Quarterly Journal of Medicine, New Series 70, No. 262, pp , February 1989

Transcription

1 Quarterly Journal of Medicine, New Series 70, No. 6, pp. 6-74, February 989 The Efficacy and Toxicity of Azidothymidine (AZT) in the Treatment of Patients with AIDS and AIDS-related Complex (ARC): An Open Uncontrolled Treatment Study DRAGO STAMBUK, MICHAEL YOULE, DAVID HAWKINS, CHARLES FARTHING, DAVID SHANSON, RICHARD FARMER*, ADAM LAWRENCE, AND BRIAN GAZZARD From St Stephens Hospital, Fulham Road, London SW0 9TH, and * Westminster Hospital, London SW Accepted 5 September 988 SUMMARY The efficacy and toxicity of oral azidothymidine has been studied in 45 patients with the acquired immunodeficiency syndrome (AIDS) or AIDS-related complex (ARC). The median survival time of AIDS patients on azidothymidine was 4.5 times higher when compared to a historical AIDS group who had not received the drug. This result must be interpreted with caution because of changes in treatment of HIV infection and growing awareness of AIDS which may have led to earlier diagnosis in the group treated with azidothymidine. The mortality was significantly higher in those patients who received transfusions and was particularly high in those who were transfused before azidothymidine. There was a significant difference in the occurrence of opportunistic infections in the patients who received transfusions compared with those who did not. AIDS patients treated immediately after an episode of Pneumocystis carinii pneumonia survived significantly longer than those in whom treatment was delayed for three months or more, and longer than those who were treated with azidothymidine because of another opportunistic infection or Kaposi's sarcoma. The T4 cell median counts increased in patients treated with azidothymidine reaching a peak at the end of the fourth month of treatment in the ARC group and at the end of the first month in the AIDS group with a subsequent fall in both groups. Sixty per cent of patients were p4 viral antigen positive at the start of treatment and 9 per cent of these patients had a fall of more than 50 per cent in antigen level while 3 per cent became antigen negative following treatment with azidothymidine. The mortality in the patients where the antigen disappeared or in whom there was a major fall of more than 50 per cent in antigen level was significantly less than in those where there was no change in antigen level. Twenty-nine patients were treated with azidothymidine because of skin Kaposi's sarcoma and in 7 tumour regressed or was stable. Thirty-two per cent of patients treated with azidothymidine became anaemic. Neutropenia Oxford University Press 989

2 6 D. Stambuk and others occurred in 3 per cent of patients. Platelets increased initially after treatment but subsequently fell to thrombocytopenic levels in eight patients. Nine of patients with thrombocytopenia before azidothymidine was commenced responded with an increased platelet count. INTRODUCTION The acquired immunodeficiency syndrome (AIDS) is characterized by severe immunodeficiency, life-threatening opportunistic infections, neoplasia and eventually death. The immune defect is caused by infection with the retrovirus human immunodeficiency virus (HIV) [-3], and most attempts at treatment have been directed against its unique reverse transcriptase enzyme. Azidothymidine (3'-azido-3'-deoxythymidine) (AZT) is a thymidine analogue which inhibits HIV replication in vitro [4]. It is phosphorylated by cellular enzymes and, as 5'- triphosphate, inhibits reverse transcriptase and so stops viral DNA-helix extension [5,6]. Azidothymidine was found to be beneficial and well tolerated over a six-week period in a phase I study [7]; there was an increase in the number of T helper cells, development of delayed hypersensitivity reactions and weight gain. Recent studies demonstrated that azidothymidine decreased mortality and the frequency of opportunistic infections in a selected group of subjects with AIDS or AIDS-related complex over eight to 4 weeks of observation [8,9]. After the first report of the efficacy of azidothymidine [7], a unicentre, open, uncontrolled treatment study was started to evaluate the efficacy and safety of azidothymidine in a welldefined group of patients with AIDS and AIDS-related complex. METHODS Patients Three groups of patients were treated, those with opportunistic AIDS and those with AIDSrelated complex with T4 cell count less than 00/ml. Patients with Kaposi's sarcoma were treated only if the T4 cell count was less than 00. Subjects with AIDS were subdivided into a Pneumocystis carinii pneumonia group where treatment was started shortly after the development of the first attack of Pneumocystis carinii pneumonia, an intermediate group where treatment was started more than three months after Pneumocystis carinii pneumonia, a group with Kaposi's sarcoma only, and a group with other opportunistic infections. A historical group of 67 AIDS patients who did not receive azidothymidine and who comprise the total experience of this unit before introduction of the drug was analysed for comparative survival. Evaluation of patients All patients were shown to be HIV antibody positive by Wellcome HIV ELISA test [0] and confirmed by Abbott recombinant DNA test []. They were assessed fortnightly by full clinical history and examination, standard laboratory tests and body weight. T4 cell count and T4/T8 ratio were evaluated monthly. Serum p4 antigen levels (Abbott Laboratories) were measured initially and twice a month thereafter []. The clinical status of the patients was assessed on each visit into one of three groups; clinically improved (well and able to work); clinically stable (unable to work but able to live an independent life) and unwell and requiring support to live outside hospital. The patients with Kaposi's sarcoma were also evaluated at each visit, the number of skin

