PHIGEN IXPHIGEN IX
`
`
`
`Exhibit 1 03 5Exhibit 1 03 5
`
`

`
`Phase I clinical trial of drug—monoclonal
`antibody conjugates in patients with advanced
`colorectal carcinoma: A preliminary report
`
`G. Teh, BSc(Hon),
`Tjandra, FRCS, G. A. Pietersz, PhD,
`J.
`A. M. Cuthbertson, FRACS, J. R. Sullivan, FRACP, C. Penfold, FRACS, and
`I. F. C. McKenzie, PhD, FRACP, FRCPA, Parlwille and East Melbourne, Victoria, Australia
`
`Melphalan (MEL), an alhylating agent, has been modified to a derivative,
`N-acetylmelphalan (N-AcMEL), which can be conjugated to anticolon cancer
`monoclonal antibodies (l\loAbs 30.6, 1- 7, and ]GT) and used for
`immunochemotherapy. The final immunoconjugates possess potent cytotoxicity and
`specificity in preclinical studies. In a phase I clinical study, N-AcMEL-Mo/lb
`conjugates were administered via the hepatic artery to 70 patients, nine of whom had
`disseminated colorectal cancer (including the liver) and one of whom had Dukes’ C
`colon cancer that had been resected.‘The, selection of Mo_Ab was based on the
`immunoperoxidase staining of the‘ primary colon cancer tissue. Thus far doses of 7000
`mg/m’ Mo/lb conjugated to 20, mg/m’ of N-/'lcMEL have been administered with no
`significant side efiects, whereas MEL unconjugated to monoclonal antibodies would
`have caused myelosuppression in a proportion of patients at the same dosage. Serum
`antimouse antibody responses were noted in all of the patients; febrile reactions were
`noted with higher doses but were easily controlled with ‘antipyretics, antihistamines
`and, if necessary, steroids. Serum sickness developed in one patient who was given a
`second course of treatment in the presence of human antimouse antibody, but the
`episode was self-limiting. Eight of the 70 patients had evaluable disease. Subjective
`improvement was noted in almost all of the patients examined, and 33%, or 3 of 9, of
`the treatments (nine courses of treatment in eight patients with eaaluable disease; one
`of the patients had two courses of treatment) led to antitumor responses (minor
`response) by objective assessment with computed tomography of the liver. It is
`important to note that treatment with N-AcMEL-Mo/ib conjugates was safe at a dose
`of 20 mg/m’ of N—AcMEL, whereas the efiicacy of such a form of treatment remains to
`be determined.
`(SURGERY 7989;706:533,-45.)‘
`
`From the Research Centre for Cancer and Transplantation, Department of Pathology, The
`University of Melbourne, and the Colorectal Unit, Royal Melbourne Hospital, Parkville, and
`the Department of ll/Iedical Oncology, St. Andrew’s Hospital, Clarendon Place, East
`Melbourne, Victoria, Australia
`
`CANCER 05 THE COLON and rectum is one of the most
`common forms of malignancy in Western countries,
`with approximately 120,000 new cases reported annu-
`ally in the United States.‘ ‘Hepatic metastases are
`present on initial diagnosis of colorectal cancer in 25%
`to 30% of patients.‘ After curative resection of colorcc-
`
`Acccptcd for publication Nov. 22, 1988.
`Reprint
`requests:
`Ian F. c. McKenzie, MD, PhD,
`Research Centre for Cancer and Transplantation, Depart-
`mcntiof Pathology, The University of Melbourne, Park-
`villc, Victoria 3052, Australia.
`
`tal primary tumors, the liver is again the most frequent
`site of relapse in 40% to 50‘7a."“ Once hepatic metasta-
`scs have developed,
`the prognosis is poor, with an
`expected median survival of 6 to 9 months,"-" the extent
`of
`the tumor being the most
`important prognostic
`factor.‘ Many different forms of trcatment,,i'nclu-ding
`systemic chemotherapy, have been used for colorcctal
`hepatic metastases, without much success.‘ The only
`patients who may achieve 5-year survival are the
`highly select-group suitable for surgical resection-‘-
`usually those patients with less than four hepatic
`metastatic lesions.“ Itshould be recognized, however,
`
`SURGERY S33
`
`PHIGENIX
`1035-01
`
`

`
`534 Yjcmdra at al.
