superantigen-targeted therapy: phase i escalating repeat ...antibody-directed, superantigen-induced...

Vol. 4, 1903-1914, August 1998 Clinical Cancer Research 1903

Superantigen-targeted Therapy: Phase I Escalating Repeat Dose

Trial of the Fusion Protein PNU-214565 in Patients with

Advanced Gastrointestinal Malignancies1

R. Katherine Alpaugh,2 Josephine Schultz,

Cecelia McAleer, Bruce J. Giantonio,

Robert Persson, Marie Burnite,

Sven-Erik Nielsen, Lauren Vitek,

Bo Persson, and Louis M. Weiner

Fox Chase Cancer Center, Philadelphia, Pennsylvania 191 1 1[R. K. A., J. S., C. M., M. B., L. M. W.]; Thomas Jefferson UniversityHospital, Philadelphia, Pennsylvania 19107 [B. J. G.]: Pharmacia and

Upjohn, Lund, Sweden 5-22007 [R. P., B. P.1: Pharmacia andUpjohn, Kalamazoo, Michigan 49001 [L. V.]; and Rigshospitalet,

Copenhagen, Denmark DK-2l00 [S-E. N.]

ABSTRACTAntibody-directed, superantigen-induced cytotoxicity

has been shown to have potent in vitro and in vivo antitumor

effects in preclinical models. In the present study, PNU-

214565, a recombinant fusion protein consisting of the Fab

of the monoclonal antibody C242 and staphylococcal enter-

otoxin A (SEA), was used in an escalating repeat dose Phase

I clinical trial in patients with advanced gastrointestinal

malignancies. A prior single-dose Phase I clinical trial had

demonstrated safety at doses of 1.5 nglkg with toxicities of

fever and hypotension that were not dose related. Twenty-

seven patients (age range, 36-75 years; median, 62; 14 males

and 13 females; 23 colorectal and 4 pancreatic) were treated

in the present study with one cycle of four consecutive daily

3-h infusions of PNU-214565 at doses of 0.15 ng/kg (n 3);

0.5 ng/kg (n = 3), 1.5 nglkg (n = 4), 2.75 nglkg (n = 12), and

3.5 nglkg (n = 5). All patients had a good performance

status [Eastern Cooperative Oncology Group: PS 0 (n

15), PS = 1 (n 12)1. As in the single-dose trial, fever and

hypotension were the most common toxicities. Dose-limiting

toxicity (DLT), consisting of transient hypotension respon-

sive to dopamine, was experienced by one patient treated at

the 2.75 nglkg dose leveL One patient with pancreatic cancer

metastatic to the liver experienced a partial response of

hepatic metastases with stable pancreatic head abnormali-

Received 9/8/97; revised 5/26/98; accepted 5/26/98.

The costs of publication of this article were defrayed in part by the

payment of page charges. This article must therefore be hereby marked

advertisement in accordance with 18 U.S.C. Section 1734 solely to

indicate this fact.

1 This study was conducted with support from Pharmacia and Upjohn

AB. B. J. G. was supported by NIH Grant Kl2 CA 01728. Additional

support was provided by NIH Grant CA 06927, the Commonwealth of

Pennsylvania, and the Bernard A. and Rebecca S. Bernard Foundation.

2 To whom requests for reprints should be addressed, at Fox ChaseCancer Center, 7701 Burholme Avenue, Philadelphia, PA 191 1 1. Phone:(215) 728-2629; Fax: (215) 728-2741; E-mail: [email protected].

ties by computed tomography scan. Further dose escalation

was suspended when two patients treated in a companion

repeat dose Phase I study experienced DLT at the 4 nglkg

dose level. Multiparameter analyses on all patients treated in

the two companion single-dose and two-repeated-dose Phase

I trials revealed that the levels of patients’ pretreatment

anti-SEA antibodies protected against toxicity at a given

drug dose. By jointly considering weight and the baseline

anti-SEA concentration in a patient, it is possible to assign a

PNU-214565 dose that will induce systemic cytokine release

(a surrogate test to assess for the presence of uncomplexed

drug and its ability to induce systemic cellular activation)

without DLT. This pharmacodynamically based dosing

scheme will be tested in future repeated-dose clinical trials

and will define maximally tolerated doses of this powerful

new immunotherapy approach.

INTRODUCTIONBiological response-modifying agents designed to augment

immune responsiveness have been the standard in preventative

medicine since the turn of the century in the form of vaccines to

enhance adaptive immune responses to infectious agents. More

recently, therapeutic strategies incorporating biological response

modifiers have aimed to enhance cell-mediated TH 1 responses

for the treatment of human diseases. Both TH1 and TH2 re-

sponses require antigen processing and presentation, are MHC

restricted, and are dependent on the initial antigen-specific ac-

tivation of T-cell subsets through the TCR.3 Antigen-specific

T-cell frequency has been estimated at one in l0�-l0” cells (1).

In attempts to increase both the frequency and activation of

tumor antigen-specific cells, treatment modalities incorporating

cytokines and growth factors have been used to up-regulate

effector cell function (2), alone or in concert with adoptive

immunotherapy and the infusion of ex vivo expanded tumor

infiltrating lymphocytes or lymphokine activated killer cells (3,

4). Immunotherapy with monoclonal and bispecific antibodies

has been used to activate and target T-cell and natural killer-cell

populations, both systemically and at tumor site (5-7).

SAgs harness potent immunostimulatory capabilities. The

activation potential of SEA is exhibited when the SEA molecule

simultaneously engages both the MHC class II molecule and

specific variable regions of the �3 chain (V�3) of the TCR, with

the induction of multiple cytokines, and T-cell activation and

3 The abbreviations used are: TCR, T-cell receptor; SAg, superantigen;

SEA, staphylococcal enterotoxin A; APC, antigen-presenting cell;SDCC, staphylococcal enterotoxin-dependent cell-mediated cytotoxic-

ity; DLT, dose-limiting toxicity; HAMA, human anti-mouse antibody;IL, interleukin; TNF, tumor necrosis factor; PE, phycoerythrin; PHA,phytohemagglutinin; FBS, fetal bovine serum.

Research. on December 20, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

1904 Phase I Repeat Dose Fab-SEA

Table 1 Dose escalation schema

Dose (ng/kg)� No. of patients Cumulative dose range (ng)

0.15 3 32.6-68.90.5 3 88.0-2091.5 4 340.2-498

2.75 12 258-1112.6

3.5 5 888-1263.6

a Administered by 3-h infusion daily for 4 consecutive days.

proliferation. SEA contains two independent but cooperative

binding sites for MHC class II (8-12) and, in addition, a binding

site for specific TCR Vf3 subsets (1, 13, 14). SEA interacts with

APCs in a MHC class 11-dependent but not MHC-restricted

manner, with MHC class II binding occurring outside the pep-

tide binding groove. SEA activity does not require APC inter-

nalization or processing before the simultaneous SEA-T cell

interaction (1 , 15-19). Optimal SEA activity is induced when

both SEA-MHC class II sites are engaged. Engagement can

occur between two SEA molecules on one APC, one SEA with

two APCs, or by cross-linking two MHC class II molecules on

the same APC, resulting not only in a stable alignment for

SEA-T cell interactions but at the same time activating the APC

(8). The number and nature of the SAg-MHC class II interac-

tions may determine the outcome of the SAg-T cell encounter,

which could induce activation, proliferation, depletion, or an-

ergy of T-cell V�3 subsets (20). SAgs have the potential to

activate more than 10-25% of the T-cell population (1, 12-14).