3 Azidothymidine in AIDS and AIDS-related Complex 63 lesions were counted and their size estimated. They were divided into four groups. Rapidly progressive was defined as production of new lesions at the rate of more than four per month or a greater than 50 per cent increase in size of existing lesions. The slowly progressive group included patients whose lesions increased at a lesser rate. The stable group had no increase in the number or in the size of lesions and the number and size of lesions in the regression group became smaller. A historical control group with skin Kaposi's sarcoma matched for age and T helper cell number had been evaluated in the same manner. Treatment regimen Patients received 00 mg of azidothymidine four-hourly throughout the day. Treatment was temporarily discontinued or stopped because of adverse reactions or at the request of the patient. The study medication was withdrawn if unacceptable toxic effects or a neoplasm requiring chemotherapy developed. Patients in whom an opportunistic infection developed stopped azidothymidine temporarily if therapy was required which might have resulted in serious additional toxic effects to bone marrow. Statistical analysis The product-limit method of probability of survival time was used for estimating median survival time in historical and current groups of patients (Kaplan-Meier) [3]. Stratified Wilcoxon's rank series analysis was used for analysing clinical data and x test for mortality rate comparisons. RESULTS Patient groups One hundred and forty-five patients were enrolled between 30 October 986 and 7 January 988. There were two females and one heterosexual male and the other patients were male homosexuals who were not drug abusers. One hundred and two had AIDS and 43 had AIDS-related complex. The mean observation period was 6±.5 months (range -4). Changes in drug treatment Ninety-five (65.5 per cent) patients received the same dose regimen of azidothymidine throughout their treatment, 34 (3.4 per cent) temporarily discontinued or stopped treatment and in 46 (3.7 per cent) the dosage was reduced. Haematological toxicity accounted for 4 of the drug withdrawals and all but one of the dose reductions. The drug was discontinued at the request of 0 patients who had read adverse reports and stopped because of abdmoninal pain in five (Table ). Mortality By 7 January 988 patients (4.5 per cent) had died, most commonly from extensive Kaposi's sarcoma or neurological disease (Table ). All the patients with ARC (43) or those treated within three months of the first episode of Pneumocystis carinii pneumonia (7) are alive ('early AIDS') and the mortality was confined to the 75 other patients with AIDS (/><0.00). The mortality in the group of patients with Kaposi's sarcoma only (eight of ;

