British Journal of Haematology, 199 3. 85, 430-4 33
`
`Correspondence
`
`DEATHS CN PATIBNTS RECEIVING ORAL IRON CHELATOR Ll
`
`Agarwal el al (1992) reported four deaths in Indian thalas(cid:173)
`saemics while on the oral iron chelator Ll or shortly after
`discontinuing it. While their conclusion was that none of the
`deaths was related to Ll in any way. despite our evidence to
`the contrary in one of them (Mehta et al. 1991 ). new evidence
`(Berdoukas et al. 1993) would appear to suggest otherwise.
`Preclinical animal studies have shown thymic and other
`organ atrophy in rats at doses of I 00 mg/kg or more
`(Berdoukas et al, 1993). One of the patients from the group of
`Agarwal et al (1992), who died of overwhelming varicella
`infection in Philadelphia. was found to have thymic atrophy
`(Berdoukas et al. 1993). It is quite likely that the fatal
`varicella infection was the result of Ll -induced atrophy of
`lymphatic organs and immunosuppression.
`Another splenectomized patient who had not been vacci(cid:173)
`nated against pneumococcus and who was not on penicillin
`prophylaxis died of pyogenic meningitis. The third patient.
`who was vaccinated against pneumococcus prior to splenec(cid:173)
`tomy and was on peniciUin prophylaxis, died of an acute
`gastrointestinal infection with possible encephalitis within
`24 h of presentation. The causative organism in either case. if
`identified. was not mentioned. Even if the organism was
`pneumococcus. it is possible that LI-induced immunosup(cid:173)
`pression (and resultant hypogammaglobulinaemia) could
`have increased the patients' susceptibility to the infections.
`The death of the fourth patient was attributed to cardiac
`failure, arrhythmia. urinary sepsis. and septicaemia by
`Agarwal et al (1992). As physicians who looked after the
`patient during his terminal illness. we have reported our
`strong belief that the patient died of consequences of Ll(cid:173)
`induced systemic lupus erythematosus (Mehta et al. 199 I).
`There was no documented cardiac arrhythmia, and the
`symptoms of the urinary tract infection had improved on
`therapy prior to death. The blood pressure of 170/ 120
`recorded 10 min prior to death is incompatible with septic
`shock. Berdoukas {l 991) initially felt that the death may
`have been due to the cardiac effects of the high-dose
`methylprednisolone that the patient was receiving. but has
`now hypothesized (Berdoukas et al. 1993) that cardiac
`dysfunction may have been precipitated or potentiated by Ll(cid:173)
`induced redistribution of iron.
`We believe that there is reasonable evidence to implicate
`L 1 directly or indirectly in at least three of the four deaths. rt is
`important to make all views pubµc now because of the debate
`on the future ofLl that is sure to follow the announcement of
`Ciba-Geigy that further development of the drug is not
`justified in view of its toxicity (Berdoukas et al, l 993 ), and the
`
`subsequent statement from the International Study Group on
`Oral Iron Chelation that this is premature (Hershko. 199 3).
`
`Blood Research Centre,
`Vivina Building 3A.
`S V Road,
`Bombay 400 058,
`India
`
`REFERENCES
`
`}AYESH MEHTA
`SEEMA SINGHAL
`B. c. MEHTA
`
`Agarwal. M.B .. Gupte. S.S .. Viswanathan. C .. Vasandani. D .. Rama(cid:173)
`nathan. J .. Desai. N .. Puniyani. R.R. & Chhablani. A.T. {l 992)
`Long-term assessment of efficacy and safety of LI. an oral iron
`chelator. in transfusion dependent thalassaemia: Indian trial.
`British Journal of Haematology, 82. 460-466.
`Berdoukas. V. ( 1991) Antinuclear antibodies in patients taking Ll.
`Lancet, 337. 672.
`Berdoukas: V .. Bentley. P .. Prost. H. & Schnebli. H.P. ( 1993) Toxieity
`of oral iron chelator f,l. Lancet. 341, J 088.
`Hershko. C. (1993) Development of oral iron chelator LI. l,am:et.
`341, 1088- 1089.
`Mehta. J .. Singhal. S .. Revankar. R .. Walvalkar. A., Chablani. A. &
`Mehta. B.C. (1991) Pata! systemic lupus erythematosus in patient
`taking oral iron chelator Ll. Lancet. 337, 298.
`
`In developing countries. iron chelation is not available to
`thalassaemics because of its high cost. lnadequate blood
`transfusions and splenectomies are common. As a result.
`most thalassaemics succumb to iron overload or infections
`during their late second decade. At our centre. analysis of
`360 thalassaemics, monitored over 8 years (1983-90).
`showed cardiac failure and infections including pyogenic
`meningitis as the main mode of death. which occurred in 52
`subjects (Agarwal, 1986 ).
`Four thalassaemic patients on Ll. who died during our
`initial trial. belonged to the age group 15- I 8 years (Agarwal
`et al. 1992aJ. Their serum ferritin ranged from 8300 to
`12800 ng/ ml (Agarwal et al. 1992a). All of them were
`splenectomized and their mode of death was either iron(cid:173)
`related cardiac failure or infection and in no way different
`from thalassaemics dying elsewhere in India.
`Mehta et al have hypothesized that these deaths may be due
`to LI-related immuno-paralysis. Their hypothesis is based on
`animal work carried out by Ciba-Geigy as well as the presence
`of thymic atrophy in one of our patients who was on Ll
`(Berdoukas et al. 1993).
`
`430
`
`
`1 of 1
`
`Taro Pharmaceuticals, Ltd.
`Exhibit 1033
`
`