SEA targets T-cell cytotoxicity in a MHC class 11-depend-

ent manner, making MHC class Il-expressing tumor cells sus-

ceptible to SEA-induced effects. MHC class Il-dependent cyto-

toxicity has been termed SDCC (21-25). Treatment modalities

exploiting the therapeutic potential of SAgs require SEA mod-

ification to decrease the systemic activation resulting from SEA-

MHC class II interaction with monocytes and B cells and to

localize the cytotoxic capabilities of SAg-activated T cells at

tumor sites. Genetic fusion of SEA to the Fab of a tumor-

targeting monoclonal antibody has resulted in molecules that

show a 10-fold reduction in SEA-MHC class II binding and

contain an antibody portion with a 100-fold stronger tumor

antigen affinity than the SEA-MHC class II affinity (26-32).

Such fusion proteins support superantigen-mediated antibody-

dependent cell cytotoxicity, with lessened MHC class II depend-

ence, and endow tumor cells with superantigenicity capable of

activating tumor infiltrating T cells exhibiting selective

SEA-V�3 specificities (27).

Phase I clinical trials using a single dose of the superanti-

gen-based agent PNU-2l4565, a recombinant fusion protein of

SEA and the Fab of the tumor targeting monoclonal antibody

C242, have been conducted at this institution (33) and at Rig-

shospitalet in Copenhagen, Denmark. The single dose trials

demonstrated the activation potency of this novel reagent, iden-

tified DLTs, and outlined a safe starting dose for the repeated

dose trial presented here. The present trial has revealed a rela-

tionship of the clinical toxicities to pretreatment anti-SEA sera

antibody concentrations and administered drug doses and has

led to the development of a pharmacodynamically based dosing

strategy.

Table 2 Patient demographics

Total entered 27Female 13Male 14

Median age (range) 62 (36-75)Baseline performance status (ECOG”)

0 15

1 12

Primary tumor siteRectal 5

Colon 18Pancreatic 4

Prior treatmentSurgery 24Chemotherapy 26Radiation 7

a ECOG, Eastern Cooperative Oncology Group.

PATIENTS, MATERIALS, AND METHODSPatient Population. The inclusion criteria for the present

repeated-dose study were consistent with those outlined in the

previously described single-dose clinical trial (33). Eligible pa-

tients were required to have a histological diagnosis of incurable

advanced pancreatic or colorectal adenocarcinoma. All patients

were 18 years of age or older with an Eastern Cooperative

Oncology Group performance status of 0 or 1 . Patients had

received no radiotherapy, chemotherapy, immunotherapy, or

biological therapy within 4 weeks of entry into this study (6

weeks for mitomycin C or nitrosoureas). All patients had ade-

quate bone marrow (WBC �3,000/mm3, absolute neutrophil

count �2,000/mm3, and platelets � 150,000/mm3) and adequate

hepatic and renal function (total bilirubin � 1 .5 times upper limit

of normal; aspartate aminotransferase and alanine aminotrans-

ferase less than or equal to twice the upper limit of normal

unless liver metastases, then �4 times the upper limit of normal;

and serum creatinine � 1 .5 mg/dl). The use of 3-blockers, either

for hypertension or glaucoma, had to be discontinued a mini-

mum of S days before treatment. Patients with pulmonary dis-

ease requiring active therapy, New York Heart Association

classes II-IV heart disease, or uncontrolled hypertension were

excluded from therapy. Written informed consent was obtained

in accordance with federal, state, and institutional guidelines.

Prior to treatment initiation, a medical history, physical exam-

ination, and clinical laboratory evaluations, including hematol-

ogy, coagulation, and chemistry profiles, were performed, and

samples were obtained for the baseline laboratory investigative

studies. Laboratory evaluations included flow cytometry analy-

ses, lymphocyte proliferation assays, sera cytokines, soluble

CA242 antigen, anti-SEA and HAMA levels, and plasma drug

levels. Serial blood samples were obtained throughout the treat-

ment course. Tumor assessment was performed at baseline and

day 28 after treatment using standard response criteria (34).

Treatment Plan. Patients were admitted for the duration

of PNU-2l4565 treatment. The patients received a daily 3-h iv.

infusion on 4 consecutive days. Therapy was administered using

a CADD-micro pump (Pharmacia-Deltec, St. Paul, MN). The

starting dose of PNU-214565 was 0.15 nglkg/day with escala-

tion as described in Table I . Three patients were accrued to each

dose level. Patients received only one 4-day course of treatment.



8000-

7000-

6000-

5000-

4000�

3000-

2000-

1000�

�Th�’I’’I’’I’�I’

Hour 051105110511051100

Day 1 2 3 4 511

A

Cl)

EEU)

ci,0

ci)

.0

Ez

B

Fig. I PNU-214565-induced hematological and cytokine effects. A,

WBC subset fluctuations occurring before and at 5 and 1 1 h after the4-day infusions for patient 12 treated at the 2.75 ng/kg PNU-2l4565

dose. 0, total WBC; #{149}, platelets; A, lymphocytes; 0, monocytes. B.TNF-a (#{149})and IL-2 (fl) profile over the 4 days of treatment.

The 2.75 nglkg dose level was expanded after DLTs were

observed in a concurrently running companion trial.4 Dose

escalation was terminated at 3.5 nglkg, when the observed

relationship between clinical toxicity and the contribution of the

baseline sera anti-SEA concentrations relative to administered

drug dose became apparent.

Expression and Purification of PNU-214565. The fu-

sion protein ofC242 Fab-SEA (PNU-2l4565) was prepared and

provided by PharmacialUpjohn (Lund, Sweden) as described

previously (33).

Clinical Cancer Research 1905

4 S-E. Nielsen, personal communication.

250-

200-

-jE-� 150-

C’J

-J

100-

50-

-- 1 I I I I I I � I I � I I ‘ ‘ ‘ IHour 035110351103511035110

Day 1 2 3 4 5

Day and hour of sampling

-500 Cytokine Determinations. Immunoenzymetric assays

were performed using Medgenix Diagnostics kits (Fleurus, Bel-

gium) following manufacturer’ s procedures for the quantitative.400 measurements of IL-2 and TNF-a. Serial samples were analyzed

daily at preinfusion, 3, 5, 1 1, and 24 h after the final drug

infusion.

.300 .�,, Flow Cytometry. Flow cytometric analysis was per-

�g:� formed on peripheral blood mononuclear cell preparations ob-

� tamed and processed using Ficoll-Paque (Pharmacia, Uppsala,-200 b Sweden) density centrifugation preinfusion days 1 and 2, 24 h

after the last PNU-2l4565 infusion, and on day 1 1. Cells were

dually labeled (FITCIPE) with CD4/CD4SRO, CD8/CD4SRO,

-100 CD14/CD16, CD4IHLADR, CD8/HLADR, and CD19/

HLADR. The fluorescein (FITC)- and PE-conjugated mono-

clonal antibodies were purchased from Becton Dickinson (San

-0 Jose, CA). Analysis was performed on a FACScan flow cytom-

eter (Becton Dickinson) equipped with Cell Quest software

600 BDIS. Gates were selected to include the CD4S positive lym-- phocyte/monocyte populations for analyses. Bridging assays

were performed by pre-incubating the patient’s lymphocytes,

-500 suspended in either PBS or autologous plasma, in the presence

and absence of PNU-214565 for 30 mm at 37#{176}Cfollowed by

400 dual labeling with CD3-FITC and CD14-PE.

