phase i evaluation of intravenous recombinant human...

Vol. 3, 409-417, March 1997 Clinical Cancer Research 409

Phase I Evaluation of Intravenous Recombinant Human Interleukin

12 in Patients with Advanced Malignancies1

Michael B. Atkins,2 Michael J Robertson,

Michael Gordon, Michael T. Lotze,

Michelle DeCoste, Jon S. DuBois, Jerome Ritz,

Alan B. Sandler, Howard D. Edington,

Pamela D. Garzone, James W. Mier,

Christine M. Canning, Linda Battiato,

Hideaki Tahara, and Matthew L. Sherman

Tupper Research Institute and the Division of Hematology/Oncology,Tufts-New England Medical Center, Boston, Massachusetts 021 1 1[M. B. A., J. S. D., J. W. M.]; Division of Hematologic Malignancies,Dana-Farber Cancer Institute, Boston, Massachusetts 021 15 [M. J. R.,J. R., C. M. C.]; Division of Hematology/Oncology, IndianaUniversity School of Medicine, Indianapolis, Indiana 46202 [M. 0.,A. B. S., L. B.]; Department of Surgical Oncology, Pittsburgh CancerInstitute, Pittsburgh, Pennsylvania 15261 [M. T. L., H. D. E., H. I.];

and Genetics Institute, Inc., Cambridge, Massachusetts 02140 [M. D.,P. D. 0., M. L. S.]

ABSTRACT

A Phase I dose escalation trial of i.v. administered recom-

binant human interleuldn 12 (rhIL-12) was performed to de-termine its toxicity, maximum tolerated dose (MTD), pharma-

cokinetics, and biological and potential antineoplastic effects.

Cohorts of four to six patients with advanced cancer, Karnof-

sky performance �70%, and normal organ function received

escalating doses (3-1009 ng/kg/day) of rhIL-12 (Genetics Insti.

tute, Inc.) by bolus i.v. injection once as an inpatient and then,

after a 2-week rest period, once daily for five days every 3weeks as an outpatient Therapy was withheld for grade 3toxicity (grade 4 hyperbilirubinemia or neutropenia), and dose

escalation was halted if three of six patients experienced adose-limiting toxicity (DLI). After establishment of the MTD,eight more patients were enrolled to further assess the safety,

pharmacokinefics, and immunobiology of this dose.Forty patients were enrolled, including 20 with renal can-

cer, 12 with melanoma, and 5 with colon cancer; 25 patients

had received prior systemic therapy. Common toxicities in-

cluded fever/chills, fatigue, nausea, vomiting, and headache.

Fever was first observed at the 3 ng/kg dose level, typically

Received 8/20/96; revised 1 1/21/96; accepted 12/2/96.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to

indicate this fact.I This work was supported in part by Grants MOl-RR-00054 and MOl-RR7SO from the Division of Research Resources, NIH, and a stipendfrom Genetics Institute, Inc.2 To whom requests for reprints should be addressed, at Division ofAdult Hematology/Oncology, Department of Medicine, New EnglandMedical Center, 750 Washington Street, Box 542, Boston, MA 021 1 1.Phone: (617) 636-6527; Fax: (617) 636-0342.

occurred 8-12 h after rhIL-12 administration, and was incom-pletely suppressed with nonsteroidal anti-inflammatory drugs.

Routine laboratory changes included anemia, neutropenia,

lymphopenia, hyperglycemia, thrombocytopenia, and hy-poalbuminemiL DLTs included oral stomatitis and liver func-lion test abnormalities, predominantly elevated transaminases,

which occurred in three offour patients at the 1000 ng/kg dose

leveL The 500 ngflg dose level was determined to be the MTh.This dose, administered by this schedule, was associated withasymptomatic hepatic function test abnormalities in three pa-tients and an onstudy death due to Closb*Iia perfringens sep-

ticemia but was otherwise well tolerated by the 14 patients

treated in the dose escalation and safety phases. The T,,� elim-

ination of rhIL-12 was CalCUlated to be 5.3-9.6 h. Biologicaleffects included dose-dependent increases in circulating IFN-�y,

which exhibited attenuation with subsequent cycles. Serum

neopterin rose in a reproducible fashion regardless of dose orcycle. Tumor necrosis factor a was not detected by ELISA.

One of 40 patients developed a low titer antibody to rhIL-12.

Lymphopenia was observed at all dose levels, with recovery

occurring within several days of completing treatment without

rebound lymphocytosis. There was one partial response (renalcell cancer) and one transient complete response (melanoma),both in previously untreated patients. Four additional patients

received all proposed treatment without disease progression.rhIL-12 administered according to this schedule is biologically

and clinically active at doses tolerable by most patients in anoutpatient setting. Nonetheless, additional Phase I studies ex-aniining different schedules and the mechanisms of the specificDLTs are indicated before proceeding to Phase II testing.

INTRODUCTION

IL-l23 is a heterodirnenc cytokine with potent immunoregu-

latory activity. It was initially given the names natural killer cell

stimulatory factor and cytotoxic lymphocyte maturation factor (1,

2) based on its stimulatory effects on these cytolytic lymphokine

populations. The cDNAs encoding the two IL-l2 subunits are

unrelated and encode for two distinct proteins having molecular

masses of approximately 35 and 40 kDa (3, 4). The p35 subunit is

distantly related to IL-6, granulocyte colony-stimulating factor, and

chicken myelomonocytic growth factor, and the p40 subunit is

related to the extracellular domain of the IL-6 receptor, as well as

to the ciliaiy neurotrophic growth factor receptor (5, 6).

In vitro studies have shown that monocytes appear to be the

major source of IL-l2 in activated PBMC suspensions (7). Al-

though EBV-transforrned B cells appear to produce IL-l2, signif-

3 The abbreviations used are: IL, interleukin; PBMC, peripheral bloodmononuclear cell; NK, natural killer; TNF, tumor necrosis factor; AST,aspartate aminotransferase; DLT, dose-limiting toxicity; MID, maxi-

mum tolerated dose; ANC, absolute neutrophil count.

Research. on September 7, 2018. © 1997 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

410 Phase I Trial of iv. rhIL-12 in Advanced Malignancies

icant production by T cells or NK cells has not been demonstrated.

