(CANCER RESEARCH 50, 7343-7350. November 15. 1990]

Interleukin 2 and Lymphokine-activated Killer Cell Therapy: Analysis of a BolusInterleukin 2 and a Continuous Infusion Interleukin 2 Regimen1

Jeffrey W. Clark, John W. Smith II,2 Ronald G. Steis, Walter J. Urba, Edward Crum, Robin Miller, John McKnight,

JoAnn Heñían,Henry C. Stevenson, Stephen Creekmore, Michael Stewart, Kevin Gonion, Mario Sznol,Peter Kremers, Paul Cohen, and Dan L. LongoBiological Response Modifiers Program, Division of Cancer Treatment [J. W. C., J. ». S.. R. G. S., E. C., R. M.. J. M., J. B., H. C. S., S. C.. M. S., K. C., M. S.,D. L. L.], Laboratory of Pathology [P. C.J, Division of Cancer Biology, Diagnosis, and Centers, National Cancer Institute, Bethesda, Maryland 20892, and ProgramResources, Inc. [W. J. U.J, National Cancer institute-Frederick Cancer Research and Development Center, and Frederick Memorial Hospital [P. K.J,Frederick, Maryland 21701

ABSTRACT

Several groups have described the efficacy of interleukin 2 (IL-2) pluslymphokine-activated killer (LAK) cells in the treatment of cancer patients with significant response rates noted in patients with renal cellcancer and malignant melanoma; however, the optimum regimen remainsundefined. The Biological Response Modifiers Program of the NationalCancer Institute conducted two consecutive Phase I, II studies evaluatingthe toxicity and clinical efficacy of different methods of IL-2 and LAKcell therapy. In the first trial, we modified the standard Rosenbergregimen by decreasing the duration of priming in an attempt to reducethe toxicity related to this phase of the therapy and thereby administermore IL-2 doses with the LAK cells. In the second trial, we used acontinuous i.v. infusion IL-2 regimen and altered both the leukapheresisprocedure and the LAK cell culture techniques based on our in vitro andpreclinical studies suggesting that 2-day LAK cells were superior. Thirtycancer patients received i.v. bolus IL-2 at 100,000 units/kg every 8 h for3 days during priming and for 5 days during LAK cell administration. Asecond group of 22 cancer patients received IL-2 by continuous i.v.infusion at 3 x III" units/m2 for 5 days during priming and an additional

5 days of IL-2 with the LAK cell phase of the treatment. The timing ofthe start of the leukapheresis procedures, their duration and number, andthe LAK cell culture techniques differed in the two trials. Overall, 52patients with various cancers were treated. The toxicities associated witheach regimen were similar to those seen in other IL-2 plus LAK celltrials. Four patients (one each with melanoma and diffuse large celllymphoma and two with renal cell cancer) exhibited partial responseslasting 2, 4, 10, and 15+ mo. Serial tumor biopsies from treated patientsdemonstrated that therapy can produce a marked mononuclear cellinfiltrate and an increase in HLA-DR expression on tumor cells. Therewas no difference in the overall response rate between the two regimens,but toxicity was less with continuous i.v. infusion IL-2. The 5-daycontinuous i.v. infusion regimen resulted in significantly higher reboundlymphocytosis, cell yield from leukapheresis, and number of LAK cellsharvested from culture.

INTRODUCTION

The initial report in 1985 of tumor responses after adoptiveimmunotherapy with IL-23 and LAK cells prompted multiple

confirmatory trials evaluating the toxicity and efficacy of various schedules of high-dose IL-2 plus LAK cell therapy in the

Received 12/18/89: accepted 8/17/90.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1This project has been funded at least in part with federal funds from theDepartment of Health and Human Services under Contract N01-CO-74102 withProgram Resources. Inc.

1To whom requests for reprints should be addressed, at Biological Response

Modifiers Program. Division of Cancer Treatment. National Cancer Institute.Frederick Cancer Research and Development Center. Building 567. Room 135,Frederick, M D 21702-1201.

3The abbreviations used are: IL-2. interleukin 2; LAK, lymphokine-activatedkiller cells; RCC, renal cell carcinoma: CT. computed tomography; ECG. electrocardiogram; DLCL, diffuse large cell lymphoma; NHL, non-Hodgkin's lym

phoma.

treatment of cancer (1-8). Rosenberg et al. (l, 2) used highdose i.v. bolus IL-2 (100,000 units/kg/8 h) for the priming andLAK cell administration phases of treatment. Although thisapproach resulted in an overall response rate (partial and complete responses) of 22%, toxicity was significant, and a majorityof patients required admission to an intensive care unit sometime during treatment. West et al. (3) administered IL-2 at 3 xIO6 units/m2/day by continuous i.v. infusion for the priming

and LAK cell phases. This change in dose and schedule significantly reduced toxicity without decreasing the overall responserate.

We conducted two sequential Phase I/II studies evaluatingthe toxicity and clinical efficacy of different methods of IL-2and LAK cell therapy. In our first trial, we modified thestandard Rosenberg regimen by decreasing the duration of IL-2 priming from 5 to 3 days in an attempt to reduce the toxicityrelated to this phase of the therapy and thereby administer moreIL-2 doses with the LAK cells. In our second study, we used acontinuous i.v. infusion IL-2 regimen and altered both theleukapheresis procedure and the conditions of LAK cell culture.We began leukapheresis 36 h after IL-2 priming ended andreduced the number of leukapheresis from five to three basedon our own observations and data from other trials (2, 9),indicating that the peak lymphocyte count occurs 48 h aftercessation of IL-2 and the yield of LAK cells decreases dramatically on Days 4 and 5 of pheresis. The length of in vitro cultureof cells with IL-2 was shortened to 2 days because of datasuggesting that these cells produce more cytokines than cellscultured for longer periods (10) and because of animal experiments suggesting that cells cultured for shorter periods of timetraffic better to tumor sites (11). This paper documents thetoxicity, immunological effects, and the antitumor activity ofthese two different IL-2 plus LAK cell regimens.

PATIENTS AND METHODS

All patients were treated at the Biological Response Modifiers Program, National Cancer Institute, Frederick, MD. Both protocols described below were approved by the Institutional Review Boards of theClinical Oncology Program, National Cancer Institute, and the Frederick Cancer Research and Development Center. All patients voluntarily gave written informed consent prior to treatment.