`
`Surgery
`September 7989
`
`Table I. Characteristics and clinical features of patients treated with N-ACMEL-MoAb conjugates
`
`Dose‘.‘(p.g/m’) of
`Age
`Perfforrnance status
`Previous therapy N- AcMEL.'MaAb
`
`Patient
`(yr)
`Sex
`(ECOG)
`1
`59
`M
`2
`HAI of cis-platinum
`5mg/m‘:l20mg/m‘
`10mg/m‘:160mg/tn’
`10mg/m2:98Omg/m’
`
`2
`
`i
`
`61
`
`M
`
`2
`
`Partial hepatectomy"
`Adjuvant chemotherapy
`
`2
`M
`57
`3
`10mg/m’:250mg/m’
`2
`M
`58
`4
`15mg/m’:340mg/In’
`3
`M
`62
`S
`15mg/m’:380mg/m’
`2
`M
`57
`6
`l 5mg/rn’:500rng/m’
`3
`F
`46
`7
`20mg/m’:440mg/m’
`3
`M
`62
`8
`20mg/m’:1000mg/ml
`1
`F
`38
`9
`20mg/m‘:820mg/m‘
`3
`M
`64
`10
`
`20mg/m’:l0O0mg/m‘
`
`HA1 of ct‘:-platinum
`
`Legend: ECOG, Eastern Cooperative Oncology Group; HA1, hepatic artery infusion.
`The amount (mg) of N-ACMEL conjugated to MnAb and administered was expressed as mg/m‘ of body surface area of the patient.
`
`that patients suitable for resection make up a very
`small percentage of all patients with colorectal hepatic
`metastases. More recently there have been encouraging
`reports of response to regional perfusion with chemo-
`therapy, especially 5-fluoro-2-deoxyuridine (FUDR);
`however, this is still limited by complications related to
`chemotherapy.‘’~ “’
`‘It is with this background of unsuccessful therapeu-
`tic maneuvers that alternative therapeutic avenues with
`monoclonal antibodies (MoAbs) are explored. By
`means of the hybridoma technique,“ murine mono-
`clonal antibodies have been produced against almost all
`of the major types of human cancer.“ However, no
`truly tumor-specific ‘MoAb has been derived thus far,
`but in most "cases the antigens recognized are present on
`tumors in‘ greater concentrations than on normal tis-
`sues.” There are several reports of clinical response to
`antitumor monoclonal antibodies used alone, mostly in
`malignant melanoma, neuroblastoma, leukemia, and
`lyrnphoma.““‘ However, the therapeutic effects are not
`dramatic, presumably because murine antibodies do
`not adequately incite appropriate host effector mecha-
`nisms to destroy tumors. It is therefore believed that the
`greatest therapeutic potential for MoAbs; lies in the
`targeting of
`anticancer
`agents
`(chemotherapeutic
`drugs,
`toxins, or radioactive" substances)
`to tumors
`rather than their use in unmodified form. By using a
`“prodrug” approach, a potent immunoconjugate was
`produced by covalently conjugating an inactive N—
`acetyl derivative of melphalan (N-AcMEL) (“pro-
`drug”) to murine MoAbs.”'The procedure removed
`the ability of the mclphalan to enter cells by its usual
`active transport via the amino acid transport systems;
`however, the MoAb provided the alternative route of
`cell entry via endocytosis, and such N-AcME‘L-MoAb
`
`conjugates, on binding to tumor antigen on the tumor
`cell surface, exert their effects after internalization and
`lysosomal degradation within the target tumor cell to
`release melphalan.” The imrnunoconjugates have dis-
`played in vitro and in vivo specificity and cytotoxicity
`and specifically inhibit
`the growth of human colon
`carcinomas xenografted in athymic mice when injected
`intravenously."
`We have described a murine MoAb 30.6that reacted
`
`with > 90% of colon cancer tissue” and could preferen-
`tially localize human colorectal
`tumor xenograft in
`nude mice” and in primary and secondary colon
`carcinoma in patients.’“" Two additional anticarci-
`nogenic embryonic antigen MoAbs (I-1 and jGT) had
`been developed, and they reacted strongly with 80% of
`colon carcinoma on immunoperoxidase staining.”
`Immunoconjugates of N-ACMEL to these MoAbs
`(30.6, I-1, JGT) have been developed by means of the
`same principles.” We now report a phase I clinical
`study with N-AcMEL—MoAb‘conjugates administered
`via hepatic artery infusion in 10 patients: nine with
`disseminated (including liver) colorectal cancer and one
`with resection of Dukes’ C colon carcinoma.
`
`MATERIAL AND METHODS
`
`Patients. Ten patients with advanced colorectal
`carcinoma were included in this study. They were
`estimatcdfto have at least a 3-month expected survival,
`a performance status’ (Eastern Cooperative Oncology
`Group) less than or equal to 3, and had no other
`cytotoxic therapy for at least 1 month before adminis-
`tration of immunoconjugate and during the 3-month
`evaluation phase of the study. Table I summarizes the
`characteristics and clinical features of the patients’.