Two low-affinity rhIL-l2 receptor subunits have been cloned,

which together constitute a high-affinity IL-12 receptor (8). High-

affinity IL-12 receptors have been identified on phytohernaggluti-

nm-activated T cells and IL-2-activated NK cells (9, 10). In addi-

tion, specific binding sites (possibly low-affinity) have recently

been detected on the Bl subset of B lymphocytes (1 1).

IL- I 2 has been shown to regulate a number of activities

central to both cellular and humoral immune responses. For

example, in vitro studies have demonstrated that IL- 1 2 can

enhance the non-MHC-restricted cytolytic activity of NK cells

( I 2, 1 3) and facilitate specific allogeneic human cytolytic T

lymphocyte responses against allogeneic irradiated melanoma

cells (14). IL-12 has been shown to be mitogenic, allowing the

continued proliferation of both activated T lymphocytes and NK

cells (15-17). IL-12 has also been shown to be a potent inducer

of IFN-�y release from both T lymphocytes and NK cells (1, 18)

while inducing much lower amounts of inflammatory cytokines,

such as TNF-a, especially when compared with IL-2 (13, 19).

Furthermore, IL-l2 promotes T helper 1 cell development while

inhibiting T helper 2 cell development and characteristic cyto-

kine expression (20-22), favoring the development of cellular

immunity and immunological memory over hurnoral immunity.

In this regard, exposure to IL-12 enhanced survival in murine

models of toxoplasmosis and leishmaniasis, two intracellular

infections that require host T helper 1 cell responses (23-25).

IL- 12 has also shown potent antitumor activity in a number

of in vivo murine tumor models, including the Bl6FlO mela-

norna, RENCA renal cell adenocarcinoma, and M5076 reticu-

lum cell sarcoma (26, 27). i.p. administration of IL-l2 sup-

pressed the growth of both pulmonary and s.c. BI6F1O

melanoma deposits and induced regression of RENCA. In the

RENCA model, long-term survival and resistance to rechallenge

were observed in mice injected peniturnorally with IL-l2. An-

titumor effects were demonstrated both in microscopic disease

models and in animals bearing large established tumors in which

IL-l2 was initiated 7, 14, or even 28 days after tumor inocula-

tion (26, 27). The antiturnor activity observed in these model

systems was more potent than that observed with IL-2 (26).

Antitumor activity was slightly reduced in cytotoxic granule-

deficient beige mice, but was nearly eliminated in T cell-defi-

cient nude mice (26). Depletion of the CD8+ (but not CD4+)

T-cell subset by systemic administration of monoclonal antibod-

ies diminished the efficacy of IL- 1 2 in the RENCA system (26),

whereas depletion of either CD8+ or CD4+ T cells reduced

antitumor activity in other models (27). In addition, most studies

have shown that administration of antibodies to IFN-’-y signifi-

cantly reduces IL-12 antitumor activity (4). These experiments

suggest that the antitumor effect of IL-l2 is largely mediated by

CD4+ and CD8+ T cells through the release of IFN--y. How-

ever, the antitumor effect of IL- I 2 in most tumor systems is

superior to that of IFN-�y, and furthermore, efficacy is absent in

nude mice despite dramatically enhanced IFN-’y production

(26), suggesting that other factors besides IFN-� must contribute

to the antitumor effects of IL-l2. These preclinical data suggest

that IL- 1 2 may have potent irnmunoregulatory and antiturnor

activities that might be useful in the treatment of human infec-

tions and malignancies and encouraged us to proceed with this

Phase I clinical trial of rhIL-12 in cancer patients.

PATIENTS AND METHODS

Patient Selection. All patients were adults with histologi-

cally proven advanced malignancy that was measurable, clearly

progressive, and refractory to conventional therapy. Patients were

required to have a Kamofsky performance status of �70% and

adequate organ function defined by white blood count > 4000/p.L,

hemoglobin > 10 g/dl, platelet count > 100,000/pA, creatinine <2

mg/dl, creatinine clearance > 60 mI/mm, bilirubin < 2.0 mg/dl,

AST < 2 times the upper limit of normal, electrocardiogram and

chest X-ray without clinically significant nonmalignant abnormal-

ities, and an FEV1 > 2.0 liters or 75% of predicted. Patients with

history of peripheral neuropathy, brain metastases, or other central

nervous system disease, prior exposure to nephrotoxic or neuro-

toxic chemotherapy agents, more than two prior systemic therapies,

history of autoimmune disease, concurrent requirement for corti-

costeroids, seropositivity for human immunodeficiency virus or

hepatitis BSAg, serious concurrent infection or other major active

illness, or who had received other chemotherapy, biological ther-

apy, or any investigational drug within the preceding 3 weeks were

ineligible.

Study Design. The study was an open-label, nonrandom-

ized, multicenten Phase 1 dose escalation trial. The treatment

protocol was approved by the Human Investigation Review

Board at the participating institutions, and written informed

consent was obtained from each patient. rhIL- 12 was supplied

by Genetics Institute, Inc. (Cambridge, MA) and administered

by rapid iv. injection into a rapidly flowing iv. line.

The treatment schedule is displayed in Fig. I. A test dose

was administered on day 1 , and patients were observed over-

night as inpatients and were seen in follow-up on days 5 and 8.

Patients who tolerated the test dose without DLT received

rhIL-l2 by rapid iv. injection daily for 5 days beginning on day

15 in an outpatient setting. Treatment cycles were repeated

every 21 days. Vital signs, including supine blood pressure,

pulse, respiratory rate, and temperature, were measured at fre-

quent regular intervals for up to 6 h after rhIL-l2 administration

on each day of the first 5-day course and for I h after IL-l2

administration on all subsequent cycles. Weight was measured

before each rhIL-l2 administration, and rhIL-l2 dosages were

recalculated before each treatment course. Patients were evalu-

ated for tumor response at the end of every second 21-day

treatment cycle, and patients with stable or regressing disease

could receive a maximum of six S-day treatment courses.

rhJL-l 2 doses were increased from 3 to 1000 ng/kg in suc-

cessive cohorts of patients. There was no intrapatient dose escala-

tion. A minimum of four patients were enrolled at each dose level,

and all patients had to have completed day 35 (end of first 21-day

cycle) before initiating enrollment to the next dose level. If any

patient developed grade 3 or greater toxicity at a particular dose

level, then an additional two patients were enrolled at that dose

level. Dose escalation was halted when three patients at a particular

dose level experienced a grade 3 or greater toxicity and the pre-

ceding dose level was designated the MTD.