4 64 D. Stambuk and others TABLE. Reasons for discontinuation of azidothymidine Reason Haematological causes Anaemia Neutropenia Pancytopenia Patient request Severe abdominal discomfort Other bone-marrow suppressive treatment Encephalitis Total No. of patients TABLE. Causes of death Cause Extensive Kaposi's sarcoma (only) 5 Neurological (dementia) 4 Sepsis Cytomegalovirus encephalitis Disseminated chickenpox Toxoplasmosis + Mycobacterium avium-intracellulare Non-Hodgkin's lymphoma Pneumocystis carinii pneumonia Cryptococcal meningitis Toxoplasmosis Mycobaclerium avium-intracellulare Unknown Total No. of patients 36.4 per cent) was similar to that in other groups, except the group who were treated early after Pneumocystis carinii pneumonia (3 of 53; 4.5 per cent). The mortality was higher in those patients who were undergoing transfusion for anaemia (5 of 47; 3.9 per cent) as a result of azidothymidine than in those who were not (six of 98; 6. per cent) {p< 0.00). Mortality was particularly high in those who had also required transfusion before treatment (five of eight; 6.5 per cent) compared to 0 of 39 (5.6 per cent) in the group who required transfusion after treatment. The time to death while receiving treatment of the former was also significantly shorter (3.3 compared to 4.9 months in the latter (p<0.05)). The time on azidothymidine was not statistically significantly longer in the transfused group of patients (6. ±.6 months) compared to the non-transfused group (5.7 ±.3 months). There was a significant 4.5-fold increase in survival for AIDS patients treated with azidothymidine compared with the historical control group with a median survival of 657 days vs. 370 days (Fig. ).

5 CUMULATIVE 00 SURVIVAL RATE Azidothymidine in AIDS and AIDS-related Complex Non-AZT AIDS patients AZT-AIDS patients 0 SURVIVAL TIME (years) FIG.. Product-limit estimates of survival time of AIDS patients who have/have not received azidothymidine. Opportunistic infections Opportunistic infections occurred in 8 of 0 (7.5 per cent) of the AIDS patients (Table 3), but in no ARC patient. Opportunistic infections developed in significantly less of the nontransfused patients (nine of 98; 9. per cent) than in those who had undergone transfusion (9 of 47; 40.4 per cent, X >P< 0.00). There was no tendency for opportunistic infections to occur more frequently with time in patients on azidothymidine. TABLE 3. Distribution of opportunistic infections in AIDS patients treated with azidothymidine Opportunistic infections No. of patients Pneumocystis carinii pneumonia Candida oesophagitis Sepsis Toxoplasma encephalitis Cytomegalovirus encephalitis Cytomegalovirus oesophagitis Mycobacterium avium complex Cytomegalovirus retinitis Cytomegalovirus pneumonia Pneumonia of unknown cause Total 3 5 3

6 66 D. Stambuk and others Kaposi's sarcoma Eight of the 9 patients (7.6 per cent) with Kaposi's sarcoma and no opportunistic infections showed regression of their lesions in six months of treatment. In nine (3 percent) they were stable, in nine (3 per cent) slowly progressive and in three (0.3 per cent) rapidly progressive (Table 4). In the historical untreated control group patients (4.4 per cent) had skin Kaposi's sarcoma which was found in the same period to be rapidly progressive and none regressed. Seven patients with skin Kaposi's sarcoma initially developed visceral disease over the period of observation and five of these died (Table 5). The transfusion requirements of the Kaposi's sarcoma group were not significantly different from groups with other AIDS diagnoses (6 of 44 vs. 6 of 58), but transfusion-dependent patients had a significantly worse prognosis (eight of 6) compared with those not transfused (two of 8, p<0.05). The mortality of the patients with Kaposi's sarcoma was higher than for other AIDS patients (0 of 44 vs. of 58), though not significantly. Clinical benefit There was a global clinical improvement in symptoms, defined as a change from one clinical grading to another, in most patients on azidothymidine (88 of evaluable patients; 7.7 per cent) over the observation period, whereas deterioration occurred in only a minority (3 of ; 0 per cent). Twenty patients (6.5 per cent) had not changed clinically. TABLE 4. The effect of azidothymidine on skin Kaposi's sarcoma in 9 patients in six months Status of disease Azidothymidine Controls group No. (%) No. (%) Fast progressing Slowly progressing 3(0.3) 9(3) (4.4) 7(4.) Stable Regressing 9(3) 8 (7.6) 0(34.5) 0 Total 9 9 TABLE 5. All Kaposi's sarcoma patients on azidothymidine Status Alive Dead Initially Skin Kaposi's sarcoma Visceral Kaposi's sarcoma 7 Developing on treatment Visceral lesions with previous skin Kaposi's sarcoma 5 Visceral or skin Kaposi's carcoma with no previous lesions 4

