been suggested for other drugs - for example, screening
`for the production of sulphoxide in patients treated with
`penicillamine. Other measures might include withdrawing
`L I during acute inflammation and infection-both of which
`activate neutrophils and monocytes11-and introducing pro(cid:173)
`tocols of low doses given often. Toxicity long term and not
`efficacy is the major issue that needs to be addressed with L I
`now and possibly other related oral chelators in the near
`future. Formal long term studies of toxicity in animals and
`clinical trials will possibly be required before all patients
`switch from desferrioxamine to L I or some other oral
`chelator. Such studies, however, are expensive and may never
`be supported by pharmaceutical companies because of the
`
`Cam elf ord revisited
`
`Still not the last word
`
`classification of LI and related chelators in the "orphan
`drugs" category.
`The potentially life saving benefits of treatment with LI in
`patients loaded with iron who are not adequately treated with
`desferrioxamine may outweigh the risks of its possible
`toxicity, and its introduction in such patients may now be
`appropriate.
`
`GEORGE J KONTOGHIORGHES
`
`Senior Research Fellow,
`Department of Haematology,
`Royal Free Hospital School of Medicine,
`University of London,
`London NW3 2QG
`
`I WorlJ Health Org:mii;a1ion . ((,mmuni1y comrol of hereditary anaemias. Bull \t'.,,JJ lfrulth Or1:.in
`1983;61:63-80.
`.? Kunzmi n NA . Toxii:ity from alummium and iron: rccogmdon, 1re~nmcn1 and prc\'cllt100. Stnun
`/\'tph,oJ 1986:6(suppl IJ: l · -11.
`3 Kontugh1orgh1:s GJ. Design. propcnies ind cff1.'<'1IW use of 1he oral chdator LI anJ (Uher
`H·kctohydro>:yp)'ndincs in 1hc treatmcnl 0(1ran:i.fus1ooal iron ()\•c:rload in thala~sacm1a. Al'fn N \'
`ik ud SCI 1990:61l:339-SO.
`'* K'mm1i;;.hiorg_hes GJ. A1douri MA , Hnffbran<l AV. Barr J. \'<'onkc B, Kt.1uroudaris T , ,., al.
`Effc.-cuvc ( hc:ladon of iron
`in
`IS
`thalas.saeinia with the oral chela1or J ,2·d1mc1h\'l·3·
`hydroxypynd4-onc. BMJ 1987;l9S: IS09-1 2.
`·
`S Komoghiorgh<S GJ, Barden AN, Hoffbrand A\1 0 Goddard JG, Sheppard L. Barr j, <1 ut. Long
`tcr-m 1rial with the oral iron chdator I ,2·d.imcthyl·3·h)·droxypyrid4-QOC ( ll ). Cl • lroo i:hd:uion
`:and metabolic studies. Br J Hucma10J 1990;76:29S·300.
`6 Tondury P, Kon1oghiorghcs GJ. l.uthy AR. Hil'1 A, Hoffbrand AV, Loucnb:.ch AM ,,., al. 1..1
`(l,2-<lime1hyl·3·hydroxypyr1d·4*0nc) for oraJ iron chda1ion in pa1jcn1s wi1h ~ chaJass.acm1a
`majo<. Br J Hamsatol 1990;16:5Sl).3
`7 Oliv·icri NF, Koren G. Hcnnann C, Bcmur Y, Chung 0, Klein J, tt al. Comparison of oral iron
`chd.ator LI and dC$fcrrioxaminc: in patient$ loaded with iron. Lanett 1990;33':127S·9.
`8 Agarwal MB. Vi.swana11wl C. Ramana1han ), Massi! DE. Shah S, Gupie SS. tt al. Oral iron
`thd:.tlon with L l. Luncd 1990;3)5:60 I.