- :i� PNU-214565 Plasma Levels. Blood samples were

-� drawn, and plasma was obtained at baseline and at hour 3 on

-300 ; days 1 and 4. Plasma was stored at -70#{176}Cfor batch analyses.

iL Infusion (final dilution) solutions were analyzed as a quality

#{149}200 control measure. Analyses were performed by Pharmacia/Up-john (Helsingborg, Sweden) as described previously (33). Anal-

� ysis of the infusion drug doses indicated that all patients re-

-100 ceived the intended dose. Circulating PNU-2l4565 levels were

. undetectable at the doses administered in this study. The range

of the cumulative doses administered in this study was 32.6 to

-0 1263.6 ng.

Humoral Immune Responses: Human Anti-Immune

Antibody (HAMA), Anti-SEA, and Soluble CA242 Antigen

Levels. Blood samples were drawn, and plasma was obtained

at baseline and days 1 1 and 28 after initiation of treatment.

Plasma was stored at -20#{176}Cfor batch analyses. HAMA was

analyzed using Immun STRIP HAMA fragment (Immunomed-

ics, Moms Plains, NJ) following the manufacturer’s procedure.

Anti-SEA was determined as described previously (33). CA242

circulating antigens levels were performed using the EIA

CanAg CA242 kit (CanAg Diagnostics, Gothenburg, Sweden)

following the manufacturer’s instructions.

Lymphocyte Proliferation Assays. PNU-2 14565-

induced lymphocyte proliferation assays were performed on

Ficoll-prepared lymphocytes from blood samples collected at

baseline, prior to the day 2 infusion, 24 h after the last PNU-

214565 infusion, and on day 1 1 as described previously (33).

Additional proliferation assays using a standard lymphocyte

donor and patient autologous plasma at baseline were performed

to assess the contribution of autologous plasma factors. All

assays used serial dilutions of PNU-2l4565. PHA was used as

a positive control stimulator. Assays were incubated 72 h at

37#{176}Cwith [3H]thymidine added during the last 4 h of incuba-

tion. Selected patient plasma samples were affinity chromatog-

raphy absorbed to remove the anti-SEA. Repeat lymphocyte




Table 3 Toxicity correlates

Mean peak

cytokine inductionToxicity Percentage of loss from baseline at nadir (pg/ml)#{176}

Grade Type WBC PMN” Lymphocytes Monocytes Platelets IL-2 TNF-a

�l 18.4 27.9 46 68.4 18.8 27.5 21.5

(n 12)2 Hypotension 19.3 30.8 29.4 70.5 16.5 36.4 13.8

(n 8)

2 Fever 38 46 66.7 70.3 36.3 94.3’ 184.3”(n 3)

2 Fever plus hypotension 39.7 32.7 90.3 98.3 44.3 193.9” 1587d

(n 3)

4 Hypotension’� 82 80 100 100 17 2798’ 1707’

(n 1)

a Significant levels for IL-2 exceed 94.3 pg/ml; significant TNF-cr levels exceed 40 pg/ml.b PMN, polymorphonuclear lymphocytes.C Increased (P < 0.05) compared with results from patients with � grade 1 toxicity or grade 2 hypotension.d Increased (P < 0.05) compared with results from patients with � grade 1 toxicity or grade 2 hypotension, but not different from Grade 2 fever.e Grade 4 hypotension was accompanied by a grade 2 fever.

�Increased (P < 0.05) compared with results from patients with � grade 1 and grade 2 fever, hypotension, and fever with hypotension.

proliferation assays were performed to assess the contribution of

the anti-SEA to the overall plasma inhibition factor.

HLA-DR Analyses. Selected patients were HLA-DR

typed as described previously (33).

Toxicity Model Derivation. The equation below used to

calculate the predicted dose for a 50% probability of cytokine

induction was derived retrospectively following analysis of data

obtained from patients treated with a single dose of PNU-

214565 (33), omitting patient 2, and those described in this

present trial to assess the association of the pretreatment base-

line anti-SEA concentration and the administered dose to the

observed treatment-induced toxicities.

Predicted dose = exp [ln (probability/lOO - probability) -

5.83 + 1 .770 x ln (pmollml anti-SEA)/0.824]

The predicted dose in nglkg required to elicit a systemic

cytokine response at a predetermined probability rate is calcu-

lated using this formula. Any desired probability (e.g., 10%,

50%, or 90%) that a systemic cytokine induction will be elicited

is inserted, as is the patient’s baseline pretreatment anti-SEA

concentration (expressed in pmol/ml) to derive the dose to be

used.

Statistics. Statistics were made using the standard Stu-

dent’s t test.

RESULTSTwenty-seven patients were treated on this study. Charac-

teristics of the patients are listed in Table 2. The age range was

36-75, with a median of 62; 14 were males, and 13 were

females. Of the 27 patients treated, 23 had colorectai cancer and

the remaining 4 had pancreatic cancer. Prior to treatment, 24

patients had undergone surgery, 26 had received chemotherapy,

and 7 had received radiation therapy. The performance status

was either 0 (n = 15) or 1 (n = 12).

Clinical Assessment. No DLTs were observed in pa-

tients treated at the 0.15-, 0.5-, or 1.5-nglkg dose levels. The

2.75-ng/kg dose was expanded to include 12 patients after DLT

(grade 4 hypotension) was experienced by one patient, as de-

scribed below. Two additional grade 4 toxicities were observed

at the 4-ng/kg dose level in a companion clinical trial performed

at Rigshospitalet.4 Five additional patients were treated at 3.5

ng/kg with no DLTs. The study was closed at this dose level

after the observation of a relationship of the pretreatment anti-

SEA sera concentration relative to the administered drug dose

and the probability of clinical toxicities.5

Twenty-five of 27 patients received the full 4-day cycle of

PNU-2l4565. Treatrnent was terminated after one dose of drug

in patient 20 (2.75 ng/kg), who developed a transient grade 4

hypotension responsive to iv. fluid and low-dose dopamine.

Day 4 treatment was not administered in patient 23 after the

development of a grade 2 skin rash on day 3 that did not resolve.

As in the single-dose trial, fever and hypotension were the

most common nonhematological toxicities experienced after

infusion of PNU-2l4565 (1). Grade 1 fever occurred in 41% (11

of 27) of the patients, and grade 2 fever occurred in 26% (7 of

27). Fever was not dose-limiting. Hypotension was encountered

in 89% (24 of 27) of the treated patients: 1 1 events (41%) were

grade 1 ; 12 (44%) were grade 2; and 1 patient (4%) exhibited a

dose-limiting grade 4 toxicity (see above). Other toxicities ex-

perienced included diaphoresis, nausea, vomiting, rigors, head-

ache, diarrhea, anxiety, and edema. Except for one episode of

grade 2 diaphoresis and one grade 2 edema of the feet, none of

the additional toxicities advanced further than a grade 1. In

general, toxicities were most prominent on day 1 . Toxicities

were low grade, transient, and subsided within 24 h of the

conclusion of treatment.