`Ages ranged from 38 years to 64 years. Nine of l0
`
`PHIGENIX
`1035-02
`
`

`
`Volume 706
`Number 3
`
`Drug-ll/IoAb conjugates and colorectzzl carcinoma
`
`535
`
`patients had extensive hepatic metastases from colorec-
`tal carcinoma, and the remaining patient had a locally
`advanced colon cancer
`(Dukes’ C)
`that had been
`resected and did not have demonstrable hepatic metas-
`tases by laparotomy or computed tomography (CT).
`Two of nine patients with hepatic metastases also had
`pulmonary metastases, and one of nine patients had a
`primary colon carcinoma that had not been resected
`because of the poor general medical condition of the
`patient. Two of the nine patients previously had failed
`intensive chemotherapy (hepatic artery infusion of
`cis-platinum), and one of these two patients had two
`courses of immunoconjugates separated by a 2-month
`interval. One of the nine patients had recurrent hepatic
`metastases after a previous partial hepatectomy and
`adjuvant 5-fiuorouracil. All patients (except patient 9)
`had progressive metastatic disease at
`the time they
`entered the study, and the hepatic metastases were too
`extensive for hepatic resection. All patients were fol-
`lowed for at
`least 3 months after therapy (except
`patient 7 who died 4 weeks after therapy with a
`generalized debility and patient 10—see below); they
`were evaluated at weekly intervals for 6 weeks, then
`monthly. This phase I study was approved by the
`Medical Research Board of the Royal Melbourne
`Hospital, and written informed consent was obtained
`from all patients.
`Monoclonal antibodies. Murine MoAbs used were
`
`an anti-
`IgG2b antibody 30.6, directed against
`gen present on human colon secretory epithelium but
`also reactive against a number of colon carcinoma cell
`lines,” and IgG1 antibodies I-1 and JGT (both
`anticarcinoembryonic antigen), which were produced
`by means of a novel
`immunization technique with
`whole serum of patients with advanced colorectal
`cancer;
`they»reacted with human colon carcinoma,
`malignant tumors of noncolonic origin (breast,
`thy-
`roid), and a number of colon carcinoma cell lines but
`not with normal tissue or benign lesions (24, unpub-
`lished observations). The antibodies were purified on
`protein A-superose (Pharmacia,
`Inc., Piscataway,
`N ._I.). After elution, MoAbs were concentrated by 45%
`ammonium sulfate precipitation, dialyzed against
`phosphate-buffered saline (PBS), aliquoted, and stored
`at -70° C. The concentration of IgG was estimated by
`absorbance at 280 nm. The prepared antibodies were
`retested for activity after all procedures (see below),
`filtered through a 0.22 urn Millex-GV filter (Milli-
`pore, Befored, Ann Arbor, Mich.), and batch tested for
`purity by sodium dodecyl sulphate polyacrylamide gel
`electrophoresis (SDS-PAGE).
`Preparation of N-AcMEL-IgG conjugates. The
`MoAbs used included 306 (IgG2b), I-1, and JGT
`
`(IgG1). The N-acetyl derivative of melphalan was
`prepared and conjugated to whole IgG.” Briefly, MEL
`was acetylated with acetic anhydride and an active
`ester of this N-ACMEL derivative was then coupled to
`the amino groups of the MoAb. Any precipitated
`protein was removed by centrifugation, and free N-
`ACMEL was removed by gel filtration chromatography
`with a Sephadex G-25 column (PD-10; Pharmacia).
`N-ACMEL incorporated in the drug-MoAb. conjugates
`was determined by absorbance spectrophotometry at
`258 nm (E15, = l X 10‘M"cm“) after subtracting the
`protein contribution following its estimation by the
`Bradford dye-binding assay.” The alkylating activity
`of the conjugate was determined by a modification of
`the method of Epstein et al."’ The final preparation
`after drug conjugation was batch tested for pyrogens
`and sterility (Department of Pharmacology, University
`of (Melbourne, and Sigma Pharmaceuticals, Clayton,
`Victoria, Australia).
`The antibody activity of N-ACMEL-IgG conjugates
`was demonstrated in a rosetting assay” and in immu-
`noperoxidase
`staining with
`formalin-fixed
`(N-
`ACMEL-I-1, N-ACMEL-JGT) and snap-frozen (N-
`ACMEL-30.6) sections of human colon cancer tissue
`(data ‘not shown).