`9 Kon1og)liorghcs GJ, Goddard JG. Bartlcll AN, Sheppard L. Pbarmaookinttic s1Udies in humans
`wilh 1he ural
`iron chdalor 1,2-dunethyl·3·hydroxypynd-4-+0ne. C/in Plwrmar(I/ Tit.er
`1990:48:2SS-61.
`10 Bardell AN. lfoflbranJ A\'. Kuntoghi0rghes (jj . Long 1cnn 1rial with the Qral iron c;hcl:uor I ~2·
`dimc1hyl·3·h)·droxypyriJ4·Qnt.: tLl). 11 Clinical obscr\'~tions. l:lr J Hatmarol 1990~76:301 ·4.
`1 l Kontoghiorghes, Gj. Ad\-a.ncc.-s 1n ural iron chelatiun 1n man. /nttm.J Harmmol ( in press).
`12 Uc1rc.~h1 JP. Drug mc1:.1001itim b)' leukocyte$ and its rok in Jn,1g induced lupus and other
`id1o:>yncr.at1(' drug reacuons. Tox1colt>JC}' 1990 ~ 20:l 13·3S.
`
`The Cornish town ofCamelford is again in the news. In July
`1988, 20 tonnes of concentrated aluminium sulphate was
`accidentally discharged into a local reservoir. Drinking water
`was heavily polluted for up to three days, not only by the
`primary contaminants but also by copper, zinc, and lead
`di~solved from domestic plumbing. The acute effects on the
`town's inhabitants included gastrointestinal disturbances and
`oral ulceration.
`In response to persistent public concern, in January 1989
`the government appointed an independent group of experts to
`advise on the possible long term consequences of the episode.
`The advisory group, chaired by Dame Barbara Clayton,
`reported in July 1989. 1 While conceding uncertainties, it
`concluded that long term toxic effects were unlikely on the
`basis of current knowledge. It did , however, support the
`collection of further information - for example, on outcomes
`of pregnancy-as well as making recommendations about the
`handling of any similar incidents in future.
`Despite the advisory group's reassuring conclusion public
`anxiety was not allayed. Furthermore, new information
`emerged which suggested that people living in Camelford at
`the time of the accident had persistent symptoms and
`clinicopathological abnormalities. In particular, there were
`reports of raised serum aluminium concentrations several
`months after the incident, of sensitivity to aluminium in some
`people, of a high prevalence of perceived difficulties with
`memory, and of abnormalities on neuropsychological testing.
`The advisory group was therefore reconvened to examine the
`
`fresh evidence, and its second report was published two weeks
`ago. 1
`Like the first, the report is carefully considered and clearly
`reasoned, and it again concludes that long term coxicity is
`unlikely. The limitations of the new data are discussed. For
`example, much of the information derives from self selected
`and highly unrepresentative samples or is inadequately
`controlled. The advisory group accepts that the accident has
`led to real mental and physical suffering but suggests that the
`excess of reported symptoms may be attributable to anxiety
`and heightened awareness rather than a direct toxic effect.
`T here are specific recommendations for further research but
`not for the large scale epidemiological survey that some have
`advocated. This conclusion is unlikely to satisfy those who
`believe that the community has been seriously poisoned.
`The task of the advisory group was always going to be
`difficult. In some ways its situation is similar to that which
`clinicians face when presented with an anxious patient whose
`symptoms seem after routine investigation to signify nothing
`sinister. How far should one go in trying to exclude improb(cid:173)
`able diagnoses before offering reassurance? Prolonged inves(cid:173)
`tigation may throw up spurious abnormalities that increase
`the patient's conviction of major pathology. On the other
`hand, just occasionally there will indeed be serious underlying
`disease.
`Epidemiological investigation of the alleged heal th effects is
`far from straightforward. There is no reliable way of measur(cid:173)
`ing exposure to the contaminated water retrospectively.
`
`1280
`
`BMJ VOLUME 303
`
`23 NOVEMBER 199\
`
`
`2 of 2
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1019
`
`