Patients experiencing a fever after the test dose could

receive prophylactic acetaminophen (up to 2600 mg/day) for

subsequent cycles. If fever recurred then indomethacin (25 rng,

p.o., every 8 h) was provided. No other concomitant rnedica-

tions were routinely administered.



test dose

rhIL.12

cycle I cycle 2

++t++TumorMeasurements

Table 1 Patient characteristics

Day I 15 19 36 40 50

. Dose levels: 3, 10, 30, 100, 250, 500, 1000 ng/kg

. 4 6 patients per dose cohort

. Maximum: test dose and six 21 day cycles

Fig. 1 Schema for this study.

No. of patients

40

52 years (25-76)24/16

13

21

6

20

12

5

2514

12

38

2

Clinical Cancer Research 411

Toxicity was assessed using the National Cancer Institute

Common Toxicity Criteria with slight modifications to address

issues of fever, neutropenia, lymphopenia, elevated bilirubin, and

capillary leak syndrome. Grade 3 elevations in bilirubin or de-

creases in neutrophil count and grade 4 lyrnphopenia were not

considered DLTs. Grade 3 capillary leak syndrome was defined as

fluid retention associated with hypotension, or symptomatic or

laboratory evidence of organ dysfunction. Patients experiencing

grade 3 or grade 4 toxicity were taken off the study, except for

those patients treated at the highest dose level administered (be-

yond the MTD), who were allowed to continue therapy at the

preceding dose level. Otherwise, no intrapatient dose modifications

were permined.

Pharmacokinetics. Serial blood specimens were col-lected for determination of circulating rhIL-l2 levels before the

administration of rhIL-l2 on day 1 and 10, 20, and 30 mm, and

1, 2, 4, 6, 10, 12, and 24 h post-drug administration. Additional

blood specimens were also collected 10 mm after rhIL-l2 ad-

ministration in subsequent treatment cycles. Serum was isolated

and analyzed by an ELISA (28) with a lower limit of detection

of 40 pg!ml.

Immunobiology. Serum for neopterin, IFN--y, and TNF-a

determinations was obtained before each rhIL-12 administration, at

all interim visits, and at the final study visit. Sera for quantitative

immunoglobulin measurements and PBMC for FACS analysis

were obtained before the test dose and immediately before each

rhIL-12 5-day treatment course, and at the final visit. PBMCs were

shipped overnight to Nichols Laboratory (San Juan Capistrano,

CA) and analyzed for the CD3, CD4, CD8, CD14, CD16, CD2O,and CD56 expression using standard techniques.

rhIL-12 Antibodies. Serum for anti-rhIL-12 antibody

determinations was obtained pretreatment, at the beginning of

each treatment cycle and at the final visit.

Assessment of Tumor Response. Tumor measurements

were obtained after the second 21-day cycle and then every

other cycle. Responses were determined according to previously

developed and published criteria (29).

RESULTS

Patient Population. Thirty-two patients participated in

the dose escalation phase of this study at seven different dose

levels. After determination of the MTD, an additional eight

patients were enrolled at that level to further assess its safety.

Table 1 depicts the relevant demographic and previous treat-

ment information for all 40 study patients. As can be seen, the

majority of patients had either renal cell cancer or melanoma

and had received prior therapy. In particular, 9 of the 20 patients

Total patientsMedian age (range)Gender (Male/Female)Performance status

(Karnofsky)

100%90%80%

Tumor typesRenal cell cancerMelanoma

Colon cancerParotid cancer

Cervical cancerAdenoid cystic cancer

Prior therapyChemo/immunotherapy(IL-2)

RadiotherapySurgeryNone

with renal cell carcinoma and S of the 12 patients with meta-

static melanoma had received prior IL-2 therapy.

Treatment. Three patients received only the test dose of

rhIL-l2 due to either the clinical appearance of previously unsus-

pected brain metastases (two patients with melanoma treated at the

250 and 500 ng/kg dose levels) or transient grade 3 hepatic trans-

aminase elevations (one patient at the 500 ng/kg dose level).

Six additional patients were not treated beyond the first

5-day treatment course due to disease progression (3 patients),

asymptornatic premature ventricular contractions ( 1 patient),

and transient liver function test abnormalities (2 patients). Five

of the remaining 3 1 patients received the test dose and all six of

the intended 21-day treatment cycles.

Clinical and Laboratory Toxicity. Common side effects

associated with rhIL-l2 included fever, chills, fatigue, headache,

and, less commonly, nausea and vomiting. The fever was first

observed at the 3 ng/kg dose level and occurred in the majority of

patients receiving � 10 ng/kg rhIL-l2. Fever was delayed at onset,

typically occurring 8-12 h after the rhIL-12 injection, and was

incompletely suppressed with nonsteroidal anti-inflammatory

drugs. Significant laboratory toxicities according to dose level are

displayed in Table 2. Frequent laboratory changes included anemia,

neutropenia, lymphocytopenia, hyperglycemia, thrombocytopenia,

and hypoalbuminemia. Most of these abnormalities appeared mdc-

pendent of dose, peaked at day 5 of a treatment cycle, and resolved

quickly. Grade 4 lymphocytopenia occurred in the majority of

patients at all dose levels and resolved without rebound lympho-

cytosis. No consistent changes in serum creatinine or neurological

toxicity were observed.