7 Azidothymidine in AIDS and AIDS-related Complex 67 TABLE 6. T4 cell counts time-profile in patients constantly on azidothymidine Patients Baseline Month No. of patients tested with ARC Median count Immunological data Patients Baseline Month with AIDS The number of T4 cells increased in patients on azidothymidine (Table 6, Fig. ), reaching a peak at the end of the fourth month of treatment in the ARC group and at the end of the first month in the AIDS group with a subsequent fall which was still above the baseline at the sixth month in the ARC group, but below the baseline by the third month in the AIDS group HIVp4 antigen data Levels of HIV p4 antigen were assessed in 87 patients who had been screened for p4 antigen before commencing azidothymidine (Table 7). At the start 53 (60.9 per cent) patients were HIV p4 antigen positive and 34 (39 per cent) were negative. Of 34 patients who were initially negative, only two (5.9 per cent) became positive, one within two months and the other within four months, of commencing treatment. Seventeen of the 53 (3 per cent) positive patients became persistently p4 antigen negative within.5 months (SD±.6). In a further 0 positive patients the p4 antigen level fell in three months by more than 50 per cent. Over the same period 6 patients had no marked reduction in antigen levels. The mortality over the period observation was similar in the group of patients who were antigen negative throughout (nine of 3; 8 per cent) and in the group with no marked fall (< 50 per cent) in p4 antigen level (seven of 6; 6.9 per cent). However the group of patients with a substantial fall in antigen level (> 50 per cent) and those in whom the antigen disappeared had a significantly lower mortality (one of 7; 3.7 per cent) than those who had no antigen at the start of treatment or those who had no marked fall in antigen levels (6 of 58; 7.6 per cent; p < 0.05).

8 68 D. Stambuk and others T CELL 300 MEDIAN COUNT 50 OARC AIDS MONTHS ON AZT FIG.. T4 cell median count in ARC and AIDS patients on azidothymidine. TABLE 7. The effect of azidothymidine on HIV p4 antigen levels HIV p4 Antigen Mortality (%) Mortality (%) Initial Follow up Negative positive /(50) Negative -» negative 9/3 (8) Positive [ positive (< 50%) 7/6(6.9) Positive i negative /7(5.9) Positive [ positive (> 50%) 0/0 0 6/58 (7.6) /7(3.7) Side-effects One hundred and eleven of our patients (77 per cent) reported at least one adverse effect which was attributed to azidothymidine, most usually nausea with or without abdominal symptoms and headaches (Table 8). These symptoms were usually worse during the first four weeks but in five cases (3.4 per cent) azidothymidine was stopped because of severe

9 Azidothymidine in AIDS and AIDS-related Complex 69 TABLE 8. Side effects of azidothymidine No. of patients (%) Nausea Headache Abdominal pain Myalgia Insomnia Dizziness Taste alteration 7 (49) 55 (38) 3 () 6() 4(0) 7 (5) 7 (5) abdominal discomfort, nausea and vomiting. Taste alteration appeared transiently during the second to third month of treatment and lasted two to three weeks. Biochemistry There were no significant changes in urine, or biochemical tests except for inconsistent rises of less than 00 per cent in aspartase aminotransferase levels in 3 patients ( per cent). Seven further patients had a greater and more persistent rise in serum aspartase transaminase but they were diagnosed as either having hepatitis B-related chronic active hepatitis or alcohol-related liver disease (3). Serum levels of vitamin B were lowered at least once in eight patients (5.5 per cent) but rose spontaneously without haematological toxic efects. Folate levels were normal in all but one patient. HAEMATOLOGY Haemoglobin Haemoglobin levels fell eventually in all patients on azidothymidine. Twelve patients had an initial rise in haemoglobin for the first month on treatment - those just recovering from Pneumocystis carinii pneumonia or chemotherapy for Kaposi's sarcoma. By six weeks 3 per cent of patients had a fall of more than g of haemoglobin. There was a significant increase in erythrocytes mean corpuscular volume (MCV) which could not be attributed to the deficiencies of serum vitamin B or folate, and was observed in 3 of 45 (9 per cent) patients treated with azidothymidine after the second week of therapy. The mean increase of MCV at three months was +. /mi 3, and MCV was eventually more than 00 /im 3 in 04 of the 45 patients (7.7 per cent). Leucocytes White cell count dropped in patients on azidothymidine mainly due to neutropenia (< 500 cells/mm 3 ). Low white cell counts and neutropenia were most marked in the patients with AIDS and low T4 cell counts at the start of treatment. By the third month 6 per cent of patients treated with azidothymidine had less than 000 leucocytes/mm 3 and 6 per cent had less than 000. A reduction in dose or discontinuation of the drug in four patients prevented worsening of neutropenia. No complications due to neutropenia were identified.

