`
`PHASE-I CLINICAL TRIAL OF
`MONOCLONAL ANTIBODY IN TREATMENT
`OF GASTROINTESTINAL TUMOURS
`
`HENRY F. SEARS
`JEFFREY MATTIS
`DOROTHEE HERLYN
`PEKKA HÄYRY
`
`BARBARA ATKINSON
`CAROLYN ERNST
`ZENON STEPLEWSKI
`HILARY KOPROWSKI
`
`The Fox Chase Cancer Center, Philadelphia, Pennsylvania, U.S.A.;
`Department ofPathology, Hospital of the University of Pennsylvania,
`Philadelphia; Centocor Inc., Malvern, Pennsylvania; and The
`Wistar Institute of Anatomy and Biology, Philadelphia
`
`Summary
`
`of a murine
`A phase-I
`clinical
`trial
`specifically
`antibody
`that
`monoclonal
`suppresses growth of human gastrointestinal tumours in
`athymic mice was conducted in four patients, who were given
`15-200 mg purified antibody. The monoclonal antibody
`persisted in the circulation for more than a week when more
`against mouse
`than
`15 mg was given.
`Antibodies
`immunoglobulin developed in three of the four patients. In
`one patient who received autologous mononuclear cells that
`had been mixed with monoclonal antibody by way of a
`hepatic-artery catheter, hepatic metastases became smaller
`and their echogenic characteristics changed, and there was
`heavier monocyte infiltration in the histological appearance
`of a resected metastasis.
`
`Introduction
`WE have developed a series of monoclonal antibodies that
`bind selectively to malignant cells of human gastrointestinal
`tract tumours.l,2 One of these antibodies, secreted by
`hybridoma 1083-17-1A (antibody 17-lA), mediates lysis of
`colorectal carcinoma cells by human or mouse effector cells3
`and specifically
`inhibits the growth of human colon
`carcinomas xenografted in athymic (nu/nu) mice.4 The
`antigen detected by antibody 17-lA is not shed during culture
`by tumour cells.5
`Monoclonal antibody 17-lA perfused through freshly
`resected human colons containing adenocarcinomas binds
`selectively to cells of some of these tumours.6 We have used
`this purified antibody in a phase-I clinical trial to assess its
`persistence in the systemic circulation, binding to tumour
`tissue, toxicity, and immunogenicity.
`
`Patients and Methods
`
`Patients
`Four patients with metastatic gastrointestinal cancer, scheduled
`for palliative surgery at Fox Chase Cancer Center, gave informed
`consent to take part. Two patients had one ureter obstructed by
`tumour, three had hepatic metastases, and one had only local pelvic
`
`28. Linos DA, Beard CM, O’Fallon WM, Brockerty MB, Beart RW, Kurland LT.
`Cholecystectomy and carcinoma of the colon. Lancet 1981; ii: 379-81.
`29. Vernick LJ, Kuller LH. Cholecystectomy and rightsided colon cancer. an
`epidemiological study Lancet 1981; ii. 381-83.
`30. Liu K, Stamler J, Moss D, Garside D, Perskey V, Soltero I. Dietary cholesterol, fat,
`fibre and colon cancer mortality. An analysis of international data. Lancet 1979; ii:
`782-85.
`31. Cruse P, Lewin M, Clark CG. Dietary cholesterol is co-carcinogenic for human colon
`cancer. Lancet 1979; i: 752-55.
`32 Hepner GW. Effect of decreased gallbladder stimulation on enterohepatic cycling and
`kinetics of bile acids Gastroenterology 1975; 68: 1574-81.
`33. Northfield TC, Hofmann AF. Biliary lipid secretion in gallstone patients. Lancet 1973;
`i: 747-48.
`
`recurrence. One patient, who died 2 months after surgery, had liver
`metastasis, obstructed ureter and colon, and an enterovaginal fistula
`after radiation therapy. Patient 4, a 54-year-old man, underwent
`subtotal gastrectomy for a poorly differentiated adenocarcinoma of
`Serial computerised
`trial.