`Administration of drug-MoAb conjugates. By
`means of the Seldinger technique,
`the catheter was
`introduced percutaneously into the left axillary or high
`brachial artery. The catheter was placed in the com-
`mon hepatic artery, and when multiple hepatic arteries
`were found supplying the liver, the catheter was placed
`in the largest vessel. The irnmunoconjugate was admin-
`istered via hepatic artery infusion with an oxymetric
`pump in 100 ml of normal saline solution for 2 hours
`per day for 2 days. All patients had three doses of the
`immunoconjugates (t = 0, t = 24 hours, t = 48 hours).
`Between infusions of the immunoconjugates, the paten-
`cy of the catheter was accomplished with heparinized
`saline solution (5000 IU aqueous heparin in 1 L
`normal saline solution at the rate of 50 ml/hr) with the
`oxymetric pump. At
`the, end of the 2-day infusion
`period, the indwelling catheter was removed. Patients
`were given dexamethasone, 8 mg intravenously, just
`before each infusion of irnmunoconjugates and oral
`prednisolone, 10 mg daily for Tdays after completion
`of infusion as prophylaxis for allergic reactions. The
`dose escalation protocols used (Table I) were as
`follows: one patient received 5 mg/m’ and 2 months
`later, 10 mg/m2; two received 10 mg/mz; three received
`15 mg/m’; and four received 20 mg/m‘ N-ACMEL
`conjugated to MoAbs. The study was closed at the 20
`mg/m2 dose of N-ACMEL conjugates because of the
`cost incurred in producing such a large quantity of
`
`PHlGENlX
`1035-03
`
`

`
`S36 Tjandra at al.
`
`Surgery.
`SC]_)l(£m/)8)’ 79897:
`
`Table II. Binding of MoAb (30.6, I-1, JGT) as detected by immunoperoxidase staining on primary
`colon cancer
`
`it
`
`Fixed
`
`Staining grade*
`MoAb: used in
`
`I—7_ JCT
`30.67‘
`Colon cancer tissue
`Patient
`z"mm zmoconjugater
`1-1
`4
`3)
`1-1, jGT
`30.6, 1-1, JGT
`1-1
`
`Fixed
`Fixed
`
`Fresh/fixed
`Fixed
`Fixed
`Fixed
`
`Fresh/fixed
`Fresh/fixed
`Fixed
`
`10
`
`3
`
`L.-J
`
`Z
`
`'
`
`\’3®\3O\iJ‘«Fla-*1.‘-3
`->-it-kb3~F-(.H4¥E\3-F
`L.a-;s.5>.u:u:IaJ-£=--:>-5>-
`
`30.6, 1-1. JGT
`1-1
`
`1-1, JCT
`30.6
`
`30.6, I-1, JGT
`1-1, JGT
`I-1, jGT
`I = up to 25%; 2 = 26% to 50%; 3 = Sl% to 75%; 4 = 76% to
`
`‘Staining score was graded based on the proportion of carcinoma cells stained: 0 =- no staining;
`l00%.
`1-30.6 MoAb tested on fresh colon cancer tissue only.
`¢Patient I had two courses of treatment.
`
`antibodies and the concern that the maximum tolerated
`
`dose of such a form of treatment may not be practicably
`achieved (see below).
`Patients were monitored clinically for changes in
`temperature, pulse, blood pressure, and respiratory
`function during and after the infusion. Blood studies
`were also done before, during, and weekly for 6 weeks
`after the therapy to assess potential hematologic (full
`blood examination), renal (urea and electrolytes), or
`hepatic toxicity (liver
`function test) and to detect
`human immune responses
`stimulated by rnurine
`immunoglobulin (human antirnouse antibody).
`Human antimousc antibody response. Human
`antibodies against the murinc MoAbs were measured
`by an enzyme-linked immunosorbent assay (ELISA)
`modified from that previously described." Ninety-six
`well flexible polyvinyl chloride plates,(Costar, Cam-
`bridge, Mass.) were coated with 50 ul/we'll of admin-
`istered MoAb (5 ug/ml of purified 30.6, 1-1, or JGT
`MoAbs) in a 0.1M carbonate buffer,
`9.6, and
`nonspecific binding blocked with 1% bovine serum
`albumin/PBS, pH 7.6. Serial dilutions of patient sera
`and pooled normal human serum (50'p,l/well) in
`PBS/0.05% Tween 20 to a final dilution of 1/256 were
`performed and added to the coated wells (50 ul/well).
`Plates were then washed with PBS/0.05% Tween 20
`and then reacted with 50 p.l/well of phosphatase
`labeled affinity purified goat antihuman IgM and IgG
`(Kirkegaard ’and Parry, Gaithersburg, Md.). The color
`reaction was developed with alkaline phosphatase
`substrate and read with an ELISA plate reader (Ti-
`tretek, Multiscan, MC) at a wavelength of 405 nm.