10 70 D. Stambuk and others Platelets Platelet numbers increased in all patients on azidothymidine with normal baseline numbers from a mean count of to at three months which was significant (p< 0.00). The increase was less marked, though substantial, in thrombocytopenic patients (less than /), one of whom had undergone splenectomy before treatment with azidothymidine but without improvement. In five of thrombocytopenic patients platelets rose within two to five weeks above and this was sustained for the observation period (- months). In a further four patients platelet counts rose to within the normal range by between three and five weeks but after a number of weeks (7-9) the count dropped below normal and in three patients below its original level before treatment with azidothymidine. Two patients with thrombocytopenia were stable while on azidothymidine with no significant change in platelet count. The count in the final patient dropped further within the first week of treatment but when the dose was reduced from 00 mg four-hourly to 00 mg four-hourly the platelet count rose to normal and remained so during the observation period. In eight patients who started with normal platelet counts these fell to thrombocytopenic levels in a median time of six months. Transfusions Forty-seven of the 45 patients (3.4 per cent) were transfused because of anaemia (Hb < 9.0 g/dl) and 34 (3.4 per cent) were transfused more than once. Transfused patients received a mean (±SD) of units per patient undergoing transfusion and 4± units per transfusion. Transfusion-dependent patients usually required 3 to 5 units of red cells every four to six weeks. The transfusion requirements increased asymptotically with time on azidothymidine (Fig. 3). Only five patients with ARC were transfused compared with 4 patients with AIDS. Although other agents potentially toxic to bone marrow were given in 57 patients there was no evidence that such treatment was given more frequently in those requiring transfusion (Table 9). All patients who were transfused before treatment with azidothymidine because of underlying disease or treatment-related anaemia needed further transfusion after azidothymidine. Mean time to subsequent transfusion following the start of AZT ( ±0.5 months) was significantly shorter than in other transfused patients (3.5 ±.5 months, p< 0.05). CONCOMITANT MEDICATION Co-trimoxazole or inhaled pentamidine or fansidar were given as prophylaxis to 57 patients who had Pneumocystis carinii pneumonia or toxoplasmosis before treatment with azidothymidine or after commencing it (Table 9). Acylovir was given to 37, ketoconazole to 53, and 5 patients were treated by chemotherapy (vincristine or bleomycin) or radiation because of Kaposi's sarcoma or lymphoma. No adverse reactions were noted. DISCUSSION Since the licensing and subsequent publication of the early experience with azidothymidine in a placebo-controlled trial in patients with AIDS diagnosed as a result of Pneumocystis carinii pneumonia and in patients with ARC with a T4 cell count of less than 400, a controlled trial in these groups of patients has not been possible [8,9]. This open prospective study confirms the value of azidothymidine in terms of survival or time from diagnosing

11 % Of transfused patients Azidothymidine in AIDS and AlDS-related Complex Length of time of AZT (months) FIG. 3. Transfusion in patients on azidothymidine. TABLE 9. Known additional treatment which was bone marrow suppressive in patients on azidothymidine. Association with transfusion Additional therapy Transfused (n = 47) Non-transfused (n = 98) On co-trimoxazole/ fia/w«/ar/pentamidine 6 On vincristine/bleomycin 7 Radiotherapy Treatment for M'ycobacterium avium-intracellulare infection Total AIDS until death occurred and numbers of opportunistic infections, compared with our historical control group. This approach has obvious limitations particularly because the historical control group were treated in different ways and may have had more serious disease as they presented earlier in the epidemic, when the index of suspicion may have been less. Nevertheless with such a newly-introduced, expensive and toxic compound confirmatory studies are important and are impossible to perform in any other way. This study was deliberately designed to treat only those who were most affected of more than 800 HIV antibody positive patients. Analysis of this study is complicated by variable observation periods and the fact that in some patients AIDS was diagnosed some months before treatment became available. However analysis of the data with a Kaplan-Meier plot has been used to compensate [3].