All treated patients experienced transient losses in WBC

5 R. Persson, 0. Nordle, et al. Predictability of immunological response

and severe toxicological effects of PNU-214565 in cancer patients,manuscript in preparation.



i’�’_

. � --.‘

r�’�

Baseline


One MonthFig. 2 Computed tomography scan of patient 12. The upper panel isbaseline, and the lower panel is one month after the initiation oftreatment. Arrow, resolution of metastases in right lower quadrant.

subsets. The degree of loss correlated with toxicity and cytokine

induction. Fig. 1 illustrates the correlation of the hematological

subset fluctuations and cytokine induction for patient 12. Table

3 shows the correlation of lymphocyte subsets decrease, toxicity

grade, toxicity type, and peak cytokine induction. For toxicities

less than or equal to grade 2, no significant difference was

observed in the loss of WBCs and polymorphonuclear lympho-

cytes. The development of fever, alone or in combination with

hypotension, correlated with lymphocyte loss and cytokine in-

duction. The IL-2 induction (grade 2 fever plus hypotension)

had greater impact on the lymphocyte and monocyte decreases

than TNF-a induction at equivalent toxicity grade. The one

patient with grade 4 toxicity had significant cytokine induction,

Table 4 HL A-DR typing

Patient no. HLA-DR allotype

1617

18

19”

20#{176}

22”2425

26

27”28”

4/701.534,53

3/701.52/53

13/701, 52/53

1/4,53

3/11,521/4,533/701,52/53

2/3,51/52

15/14,51/521/2,51

(I Patients with g rade �2 toxicity.

and all white cell subsets were drastically reduced. However,

platelet decreases were not significant in any of the patients.

Clinical Outcomes. An overall response lasting 2.5

months was experienced by patient 12, who had been diagnosed

with pancreatic cancer metastatic to the liver. This patient was

treated at 2.75 ngfkg. At 28 days, a partial response of the

hepatic metastases with stable pancreatic head abnormalities

was demonstrated by computed tomography scan (Fig. 2).

Laboratory Assessment: PNU-214565-associated Ef-fects. Samples from 16 of 27 patients (59%) showed a 1.1-3-

fold increase in lymphocyte counts on day 1 1 when compared

with baseline values. Flow cytometry analysis showed some

evidence of activation during and after the course of treatment in

the majority of patients. Activation was indicated by increases in

the total percentage of dually labeled CD14+/CD16+ mono-

cytes, CD8+ lymphocyte populations, or CD16+ natural killer

populations, accompanied by mean fluorescent index increases

for HLA-DR and CD45RO both on CD4+ and CD8+ cells

(data not shown).

Humoral immune responses were assessed to determine the

response to the Fab C242 mouse-derived portion of the fusion

protein. All patients were followed a minimum of 1 month after

treatment. As predicted by the low protein level administered

and the Fab nature of the mouse component, no patient showed

a HAMA response. One patient (patient 22) demonstrated 39.5

ng/ml HAMA to Fab mouse fragment pretreatment with a drop

to 4. 1 ng/ml at the 1 month interval. The pretreatment value was

unexplained. The patient’s medical history gave no indication of

prior murine monoclonal antibody exposure.

Anti-SEA concentrations were evaluated at baseline, day

1 1 , and 1 month after treatment to investigate immune responses

to the SEA portion of the fusion protein. Baseline anti-SEA

concentrations ranged from 15.4 to 2091 pmol/ml. No patients

at the 0. l5-nglkg dose level showed increases over baseline in

their anti-SEA concentrations. Two patients treated with 0.5

ng/kg dose level showed a 2-fold increase at 1 month over

baseline. Fifteen patients treated at the l.5-nglkg and higher

dose levels had increases in their anti-SEA concentrations of

2-381-fold at 1 month after treatment (mean, 39.5-fold; median,

3.3-fold). The greatest increase was observed in a patient treated

at 3.5 nglkg with a baseline concentration of 24 pmol/ml. This

patient’s anti-SEA concentration increased to 9152 pmol/ml at 1

month after treatment. The “threshold” for anti-SEA induction



A 5

4

0

x 3 o 000

0

#{149}0 2(5

0

0 � �__

C 2000

�. 1500

E

.2�

U-zI-. 500


0

B 3000

�. 2000

Cs,’

� 1000

t�rTI1

Anti-SEA/dose ratio

Fig. 3 Association of the anti-SEA/dose ratio to toxicity. For eachtreated patient, the highest grade ofclinical toxicity (A) exhibited during

treatment, peak IL-2 levels (B), and peak TNF-a levels (C) are plottedagainst the ratio of the anti-SEA concentration (pmollml) to adminis-tered dose (ng/kg). Note: graphed results represent a composite of the 84patients treated to date in four separate studies of PNU-214565. Some

points represent multiple patients exhibiting the same anti-SEA drug

ratios and equivalent toxicity grades or cytokine induction.

appears to be at or near a dose of 0.5 nglkg. Once the threshold

is reached, the factors determining the extent of anti-SEA in-

duction are not entirely dose dependent.

Soluble CA242 antigen levels were also followed up to 1

month after treatment. Evaluation of CA242 levels in patients

treated in the single dose trial (33) showed a trend suggesting

that patients with high circulating antigen levels had less toxic-

ity. A similar evaluation of patients in this study showed no

differences in circulating CA242 levels at baseline in those

patients with grade 1 (1420 ± 652 units/ml), grade 2 (1698 ±

947 units/mi), and grade 4 (1621 units/mi) toxicity. Levels

ranged from undetectable to 10,087 units/nil at baseline. No

patient in this study had greater than a 4.7-fold increase at 1

month, 12 patients had 1.5-fold or less, 10 patients demonstrated

a 1.6-3.0-fold increase, and five patients had a 3.3-4.7-fold

increase. There were no noted correlations of circulating CA242

antigen levels and disease status.

Table 4 gives the HLA-DR allotypes for the patients eval-

uated. A number of the allotyped patients expressed the DR53

allele. The DR53 allele has an amino acid alteration at position

8 1 , changing the histidine to tyrosine. This histidine at position

81 is responsible for the SEA-DR1 moderate affinity binding

(35-37). Although investigations have not shown SEA-MHC

class II binding to be affected significantly by the HLA-DR

allotype (37, 38), augmentation of this moderate affinity MHC

class LI-SEA binding could contribute to the outcome resulting

from the subsequent SAg-TCR V�3 interaction. However, no

associations between allotypes and toxicities have been ob-

served.

Pretreatment Anti-SEA Concentration and Association

with Toxicity. Evaluations following the single-dose trial (1)

and the initial part of the present repeated-dose trial indicated

that the observed clinical toxicities were not solely dependent on

the weight-based dose assignment. When baseline anti-SEA

concentrations in relation to administered dose were considered,

an association with toxicity was revealed. Fig. 3 shows data for

the 84 patients treated in the single- and repeated-dose trials at

this facility and Rigshospitalet. The highest grade of clinical

toxicity experienced by each patient during the course of treat-

ment is plotted against the ratio of the baseline anti-SEA con-

centration (pmol/ml) to the daily administered dose (nglkg).

Patients with grade 3 toxicities all had ratios less than 30: 1, with

all three grade 4 toxicities occurring in patients having a ratio of

6: 1. Likewise, significant IL-2 and TNF-a induction was de-

tected in those patients with ratios less than 10:1.

Calculations were derived (see “Materials and Methods”)

to define a dose with 50% probability of inducing systemic

cytokines, which is used as an indication of the presence of

uncomplexed drug and its ability to induce systemic cellular

activation, or DLT. As the calculation indicates, the relationship

between the predicted dose and the anti-SEA antibody concen-

tration is not linear. Using a 50% probability of systemic cyto-

kine induction (IL-2 and TNF-a), Fig. 4 (upperpanel) illustrates

the ratio of actual to predicted administered dose for the treated

patients in this study and indicates that 7 of 27 patients received

a ratio greater than 1. Fig. 4 (lower panel) gives the peak IL-2

(0) and TNF-a (#{149})levels for those patients. Three of the 7 (or

50%, as predicted by the calculation) did in fact exhibit a

cytokine induction. Patient 20, who experienced a grade 4

hypotension with significant levels of circulating cytokines,

received 9.2 times the dose predictive for cytokine induction

(Table 5). Cytokine induction was detected in Patient 12 (grade

2 fever) and patient 23 (grade 2 fever and hypotension). These

patients received 3 and 1.3 times, respectively, the predicted

dose. Although there was sufficient uncomplexed drug to induce

systemic cytokine, the degree of cytokine induction in these

patients did not induce DLT. However, patient 28 (4.5 times

predicted dose and grade 2 hypotension), patient 17 (3.0 times

predicted dose, no toxicity), patient 19 (2.3 times predicted

dose, grade 2 fever and hypotension), and patient 14 (1.3 times



A

0)-�

C)C

ci,C,)

0a�0ci)0

�0ci)

CS

0

0

(Scc

B

0

-I

EC)0.