Patients. Eligibility criteria for both protocols were the following: (a)documented cancer with évaluableor measurable disease; (b) failedstandard treatment; (c) age >15 yr; (d) expected survival of greaterthan 3 mo; (e) diffusion capacity >75% of predicted; (/) no centralnervous system métastases;(g) white blood cell count >2000/mm'; (h)

no antineoplastic therapy within the month prior to study entry; and(i) no active systemic infection, coagulation disorders, or major cardiovascular or pulmonary' disease.

Prestudy evaluation included history and physical examination, complete blood cell count, serum chemistry profile, prothrombin time and

7343

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

IL-2/LAK THERAPY: ANALYSIS OF TWO DIFFERENT REGIMENS

partial thromboplastin time, urinalysis and culture, serum iron bindingcapacity, hepatitis B surface antigen, antibodies to human immunodeficiency virus 1, chest X-ray, CT scan of the brain, other appropriateimaging studies to measure disease, and pulmonary function testsincluding carbon monoxide diffusing capacity and arterial blood gases.Many of the patients also underwent an exercise treadmill test to assesscardiac capacity.

Patient characteristics are listed in Table 1, and diagnoses are listedin Table 4. The characteristics of the patients enrolled in the two studiesare similar except for a higher percentage of patients who received anyprior treatment on the first trial (80% versus 59%) as well as a higherpercentage of patients who received chemotherapy and/or radiationtherapy (73% versus 50%). Initial patients enrolled onto the first schedule primarily had renal cell carcinoma, malignant melanoma, or coloncarcinoma. After treating 19 patients, efforts focused on entering patients with relapsed Hodgkin's disease or relapsed non-Hodgkin's lym-

phoma. Patients were enrolled onto the second trial without regard totheir diagnosis.

Treatment Schema. The two treatment regimens are depicted in Figs.1 and 2. The recombinant IL-2 used in these trials was produced inEscherichia coli transfected with the gene from the Jurkat cell line andwas generously supplied by Cetus Corporation, Emeryville, CA. Itsspecific activity was approximately 3x10' units/mg of protein (Cetus

units).Schedule I is identical to the regimen reported by Rosenberg et al.

(1) except that the initial IL-2 priming period was 3 days instead of 5,and a repeat cycle of leukapheresis not preceded by IL-2 primingfollowed by an additional 5 days of IL-2 and LAK cell infusion wasgiven. Recombinant IL-2 (100,000 units/kg) was diluted in 50 ml of5% dextrose in water and administered as an i.v. bolus over 15 minevery 8 h for a maximum of 9 doses as priming therapy and likewisewith the LAK cells for a planned 5 to 7 days. IL-2 was either given infull dose on schedule or omitted because of toxicity based on criteriasimilar to those described elsewhere (6).

In Schedule I, leukapheresis started at 8:00 a.m. on Monday approximately 56 h after the last priming dose of IL-2 at midnight on Friday.Leukapheresis was performed for 4 h per day for 5 days using an IBM-2997 continuous flow cell separator (Cobe Laboratories, Lakewood,

Table 1 Patient characteristics

No. ofpatientsMale/femaleMedian

age(range)Karnofskyperformancestatus100%80-90%Prior

treatmentNoneChemotherapyRadiation

therapyChemotherapyor radiationtherapyHormonal

therapyImmunotherapySchedule

13021/942(13-68)2436

(20%)17(57%)12(40%)22

(73%)2(13%)4

(7%)Schedule

II2214/849

(20-63)1839(41%)10(46%)7

(32%)11(50%)2(18%)4

(9%)

CO) at a centrifuge speed of 920 rpm and WBC collection rate of 2ml/min. Twelve liters of blood were processed in 4 h at a flow rate of50 ml/min using acid citrate dextrose (NIH Formula A) as an anticoagulant. The final volume collected was approximately 500 ml with ahemoglobin of 0.8 to 1.2 g/dl and a lymphocyte content of 75 to 95%.In almost all the patients, this procedure was performed using a doublelumen central catheter designed for hemodialysis (Quinton).

In Schedule II, IL-2 was administered by continuous i.v. infusion at3 x IO6 units/nr for 120 h as priming and for 120 h with the LAK

cells, and an identical course was repeated after 2 wk of rest (Fig. 2).The timing of leukapheresis in Schedule II was 36 h after IL-2 primingended, rather than the 56-h interval in Schedule I. The number ofleukaphereses was reduced to three, but their duration was increased to5 h using the same machine and technique as previously described.

LAK Cell Generation. For Schedule I, patients' mononuclear cells

were separated by centrifugation through a Ficoll-Hypaque densitygradient and washed with Hanks' balanced salt solution. Cells were

cultured in RPMI-1640 medium containing 50 Mgof streptomycin, 5units/ml of penicillin, 5 Mg/ml of gentamicin, 2 mM glutamine, 2%heat-inactivated human AB serum (M. A. Bioproducts, Walkersville,MD), and 1000 units/ml of IL-2 at a concentration of 1 to 2 x IO6cells/ml. Penicillin was excluded from cultures of cells from penicillin-allergic patients. Cells were cultured in 2-liter roller bottles at 1 rpm at37°Cfor 3 or 4 days. LAK cells were collected by centrifugation,washed in Hanks' salt solution, and resuspended in 200 ml of normal

saline with 5% human serum albumin and 75,000 units of IL-2. Thismixture was filtered through Nytex filters (Tetko, Inc., Zurich, Switzerland) and transferred to a 600-ml transfer unit pack. Aliquots wereexamined microscopically, cultured for bacterial and fungal contamination, and tested for LAK activity. LAK activity was assessed in vitrousing Daudi cells as targets in a 4-h chromium release assay that wehave described previously (12).

Autologous LAK cells were administered in Schedule I in threeinfusions on Days 10 (cells from Days 6 and 7 leukaphereses), 11 (cellsfrom Day 8 leukaphereses), and 13 (cells from Days 9 and 10 leukaphereses). LAK cells were infused i.v. over 30 to 60 min starting within45 min from the end of cell harvest.