12 7 D. Stambuk and others The original study [8,9] only treated AIDS patients after Pneumocystis carinii pneumonia which is probably an 'early' sign of AIDS occurring before more profound disintegration of the immune system leading to other opportunistic infections [4,5]. Our experience of AIDS patients other than those who have just recovered from Pneumocystis carinii pneumonia indicates that they have a poor prognosis even when treated with azidothymidine. While it is not possible to state whether survival is prolonged compared with untreated patients it may be that survival is so short that the use of this drug with its serious side effects might not be in their best interests. In patients who received azidothymidine there was only one death from Pneumocystis carinii pneumonia which is a common cause of death in patients who have not received the drug [4,5]. Although there is no evidence that azidothymidine has a direct effect on Pneumocystis carinii pneumonia this difference is striking. Also we have previously shown [6] that the only agent associated with complete eradication of cryptosporidium from the stools is azidothymidine and indeed there were no deaths from this organism in the series reported here. Patients with Kaposi's sarcoma only were not treated in the initial American study [8,9]. Controlled trials of azidothymidine in Kaposi's sarcoma are under way but our studies would certainly indicate that regression occurred in a much larger proportion of patients than would be expected from our previous experience, although survival was not greatly prolonged. The poor survival of our patients with Kaposi's sarcoma is likely to be related to the fact that only those with pronounced immune deficit (T4 cell count below 00/ml) have been included. Patients who were anaemic before treatment were also excluded from the American study [8,9] and it can be seen from our results that transfusion before treatment was a poor prognostic feature both for survival and further opportunistic events and also for subsequent transfusion requirements. It is likely that the need for transfusion before treatment selects a group of patients who are at a later stage in the disease spectrum. It seems from this study that treatment of early disease is associated with fewer side effects and a greater improvement in survival. T4 counts in our patients were striking in this regard with a sustained improvement in ARC patients but only a transitory increase in AIDS patients, which should give further impetus to controlled trials in even earlier disease. However, a rise in MCV is almost inevitable and such trials will be difficult to design in a blind way. Our experience with transfusion requirements where there was asymptotic rise in the proportion of patients requiring transfusion over the months of treatment is different from previous studies which indicated a constant proportion of patients requiring such treatment [9,7]. Patients who needed transfusion whilst on azidothymidine had a worse prognosis and as a group they may have been more ill at recruitment and therefore more likely to develop bone marrow suppression. The cause of bone marrow suppression in patients on azidothymidine is unclear but is likely to be related to interference with DNA synthesis, probably as a cumulative effect. Although neutropenia is common it does not appear to be clinically significant. The changes in platelets are complex, with thrombocytosis being common in the early months, followed by thrombocytopenia in a small proportion of patients on further treatment. Our study shows that azidothymidine may be beneficial in treatment of AIDS-associated thrombocytopenia. There is evidence that the appearance of p4 antigen in the bloodstream is associated with poor prognosis [8]. This study again emphasizes the high proportion of patients (one in three) who have AIDS but are antigen negative. Such patients frequently die without developing antigenaemia. Previous reports have indicated that viral antigen levels diminish with azidothymidine [9-] and our study confirms this. Furthermore a new finding was that in those patients who were antigen positive the prognosis was related to whether or not