Dose-dependent abnormalities included stomatitis, which

first became apparent during the sixth 5-day treatment cycle in

one patient at the 100 ng/kg dose level and was evident in three

of four patients at the 1000 ng/kg dose level, and liver function

test abnormalities, which first became evident in one patient at

the 250 nglkg dose level and were universal by the 1000 ng/kg

dose level. Liver function test abnormalities consisted predom-

inantly of elevations in hepatic transaminases and lactate dehy-



412 Phase I Trial of iv. rhIL-l2 in Advanced Malignancies

Table 2 Number of pati ents experiencin g selected Ia boratory toxicity in all cycles (grad es 2, 3)

Dose level (ngfkg)(n)

3 10 30 100 250 500 1000Toxicity (4) (4) (4) (4) (6) (14) (4)

Hepatic

Alkaline phosphatase ( 1 , 0) ( 1, 0) (7, 2) (1, 1)AST (1, 0) (0, 1) (2, 2) (6, 3) (2, 2)

Bilirubin (1, 1) (1, 0) (1, 6) (0, 1)

Hyperglycemia (1, 0) (1, 0) (2, 0) (2, 0) (2, 1) (3, 1) (1, 0)Creatinine (2, 0)Hematologic

ANC (1, 0) (0, 3) (3, 1) (3, 5) (3, 1)

Hemoglobin (3, 0) (1, 0) (2, 0) (0, 1) (2, 0) (3, 1) (1, 1)

Platelets (1, 0) (1, 0)

drogenase, but alkaline phosphatase and bilirubin were occa-

sionally elevated as well. Stomatitis usually resolved over 3

days, whereas liver function test abnormalities resolved over

7-10 days. Mucosal biopsies were not performed, but herpes

simplex virus cultures from the oral lesions were negative on

multiple occasions in several patients.

Determination of MTD. During the dose escalation

phase, one patient each at the 250 and 500 ng/kg dose levels had

DLT (grade 3 nausea and vomiting secondary to previously

undiagnosed brain metastases and grade 3 elevations of AST,

respectively). Consequently, two additional patients were added

to each dose level and tolerated therapy without difficulty.

Three of four patients treated at the 1000 ng/kg dose level

experienced DLT. One patient developed grade 3 elevation of

hepatic transaminases, grade 3 stomatitis, and a mild capillary

leak syndrome after three doses. Another patient developed

grade 3 elevation of hepatic transaminases and grade 2-3 sto-

matitis after four doses, and the third patient developed grade 3

elevation of alkaline phosphatase after completing all five doses.

The fourth patient at this dose level experienced both grade 2

stomatitis and grade 2 elevations in hepatic transaminases. On

the basis of this experience, the DLTs were defined as liver

function test abnormalities and stomatitis, and the MTD for this

schedule was determined to be 500 nglkg.

Safety Experience. After determination of the MTD, an

additional eight patients were treated at this dose level to verify

its safety. The results of all 14 patients treated at the 500 ngfkg

dose level are displayed in Table 3. One patient went off study

after only receiving the test dose for early disease progression

and was considered inevaluable for toxicity. Nine patients re-

ceived all of the first 5-day treatment course without significant

toxicity. Two of these nine patients experienced potential DLT

during later cycles: an upper gastrointestinal bleed in one patient

after three cycles and an admission for dehydration and ortho-

stasis in another patient after the second 5-day treatment course.

Both of these toxicities were believed not to be directly related

to rhIL-12. Three of the remaining four patients had doses held

for hepatotoxicity: grade 3 elevations in transaminases or alka-

line phosphatase that were largely transient and asymptomatic.

One of these patients was able to continue to receive therapy

without additional grade 3 hepatic toxicity. The final patient

developed grade 3 elevation of alkaline phosphatase after four

doses in the first 5-day treatment course and had his treatment

stopped. Subsequently, he developed abdominal pain and ane-

mia and was admitted for evaluation of potential gastrointestinal

bleeding. No mucositis was documented. Within 24 h he devel-

oped fever, followed by rapid onset of irreversible shock and

death. Blood cultures grew Clostridia perfringens from an un-

known source. No autopsy was performed. The death was felt to

be secondary to sepsis, presumably from necrotic intra-abdom-

inal tumor, and was classified as “possibly related to rhIL-l2.”

Toxicity was fairly consistent from one cycle to the next

within individual patients without evidence of either cumulative

effect or attenuation corresponding to the changes observed in

some biological parameters (see below). Fig. 2, A and B, dis-

plays hepatic transaminase levels and neutrophil counts follow-

ing the test dose and subsequent cycles for four patients who

received at least two 5-day courses of therapy at the 500 ng/kg

dose level. AST levels were maximal at day 5 of the first 5-day

treatment course, with possible attenuation with subsequent

cycles. ANC nadirs occurred around day 5 of each cycle without

apparent cumulative toxicity. ANC recovery occurred by day 15

of each cycle without rebound neutrophilia.

Tumor Response. Objective tumor responses were ob-

served in two patients. One patient with metastatic renal cell

carcinoma with small pulmonary nodules and hilar adenopathy

experienced a partial response that is still ongoing at 22+ months.

The second patient, a 57-year-old woman with melanoma, experi-

enced a complete regression of small pleural-based nodules and

cervical adenopathy lasting approximately 4 weeks. Four additional

patients completed six 21-day treatment cycles with stable disease.

Of the 14 patients treated at the MTD (6 renal, 4 melanoma; 3

without previous therapy), no responses were seen.

Biological Parameters. Mean circulating IFN-�y concen-

trations for the top three cohorts are displayed in Fig. 3. IFN--y

appeared to be induced in a dose-dependent manner with peak

concentrations exceeding 1000 pg/mi observed 24 h after a test

dose of 1000 nglkg rhIL-12, with return to baseline by day 8.

During the 5-day treatment courses, IFN--y concentrations were

highest at 48-72 h after the first rhIL-12 injection and returned

to baseline by 72 h after the last dose. Peak IFN-�y concentra-

tions in general were higher during the first 5-day treatment

course than after the test dose. Peak IFN--y concentrations dur-

ing subsequent cycles tended to progressively decline.

Mean serum neopterin concentrations are displayed in Fig.

4. Neopterin tended to rise in a non-dose-dependent manner,




Table 3 Results of patients treated at the 500 ng/kg dose level ofrhIL-12

Patientno.

No. of cyclescompleted Results

Dose escalation 023 2 No DLTphase

Safety phase

024025

026027

028

033

034035

036

037

038

039

040

22

Test doseTest dose

2

1”

2�2”

3

4’

2

2

1”

No DLTNo DLT

Grade 3 AST, discontinuedBrain metastases,

discontinued

No DLI

Grade 3 alkalinephosphatase.sepsis/deathe

No DLIGrade 3 orthostasis,1

discontinued

Gastrointestinal bleed!discontinued

Grade 3 liver function test,dose held in cycle 2,continued on therapywithout subsequent DLT

No DLI

No DLT

Grade 3 alkalinephosphatase,discontinued

a Received four doses in cycle 1.