LAK cell generation in the second trial differed in several ways.Patients' mononuclear cells were not centrifuged over Ficoll-Hypaque

because a detailed analysis of the cells obtained from leukapheresisrevealed that less than 5% of the cells were granulocytes and that thislevel of granulocyte contamination did not adversely affect LAK cellgeneration. Instead of using culture medium containing human ABserum, serum-free medium (AIM V; Gioco) with glutamine, streptomycin, and gentamicin was used after experiments showed that thismedium supported the growth of LAK cells as well as the serum-containing medium for the 2-day period of culture. Cells from patientstreated on Schedule II were cultured at a density of 2.5 to 5.0 x IO6cells/ml instead of 1 to 2 x 10' cells/ml, because experiments indicated

similar LAK cell yields and activity under these conditions. Lastly, theduration of culture was shortened to 2 days.

For patients treated according to Schedule II, autologous LAK cellswere administered in six infusions on Days 10, II, and 12 with cells

Schedule I

IL-2 Priming

9 boluses1234

I I I IW Th F S

LAK Cells

LeukapheresisIL-2

LAK Cells

LeukapheresisI I T

IL-2

12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

I I I I I I I I I I I I I I I I I I I I I

WThFSSMTWThFSSMTWThFS

LAK Cells LAK Cells

Fig. 1. First IL-2/LAK clinical protocol.

7344

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

IL-2/LAK THERAPY: ANALYSIS OF TWO DIFFERENT REGIMENS

Schedule II

Leukapheresis,.

1121120hrs345I

! I61|

1 ; 120hrs7I8I9_J-10I11:12113:14

15

IL-2 Priming Leukapheresis,. IL-2

120 hrs120 hrs ••I •

Rest 29 30 31 32 33 34 35 36 37 38 39 40 41 42 432 I—I—I—I—I—I—I—I—I—I—I 1 1—I—I

MTWThFSSMTWThFSSM Weeks MTWThFSSMTWThFSSM

LAK Cells

Fig. 2. Second IL-2/LAK clinical protocol.

LAK Cells

from Days 8, 9, and 10 leukaphereses, respectively, and on Days 38,39, and 40 with cells from Days 36, 37, and 38 leukaphereses (see Fig.2).

In Schedule I, patients routinely underwent IL-2 priming on a regularinpatient oncology floor but received their LAK cells in a unit with thecapability for continuous ECG and pulse oximeter monitoring andintensive care unit-level nursing care. Patients treated according toSchedule II were transferred to the monitoring unit only when clinicallyindicated.

Patients were evaluated for response 4 wk after the completion oftherapy. Any objective antitumor response, even if it did not meet thecriteria for a partial response, qualified the patient for another treatment cycle.

Response Criteria. Standard response criteria were used. A completeresponse was defined as the disappearance of all tumor for at least 4wk with no new lesions developing. Partial response was defined as50% or greater reduction in the sum of the products of the largestperpendicular diameters of all measurable disease maintained for atleast 4 wk without any new lesions appearing. Progressive disease wasan increase of greater than 25% in the sum of the products of thelargest perpendicular diameters or the appearance of any new lesions.Stable disease was defined as disease not meeting the above criteria forresponse or progression.

Supportive Care. In both treatment regimens, patients received in-domethacin (25 to 50 mg p.o. or per rectum every 8 h), ranitidine (150mg p.o. twice a day or 50 mg i.v. every 6 h), and acetaminophen (650mg p.o. or per rectum every 4 h). In addition, meperidine for chills,hydroxyzine hydrochloride or diphenhydramine for pruritus, low-dosedopamine for oliguria, phenylephrine for blood pressure support, andantidiarrheal and antiemetic treatment were administered as needed.The first nine patients without a diagnosis of Hodgkin's disease or

lymphoma treated on Schedule I received dexamethasone (4 mg) p.o.every 6 h during the LAK cell infusion phase of the protocol because itwas thought at that time that steroids would reduce IL-2-related toxicityand allow more IL-2 to be administered. Subsequently, concernemerged that the steroids were interfering with the antitumor efficacyof the regimen, so no other patients on either study received steroids.

Immunopathological Studies. Histopathological and immunopheno-typic analyses of sequential tumor biopsy specimens from three patientswith accessible metastatic lesions were performed according to thetechniques described previously (13).

RESULTS

Twenty-seven of 30 patients who began treatment accordingto Schedule I are évaluablefor toxicity and response assessment.Three patients were removed from study during priming because of intercurrent medical problems [hypercalcemia (onepatient), vertebral metastasis (one patient)] requiring treatmentor because of patient refusal (one patient). Twenty-three of the27 évaluablepatients completed the entire 30 days of treatment;four patients received treatment through their first LAK cell

phase only, because of toxicity described below.All patients are évaluablefor toxicity, and 21 of 22 are

évaluablefor response on Schedule II. One patient completedIL-2 priming on Schedule II but refused to continue. Fourpatients did not receive their second round of IL-2/LAK treatment (Days 29 to 43); two because of progressive disease andtwo because of toxicity.

Number of IL-2 Doses and Cells Administered. The meantotal number of IL-2 doses administered on Schedule I was 29.Patients received an average of 8.6 of the planned 9 IL-2priming doses (27 patients), a mean of 10 doses (range, 1 to15) with the first LAK cell infusions (27 patients), and a meanof 10.4 doses (range, 4 to 23) with the second round of LAKcell infusions (23 patients). Only three patients received morethan 15 doses of IL-2 with LAK cells (one patient, 16 doses;one patient, 17 doses; one patient, 23 doses). The number ofIL-2 doses administered with the LAK cells was not significantly different in the patients who also were treated withsteroids (total of 21 with steroids versus 22.3 without). One of9 patients (11.1%) treated with steroids and 3 of 18 patients(17.6%) treated without steroids did not receive the second halfof their planned therapy due to the toxicity of treatment.

The mean lymphocyte rebound, number of cells obtainedfrom leukapheresis and placed into culture, and the number ofLAK cells infused per leukapheresis cycle for both schedulesare listed in Table 2. Despite the fact that IL-2 usually ended 5to 7 days before the second set of leukaphereses in Schedule I,the lymphocyte count on the first day ofthat leukapheresis, theleukapheresis yield, and the number of LAK cells harvestedwere essentially the same as those observed during the first setof leukaphereses after IL-2 priming. The number of LAK cellsharvested after culture in vitro represented 48% of the cells(lymphocytes and monocytes) obtained from leukapheresis.