`the stomach 6 weeks before this
`tomography (CAT) scans showed enlarging liver metastases.
`Preparation of Monoclonal Antibody
`against human colorectal
`antibody
`Murine
`monoclonal
`carcinoma (antibody 17-lA) of y2a isotype has been described
`previously.1,2 Ascitic fluid was collected aseptically, was allowed to
`clot at 37 °C, and was then centrifuged and filtered under sterile
`conditions through 0 . 22 j-tm ’Millex’ filters (Millipore, Bedford,
`Massachusetts). The filtrate was diluted with an equal volume of
`sterile 0-11 mol/1 "tris"-buffer, pH 8-0, and applied to a sterile
`’protein-A-Sepharose’ (Pharmacia, Piscataway, New Jersey) column
`(10 ml) for isolation ofthe IgG2a immunoglobulin. The column was
`then washed thoroughly with 0.1mol/1 "tris" buffer, pH 8 - 0; the
`adsorbed IgG2a was eluted with 0 - 11 mol/1 citrate, pH 4 - 5. The pH
`of the eluate was adjusted to neutral, and the eluate was dialysed
`against saline. The immunoglobulin was judged to be 95% pure in
`sodium-dodecyl-sulphate/polyacrylamide-gel electrophoresis and
`gave negative results in the Limulus amoebocyte lysate assay (M.A.
`Bioproducts, Bethesda, Maryland) at a concentration of 500 g/m1.
`The immunoglobulin was quantified by absorbance at 280 nm.
`Treatment with Monoclonal Antibody
`hypersensitivity
`for
`Patients
`tested
`to
`mouse
`were
`immunoglobulin; patient 1 received 15 mg purified, pyrogen-free
`monoclonal antibody 17-lA intravenously, and patients 2 and 3
`received 180 mg and 150 mg, respectively. Patient 4 was given a first
`injection of 200 mg antibody intravenously on day 0. On day 1
`mononuclear cells (approximately 7 - 5 x 10 ), separated from one
`unit ofhis blood by gradient centrifugation, were incubated with 67
`mg antibody 17-lA for 30 min at room temperature and returned to
`patient 4 by way of a hepatic-artery catheter. He was given a further
`38 mg antibody 17- 1A on day 3 and another 30 mg on day 7; the final
`injection was attempted on day 10, but the patient received only half
`the 30 mg dose.
`Detection of Immunoglobulins
`Radioimmunoassay was carried out as described before.I,2,5 To
`detect mouse immunoglobulin, rabbit anti-mouse-IG antibody
`was exposed to patients’ serum or urine samples, and the binding
`was determined by 1-labelled rabbit
`anti-mouse F(ab’)z
`immunoglobulin. Circulating specific anti-colorectal-carcinoma
`activity of mouse immunoglobulin in patients’ serum was detected
`with live SW 1116 colon carcinoma cells as the target cells.7 To
`detect human antibodies against mouse immunoglobulin in
`patients’ serum, mouse monoclonal anti-colon-carcinoma antibody
`was allowed to react with patients’ serum samples, and 121 I-labelled
`rabbit antibodies against human F(ab’)z immunoglobulin were used
`to detect the binding.
`Immunoperoxidase Assay with Monoclonal Antibodies
`The immunoperoxidase assay was carried out by the method of
`Kolcher et al. 8 Fixed, deparaffinised tissue samples were assayed for
`binding with monoclonal antibodies (see table):
`17-1A, 19-9
`
`BINDING OF MONOCLONAL ANTIBODIES AS DETECTED BY
`IMMUNOPEROXIDASE ASSAY ON TISSUE SPECIMENS FROM
`PATIENT 4
`
`1 of 4
`
`BI Exhibit 1028
`
`
`
`763
`
`The serum of patient 4 also showed a strong binding
`capacity of circulating mouse monoclonal antibody 17-lA to
`cultured colon carcinoma SW 1116 cells (fig. 2). The binding
`increased greatly immediately after each administration of
`the monoclonal antibody (except on day 7 when no blood
`sample could be obtained). The highest values for binding of
`circulating monoclonal antibody to cultured SW 1116 cells
`by sera of patients 2 and 3 were observed 1-3 days after
`administration of monoclonal antibodies.
