Reliance on one methodology without verification via other
`important sources of data, such as the Somali Red Crescent death
`cart, greatly reduced the usefulness of Moore and colleagues’ visit.
`Preliminary analysis of data taken at the Concern Worldwide adult
`therapeutic centre in Baidoa underlines the poor state of knowledge
`about adult malnutrition. The average body mass index of the first
`650 severely malnourished adults and adolescents was 13. This is
`substantially lower than 16, recently proposed as the level to mark
`the most severe grade of adult starvation.3 Others survived with a
`body mass index below 10 on admission. Famine oedema occurred
`in 30% of admissions and was associated with a poorer prognosis.
`
`Concern Worldwide,
`248-250 Lavender Hill,
`London SW11 1LJ, UK
`
`STEVE COLLINS
`
`1. Mercer A. Medecins Sans Frontieres (Holland), Health surveillance review, Baidoa.
`Somalia, 1993.
`2. Collins S. Concern Worldwide Survey of nutrition and food distribution in North
`West Baidoa district February 1993.
`3. Ferro-Luzzi A, et al. A simplified approach to assessing adult chronic energy
`deficiency. Eur J Clin Nutr 1992; 46: 173-86.
`
`SIR,-Your April 23 editorial "Do epidemiologists cause
`epidemics?" brought to mind a similar speculation: "Do
`nutritionists cause mass starvation?" Your April 9 issue carries a
`paper by Dr Moore and colleagues on microdemographic research
`in Somalia undertaken by the Centers for Disease Control (CDC).
`This study detailed high mortality among two populations of
`displaced people, with death rates of 4-7 and 16-8 per 10 000 per
`day, respectively, over a period of 8 months. These rates are high,
`but well within the recorded range for famine relief shelters,
`particularly in view of the fact that the displaced people in Baidoa,
`who suffered the higher rate, were the remnant of a larger
`population subject to much out-migration. As in all recorded
`famines, the overwhelming causes of death were preventable
`infectious diseases such as measles and diarrhoea.
`However, when these results were first announced in December,
`1992/ the media claimed the rate to be the highest ever recorded and
`attributed the deaths exclusively to starvation. This was because, in
`addition to the 8 month death rate, CDC included death rates in the
`month before, when, in Baidoa, 5 children out of 21 had
`died-equivalent to 69 4 per 10 000 per day. These five unfortunate
`children received a wholly unwarranted international notoriety:
`"Somalia has amongst the highest death rate from starvation ever
`recorded... deaths among children under five in one village reached
`69-4 per day for every 10 000 children-more than three times that
`of the 1984-85 Ethiopian famine".2 This claim is absurd, but,
`coming on the day after US Marines stormed ashore in Somalia
`with the stated aim of saving two million Somalis from death by
`starvation, it helped justify that unprecedented action, and also the
`continuing preoccupation with delivering food among most of the
`intematinal agencies active in the country.
`By December, 1992, the price of staple grain in Somalia’s markets
`had fallen below the cost of production, and farmers were beginning
`to complain that the imported food aid was making it impossible for
`them to sell their harvests. Two million Somalis did not face death
`by starvation; the food scarcity was almost over. Nonetheless, relief
`agencies continued to open feeding centres and distribution points:
`the World Food Programme opened 35 distribution points in
`Mogadishu in February alone. Meanwhile, contrary to the
`warnings of Somali health professionals, the public health
`requirements of the country have remained largely neglected.
`Measles, malaria, tuberculosis, dysentery, and acute respiratory
`infections are still killing large numbers of Somali children. As a
`result, we concluded in our recent evaluation of Operation Restore
`Hope3 that the international intervention had done little to save lives
`in Somalia.
`
`African Rights,
`11 Marshalsea Road,
`London SE1 1EP, UK
`
`ALEX DE WAAL
`RAKIYA OMAAR
`
`1. Population-based mortality assessment—Baidoa and Afgoi, Somalia, 1992. MMWR
`1992; 41: 913-17.
`2. The Guardian London, UK. Dec 11, 1992.
`3. Operation Restore Hope: a preliminary assessment. London: African Rights, 1993.
`
`1479
`
`Informed consent
`
`SIR,—Your news item on informed consent (May 1, p 1141) can
`be read as giving currency to the notion that there are grades of
`consent--consent and informed consent. There are not.