Patients treated with continuous infusion IL-2 on ScheduleII received a mean of 119.5/120 h (99.6%) of IL-2 primingduring the first half of their treatment course (22 patients) and117.5 of 120 h (98%) of IL-2 priming during the second half(17 patients). They received a mean of 116 of 120 h (96.7%)with their first set of LAK infusions (21 patients) and 112.6 of120 h (93.8%) with their second set (17 patients).

The lymphocyte rebound, yield from leukapheresis, and number of LAK cells harvested for Schedule II are listed in Table2. The results in each of these areas were approximately 2 timesgreater than with Schedule I (P < 0.0001). In Schedule II, thenumber of LAK cells harvested represented 64% of the cellsobtained from leukapheresis.

Toxicity. The toxicities associated with treatment are listedin Table 3. These are very similar to those reported for other

7345

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

IL-2/LAK THERAPY: ANALYSIS OF TWO DIFFERENT REGIMENS

Table 2 Rebound lymphocytosis and L.4K cell culture data

Lymphocyterebound"Schedule

ISchedule IIFirst

pheresis3577±Ì55"1

7623 ±545(113t>'-/Second

pheresis3901

±3999395 ±664 (124|)/Leukapheresis

yield*First

pheresis9.7

x 10'°±0.617.1 x 10'°±1.2(76|)/Second

pheresis9.1

x 10'°±1.019.3 x 10'°±1.5(112No.

of LAK cellsharvested'First

pheresis4.9x 10'°±0.4

:])' 11. Ox 10'0±0.8(124|)/Second

pheresis4.2

x 10'°±0.512.2 x 10'°±0.9 (190|)/

.* Total number of cells (lymphocytes plus monocytes) obtained from the set of leukaphereses.' Total number of cells obtained after in vitro culture.' Mean ±SEM.' Numbers in parentheses, percentage of increase compared with Schedule I.^ Significance level of each parameter in Schedule II compared with Schedule 1. P< 0.00001 (unpaired Student's r test).

Table 3 Systemic LAK + IL-2 loxicities

Schedule I Schedule II

FatigueFever/chillsNausea/vomitingDiarrheaDiffuse

erythema/pruritusWtgain (>10% bodywt)Myalgia/arthralgiaNasal

congestionCatheter-relatedinfectionsSevere

mucositis (GradeIII)"Severe

respiratory distress (GradeIV)Hypotensionrequiring pressors(GradeIII)Central

nervous system disturbance(anxi-ety/somnolence/disorientation/inap-propriate

behavior/headache)Generalizedseizure (GradeIV)Cortical

blindness (GradeIV)Cardiacarrhythmias (primarily atrial al

though ventricular alsoseen)Death(pulmonaryinsufficiency)Abnormal

liver function,hyperbilirubine-mia

(>3.0) (Grade IV) orelevatedserumglutamic-oxaloacetictransami-nase

(>5 times normal) (Grade111)Abnormalrenal function, oliguriaand/orcreatinine

>2.0Creatinine>5.0Anemia

requiring transfusion (GradeIII)Totalno. of units of RBCstransfusedThrombocytopeniaPlatelets

<100,000Platelets<50.000 (GradeIII)%

of patients requiring platelet transfusionsTotal

no. of units of plateletstransfusedEosinophilia1009292929263603322151181604033448924100140"1003330246*9210095100951006131945055932553302471106453*6819510*100

' Grade of toxicity according to Common Toxicity Criteria of the Cancer

Treatment Evaluation Program.* Number.

high-dose IL-2 plus LAK cell therapy treatment regimens (1-6, 13-16). On both schedules, all patients had flu-like symp

toms; fevers; chills; malaise; and fatigue. The majority of patients also had fluid retention with weight gain, erythematousskin rash (often pruritic and desquamating), mild respiratorydistress, hypotension, nausea, vomiting, diarrhea (often severe),and psychiatric disturbances (ranging from anxiety to inappropriate behavior). Less commonly, cardiac arrhythmias (primarily atrial), myalgia, arthralgia, and mucositis also occurred.Moderate to severe rigors occurred during LAK cell infusions,especially in patients treated according to Schedule II. Infections related to central venous catheters were a significantproblem in patients treated according to either schedule. Thehigher incidence of infection during the second schedule (45%versus 22%) is due partially to the fact that routine surveillanceblood cultures were obtained from the patients' catheters. On

Schedule I, respiratory distress requiring intubation occurredin two patients, and a grand mal seizure and angina wereobserved in one patient each. On Schedule II, no patientsrequired intubation, although one had severe respiratory distress, and one each had a seizure and angina. One patient onSchedule II developed transient cortical blindness lasting lessthan 24 h. Extensive neurological workup including a magneticresonance imaging scan of the brain was negative. The onlydeath during treatment was a patient with Hodgkin's disease

on Schedule I who had parenchymal lung lesions and had beenheavily pretreated with chemotherapy, radiation therapy, andan autologous bone marrow transplant. This patient requiredintubation for respiratory distress after the third dose of LAKcells and could not be extubated. Bronchoscopic biopsies ontwo occasions showed marked pulmonary fibrosis which led toher demise. Although the toxicity on Schedule II was overallqualitatively very similar to that on Schedule I (Table 3),quantitatively it was much less severe as shown by the lowerincidence of Grade III or Grade IV toxicity. All toxicities onboth treatment regimens resolved within 2 wk of stoppingtreatment except for occasional patients who had persistentfatigue, myalgias, or arthralgias as long as 4 wk after stoppingtreatment.

Clinical Response. The diagnosis of patients and the responseto treatment for both trials are listed in Table 4. Two partialresponses (melanoma and DLCL) were seen in patients following treatment according to Schedule I, and two partial responses

Table 4 Patient characteristics and response to therapy

DiagnosisSchedule

IRenalMelanomaNHLColorectalHodgkinsPancreasAdrenocorticalTotalSchedule

11MelanomaRenalColorectalOvarianPancreasFSCCLCervical

clearcellTotalNo.

ofpatients965441130944112122No.

ofpatients

évaluableforresponse955421127943112121Response01

PR"(10)"1PR(2)*0(1002

(7.4%)0(1

MR)"2PR (4.15+)*000002<9.5r¿)

" PR, partial response; MR, mixed response in a patient with a 50% reduction

in subcutaneous disease but slight progression of pulmonary métastases; FSCCL,follicular small cleaved cell lymphoma.