`
`Results were expressed as the absorbance value of
`patient serum compared with pooled normal human
`serum, and a positive test result was considered to be a
`value at least twice the control.
`
`Immunoperoxidase
`Immunoperoxidase staining.
`staining was performed” on 6 mn tissue sections of
`colon cancer tissue from all patients with I-1 and _]GT
`MoAbs; if possible staining was also performed with
`30.6 MoAb. The 30.6 MoAb only reacts with snap-
`frozen but not
`for-malin—fixed colon cancer
`tissue,
`whereas I-1 and JGT MoAbs react with both snap-
`frozen and formalin- fixed sections. A nonreactive con-
`
`trol antibody was used in all cases. The sections were
`then assessed by light microscopy to estimate the
`~ percentage of colon carcinoma cells stained with each of
`the antibodies; results were expressed on a scale of 0 to
`4 according to whether nil (0), up to 25% (1), 26% to
`50% (2), 51% to 75% (3), or >75% (4) of carcinoma
`cells stained. This is a serniquantitative assay and is
`highly reproducible.” The intensity of stain, the distri-
`bution of stain in the cancer cells, and the staining of
`extracellular material were not taken into account. The
`
`MoAbs selected for use in drug conjugation for an
`individual patient had to have a staining score of 3
`or 4.
`_
`‘
`
`Evaluation of tumor responses. Patients were
`evaluated clinicallyand biochemically (liver function
`test, carcinoembryonic antigen [GEM level), and the
`measurable lesions were measured at 1 and 2 months
`
`after therapy by CT scans of the abdomen performed
`with the same technique by the same radiographers
`and radiologists as that used for the pretherapy evalu-
`
`PHIGENIX
`1035-04
`
`

`
`Volume 106
`Number 3
`
`Drug-MoAb conjugates and colorectal carcinoma
`
`537
`
`ation (performed within 2 weeks before therapy).
`Complete response is defined as the disappearance of _
`all evidence of tumor. Partial response is defined as a.
`reduction of at least 50% in the sum of the products of
`the two greatest diameters of measured lesions.’ Minor
`response is a reduction of more than 25% but less than
`50% in the size of measurable tumors in the absence of
`
`progression or occurrence of new lesions elsewhere.
`Stable disease is an objective regression of measurable
`lesions less than that required to meet the criteria for
`minor or partial response or an increase of less than
`25% in the size of one or more measurable lesions for at
`
`least 4 weeks. Progressive disease is the appearance of
`new lesions or increase in size of one or more measur-
`
`able lesions by at least 25%.
`
`RESULTS
`
`Immunohistochemical testing on primary colon
`cancer. A selection of MoAbs for conjugation with
`N-ACMEL was made for each patient, based on the
`binding of the particular MoAb (30.6, 1-1, jGT) to
`sections of the primary colon cancer by'irnmunoperox-
`idase staining (Table II).
`In general, MoAb was
`selected only if it had a staining score of 3 or 4. When
`multiple antibodies were used for drug conjugation, the
`final preparation of the immunoconjugates had equal
`proportions of the MoAbs. A combination of at least
`two MoAb conjugates (N -ACMEL-30.6, N7ACMEL'
`I-1, N-AcMEL~jGT) was used in 7 of 11 treatments
`(Table II).
`It was considered that
`the use of a
`combination of MoAb conjugates would ensure
`maximal immunoreactivity and help to overcome the
`potential problem of tumor heterogeneity within and
`between tumor masses. We were unable to obtain
`tissue from the liver metastases itself for immunohisto-
`
`chemical testing before treatment.
`Toxicity. Tables III and IV summarize the effects
`of hepatic artery infusion of N-ACMEL-MoAb conju-
`gates. In general, the therapy was well tolerated with
`no disturbance in gastrointestinal, renal, or cardiac
`parameters, and there was no evidence of myelosup—
`pression. There was neutrophilia during the time of
`treatment, which restabilized after completion of treat-
`ment. The extcrnal arterial catheter was well tolerated
`with no complications, and all patients maintained
`good mobility for the duration of therapy. Patient 1 had
`two courses of therapy separated by a 2-month interval,
`despite the presence of a high titer of human antimouse
`antibody (see below). During the second course of
`therapy he had pain in the lower back, fever (39° C),
`urticaria, and bronchospasm. These reactions occurred
`about
`1 hour after immunoconjugate infusion was
`
`Table III. Toxicities
`
` Parameters No. of patient:
`
`Pain
`-
`l
`Febrile >38” C
`5
`
`Allergic phenomena
`(urticaria, bronchospasm)
`Hematologic
`White cell count
`
`<4000/mm’
`Platelets <lO0,(]00/mm’
`Gastrointestinal
`
`Nausea/vomiting/
`dyspepsia
`Diarrhea
`Bilirubin elevation >50%
`AST/ALT elevation
`>50%
`
`Alkaline phosphatase
`elevation >25%
`Renal
`Urea elevation >25%
`Creatinine elevation
`>25%
`
`Proteinuria/hematuria
`Cardiac
`
`1
`
`0
`
`0
`
`0
`
`1
`0
`2
`
`0
`
`0
`0
`
`0
`
`0
`O
`0
`
`Rhythm changes
`Rate <60 or >110/min
`Diastolic blood pressure
`elevation >30%
`Catheter-related,
`(thromboembolism,
`hemorrhage,
`displacement, intimal
`tears)
`Legend: AST, Aspartatc transaminaxe; ALT, alanine transaminase.