`The Royal College of Physicians explains in its guidelines:1
`"Potential research subjects are entitled to choose whether or not
`they will participate in research, and obtaining valid (informed,
`understanding, voluntary) consent is central to the ethical conduct
`of clinical investigation. The terms ’valid’, ’informed’ and
`’voluntary’ imply that subjects have enough information, in a form
`that is comprehensible, to enable them to make an autonomous,
`deliberated (proper) judgement whether or not to participate. The
`word ’consent’ encompasses these requirements, for if they are not
`met there is no consent. The use of qualifying adjectives is
`unnecessary and may even be confusing. Therefore the word
`’consent’ will be used without qualification in this document".
`However, the term informed consent is now standard usage and it
`is a continuing reminder that participants in research trials must be
`given adequate information. But there is no such thing as
`uninformed consent. Consent is consent is consent, as the late
`Gertrude Stein might have said.
`37 Denning Road,
`London NW3 1ST, UK
`
`D. R. LAURENCE
`
`1. Royal College of Physicians. Guidelines on the practice of ethics committees in medical
`research involving human subjects. London: RCP, 1990.
`
`Future of oral iron chelator deferiprone (L1)
`SIR,—The decision by Ciba-Geigy to stop development of the
`oral iron chelator L 1 (deferiprone) has already prompted a response
`from the International Study Group on Oral Iron Chelators (April
`24, p 1088). After the 4th International Conference on Oral
`Chelators (ICOC) in Cyprus on March 26-29, at which the
`controversy was aired, the following recommendations were made
`by the ICOC committee.
`Some patients receive no chelation therapy, because of non-
`compliance with desferrioxamine or cost. Over 90% of the world’s
`patients with thalassaemia cannot afford desferrioxamine. The
`options seem to be that these patients receive no treatment or that,
`where clinical trials are feasible with Ll, these should be set up and
`patients recruited, with the permission of the local health authorities
`or, in view of the vast amount of information on this drug,l,2
`attempts should be made to assist organisations or companies to
`register and use this drug under surveillance. Regularly transfused
`patients are at risk, in the absence of chelation therapy of irreversible
`organ damage due to iron overload. Marketing of Ll under
`controlled clinical use should be considered once preclinical and
`clinical data satisfy the local health authorities, and this should also
`apply where desferrioxamine is freely available but cannot be used
`because of toxicity or non-compliance.
`Collected data on all aspects of Ll, including chemistry,
`toxicology, and pharmacology, should be made available, and
`information on adverse effects of L and other iron chelators should
`be sent to the WHO Collaborating Centre for International Drug
`Monitoring, Uppsala, Sweden.
`The ICOC committee took this view in the light of preclinical
`studies and clinical investigations in over 450 patients who have
`taken Ll in fifteen countries. Some patients have taken L1 daily for
`over 4 years, over 100 having taken it daily for more than 3 yeras.3
`All the centres have reported that compliance with Ll is high, iron
`excretion is comparable with that caused by desferrioxamine, while
`liver iron and serum ferritin is being reduced substantially in most
`cases after 6 months to a year. Side-effects during the trials have
`usually been transient. They include 5 cases of transient
`agranulocytosis, transient musculoskeletal/joint pains in about
`20%, gastric intolerance in 6%, and zinc deficiency in 2%. Most
`toxic effects were associated with daily doses of 100 mg/kg but were
`substantially reduced by lowering the dose to 75 mg/kg. Although a
`few patients have had to stop taking Ll because of side-effects no
`clinical trial has been suspended because of toxicity or lack of
`efficacy. The therapeutic margin, in animal models and in patients,
`does not seem to be different for Ll and desferrioxamine.
`
`Apotex Tech.
`Ex. 2021
`
`

`

`1480
`
`Controlled clinical use of L1 should continue and WHO, the
`Thalassaemia International Federation, and other interested parties
`should be encouraged to get involved with the development of Ll
`until a parent for this orphan drug is found. It may prove possible to
`control the toxicity of L in future by identifying the mechanisms of
`the adverse effects and the patients who may be susceptible to such
`toxicity.
`
`Department of Haematology,
`Royal Free Hospital School
`of Medicine,
`London NW3 2QG, UK
`
`GEORGE J. KONTOGHIORGHES
`M. B. AGARWAL PETRIGN TONDURY
`MICHAEL JAEGER
`M. J. KERSTEN
`G. VREUGDENHIL A. VANIA
`Y. E. RAHMAN, for the ICOC committee
`
`1. Oral chelation in the treatment of thalassaemia and other diseases. Drugs Today 1992;
`28 (suppl A): 1-187.