`In patients 1, 2, and 3 there were no immediate or delayed
`side-effects
`after administration of murine monoclonal
`antibody 17-1A. It was not possible to measure the effect of a
`single injection of monoclonal antibodies on the tumour
`because of the need for surgical intervention. Since we were
`seeking data to indicate the lack of adverse effects of murine
`immunoglobulins, we will not give a detailed description. In
`patient 4, however, the study was extended to include an
`therapeutic
`repeated
`effect
`of the
`after
`evaluation
`administration ofmonoclonal antibody, given either alone or
`together with, the patient’s peripheral-blood mononuclear
`cells.
`Levels of carcinoembryonic antigen were normal, and
`levels of a circulating tumour antigen detected with antibody
`monosialoganglioside9,10 were high and
`against
`19-9
`remained high during the course of immunotherapy in
`patient 4.
`Patient 4 was given a first injection of 200 mg purified
`antibody 17-1A intravenously over 30 min. The next day (day
`1) the mixture ofmononuclear cells and antibody was infused
`through a hepatic-artery catheter over 15 min. Small
`aggregates in the preparation were noted towards the end of
`the injection. The flow in the hepatic artery, which was
`sluggish at the start of the infusion, temporarily stopped at
`the end of the infusion. The next day (day 2) the patient’s
`temperature was 38’5°C and he complained of right
`epigastric discomfort and hiccups. Abdominal examination
`was unremarkable; however, the patient’s serum aspartate
`aminotransferase (AAT) level, which had been 86 IU on
`admission, rose to 259 IU then fell rapidly to 195 IU by that
`afternoon and continued to decrease throughout the
`remainder ofthe treatment. Lactic dehydrogenase levels also
`increased (to 429 IU) on day 2 but were almost normal by day
`7.
`
`On day 3, 38 mg 17-lA antibody was given intravenously.
`At laparotomy on day 4 three hepatic metastases with
`surrounding normal hepatic parenchyma were resected, and
`the nodal metastases in the retrocaval area were biopsied. On
`day 7 another 30 mg antibody was given. On day 10 the
`patient received less than half of the 30 mg dose, since he
`became flushed and complained of mild bronchospasm.
`Symptoms were relieved when administration of antibody
`was discontinued, and 0-33 ml adrenaline (1:10000) was
`given intravenously. 4 days later the clavicular metastasis was
`biopsied. The patient showed no signs of serum sickness
`when examined for the next 2 weeks as an outpatient. He had
`no proteinuria, and renal function was normal. Liver
`ultrasound examination 3 weeks after administration of
`monoclonal antibodies showed that the metastases were
`much smaller, and their echogenic characteristics had
`changed. No change was noted in the bone metastasis during
`the same period.
`The material from the original gastric resection ofpatient 4
`showed a poorly differentiated adenocarcinoma that widely
`infiltrated the mucosa, muscularis, and serosa. The linitis
`plastica invasion of the tumour extensively involved nerves
`
`1—Presence of mouse immunoglobulin (-)and of
`Fig.
`antibody against mouse immunoglobulin in serum of
`patients 1 (0), 2 (A) and 3 ( f ).
`
`(directed against a monosialoganglioside present in serum of
`patients with gastrointestinal tumours),9no antibody 10-17 with Leb
`specificity,li and 29-1 (raised against freshly isolated gastric
`carcinoma cells and directed against [a-1,3] fucosyl-p-globoside
`present in gastrointestinal tumour cells). Immunoglobulin of
`P3 x 63Ag8 mouse myeloma was used as a control.