/) Received two doses in cycle 2.C Received four doses in cycle I.d Received three doses in cycle 1.e Possibly related to rhIL-l 2.

1Not directly related to rhIL-12.

reaching a maximum at day 5 and declining slowly thereafter. In

contrast to IFN--y, no attenuation in peak concentrations was

observed with successive rhIL-12 cycles. TNF-a was not de-

tected in any of the plasma specimens at any dose level.

Immunophenotyping. Mean lymphocyte and monocyte

counts fell promptly during therapy and recovered quickly after

treatment without significant rebound or dose dependence. Phe-

notypic analyses showed reduction in all lymphocyte subsets

(data not shown).

IL-12 Antibodies. One patient treated at 500 ng/kg de-

veloped low titer antibodies to rhIL-12 after 2 weeks of therapy.

No other patient developed antibodies to rhIL-l2.

Pharmacokinetics. rhIL- 12 concentrations in patients

receiving either 3 or 10 ng/kg were below the limit of assay

quantitation. The pharmacokinetic parameters for all other pa-

tients are shown in Table 4. Both maximum concentration

(Cmax) and area-under-concentration versus time curve extrap-

olated to infinity (AUC0...,�) were proportional to the dose

administered, indicating that over the dose range of 30-1000

ng/kg, the pharmacokinetic characteristics of rhIL-12 were un-

ear and not dependent on dose. The elimination half-life was

calculated to be between 5.3 and 10.3 h. Fig. 5 displays the

concentration versus time profiles of patients receiving 500

ng/kg, the MTh in this study. All patient profiles fell well

within the range around the mean expected for this limited

number of samples, and there was no correlation observed

between rhIL- 1 2 elimination half-life and toxicity.

DISCUSSION

IL-12 has potent immunomodulatory and antitumor effects

in preclinical studies. This report describes the results ofthe first

clinical trial of this promising cytokine. In this study of rhIL- I 2

administered by bolus iv. injection initially as a single test dose,

then daily for 5 days every 3 weeks, we defined the MTD for

rhIL-12 to be 500 ng/kg. DLT, transient reversible stomatitis

and/or hepatic dysfunction, occurred in three of four patients

treated at the 1000 ng/kg dose level. One patient treated at the

500 ng/kg dose level died of C. perfringens sepsis. It is possible

that some effect of rhIL-12 on intestinal mucosa or bacterial

surveillance may have contributed to this event; however, be-

cause mucositis was not a prominent toxicity at this dose level,

and because this dose and schedule were well tolerated by most

other patients in an outpatient setting, we felt this dose did not

exceed the MTD. Other prominent clinical abnormalities in-

cluded fever and chills, fatigue, and, less commonly, nausea,

vomiting, and headache. Commonly observed laboratory abnor-

malities included anemia, neutropenia, lymphopenia, thrombo-

cytopenia, hyperglycemia, and hypoalbuminemia. The fever

was unusually delayed, typically occurring 8-12 h after an in-

jection rather than at 2-4 h as is typically seen with other py-

rogenic cytokines such as IL-i, IL-2, IL-4, IL-6, and the inter-

ferons (30-34). The half-life of rhIL-l2 was also unusually

prolonged relative to these other cytokines (5-10 h versus 20

mm) perhaps explaining the delay in onset of the fever.

Similar to preclinical observations, rhIL- 12 was found to

be a potent inducer of IFN--1’. IFN--y was observed to be induced

in a dose-dependent manner, but to a progressively lesser degree

with each successive 5-day treatment course. Many of the side

effects observed with rhIL-1 2 administration, in particular the

liver function test abnormalities, fever, flu-like symptoms, fa-

tigue, hernatological changes, and hyperglycemia, could be di-

rectly or indirectly attributable to IFN-�y production because

they have been observed in clinical trials with this cytokine (35).

The etiology of the stomatitis is less clear. It was a sur-

prising DLT because it had not been observed in preclinical

toxicology studies. Potential mechanisms include: decrease in

mucosal IgA immunity, an autoirnmune phenomenon, viral ac-

tivation, or inhibition of constitutive mucosal epithelial prolif-

eration. Although in the absence of histological confirmation it

is difficult to distinguish between these possibilities, no unusual

bacterial, fungal, or viral organisms were obtained from any

cultures, and no other evidence of autoimmune phenomena was

observed, making the first three possibilities unlikely. IL-12 has

been observed to enhance the mucosal toxicity associated with

radiation therapy, presumably through an IFN--y-mediated

mechanism (36), and IFN-�y has been shown to inhibit TGF-�3

production (37), a factor purported to be important in maintain-

ing mucosal integrity, perhaps implicating IFN-’-y in this toxicity

as well. However, IFN-y did not cause mucositis in Phase I

trials (35), arguing against it having a sole causative role in the

stomatitis associated with rhIL-l2. Perhaps the more sustained

release of IFN-y in response to prolonged presence of IL- I 2 or

the concomitant effect of other factors may have enabled this

new toxic effect to become manifest. This side effect limited

further escalation of rhIL-12 on this schedule. Whether a better

understanding of the underlying mechanism for this toxicity or




x xxxxx

AST (SGOT)

xxxx x

A

-I

I-0C,Cl)

I-Cl)4

xxxxx400�

350�

300�

250�

200�

150�

100�

50#{149}

B

0 10 20 30 40 50 60 70 80

DAY

xxxxx

ANC

xxxxx xxxxx

10 20 30 40 50 60 70 80

13

12

11

10

DAY

Fig. 2 Serial AST (SGOT; A) and neutrophil (ANC; B) levels for four individual patients over the first 70 days of therapy at the 500 ng/kg dose level.x,rhIL-l2 administration (500 ng/kg).

aggressive preventative treatment will enable higher doses of

rhIL-l2 to be safely administered remains to be determined.

Other cytokines have been shown to be potent inducers of

TNF-a (19), but despite the ability of IL-12 to induce some TNF

in vitro and the appearance of some side effects potentially

attributable to TNF, no TNF was detectable in the plasma of

patients receiving rhIL- 12. Therefore, in contrast to mechanisms

for toxicity proposed for other cytokines, toxicities observed in

response to rhIL-l2 administration are unlikely to be related to

secondary TNF production.