-I

(S

0�

3000

2500

2000

1500

1000

500

.-J

E

U.

zI-

ci,

Patient #


Fig. 4 A, actual/predicted dose

ratio. The predicted drug dose isbased on a 50% probability forcytokine induction (see text).---‘ 1:1 ratio, where the pa-tient received the predicted drug

dose. B, peak IL-2 (E’ left axis)

and TNF-a (S, right axis) cyto-

kine induction detected during

each patient’s treatment course.

predicted dose, no toxicity) had only minimal detectable levels

of cytokines. These data illustrate a relative association between

the anti-SEA concentration and the administered drug dose with

respect to toxicity. However, components of the total anti-SEA

concentration may play a more toxicity “protective” role as

shown by the outcome of patients 12 and 23 compared with

patients 28, 17, 19, and 14.

Lymphocyte proliferation assays provide a functional

means to assess the impact of inhibitory anti-SEA components

in patient plasma. Fig. 5 compares the PNU-2l4565 molar

concentrations required for half-maximal stimulation plotted

against the baseline anti-SEA concentrations (pmol/ml) for each

treated patient. These assays were performed using a single

healthy donor’s lymphocytes to assess the impact of patients’

circulating anti-SEA concentration. Donor lymphocytes assayed

in FBS were half-maximally stimulated at a PNU-2l4565 con-

centration of 9 x l0�� M. Thus, half-maximal PNU-214565

concentrations approaching 9 X 10 “ M would indicate a lower

inhibitory/neutralizing component in a given patient’s plasma.

Table 5 shows that the plasma of six patients (patients 2, 4,

12, 17, 20, and 23) had lowered half-maximal PNU-2l4565

concentration requirements for lymphocyte proliferation. Con-




Table 5 Toxicity predictors

Baseline Ratio Lymphocyte proliferation”Dose Patient anti-SEA anti-SEA! Ratio actuaL’ Clinical half-maximal stimulation

(ng/kg) no. (pmollml) dose predicted dose” toxicity (molar concentration of drug)

0.15 1

23

149

1697

993

107647

0.00381

0.4420.0097

G2HC

6 x l0”

2 x 10 �1 x �

0.5 45

6

33262

26

66524

52

0.3320.003780.535

G2FFI 2 x l0�”

4 x l0�5 x 10’ �

1.5 789

10

15072

2771062

10048

185708

0.03750.1830.01010.00056

G2HG2HG2HG2H

5 x l0’�6 x l0’�6 x l0’�7 X iO�

2.75 11

12

1314

1516

17

18

19202223

749

26

75238

77172

26

172

30155538

272

9

27314

2863

9

63

115

2014

0.00218

3.07

0.002161.29

0.2870.0513

3.10

0.0513

2.269.150.6031.31

G2F

G2H

G2F

G2FHG2F, G4HG2HG2FH

6 x l0’�

5 X 10#{176}4 x iO�4X iO�

2.5 x 10_b

1.5 x l0��

6.5 x l0�”

6 x l0��

3 x l0�’#{176}4 x 10 �4 x l0�7 x l0#{176}

3.5 2425

262728

2091120

852059

24

59734

24588

9

0.0003050.141

0.2970.0003154.49

G2HG2H

6 x l0�5 x l0�

1.7 x l0��6 X l0�5 x 10_co

a Ratio obtained by dividing the administered dose (ng/kg) by the predicted dose for a 50% probability of cytokine induction (see “Materials and

Methods”).b Assay used healthy donor lymphocytes and patient plasma; half-maximal stimulation was 9 X 10 13 M for the healthy donor lymphocytes

assayed in FBS.C G, grade; F, fever; H, hypotension; all patients with blanks had � grade 1 clinical toxicity.

centrations in the 10 ‘ ‘ M range indicate minimal inhibition.

Although patients 2 and 4 had half-maximal drug concentrations

in this range of l0 ‘ ‘ M, their lack of severe clinical toxicity and

a lack of cytokine induction was most likely related to the low

drug doses these patients received. In contrast, patients 12, 20,

and 23 also had half-maximal drug concentration requirements

in this range, but all experienced clinical toxicity and significant

cytokine induction (Fig. 5, .1.). All of these patients received

greater than the predicted drug dose based on their total baseline

anti-SEA concentration (Fig. 4). Half-maximal concentrations

and anti-SEA/drug dose ratios would have predicted that patient

17 would experience severe toxicity and cytokine induction.

However, this was not observed. Five patients demonstrated

half-maximal PNU-2l4565 concentration requirements of

l0 ‘#{176}M (patients 3, 6, 15, 19, and 28). Toxicity would not have

been predicted for patients 3 and 6. These two patients demon-

strated an intermediate level of inhibition and received low drug

doses. The lack of toxicity experienced by patients 15, 19, and

28, who received 0.3, 2.3, and 4.4 times the predicted dose,

respectively, demonstrates that this intermediate level of inhi-

bition is sufficient to protect against severe toxicity at the drug

doses received. All remaining patients contained a sufficient

level of neutralizing anti-SEA to protect against toxicity at the

doses administered.

Affinity chromatographic absorption was performed on

selected patient plasma samples to remove the anti-SEA to

definitively assess its contribution to the half-maximal drug

concentrations required for proliferative stimulation. Table 6

indicates that lowered drug concentrations are capable of half-

maximally stimulating donor lymphocytes after anti-SEA ab-

sorption. Patients 8 and 12 had similar preabsorption anti-SEA

concentrations. The plasma from patient 12 exhibited little in-

hibition of PNU-214565-induced lymphocyte proliferation. This

patient exhibited grade 2 fever and cytokine induction after

treatment with 2.75 nglkg PNU-214565. In comparison, lym-

phocyte proliferation was inhibited by plasma from patient 8,

who exhibited only grade 2 hypotension and had no detectable

cytokine induction. Patients 16 and 24 had neither clinical

toxicity nor cytokine induction, suggesting the presence of ad-

equate PNU-2l4565-neutralizing anti-SEA for the drug dose

administered.

Fig. 6 depicts the SEA-induced cell-cell interaction that can

occur systemically. It illustrates the variable capacity of the

plasma of two patients with comparable “low” anti-SEA con-

centrations, and who received the predicted drug dose, to block

the PNU-214565-induced bridging of monocyteslB cells and T

cells resulting from MHC class 11-SEA and TCR Vp-SEA

interactions. Fig. 6A shows that PNU-2l4565-induced bridging

was reduced from 5.61% in PBS to 1.25% in baseline plasma for

patient 19. This patient had a baseline anti-SEA of 30 pmol/ml



-IE0

E0.C0

CS

Cci)0C00

w

C

10.11 10.10 i�-� 10.8


Fig. 5 Relationship of anti-SEA to PNU-214565 concentrations required for half-maximal stimulation. The assay was per-formed with donor cells and 10% patientplasma. lO_8 M PNU-214565 was the high-est concentration used in the assay. Toxicitygrade for each patient: #{149},�grade 1 . C, grade2 hypotension alone; �, grade 2 fever aloneor in combination with hypotension: �, grade4 hypotension. � , patients showing signifi-cant cytokine induction.