`
`0
`
`begun on the second day of the second course of
`therapy, despite prior administration of dexametha-
`sone; treatment was required with antihistamine and
`an additional dose of dexamethasone. The bronchu-
`
`spasm, urticaria, and pain rapidly resolved with such
`additional measures, but the fever persisted for another‘
`4 hours after the completion of infusion of immunocon~
`jugates. There was, however, no reaction when further
`infusion of immunoconjugates was given.
`In four
`patients (patients 7, 8, 9, and 10) a temperature of 38
`to 38.5° C was noted during the second and third day
`of antibody infusion, starting about 1 hour after the
`infusion was begun and continuing for 1 hour after the
`infusion had ended. It therefore appears that febrile
`reactions were more common in patients receiving
`higher doses of N-AcMEL-MoAb conjugates‘.
`
`PHlGENlX
`1035-05
`
`

`
`
`
`
`
`S38 Tjandra et al.
`
`Surgery
`September 7989'
`
`
`Table IV. Results of hepatic artery infusion of N~AcMEL conjugates
`Serum CEA* level
`
`its/17,3;,;;g;e'~e to
`
`Response
`
`Re-
`
`S“;:::;:i{::"‘
`
`of liver
`Status
`sporue
`by CT
`......_.———.—_...._—:...__.
`metastases
`(time from
`duration
`scan of
`After
`Alteration
`Known disease
`
`
`(4 wk)Beforesite:Patient (ma) (%) HAMAf liver (mo) treatment)
`
`
`
`
`
`
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`Hepatic
`metastases
`Hepatic
`metastases
`Hepatic and
`pulmonary
`metastases
`
`Hepatic and
`pulmonary
`metastases
`
`Hepatic
`metastases
`
`Hepatic
`metastases
`Hepatic
`metastases
`
`Hepatic
`metastases
`and
`unresected
`
`primary
`colon
`carcinoma
`Resectcd
`Dukes’ C
`
`3920
`1,160
`200
`
`26
`
`270
`
`920
`1,050
`221
`
`15
`
`50
`
`5,425
`
`4,300
`
`32
`
`62
`
`325
`
`12
`
`144
`
`42
`
`77
`9
`10
`
`42
`
`81
`
`21
`
`38
`
`132
`
`87
`
`5.0:1.0
`6.2110
`32:10
`
`MR
`SD
`SD
`
`3.0:1.0
`
`MR
`
`2.4:1.0
`
`SD
`
`5.0.1.0
`
`PD
`
`4-.6:1.0
`
`MR
`
`4.1:1.0
`
`PD
`
`4-.6:1.0
`
`——
`
`3
`2
`3
`
`12
`
`6
`
`———
`
`6
`
`—
`
`-—
`
`Deceased
`(12 mo)
`Alive with disease
`(9 mo)
`Alive with disease
`('12 mo)
`
`'
`
`Alive with disease
`(10 mo)
`
`Deceased (6 mo)
`
`Alive with disease
`(6 mo)
`Deceased (5 mo)
`
`Deceased (1 mo)
`
`<1
`
`<1
`
`0
`
`2.0:1.0
`
`——
`
`-—-
`
`Alive with disease
`(10 mo)
`
`24
`
`17
`
`9
`
`22
`
`12.
`
`8
`
`6
`
`3
`
`ll
`
`4
`Alive with disease
`1
`SD
`4.0210
`8
`1,850 _ 2,000
`Hepatic
`10
`
`.metastases (6 wk)
`Legend: MR, Minor response; SD, stable disease; PD, progressive disease.
`‘Cancinuembryonic antigen (normal <5 ng/ml).
`’
`fl-luman antimousc antibody (IgG) response at 4 weeks after treatment expressed as the ratio of ahsorhance value of patient serum compared with pooled nonrtal human
`serum; the higher the ratio, the higher the HAMA response. A positive test result was considered to be a value z2.0:l.0.