* Duration of response in months.

7346

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

IL-2/LAK THERAPY: ANALYSIS OF TWO DIFFERENT REGIMENS

(both RCC) were seen in patients treated according to thesecond schedule. The patient with melanoma received only 10doses of IL-2 and one infusion of LAK cells because of thedevelopment of angina accompanied by ischemie changes onECG. Her followable disease consisted of four subcutaneousnodules and a pulmonary nodule. When treatment was stoppeddue to angina, she had the four subcutaneous nodules (that hadnot changed in size) resected for evaluation of immunohisto-chemical changes (described below). At her return evaluation 1mo later, the nodule on chest X-ray had resolved and was notdetectable by CT scan. She had previously received radiationtherapy for a left humoral metastasis with a residual bone scanabnormality in that area which persisted unchanged throughouther course. Her response persisted for 10 mo when she relapsedwith a subcutaneous metastasis without recurrence of her lungnodule or the development of any new pulmonary lesions.

The patient with DLCL had progressive disease post chemotherapy. He had complete resolution of two subcutaneous nodules during the IL-2 priming phase of his treatment withoutany change in a left paraaortic lymph node. The response lastedfor 2 mo before disease progression occurred, manifested bythe development of a new 1-cm x 1-cm subcutaneous nodule,an increase in the size of the paraaortic lymph node, and thedevelopment of a new paraaortic lymph node.

On Schedule II, one patient with RCC had a partial responseconsisting of nearly complete regression of a large parenchyma!lung mass and disappearance of a second pulmonary nodulewithout any change in paratracheal lymphadenopathy. An additional treatment cycle failed to improve his partial remissionwhich lasted 4 mo when he had progression of disease in hislargest lung metastasis. The second patient with RCC had a30% reduction in the size of a large biopsy-proven paraaorticlymph node metastasis after one cycle of therapy and a furtherreduction to >50% after a second treatment cycle (Table 4).The patient removed himself from treatment after approximately 85 h of IL-2 during a third treatment cycle. CT scans 9mo post therapy continue to show nearly total resolution of hisretroperitoneal adenopathy.

Evidence of a transient minor response was seen during theIL-2 priming portion of the bolus regimen in two patients(disappearance of a lymph node in a patient with follicularsmall cleaved cell lymphoma and decrease in the size of subcutaneous lesions in a patient with melanoma), but both ofthese patients had progressive disease at the time of theirevaluation 1 mo after treatment. One patient with melanomatreated on the continuous infusion regimen had a mixed response with >50% reduction in subcutaneous métastaseswithprogression of pulmonary métastases.He was retreated withfurther response of his cutaneous disease but not of his pulmonary disease.

Characteristics of Renal Cell Cancer Patients. Six of the ninepatients with RCC treated according to Schedule I had prognostic features associated with a poor response (less than orequal to 10%) to LAK plus IL-2 in other studies (4). Threepatients still had their primary tumor in place (two were > than100 cm2 in the product of longest perpendicular diameters),

and three had liver métastases.One of the remaining three was65 yr old, and three patients overall were 63 yr old or older.This is older than any reported responders from other IL-2 plusLAK trials, although the ages of most responders has not beenstated in these reports.

The four RCC patients treated with continuous infusion IL-2 (two of whom responded) had relatively good prognostic

factors for response (métastasesto lung plus bone in one, lungplus subcutaneous tissue in a second, lung only in the third,and retroperitoneal lymph nodes in the fourth).

Immunopathological Changes. One patient with malignantmelanoma treated according to Schedule I had subcutaneousnodules that were biopsied before treatment, 1 wk after treatment, and at relapse more than 1 yr later. Pathological examination showed a marked T-cell infiltrate and an increasednumber of cells bearing the monocyte marker Leu M3 in thebiopsy obtained 1 wk after treatment. No Leu 19-positive cellswere detected in any of the biopsies, suggesting that the adoptively transferred LAK cells did not traffic to these tumor sites.This patient's tumor cells showed a marked increase in HLA-

DR expression from <20% of the cells showing weak (1+)positivity to all of the cells showing strong (3+) positivity aftertreatment (Fig. 3). Strong expression of HLA-DR was also

noted on the biopsy obtained at relapse 10 mo later. The patientwas not offered another course of LAK + IL-2 at relapsebecause of the treatment-induced myocardial ischemia documented on the first cycle.

Two other patients with malignant melanoma treated according to Schedule I had serial biopsies. One patient who had aminor tumor regression had a small T-cell and monocyte infiltrate pretreatment that increased slightly after therapy. No Leu19+ cells were noted before or after therapy. This patient's

biopsies before, during, and immediately post treatment allstained strongly (3+) positive for HLA-DR, but the biopsyobtained 10 days post treatment showed less than 10% of thetumor cells positive for HLA-DR. The other patient did notrespond to treatment and had no T-cell infiltrate before or aftertreatment. An infiltration of Leu M3- and Leu MS-positivecells was noted pretreatment, but did not increase post treatment. Less than 10% of this patient's tumor cells were HLA-

DR positive before and after treatment.

DISCUSSION

A number of studies have shown that high-dose IL-2 aloneor in combination with LAK cells has antitumor activity againstRCC, melanoma, NHL, and to a lesser extent, colon cancer. Avariety of schedules of IL-2 administration have been used, butit is still not clear what the optimum treatment schema is. Todetermine if modifications in the regimen might improve efficacy, 52 patients with various cancers were treated with eithera bolus IL-2 regimen in which the IL-2 priming phase consistedof 9 doses [versus 15 in the Rosenberg regimen (1)] and with arepeat cycle of IL-2 plus LAK cells without priming (30 patients) or a continuous infusion IL-2 regimen using 3 days ofleukapheresis and a shorter duration of in vitro LAK cell culturewith IL-2 than that reported by West et al. (22 patients) (3).