IL- 12 was shown in preclinical studies to allow the con-

tinued proliferation of preactivated NK and T cells, to enhance



MEAN INTERFERON GAMMA

�

L

17000 x

16000

15000

14000

�. 13000

E.�l 12000a.

.� 11000

�� 9000

C, 8000

z 7000 �\� 6000 �wU- 5000 :� � 1%.� 4000

z 3000 .

- � �0 � 40 5� 60 70 80 90 100

DAY

Fig. 3 Serial mean serum IFN--y levels for patients treated at the top three rhIL-l2 dose levels. �, 250 ng/kg; #{149}, 500 ng/kg; 0, 1000 ngfkg. x,

rhIL-12 administration.

MEAN NEOPTERIN

300

250

-J� 2000EC

z 150

wI-0.

0wz

50

0

100

DAY

Fig. 4 Serial mean serum neopterin levels for patients treated at the top three rhIL- 12 dose levels. E, 250 ng/kg; #{149}, 500 ng/kg; 0, 1000 ng/kg. x,rhIL-l2 administration.

x�0tx �

0 10 20 30 40 50 60 70 80 90


cytotoxic activity of T cells and NK cells, and to induce T helper

1 cell and inhibit T helper 2 cell differentiation (20-22). Special

studies to determine potential effects of rhIL-l2 on TH cell

distribution were not performed and, therefore, must await fu-

ture rhIL-12 studies. PBMC phenotypic analyses showed reduc-

tion in all lymphocyte subsets.

Tumor responses were observed in one patient each with

melanoma and renal cell carcinoma treated near the MTD.

Because this study involved few patients per dose level and

many of these patients had previously received other immuno-

therapy agents (most commonly IL-2), it is impossible to deter-

mine the cancer immunotherapeutic potential of rhIL-12 or the

most appropriate dose for Phase II testing. However, based on

this study, we feel that Phase II testing eventually will be



Parameter

c,,,a� (pg/mI)

30 100

744.3 ± 313.1 2877.8 ± 904.7 6640.0 ± 1385.3 1 1 136.8 ± 3529.8 19575.0 ± 5269.0

v� (mI/kg) 48.4 ± 29.7 43.4 ± 13.3 47.8 ± 23.2 56.7 ± 8.6 62.5 ± 14.3

T112�.1,�, (hours) 5.3 ± 0.69 5.6 ± 1.1 7.2 ± 2.5 9.6 ± 2.6 9.1 ± 2.1

CL (mI/hr/kg) 6.8 ± 5.3 5.6 ± 2.0 5.8 ± 5.5 4.4 ± 1.5 5.2 ± 2.5

AUC��. (pg X hr/mI) 61 19.2 ± 3232.9 19821.0 ± 7162.9 64276.5 ± 30675.8 123864.4 � 37559.8 220096.9± 75237.5

250 500 l00()

20


Table 4 Mean pharmacokinetic parameter estimates in patients receiving rhIL-12 as single iv. dose on day I

All values are mean ± SD. Cm�x, maximum concentration; V�, the volume of distribution at steady state: T112�11,,, the terminal or elimination

half-life: CL, clearance: and AUC0_,�, area-under-concentration versus time curve extrapolated to infinity.

Dose (ng/kg)

.�1

E0)

0.

c.,J

0C0

(0

C

C0

C)

100000

10000

I 000

I 00

0 5 10 15

Time After Administration (hrs)

Fig. 5 Concentration versus time profiles of all patients receiving 500ng/kg IL-l2 as a single iv. dose on day I. . individual patients:-, mean of all patients.

warranted, and if this schedule is to be used, that doses in the

range of 500 ng/kg should be tolerated by most patients.

Of note, a Phase II trial of rhIL-l2 in metastatic renal cell

carcinoma using this schedule (without the test dose) at the 500

ng/kg dose level was initiated, but had to be closed because of

unexpected severe toxicity (38). Although the reason for the

severe toxicity in this Phase II trial is still being explored, initial

results indicate that the omission of the test dose was a major

contributing factor.4 The observation from this study that IFN-�y

levels progressively declined with each successive cycle of

rhIL-l 2 supports the contention that prior rhIL-l2 administra-

tion may in some way exert a protective effect against subse-

quent toxicity and that this effect may be sustained over at least

4 J_ P. Leonard, M. L. Sherman, 0. L. Fisher. L. J. Buchanan, 0. Larsen, M.B. Atkins, J. A. Sosman, J. P. Dutcher, N. J. Vogelzang, and J. L. Ryan,Effects of single-dose interleukin 12 (IL-12) on IL-l2-associated toxicityand interferon--y production. manuscript in preparation.

a 10-17-day interval. Ifthis hypothesis is confirmed in ongoing

preclinical and subsequent clinical testing, then this would have

a major impact on the design of future rhIL- I 2 studies. Because

of the severe toxicity associated with minor schedule changes,

additional Phase I studies are necessary to examine alternative

schedules and further evaluate the significance and mechanism

underlying this “desensitization effect.”

ACKNOWLEDGMENTS

We thank Kerry Kappler, R. N., M. S., N. P., Sandy Dunstan, R. N.,

Lynn Colicchio, R. N., Lisa O’Connor, R. N., and Donna Kinzler, R. N.,for nursing support; Deborah Rovner, Steven Chartier, and Dr. Beth

Harrison-Mann for data management; Christine Cameron, Dr. Zhiling

Yu, and Kyle McCarthy for laboratory assistance: and Jacqueline L.

Myers for assistance in preparing the manuscript.

REFERENCES

I. Kobayashi, M., Fitz, L., Ryan, M., Hewick, R. M., Clark, S. C.,Chan, S.. Loudon, R.. Sherman, F., Perussia, B., and Trinchieri, 0.Identification and purification of natural killer cell stimulatory factor(NKSF), a cytokine with multiple biologic effects on human lympho-cytes. J. Exp. Med., 170: 827-845, 1989.

2. Stem, A. S., Podlaski, F. J., Hulmes, J. D., Pan, Y-C. E., Quinn, P. M.,

Wolitzky, A. 0., Familletti, P. C., Stremlo, D. L., Truitt, I., Chizzonite, R.,

and Gately, M. K. Purification to homogeneity and partial characterizationof cytotoxic lymphocyte maturation factor from human B-lymphoblastoid

cells. Proc. Natl. Acad. Sci. USA, 87: 6808-6812, 1990.