Table 6 Lymphocyte proliferation after affinity chromatographyabsorption

Preabsorption Postabsorption

Half-maximal Half-maximalPatient Anti-SEA stimulation Anti-SEA stimulation”

no. (pmol/ml) (molar drug) (pmol/ml) (molar drug)

8 66 5.5 x 10� 25 3 x 10- 12

12 65 4Xl01’ 48 <i#{128}r’416 203 5 x l0’#{176} 46 <10- 14

24 1504 6 X l0� 34 2.3 X 10- 12

a Half-maximal stimulation for healthy donor cells in FBS was

4.5 X 1013 M.

with a half-maximal drug stimulation concentration in plasma of

3 X l0 �#{176} M. After treatment with 2.75 nglkg PNU-2 14565,

patient 19 exhibited grade 2 fever and hypotension with minimal

TNF-a induction (81 pg/mI). The bridging assay depicted in

Fig. 6B showing no or minimal plasma blocking capability

(1.55% cell bridging in PBS versus 1.33% in plasma) is for

patient 20 treated at the 2.75 nglkg dose level and having a

baseline anti-SEA of 15 pmol/ml and a half-maximal lympho-

cyte proliferation drug concentration requirement of 4 X l0 “

M. This patient exhibited grade 2 fever and grade 4 hypotension

and had significant cytokine induction (1707 pg/ml TNF-a;

2822 pg/ml IL-2). Correlation of the systemically SEA-induced

cell-cell interaction and toxicity could only be demonstrated in

patients with low anti-SEA concentrations who received suffi-

cient drug to exceed the blocking/inhibitory component in their

total anti-SEA.

PNU-214565 Molar concentration for half maximal stimulation

DISCUSSION

The present clinical trial is the first report of repeated

dosing with an SEA-based fusion protein and shows that PNU-

214565 can be administered safely in a repeated-dose regimen.

No cumulative dose-related toxicities were observed. The tox-

icities encountered were manageable and of short duration.

Laboratory and clinical findings showed no detectable HAMA

to the C242 Fab portion at the doses administered, but there

were variable responses to the SEA portion of the fusion protein.

Toxicity correlated with systemic cytokine induction and the

extravasation of WBC subsets. The required PNU-2l4565 con-

centration for half-maximal lymphocyte proliferation stimula-

tion correlated with treatment-induced cytokine release, the type

and grade of toxicity, and the ratio of baseline circulating

anti-SEA concentration to administered drug dose.

T-cell activation using immunoconjugates of SAg has ad-

vantages over other monoclonal antibody-based strategies (39).

SAg activates a limited, although significant, fraction of T

lymphocytes, and in contrast to pan T-cell activators, can be

expected to leave the majority of the T-cell repertoire intact,

thereby avoiding general immune suppression. In addition,

whereas monoclonal antibodies penetrate tumors poorly, T-

lymphocytes are not limited by such factors as interstitial pres-

sure and, depending on the relevant chemotactic signal, can

move freely throughout large tumor masses.

Cytokine induction was the most reliable indicator of a

systemic immunological response and was used as a surrogate

measure of uncomplexed circulating drug and SEA-induced

cell-cell activation mediated by the SEA portion of the agent in this

study. Such responses indicate that circulating drug levels are

sufficient to exceed the circulating inhibitory anti-SEA plasma



PBSA

B

Autolog Dus �asma

#{149}1.

w

1�

0

�3

No PNU-214565 PNU-21 4565

CD3FITC

6 J� Babb, A. Rogatko, and S. Zacks. Controlling decision-error proba-

bilities with Bayesian test procedures, submitted for publication.7 S. Zacks, A. Rogatko, and J. Babb. Optimal Bayesian-feasible doseescalation for cancer Phase I trials, submitted for publication.


1.39’� 1.25

�

Fig. 6 Flow cytometric bridging assays. A, a patient (patient 19) witha low anti-SEA concentration containing a significant blocking/inhibi-tory/neutralizing component preventing bridging via PNU-214565 (low-

er right) compared with assays in PBS (upper right). Upper and lower

left contours represent PBS and plasma controls with no PNU-2 14565.B, upper and lower right contours depict a patient (patient 20) withminimal plasma blocking capacity.

component, allowing drug to intravascularly stimulate patient cells

via SDCC. Also, systemic IL-2 induction would suggest the pres-

ence of the appropriate SEA-TCR V�3 subsets. Circulating drug

plasma levels at the doses administered in this trial are estimated to

be in the attomolar range (l0 16 to 10 18 M), and the observed

toxicities underscore the remarkable activation potency of this

agent. In vitro testing shows SEA-induced lymphocyte prolifera-

flon to be half-maximal at drug concentrations in the pico-femto-

molar range when assays are performed in FBS and in the nano-

molar range in the presence of plasma.

The relative ratio of the baseline circulating anti-SEA

concentration to drug dose, although not linear, has been the

best predictor of toxicity and showed patients having ratios

<10-15 at the most risk of toxicity. Drug doses must be

balanced to achieve circulating levels in excess of anti-SEA

with acceptable toxicities. In this trial, this balanced excess of

drug was accomplished in only 7 of 27 patients (patients 12, 14,

17, 19, 20, 23, and 28; Table 5; Fig. 4). In the present clinical

trials that build on these results, the assigned dose is based on

weight and uses the baseline anti-SEA concentration as a co-

variate predicting for probability of cytokine induction/DLT.

The dose escalation between patients uses a Bayesian dose

escalation scheme. The dose for each patient will be determined

so that, on the basis of all available data, the probability of DLT

is 0.1 and the probability that the dose exceeds the MTD is equal

to a, which gradually increases from the current level of 0.25 to

0.50.67

In patients demonstrating relative anti-SEA/drug ratios be-

low 30, the anti-SEA impact on toxicity has been variable. In

vitro lymphocyte proliferation assays are capable of revealing

the level of the inhibitory component in plasma with a given

total anti-SEA concentration. This functional assessment of the

total anti-SEA concentration was predictive in 26 of the 27

patients treated in this study (Fig. 5) and could serve as an

associated predictive factor for toxicity in future clinical trials,

where drug dose is calculated to be in excess of the total

anti-SEA in circulation. The data presented here do not exclude

the possibility that factors other than anti-SEA antibodies mod-

ulate the SEA-induced immune responses observed systemically

and potentially at the tumor site. However, determining the

contribution of the neutralizing anti-SEA component relative to

the total anti-SEA concentration is likely to enable accurate

dosing of this agent.

The clinical toxicity and cytokine induction are reflective

of the systemic intravascular cellular activation resulting from

the SEA-induced cell-cell interaction. However, an individual’s

innate cellular capability must be considered when assessing the

overall effectiveness of such biological response-modifying

agents. Additional factors could affect PNU-2 14565-induced

immune responses at tumor sites. Of the 27 patients entered into

study, 18 patients (67%) demonstrated an adequate lymphocyte

mitogenic stimulatory (PHA) capacity when assayed in FBS

(data not shown). Ten of these 18 patients maintained that

capacity when assayed in autologous plasma. No patient had a

greater mitogenic stimulation in plasma compared with FBS-

containing assay systems. These results suggest that autologous

plasma contains as yet unidentified inhibitory factors that could

modify the proliferation/activation potential at tumor site. When

SEA-induced mitogenic stimulation was examined, it was found

that 16 of 27 patients (59%) had significant SEA-induced lym-

phocyte proliferation in FBS-containing assay systems (data not

shown). Plasma anti-SEA concentrations account for a majority

of the inhibition of SEA-induced proliferation in assays per-

formed in plasma (Table 6). Diminished PHA mitogenic re-




sponses, as revealed in FBS-containing assay systems, could

indicate an innate decreased lymphocyte proliferative capacity.