Toxicities of these two regimens were very similar to thosereported with other IL-2 plus LAK regimens. Although qualitatively similar, toxicity was quantitatively significantly lesssevere with continuous infusion IL-2 than with bolus administration. This difference in toxicity was due to the fact thatconsiderably less IL-2 was given on a daily basis during continuous infusion as opposed to bolus injection. Patients on thecontinuous infusion IL-2 regimen received only one-fourth toone-third the daily dose of IL-2 compared with the patients onthe bolus injection regimen. The continuous infusion methodof administration is actually more toxic; the maximum tolerateddose for IL-2 is much lower for a 24-h continuous i.v. infusion

than for the same dose given by i.v. bolus (8).7347

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

•-i

>:>•;... ' -'";..;_.,.-.

* ••". i "' ; ' ,», >,»•»- .

A ''v ' -. . -'V^

f-

Fig. 3. Tumor biopsy specimens of a patient with malignant cutaneous melanoma before (A, C, E. and G') and after (B, D, F, and //) treatment with IL-2/LAK.

Staining (H & E) shows malignant melanoma cells without a significant lymphoid infiltrate before therapy (A). Four days after the end of treatment, a markedmononuclear cell infiltrate is obvious with H & E(Ä). Before therapy, Leu-M.V cells (f') and CD3* cells (E) were rare, but after treatment they became more numerous

(D and F). Before therapy, less than 20S of the tumor cells stained positively for HLA-DR (G). After treatment (//). all the tumor cells stained strongly positive.This patient had a partial response to therapy. (Original magnification: A. x 100; B. x 200: C to H, x 100).

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

IL-2/LAK THERAPY: ANALYSIS OF TWO DIFFERENT REGIMENS

An aspect of our second schedule that differed from otherIL-2/LAK regimens was the use of LAK cells that had beencultured for only 2 days instead of 3 to 4 days. We showed inthis study that large numbers of these 2-day LAK cells couldbe given on 3 consecutive days with acceptable toxicities.

One of the unique features of the first regimen was a 3-dayIL-2 priming period which we hoped would produce comparablerebound lymphocytosis and LAK cell yield with less toxicityand thereby permit more IL-2 to be given with the LAK cells.Unfortunately, shortening the IL-2 priming to 3 days did notpermit delivery of significantly more bolus IL-2 with the LAKcell infusions as was hoped. The mean total number of bolusIL-2 doses (29 doses) in this study is greater than that administered by the Extramural IL-2/LAK Working Group (20 to 24doses) and by Rosenberg (20 doses); however, our doses weregiven over 30 days in 2 cycles, instead of 16 days in one cycle.The mean number of LAK cells infused per pheresis cycle inSchedule I (4.6 x 10'°)is lower than that reported by otherinvestigators (7.6 x 10'°);however, the mean total dose of LAKcells given (9.1 x 10'°)is comparable, although it was spread

over two cycles instead of one.The mean rebound lymphocytosis, leukapheresis yield, and

number of LAK cells harvested were approximately 2 timesgreater for the continuous IL-2 infusion regimen comparedwith the bolus IL-2 regimen (P < 0.0001) (Table 2). Whiledifferences in baseline lymphocyte counts and prior treatmentwith chemotherapy and/or radiation therapy could all contribute to differences in lymphocytosis or LAK cell yields, it probably does not account for the magnitude of the differencebetween the two schedules. Boldt et al. (9) noted that the peakrebound lymphocyte count following IL-2 priming correlatedwith age, number of IL-2 priming doses, prior treatment, andbaseline lymphocyte count. The patients without prior chemotherapy or radiation therapy had a mean rebound lymphocytosisof 5444 cells per ^1 versus 4084 cells per ß\in patients withprior treatment. A higher percentage of patients treated in ourfirst study had prior chemotherapy or radiation therapy (73%)versus our second trial (50%); however, the difference in meanrebound lymphocytosis in the two trials (7623 versus 3577 cellsper /il, P < 0.0001) is greater than the difference expected fromprior treatment alone. Eliminating the patients with non-Hodg-kin's lymphoma or Hodgkin's disease who were treated accord

ing to Schedule I from the analysis did not change the magnitude or significance of the difference between the two schedulesin any of the above-mentioned parameters.

It appears that 3-day bolus IL-2 priming does not produceeither as pronounced a rebound lymphocytosis or LAK cellyield as do 5 days of continuous IL-2 infusion priming; however,other factors may be influencing this difference. One is thetiming of leukapheresis (20 h earlier and 5 h/day for 3 days onSchedule II). The other is differences in culture technique (theuse of serum-free medium, the omission of Ficoll-Hypaque,and the shorter duration of culture with Schedule II). The LAKcell culture efficiency increased from 48% [similar to the 53%efficiency noted by others (9)] in Schedule I to 64% in ScheduleII. The lytic activity per cell of the LAK cells from bothregimens was similar and equivalent to that of cells generatedby 5 days of bolus IL-2 priming (data not shown).

The contribution of the reinfused ex v/vo-generated LAK cellsto the antitumor effect of IL-2 plus LAK is still not clear, sinceresponses to high-dose IL-2 alone have been reported and therehas not been a correlation between the number of LAK cellsinfused and response in most studies (1-3, 5, 9). This was also

true in our study where three of the four responders receivedequal to or less than the mean number of LAK cells of theirrespective regimens. Because there are extensive animal datasuggesting the importance of LAK cells in the IL-2 response, it

would appear desirable to attempt to maximize LAK cell number and activity in those regimens using ex v;Vo-generated LAKcells. Therefore, 5-day priming with either continuous infusionor bolus IL-2 would appear preferable to 3-day priming.