`
`Results
`Mouse immunoglobulin was found in the circulation of
`patient 1 for only 48 h after he received 15 mg monoclonal
`antibody (fig. 1). Mouse immunoglobulin was detectable for
`considerably longer in the blood of patients 2 and 3, who
`received 180 mg and 150 mg antibody, respectively (fig. 1).
`against mouse immunoglobulin were first
`Antibodies
`detected 6 to 8 days after treatment in patients 1 and 3 and
`reached peak levels 11 - 14 days after treatment; they dropped
`to zero 110 and 50 days after treatment, respectively, in
`patients
`1 and 3 (fig.
`1). Antibodies against mouse
`immunoglobulin did not develop in patient 2 during the 40
`days after treatment. Patient 4, who received repeated
`injections of antibody 17-lA (fig. 2), had the highest serum
`levels of mouse immunoglobulin 24 h after administration of
`monoclonal antibody. The injections 3 and 10 days after
`treatment were followed by rises in circulating mouse
`immunoglobulin: the levels then fell in a linear fashion from
`the 11th until the 14th day after treatment. Antibodies against
`mouse immunoglobulin were first detected 9 days after
`treatment and increased steadily throughout 21 days after
`treatment (fig. 2).
`
`Fig. 2-Presence of mouse immunoglobulin (0-0) and of
`antibody against mouse immunoglobulin (0""0) and binding
`activity to SW 1116 target cells (————) in serum of patient 4.z
`
`2 of 4
`
`BI Exhibit 1028
`
`
`
`764
`
`and vascular spaces. The liver metastasis resected 4 days after
`the first infusion of antibody was a well-defined nodule with
`total necrosis in the centre. At the periphery of the nodule
`there was a rim ofviable tumour cells interspersed among an
`inflammatory infiltrate composed mainly of mononuclear
`cells. Sections from the bone metastasis resected 14 days after
`the first infusion of antibody showed bone marrow with
`foci
`of
`poorly
`metastatic
`differentiated
`several
`adenocarcinoma similar to the primary tumour but without
`tumour necrosis and only focal mononuclear inflammatory
`infiltrate.
`Sections from the stomach tumour and liver and bone
`metastases were studied by immunoperoxidase assay for
`binding of four monoclonal antibodies with a variety of
`specificities. Antigen detected by monoclonal antibody
`17-lA was present on the tumour cells of the primary gastric
`carcinoma and in the liver and bone (clavicle) metastases (see
`table); the staining of the specimens, however, was very weak.
`Antigens detected by antibody 10-17 (which defines an Lebb
`specificity") and antibody 29-1 (against [a-1,3] fucosyl-p-
`globoside) were present in all three specimens, and the
`immunoperoxidase
`The
`reaction
`strong.
`very
`was
`detected by monoclonal
`monosialoganglioside
`antigen
`antibody 19-99,10 was expressed by the tumour cells of the
`primary stomach lesion (fig. 3) and its liver metastasis (fig. 4),
`but not by tumour cells of the bone metastasis (see table).
`
`Fig. 3-Original biopsy of gastric adenocarcinoma.
`Poorly
`haematoxylin
`Immunoperoxidase counterstamed with
`only.
`differentiated tumour nests infiltrating the gastric muscularis demonstrate
`antibody.
`Original
`monoclonal
`(arrow)
`with
`straining
`19-9
`strong
`magnification x 640, reduced by one third.
`
`Fig. 4-Liver metastasis.
`Immunoperoxidase counterstained with haematoxylin only. Cytoplasmic
`localisation of monosialoganglioside detected by 19-9 monoclonal antibody in
`malignant cells. Original magnification x 640, reduced by one third.