`
`Patient 1 also had mild diffuse arthralgia and fever
`that developed 11 days after completion of the second
`course of therapy; both slowly resolved over 2 months.
`Patient
`8, who had unresected primary colon
`carcinoma and multiple hepatic metastases, died 4
`weeks after receiving 20 mg/m’ N-ACMEL conjugated
`to 1000 mg/ml-MoAb of general debilitating disease.
`There was worsening of diarrhea that started 2 weeks
`aftcr the treatment was given; however, the patient had
`unresected advanced colon cancer causing subacute
`bowel obstruction, which could have accounted for the
`diarrhea. Batients 7 and 10 had transient increases in
`the aspartate transaminase level by 70% (from 71
`IU/L to 120 IU/L) and 100% (40 IU/L to 81 IU/L),
`respectively, during treatment, which rapidly returned
`
`to pretreatment levels within 3 days of completion of
`treatment.
`All patients had human antimouse antibody with a
`raised level (absorbance value twice normal) as from
`the twelfth day after the first antibody exposure and
`was ofi IgM as well as IgG response. The peak
`response usually occurred within 30 days after therapy.
`The geometric means of the human antimouse anti-
`body titer were not higher in patients receiving higher
`doses of MoAbs.
`‘I
`»
`Antitumor effects. Table IV summarizes the tumor
`responses evaluated by CT scan in patients treated
`with N-ACMEL-MoAb conjugates. Minor antitumor
`responses were seen in three patients (patients 1, 3, and
`6).
`
`PHIGENIX
`1035-06
`
`

`
`Volume 706
`Number 3
`
`Drug-Mo/ib conjugates and colorectal carcinoma
`
`539
`
`
`
`Fig. 1. CT scans of patient l with multiple hepatic metastases before treatment (A) and 1 month after treatment
`(3)
`
`Patient 1 had subtotal colectomy 24 months previ-
`ously for synchronous carcinoma of the transverse and
`sigmoid colon. At laparotomy multiple large metasta-
`ses in the liver were noted. The hepatic metastases
`were too extensive for surgical
`resection and were
`treated with two courses of hepatic artery infusion of
`ci:-platinum with no response. The clinical condition
`continued to deteriorate over the following 10 months,
`with anorexia, nausea, lethargy, and severe hepatic"
`pain. Results of
`liver
`function tests were grossly
`‘deranged with elevated transaminase (AST = 136 IU/
`L) and C-EA (3920 ng/ml) levels. The abdominal CT
`scan showed multiple large metastases in both lobes
`occupying about 60% of the liver. Within 3 weeks after
`treatment with N-ACMEL-MOAB conjugates,
`there
`was a dramatic improvement in his constitution; he
`regained his appetite and had good relief of the hepatic
`pain; he felt so well that he returned to work as a
`headmaster. There was an improvement in liver func-
`tion (AST == 60 IU/L), a decrease in the CEA level
`(CEA = 920 ng/ml), and an improvement
`in the
`abdominal CT scan with a reduction in the size of the
`
`hepatic metastases (25%) (Fig. 1). A repeat abdominal
`CT scan performed 5 months after treatment showed
`that the hepatic metastases had not progressed since
`treatment. The patient subsequently died 12 months
`after treatment of progression of disease.
`Patient 2 had a previous left-sided hemicolectomy
`for carcinoma of the colon, partial hepatectomy for
`hepatic metastases, and adjuvant 5—HuorouraciI some
`18 months earlier. He subsequently had recurrent
`hepatic pain from recurrent hepatic metastases, which
`did not respond to radiotherapy, and the CEA level
`was rapidly rising when he entered the study. There
`was a subjective improvement in his general well-being
`
`with improved appetite, the CEA level remained stable
`after treatment, and the CT scan performed up to 3
`months after treatment showed that the lesions in the
`liver had remained unchanged in size.
`Patient 3 had hepatic and pulmonary metastases
`with a pretherapy CEA level of 26 ng/ml; he was
`lethargic and was losing weight. Within 2 weeks of
`treatment he became more energetic,
`regained his
`appetite and weight, and the CEA level fell
`to 15
`ng/ml. The CT scan of the liver performed 1 month
`after treatment showed increased calcification in parts
`of the hepatic metastases as evaluated by CT scan,
`although the CT scan sections were not exactly compa-
`rable. Another CT scan performed 2 months after
`therapy showed a further increase in areas of calcifica-
`tion in the hepatic metastases. By that stage, although
`the reduction of the area of the lesions was still <SO%,
`calcification had replaced most (50%) of the remaining
`hepatic metastases (Fig. 2). The pulmonary metastases
`remained unchanged in size for 7 months.