The continuous infusion IL-2 regimen produced a markedincrease (approximately 2-fold) in rebound lymphocytosis andLAK cell yield. Because of the many differences between thetwo schedules, it is difficult to define the contribution of anyone change. Recently, a randomized trial was reported comparing IL-2 given i.v. by continuous infusion versus bolus, bothwith LAK cells, in patients with renal cell carcinoma (14). Thestart of leukapheresis (36 h after IL-2 priming) was earlier thanprevious trials, and the technique was changed from prior trialsto one where they collected for 5 h/day for 4 days using acentrifuge speed of 1400 rpm, WBC collection rate of 4 ml/min, and flow rate of 80 ml/min, yielding a 1-liter final volumeof the Leukopak. LAK cell culture was performed withoutFicoll-Hypaque separation of cells and with serum-free mediumfor 3 to 4 days. The LAK cell harvest in the bolus arm wasconsiderably higher (2-fold) than earlier studies using the samebolus IL-2 regimen (100,000 units/kg every 8 h) despite thefact that patients received a mean of 10.8 priming boluses, lessthan the number of priming doses reported in earlier trials bythe same group [12.9 doses (4) and 14 doses (6)]. This indicatesthat the timing and duration of pheresis, elimination of Ficoll-Hypaque centrifugation, and use of serum-free medium probably account for most of the increase in LAK cell harvestobserved in their trial as well as ours.

The striking T-cell and macrophage infiltrate and the absenceof cells bearing natural killer or LAK phenotypic markers notedin the tissue biopsies of the responding patients are consistentwith the observations of Cohen et al. (13), who noted a correlation between such an infiltrate and response to treatment.These investigators also suggested that HLA-DR expression ontumor cells, whether it was present on the tumor before therapyor was induced as a result of treatment, correlated with tumorregression clinically. The data from our patients support thisfinding also; one patient with a partial response to therapy hadhigh levels of HLA-DR induced by treatment, another patientwith a minor tumor regression had high levels present beforetreatment, and a nonresponding patient had almost no HLA-DR expression on tumor cells before and after therapy. Theseobservations suggest that the tumor regressions noted in vivoare not mediated directly by LAK cells; instead, they could beascribed to an indirect effect of the IL-2 or LAK cells (inductionof secondary cytokines) or be a result of the influx of T-cells ormacrophages.

The diseases that responded to LAK and IL-2 in our twostudies (melanoma, RCC, DLCL) are the same tumors in whichresponses were seen with other IL-2 plus LAK regimens. Although the small number of patients, potential differences inpatient characteristics between different trials, and the lack ofinformation from randomized trials make the comparison ofresponse rates between different trials problematic at best, the20% response rate (one of 5) of the melanoma patients onSchedule I is comparable to that seen in other IL-2 plus LAKstudies (1-3, 5-7). Among the nine patients with melanomatreated with continuous infusion IL-2, seven had progressivedisease, one had stable disease, and one had a mixed response.

7349

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

IL-2/LAK THERAPY: ANALYSIS OF TWO DIFFERENT REGIMENS

Although the marked reduction in cutaneous métastasesseenin one patient with melanoma indicates antitumor activity, nocomplete or partial responses were seen in melanoma patientstreated on the continuous infusion regimen. This result isconsistent with the experience of the Extramural IL-2/LAKWorking Group (14, 15) and Thompson et al. (16), wherecontinuous infusion IL-2/LAK regimens produced negligibleresponse rates in melanoma (one of 39 and zero of 11, respectively). The reason for the difference in response rates of melanoma to bolus versus continuous infusion IL-2 regimens remains unexplained.

Two of four RCC patients treated with continuous infusionIL-2 had a partial response, a third patient had a slight reduction in the size of pulmonary lesions while his overall diseaseremained stable, and the fourth patient had to be removed fromstudy after he had received only one of the two planned coursesof IL-2 plus LAK cells because he developed a femoral metastasis requiring radiation. In contrast, although two of the RCCpatients treated on the bolus regimen had stable disease for aprolonged time (14.5 and 8.5 mo), there were no responses. Itis difficult to know whether this is related to the treatmentregimen or to other factors. Six of the nine patients treatedwith the bolus regimen had one or more of the features associated with a <10% response rate (liver métastases,primary inplace, or mass >100 cm2) when patients from other bolus IL-2

plus LAK cell trials were analyzed (4). Four of these patientsreceived steroids during the LAK cell portion of their treatmentincluding one of the three patients without the poor prognosticfactors listed above. Although there has been no randomizedtrial testing the effect of steroids, there are limited data suggesting that steroids may have adverse effects on the clinicalefficacy of treatment (17). Finally, three of these patients were63 yr of age or older and received 14, 21, and 24 total doses ofIL-2, respectively, which are all less than the mean number ofIL-2 doses given on this regimen (29). Therefore, there are anumber of possible explanations for the absence of response inthese patients. In contrast to the experience in melanomapatients, there has not been a significant difference in responserates for renal cell cancer patients treated with bolus versuscontinuous infusion IL-2 in other trials (2-4, 7, 16) and in theone randomized trial noted earlier (14).

These trials confirm the antitumor activity of IL-2 plus LAKagainst RCC, melanoma, and NHL, cancers previously reportedto be responsive to this immunotherapy. Although not conclusive, these studies indicate that 5 days of IL-2 priming bycontinuous i.v. infusion produced a significantly greater lymphocyte rebound, leukapheresis yield, and number of LAK cellsharvested than 3 days of bolus IL-2 and did so with substantiallyless toxicity. Efforts at improving IL-2/LAK therapy by combining it with other active agents such as «-Interferon, monoclonal antibodies, and chemotherapy against these diseases areongoing.

ACKNOWLEDGMENTS

We would like to thank Susan Strobl and Jo Ellen Tase of the LAKLaboratory for the preparation of the LAK cells and Judy Beacht forthe preparation of this manuscript. We are also grateful to SharonMarshall and her administrative staff, to all the Biological Response

Modifiers Program nurses for the excellent care provided to thesepatients, to Sue Marcus, R.N., and Joy Beveridge for their help in datamanagement, and to Greg Alvord, Ph.D., who provided helpful statistical analysis.

REFERENCES

1. Rosenberg. S. A., Lotze. M. T.. Muul. L. M., Leitman. S.. Chang. A. E..Ettinghausen, S. E., Matory, V. L., Skibber. J. M.. Shiloni. E., Vetto, J. T.,Seipp. C. A.. Simpson, C.. and Reichert. C. M. Observations on the systemicadministration of autologous lymphokine-activated killer cells and recombi-nanl interleukin-2 to patients with metastatic cancer. N. Engl. J. Med.. 313:1485-1491. 1985.