`
`a
`
`Discussion
`Our aim was to identify potential hazards of further
`immunotherapy or immunodiagnostic efforts by means of a
`monoclonal antibody that specifically destroys human
`gastrointestinal tumours implanted in animals.4 We were
`particularly concerned with binding of the antibody to
`tumour and to normal tissues, sensitisation of the host to
`mouse immunoglobulin, and potential antigenic modulation
`secondary to exposure to antibody. Though the patients
`showed no evidence of serum sickness, the data suggest that
`whole mouse immunoglobulins will induce an anti-mouse-
`immunoglobulin response.
`In other attempts at immunotherapy against human
`tumours’2-14 antibody against a normal lymphocyte antigen
`was used in smaller amounts; it may therefore have been
`bound rapidly by antigen on circulating cells. 13 By contrast,
`antibody 17-lA does not react with antigens shed by the
`tumour cells.5 This may explain why functional antibody
`for
`could
`detected
`considerable
`after
`be
`time
`administration. Administration of 15 mg antibody 17-lA
`results in the transient appearance of mouse immunoglobulin
`in the patient’s circulation immediately after injection. When
`larger amounts (150 mg) of antibody 17-lA were injected, the
`intact mouse immunoglobulin was present in the circulation
`for longer periods of time and was also found transiently in
`the urine of one patient.
`The fraction of the circulating mouse immunoglobulin that
`binds in vitro to colorectal carcinoma target cells and
`represents the active 17-lA antibody persisted in the serum of
`patients2 and 3 for as long as the mouse immunoglobulin did.
`In patient 4 the specific binding decayed by day 10 after
`treatment, whereas mouse immunoglobulin persisted for a
`longer time. As with mouse immunoglobulin, binding
`activity to colorectal carcinoma cells was highest 2 to 4 days
`after treatment started.
`In three of our four patients an antibody response to the
`mouse immunoglobulin developed within 6 to 10 days. The
`lack of antibody response in patient 2 might be attributed to
`her debilitated condition and to radiation and chemotherapy
`before the antibody therapy. Development of antibody
`against mouse immunoglobulin in patient 4 led to a large fall
`circulating mouse
`immunoglobulin;
`change
`in
`this
`accompanied the patient’s adverse clinical reaction to the last
`injection of antibody 17-lA.
`Miller et al. 12,14 have described a T-cell leukaemic patient
`in whom antibody against mouse immunoglobulin was
`days
`of
`detected
`after
`transiently
`administration
`5
`monoclonal antibody, but who showed no clinical signs after
`a second dose of antibody 7 days after the first. The lack of
`clinical signs may be due to the smaller dosage ofmonoclonal
`antibody used (1-5mg) or to the reduced ability ofa patient
`with advanced leukaemia to mount an adequate immune
`response.
`Purified peripheral-blood mononuclear cells exposed to
`17-1A monoclonal antibody effectively destroy colorectal
`carcinoma cells.4 Destruction of colorectal carcinoma cells in
`athymic nude mice injected with antibody 17-lA is attributed
`to effector cells exposed to circulating antibody.15 As an
`adjunct to our immunotherapeutic trial we therefore isolated
`peripheral-blood lymphocytes of patient 4, exposed them to
`antibody 17-1A, and returned them to the patient. Liver
`metastases of patient 4 were affected by the treatment, as
`
`3 of 4
`
`BI Exhibit 1028
`
`
`
`shown by histology of the resected metastases and by
`ultrasound scanning of the two metastases remaining in situ.
`During the administration ofmononuclear cells and antibody
`17-lA, small aggregates formed which interfered with arterial
`blood supply to liver; lysis of tumour cells may therefore have
`been due to ischaemia. The transient elevation of AAT levels
`immediately after infusion of the mixture of peripheral-blood
`mononuclear cells and antibody may also have indicated
`hepatic dysfunction resulting from ischaemia. However,
`hepatic-artery flow observed 2 days later at surgery appeared
`normal, and the liver tissue appeared to be well vascularised.
`Furthermore, histology showed heavy infiltration of the
`necrotic area by mononuclear cells, implying that the
`peripheral-blood mononuclear cells played an active part in
`the destruction of tumour metastases.