`2. Kontoghiorghes GJ. Advances in oral iron chelation in man. Int J Haematol 1992; 55:
`27-38.
`3. Abstracts of 4th International Conference on Oral Chelators for the Treatment of
`Thalassaemia and Other Diseases. March, 26-29, 1993, Limassol, Cyprus.
`
`SIR,—The International Study Group on Oral Iron Chelators
`disagrees with Ciba-Geigy’s decision to stop. They question the
`relevance of toxicity studies in normal (non-iron-loaded) animals to
`the situation in iron-overloaded patients. This argument seems
`logical at first glance but proves incorrect once the safety of Ll is
`compared with that of desferrioxamine. Pharmacological and
`toxicological studies in rats show that there is no safety margin for
`Ll, while desferrioxamine displays an acceptable safety margin:
`Subcutaneous
`desferrioxamine
`(mg/kg)
`
`Oral L 1
`(mg/kg)
`
`50
`
`> 100
`
`Dose effecting excretion of
`0-5 mg Fe/kg/day (single day
`pharmacology)
`Dose causing serious bone
`marrow lesions (3-month
`toxicity study)
`Dose causing:
`> 500t
`serious maternal toxicity
`100
`100
`> 500t
`serious embryotoxicity
`No effect#
`serious teratogenicity
`25
`*Cataract formation 3/30 at 250 mg/kg and 14/30 at 500 mg/kg, but no
`bone marrow effects at maximum tested dose of 500 mg/kg.
`t1096 of animals at 540 mg/kg.
`*Highest dose tested 540 mg/kg
`Desferrioxamine is impermeable to most cells whereas Ll, which
`rapidly penetrates all cells and tissues, is toxic exactly in its effective
`dose range.
`Porter et all have shown that iron overload does not protect
`against bone marrow toxicity of L 1. In a 60-day toxicity study in
`iron-overloaded and non-overloaded mice, Ll (200 mg/kg) but not
`desferrioxamine at the same dose, caused significant decreases in
`white cell counts and haemoglobin. These effects were identical in
`iron-loaded and non-loaded mice, though there was a difference in
`mortality (not significant).
`The major toxic effect in animals (bone marrow depression and
`decreased white cell counts) has been reported in 5 patients. The
`frequency of reporting of this side-effect is lower than predicted
`from animal studies, but there is no "sharp contrast between serious
`toxicity in normal animals and the limited toxicity in iron-
`overloaded patients". In none of the 5 cases of agranulocytosis did
`the patient have an especially low iron load, and the most frequent
`side-effect with L1 (arthropathy) seems to occur mainly in heavily
`loaded patients.
`It is the lack of a therapeutic margin of Ll and the consistency of
`the toxicity pattern in all animal species investigated, including
`primates, which has led us to discontinue the development of this
`agent.
`
`No effect*
`
`100
`
`Pharmaceuticals Division,
`Ciba-Geigy Ltd,
`CH-4002 Basel, Switzerland
`
`FRIEDLIEB PFANNKUCH
`PHIL BENTLEY
`HANS PETER SCHNEBLI
`
`1. Porter JB, Hoyes KP, Abeysinghe RD, et al. Comparison of the subacute toxicity and
`efficacy of 3-hydmacypyridin-4-one iron chelators in overloaded and non-
`overloaded mice. Blood 1991; 78: 2727-34.
`
`SIR,-In view of the serious humanl-4 and animal toxicity
`reported with Ll we are surprised that the International Study
`Group on Oral Iron Chelators feels that Ciba-Geigy’s
`discontinuation of further development of this agent is
`"premature". Before their appeal for further trials is considered it
`may be worthwhile drawing attention to the inconsistencies and
`omissions in published trial reports.’
`Olivieri et al5 did not describe any immunological abnormalities
`in their paper but in response to our report of drug-induced lupus8
`they noted antinuclear antibodies (ANA) in 5 of 12 thalassaemic
`patients with absent antibodies to nuclear histones (AHA) before
`starting L 1.9 The evolution of these abnormalities during and after
`Ll administration was not described. Al-Refaie et al4 reported
`ANA in 2 of 12 patients before L 1, and in 4 of 12 after conclusion of
`the trial. AHA, which are more specific for drug-induced lupus,
`were not mentioned. Agarwal et alb reported no significant toxicity
`on "extensive" monitoring of the immune system but they later
`noted ANA and AHA in L 1 recipients and in thalassaemic patients
`not taking L1.3,7 Our work, on patients from the same
`socioeconomic and ethnic background in Bombay, shows that
`against some background ANA positivity in thalassaemias, there is a
`significantly higher incidence in patients taking Ll,2 We detected
`AHA in patients receiving Ll but not in those not on the drug.