2. Rosenberg. S. A., 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. T.,Seipp. C. A.. Simpson. C. G.. and White. D. E. A progress report on thetreatment of 157 patients with advanced cancer using lymphokine-activatedkiller cells and interleukin-2 or high-dose interleukin-2 alone. N. Engl. J.Med., 316: 889-897. 1987.

3. West. W. H., Tauer. K. W.. Yannelli, J. R., Marshall. G. D., Orr, D. W.,Thurman. G. B., and Oldham. R. K. Constant-infusion recombinant interleukin-2 in adoptive ¡mmunotherapy of advanced cancer. N. Engl. J. Med..316: 898-905, 1987.

4. Fisher. R. !.. Coliman. C. A.. Doroshow, J. H., Rayner. A. A.. Hawkins, M.J.. Mier. J. W., and Wiernik P. A Phase II clinical trial of interleukin-2 pluslymphokine activated killer cells in metastatic renal cancer. Ann. Intern.Med.. 108: 518-523. 1988.

5. Dutcher, J. P., Creekmore. S.. Weiss, G. R.. Margolin, K.. Markowitz. A.B., Roper, M. A., and Parkinson. D. A Phase II study of interlcukin-2 andlymphokine activated killer cells in patients with metastatic malignant melanoma. J. Clin. Oncol., 7: 477-485. 1989.

6. Margolin, K. A., Rayner, A. A.. Hawkins. M. J.. Atkins. M. B., Dutcher, J.P., Fisher. R. I.. Weiss, G. R., Doroshow. J. H., Jaffe, H. S., Roper, M..Parkinson. D. R.. Wiernik. P. H., Creekmore, S. P.. and Boldt, D. H.Interleukin-2 and lymphokine-activated killer cell therapy of solid tumors:analysis of toxicity and management guidelines. J. Clin. Oncol.. 7:486-498.1989.

7. Wang, J.. Walle. A.. Gordon, B.. Novogrodsky, A., Suthanthiran. M., Rubin.A. L., Morrison. H., Silver. R. T.. and Stenzel. K. H. Adoptive immunotherapy for Stage IV renal cell carcinoma: a novel protocol utilizing periodateand interleukin-2 activated autologous leukocytes and continuous infusionsof low-dose interleukin-2. Am. J. Med.. 83: 1016-1023. 1987.

8. Thompson, J., Lee, D., Benz, L., Lindgren, C., Collins, C, Shuman, W. P.,Levitt, D., and Fefer, A. Influence of schedule of interlcukin-2 (IL-2) administration on therapy with interleukin-2 and lymphokine activated killer cells.Cancer Res., 49: 235-240, 1989.

9. Boldt. D. H.. Mills. B. J.. Gemlo, B. T., Holden, H., Mier, J., Paietta, E.,McMannis. J. D.. Escobedo. L. V.. Sniecinski, I.. Rayner, A. A., Hawkins.M. J., Atkins, M. B.. Ciobanu, N., and Ellis, T. M. Laboratory correlates ofadoptive immunotherapy with recombinant interleukin-2 and lymphokine-activated killer cells in humans. Cancer Res.. 48: 4409-4416. 1988.

10. Kovacs. E. J.. Beckner. S. K., Longo, D. L.. Varesio, L., and Young. H. A.Cytokine gene expression during the generation of human lymphokine-activated killer cells: early induction of interleukin-1 fi by interlcukin 2. CancerRes.. 49: 940-944. 1989.

11. Hornung. R. L.. Salup. R. R.. and Wiltrout. R. H. Tissue distribution andlocalization of IL2-activated killer cells after adoptive transfer in vivo. In: E.Lotzova and R. B. Herberman (eds.). The Role of IL2 and IL2 ActivatedKiller Cells in Cancer, pp. 245-258. Boca Raton. FL: CRC Press, Inc., 1990.

12. Beckner. S. K.. Maluish. A. E.. and Longo, D. L. Lymphokine-activatedkiller cells: culture conditions for the generation of maximal in vitro cytotox-icity from normal donors. Cancer Res.. 47: 5504-5508. 1987.

13. Cohen. P. J.. Lotze. M. T.. Roberts. J. R.. Rosenberg, S. A., and Jaffe, E. S.The immunopathology of sequential tumor biopsies in patients treated withinlerleukin-2. Am. J. Pathol.. 129: 208-216. 1987.

14. Weiss, G. R., Margolin. K.. Aronson, F. R., Sznol, M.. Atkins, M. B.,Gucolp, R., and Fisher, R. I. A randomized Phase II trial of continuousinfusion intcrleukin-2 or bolus injection of IL-2 plus lymphokine activatedkiller cells for advanced renal cell carcinoma. Proc. Am. Soc. Clin. Oncol..8: 131, 1989.

15. Dutcher, J. P., Gaynor, E., Boldt. D. M.. Doroshow, J., Rayner. A. A.. Sznol.M.. and Micr, J. Phase II study of high dose intravenous continuous infusionIL-2 and LAK cells in patients with metastatic melanoma. Proc. Am. Soc.Clin. Oncol., 8: 282. 1989.

16. Thompson, J.. Lee, D.. Benz. L.. Lindgren. C.. Collins. C.. Levitt, D.. andFefer, A. High dose continuous infusion IL-2 and LAK cell therapy for renalcell carcinoma and melanoma. Proc. Am. Soc. Clin. Oncol.. 8: 181, 1989.

17. Vetto. J. T.. Papa. M. Z., Lotze, M. T., Chang, A. E.. and Rosenberg, S. A.Reduction of toxicity of intcrleukin-2 and lymphokine-activated killer cellsin humans by the administration of corticosteroids. J. Clin. Oncol., 5: 496-503. 1987.

7350

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1990;50:7343-7350. Cancer Res   Jeffrey W. Clark, John W. Smith II, Ronald G. Steis, et al.   Interleukin 2 RegimenAnalysis of a Bolus Interleukin 2 and a Continuous Infusion Interleukin 2 and Lymphokine-activated Killer Cell Therapy:

  Updated version

  http://cancerres.aacrjournals.org/content/50/22/7343

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

  .pubs@aacr.orgDepartment 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://cancerres.aacrjournals.org/content/50/22/7343To request permission to re-use all or part of this article, use this link

on July 9, 2020. © 1990 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from