`Although the bone metastasis apparently became smaller
`during the treatment, there was no histological evidence of
`tumour destruction. The lack of evidence of tumour
`destruction might be attributed to antigenic modulation of
`the metastatic cells, as indicated by the absence of 19-9
`antigen in this lesion.
`We have demonstrated that a mouse monoclonal antibody
`against a human tumour antigen can be safely administered
`directly to the affected organ and that the antibody persists in
`the circulation for long periods of time. An anti-mouse-
`immunoglobulin response develops; this may limit repeated
`administration
`of
`whole
`of
`molecules
`mouse
`immunoglobulins. The efficacy of this immunotherapeutic
`approach may be enhanced by exposing the patient’s own
`effector cells to mohoclonal antibody and administering these
`cells directly into the metastatic site.
`
`We thank J. Smith and K. O’Neill for technical assistance. The study was
`supported by grants CA-10815 and CA-21124 from the National Cancer
`Institute and grant RR-05540 from the Division of Research Resources.
`
`Correspondence should be addressed to H. K., The Wistar Institute, 36th
`Street at Spruce, Philadelphia, PA 19104, U.S.A.
`
`REFERENCES
`
`1. Herlyn M, Steplewski Z, Herlyn D, Koprowski H. Colorectal carcinoma-specific
`antigen: detection by means of monoclonal antibodies. Proc Natl Acad Sci USA
`1979; 76: 1438-42.
`2. Koprowski H, Steplewski Z, Mitchell K, Herlyn M, Herlyn D, Fuhrer P. Colorectal
`carcinoma antigens detected by hybridoma antibodies. Somat Cell Genet 1979; 5:
`957-72.
`3. Herlyn D, Herlyn M, Steplewski Z, Koprowski H. Monoclonal antibodies in cell-
`mediated cytotoxicity against human melanoma and colorectal carcinoma. Eur J
`Immunol 1979; 9: 657-59.
`4 Herlyn D, Steplewski Z, Herlyn M, Koprowski H. Inhibition of growth of colorectal
`carcinoma in nude mice by monoclonal antibody. Cancer Res 1980; 40: 719-21.
`5. Steplewski Z, Chang TH, Herlyn M, Koprowski H. Release of monoclonal antibody
`defined antigens by human colorectal carcinoma and melanoma cells. Cancer Res
`1981; 41: 2723-27.
`6. Sears HF, Herlyn D, Herlyn M, et al. Ex vivoperfusionof tumor-containing colon with
`monoclonal antibody J Surg Res 1981; 31: 145-50.
`7 Koprowski H, Steplewski Z, Herlyn D, Herlyn M. Study of antibodies against human
`melanoma produced by somatic cell hybrids. Proc Natl Acad Set USA 1978; 75:
`3405-09.
`8 Kolcher D, Hovanhand P, Teramoto YA, Wunderlich D, Schlom J. Use of monoclonal
`antibodies to definea diversity ofmammary tumorviral geneproducts invirionsand
`mammary tumors of the genus MUS. Cancer Res 1981; 414: 1451-59.
`9. Koprowski H, Herlyn M, Steplewski Z, Sears HF. Specific antigen in serum of patients
`with colon carcinoma. Science 1981, 212: 53-55.
`10 Magnani JL, Brockhaus M, Smith DF, et al. A monosialoganglioside is a monoclonal
`antibody defined antigen of colon carcinoma. Science 1981; 212: 55-56
`11 Brockhaus M,MagnaniJL, BlaszczykM, et al. Monoclonal antibodiesdirectedagainst
`the human Leb blood group antigen. J Biol Chem 1981; 256: 13223-25.
`12 Miller RA, Levy R Response of cutaneous T cell lymphoma to therapy with
`hybridoma monoclonal antibody. Lancet 1981; ii: 226-30.