The level of immunoglobulin and other potential inhibitory

factors at the tumor site milieu is unknown. Factors affecting the

SEA-induced proliferation, in addition to immune suppressive

factors, could include variability in the expression of TCR

V�3-SEA reactive subsets, improper presentation of SEA, MHC

class II polymorphisms, or a lack of accessory molecule engage-

ment. Any of these factors could, in combination with the

inhibitory/blocking anti-SEA, modulate lymphocyte responsive-

ness to SEA (19, 35, 36).

Systemic activation resulting from MHC class Il-depend-

ent SDCC most likely will be dose-limiting for this agent, and

the anti-SEA/fusion protein complex formation could poten-

tially limit the delivery of therapeutic drug to tumor sites.

Preclinical studies are presently in progress with mutant SEA

molecules that demonstrate decreased MHC class II binding

(9, 12, 40, 41). Strategies that minimize SDCC and maximize

staphylococcal enterotoxin antibody-dependent cell-mediated

cytotoxicity could greatly enhance the potential for therapeutic

effect of this potent and novel immunoregulatory treatment

modality.

This trial has led to the development of a novel pharma-

codynamic dosing model encompassing both weight-based dos-

ing and a dosing schema based on the relationship of the

pretreatment anti-SEA concentrations to a dose sufficient to

overcome blocking anti-SEA and provide free drug capable of

inducing a systemic response while minimizing the probability

of DLT. Present trials using this dosing model are in progress.

Refinement of the model will result in individualized Phase II

treatment doses that provide the needed balance between the

plasma anti-SEA inhibitory component and the agent’s SEA

immunostimulatory capability. It is anticipated that by adjusting

the doses administered to account for the anti-SEA concentra-

tion, other factors contributing to toxicity can be identified and

more fully explored.

ACKNOWLEDGMENTS

We thank Cathy Thompson for excellent secretarial assistance. Wewant to acknowledge the care. compassion, and dedication to detail of

the Mary S. Schinagl Clinical Studies Wing nursing unit. We thankMalak Kotb and the laboratory staff at the University of Tennessee for

the HLA-DR and ICR V�3 analyses.

REFERENCES

1. Kotb, M. Bacterial pyrogenic exotoxins as superantigens. Clin. Mi-

cr0. Rev., 8: 41 1-426, 1995.

2. Wynn, T. A., Reynolds, A., James, S., Cheever, A. W., Caspar, P.,Hieny, S., Jankovic, D., Strand, M., and Sher, A. IL-l2 enhancesvaccine-induced immunity to schistosomes by augmenting both humoral

and cell-mediated immune responses against the parasite. J. Immunol.,

157. 4068-4078, 1996.

3. Melder, R. J., Rosenfeld, C. S., Herberman, R. B., and Whiteside,

T. L. Large scale preparation of adherent lymphokine-activated killer(A-LAK) cells for adoptive immunotherapy in man. Cancer Immunol.

Immunother., 29: 67-73, 1989.

4. Rosenberg, S. E.. Lotze, M. T., Muul, L. M., Chang, A. E., Avis,

F. P., Leitman, S., Linehan, W. M., Robertson, C. N., Lee, R. E., Rubin,J. 1., Seipp, C. A., Simpson, C. G., and White, D. E. A progress reporton treatment of 157 patients with advanced cancer using lymphokine-

activated killer cells and interleukin-2 or high-dose interleukin-2 alone.

N. Engl. J. Med., 316: 889-897, 1987.

5. Chatenoud, L., Ferran, C., Legendre, C., Thovard, I., Merite, S.,Reuter, A., Gevaert, Y., Kreis, H., Franchimort, P., and Bach, J. F. In

vivo activation following OKT3 administration. Transplantation (Balti-

more), 49: 697-702, 1990.

6. Weiner, L. M., Clark, J. I., Davey, M., Li, W. S., Garcia de Palazzo,

I., Ring, D. B., and Alpaugh, R. K. Phase I trial of 2B1, a bispecificmonoclonal antibody targeting c-erbB-2 and Fc’yRIII. Cancer Res., 55:

4586-4593, 1995.

7. Bodey, B., Siegel, S. E., and Kaiser, H. E. Human cancer detectionand immunotherapy with conjugated and non-conjugated monoclonal

antibodies. Anticancer Res., 16: 661-674, 1996.

8. Tiedemann, R. E., and Fraser, J. D. Cross-linking of MHC class II

molecules by staphylococcal enterotoxin A is essential for antigen-

presenting cell and T cell activation. J. Immunol., 157: 3958-3966,1996.

9. Abrahms#{233}n, L., Dohlsten, M., Segren, S., Bjdrk, P., Jonsson, E., and

Kalland, T. Characterization of two distinct MHC class II binding sitesin the superantigen staphylococcal enterotoxin A. EMBO J., 14: 2978-2986, 1995.

10. Kozono, H., Parker, D., White, J., Marrack, P., and Kappler, J.Multiple binding sites for bacterial superantigens on soluble class II

MHC molecules. Immunity, 3: 187-196, 1995.

I I. Hudson, K. R., Tiedemann, R. E., Urban, R. G., Lowe, S. C.,Strominger, J. L., and Fraser, J. D. Staphylococcal enterotoxin A has

two cooperative binding sites on major histocompatibility complex class

II. J. Exp. Med., 182: 711-720, 1995.

12. Newton, D. W., Dohlsten, M., Olsson, C., Segr#{233}n,S., Lundin,K. E. A., Lando, P. A., Kalland, T., and Kotb, M. Mutations in the MHCclass II binding domains of staphylococcal enterotoxin A differentiallyaffect T cell receptor V�3 specificity. J. Immunol., 157: 3988-3994,1996.

13. Fleischer, B., Gerardy-Schan, R., Metzroth, B., Carrel, S., Gerlach,

D., and Kohler, W. An evolutionary conserved mechanism of T cellactivation by microbial toxins. Evidence for different affinities of T cell

receptor-toxin interaction. J. Immunol., 146: 11-17, 1991.

14. Irwin, M. J., Hudson, K. R., Fraser, J. D., and Gascoigne, N. R. J.

Enterotoxin residues determining T-cell receptor V�3 binding specificity.Nature (Lond.), 359: 841-843, 1992.

15. Fraser, J. D. High affinity binding sites of staphylococcal entero-

toxins A and B to HLA-DR. Nature (Lond.), 339: 221-224, 1989.

16. Fischer, H., Dohlsten, M., Lindvall, M., Sjogren, H. 0., and

Carlsson, R. Binding of staphylococcal enterotoxin A to HLA-DR on

B cell lines. J. Immunol., 142: 3151-3155, 1989.

17. Dellabona, P., Peccoud, J., Kappler, J., Marrack, P., and Mathis, D.Superantigens interact with MHC class II molecules outside of the

antigen groove. Cell, 62: 1115-1121, 1990.

18. Jardetzky, T. S., Brown, J. H., Gorga, J. C., Stern, L. J., Urban,

R. G., Chi, Y., Stauffacher, C., Strominger, J. L., and Wiley, D. C.Three-dimensional structure of a human class II histocompatibility

molecule complexed with superantigen. Nature (Lond.), 368: 71 1-7 18,1994.