3. Wolf, S. F., Temple, P. A., Kobayashi, M., Young, D., Dicig, M.,

Lowe, L., Dzialo, R., Fits, L., Ferenz, C., Hewick, R. M., Kelleher, K.,Herrmann, S. H., Clark, S. C., Azzoni, L., Chan, S. H., Trinchieri, 0.,

and Perussia, B. Cloning of cDNA for natural killer cell stimulatory

factor. a heterodimeric cytokine with multiple biologic effects on I andnatural killer cells. J. Immunol., 146: 3074-3081, 1991.

4. Sieburth, D., Jabs, E. W., Warrington, J. A., Li, X., Lasota, J., LaForgia,

S., Kelleher, K., Huebner, K., Wa.smuth, J. J., and Wolf, S. F. Assignmentofgenes encoding a unique cytokine (IL-IL-l2) composed oftwo unrelated

subunits to chromosomes 3 and 5. Genomics, 14: 59-62, 1992.

5. Merberg. D. M., Wolf, S. F., and Clark, S. C. Sequence similarity

between NKSF and the IL6/G-CSF family. Immunol. Today, /3: 77-78,

1992.

6. Gearing, D. P., and Cosman, D. Homology of the p40 subunit of

natural killer cell stimulatory factor (NKSF) with the extra cellular

domain of the interleukin-6 receptor. Cell, 66: 9-10, 1991.

7. D’Andrea, A.. Rengaraju, M., Valiante. N. M., Chehimi, J.. Kubin,

M.. Aste, M., Chan, S. H., Kobayashi. M., Young, D., Nickbarg, E.,Chizzonite, R., Wolf, S. F., and Trinchieri, 0. Production of naturalkiller cell stimulatory factors (interleukin- I 2) by peripheral blood mono-nuclear cells. J. Exp. Med., 176: 1387-1398, 1992.

8. Chua, A. 0., Chizzonite, R., Bhupesh, B. D., Truitt, I. P., Nunes, P.,

Minetti, L. J., Warner, R. R., Presky, D. H., Levine, J. F., Gately, M. K.,

and Gubler, U. Expression cloning a human IL- 12 receptor component:




a new member of the cytokine receptor superfamily with strong homol-

ogy to gpl3O. J. Immunol., 153: 128-136, 1994.

9. Chizzonite, R., Truitt, I., Desai, B. B., Nunes, P., Podlaski, F. J.,Stem, A. S., and Gately, M. K. IL-l2 receptor. I. Characterization of thereceptor on phytohemagglutinin-activated human lymphoblasts. J. Im-munol., 148: 3117-3124, 1992.

10. Desai, B. B., Quinn, P. M., Wolitzky, A. 0., Mongini, P. K. A.,Chizzonite, R., and Gately, M. K. IL-l2 receptor. II. Distribution andregulation of receptor expression. J. Immunol., 148: 3125-3132, 1992.

1 1. Metzger, D. W., Vogel, L. A., Van Cleave, V. H., Lester, I. L., andBuchanan, J. M. The effects of IL-l2 on B-cell subset function. Res.Immunol., 146: 499-505, 1995.

12. Robertson, M. J., Soiffer, R. J., Wolf, S. F., Manley, I. J.,

Donauhue, C., Young, D., Herrmann, S. H., and Ritz, J. Responses ofhuman natural killer (NK) cells to NK cell stimulatory factor (NKSF):cytolytic activity and proliferation of NK cells are differentially regu-lated by NKSF. J. Exp. Med., 175: 779-788, 1992.

13. Naume, B., Gately, M., and Expevi, I. A comparative study of

IL-l2 (cytotoxic lymphocyte maturation factor), IL-2, and IL-7 inducedeffects on immunomagnetically purified CD56+ NK cells. J. Immunol.,143: 127-142, 1992.

14. Gately, M. K., Wolitzky, A. 0., Quinn, P. M., and Chizzonite, R.

Regulation of human cytolytic lymphocyte responses by interleukin-l2.Cell. Immunol., 143: 127-142, 1992.

15. Gately, M. K., Desai, B. B., Wolitzky, A. 0., Quinn, P. M., Dwyer,C. M., Podlaski, F. J., Familletti, P. C., Sinigaglia, F., Chizzonite, R.,Gubler, U., and Stern, A. S. Regulation of human lymphocyte prolifer-ation by a heterodimeric cytokine, IL-l2 (cytotic lymphocyte maturationfactor). J. Immunol., 147: 874-882, 1991.

16. Perussia, B., Chan, S. H., D’Andrea, A., Tsuji, K., Santoli, D.,Pospisil, M., Young, D., Wolf, S. F., and Trinchieri, 0. Natural killer

(NK) cell stimulatory factor or IL-l2 has differential effects on theproliferation of TCR-a�3+, TCR--y� + I lymphocytes, and NK cells.J. Immunol., 149: 3495-3502, 1992.

17. Bertagnolli, M. M., Lin, B-Y., Young, D., and Herrmann, S. H.

IL- 12 augments antigen-dependent proliferation of activated I lympho-cytes. J. Immunol., 149: 3778-3783, 1992.

18. Chan, S. H., Kobayashi, M., Santoli, D., Perussia, B., andTrinchieri, 0. Mechanisms of IFN--y induction by natural killer cellstimulatory factor (NKSF/IL-12). Role of transcription and mRNAstability in the synergistic interaction between NKSF and IL-2. J. Im-munol., 148: 92-98, 1992.

19. Mier, J., Vachino, 0., Van der Meer, J., Numerof, R., Adams, S.,Cannon, J., Bernheim, H., Atkins, M., Parkinson, D., and Dinarello, C.Induction of circulating tumor necrosis factor (TNF) as the mechanismfor the febrile response to Interleukin-2 (IL-2) in cancer patients. J. Clin.Immunol., 8: 426-436, 1988.

20. Manetti, R., Parronchi, P., Giudizi, M. 0., Piccinni, M-P., Maggie,

E., Irinchieri, 0., and Romagnani, S. Natural killer cell stimulatoryfactor (interleukin 12 [IL-l2] induces I helper type 1 (IH 1)-specificimmune responses and inhibits the development of IL-4-producing Th

cells. J. Exp. Med., 177: 1199-1204, 1993.