`Patient 4 is alive with disease (hepatic and pulmo-
`nary metastases) that has remained stable since treat-
`ment (6 months), and the CEA level has been falling
`gradually since treatment from a pretreatment value of
`270 mg/ml
`to 50 mg/ml 4 weeks after treatment.
`Patient 5 had rapidly progressive disease when entered
`into the study andwas previously treated with cis-
`platinum via hepatic artery infusion with no response. i
`There was a transient (6 weeks) improvement in the
`constitutional symptoms. Abdominal CT scan repeated
`1 month after the treatment showed some insignificant
`regression of the extensive hepatic metastases. Both the
`liver function test results and CEA level remained
`
`unchanged for 3 months. Six weeks after treatment
`patient 5 had clinical evidence of progression of disease
`Q
`
`PHIGENIX
`1035-07
`
`

`
`540 Tjandra at al.
`
`Surgery
`September 7989
`
`
`
`Fig. 2. CT scans of patient 3 with hepatic metastases before treatment (A) and 1 month (B) and 2 months (C)
`after treatment. Presence’ of increased calcification is indicated by arrow.
`
`with increasing hepatomegaly and development of
`ascites.
`Patient 6 had an anterior resection of the rectum
`because of rectal carcinoma and was found to have
`
`multiple bilobar hepatic metastases at laparotomy. An
`uneventful recovery from surgery was made, although
`there were complaints of frequent discomfort over the
`right upper quadrant of the abdomen. Within 4 weeks
`of treatment with the imrnunoconjugates, the abdomi-
`nal pain disappeared and he returned to work as an
`engineer. The CT scan performed 1 month after
`treatment showed complete disappearance of two of the
`metastatic lesions (Fig. 3), although the overall reduc-
`tion‘ in tumor size only qualified as a minor response
`(>25% but <50%). However, one of the lesions (Fig. 3,
`A) was small and could have been missed in between
`slices of the CT scan study._There was a slight decrease
`in the CEA level from a pretreatment level of 32 ng/ml
`to 12 ng/ml 4 weeks after treatment.
`Patient 7 had rapidly progressive disease (hepatic
`metastases) when entered into the study. There was a
`transient improvement in the general well—being with a
`notable reduction of nausea. An abdominal CT scan
`performed 4 weeks after treatment showed no signifi-
`cant change, and soon there was obvious clinical
`evidence of disease progression with worsening hepatic
`pain,
`increasing hepatomegaly, and a rising GEA
`level.
`
`Patient 8 had widespread bilobar hepatic metastases
`from a locally advanced colonic carcinoma that was not
`resected because of the patients poor general medical
`condition. After treatment with immunoconjugates via
`hepatic artery infusion, there was a transient improve-
`ment
`in the general well-being and there was a
`significant fall in the CEA level (325 ng/ml to 42
`ng/ml), but he rapidly deteriorated with fluid and
`electrolyte imbalance as a result of worsened diarrhea,
`which has already been discussed, and also from
`progression of generalized disease. He died 4 weeks
`after treatment before a repeat CT scan could be
`performed to evaluate the hepatic metastases. The
`response of the hepatic metastases to treatment was
`therefore not evaluable in this patient.
`Patient 9 had a locally advanced colon cancer
`(Dukes’ C) that was completely resected. She had no
`evidence of hepatic metastases on preoperative CT scan
`of the liver or at
`laparotomy. The CEA level had
`remained normal (<1. mg/ml). She was entered into
`the study to have the treatment in the form of adjuvant.
`She remained 'well and free of clinical recurrent disease
`until 7 months after treatment when three metastatic
`deposits in the liver were noted on CT scan. She
`underwent right hepatic lobectomy and the metastatic
`lesions were confirmed histologically. Immunoperoxi-
`dase‘. staining of the metastatic lesions showed a poorer
`reactivity with the administered antibodies when com-
`
`PHIGENIX
`1035-O8
`
`

`
`Volume 706
`Number 3
`
`Drug-A/IoAb conjugate: and coloreclal carcinoma
`
`54!
`
`Fig. 3. Two representative CT scan sections of patient 6 with multiple hepatic metastases before treatment (A)
`and 1 month after treatment (B). Complete disappearance of two small metastatic deposits is indicated by
`HTTOLUS.
`
`pared with the primary colon carcinoma tissues (Fig.
`4). This illustrates the problem of tumor heterogeneity
`between primary tumor and its metastases.
`Patient 10 had metachronous colonic carcinomas 15
`
`years apart, and both were successfully resected. At the
`time the second colonic carcinoma was first seen,
`
`multiple large metastatic deposits in the liver were
`noted, and they increased in size rapidly over a 6-week
`period before treatment with monoclonal