19. Malchiodi, E. L., Eisenstein, E., Fields, B. A., Ohlendorf, D. H.,

Schlievert, P. M., Karjalainen, K., and Mariuzza, R. A. Superantigen

binding to a T cell receptor �3 chain of known three-dimensional struc-

ture. J. Exp. Med., 182: 1833-1845, 1995.

20. Seth, A., Stern, L. J., Ottenhoff, T. H. M., Engel, I., Owen, M. J.,Lamb, J. R., Klausner, R. D., and Wiley, D. C. Binary and ternary

complexes between T-cell receptor, class II MHC and superantigen in

vitro. Nature (Lond.), 369: 324-327, 1994.

21. Dohlsten, M., Lando, P. A., Hedlund, G., Trowsdale, J., andKalland, T. Targeting of human cytotoxic T lymphocytes to MHC classII expressing cells by staphylococcal enterotoxins. Immunology, 71:

96-100, 1990.

22. Herrmann, T., Maryanski, J. L., Romero, P., Fleischer, B., andMacDonald, H. R. Activation of MHC class I restricted CD8+ CTL by




microbial T cell mitogens. Dependence upon MHC class II expressionof the target cells and V�3 usage of the responder T cell. J. Immunol.,144: 1181-1186, 1990.

23. Kalland, T., Hedlund, G., Dohlsten, M., and Lando, P. A. Staphy-lococcal enterotoxin-dependent cell-mediated cytotoxicity. Curr. TopicsMicrobiol. Immunol., 174: 81-92, 1991.

24. Hedlund, G., Dohlsten, M., Petersson, C., and Kalland, T. Superan-

tigen based tumor therapy: in vivo activation of cytotoxic T cells. CancerImmunol. Immunother., 36: 89-93, 1993.

25. Fleischer, B., and Schrezenmeier, H. Stimulation by staphylococcal

enterotoxins. Clonally variable response and requirement for MHC classII molecules on accessory cells. J. Exp. Med., 167: 1697-1708, 1988.

26. Dohlsten, M., Sundstedt, A., Bj#{246}rklund, M., Hedlund, G., and

Kalland, T. Superantigen-induced cytokines suppress growth of humancolon-carcinoma cells. mt. i. Cancer, 54: 482-488, 1993.

27. Dohlsten, M., Abrahms#{233}n,L., Bj#{246}rk,P., Lando, P., Hedlund, G.,Forsberg, G., Brodin, T., Gascoigne, N. R. J., Forberg, C., Lind, P., andKalland, T. Monoclonal antibody-superantigen fusion proteins: tumorspecific agents for T-cell based tumor therapy. Proc. Natl. Acad. Sci.USA, 91: 8945-8949, 1994.

28. Dohlsten, M., Hedlund, G., Akerblom, E., Lando, P. A., and Kalland,T. Monoclonal antibody-targeted superantigens: a different class of anti-tumor agents. Proc. Nail. Acad. Sci. USA, 88: 9287-9291, 1991.

29. Lando, P. A., Dohlsten, M., Hedlund, 0., Akerblom, E., and Kal-

land, T. T cell killing of human colon carcinomas by monoclonal-antibody targeted superantigens. Cancer Immunol. Immunother., 36:

223-228, 1993.

30. Lando, P. A., Hedlund, G., Dohlsten, M., and Kalland, T. Bacterialsuperantigens as anti-tumor agents: induction of tumor cytotoxicity inhuman lymphocytes by staphylococcal enterotoxin A. Cancer Immunol.Immunother., 3: 231-237, 1991.

31. Dohlsten, M., Lando, P. A., Bj#{246}rk,P., Abrahms#{233}n, L., Ohlsson, L.,

Lind, P., and Kalland, T. Immunotherapy of human colon cancer byantibody-targeted superantigens. Cancer Immunol. Immunother., 41:

162-168, 1995.

32. Gidl#{246}f,C., Dohlsten, M., Kalland, T., and T#{246}tterman,T. H. Antibodiesare capable of directing superantigen-mediated T cell killing of chronic Blymphocytic leukemia cells. Leukemia (Baltimore), 9: 1534-1542, 1995.

33. Giantonio, B. J., Alpaugh, R. K., Schultz, J., McAleer, C., Newton,

D. W., Shannon, B., Guedez, Y., Koth, M., Vitek, L., Persson, R.,Gunnarsson, P-O., Kalland, T., Dohlsten, M., Persson, B., and Weiner,

L. M. Superantigen-based immunotherapy: a Phase I trial of PNU-214565, a monoclonal antibody-staphylococcal enterotoxin A recombi-

nant fusion protein, in advanced pancreatic and colorectal cancer.

J. Clin. Oncol., 15: 1994-2007, 1997.

34. Miller, A. B., Hoogstraten, B., Staquet, M., and Winkler, A. Re-

porting results of cancer treatment. Cancer (Phila.). 47: 207-214, 1981.

35. Wen, R., Blackman, M. A., and Woodland, D. L. Variable influenceof MHC polymorphism on the recognition of bacterial superantigens byT cells. J. Immunol., 155: 1884-1892, 1995.

36. Ohnishi, H., Ledbetter, J. A., Kanner, S. B., Linsley, P. 5., Tanaka,T., Geller, A. M., and Kotb, M. CD28 cross-linking augments TCR-

mediated signals and costimulates superantigen responses. J. Immunol.,

154: 3180-3193, 1995.

37. Herman, A., Labrecque, N., Thibodeau, J., Marrack, P., Kappler,J. W., and Sekaly, R. P Identification of the staphylococcal enterotoxinA superantigen binding site in the beta I domain of the human histo-compatibility antigen HLA-DR. Proc. NatI. Acad. Sci. USA, 88: 9954-

9958, 1991.

38. Karp, D. R., and Long, E. 0. Identification of HLA-DR1 beta chain

residues critical for binding staphylococcal enterotoxins A and E. J.

Exp. Med., 175: 415-424, 1992.

39. Kalland, T., Dohlsten, M., Lando, P. A., Abrahms#{233}n,L., Bj#{246}rk,P.,Hedlund, G., Ohlsson, L., Brodin, T., and Hansson, J. Antibody targetedsuperantigens in experimental tumor therapy. In: J. Thibodeau and R. P.Sekaly (eds.), Bacterial Superantigens: Structure, Function and Thera-peutic Potential, pp. 219-238. Austin, TX: Landes Co., 1995.

40. Holzer, U., Orlikowsky, T., Zehrer, C., Bethge, W., Dohlsten, M.,Kalland, T., Niethammer, D., and Dannecker, G. E. T-cell stimulationand cytokine release induced by staphylococcal enterotoxin A (SEA)and the SEAD227A mutant. Immunology, 90: 74-80, 1997.

41. Lando, P. A., Olsson, C., Kalland, T., Newton, D., Kotb, M., and

Dohlsten, M. Regulation and superantigen-induced T cell activation in

the absence and presence of MHC class II. J. Immunol., 157: 2857-2863, 1996.



1998;4:1903-1914. Clin Cancer Res R K Alpaugh, J Schultz, C McAleer, et al. gastrointestinal malignancies.trial of the fusion protein PNU-214565 in patients with advanced Superantigen-targeted therapy: phase I escalating repeat dose

Updated version

http://clincancerres.aacrjournals.org/content/4/8/1903

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/4/8/1903To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



superantigen-targeted therapy: phase i escalating repeat ...antibody-directed, superantigen-induced...

Documents