21 . Scott, P. IL- 12: initial cytokine for cell-mediate immunity. Science(Washington DC), 260: 496-497, 1993.

22. Hsieh, C-S., Macatonia, S. E., Tripp, C. S., Wolf, S. F., O’Garra, A.,

and Murphy, K. M. Development of I�l CD4+ I cells through IL-12produced by Listeria-induced macrophages. Science (Washington DC),260: 547-549, 1993.

23. Gazzinelli, R. I., Hieny, S., Wynn, I. A., Wolf, S. F., and Sher, A.

Interleukin 12 is required for the I-lymphocyte independent induction of

interfemn -y by an intra-cellular parasite and induces resistance in I-celldeficient hosts. Proc. Nail. Acad. Sci. USA, 90: 6115-6119, 1993.

24. Sypek, J. P., Chung, C. L., Mayor, S. E. H., Subramanyam, J. M.,

Goldman, S. J., Sieburth, D. S., Wolf, S. F., and Schaub, R. 0. Reso-lution of cutaneous leishmaniasis: interleukin- I 2 initiates a protective T

helper type 1 immune response. J. Exp. Med., 177: 1797-1802, 1993.

25. Afonso, L. C. C., Scharton, T. M., Vieira, L. Q., Wysocka, M.,Trinchieri, 0., and Scott, P. IL-l2 functions as an effective adjuvant ina vaccine against Leishmania major by directing the development ofleishmanial specific CD4+ Thl cells. Science (Washington DC), 263:

235-237, 1994.

26. Brunda, M. J., Luistro, L., Warner, R. R., Wright, R. B., Hubbard,B. R., Murphy, M., Wolf, S. F., and Gately, M. K. Antitumor andantimetastatic activity of interleukin- 12 against murine tumors. J. Exp.Med., 178: 1223-1230, 1993.

27. Nastala, C. L., Edington, H. D., McKinney, I. 0., Tahara, H.,

Nalesnik, M. A., Brunda, M. J., Gately, M. K., Wolf, S. F., Schreiber,R. D., Storkus, W. J., and Lotze, M. I. Recombinant IL-12 administra-tion induces tumor regression in association with IFN Gamma produc-tion. J. Immunol., 153: 1697-1706, 1994.

28. Murray, K. S., Garzone, P., Wiencis, A. M., Kolowsky, M. L., Velick-ovic, A., Celniker, A., and Van Cleave, V. H. Chemiluminescent-based

ELISA for IL-l2: high sensitivity permits pharmacokinetic analysis. Ab-stract of the Ninth International Congress of Immunology, p. 23, 1995.

29. Oken, M. M., Creech, R. H., Tormey, D. C., Horton, J., Davis, I. E.,

McFadden, E. I., and Carbone, P. P. Toxicity and response criteria ofthe Eastern Cooperative Oncology Group. Am. J. Clin. Oncol., 5:

649-655, 1982.

30. Dinarello, C. A., Cannon, J. 0., Mier, J. W., Bernheim, H. A.,Lo-Preste, 0., Lynn, D. L., Love, R. N., Webb, A. C., Auron, P. E.,Reuben, R. C., Rich, A., Wolff, S. M., and Putney, S. D. Multiplebiological activities of human recombinant interleukin-l . J. Clin. In-vest., 77: 1734-1739, 1986.

31. Atkins, M. B., Gould, J. A., Allegretta, M., Li, J. J., Dempsey,R. A., Rudders, R. A., Parkinson, D. R., Reichlin, S., and Mier J. W.Phase I evaluation of recombinant interleukin-2 in patients with ad-vanced malignant disease. J. Clin. Oncol., 4: 1380-1391, 1986.

32. Atkins, M. B., Vachino, 0., Tilg, H. J., Kappler, K., Karp, D. D.,Robert, N. J., and Mier, J. W. Phase I evaluation of thrice dailyintravenous bolus Interleukin-4 (IL-4) in patients with refractory malig-nancy. J. Clin. Oncol., JO: 1802, 1992.

33. Weber, J., Yang, J. C., Topalian, S. L., Parkinson, D. R.,Schwartzentruber, D. S., Ettinghausen, S. E., Gunn, H., Mixon, A., Kim,H., Cole, D., Levin, R., and Rosenberg, S. A. A phase I trial ofsubcutaneous interleukin-6 in patients with advanced malignancies.J. Clin. Oncol., 11: 449, 1993.

34. Sosman, J. A., Aronson, F. R., Sznol, M., Atkins, M. B., Dutcher,J. P., Fisher, R. I., Weiss, 0. R., Isaacs, R. E., Margolin, K. A., Ernest,M. L., Mier, J. W., Oleksowicz, L., Eckhardt, J. R., Levitt, D., andDoroshaw, J. H. Interleukin-6: Phase I trials of 1 and 120 h intravenousinfusions. Proc. Am. Soc. Clin. Oncol., 12: 292, 1993.

35. Quesada, J. R. Biologic therapy with interferon--).. in: V. I. DeVita,S. Hellman, and S. A. Rosenberg (eds.), Biologic Therapy ofCancer, pp.435-442. Philadelphia: J. B. Lippincott Co., 1995.

36. Neta, R., Stiefel, S. M., Finkelman, F., Herrmann, S., and Ali, N.IL-12 protects bone marrow from and sensitizes intestinal tract toionizing radiation. J. Immunol., 153: 4230-4237, 1994.

37. Planchon, S. M., Martins, C. A. P., Guerrant, R. L., and Roche, J. K.Regulation of intestinal epithelial barrier function by TGF-�3l : evidencefor its role in abrogating the effect of a I cell cytokine. J. Immunol.,153: 5730-5739, 1994.

38. Cohen, J. IL-l2 deaths: explanation and a puzzle. Science (Wash-ington DC), 270: 908, 1995.



1997;3:409-417. Clin Cancer Res M B Atkins, M J Robertson, M Gordon, et al. interleukin 12 in patients with advanced malignancies.Phase I evaluation of intravenous recombinant human

Updated version

http://clincancerres.aacrjournals.org/content/3/3/409

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/3/3/409To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



phase i evaluation of intravenous recombinant human...

Documents