`13 Ritz J, Pesando JM, Sallan SE, et al. Serotherapy of acute lymphoblastic leukemia with
`monoclonal antibody. Blood 1981; 58: 141-52.
`14. Miller RA, Maloney DG, McKillop Y, Levy R. In vivo effects of murine hybridoma
`monoclonal antibody in a patient with T-cell leukemia. Blood 1981; 58: 78-86.
`15. Herlyn D, Koprowski H Monoclonal antibodies against solid human tumors inhibit
`tumor growth in nude mice. Hybridoma 1982; 1: 206.
`
`765
`
`MORPHOLOGICAL IDENTIFICATION OF THE
`AGENT OF KOREAN HAEMORRHAGIC FEVER
`(HANTAAN VIRUS) AS A MEMBER OF THE
`BUNYAVIRIDAE
`
`J. B. McCORMICK
`D. R. SASSO
`E. L. PALMER
`M. P. KILEY
`Viral Disease Division, Center for Infectious Diseases,
`Centers for Disease Control, Atlanta, Georgia, U.S.A.
`
`Summary
`
`(KHF)
`haemorrhagic
`fever
`Korean
`(Hantaan virus), a rodent-borne viral illness,
`is an important cause of human disease throughout much of
`Asia and Eastern Europe. The agent responsible for KHF has
`not yet been conclusively identified. Plaque-purified KHF
`virus was concentrated and then banded in a potassium
`tartrate gradient. Material from the 1·
`17-1· 19 g/ml band
`was examined by electron microscopy and particles with a
`morphology identical to that ofthe family Bunyaviridae were
`found. The particles were aggregated by KHF serum but not
`by saline solution or non-immune serum. Identification of
`KHF virus as a member of the family Bunyaviridae
`implies a potential for spread by arthropod vectors.
`
`Introduction
`KOREAN haemorrhagic fever (KHF) was described several
`decades ago and is known by many names throughout Asia
`and Europe.’ It is a severe, not uncommon disease, found in a
`geographic area from Japan to Europe which is occupied by
`about one-half of the world’s population. The agent was
`isolated in 1978 by Lee et al. and was grown in tissue culture
`by French in 1981.3 The virus has not, however, been
`satisfactorily purified for morphological identification. We
`describe the purification and morphological characteristics of
`Hantaan virus.
`The strain of Hantaan virus used for this study, designated
`76-118, has been registered in the Working Catalogue of
`Arthropod-Borne Viruses. Since it is a direct descendent of
`the 76-118 isolate described by Lee et al. in the original
`isolation of the virus,2 and by French for growth in A-549
`cells,3 it has a well-defined pedigree. It is also 1 of the 3 strains
`studied by Lee and Cho in their effort to characterise the
`virus.4
`Several important characteristics of the 76-118 strain have
`been identified. (a) It was initially isolated from the rodent
`Apodemus agrarius corae.
`2 (b) Antibody titres to 76-118 strain
`rise 4-fold or more in persons with typical KHF illness.2,3
`(c) Antibody titres to strain 76-118 have been found in sera
`from patients with nephropathia epidemica in Scandinavia
`and epidemic haemorrhagic fever with renal syndrome in
`China and Japan.5-8 (d) Strain 76-118 is sensitive to lipid
`solvents and is acid labile. 3
`
`Methods
`
`Virus Cultivation
`A fifth A-549 cell passage of strain 76-118 virus was inoculated
`into 30 mm roller tubes of E-6 cells, a cloned line of Vero cells
`maintained at the Centers for Disease Control (CDC) and available
`from the American Type Culture Collection. These infected E-6
`cells were maintained in minimal essential medium (MEM) with
`1007o fetal calf serum (FCS) and virus was harvested after 12 days.
`Six further passages were made in 75 cm2 (surface) flasks by
`inoculation at a multiplicity of infection of approximately 1 tissue
`culture infective dose (TCID5o) per cell.
`
`4 of 4
`
`BI Exhibit 1028
`
`