13 Azidothymidine in AIDS and AIDS-related Complex 73 there was a response of antigen to azidothymidine. The prognosis of antigen negative patients was poor and it remains to be determined whether treatment with azidothymidine is beneficial in this group. Few data are available at present on other drug interactions with azidothymidine. About half of the the patients in this study were receiving concomitant therapy without identifiable additional side effects. The precise role of azidothymidine in the treatment of AIDS remains to be determined. The drug is more toxic and is associated with a higher mortality in many subgroups of patients with AIDS than might be inferred from the inital studies [8,9]. ACKNOWLEDGEMENTS We thank Carolyn Ball for her secretarial assistance. REFERENCES. Barre-Sinoussi F, Chermann JC, Rey F, et al. Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome (AIDS). Science 983; 0: Popovic M, Sarngadharan MG, Read E, Gallo RC. Detection, isolation and continuous production of cytopathic retroviruses (HTLV-III) from patients with AIDS and pre-aids. Science 984; 4: Gallo RC, Salahuddin SZ, Popovic M, et al. Frequent detection and isolation of cytopathic retroviruses (HTLV-III) from patients with AIDS and at risk for AIDS. Science 984; 4: Mitsuya H, Weinhold KJ, Furman PA, et al. 3'-azido-3'-deoxythymidine (BW A5O9U): an antiviral agent that inhibits the infectivity and cytopathic effect of human T-lymphotropic virus type III/lymphadenopathy-associated virus in vitro. Proc Natl Acad Sci USA 985; 8: St Clair MH, Weinhold K,Richards CA, et al. Characterization of HTLV-III reverse transcriptase and inhibition by the triphosphate of BW A509U. In: Program and abstracts of the Twenty-fifth Interscience Conference on Antimicrobial Agents and Chemotherapy, Minneapolis, October 985. American Society of Microbiology, 985: 7, abstract. 6. Furman PA, St. Clair M, Weinhold K, et al. Selective inhibition of HTLV-III by BW A509U. In: Program and abstracts of the Twenty-fifth Interscience Conference on Antimicrobial Agents and Chemotherapy, Minneapolis, October 985. American Society of Microbiology, 985: 7, abstract. 7. Yarchoan R, Klecker RW, Weinhold KJ, et al. Administration of 3'-azido-3'-deoxythymidine, an inhibitor of HTLV-II/LAV replication, to patients with AIDS or AIDS-related complex. Lancet 986; : Fischl MA, Richman DD, Grieco MH, et at. The efficacy of 3'-azido-3'-deoxythymidine (azidothymidine) in the treatment of patients with AIDS and AIDS-related complex: a doubleblind placebo-controlled trial. N Engl J Med 987; 37: Richman DD, Fischl MA, Grieco MH, et al. The toxicity of azidothymidine (AZT) in the treatment of patients with AIDS and AIDS-related complex: a double-blind, placebo-controlled trial. N Engl J Med 987; 37: Wellcome Diagnostics. Wellcozyme anti-htlv III. (Enzyme immunoassay for detection of antibody to Human T cell lymphotrophic type III [HTLV III]. Lymphadenopathy Virus [LAV]. January Mortimer PP, Parry JV, Mortimer JY. Which anti-htlv I/LAV assays for screening and confirmation testing? PHLS Virus Reference Laboratory, Central Public Health Surveillance Centre, London NW9. Lancet 985; : Goudsmit J, de Wolfe F, Paul DA, et al. Expression of human immunodeficiency virus antigen (HIV-Ag) in serum and cerebrospinal fluid during acute and chronic infection. Lancet 986; : Kaplan EL, Meier P. Non-parametric estimation from incomplete observations. J Am Stat Assoc 958; 53:

14 74 D. Stambuk and others 4. Update on Acquired Immune Deficiency Syndrome (AIDS) - United States. MMWR 98; 3: Update: Acquired Immune Deficiency Syndrome (AIDS) - United States. MMWR 98; 3: Connolly GM, Dryden MS, Shanson DC, el al. Cryptosporidial diarrhoea in AIDS and its treatment. Gut 988; 9(5): Walker RE, Parker RI, Kovacs JA, el al. Anaemia and erythropoiesis in patients with the acquired immunodeficiency syndrome (AIDS) and Kaposi's sarcoma treated with zidovudine. Ann Intern Med 988; 08: Allain J-P, Laurian Y, Paul DA, et al. Long term evaluation of HIV antigen and antibodies to p4 and gp4 in patients with hemophilia. N Engl J Med 987; 37: Chaisson RE, Allain J-P, Leuther M, et al. Significant changes in HIV antigen level in the serum of patients treated with azidothymidine. (Letter) N Engl J Med 986; 35: de Wolf F, Lange JM A, Goudsmit J et al. Effect of zidovudine on serum human immunodeficiency virus antigen levels in symptom-free subjects. Lancet 988; : Reiss P, Lange JMA, Boucher CA, et al. Resumption of HIV antigen production during continuous zidovudine treatment. (Letter) Lancet 988; : 4.