`Infection-related deaths of 3 patients in India while on Ll or
`shortly after discontinuation, while felt by Agarwal et al3 to be
`unrelated to the drug, need to reassessed in view of the atrophy of
`lymphatic organs seen in animal studies (April 24, p 1088).
`
`Blood Research Centre,
`Vivina Building 3A,
`S V Road,
`Bombay 400 058, India
`
`JAYESH MEHTA
`SEEMA SINGHAL
`B. C. MEHTA
`
`1. Veys PA, Wilkes S, Al-Refaie FN, et al. The mechanism of agranulocytosis mediated
`by the oral iron chelator L1. Br J Haematol 1993; 84 (supp 1): 64.
`2. Mehta J, Chablani A, Reporter R, Singhal S, Mehta BC. Occurrence of autoantibodies
`in thalassemia major and their possible relationship with oral iron chelator L1. Br J
`Haematol 1993; 84 (suppl 1): 64.
`3. Agarwal MB, Gupte SS, Viswanathan C, et al. Long-term assessment of efficacy and
`safety of L1, an oral iron chelator, in transfusion dependent thalassaemia: Indian
`trial. Br J Haematol 1992; 82: 460-66.
`4. Al-Refaie FN, Wonke B, Hoffbrand AV, Wickens DG, Nortey P, Kontoghiorghes
`GJ. Efficacy and possible adverse effects of the oral iron chelator 1,2-dimethyl-3-
`hydroxypyrid-4-one (L1) in thalassemia major. Blood 1992; 80: 593-99.
`5. Olivieri NF, Koren G, Hermann C, et al. Comparison of oral iron chelator L1 and
`desferrioxamine in iron-loaded patients. Lancet 1990; 336: 1275-79.
`6. Agarwal MB, Gupte SS, Viswanathan C, et al. Phase II trial of 1-2,dimethyl-3-
`hydroxypyrid-4-one (L1) the oral iron chelator in 52 patients of transfusion
`dependent thalassaemia. Blood 1990; 76 (suppl 1): 52a.
`7. Agarwal MB, Gupte SS, Viswanathan C, et al. Effective oral iron chelation in patients
`of thalassemia. In: Proceedings of the International Symposium cum Workshop on
`Anemia in Children (Bombay, February, 1991): 13.1-13.13.
`8. Mehta J, Singhal S, Revankar R, Walvalkar A, Chablani A, Mehta BC. Fatal systemic
`lupus erythematosus in patient taking oral iron chelator L1. Lancet 1991; 337: 298.
`9. Olivieri NF, Koren G, Freedman MH, Roifmart C. Rarity of systemic lupus
`erythematosus after oral iron chelator L1. Lancet 1991; 337: 924.
`
`Absence of association between HLA-DR7
`and cytomegalovirus infection in renal
`transplant patients
`SIR,-Dr Kraat and colleagues (Feb 20, p 494) report a higher
`frequency of cytomegalovirus (CMV) infection in HLA-DR7-
`positive recipients of kidney grafts. Since 1987 we have been
`prospectively following 206 consecutive patients undergoing renal
`transplantation, with the aim of detecting CMV infection or disease.
`We have been unable to confirm Kraat and co-workers’ findings.
`Virological studies included conventional and shell vial cultures
`in blood and urine, measurement of antigenaemia by indirect
`immunofluorescence with monoclonal antibodies in buffy coat,’
`and detection of IgG and IgM antibodies (complement binding and
`enzyme-linked immunoabsorbent assay). Specimens were obtained
`preoperatively, weekly during the first 3 months, and then at each
`outpatient visit until 12 months after transplantation. CMV
`infection was defmed as CMV isolation in any specimen with or
`without seroconversion or a substantial increase in IgG antibodies.
`CMV disease was diagnosed if at least two tests were positive and
`the patients had two or more symptoms (fever, leucopenia,
`hepatitis, pneumonitis, or intestinal damage without other cause).
`
`Apotex Tech.
`Ex. 2021
`
`